Hate OJolbgc of ^Agriculture Sit (lorttcU IntMeraitB Miiuta. BT. H- Stibrarg ^*'"lli llilimiltm '* disposal; a textbook, 3 1924 003 634 767 „ Cornell University Library The original of this book is in the Cornell University Library. There are no known copyright restrictions in the United States on the use of the text. http://www.archive.org/details/cu31924003634767 SEWERAGE AND SEWAGE DISPOSAL A TEXTBOOK AMERICAN SEWERAGE PRACTICE THREE VOLUMES BY METCAT.F AND EDDY Vol. I — Design of Sbwebs 747 pages, 6X9, 328 illus., 72 tables Vol. II CONSTBUOTION OF SeWERS 664 pages, 6X9, 181 illus., 81 tables Vol. III — Disposal of Sewage 877 pages, 6X9, 234 iUus., 210 tables SEWERAGE AND SEWAGE DISPOSAL A TEXTBOOK BY LEONARD METCALF AND HARRISON P. EDDY First Edition Second Impression McGRAW-HILL BOOK COMPANY, Inc. NEW YORK: 370 SEVENTH AVENUE LONDON; 6 & 8 BOUVERIE ST., E. C. 4 1922 coptbight, 1922, by the McGraw-Hill Book Company, Inc. PRINTED IN THE UNITED STATES OP AMERICA oj I'^H I 7 THE MAPUc PRESS - YORK PA PREFACE Since the three volumes of "American Sewerage Practice" were published in 1914-1915, the authors have been urged to prepare a single-volume abridgment of them for class use in engineering schools. The authors were reluctant to do this, because of two conditions. The first is the great variation in the time devoted to sewerage engineering in different schools. This has the result that a- textbook so brief as to verge on the super- ficial from the viewpoint of one class may be too detailed for another. The second condition is the fact that some of those features of sewerage engineering which in practice it is most important to master are apparently held by those laying out college courses to deserve little attention from students. For example, a wise estimate of the quantities of sewage and storm water to be handled is the basis of design of a sewerage system, yet the methods of preparing such estimates and actual practice in their use receive relatively little attention in college courses. Furthermore, the best arrangement of the main sewers in a sewerage plan may depend largely upon the treatment, if any, that the sewage must receive before disposal and upon the place of disposal, subjects involving a knowledge of the nature of sewage and the changes which it undergoes under natural con- ditions or those involved in specific methods of treatment. Yet some college classes in sewerage engineering receive no instruc- tion in sewage treatment. As the requests for a textbook continued, the authors finally decided to prepare a book giving the information which they consider it desirable for the young student to acquire before taking up work in this field. In a sense, .they have felt under obligations as professional engineering specialists to do their part in smoothing the way for the first steps taken by engineering students in this branch of engineering. This book can be used in two ways. The text can be read and mastered within the time allotted to the subject even when the course is very short. Where more time is available, the authors believe that the student will gain a much better grasp of the fundamentals of the subject by devoting most of this addi- VI PREFACE tional time to the solution of problems. For example, a student should, if possible, test his ability to predict the growth of popu- lation by forecasting the 1920 population of cities by the methods described in Chapter II, and checking his figures by the 1920 census returns. He should determine the carrying capacity of sewers of different cross-sections on the same grades, and the sizes needed when the plan remains the same, but the grades are different. In short, he should test his grasp of the qualitative information which must necessarily form the larger part of a sewerage textbook, by applying this information to quantitative problems based on assumed conditions or data from places with the local conditions of which he is familiar. He should also lose no opportunity to see sewerage construction in progress and sewage treatment works in operation. There is little in this textbook not covered in much greater detail in "American Sewerage Practice." The three volumes of that treatise have elaborate indices, so that it seemed unneces- sary to insert references to the treatise in this textbook. The student is assumed to have access to textbooks on hydraulics, masonry and reinforced concrete. No cost data are given, because the great changes in costs induced by the war make it necessary to employ older data with a great deal of caution. They must be supplemented, too, by practical knowledge of current changes which very few undergraduate students possess. A study of fluctuations in commodity prices since the war, based upon Dun's and Bradstreet's characteristic numbers, will be helpful to the student, however, in connecting up, approximately, pre-war a,nd post-war costs, and in bringing home to him the necessity of recording with his cost data the unit prices paid to labor and for materials upon the construction work under con- sideration. In this connection the accompanying diagram of fluctuations in commodity prices, in percentages of 1910 prices, as determined by Dun's index numbers, prepared by the authors for their office use, may be of interest. Dun's index numbers represent the weighted average wholesale quotations of cost per pound, on the first business day of each month, of about 300 commodities, the actu'al cost of each commodity being weighted according to its annual per capita consumption in the United States. The reasons for making the percentage comparison with the prices of 1910 as a base were that the figures of that year approximated the average condition over the period covered PREFACE Vll xspuioiei J-0+U3D -isd o o o S ro cvj _ O OJ f\j OJ CJ «puiOI6iP+"33-'^d Vlll PREFACE by a number of different records, were uninfluenced by the war, and yet were near enough the war period to approximate, though they were slightly below, the costs prevailing just before the war. Reference to sewage disposal for detached residences and institutions has been omitted in this book because the principles involved are fully explained. The practice in their application varies not only with the local conditions, the size of the project and the influence upon design of the necessity for simplicity, but also with the standards set by the local health authorities. Standards have been developed by the U. S. Public Health Service, by the State Colleges, and by State Boards of Health of various states The authors are engineers, not teachers, hence this text book reflects the engineer's rather than the teacher's viewpoint. They will appreciate the more, therefore, any suggestions for increasing its usefulness to undergraduate students and their instructors. Constructive criticism is invited, because through it the experi- ence of the authors and other engineers in designing, building and operating sewerage and sewage treatment works, can be made more helpful to the students who will eventually carry the responsibihty for such undertakings. The authors take pleasure in giving recognition to Mr. John M. Goodell, for many years Editor of "Engineering Record," who assisted them in the preparation of "American Sewerage Prac- tice" and who has done the greater part of the work of condensing its three volumes into this book. Credit is also due to the junior partners and members of the staff of Metcalf and Eddy, and to the publishers, for their helpful assistance. As the book reflects the past experiences and current opinions of many practicing engineers, it should be of service to the student. ,. „ „ Leonard Metcalf. 14 Beacon Street, t. t-, Boston, Mass. Harrison P. Eddy. December, 1921. CONTENTS CHAPTER I Page Preface y General Features op Sewerage and Sewage-Treatment Works 1 Definitions — Sewer Appurtenances — General Conditions Govern- ing Sewer Systems — Respective Usefulness of Separate and Com- bined Sewers — Influence of Topography on Sewerage Plan — Sewage Disposal and Treatment — Why Sewage Treatment is Necessary at Times — Influence of Method of Disposal and Treat- ment upon General Sewerage Plan — Municipal Liability for Inade- quate Sewers. CHAPTER II Quantity of Sewage. . 23 Population Studies Including Assumed Uniform Percentage, Assumed Curvilinear and Assumed Arithmetical Rates of Growth, and Assumed Decreased Rate of Growth — Increase in Area — Density of Population — Proportion of Municipal Water Supply Reaching Sewers — Water Consumption in Cities — Rate of Water Consumption in Different Parts of a City — Fluctuations in Water Consumption — Ground Water — Sewage per Day — Sewage per Hour — Sewage from Residential Districts — Sewage from Com- mercial Districts — Sewage from Industrial Districts — Provision for Storm Water in Intercepting Sewers — Caution Regarding Allowance for Storm Water Entering Separate Sewers — The Design of a Separate Sewer System; Problem Worked Out in Detail. CHAPTER III Quantity of Storm Water .... 55 Automatic Rain Gages — Intensity of Precipitation — Form of Rainfall Curve — Phenomenal Rainstorms — Rational Method of Estimating Run-off — Time of Concentration — Run-off Factor — Values of Run-off Factor — Relative Adequacy of Laterals and Mains — Some Examples of the Use of the Rational Method, Including Hypothetical Problems and tlieir Solution — Run-off Formulas — Flood Flows from Large Areas — Sewer Gaging — Gages — Setting Sewage Gages. CHAPTER IV Hydraulics of Sewers. . . 92 Quantities of Sewage to be Provided for — Velocity of Flow — Trans- porting Power of Water — Minimum Velocities — Examination of ix X CONTENTS Page Sewer Design with Reference to Minimum-flow Conditions — Maximum Velocities — Formulas for Velocity — The Limitations of Kutter's Formula — Flow in Partly Filled Sections — Discharge and Velocity at Different Depths — Selection of Sewer Sections — Compensation for Curvature — Velocity Head — Recovery of Velocity Head — Twelve Problems. CHAPTER V Sewer Appubtenances and Special Structures ... . 126 Street Inlets — Catch Basins — Castings for Inlets and Catch Basins — Manholes — Drop Manholes — Wellholes — Flight Sewers — Manhole Castings — Steps — Junctions — General Features of In- verted Siphons — Size of and Velocity through Inverted Siphons — Means of Flushing Inverted Siphons — Manholes or Clean-out Chambers on Inverted Siphons — Materials of Construction of Inverted Siphons — Size of Pipe for Inverted Siphons — Essential Principles in Design of Inverted Siphons — Solution of a Siphon Problem Including Design of Inlet and Outlet Chambers — Bridges — Flushing Devices including Flushing from Brooks, Flushing Manholes and Automatic Flush Tanks — Regulating Devices including Regulators, Storm Overflows, and Tide Gates — Outlets — Pumping Sewage — Storage and Screening — Waltham Pumping Station — Summer Street Station, Boston. CHAPTER VI Preliminary Investigations, Surveying, Inspection and Excava- tion ... 178 Field Work — Office Work — Reference Marks — Transferring the Line into the Trench — Notes — Quantity Surveys, Location of Specials, etc. — Structural Surveying — Inspection of Excavation, Concrete Materials, Cast-iron Pipe, and Vitrified Pipe — Classifica- tion of Excavation — Opening the Trench — Hand Excavation — Trenching Machinery — Rock Excavation — Hand Drilling — Spac- ing of Drill Holes — Cleaning Drill Holes — Drilling by Machinery — Hammer Drills — Explosives and Blasting — Freezing of Dynamite — Exploders and Fuses — Pumping — Underdrains — Quantities of Excavation. CHAPTER VII Sheeting and Bracing op Trenches 212 Shoring Trenches — Behavior of Different Materials in Excavation — Earth Pressures — Vertical Sheeting — Box Sheeting — Poling Boards — Stay Bracing — Skeleton Sheeting — Bracing Curves — Jetting — Practical Suggestions on Timbering — Lumber and Tools. CONTENTS xi CHAPTER VIII Page Pipe, Brick and Concrete Sewers ... . . . . 237 Loads — Strength of Pipe Sewers — Vitrified Sewer Pipe — Rejec- tion of Pipe — Concrete Sewer Pipe — Laying Sewer Pipe — Cement Joints — Sulphur and Sand Joints — Asphalt and Tar Joints — Lin- seed Oil Jointing Preparations — Platforms and Cradles — Branches — Backfilling — Sections for Concrete and Brick Sewers — Analysis of Masonry Arches — Influence of Construction Conditions on Section — Cost of Excavation and Materials — Stability — Imper- viousness — Materials for Arches — Lining for Concrete Construction — Brick Laying — Profiles and Centers for Brick Sewers — Placing Concrete — Placing Reinforcement — Forms in General — Building Concrete Sewers — Full Round Wooden Forms — Forms for Sewer Built in Two Operations — Manholes. CHAPTER IX Chemical and Biological Characteristics of Sewage /^ 280 Sampling — General Characteristics of Sewage — Solids or Residue on Evaporation — ^Suspended and Dissolved Solids — Settling Solids — Colloidal Matter — Fixed Residue, .Fixed Solids or Mineral Matter — ^Loss on Ignition, or Organic Matter — Nitrogen Cycle — Organic or Kjeldahl Nitrogen — -Albuminoid Ammonia — Free Ammonia — Nitrites and Nitrates — Oxygen Consumed — Putresci- bility Tests — Biochemical Oxygen Demand — Dissolved Oxygen — Chlorine — Appearance — Odor — Changes in Sewage Caused by Bacteria — Number of Bacteria in Sewage — Breaking Down of Complex Substances — Decomposition of Sewage with Diminution of Oxygen — Oxidation Processes — Detection of Contamination by Test Bacteria — Typhoid Fever — Plankton — Composition of Sew- age, Including Sources of Total Solids, Weight of Constituents in Grams per Capita, and Suspended Matter. CHAPTER X ,/ Sewage Disposal by Dilution ... .... 310 Methods of Sewage Treatment — Nature of Disposal by Dilution — Decomposition of Organic Matter — Investigations by Stearns and Goodnough — Pollution of Ohio Rivers — New York Harbor Investi- gation — Temperature — Fresh and Salt Water — Tides — Floats — Depth of Outlet in Tidal Waters— Salinity— Effect of Wind- Milwaukee Investigation of Lake Currents — Temperature Changes in Deep Ponds — Quiet and Running Water — Basic Information for Planning Dilution Projects — Sewage Dilution at Boston — Dilution at Cleveland — Chicago Drainage Canal. xii CONTENTS CHAPTER XI Page Geit Chambers, Racks and Screens 330 Objects of Grit Chambers, of Racks and of Screens — Theory of Grit Chambers — Quantity and Character of Grit — Design of Grit Chambers — Cleaning Grit Chambers — Theory of Screening — Racks or Parallel-bar Screens — Fixed Racks — Cleaning Fixed Racks — Cage Racks — Wing Racks — ^Link-bar Racks — Drum Screens — Disc Screens — Belt Screens — Screenings. CHAPTER XII Sedimentation and Septic Tanks; Chemical Precipitation . 351 Theory of Sedimentation — Types of Sedimentation Basins — Deten- tion Period — Design of Horizontal-flow Tanks — Sedimentation in Vertical Tanks — Design of Vertical-flow Tanks — Septic Action — Septic Sludge — Scum — Volume of Sludge for which Provision should be Made — Number of Septic Tanks — Roofs — Inlets and Outlets — Bottom Slopes — Chemical Precipitation — Quantity of Lime required — Slaking Lime — Mixing Chemicals and Sewage — Detention Period — Design of Precipitation Tanks. CHAPTER XIII The Imhoff or Emscher Tank ... 377 Diffusion between Chambers — Quantity of Sludge — Sludge Storage Capacity Required — Radial-flow Tanks — Parallel-flow Tanks — Inlet and Outlet Channels — Sedimentation Chamber — Gas Vents — Sludge Removal. CHAPTER XIV Activated Sludge ... .... 395 Description of Process — Theory of Purification — Establishment and Maintenance of Sludge Activity — Physical Properties of Activated Sludge — Reactivating Sludge — Bacterial Digestion of Sludge — Winter Operation — Extent of Purification Secured — Preparatory Treatment of Sewage Required — Proportion of Return Sludge — Period of Aeration — Quantity of Air Required — Measure- ment of Sludge — Importance of Measurements of Sewage, Sludge and Air — Aeration Tanks — Air Diffusers — Power Required for Compressing Air — Sedimentation Tanks — Volume and Character of Sludge Produced — Disposal of Sludge — Preparation of Sludge for Market — Management. CHAPTER XV Sludge . ... 427 Volume of Sludge — Measuring Sludge — Characteristics of Sludge — Removing Sludge from Tanks after Draining off the Sewage — Sludge Removal without Interruption of Tank Service — Disposal of Sludge — Dewatering Sludge Mechanically — Drying Sludge by Heat — Fertilizing Value of Sludge. CONTENTS xiii CHAPTER XVI Page Contact Beds 448 Filling and Draining — Standing Full — Standing Empty — Matur- ing of Beds — Loss of Capacity — Multiple Contact — Kinds of Contact Material — Size of Contact Material — Depth and Drainage of Contact Beds — Size of Beds — Construction — Automatic Control — Air-lock Feeds — Timed Siphons — Operating Results. CHAPTER XVII Trickling Filters . .... 467 Theory — Unloading of Trickling Filters — Results — Character of Filtering Medium — Size of Filtering Medium — Depth of Trick- ling Filters — British Opinions on Depths — Investigations at Lawrence Experiment Station — Motion of Sewage in Bed — Revolving Distributors — Travelling Distributors — Dash Plates — Spraying Nozzles — Efficiency of Fixed Nozzles — Spacing of Noz- zles — Relative Advantages of Intermittent and Continuous Operation — Distribution Systems — -Underdrainage Systems — Ven- tilation — Filter Units — Preliminary Treatment — Sedimentation of Effluent — Siphonic-dosing Apparatus — Head-varying Valves — Organic Growths and Insects — Comparison of Trickling Filters and Contact Beds. CHAPTER XVIII Intermittent Sand Filters. .50^ Mechanical Straining — Oxidation — Filter Sand — Effective Size and Uniformity Coefficient — Mechanical Analysis of Sand — Depth of Bed — Rate of Filtration — Size, Shape and Grouping of Beds — Surface of Bed — Banks and Roads — Underdrainage — Distribution of Sewage — Automatic-dosing Apparatus — Air-controlled Siphonic Apparatus — Mechanically -controlled Siphonic Apparatus — Mechanical-dosing Apparatus — Efficiency of Intermittent Sand Filters — Odors — Operation — Clogging of Beds — Objections to Disturbing Surface of Filters — Removal of Dirty Sand — Resting — Replacing Sand — Maintenance in Winter — Winter Cleaning. CHAPTER XIX Disinfection op Sewage and Sewage Effluents 62' Disinfection by Acids — Disinfection by Copper Sulphate — Disinfection by Chlorine and its Compounds — Disinfection with Hypochlorites — Providence Hypochlorite Plant — Emergency Plant — ^Liquid Chlorine. CHAPTER XX Irrigation with Sewage ... .54: English Experience — French Sewage Farming — German Sewage Farming — American Sewage Farming, XIV CONTENTS CHAPTER XXI Page Cost Estimating. . 551 Bookkeeping vs. Cost-keeping — Purposes of Cost-keeping — Practical Advantages of Cost-keeping as Developed by a Success- ful Building Concern — The Methods of Construction — Day Work vs. Contract — Simplicity— Local Conditions — Two Methods of Estimating — ^Contract Prices to be Used with Caution — Weighted Price Method of Estimate — The Ratio Method — Effect upon Cost Estimates of Length of Construction Period — Extra Cost Involved by Complexity and Novelty of Design — Classification of Items — Methods of Applying Data on Cost — Progress Reports — Final Estimate. Index. . 569 SEWEBAGE AND SEWAGE DISPOSAL A TEXTBOOK CHAPTER I GENERAL FEATURES OF SEWERAGE AND SEWAGE- TREATMENT WORKS Sewerage provides the means for removing the liquid wastes from a building to some more or less remote place of disposal. In its simplest form, it is merely the extension of the discharge pipe of the plumbing system to a neighboring body of water, to a cesspool,^ or to some plot of land which will absorb the wastes. Each of these methods of disposal may be objectionable under some conditions. The water into which the sewage is discharged may be contaminated by the sewage so as to lose part or all of its usefulness to others; the cesspool may overflow and become an offensive breeding place for flies; and the land may be made foul by the sewage discharged upon it. From the technical and sanitary viewpoints, the construction and operation of these small isolated sewerage systems calls for the same kind of knowledge involved in the design and operation of the municipal sewerage plants. In taking up the study of sewerage and sewage treatment for the first time, the clearest conception of the field to be covered can be formed from an examination of the works in a city, because of their greater size. Definitions. — At the beginning of such an examination, a clear conception of the technical terms employed in sewerage engineering is helpful. These terms are used very loosely by the general public and engineers should avoid such carelessness, in order that what they say and write may be free from ambi- guity. The Committee on Sewerage and Sewage Disposal of the American Public Health Association, recommended certain defi- nitions which will be quoted in this chapter: 1 Cesspools are usually shallow pits In the ground, lined with stone set without mortar, from which the Uquids percolate into the ground and mingle with the ground water. They are usually not wholly satisfactory except when they are surrounded by porous material, such as sand, and great care should be taken that the liquids leaching from them shall not contaminate wells in the neighborhood. 1 2 SEWERAGE AND SEWAGE DISPOSAL "Sewage is a combination of the liquid wastes conducted away from residences, business buildings and institutions, together with those from industrial establishments; and with such ground, surface and storm water as may be present. " Domestic sewage is that from residences, business buildings or institutions. "Industrial wastes^ are the liquid wastes resulting from the processes employed in industrial estabhshments. "Surface water is that portion of the precipitation which runs off over the surface of the ground. "Storm water is that portion of the precipitation which runs off over the surface during a storm and for such a short period following a storm as the flow exceeds the normal or ordinary run-off. "Ground water is that which is standing in, or passing through, the ground. "A sewer is a conduit for carrying off sewage. "A common sewer is a sewer in which all abuttors have equal rights of entrance and use. "A house connection is a pipe leading from a building to a common sewer. "A lateral sewer is a sewer which does not receive the sewage from any other common sewer. "A sub-main^ or branch sewer is a sewer into which the sewage from two or more lateral sewers is discharged. "A mam or trunk sewer is a sewer into which the sewage from two or more sub-main sewers is discharged. "An outfall sewer is a sewer extending from the lower end of the collecting system to a point of final discharge into a body of water, or to a sewage- treatment plant. "A separate sewer^ is a sewer intended to receive domestic sewage and industrial wastes without the admixture of surface or storm water. "A combined sewer is a sewer intended to receive domestic sewage, in- dustrial wastes and surface and storm water. "An intercepting sewer is a sewer generally laid transversely to the general sewer system to intercept all the sewage collected by the sewers of a separate system or the dry-weather flow of sewage and such additional surface and storm water as may be determined from a combined system. "A relief sewer is a sewer designed to carry a portion of the flow from a district already provided with sewers of insufficient capacity, and thus prevent overtaxing the latter. "A storm-water overflow sewer is a sewer designed to carry the excess over a certain volume of combined sewage from a main or intercepting sewer to an independent outlet. "A drain is a conduit for carrying off storm water, surface water, subsoil or ground water. "A storm drain is a conduit for carrying off surface water and storm water. 1 In the opinion of the committee, the term "industrial wastes" is preferable to "trade wastes." 2 The committee is of the opinion that preference should be given to the term "submain sewer" rather than to "branch sewer." 3 In the opinion of the committee, the use of the term "separate sewer" is preferable to "sanitary sewer," which latter term should be discontinued. SEWERAGE AND SEWAGE-TREATMENT WORKS 3 "A land drain is a conduit for carrying off subsoil or ground water, and for draining land. "A sev)er system is the collecting system of sewers and appurtenances, together with such small pumping stations as may be required to lift the sewage from low-level districts. "A combined system is a system of combined sewers. ''A separate system is a system of separate sewers. "Sewerage works comprise the sewer system, main pumping stations, treatment works, means of disposal of effluent and sludge, and all other works necessary to the complete collection, treatment and disposal of sewage." Before continuing the committee's definitions it is desirable to supplement those already given with others. Plumbing inside a building, as the term is ordinarily used, comprises the pipes and fixtures used to supply water and to remove it when used. Plumbing connects with both the sewer- age and water-supply systems of a city, just as the capillaries in the human body connect the veins and arteries. The main or house drain, or house sewer, better characterized as the house connection, is the pipe connecting the sewers or drains within the house with the common sewer in the street. It should not be too small. The wastes removed from a store, apartment house or office building are usually much the same as those from houses. Sometimes large quantities of dirt and grease are washed from an automobile into the house connection from a garage; bundles of rags, towels, and occasionally old hair brushes and even larger objects get into the pipes and stop them if they are not large enough to give such rubbish free passage. Because of the misuse of plumbing neither the main drain nor the house connection should be less than 4 in. in diameter and 5 or 6 in. (the most common size) is a preferable size for the house connection, which is more difficult to clean than the main drain. The main trap is a small inverted siphon sometimes placed in the main drain near the building wall to keep air from the sewer system out of the building. The bend of the trap is filled with water while the house is occupied and this water acts as a seal against the passage of air through the trap. There has been a great deal of controversy among engineers regarding the use of a main trap. When sewer systems were comparatively novel and often badly designed, so that solids collected in them and gave off foul odors while decomposing. 4 SEWERAGE AND SEWAGE DISPOSAL there was a general belief that sewer air was dangerous to health and should be excluded from buildings by the use of the main trap. In the course of time sewers were better designed and built and examinations of the air within such sewers showed no traces of injurious gases or bacteria. Meanwhile the plumbing within buildings was improved and building codes compelled the trapping of all fixtures and the venting above the roof of all stacks. Many sanitary engineers held, therefore, that it was desirable to ventilate the sewers through the main drains and stacks, by omitting the main traps. At first there was great opposition to this, particularly by physicians serving as health officers, but the opposition has practically died out with the spread of definite knowledge regarding sewer air. Traps are now used in cities where all sorts of wastes are discharged into the sewers and there is danger of odors coming from some of these wastes before they leave the sewer system. But in residential districts where the sewers are in good condition, it is customary to omit the main traps. Stacks are vertical pipes rising from the main drain. A soil pipe, strictly speaking, is any pipe through which pass the wastes from a water closet, but in the plumbing trade the term also designates cast-iron pipe of any size used in the plumbing system. An inverted siphon is a portion of a sewer which drops consider- ably below the average grade and then rises again, in such a way as to run under pressure. A waste pipe is one receiving wastes from other fixtures than water closets. A dead end is the upper extremity of a sewer, above the point where the first house connection or other sewer enters it. Roof water is the storm water running off roofs. Street wash is the water flowing from the surface of a street into a sewer or drainage system. Street wash may be rain water or water used in flushing the streets from hydrants or sprinkling carts, or melted snow. The first flow from a street after dry weather is often grossly polluted and liable to cause offensive conditions if not removed promptly. Run-off — in a narrow sense — is the portion of the storm water that reaches the sewers and drains. The term is also used to denote the portion of the rainfall that passes away in brooks and streams from a designated area. SEWERAGE AND SEWAGE-TREATMENT WORKS 5 Infiltration or leakage Is tlie ground water that finds its way into sewers. In sewerage engineering "leakage" usually has an opposite meaning to its significance in water supply, for it designates the water entering a conduit, rather than the liquid escaping from it. Infiltration sometimes presents problems difficult for the engineer to solve. The authors have known storm and ground water to leak into sewers to such an extent that the normal flow of 300,000 gal. per day during dry weather was increased to 3,000,000 gal. by heavy rainfall, although all storm and ground water was theoretically excluded from the sewers. Underdrains are drains to remove ground water from engineer- ing works above them, such as sewers, pavements, filters and foundations. Pipe sewers are those constructed of relatively short lengths of pipe, jointed together. A cradle is a support for a sewer, usually constructed of wood or concrete, which distributes the weight of the sewer over a wider portion of the bottom of the trench or other support than the sewer itself would cover. The invert of a sewer is technically the lowest point of its interior at any cross-section; the term is also applied loosely to the whole inside bottom surface of a sewer. The crown of a sewer is the highest point of its interior at any cross-section. Sewer Appurtenances.— There are a number of appurtenances of sewers which are so generally used that they are enumerated here. The Committee on Sewerage and Sewage Disposal of the American Public Health Association recommended the following definitions in 1917: "A manhole is a shaft, or chamber, leading from the surface of the ground to the sewer, large enough to enable a man to gain access to the latter. "A lamphole is a small vertical pipe or shaft leading from the surface of the ground to the sewer, for admitting a lantern or reflected light for pur- poses of inspection. "A wellhole or drop manhole is a vertical shaft in which sewage is allowed to fall from one sewer to another at a lower level. "An inlet is a connection between the surface of the ground and a com- bined sewer or drain for the admission of surface or storm water. "A catch basin is a chamber inserted on an inlet to prevent the admission of grit and other coarse material into the sewer or drain. "A flush lank is a tank in which water or sewage is accumulated to be quickly discharged later, for the purpose of flushing the sewer. SEWERAGE AND SEWAGE DISPOSAL "A regulator is a device for controlling the quantity of sewage admitted to an intercepting sewer. "An outlet is the end of a sewer or drain from which its contents are finally discharged. "A storm overflow is a weir, orifice or other device for permitting the discharge from a combined sewer of that portion of the storm flow in excess of that which the sewer is designed to carry." Other appurtenances which are quite generally used and need definition here are the following: Fittings are forms of pipe in which the cylindrical form between the spigot and socket ends is modified, in such a way as to accom- plish a specific purpose, such as the joining of two or more sewers, or changing the diameter or direction of the sewer. They are best defined by the diagram of their outlines, Fig. 1. Some of them. 90 Deg. Curve K^ <^ 45 Deg. Curve O B ff o Double Y Branch Y Branch T Branch DoubleT Branch SIcinf Fig. 1. — Vitrified pipe fittings. Reducer Increaser such as the slant, are also used in connecting pipe sewers with masonry or concrete sewers. Such fittings were formerly called "specials" because each engineer designed his own forms, but as they are now standard products "fittings" is a better term for them. A chimney is a vertical pipe, usually encased in 6 in. or more of concrete, which is sometimes run up from every branch or slant of a sewer in a deep trench, to afford an easy connection for a house sewer. These chimneys usually terminate at a uniform depth below the surface. A manhole casting comprises a heavy cast-iron frame set on top of the manhole masonry and a cover or Hd, closing the man- hole, fitting rather loosely on the top of the frame. If it is desired to ventilate the sewer through the manhole, as when main traps are used, the cover may have holes through it and it is then sometimes called a "perforated" or "ventilating" cover. The SEWERAGE AND SEWAGE-TREATMENT WORKS 7 present tendency is to minimize the number of perforated covers, as they permit the entry of storm water and earth, grit, and other foreign material into the sewer. General Conditions Governing Sewer Systems. — Experience indicates that in order to avoid frequent clogging, laterals should not be smaller than 8 in. in separate sewerage systems, except in very small towns. In most cases the smallest size suitable for a storm-water drain or combined sewer is 12 in. The size of the sub-main and main sewers should be selected so that they will be able to carry off the domestic sewage determined by the methods explained in Chapter II, the industrial wastes estimated after making an investigation of the local conditions, and, in some cases, the storm water determined by the methods explained in Chapter III. To save needless expense both sewers and house connections should be laid no deeper than necessary, but in order to carry off the sewage from the laterals the sub-main sewers must be deeper than local house drainage alone demands. The grades of all sewers should be sufficient to give an adequate scouring velocity to the sewage, as explained in Chapter IV. Anything more than this in relatively flat country involves useless increase in cost; anything less, increase in cost of maintenance. The grades must be flat enough to allow the main sewers to discharge into the out- fall sewer without pumping, if this is practicable; but it is some- times necessary, in order to avoid excessively deep sewers, to construct small pumping stations where the main sewers reach considerable depths and to lift the sewage at these small stations to such a height that it will be able to flow away by gravity through sewers at reasonable depths. The elevations and grades of the outfall sewers are fixed by the local conditions. The net- work of lateral, sub-main and main sewers must be adjusted to them. It is in making such connections of lateral, sub-main and main sewers that drop manholes often prove useful. The main stems of a sewerage system are the outfall sewers and the larger main sewers. In small cities there may be but one outfall sewer, while in large cities there may be several, sometimes discharging independently and sometimes radiating like the ribs of a fan. With combined sewers, the volume of storm water is generally so much greater than that of house sewage and industrial wastes that the capacity of the large sewers is determined by it. The only influence of the volume of 8 SEWERAGE AND SEWAGE DISPOSAL house sewage and industrial wastes in such a case is to modify the shape of the invert of some sewers, so as to afford a self -cleansing channel for the dry-weather flow, a subject explained in Chapter IV. Where one sewer joins another, particularly when they are large, the angle between the axes of the two currents at the junc- tion should be as small as possible in order to prevent any check- ing of the velocity of flow. For the same reason, the elevation of the surface of the sewage in both sewers at the junction should be as nearly the same as practicable. The cost of outfall sewers, usually a large item in the cost of the sewerage system, may be reduced by various expedients. Sometimes there are several points along a main sewer of a combined system where storm water can be discharged into short channels or conduits leading to neighboring bodies of water. In London part of the storm water entering the outfall sewers is pumped from them mto the Thames. These expedients, by relieving the outfall sewer of excessive quantities of storm water, enable its size and hence cost to be kept within reasonable limits. Another expedient is to run the combined sewers to neighboring bodies of water and provide intercepting sewers near their outlets which will take from them all the dry- weather sewage and conduct it to a more distant point where its discharge will be less objectionable. In this way the stronger dry-weather flow only is carried to the more distant disposal point, the more dilute storm-water flow finding its way directly to the nearest water course. Intercepting sewers are used not only as explained in the previous paragraph but also to collect all or part of the sewage above a given elevation, either to permit flow by gravity to the place of treatment or disposal, or to prevent the accumulation of so much sewage at lower levels that the intercepting sewers there would be excessively large. Respective Usefulness of Separate and Combined Sewers. — A combined sewerage system requires but a single sewer in a street, thus saving a portion of the space required by the two conduits of the separate system. This is important in streets congested with underground structures. Usually the quantity of domestic and industrial sewage is not greater than the margin of error in estimates of the quantity of storm water to be removed. There- fore the inclusion of the sewage with the storm water does not SEWERAGE AND SEWAGE-TREATMENT WORKS 9 require a combined sewer appreciably larger than the drain for serving the same district adequately. Even if it does require an appreciably larger section, the additional cost of building the combined sewer as compared with the drain will be practically negUgible in most cases. Hence the cost of a combined system is less than that of both a separate system and a drainage system, where complete underground removal of both storm water and sewage is necessary. It may be undesirable to allow storm water from urban dis- tricts to contaminate neighboring bodies of water, not only because of sanitary objections but sometimes because of the disagreeable appearance of accumulations of refuse or deposits of sediment forming on the bottoms of these water courses, floating matter and sleek. Such conditions in any case make necessary an extensive drainage system which might as well be made to remove sewage, thus avoiding the additional expense of dual systems. Where a separate system of sewers is paralleled by a drainage system and roof water cannot be allowed to flow over the surface of the ground, a system of collecting drains is required outside the buildings in addition to the usual inside plumbing, or two systems of plumbing are necessary within the buildings, one to take care of the sewage and the other to remove the roof water. This is not only expensive to the property owner but it introduces a dangerous condition. There have been sewerage and drainage systems administered so badly that the surface water was dis- charged by incorrectly laid connections into sanitary sewers, forcing the sewage back into the cellars and out through manholes upon the streets; in other cases sewage has been discharged into drains resulting in unnecessary pollution of waters into which they emptied. Notwithstanding these disadvantages of the separate system there are certain conditions under which it may be advantageous. Dr. Hering summarized' some of them as follows: "The separate system is suitable — "Where rain-water does not require extensive underground removal and can be concentrated in a few channels slightly below the surface, or where it can safely be made to flow off entirely on the surface. Such conditions are found in rural districts where the population is scattered, on small or at least short drainage areas, and on steep slopes or side hills. 1 Report to National Board of Health on European methods of Sewerage and Sewage Disposal, 1881. 10 SEWERAGE AND SEWAGE DISPOSAL "Where an existing system of old sewers, which cannot be made available for the proper conveyance of sewage, can yet be used for storm-water removal. "Where purification is expensive, and where the river or creek is so small that even diluted sewage from storm-water overflows would be objectionable, especially when the water is to be used for domestic purposes at no great distance below the town. "When pumping of the sewage is found too expensive to admit of the in- creased quantity from intercepting sewers during rains, which can occur in very low and flat districts. "Where it is necessary to build a system of sewers for house drainage with the least cost and delay, and the underground rain-water removal, if at all necessary, can be postponed. "The principle of separation, although often ostensibly preferred on sanitary grounds, does not necessarily give the system in this respect any decided advantage over the combined, except under certain definite condi- tions. Under all others, preference will depend on the cost of both con- struction and maintenance, which only a careful estimate, based on the local requirements, can determine." Influence of Topography on Sewerage Plan. — The arrangement of the small and large sewers which make up a sewer system is influenced largely by the topography of the city. In a large city situated on a flat plain without any neighboring rivers or lakes into which the sewage can be discharged without elaborate treatment, the radial system may prove best. This has its most elaborate development in Berlin, where it was introduced by Hobrecht. The city is divided into a number of sectors and the sewage of each sector is carried outward by pumping to its independent disposal farm, or the trunk sewers of two or more sectors may be connected to a farm. The advantage of this system is that most of the sewers are likely to be of adequate capacity for a long period, and the large expensive sewers are reduced to their minimum length. In most cases such an arrangement is rendered impracticable by the existence of hills, water courses and other topographical conditions. Usually, moreover, old sewers complicate the pro- blem, for it is always desirable to utilize existing structures so far as practicable. Only in rare cases does the engineer have an opportunity to design a complete sewerage system for a large city, as was the case in New Orleans and in Baltimore. In Baltimore, where the sewage had to be taken 5^ miles outside the city for treatment, it was apparent that the storm water should be collected separately, for there was no objection SEWERAGE AND SEWAGE-TREATMENT WORKS 11 to its discharge into the nearest water courses adapted to re- ceiving it. The city is intersected by four streams, which dis- charge into branches of the Patapsco River. One of these streams receives so much foul run-off that it has been covered over; the others are open. The Patapsco and its branches are tidal arms of Chesapeake Bay. The drainage area was divided into 28 districts, and the storm-water drains in each one were planned independently of the rest, to fit the topography and arrangement of streets in the best way. These drains were kept as close to the surface as possible, in the interest of economy and \ ^^Hampdin/ResmWr 1 \ =» ) MomRiLia \ i O ^^%ia-(i ^^ ?r Reservoir 'ttsft 'ffi^^ UKE® K~ '^^ CUFTOH Reservoir •^^. S^SEte ^i faSlj^j il ^\^ S Mblbmok V \ Testing Station J^ S, ^•Fbn^MainK' Sewer \ i '""J^^S^X. JcH Sewer W^ 7M \ \ll'i ^__/«^ \awm>,S^™ \ Hw^^^ ^^^ a^f-aiferrt ^PV/.^ i m mo I.I.I 6twage to be Pumped fmm Area ij ^<&^^^ ' Stations la with Shatiec/ eaum/ary. (L^ '''O^Wi fmpscomnn Fig. 2. — Baltimore intercepting sewer system. in order not to force the sewers so low that it would be difficult to connect the houses with them. In one low-lying district where the drainage problem was particularly difficult, the plans called for raising the street grade and building a drain to carry the storm water into the Patapsco River instead of a nearer stream which was liable to have its surface raised considerably during floods, a condition which might cause a surcharge of the drains emptying into it. The removal of the house sewage was a much more complicated problem. Part of it comes from districts which are high enough to enable the sewage to flow by gravity to the treatment works, but a large part has to be pumped. The dividing contour line between these two service districts was determined by two 12 SEWERAGE AND SEWAGE DISPOSAL factors, the elevation at which the sewage must be discharged at the treatment works and the minimum safe grade of the outfall sewer from the city to the works. The accompanying plan of the intercepting sewers, Fig. 2, from Engineering Record, Dec. 5, 1908, shows where the outfaU sewer reaches the eastern boundary of the city and is continued through it toward the western boun- dary as a high-level intercepter, receiving all the sewage that can be delivered by gravity to -the disposal works. The sewage of the low-lying portions of the city is collected by four intercepting sewers, two of which contain small pumping plants to lift the sewage enough to prevent a deep position of these sewers, which is undesirable on account of the high cost of construction in deep trenches in water-bearing soil, and the difficulty of connecting the tributary sewers satisfactorily with deep-lying intercepters. All these intercepters run to a station containing five pumps, each with a nominal rating of 27,500,000 gal. a day against a total head of 72 ft. These pumps force the sewage through two lines of 42-in. cast-iron mains 4,550 ft. long into a sewer about a mile long, discharging by gravity into the main outfall sewer. Between the arrangement of sewers in the Borough of Man- hattan, discharging both storm water and house sewage through short Unes into the nearby rivers, through many outlets, and the arrangement at Baltimore, with its separation of the storm water and the sewage, its high and low levels, pumping stations, long outfall sewer and elaborate sewage treatment works, there is an infinite variety of combinations practicable. In every case, how- ever, the topography suggests the natural drainage and the street plan exercises a more or less strong modifying influence. Special attention should be paid to the low-lying districts, for it is there that the largest sewers must be built in many cases, and the difficulties of construction are the greatest. It may be found advisable to reduce such work to a minimum by construct- ing an intercepting sewer at a somewhat higher level and thus restrict the construction in the low-lying sections to small sewers only deep enough to serve the property of those districts. Another influence of topography on sewerage plans, often overlooked, was stated as follows by Dr. Hering in his report of 1881 to the National Board of Health: "In case of sudden showers on a greatly inclined surface which changes to a level below, the sewers on the latter will become unduly charged, because a greater percentage flows off from a steeper slope in a certain time. To SEWERAGE AND SEWAGE-TREATMENT WORKS 13 avoid this uneven reception, the ahgnment should, as much as possible, be so arranged as to prevent heavy grades on the sloping surface, at the expense of hght ones on the levels. In other words, the velocity should be equalized as much as possible in the two districts. This will retain the water on the slopes and increase its discharge from the flat grounds, thus corresponding more to the conditions implied by the ordinary way of calculating the capacity of sewers. It will therefore become necessary not to select the shortest line to the low ground, but, like a railroad descending a hill, a longer distance to be governed by the gradient. This does not necessarily imply a longer length of sewers for the town, because more than one sewer for a street is not required by it." Still another decided influence of topography is shown where the configuration and surroundings of the city are such that it is advisable to employ combined sewers in all parts of the city down to the lowest contour line which will permit storm-water overflows to be used. This is the rule adopted by E. J. Fort for the new sewerage works of Brooklyn. Below this contour line, the com- bined sewers become storm-water drains, run at a higher level than the house sewers, so as to have a free outlet to tide water, and the house sewage of the low districts is pumped to points of disposal. After the most favorable location of the sewers has been deter- mined from an exclusive sewerage viewpoint, the desirability of minor changes of position to avoid needless interference with travel through busy streets should receive attention. The con- struction of a sewer in a narrow or crowded street costs the community a considerable sum in indirect damages and directly affects those having places of business on the streets. Sewage Disposal and Treatment. — Having collected the sewage by a sewer system, it becomes necessary to dispose of it in some way which will not cause objectionable conditions. In connection with such work certain terms are used frequently, and accordingly their definitions, as recommended by the Com- mittee on Sewerage and Sewage Disposal of the American Public Health Association, are given here : "Sewage disposal is a generic term applied to the art of disposing of sewage by any method. "Sewage treatment^ is the process to which sewage is subjected in order to partially remove its impurities or so to change them as to render the effluent fit for final discharge. "Contamination is the introduction into a water of bacteria or other substances which tend to render it unsuitable for domestic use. 1 The term "sewage purification" should be abandoned. 14 SEWERAGE AND SEWAGE DISPOSAL "Pollution is the introduction into a watei' of substances of such character and in such quantity that they tend to render the body of water or river objectionable in appearance, or to cause it to give off objectionable odors. "Influent is sewage, or partially treated sewage, flowing into any sewage treatment device. Effluent is partially or completely treated sewage flowing out of any sewage treatment device. "Putrescibility is the susceptibility of sewage, effluent, or wet sludge, to putrefaction under the conditions to which it is subjected. "Stability is the capabihty of sewage or effluent to resist putrefaction under the conditions to which it is subjected. "Relative stability is the ratio of available oxygen to the oxygen required to prevent putrefaction, expressed in per cent. "Suspended solids are those which are removed from sewage or effluent by standard laboratory methods of filtration. "Settling solids are those suspended matters which will subside in quiescent sewage in 2 hr. "Clarification is a relative term denoting the partial removal of suspended and colloidal matter by straining, sedimentation, or by coagulation and sedimentation. "Colloidal matter (conveniently defined) is the suspended matter which is so finely divided that it will not subside in 2 hr., yet will not pass through a parchment membrane in the ordinary process of dialysis. "Screens. — A screen is a device containing openings of proper size to retain a part of the suspended matter of sewage. "A coarse screen is one having openings in excess of 1 in. in least dimension. "A fine screen is one having openings of J4 in., or less, in least dimension. "A medium screen is one having openings intermediate between a coarse screen and a fine screen. "A bar screen is one composed of parallel bars or rods. "A mesh screen is one composed of a fabric,usually wire. "A grating consists of two sets of parallel bars in the same plane, the sets intersecting at right angles. "A band screen is one consisting of an endless band or belt of wire mesh or other screening material which passes over upper and lower rollers. "A wing screen is one having radial or curved vanes, usually composed of uniformly spaced bars, which rotate on a horizontal axis. "A drum screen is one in the form of a cyUnder or cone, consisting of perforated plates or a wire mesh which rotates on an approximately hori- zontal axis. "A disc screen consists of a rotating circular perforated disc, with or without a concentric truncated cone of similar material. "A cage screen consists of a cage, usually with sides of parallel bars or rods, so arranged that it may be lowered into the sewage and raised there- from for cleaning. "Tank treatment is the detention of sewage or sewage sludge in tanks, either quiescent or with continuous fiow. "A grit chamber is a chamber or enlarged channel in which the cross- SEWERAGE AND SEWAGE-TREATMENT WORKS 15 section is so designed that the velocity is such that only heavy solids, such as grit and sand, are deposited, while the lighter organic solids are carried forward in suspension. "A sedimentation tank is a tank for the partial removal of suspended matter either by quiescent settlement or by continuous flow at such velocity and time of retention as to allow deposition of suspended matter. "A Dortmund tank is a vertical sedimentation tank, usually cylindrical above and with a conical or hopper shaped bottom, into which the sewage or partially treated sewage is introduced near the center and after rising through the tank passes out at the surface — ^the sludge being drawn off, without emptying the tank, before it becomes septic. "A hydrolytic tank is a sedimentation tank in which by biological pro" cesses a portion of the suspended matter is converted into hquid and gaseous form. "A septic tank is a horizontal, continuous-flow, one-story hydrolytic tank in which the suspended matter is retained until anaerobic decompo- sition is to a considerable extent effected. "A Travis tank is a two-story hydrolytic tank consisting of an upper or sedimentation chamber, with steeply sloping bottom, terminating in one or more slots through which the solids may slide as deposited into the lower or sludge digestion chamber, through which a predetermined portion of the sewage is allowed to pass for the purpose of seeding and maintaining bacterial life in the sludge and carrying away decomposition products, thus inducing digestion of the sludge attended by its reduction in volume. "An Imhoff tank is a two-story hydrolytic tank, consisting of an upper or sedimentation chamber, with steeply sloping bottom, terminating in one or more slots through which the soUds may sUde as deposited into the lower or sludge digestion chamber — these slots being trapped so as to prevent the rise of gas and solids from the lower chamber — the lower chamber being provided with vents for the escape of the gases, the tank being so constructed as to facUitate the passage of the sewage quickly through the upper chamber and prevent the flow of sewage through the digestion chamber, and intended to be so operated that the sludge may be thoroughly decomposed, rendered practically free from offensive odor and so filled with gas that it can be readily drawn off and dried. "Activated sludge process is the agitation of a mixture of sewage with about 15 per cent or more of its volume of biologically active liquid sludge in the presence of ample atmospheric oxygen, for a sufficient period of time at least to coagulate a large proportion of the colloidal substances, followed by sedimentation adequate for the subsidence of the sludge flocculi; the activated sludge having been previously produced by aeration of successive portions of sewage and maintained in its active condition by adequate aeration by itself or in contact with sewage. "Chemical precipitation is the addition to, and thorough mixing with, the sewage of such chemicals as will, by reaction with each other or with the ingredients of the sewage, produce a flocculent precipitate; and subsequent sedimentation. "Sludge is the suspended solids of the sewage deposited in tanks or intercepted at the surface of filters, mixed with more or less water. 16 SEWERAGE AND SEWAGE DISPOSAL "Sludge digestion is the biological process by which organic matter in sludge is gasified, hquified, mineralized, or converted into stable organic matter. "Separate sludge digestion is the digestion of sludge in tanks entirely independent from the tanks in which it is produced. "A sludge drying bed is a natural Or artificial layer of porous material upon which sludge is dried by drainage and evaporation. "Sludge concentration is the process of reducing the volume of sludge, and increasing its proportion of solids by allowing it to stand in a suitable tank until the solids settle down, and drawing off the relatively clean water at the top. "Sludge drying is the process of drying sludge by drainage and evaporation. "Sludge dewatering is the process of removing a portion of the water contained in sludge by draining, pressing, centrifuging, or by other natural or mechanical processes. "Sludge pressing is the process of dewatering by subjection to pressure, the solids being retained by a cloth fabric which permits the water to pass through it. "Sludge cake is the mass of dewatered sludge resulting from sludge pressing. "Commercially dry sludge is sludge containing not more than 10 per cent of water. "Scum is a mass of sewage solids, buoyed up in part by entrained gas or grease, and which, consequently, floats at the surface of the water. "Screenings constitute the material removed from sewage by screens. "Grit is the heavy mineral matter deposited from sewage. "Sleek is the oily film of microscopic thickness present on the surface of waters about and often extending a considerable distance from sewer outlets. "Sewage oxidation is the process whereby through the agency of living organisms in the presence of air, the organic matter is converted to a more stable condition or into mineral matter. "Irrigation is the process of sewage treatment in which the sewage is apphed to land for the primary purpose of purifying the sewage and the secondary purposes of supplying water and fertilizer to crops. "Surface irrigation is the process in which sewage is applied to and distributed over the surface of cropped ground. "Sub-surface irrigation is the process in which sewage is distributed beneath the surface of the ground by means of open-jointed pipes. "An intermittent filter is a natural or artificial bed of sand or other fine- grained material to which sewage is intermittently applied in doses, and which, by its capillarity, holds the sewage for a time sufficiently long, in the presence of air, to effect by biological processes a high degree of purification. "A contact bed is a watertight basin filled with coarse material, such as broken stone, in contact with which the sewage is held for a time by control of the underdrains — the cycle of operation involving periods of filling, stand- ing full, emptying and resting empty; so regulated as to secure such contact with the bacterial films adhering to the surface of the coarse material, and SEWERAGE AND SEWAGE-TREATMENT WORKS 17 such aeration of the bacterial surfaces, as may be required to oxidize the sewage. "A slate bed is a watertight tank filled with slabs of slate or other similar material, laid horizontally and spaced an inch or more apart vertically, equipped so that it may be filled with sewage, allowed to stand full for a definite period of time, drained and allowed to stand empty for a time, for the purpose of oxidizing the organic matter deposited upon and adhering to the slates. "A trickling filter^ is an artificial bed of coarse material, such as crushed stone or clinkers, over which the sewage is distributed as a spray from fixed nozzles or as a film from moving distributors, through which it trickles to the underdrain system, coming in contact with the bacterial fihns adhering to the surface of the stones, and in which such aeration of the bacterial surfaces as may be required to oxidize the sewage is afforded. "A dosing apparatus is the apparatus used for regulating the application of sewage to filters or for applying the required quantity of chemicals to sewage. "A dosing tank is a tank into which raw or partially treated sewage is introduced and held until the desired quantity has been accumulated, and then discharged at such a rate as is necessary for the distribution essential to the subsequent treatment. "A sprinkler nozzle is a nozzle used for applying sewage in the form of a spray to trickling filters. "A distributor is a movable perforated pipe, channel or waterwheel which distributes sewage upon the surface of a trickling filter. There are two types of distributors — the rotary and the traveling; the rotary moves about a central axis with delivery to a circular filter; the traveling moves back and forth the length of a rectangular filter. "Disinfection is the destruction, by the agency of some chemical, of a large percentage of the bacteria in sewage or contaminated water, so as to materially reduce the danger of infection. "Sterilization is the destruction, by the agency of some chemical, of all the bacteria in sewage or contaminated water, including their spores." Why Sewage Treatment is Necessary at Times. — So far as rainfall run-off is concerned, it is only when its quantity is large that the disposal of it presents any difficulty, for there are usually opportunities for discharging unpolluted water into bodies of water or on land in the neighborhood. The main difficulty in sewage disposal is due to the substances present in sewage and in the portion of the street wash which flows away during the early part of a rain. While these deleterious organic substances are very small in amount — generally less than one- tenth of 1 per cent, the remaining 99.9 per cent of the sewage iThe committee is of the opinion that the term "trickUng filter" is preferable to the terms "sprinkUng filter" or "percolating filter" which are frequently used as synonymous terms. 2 18 SEWERAGE AND SEWAGE DISPOSAL being water — their successful handling constitutes the essential problem in sewage disposal. The part of the street wash which enters the sewers and drains after the refuse on the pavements has been carried off is rela- tively much less polluted, and on this account it can often be discharged into waters or over land unfitted to receive seriously polluted water. Where the streafns into which sewage must be discharged are small and it is very necessary to keep them clean, as is generally the case in Great Britain, the chief objection to the discharge of storm water into them is due to the suspended solids carried by the flowing water. These may form undesirable deposits on the banks and the bed of a stream and the refuse floating on the surface may give an objectionable appearance to the water. Consequently it is the general engineering rule in Great Britain that in places where the dry-weather sewage must be treated before it is discharged into a body of water, the storm water must also be treated before discharge into the same body of water, until its volume amounts to about three times the normal flow of dry-weather sewage. It is not unlikely that some American cities with separate sewers will later be faced with the necessity of installing works for removing part of the suspended matter from the storm water now discharged without treatment into neighboring rivers and lakes. The necessity for a large part of the treatment which sewage should receive before its final disposal is due to the organic matter it contains. Under normal conditions, this changes fairly rapidly into other substances, some of them offensive, and the change is accompanied by a rapid growth of bacteria, some species of which may be objectionable in the water receiving the sewage. The purpose in treating sewage is to remove the solids which cause deposits in the rivers and ponds, as well as undesirable surface conditions, and to carry out certain changes in the organic matter, under the control of the operator of the treatment works, until the final effluent from the works has its organic matter in such a state that any further changes which it may undergo can proceed in the river or lake without detriment to the quality of the water for the purposes for which it is used. There are many different methods of treatment, such as dilu- tion in a large body of water, removing the solids by screens or by sedimentation in small or large tanks, allowing the sewage or the organic solid matter settling from it to remain in tanks under SEWERAGE AND SEWAGE-TREATMENT WORKS 19 conditions which will cause various changes in it, changing the character of sewage by aeration, and filtering the sewage or the effluent from some previous form of treatment. In each of these methods of treatment a certain minimum fall through the tanks or filters and the outfall sewer is necessary and must be provided, naturally or artificially by pumping, and consequently the method of treatment and disposal of sewage affects not only the direction from the city which the outfall sewer or sewers must take but also the slopes or grades of such sewers. Influence of Method of Disposal and Treatment upon General Sewerage Plan. — There are three general methods of disposal that affect the general design of a sewerage system. The first of these disposal methods is directly into a river or other body of water on the shore of which the city lies. Prob- ably the Borough of Manhattan, New York City, offers the best example of this among large cities, with its numerous main sewers running east and west to outlets at the Hudson and East Rivers. The second method of disposal is to intercept the sewage and carry it to a point in the adjoining body of water where it will not cause trouble ; this may not be necessary at first, but in most cases it is inevitable if the city grows as rapidly as do most Amer- ican municipalities, and attention must be paid to it, particularly to the future desirability of separating the house .and industrial sewage from part of the storm water. A proposed Cleveland system, shown in Fig. 3, from Engineering News, March 28, 1912, is an example of this intercepting plan. The third method of disposal is by some treatment of the sewage which will materially change its character before it is discharged into a body of water. This makes it necessary to deliver the sewage to treatment works, suitable sites for which are difficult to procure in many cases, particularly where the country is well built up, not enough open land properly located is available in the city, and neighboring towns object to the plant being located within their limits. The separation of storm water from the house sewage often becomes financially advisable, so as to permit the former to be discharged by short, direct lines into the river, lake or bay nearby, and to keep down the cost of the long sewer to the disposal works, and the disposal costs as well. In the case of combined sewers, the same end is attained by making provision at one or more points for the discharge of the storm water in excess of a predetermined amount, through overflow 20 SEWERAGE AND SEWAGE DISPOSAL weirs or chambers into channels or other outlets leading directly to the river or lake. In cities situated on rivers it is customary to convey the sewage to a point below the thickly settled district, whether the sewage is to be discharged in its raw state or is to be treated by some artificial means before discharge. This plan avoids the danger of causing objectionable conditions along the water front and, perhaps, the contamination of the local water supply by the dis- Infercepters ' Sewer Outlets . * . • Low Level InfercepttjrArsa Shown Hatched. "-■iiOld Intake Crib \Newlntake ^""^-^ ^ FiQ. 3. — Cleveland intercepting sewer system. charge of sewage into the river. Similarly, cities situated on the Great Lakes may convey their sewage in the direction of the trend of the waters to a point well beyond the populous districts. This policy is not so well established nor so effective as in the river cities, for the lacustrine current is not so marked as most river currents and it may be altered temporarily by adverse winds. Cities situated on tidal waters often build long inter- cepting sewers to convey the sewage to a favorable location for discharge. SEWERAGE AND SEWAGE-TREATMENT WORKS 21 Each method of treatment has certain features materially affecting if not actually determining the location at which it can be practiced. Intermittent sand filtration requires large areas of suitable sand. Rapid filtration through deep beds of coarse aggregates requires more head than other methods involve. Aeration requires a power plant to supply compressed air. Some tank treatments are accompanied by offensive odors. These conditions, explained in later chapters, make it particularly desirable to study the practicability of discharging sewage at several nearby points, after suitable treatment to fit it for the conditions at each outlet, rather than the delivery of all of the sewage through a long, expensive outfall sewer to a single treat- ment plant. It should be understood at the beginning of any study of sewage disposal that the purpose of treatment is to protect the water or land receiving the sewage. Changing the character of the sewage merely for the sake of making it less offensive or dangerous is a waste of money unless it is necessary, just as it is a waste of money to carry the sewage farther than the local conditions in each case require. It should also be understood that where the conditions warrant treatment by dilution, which means the discharge of sewage into large bodies of water, that method of treatment is not only the cheapest in first cost in most cases but is just as well established as a truly scientific process as the most elaborate artificial treatment by sedimenta- tion, aeration, filtration and sterilization, needed by a city not favored by Nature with conditions admitting the use of less expensive methods. Municipal Liability for Inadequate Sewers. — There is a quite general belief that a city is never liable for damages caused by the inadequate capacity of sewers. As a matter of fact, a city may be liable for such damages under certain conditions, and even where there is no legal liability a grave moral responsibility rests upon the engineer who designs a sewerage system. The present legal aspect of the subject places the engineer upon a strictly professional plane like that of a physician or lawyer, and it is one of the few instances in which the courts have recognized the professional nature of engineering. It has been held that if a city has a collection of sewers and it cannot prove that this collection was built or rebuilt in accordance with a comprehensive plan prepared by an engineer presumably competent to design such works, the city is liable for any damages 22 SEWERAGE AND SEWAGE DISPOSAL which may result from the inadequate capacity of the sewers. ^ This makes it necessary for the city to employ competent engineers to design its sewers and to approve the sewers in any additions incorporated within the municipal boundaries from time to time. In some cases the courts have held that if a city employs a presumably competent engineer to design its sewerage system and the construction proceeds in conformity with his plan and subsequent modifications of it by presumably competent engineers, it is not liable for damages due to a fault in the plan. In Massachusetts, it is not clear that the sewer system must be designed by an engineer, the law being stated^ as follows, "Cer- tain governing principles of law are well established. A munici- pality is not responsible for any defect or want of efficiency in a general plan for drainage adopted by a board of public officers, even though private individuals may be exposed to great inconvenience and loss." The city is liable, however, for damages due to the known lack of repair of sewers or its failure to maintain them to the standard of efficiency of the original construction. The damages caused by inadequate sewers are generally due to an insufficient estimate of the quantity of sewage for which provision must be made, particularly where the combined system is used. Too large capacity not only causes needless expense but develops a tendency for the sewage to flow so slowly and in such shallow streams that part of the suspended solids will settle on the inverts, forming offensive deposits. If the sewers are too small, cellars may be flooded with sewage, which may even rise from manholes and gutter inlets and flow over the surface of the ground, resulting in damage and litigation. The subject is, in some respects, the most important in the whole field of sewerage engineering, one calling for sound judgment as well as knowledge. ' A typical ruling of this sort was made in the case of Hart vs.. City of Neillsville, 104 N. W. Rep. 699, where the Wisconsin Supreme Court made the decision. 2 Diamond vs. North Attleborough, 219 Mass. 590; see, also, 4 N.E. Rep. 321 (New York Court of Appeals); 61 Atl. Rep. 180 (Maine Supreme Court). CHAPTER II QUANTITY OF SEWAGE In this chapter the term "sewage" will be restricted to the liquids which are ordinarily discharged by separate sewers, comprising (1) domestic and industrial wastes derived primarily from the public water supply, (2) industrial wastes derived pri- marily from privately owned water supplies, and (3) infiltration. The main sewers for removing this sewage must be designed to meet future rather than present conditions. The ideal plan is one which will provide satisfactory sewerage facilities during such a period of years that the average annual cost of this service shall be a minimum. Smaller sewers than the economic size for the first part of this period and additional sewers for the last part might give equally satisfactory service, but usually at a higher cost. Larger sewers than the economic limit would not only cost more, but, on account of the small proportion of their capacity utilized during the early years and the consequent low velocities of flow and resulting probability of forming deposits on the flat grades, they might prove less satisfactory during the early part of their service. In determining the period for which sewerage facilities must be provided, assumptions must be made regarding a number of future conditions, such as changes in the rate of interest which a city must pay for money borrowed to build the sewers, increase in the population, incorporation of additional territory within the city, development of industries, and changes in the habits of the residents in respect to the use of water. The fluctuations in rates of interest fall within the field of economics, but the other influences lie within the field of the engineer who must rely mainly on his own knowledge and judgment in reaching decisions, after careful study and analysis of them. The as- sumptions made may have a serious future result, as when the volume of sewage is so underestimated that either the main trunk sewer shall prove inadequate or the treatment plant to which it runs, be outgrown in a few years and be impossible of 23 24 SEWERAGE AND SEWAGE DISPOSAL adequate enlargement, on account of lack of available land. In such case, the expense of intercepting part or all of the sewage and taking it to some other place of disposal may add greatly not only to the total cost of the works necessary for satisfactory service, but also to the annual cost of their operation and maintenance. Sound judgment in such matters can only come from observation and study of the actual growth and development of individual cities and the reasons therefor, somewhat along the lines con- ducted by the traffic departments of some railways and the economists of some large banks. Population. — The increase in the population of a city is some- times affected by exceptional conditions, which may be tem- porary, as was the case in Washington during 1917 and 1918, or permanent, as where a new industry develops phenomenal growth, like the rubber industry in Akron or the automobile industry in Detroit. Sometimes, as at Gary, Ind., practically a new city develops in a few years about a great industry. But in the average case, the past records of growth of a city and the records of the growth of similar places furnish fairly reliable guides for estimates of future development for periods of 25 to 40 years. The various available records are used in four ways, according to the assumptions made regarding the rate with which the population increases. These four assumptions are: 1. Assumed Uniform Percentage Rate of Growth. — When cities are young and thriving, it often happens that their rate of growth is high for some years, but an assumption that this uni- form percentage rate will persist for a long period will certainly lead to an over-estimate of the population in most cases. This is clearly shown in Fig. 4, where the line A indicates the increase in population which would have taken place had the growth of the cities been 31 per cent per 10 years, while the lines marked "Average of 6 Cities" and "Average of 4 Cities" show what it actually was. The assumption of a uniform percentage rate of increase is apparently most reliable in the case of old, large cities not subject to periods of great commercial or industrial activity. 2. Assumed Curvilinear Rate of Growth. — The information furnished by diagrams of the past growth of cities is very in- structive, but an attempt to predict the future growth of a city from its past development alone, by extending the curve of that QUANTITY OF SEWAGE 25 development, is likely to give misleading results, as will be shown later. Diagrams have a useful place in the study of changes in population, but they are not a substitute for an investigation of the various influences which have affected the city's growth in the past and may affect it in the future. 3. Assumed Arithmetical Rate of Increase. — Occasionally it is assumed that a city will gain approximately the same number of inhabitants annually, leading to a straight-line increase of population like that shown by line B, Fig. 4. This assumption may be warranted only in the case of a few large cities with good I30O 1910 I9Z0 1930 The&e Dates Apply to Milwaukee only. Fig. 4. — Growth of large American cities. transportation facilities into suburban towns, which result in many persons doing business in the large city being enumerated as suburban residents. As their business life is spent in the large city, they must be classed for purposes of water supply and sewerage as inhabitants of both places, a fact often overlooked in engineering discussions of population. Commercial districts often require large sewerage capacity, yet the census taker will report a very small population in them. Small cities without industries, which are the commercial centers of a prosperous agricultural region, occasionally show a tendency toward this 26 SEWERAGE AND SEWAGE DISPOSAL rate of growth, but if they become producers as well as traders, through the development of manufacturing, they are likely to grow for a time more rapidly than the assumption of arithmetical progression wiU indicate. 4. Assumed Decreased Rate of Growth. — As a general rule it is found that the larger a city becomes, the smaller will be its percentage rate of growth from year to year. Fig. 4 shows this clearly. The growth of Chicago has been abnormal and that of New York (Manhattan) and Brooklyn has been influenced by unusual conditions, so that these cities are not reliable guides j 1 1 1 i/ / / >' I ) W\ ^ #. ^f <—Z6% -* Increase in 1 I'fn m y^ '/¥" \^-Z2%-f> } <-.^z ;%-> i^ „■*'■ Increase m/O/ea "S y Assumed Average -^ .^ <-6 j»^> 'Increase m 10 years _ 1850 '66 '60 '65 '70 1875 '80 '86 '90 '96 1900 '05 '10 '15 'ZO ijjj '30 '35 1940 Years. Fig. 5. — Growth of population of New Bedford. 280 260 240 220 200 . % 180 5 160 o 140 £ 120 1 o 100 = D. O 80 "^ 60 40 20 in general investigations of changes in population. There is not only decrease in the percentage rate of growth of cities as they grow in size, but there is also a general decrease in the percentage rate of growth of the entire population of the country. As a rule, this assumption is the most reliable of the mathematical methods of estimating future populations, particularly if it is checked by basing the prediction on the experience of comparable cities which have already passed the present population of the city under consideration. This is done, as shown in Fig. 4, by arrang- ing the lines indicating the change in population • of different QUANTITY OP SEWAGE 27 cities so that when they reach the present population of the city under consideration, they all pass through the same point. This method will tend to give slightly too high a percentage rate of in- crease, since it does not allow for the average progressive decrease throughout the country as the years go by. It is interesting to note in connection with Fig. 5, showing the growth of population in New Bedford, that this diagram, pre- pared in 1911 to accompany a report by the authors, shows a forecast for the population of 1920 of about 128,000. The 1920 U. S. Census gives a population of 121,217. Thus the forecast for a nine-year period was within about 5 per cent of the actual conditions realized. Increase in Area. — The tendency of many cities to an increase of area by the absorption of suburbs is shown by the growth of Cincinnati since it was incorporated as a city in 1819: Year 1819 1850 1870 1890 1900 1910 1913 Area, square miles 3.00 6.16 19.05 23.73 35.27 50.26 69.85 Such enlargements of area may cause large and sudden increases in population, which if not anticipated, may result in overtaxing intercepters within the period for which they were intended to be adequate. Increases in area may also require long extensions of main sewers and may result in greatly increased quantities of ground water made tributary to the intercepters. Where the community is served by combined sewers, there is also the probability that for considerable periods in the future, or until the population becomes quite dense, brooks will be turned into the main sewers, thus adding materially to the nominal dry-weather flow of sewage. It is also of vital importance to consider where the estimated increase in population will occur in order that the lower sections of the intercepter may be placed at elevations from which it will be possible to make extensions into new territory that may become populated within the period for which the inter- cepter is designed. Density of Population. — An idea of the density of population, in persons per acre, in American cities is given by the following figures showing a range in 1910 of from 85 in Hoboken, N. J., downward : 28 SEWERAGE AND SEWAGE DISPOSAL Persons FEB Acre Places 85 Hoboken 29 Baltimore 27 Boston 26 New York 25 Milwaukee 23 Newark 2D Pittsburgh 19 Cleveland, Philadelphia, Chicago, Harrisburg, Providence 18 Detroit 17 St. Louis, Buffalo, Louisville, Rochester 16 Savannah 14 San Francisco, Columbus, Albany. The density within a city varies greatly, and it is difficult to estimate its probable future changes. A residential section of the present decade may become a commercial or manufacturing district in the next decade, or the detached homes of persons of means may be replaced by crowded tenements or commercial developments. As estimates of the future population in the different parts of a city are part of the basic data upon which sewerage works are designed, and they must be determined by the exercise of sound judgment, the importance of studying present tendencies toward growth or stagnation, and their causes, is evident; The difference in policy of our cities, with respect to the ex- tension of their boundaries, has been very marked. In our Eastern communities the city limits have usually been rather more restricted than in our western cities. This tendency has led to rather dense population in the Eastern cities and sparse population in Western cities of like size. Thus, for example, we find from the 1915 U. S. Census Burean, Bulletin 133, "Estimates of Population," with respect to cities having a population of 250,000, more or less, the facts given in Table 1. Such conditions must be noted. The scattered population involves greatly increased cost of the public utilities or services of various kinds and far greater difficulty of accurate forecast of the probable future direction and degree of development and of increase in population. Proportion of Municipal Water Supply Reaching Sewers. — As the wastes which form sewage are mainly water from the water works supplying the city, the proportion of the water supply which will reach the sewers must be estimated. A considerable QUANTITY OF SEWAGE 29 Table 1. — Population and Area of Certain American Cities in 1915 City Popula- tion Land area in acres Acres per 1,000 of popu lation Density of popu- lation per acre Oakland, Cal Denver, Col St. Paul, Minn.. . Kansas City, Mo Portland, Ore. . . . Seattle, Wasli. . . , Toledo, Ohio .... Indianapolis, Ind Columbus, Ohio. Louisville, Ky. . . . Rochester, N. V. Providence, R. I. Jersey City, N. J Average 190,803 253,161 241,999 289,879 272,833 330,834 187,840 265,578 209,722 237,012 250,747 250,025 300,133 31,591.0 37,028.0 33,388.0 37,555.8 32,748.8 37,481.0 16,025.6 22,165.1 14,149.6 15,368.4 14,876.3 11,353.0 8,320 . 165.5 146.3 137.9 129.6 120.1 113.3 85.3 83.4 67.5 64.8 59.4 45.4 27.7 6.0 6.8 7.3 7.7 8.3 11.7 12.0 14.8 15.4 16.9 22.0 36.1 252,351 24,003.9 95.9 13.3 part of the water supply used by railroads, by manufacturing establishments and power plants, in street and lawn sprinkling, in extinguishing fires, and by consumers not connected with sewers, does not reach the sewers. There is also some leakage from the water mains and service pipes which does not reach the sewers. In 19 11 the Milwaukee Sewage Disposal Commission estimated that of the total daily supply of 105 gal. per capita 40 gal., or 38 per cent of the supply, never reached the sewers. This 40 gal. included: Gallons Steam railroads 5 Industrial uses 5 Street sprinkling 5 Lawn sprinkling 2 J^ Consumers not connected with sewers 7^2 Leakage from water mains and services 15 In a few cities where the infiltration of ground water and the discharge of roof water into sewers is low, the quantity of sewage is actually less than the quantity of water supplied, as at Brock- 30 SEWERAGE AND SEWAGE DISPOSAL ton, Mass., where the sewage has ranged from 56 to 73 per cent of the water supply. Often, however, infiltration, roof water, and water used in industries and obtained from privately owned sources, make the quantity of sewage larger than that of the public water supply. This is shown by the ratio of sewage to water supply in the following places: Locality Water to Sewage Massachusetts North Metropolitan Sewerage District 1 to from 1 . 16 to 1 . 29 Worcester, Mass 1 to from 1 . 09 to 1 , 73 Quincy, Mass 1 to from 1 . 05 to 1 . 43 Providence, R. 1 1 to from 1 . 14 to 1 . 56 With well-built sewers and with roof water excluded fron; the sewage, the variation from year to year in the ratio of sewage to water supply in a city is not great, unless there is a substantial change in the industrial uses of water. For this reason, the consumption of water in a city affords basic data for sewerage as well as water-supply engineering. Water Consumption in Cities. — The consumption of water in cities is usually expressed in gallons per capita daily. A com- parison of such records from different cities is likely to be very misleading because in some cities large quantities of water used industrially are obtained from privately owned supplies, as at Fall River, and in other cities, as Philadelphia and Buffalo, the industries use the municipal supply mainly. Furthermore, the care taken to reduce the waste of water through leaks in mains, services and plumbing has a decided effect on the per capita consumption. Waste is prevented mainly by metering the consumers' supplies, which tends to reduce them as shown in Fig. 6, and by inspection of the condition of the piping and plumbing. The waste .and unaccounted-for water may range in practice from 20 to 40 per cent, more or less, of the total water entering the supply-pipe system. In any particular case, there- fore, it is important to ascertain the present practice in preventing waste of water, the probability that more attention will be paid to such work, and the probable development of heavy industrial uses. In such studies valuable aid can be obtained from a re- port^ on water consumption by Metcalf, Gifford and Sullivan. From the statistics available two conclusions seem warranted: First, there is a tendency toward a gradual increase in the quan- 1 Jour. N. E. Water Works Assoc, March, 1913. QUANTITY OF SEWAGE 31 tity of water used per capita of population. This is undoubtedly due, so far as it relates to domestic uses, to more elaborate plumb- ing. The number of fixtures per person, as well as the quantity of water required per fixture, has greatly increased in recent years. In the larger cities, the increased consumption may be due in part to the difficulties surrounding the management of the water departments, which are usually much greater than in the smaller cities and towns. Second, the evidence furnished by such cities as Providence, Worcester, Fall River and Lawrence, indicates that with careful management, aided, perhaps, by thorough metering, it is possible to hold down the increase in quantity of water consumed to reasonable proportions. 200 190 O180 Q ?■ S-160 5 fe 70 a. g 60 o 1 50 40, ~^ —\ n — 1 n n n n n -1 n Average of 14 CiHes , Having less than IS% _ Services Metered. . 1 / J ^ / _ / X ■ ' ' ■ Average of 16 Cities - Having more fhan 50% - \ L «- ~ r s "^ ^ 200 190 180 170 160 70 60 50 40 Fig. 6. — Composite curves of water consumption in cities with small and large percentages of services metered. Rate of Water Consumption in Different Parts of a City. — The rate of water consumption per capita varies greatly in differ- ent parts of a city. Investigations in 19 cities made in 1910 and 1911 showed^ that the range was as follows: Water Consumption in gal. per Day per Capita Class of Buildings Range Average Apartment houses 24 to 135 62 First-class dwellings 15 to 75 64 Middle-class dwellings 11 to 66 34 Lowest-class dwellings 4 to 37 15 The industrial uses vary greatly, according to the nature of the manufacturing. Fuertes reported^ in 1906 that the range '■Jour. N. E. Water Works Assoc, March, 1913. 2 Waste of Water in New York. 32 SEWERAGE AND SEWAGE DISPOSAL in 11 cities was from 0.4 gal. per capita daily in Wellesley, Mass., to 81 gal. in Harrisburg, Pa. It was 30 gal. in Boston, 40 gal. in Cleveland and 45 gal. in Milwaukee. These figures are based on the entire population of the city, but as manufacturing is carried on only in certain districts, as a rule, the per capita con- sumption for manufacturing in those districts is much higher than the average for the city. In practical designing work it is very desirable to make an actual inspection of the industries and a careful estimate of the quantities of water they use from all sources. The same is true of the consumption in business districts. Fluctuations in Water Consumption. — While it is important to know the average quantity of water consumption, it is of still greater value to have data relating to the fluctuations, as a sewer must be designed to take the sewage when flowing at its maximum rate. The maximum rate ■ of water consumption usually occurs during summer months when water is in demand for street and lawn sprinkling and the excess is not likely to reach the sewers, or in the winter when large quantities are allowed to run to prevent freezing of pipes and fixtures, this excess usually finding its way into the sewers. Records of maxi- mum water consumption for 67 Massachusetts cities and towns (1910) have been compiled by a Committee on Water Consump- tion Statistics and Records.^ The average figures are as follows: Average water consumption, gallons per capita per da}' 63 Daily average in maximum month, percentage of average for year .... 128 Daily average in maximum week, percentage of average for year 147 Maximum consumption in one day, percentage of average for year. . . . 198 There were, however, instances in which the maximum rates greatly exceeded these averages. For example, in Manchester and Mansfield, Mass., the maximum daily consumption was 302 and 368 per cent of the average for the year, respectively. These high rates of consumption, however, almost always occur at times when the usual proportion of the flow does not reach the sewers, as in the driest portion of the summer, or in winter when water from other sources, as for example, ground water, is likely to be at a minimum. In addition to the fluctuations in flow already discussed, there is an important variation from hour to hour each day, as illus- 1 Jour. N. E. Water Works Assoc, March, 1913. QUANTITY OF SEWAOE 33 trated by Fig. 7 and 8 taken from the same report. It will be seen from Fig. 8 that the maximum peak flow during the week 130 \Z5 120 115 ^110 ■6 ■ Q 105 •t 100 c 95 S. 90 1 85 O 80 O 75 70 65 60 " " / ^^\ / t ":::::] t\ ._/ \ t ^-— ~t~ Aience^ ~ i 7 :::: ___t- "" / .s__ ^~ ___A_ ""T'~V~ ' \ "s--- 130 125 120 115 110 105 100 95 90 85 80 75 70 65 60 1 2 3 4 5 6 7 8 9 10 1112 1 2 3 4 5 6 7 8 9 1011 12 \i—. -A. M ->j<- P. M. >[ Hours of Day. Fio. 7. — Hourly water oonsun;Et;on for average day in Holyoke in November, 1905. Estimated population supplied, 51,000. l<--Monday--4«--Tue5day--»fi-Wedne5day»j<--Thur5doy-->j<--'^i'i<''oy--->f--5'»Turday--s«--Sunday-i 6 12 6 12 6 12 6 IZ 6 12 6 12 6 12 6 IZ 6 12 6 12 6 12 6 IZ 6 12 6 12 M. M. M. K M. M. M. FiQ. 8. — Fluctuations in water consumption in Holyoke during week ending November 17, 1905. occurred on Monday, when the draft was about 135 per cent of the average for the day, and the minimum peak draft was on 34 SEWERAGE AND SEWAGE DISPOSAL Sunday when it was 146 per cent of the average for the day, these rates being 139 per cent, and 132 per cent, respectively, of the average rate of draft for the week. The hourly fluctuation in rate of water consumption has a decided effect upon the rate of sewage flow, as discussed later in this chapter. It is not, however, entirely responsible for the fluctuation in the rate of flow of sewage, for in some places large quantities of ground water are pumped by industrial establishments and discharged into the sewers during the working hours of the day, thus tending to increase the peak flow beyond the amount resulting from the normal fluctuation in the draft on the municipal water supply. In the absence of more authoritative information an addition of 50 per cent over the average rate of water demand during any 24 hr., may be made as a fair allowance for the excess of the maximum demand during the 24 hr., from all sources, over the average daily demand, at the time when the flow of sewage is at its maximum. This rate, however, will vary in different places. If this peak consumption is applied to the maximum draft for a single day, of 198 per cent of the average annual consumption, and it is assumed that the portion of the annual consumption which finds its way into the sewers averages 50 gal. per day, we have a maximum rate of contribution from the public and private water supplies of about 150 gal. per capita daily (50 X 1-98 X 1.50 = 148.5). This will serve to illustrate the theory of the yield of sewage based on water consumption, but should not be applied in design unless local conditions are found to warrant it. Ground Water. — A portion of the rainfall ordinarily runs quickly into the storm-water drains or other drainage channels. Another portion is evaporated or is absorbed by vegetation and the remainder percolates into the ground becoming ground water. The proportion which thus percolates into the ground depends upon the character of the surface and soil formation, upon the rate and distribution of the precipitation according to seasons, and the time of its occurrence. Any reduction in permeability, as, for instance, that due to buildings, pavements, or frost, decreases the amount of ground water and increases the surface run-off correspondingly. The amount of ground water flowing from a given area may vary from a negligible amount for a highly impervious district or a district with a dense subsoil, to 25 or 30 per cent of the rainfall for a semi-pervious district with a QUANTITY OF SEWAGE. 35 sandy subsoil permitting rapid passage of water into it. The percolation of water through the ground from rivers or other bodies of water sometimes has considerable effect on the ground- water table, (the ground-water table is the surface of the water in the soil) which rises and falls continually. The presence of large quantities of ground water results in leakage into the sewers and an increase in the quantity of sewage and the expense of disposing of it. This infiltration from water- bearing earth may be as low as 5,000 gal. per day per mile of sewers and may be as high as 40,000 gal. or more. During heavy rains, when there is leakage through manhole covers as well as infiltration into the sewers themselves, the quantity may exceed 100,000 gal. per day per mile of sewer. It is a very vari- able part of the sewage, depending upon the quality of the materials and workmanship in the sewers, the character of the maintenance, the height of the ground-water table and the extent the sewers have been injured when house sewers were connected to the common sewers. The early sewers first built in a district usually follow the water courses in the bottoms of valleys, close to and occasionally in the beds of brooks. The water of the brooks is taken into these sewers. The discharge of the brooks grows gradually smaller after the immediate run-off from a storm has passed by, and the flow until the next storm occurs is mainly ground water draining out of the earth. As a result, these old sewers receive comparatively large quantities of ground water, while sewers built later at higher elevations in the districts will receive rela- tively smaller quantities of ground water. With an increase in the percentage of area in a district which is paved or built over, comes an increase in the percentage of storm water which is conducted rapidly to the drains, sewers and water courses, and a decrease in the percentage of the storm water which will perco- late into the earth and tend to leak into the sewers. A sharp distinction is to be made between maximum rates of leakage or infiltration into sewer pipe systems and average rates. The former are necessary in determining required sewer capacities: the latter, in the discussion of operating problems such as the annual cost of pumping sewage. The records of measured leak- age of ground water into sewers given in Tables 2, 3 and 4 are significant. 36 SEWERAGE AND SEWAGE DISPOSAL m m o Eh M o <; M i l> Qi 00 5 Oi 00 S S . >>^ i> . u apn' » » » » ^ » o o o o o o qa (0 q:l q^ « ca mill tttitt mill a 4 |>H Ph IH JH >. ■e^ ■1 s T ■£ -S a a, 1 1 a a a a i i 1 1 s m t^ n S3 o < K O iz to o CO o> 8§§g§§ o o o o o o U3 CO" rH tH od" N ui ti ui tn th '•n +1 o cq § o» o CO 2 s § o § so s +1 2 CO cq +1 1 M fa 7 +[ CO o o CO ■§5 2 M a °-l i S cq (M o o o o « o O ■* « Oi (N to M cq M (N « ec lO U5 CM CI CO CO '3 CO B CO 5D O d "^ CO o OJ 00 5 0) -3 : ; ; : j ; 3 : > ■ o u ] u O :0 ids- 1 ll ?! --0 as 1 i 1 1 38 SEWERAGE AND SEWAGE DISPOSAL Table 3. — Estimate of Infiltration into Sewbks at Worcester, Mass., 1890-1897 (Million gallons per day) Month 1890 1891 1892 1893 1894 1895 1896 January. . . February. . March April May June July August .... September. October . . . . November. December. Averages .... 1 . 466 1.254 1.003 0.964 1.325 1.334 0.906 1.089 1.508 1.014 1.275 2.701 1.628 1.350 1.080 1.038 1.427 1.435 1.975 1 172 1.466 1.610 1.083 1.361 2.885 1.738 1.442 153 109 524 533 1.042 1.252 1.478 719 156 1.453 3.080 1.856 1.539 1.231 1.183 1.628 1.637 1.112 1.338 1.578 1897 1.834 1.234 1.550 3.286 1.980 1.643 1.314 1.263 1.736 1.747 1.187 1.427 1.685 1.966 1.323 1.663 3.524 2.122 1.761 1.409 1.354 1.861 1.873 1.272 1.530 1.805 2.056 1.383 1.738 3.684 2.220 1.840 1.473 1.416 1.946 1.958 1.330 1.600 1. 2.139 1.439 1.808 3.832 2.309 1.915 1.532 1.472 2.024 2.037 1.384 1.663 1.963 Total mileage Average gallon per mile per day .... Population on line of sewers Gallon per capita per day .63.90 22,960 75,644 19.4 68.22 21,650 79,015 18.7 72.78 21,690 83,272 18.9 78.03 21,580 83,588 20.3 81.60 22,120 89,195 20.2 84.87 22,250 90,754 20.8 88.96 22,080 93,737 20.9 From 1891 to 1897, inclusive, the average leakage per day was 22,047 gal. per mile or 19 . 9 gal. per capita. The quantity of leakage depends on the length of sewer, and to a certain extent on the population, which affects the number of connections to buildings and the lengths of the sewers. The authors have found that the leakage through defective joints, porous concrete and cracks is large enough in most cases to lower the ground-water table to the crown of the sewer and usually well down toward the invert, although its elevation varies greatly with the quantity of rain and snow water percolat- ing into the ground. QUANTITY OF SEWAGE 39 Table 4. — Leakage op Ground Water into Sewers Place Gal. per day per Extent of sewers mile of sewers considered Alliance, Ohio 195,000 41,000 86,000 264,000 45,000 61,000a 178,0005 26,000 32,500 30,000 22,000c 9,000 35,000 45,000 25,000d 48,000 50,0.00 50,000 25,000e 40,000/ 80,000 to 100,000 32,000 to 60,000 24,000 100,000 5,000 1,072,000 32,000 Altoona, Pa 1 . 2 miles . 6 miles 0.95 miles 2,000 ft. 10,400 ft. 10,400 ft. 11 miles Altoona, Pa. ... Altoona, Pa Brockton, Mass Brockton, Mass Brockton, Mass Canton, Ohio .... Clinton, Masa Concord, Mass East Orange, N.J whole system 29 miles 25 miles whole system whole system 150 miles East Orange, N. J. Gardner, Mass. . . . whole system whole system whole system 137 miles 8 . 58 miles Marlboro, Mass Medfield, Mass Metropolitan System New Orleans La North. Brookfield, Mass .... 1.41 miles 3,010 ft. Worcester, Mass a. Water in river low. b. Water in river high. c. Great precautions taken to prevent leakage, as construction was carried on in quicksand and the ground-water table was naturally 10 ft. or more above the sewer, d. This relates to the sewer serving parts of Newark and Elizabeth, N. J., and smaller places westward to Summit, e. Before house connections were made. /. Before any connections were made. The Maiden figures are from Eng. News, Aug. 27, 1903; the Concord figures from the 1900 report of the Sewer Commissioners, and the remainder from reports of the Mass. Board of Health and Trans. Am. Soc. C. E., vol. ixxvi, p. 1909, et seg. Sewage per Day. — In Table 5 are given average statistics of the sewage of a number of Massachusetts towns and small cities having separate sewers. The gagings of the sewage from 40 SEWERAGE AND SEWAGE DISPOSAL S 6 O m o 00 o 1^ a ■< a f. n g is Ck o c ^ a , Xi a a on Oh V 1 * FLh ^ c •43 Jil 0) 04 , C3 Ph o o o o o o Tj( W OS G o CQ •-( <-■ !£> CD O to lO o o o o o o o o CO ■* r- CO cq IQ : o» 0> 00 l> t> CO GG G r* CO i-H lo oj N >C W (N l> »C Oi oooooooo O O O O O O O < 1 O'nooM'^t-— '* cDC0i0tN.(£i0»c00it-l>OWI>'OC0t>-(NC0C0»0 00 oi" to" (m' to" i-T t>-" ca CD m" t> »o" co" c? O) o (m" rH ira" o i. ^ . -a:3§adoiS^ft-o'S»-^-g-eE51?£ ,13 »*' 1-3 2 o A 4 Q ^ s^ 01 (7) ^ , O 03 03 ^ s ^ ll s ° P5 a O tH bO ^ CQ 09 0) -tJ " a I -I o 3 S ■o 2 ^ Oi » ^ o •a a -o s .2 a o a) +* ^ 8 ~ .2 >. S fill- S w iJ -< CO QUANTITY OF SEWAGE 41 several districts in Chicago, Table 6, exceed greatly the Massa- chusetts records. The Chicago sewers are built on the combined system, but the gagings were made in dry weather. The large flow is due mainly to the heavy consumption of water in the city, which averaged 242 gal. per capita daily in 1910. Table 6. — Typical Drainage Areas and Dry Weather Run-opps, Chicago, 1910 and 1911 (Prom Wlaner's Beport on Sewage Disposal. San. Dist. of Chicago, 1911) Sewer outfalls 1 n .2 "^ OS Dry weather run- jffs. Period covered by observation 6 I o 6 a p. 6 3 P, "1 •3 O Diversey Boule- 890 23,550 8.65 0.0097 0.22 238 26.4 Aug. 15-17, 1911, vard (W). 2 days. Eandolph St. (W) . 240 11,368 6.10 0.0254 16.25 348 47.4 Aug., 3 days. Robey St. (S) 2,500 38,728 10.1 0.0040 2.58 169 15.5 June 1-3, 1911, 2 days. Ashland Ave., (S).. 980 44,581 23. 2^ 0.0237 15.1 338 45.5 May 18-20, 1911, 2 days. Center Ave., (S)... 660 23,463 20.92 0.0317 20.3 578 35.6 May 16-18, 1911, 2 days. Thirty-ninth St., 14,340 285,900 140.0' 0.0098 6.25 .'(18 20.0 pumping station. 100.0* 0.0070 4.46 227 f 209 days. Ninety-second St... 98 3,666 1.84 0.0188 12.0 325 37.4 Aug. 1, 1910. Mar. 31, 1910. ■ Aug. 1, 1910. Wentworth Ave., 5,300 30,464 12.4 0.0023 l.SS 284 5.8 ] July 31, 1911. (S), (Calumet). 253 days. 1 Daily variation average 8 A.M. to 8 P.M. 28.5 c.f.p.s. contains large amount of industrial waste. 8 P.M. to 8 A.M. 18.6 c.f.p.s. * Daily variation average 8 A.M. to 8 P.M. 25.7 c.f.p.s. contains large amount of industrial waste. 8 P.M. to 8 A.M. 17.0 c.f.p.s. • This run-off or more for 76 days in 1909. « This run-off or more for 276 days in 1909. 8 2.4 c.f.p.s. per square mile occurred 329 days in the year. The flow on different days of the week varies considerably, as might be expected from the daily variations in the consumption of water. It is larger on Monday and smaller on Sunday than on other days. There is also a difference between the average daily volume of sewage at different seasons of the year, due in part to variations in water consumption and in part to ground- water conditions. The average rates of flow upon which estimates of cost of 42 SEWERAGE AND SEWAGE DISPOSAL pumping and treatment may be based are much below the maximum rates, and, from the data available, appear to range in a general way between 100 and 125 gal. per capita per day for the larger cities. For small towns average rates appear to range from about 25 to 60 gal. per capita per day. Sewage per Hour. — The flow of sew.age per hour is more important to the designing engineer than the flow per day, 180 170 160 ISO 140 130 iieo 110 o ■■'6 a: §100 X 90 -1 - ( FullLine isAveraai r r ... ino 5, - / \\ Points shown ffius: o / \ . Dotted Line is Average of Milwaukee . Cagmgs, District No.47. - 1 \ ! \ taints shown thus: ® 1 1 I 1 t 1 1 1 1 1 1 1 \ ---< 1 s f — A s \ \ 1 1 ^ \ \ \ i .\ \ \ \ ( M K ) \ \ \ \ S / S \i \ s 1 r \ ^■< / M k / V \ 1 \ N s i^ \ > s ~f\ H ^ 1 ■^ - - 1 \ \ 1 — < / 80 S 70 60 12 I Z 3 4 5 6 7 8 9 10 II It 1 ? 3 4 5 6 7 8 9 16 II 12 •£ 3 k- - -.-A.M.-— ..v"-J<-— — P.M.- -J Hours of Day Fig. 9. — Hourly variation in flow of sewage in various cities. The records used in preparing this diagram were from the following sources: Birmingham, England, average of two years, 1906-7; East Orange, N J., March 16-17, 1910i GloveraviUe, N. Y., Oct. 30, 1906, and Sept. 12, 1907; City A, of 15,000 population, typical average curve; Milwaukee, Wis., Oct. 24-28, 1920; Toronto, Ont., 1900 and 1908; Worcester, Mass., Nov. 13, 1909, and March 21-27, 1910. because he has to provide sufficient sewer capacity to carry oft' the largest quantity of sewage which it is reasonable to expect. In Fig. 9 are plotted the hourly rates of flow in several cities, expressed as percentages of the daily average. The smooth QUANTITY OF SEWAGE 43 curve is drawn through points obtained by averaging points taken from curves representing the hourly rates of flow in the cities named in the note below the illustration. In preparing curves of this nature it is desirable to synchronize them by mak- ing an allowance for the time required for the sewage to flow from the city to the gaging point. This time differs, and if an allowance is not made for it the data from different cities are not easily compared. The dotted line in Fig. 9 represents the flow of sewage from a large Milwaukee residential district. It is interesting to compare the hourly fluctuation shown in Fig. 9 with the fluctuations in water consumption referred to previously in this chapter (see Fig. 7). Sewage from Residential Districts. — The quantity of sewage from a residential district will depend upon its area, density ..140 .120 100 \\ V \ \ "^'^ « \ ^■v -47 k. %,-» K°^^s^,sW, =■<; /S-£ n ^14^ v^;,ri -■i: J — __ ^ ^ Sg — ■— -o -i 80 60 .E o 20 SO 500 1000 1500 EOOO 2500 3000 3500 4000 4-500 6000 Area in Acres Fig. 10. — Maximum density of population, Lynn, Mass. of population, per capita water consumption and the quantity of ground water leaking into the sewers. The future character of such a district should be considered carefully, keeping in mind that there is often a lowering of the per capita consumption of water as widely separated homes of wealthy families give way to more numerous and modest homes or to tenements. In making estimates of the quantity of sewage from such a district, it is incorrect to use the average density of population of the entire city, which rarely exceeds 25 persons per acre, because there are many residential districts where it is much higher. For example, it was found in Philadelphia in 1910, that the population per 44 SEWERAGE AND SEWAOE DISPOSAL acre was 58 to 64, where the prevailing building type was pairs of two- and three-story houses, and 97 to 123 where the buildings were solid rows of houses. The sewage from the former type of district amounted to 9,800 to 14,200 gal. per acre daily, and from the latter type 10,500 to 26,300 gal. The decrease in average density of population with increase in area is a factor of importance in the design of separate and intercepting sewers. It is impossible to predict with certainty the locations of future areas of high density. This Is a further reason for liberality in the design of the laterals. Fig. 10 10 10 dO 70 80 90 100 40 SO 60 Area in Acres Fig. 11. — Maximum rate of discharge of residential sewage, Lynn, Mass. shows the actual maximum densities for different areas found in Lynn, Mass., in 1916 and the densities estimated for the year 1950, during which period it was assumed that the total popula- tion would double. Data sufficiently accurate for making and checking such curves may be obtained by actual count, by the use of assessors' estimates, from the census reports where the population is sufficiently large so that it is divided into wards and precincts, or from data contained in directories, post office counts and lists of polls. It is evident from the foregoing that the maximum rate of discharge of residential sewage per unit of area may be much higher for small areas than for larger ones. Fig. 11 shows the QUANTITY OF SEWAGE 45 estimated maximum rate of discharge of sewage for different given areas based upon densities of population shown in the higher curve of Fig. 10 and the maximum per capita rate of sewage flow of 220 gal., otherwise determined for the district under consideration. The rate per hour fluctuates greatly in the case of moderately- developed medium and high-class residential districts, because of the great variation in the hourly consumption of water. Fig. 12, curve A, gives the average hourly rate determined from two sets of gagings of a Cincinnati district of 1,617 acres with a population of 11.1 persons per acre. This district is very largely residential, and the maximum observed hourly flow was 273 per cent of the average although the district is so large that such a wide range in the rate of flow in its main sewer is unusual. In the usual case of a district of this size, the maximum hourly rate may be expected to reach 160 to 170 per cent of the average. Sewage from Commercial Districts. — A commercial district yields much more sewage per acre than a residential district, because of the concentration there during the day of many per- sons in stores, office and public buildings, hotels and often some apartment houses. Such a district is relatively small compared with the other districts in a city, but its growth should be fore- casted carefully on account of the large quantity of sewage coming from it. The allowance for used water from a mercantile district is more difficult to estimate than that from a residential district. If the estimate is made in connection with the design of inter- cepting sewers, pumping stations or treatment works for use with existing main district sewers, the latter should be gaged and a suitable allowance made for future development. It may also be helpful to ascertain the water supply of the district from both municipal and private sources. The average quantity of sewage from a 123-acre commercial district in Philadelphia was 92,800 gal. per acre per day. This district was entirely built over, the area occupied by buildings being 80 acres and that by streets 32 acres. This is a rather large quantity of sewage from a commercial district. Gagings of the flow from eight such districts in Cincinnati ranged from 14,000 to 72,000 gal. The hourly rate of flow from districts of this character varies greatly, as is to be expected, but the high rate of discharge lasts 46 SEWERAGE AND SEWAGE DISPOSAL 16! — -1785 Hours of the Day. Hourly Variation in Flow of Sewage , Curve A. Ross (.Bloody) Run Sewer. 6 8 10 12 Z 4 6 8 10 12 •-A.M. ■>]< _._.-pM..._„. ->j Hours of the Day. Curve 5. Vine Street Sewer. 8990---1I1900 7490 b-eS'ESO g ° S 5390 g---74600.a 4490 g-55950 g 2993--37300^ 150 125 100 75 50, X / \ \. .^ / / /n dush hit isfric t \ N — > — 12 2 4 6 k. -A.M 10 12 >l^- 6 8 10 12 -P.M.— -iJ 545 — o 4541 10394 o -8661 fe 363 r-6929j 272 £---5197 i ,182 3465 Hours of the Day. Curve C.Marshall Ave.Sewen- Fig. 12. — Hourly variations in flow of sewage Cincinnati. QUANTITY OF SEWAGE 47 much longer than does the early morning peak in the curve of flow from residential districts. Curve B in Fig. 12 illustrates this, and shows the hourly rate in a district of 23.6 acres with 12.5 persons per acre, the center of business in Cincinnati. The average daily flow in this case was 72,000 gal. per acre and the maximum during the gaging period was 139,000 gal. Sewage from Industrial Districts. — The quantity of industrial wastes not originating in the public water supply should be studied individually in each case. The following estimates, in gallons per capita daily, have been made in connection with the design of sewerage works: Cincinnati, Oliio 50 Fitchburg, Mass., maximum 81 Louisville, Ky 57 Neponset Valley intercepter, Mass 25 Milwaukee, Wis 57 Passaic Valley intercepter, New Jersey, maximum 38 Paterson, N. J., maximum 18 Providence, R. I., maximum 42 The variation in the hourly rate of flow from a typical indus- trial district with some residences, in Cincinnati, is shown by curve C, Fig. 12. It had an area of 294 acres and a population of 19.1 persons per acre. The curve is based on 3 days' gagings during which the average rate was 6,787 gal. per acre and the maximum rate 13,485 gal. This maximum is equivalent to an hourly rate of 708 gal. per capita daily, 198 per cent of the average daily rate. Provision for Storm Water in Intercepting Sewers. — There is a general impression that it is wise to provide in intercepting sewers for a small quantity of storm water, expressed often as being suflScient for the "first flushings" of street surfaces and sewers. This impression is based upon the assumption that there are accumulations of sewage sludge in the sewers and quantities of filth on the streets which will be immediately flushed into the intercepting sewers with the first run-off due to rain. In some sewers laid on very flat grades and where sewers have settled or have been built with depressions in them, there may be such deposits, but where sewers are laid on grades which give satisfactory velocities such deposits are believed to be exceptional. Where deposits do occur they are generally found to consist largely of sand and other heavy detritus, dropped by the current, 48 SEWERAGE AND SEWAGE DISPOSAL as the velocity of flow decreased with the decrease in volume and hence depth of flow in the sewer. It is apparent, therefore, that these deposits will not be picked up again by the current until a very substantial rate of discharge is reached. Intercepters are fed by trunk sewers serving rather large districts. Considerable time is required to flush the major part of the systems to the intercepter, during which a large flow is likely to reach them from the nearer portions. Unless consider- able surplus capacity is provided, the intercepters will often be running full before the flushings from much of the tributary area can reach them. Therefore, too much stress should not be laid on their ability to care for the "first flushings," although as ordi- narily designed they can accomplish something in this direction. Assuming that the average daily flow of sewage is 100 gal. per capita and that the capacity of the intercepter is 300 gal. per capita, there will be a surplus capacity available for "first flushings" equivalent to twice the average flow of sewage, if such flushings come at a time when the flow is at the average rate. The maximum rates of flow generally occur in the spring when ground water is high and at other times there will always be some surplus capacity. Furthermore, as intercepters are built for many years, there will be a considerable excess capacity during the earlier years, although this should not usually be counted upon to care for storm water, for it is a gradually diminishing allowance accompanied by a gradually increasing need, if need there should be. •Under the foregoing conditions the sewage will be diluted to three times its normal flow. Furthermore, consideration must be given to the excess of water used in this country, giving a more dilute sewage, and to the quantity of ground water which leaks into the sewers. With these eliminated the quantity of sewage would be comparable with that obtained in Europe. Taking all these conditions into account, it is evident that the dilution approaches the standard of the Royal Commission on Sewage Disposal, which is six times the dry-weather flow. In view of these conditions, it may be assumed that the 300- gal. daily per capita intercepter capacity, allowed in a previous paragraph, contains a sufficient allowance for storm water. Caution. — The foregoing outline of a rational method of estimating the quantity of sewage to be provided for applies to the design of intercepters, pumping equipment, grit chambers, Portion of Carus^ Brook DRAiFMOEDisnucr Fig. 13. — Map for problem of separate sewer design. y^acma Page 48) QUANTITY OF SEWAGE 49 etc., and to large trunk sewers and the units adopted are for maximum rates of flow when the sewers are running full. These principles, also, constitute the basis of design for a separate system of sewers, or a system from which it is intended that storm water shall be excluded. It is important to note, however, that in practice it is not uncommon to find a consider- able amount of surface run-off finding its way into manholes and numbers of storm water roof conduits connected to separate sewers, in spite of rules or ordinances prohibiting such connection. It is wise, therefore, to make some allowance in the maximum rate of sewage flow, for a certain amount of storm water. The steps and basic data used in the design of intercepting sewers are similar to those used in the design of a separate system, an example of which follows, worked out in detail. The Design of a Separate Sewer System. — The district shown in Fig. 13 is a. portion of the Carlisle Brook District of the City of Springfield, Mass. This is a residential district, which, in 1920, was about two-thirds developed with detached houses. This district is so nearly fully developed that the probable future density of population can be estimated with a fair degree of certainty and it is not necessary to make a study and curve such as is shown on Fig. 10. It is estimated that the future average density of population will be 65 persons per acre. The maximum rate of flow of sewage is estimated at 250 gal. per capita per day, including 30 gal. per capita per day allowance for storm water. The subsoil of the district is sand and will permit a large portion of the rainfall to percolate into the ground. The maximum rate of infiltration of ground water into the sewers, to be provided for, is 2,000 gal. per acre per day. The minimum size of sewer is to be 8 in. The minimum velocity of flow in the sewer when flowing full is to be 2 ft. per second (the more desirable limit of 2.5 ft. was impracticable because of the prohibitive expense that would have resulted from its adoption). The minimum depth of cover allowable is 7 ft. Problem. — Design a system of separate sewers for the area included within the dash-dot Une. This system is to discharge into the Carlisle Brook main sewer at the right-of-way leading from Shattuck St. about opposite Courtland St. to the main sewer. Prepare a plan showing the location, size and slope of each sewer section between manholes in each street. Prepare a profile of each sewer showing all connections with each lateral sewer. On the plan show the location of manholes. The profile should show the street surface; the manholes; the proposed sewers, their sizes and slopes; and the invert elevations in figures, at each manhole. 4 50 SEWERAGE AND SEWAGE DISPOSAL Determine the required elevation for the Carlisle Brook main sewer at the right-of-way. This problem may be combined with the problem for the design of the storm-water drainage system. This would, however, involve additional complications, as precaution would have to be taken that proper connections of sewers and drains could be made without interference with each other. It often happens in practice that this feature of the design of a separate system is very troublesome, particularly where a system of sewers has been installed without particular study of the desirable relative position of the pipes in the streets when the drains shall have been added. Having determined the basic data for any problem, the different steps required to reach a rational solution are as follows, in the order given : First. — ^Draw a line to represent the sewer in each street or alley to be served. Near the hue place an arrow pointing in the direction in which the sewage is to flow. Except in special cases, the sewer should slope with the surface of the street. It is usually more economical to plan the system so that the sewage from any street wiU reach the point of disposal by the most direct and consequently the shortest route. If the lines are properly drawn, the system will resemble a tree and its branches, all the laterals connecting with the sub-mains and these in turn with the main or trunk sewer, which leads to the point of discharge. Second. — ^Locate the manholes, giving each an identification number. Manholes should generally be placed at all angles, or bends, changes in grade, at all junctions of common sewers, at the upper ends of all laterals and at intermediate points where the distance exceeds 400 ft. or such other maximum hmit as may be estabMshed. Third. — Sketch the limits of the drainage areas for each lateral, unless a single lateral wUl be required to accommodate an area larger than can be drained by the minimum size of sewer with the minimum slope, in which case a further subdivision must be made. Where the streets are laid out, the drainage hues may be assumed as being midway between them. If the street layout is not shown on the plan, the hmits of the different areas cannot be determined as closely and the topography must serve as a guide. Fourth. — Measure the area of the different drainage areas. For this a planimeter will give results within the accuracy required. Fifth. — Prepare a table, see Table 7a, for the different steps in the com- putation and for each length of sewer between manholes. This is the most concise, time-saving method and keeps the results in better shape for the subsequent use of the draftsman. Use column 1 for numbering the hnes of the table, for ready reference. Determine by inspection the manhole which is farthest from the point of disposal and enter its identification number in the first fine of column 2, and the number corresponding to the manhole next on the Line toward the trunk sewer in column 3. Enter the name of the street or alley in column 4, the length between manholes in column 5, and the area in acres to be drained by the sewer at a point just above the lower manhole in column 6. On the next line enter the same data for the next stretch of sewer and in column 7 enter the sum of the drainage areas listed in column 6. The area in column 7 is the basis for computing the required capacity of the sewer. Enter the data for each stretch of sewer in the QUANTITY OF SEWAGE 51 0.3 i S g S S-o pj V ^ o a a V 1 g.-3 ■s lilo 10 M-O B 03 a O tk. « !:; 2 «■« 3,2 Q." .4. S 3 S g EHga O ^^ OS 13 O a - a) MS OS'S 03 «-a a:£ OS 3 1^- t^'*r-co>-icot*r-0030aiO« o ■tjH w 1-H oj ^•' ^-^ o ic ■<}<■'*< CO (N w f-J .-H o OIOJOOIQOOOXWOOMMWOOWWCOOO I— I^Hi-HTHr-lr-li-lrHr-lT-li-Hi-li— li— ItMi-lrH Oh-'*'HOC«D<-ttDi-Ha>t^iMO0lO"*O OrHCCOOOCOCJCOi-if-OltOfflt^t^i-i OOtDrJttNi-HOit-^h-cDiOTi^TiHCOWciiHi-i ©OSOSOSOIOOOOCOOOMOOMOOWOOOOM (NUicD^-^i-^tNNOicOi-iXfNtDt-CCO CD (D CD t^' ^*' CO W M ,-J i-I M M CO CO CO (N «3 00000000000000000 «NC<)CgN«MNMMMMNNCqM--H COOOMOOWOOWiMlNiMCD iOiOiOinu5»0»OTjH-^COCO(NW(NNrt'-H 00000000000000000 00000000000000000 ddddoddddoddo'dddd oooooooooocqooocooo (N «" w -i '-H .-I rH cm' (N (N (N CO CO oooooooocooooooowwiraioiciiooow Wi0'<*-Ni-H .■-t-a>0'OM03l>Nt-«.-i . W CO lO r- CTi i-H CM CO 01 Ol M CO . .OOOOOrHr-(W.-I.HCq(N • d d d d d odd ci<6d><6 .wCOO'Mb-OCDOiOOO'-' .(Ni-i«l^tDCDTt d »o b-' 01 ■* CO >c 00 o 10 ..-ItHC^C0'*»CCO00OJOJ'-''-h "^ .OSOWb-OOCOTfincOCM . . 00 O CD O 00 CO CM --I l> T}< (N ! ! 10 d »o CM >-< w d c^' (N N -(jH ood'^ 000 CSINCM MMOO ICHOIC 000 000 000 N W (N <6c6<6 icioiocowoT-fioiraoNOiocomioo ^^^iOOOOCCCMCMeO«OOt^«-<"*t^CO CO?0COi-)Tt< NCMi-i CMCqi-HCMCOi-i 000 lOioin > > « 03 03 > >. t>. > O oMja^MMu^ 0055535^ IV 0) O) bD bo be ,D^J3 00a CO^'OCDb-OOOlO^WCOTt^lOCOt^OOOl iMCOT*HiO OOOOOOOI;^ Slope, feet per foot lO 00 00 00 00 00 (N »n lo lo lO lo Tj, . . o o o o o o o o o o o o o -g 066006 Veloc- ity, feet per second '*' cj (N M 1m' IN im' oi "S : Capac- ity, cubic feet per second CO t^ t^ t^ t* t^ rH Q . 6 6 6 6 6^" ■ Size of sewer, inter- nal diam- eter in inches (N 00 00 00 00 00 si . ^ ' Total maxi- mum flow, sewage and ground water Cubic feet- per second - . . . . ,-H Tj, . . . . . 10 00 . . . . -^ <» Million gallons daily o 0.328 0.539 0.642 etc. Ground water at 2,000 gallons per acre daily G) 0.036 0.059 0.070 etc. Sew- age 1 million gallons daily « 0.292 0.480 0.572 etc. OS 1 3 . . . 00 CO . . . 1-H CV OOmiCiCOOMt^ON i-IWOCOlOt-rJfOSrH o "Ct-ffl-^coiraooom CO-<}TH-cy3 COC0C0-iMC0"*»0«0t^000) e o M (M CO ■* lo OS m O,HNc0Tt(iCeD!^« Num- ber of line - tHiMCO'^iOOb-OOOSO'-HiMCOTfiCfCi 54 SEWERAGE AND SEWAGE DISPOSAL Cambridge St. would be about 3.5 ft. deeper for 300 ft., (3.5 X 3.5 X 5) ^ 27 = $2.27; for 300 ft., this will be $680, and the sewer in the right-of-way would cost say $6 per foot for 286 ft., or $1,720; the cost of the right-of- way wiU be about $500. The net saving, then, would be about $3,200. SUMMAKY OF SaVING EFFECTED BY PhOJECT B 1. Saving in trunk line changes west of Cambridge St $2,265 2. Saving in Andrew and Burr Streets 3,835 3. Gross saving $6,100 4. Increase in Cambridge Street $680 5. Sewer in Right-of-way 1,720 6. Cost of Right-of-way 500 $2,900 7. Resulting net saving of Project B over Project A.. . $3,200 Whether or not this saving is justified in spite of the disadvantage of laying the sewer partly in a private right-of-way, may depend upon other considerations than relative cost. CHAPTER III QUANTITY OF STORM WATER The rate at which storm water, including melted snow and ice, flows from the surface of the ground into drains and combined sewers must be estimated before designing these conduits. The quantity of storm water is usually so much larger than the quantity of sewage for which combined sewers are designed that their size is determined primarily by the rates of rainfall and run-off and the slopes of the sewers. The amount and intensity of rainfall can be measured. The run-off or part of the rainfall which reaches the drains or combined sewers depends upon the ratio of the impervious to the relatively absorbent surface, the intensity and duration of the rainfall, the character, shape, and slope of the drainage area, etc. It is manifest that a smaller proportion of a light rainfall on a dry sandy tract will reach the sewers than in the case of a heavy rainfall on a clay tract of the same size and topography. In sewerage engineering the rate of rainfall during periods just long enough to produce maximum run-off conditions is more important than the total daily rainfall. Such rates can only be measured practically by automatic recording rain gages, and on account of the small number of these instruments in use until recently, there are few long-time records of this character. Automatic Rain Gages. — The gages which give the records most helpful to sewerage engineers make charts on which it is possible to determine accurately the precipitation in time inter- vals as short as 2 and 5 min. In all such instruments the rainfall on a circular collecting area or opening, from 8 to 12 in. in diameter in different instru- ments, passes to an operating mechanism, of which there are three general types. The first type receives the water in a container supported on a weighing device which actuates record- ing mechanism that moves a pen pressing on a chart held by a drum revolved by clockwork. In the second type, the rain water is delivered to tilting pans, which tip when full and thus discharge their contents. Every act of tipping actuates an electrical or mechanical device which makes a record of the 55 S6 SEWERAGE AND SEWAGE DISPOSAL discharge on a chart moved by clockwork, and as the quantity of water discharged by each tipping is known, the rate of rainfall is thus determined. In the third type, the water flows into a container in which there is a float with a vertical arm carry- ing a pen. This pen records the height of water in the container on a chart moved by clockwork, and the container is usually of such size that the float will rise at a faster rate than the rate of rain- fall. Fig. 14 illustrates a tipping-type of gage made by Julien P. Friez, Balti- more. The rain is collected in a funnel 12 in. in diameter and conducted through a tube into a bucket with two compart- ments, each holding the equivalent of 0.01 in. of rainfall. The bucket is sup- ported on trunnions in such a manner that as soon as a compartment is full the bucket tips, the com- partment is emptied and the other compartment brought into Fig. 14. — Friez tipping bucket gage. Fig. 15. — Register for Friez gage. position for filling. Each time the bucket tips, it closes an electric circuit which causes a pen to record a step on a chart carried by a revolving cylinder, Fig. 15. A sample chart from this instrument QUANTITY OF STORM WATER 57 is shown in Pig. 16. The curve traced does not represent directly the progress of the storm, the motion of the pen being reciprocat- ing, up for 0.05 in. and down for 0.05 in. ./l.-.iS... R 1 \ 32 It ! 7 \ i ; 1 $ i ^ \ =i= 1 ; ^ £ = ^ ; 3 ; ^ f ^ = ^ g 1 2 ^ ] 1 p : T « 5 KUMARIUi •M »»t r..- T. : ~"U.,^^(!.a?r..,..,«. _. " " ■■ ffi ri = ^; — p — fVi sntJ ± C ?; ■& >..^™^ -. 0< »«f at - -TT^ =|r>. ...^ ^ ^- - t"- _-_. : H- -qs i^ ai T" = -- 4- y = : = £: = + : 6 :r= - , RBMARHA- 1 ta^ r*. To i — li>>tiiiii..£'.> -«0J — p LO*( -— — — — odoooc 00 jnoH -Jsd ssLjDui - fi4.isua+ui !m o ■ /f 1 1- r ^ Pa h / 1 1 ' Is o c =^ /\jl ^Iji ".9 < -^^ *l 1 r 4 AT ^L < ^^A .'0? y 1 , ^^ ^ 10 1^ /" ^ Vi^S o ^ ..'< 8^=^ 'y^ ^'t- '^^'^ -• 5 ^ ^S' 4 " ..*^=-''' ""2^\ ^^^: 1 •■■f' a p> QUANTITY OF STORM WATER 61 S \ ^ g o lO m + E o 8 O Si s 1 f ' ^ OS 5 1 1 » fi'l 1, T) '/; i O 1 1 / 1 / 1 // f. s? /; 's '1 n k ';'; -4' S g /< ^/ li 1 i 1 // / 1 ^ } ]? /. ^ $ o 8 pq / ^1 1 ^^-1 y ^^ Ij o t / / ^r '/< t 1? Pi 2'. - I / 5 Ha .1/ J D t- ^D LO"''^ lO N — O <5> <0 r- - ^_:_ _-_ _ -ooc M jnoH-i3d saqaux-h+isua^uj 1 • ' -4^ o 3 ■i 40 50 , 60 7 Dura+ion - Mmu+es -Rainfall curves for Chest fti Wi 1 h 'i y s) y r f % /. V ' A .^^ K> 1 o 1 y "J^ r? ^ 1 f' ^ i^ % •^ ^ > «^ ^ o ^L, ___^ ^ fftf -ii-i* 1 N Si ^. o CO r- iD lO -^ K> 62 SEWERAGE AND SEWAGE DISPOSAL ton, Tex., based upon a 30-year record from 1890 to 1919, is shown in Fig. 21 Curves drawn through the points representing the maximum rates of precipitation are obviously irregular in form. From suitable data, however, approximate smooth curves can be drawn and many of them may be mathematically expressed. Such curves showing the relation between intensity and duration of precipitation for extreme and other storms at several places are reproduced in Figs. 22 to 26. u7 i 1 \ .■XYEARFRB9UBNO ^^ N ^ ^ ^ ^ -.^'•' ■■yl5 ,. ^ ^ ^ ^ ^-< :^ - 6 5 '£AR FRBQU f/vcy. pSss ==*=: i= — == " »* -'' =^ in o c -♦- I 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 Dora+i'on in Minu+es Fig. 21. — Rainfall curves for Galveston, Tex., from 30-year record (1890-1919). The selection of the rainfall curve to be adopted as the basis for design of sewers in any community is a very important matter. While it would be desirable to provide drainage for the greatest storms of which there are local records, such a course will usually require very large expenditures which in many cases will be prohibitive, particularly if proper provision be made for increases in the proportion of impervious area likely to take place within the next 30 to 50 years in a rapidly growing community. It may be necessary, therefore, to reduce the allowances and one of the most logical ways in which this can be done is by the adop- tion of a rainfall curve having a lower frequency, as, for example, 15, 10, or even as low as 5 years. In making the selection, con- sideration should be given to the several economic conditions of the case, as the funds reasonably available for construction and QUANTITY OF STORM WATER 63 5 ~ ~ ~ ~ ~" "■ — ^ >» "" IS Relation between ~ TImB and Intensity of Rainfall according 1o Talbot's formulas For Eastern U.S. CurvBAt Eittreme SformSt /'• ^f^ ■ Curve B, Heavy Storms, i- jjif Formulas eterivedl byAM.TalbofJSU Ploited Poirrh are from Tram JmSotCt, Vol.S4(li05),p.ZIZ. - - n \ 1 II \ » in [ ;;■"' \ \ 9 \ • 8 r \7 \ \ i-6 s " : 5 \ V \ V, -/ 3 ^4 s ~ \ ^J ! 3 s ^~' *- ^ s "" --• •^ ■ t ^ .<* ""■ — -. ■^ «.. ■*~ "- — L« - -J •-'- Sa - — ^ ^ ' — — — -H L_ — 1 " 170-110 I9Q ZOO Tlm6(f)in Minutes Fig. 22. — Talbot's intensity of rainfall curves. If 10 U 1 |8 O . V- Q) k- Q) S «> Q. ^ ^- o O) C ft t ^ -£ I ~ ^~~ *"— ■ ' ■" r— p— n _ Relation between Intensity and \ Durationof Rainfall as Shown by Pluviometer Records During theYear 1911. Phil»delphia,Pa. r.irrues Deduced from Records of \ ' 1 \ i \ \ ^^ V rv ?v r \= ^ 25 Years. 9-n N^ ^ \ ii> s \ t. X vJ>^ L \ > L \ hv r.* ■"" _£^ ^i-p _ s r^" "i/, — ■— ~J I CI ; S-^" •^ p — . -ye {• /? ^~ ~- — — — f's _ 1— r— "^ — ^ Curve ^ _ ^ -'■■ih- _ I — — ' — — — ' — — — — — 10 ZO 30 40 60 60 70 80 90 100 110 120 130 140 160 Duration of Rainfall in Minutes. (t> FiQ, 23. — Philadelphia intensity of rainfall curves. 64 SEWERAGE AND SEWAGE DISPOSAL W n p r p m r-1 r- — — — — — - — — — - - - ' Time and Intensity of Rainfall at St. Louis, Mo. g t, X " S ,/> = -, Curve ^^ o 9 1 /■ = - 1 {W.W.Horner 1910) j - Observations bfU.S.Weather Bureau 1872-1909. 1 1 1 1 ) ) 1 1 1 r 1 1 1 1 1 1 1 t t f ° 1 fe . I \ * n \ •it'' I 5 -< K \ \ ^4 -< \ ■=1-1 r~ s $ 3 i ^, -^ , ■* ^ a I — -4- _ _ — — = ^ _ — _ _ _ _ _ _ r - — — - — n - — - ~ r r r 10 so 30 40 50 60 70 60 90 100 110 IZO 150 140 ISO 160 170 Time lt> in Minutes. 190 ZOO Fig. 24. — St. Louis intensity of rainfall curve. In 1916 a curve of somewhat higher intensities was adopted by the city for the design of drains. 10 20 30 40 50 60 _ 70 60 90 100 110 120 130 140 150 1^9 170 J89 190 200 Time it) in Minutes. Fig. 25. — Denver intensity of rainfall curves. QUANTITY OF STORM WATER 65 the probable amount of damage and inconvenience caused by occasional flooding. Fonn of Rainfall Curve. — Most equations of rainfall curves have been written in the form i = a/{t + b). It usually ex- presses the actual observations with a fair degree of accuracy for rains of 10 min. to 2 hr. duration. Beyond these time limits, the results are generally too low. In many cases the exponential equation, i = a/t'' has been found accurate. The exponent b is usually between 0.5 and 0.7. The best practice is probably to use a curve of rainfall plotted from actual records. ~ - "~ "~ ~" ~" ~* "■ — T- - *• ' ■ ' . r . 1 1 1 1 l fiela+ion between 1 Intensity ot Rainfall at Ff^ANCISCO, CAL. B,fioformi//a,(C£.6ri/ffsfy/SiX0 * tsenfprfncipatsfomts /B$$-/903 I5.yfeafher Bureau m.SocCE.Vol si^zr.m. 1 u », n es.p.3l6. " III ^10 5>AN " .«" ■F. a y Cvrvf g' " C8 S7 s ie - f - * " ^ . . 'S.-1 ■, •S^^ £S - [- ' " ~ ~ ~ ~ ~ ~ — 1 -^ ? r^s ° S V:S ■■ " 1 "^'i^ '^'*-" ' 1 __ M r - — ^. ■■' ._ . " ~" ~ "' ~ r- "■ — — « 10 ZO 30 40 60 60 70 80 90 100 110 120 130 W 160 160 170 ._. TimeWin Minutes. FiQ. 26. — San Francisco intensity of rainfall curves. without attempting to express it by an equation. The intensity to be expected for any given duration of time would then be taken directly from the curve, instead of being obtained by solving an equation. There is no apparent reason why the relation between the intensity and duration should follow any mathematical law. Table 8 shows how the results of different formulas differ. Phenomenal Rainstorms. — Storms of extreme intensity, com- monly called "cloud-bursts," are occasionally experienced. They are usually of such rare occurrence as to be classed as "Acts of God," for which it would not be reasonable to provide SEWERAGE AND SEWAGE DISPOSAL Oi-cOO O 1 ift OO O** co-tjt CO ■—I WOO o OO 5 s t- OJ ■>«lCQ«Si-tt-CO s t^ CO CO CO o BSSS K OOS s o o o o o OOOt-iOO o o o o o o oooo o OO o ooood § s? ^ ^ 00 CO ssssss g ^ o s ■^ CO OOOcCiM S3 OMTO e S?^SgS5 ^ o o o o OOOi-i wo o o o o o o OwOO o OO o ooooo o o' o o c K S O tH jH--l eg i-» 1-1 w O O O i-H ^ M '* ^H (M OO Tt' Oa 00 00 CO O eoco C^ "* ^ t--^ c»c CO ec (N c. u fl CI 3 SESS-st-a f- h ^ 3 o a ^ COCQrt- . ' H M M g I L-iO ffoco +S+2 I- + +S -COOJ U-c-* wM M +S+S+' rHiHlOOO CO II II II II II II II II II P P g S S g a a" a H" " o o QUANTITY OF STORM WATER 67 in designing storm drains. They are not to be confused, how- ever, with special topographical conditions, more particularly, in the mountainous country, producing records abnormally high according to usual standards but not unusual for the specific locality under consideration. These cases are rare, however. During 1913 New York City experienced four storms, in all of which the intensity of precipitation, practically throughout the storm, was greater than that given by the equation i = IS/i"-*. The significant facts relative to these storms and the intensities obtained by this formula are given in Table 9. Table 9.— Phenomenal Rainfall in New York City, 1913 Date July 10 July 28 Sept. 5 Oct. 1 15 Place 100 Broadway CeDtral Park Central Park Richmond i-minutea 1 ntensity i-inche 3 per hour 1 2 4 5 7 10 15 19 30 37 49 59 60 85 106 120 123 8.40 8.10 6.45 15.00 10.60 7.50 6.72 5.68 4.75 3.88 3.45 2.74 2.47 2.15 1.95 1.94 1.63 1.46 1.37 1.35 9.88 6.12 7.20 6.24 5.64 5.16 5.05 7.56 6.52 5.76 4.80 6.90 6.36 4.18 2.96 5.24 4.84 4.75 4.44 2.30 2.73 3.31 3.80 3.37 1.28 1.56 1.85 3.06 Rational Method of Estimating Run-off. — Any method of estimating run-off requires the engineer to rely upon his judgment. No two engineers working independently are likely to reach precisely the same conclusions regarding the extent of the improvement of a drainage district within the economic period for which drains should be built now, the rate of rainfall for which drainage should be provided, and the rate" at which storm water will reach the sewers. The rational method of estimating run-off has come into favor because it enables the engineer to apply judgment directly to specific allowances which are subject to analysis, measurement and estimate. It requires him to exercise judgment logically, 68 SEWERAGE AND SEWAGE DISPOSAL after an analysis of the local conditions. The rational method is based on the direct relation between the rainfall and run-off expressed by Q = CiA, in which Q — the run-off in cubic feet per second from a given area; C = a coefficient representing the ratio of run-off to rainfall, generally called "run-off coefficient;" i = the intensity of rainfall in cubic feet per second per acre (or nearly enough, its approximate equivalent, the rate of rainfall in inches per hour); A = the drainage area in acres. The engineer can measure A, but he must determine the proper values of the other factors. The value of i to be used depends not only on the curves of the intensity of rainfall which fit the local conditions and the assumed period of recurrence, but also on the "time of concentration" it takes water to flow from the most distant point in the district to the nearest sewer inlet and thence through the sewers to the point in the sewerage system for which the maximum quantity of flow is to be estimated, as a basis for design of the sewer section or size. It is the greatest uniform intensity of rainfall during this time of concentration for which provision must be made, not the intensity for a shorter period. The value of C must be estimated from a study of the soil, slope and character of the surface, and a consideration of probable future development. Time of Concentration. — Time of concentration is made up of "inlet time," the period consumed by water in flowing from the most distant point in the drainage area to the sewer, and "time of flow" in the sewer. The latter is readily estimated by hydraulic computations explained in Chapter IV. Inlet time is frequently the most important factor in determining the probable run-off, particularly in small districts or in fairly large districts with steep slopes. In cities it is seldom less than 3 or more than 20 min. Horner's investigations' led him to conclude that along improved streets with grades of 0.5 to 5 per cent, the water from streets, sidewalks and roofs will reach the sewer in 2 to 5 min., but the velocity over grass plots is so low that even in heavy rains water will take 10 to 20 min. in flowing 100 ft. The sewers are designed to meet the most serious conditions which are anticipated during the economic period of service for which they are planned. They are assumed in the designing work to run full, with the velocity of flow practically a maximum, and the run-off from the roofs and streets and the flow in gutters 1 Eng. News, Sept. 29, 1910. QUANTITY OF STORM WATER 69 at maximum rates, giving a minimum time of concentration which is a constant for a given sewer district in a particular state of development. When gagings of the storm-water flow are made, the conditions are usually different from those assumed in designing, and it is the actual time of concentration which fixes the period of rainfall with which the resulting flow must be compared. During a moderate storm, the sewer may be but partly filled and the velocity of flow may, therefore, be considerably less than the maximum. Moreover, unless rain has previously been falling for some time, the filling of depressions and the accumulation of sufficient head to cause flow over rough or nearly flat surfaces will require an appreciable amount of time. The actual time of concentration will, therefore, exceed the minimum in all cases except those which produce the same or a greater rate of run-off than that for which the sewer was designed. Run-off Factor. — The run-off factor depends upon a large number of elements and is not a constant for a giiren area, even during a single storm. It is seldom unity, even when the entire surface is covered with roofs and pavements, for some evapora- tion takes place, even during a storm, nominally impervious surfaces absorb some moisture, and irregularities of the surface tend to hold back some of the water. The run-off factor gradu- ally increases for some time after the beginning of a rain until the soil has become thoroughly saturated, the impervious surfaces thoroughly wetted and the depressions filled. After that time the factor remains substantially constant for a given area. It therefore makes considerable difference in the amount of run-off whether the critical precipitation comes near the beginning of a storm or after rain has been falling for some time. Prof. A. J. Henry of the U. S. Weather Bureau prepared' Table 10 which is helpful in this connection by showing the percentage of cases of downpour in Washington, Savannah and St. Louis, in which the maximum rate of precipitation occurred within various periods after the beginning of the storm. If a warm rain falls when the surface is covered with ice, the run-off factor may even exceed unity. It is evident that an exact determination of the run-off factor for conditions which will exist in the future is not possible. ^Bulletin D, "Rainfall in the United States" U. S. Weather Bureau, Jour. West. Soo. Engrs., April, 1899. 70 SEWERAGE AND SEWAGE DISPOSAL The engineer's problem is to forecast to the best of his ability the changes which will take place in the district and to estimate the rate of precipitation for which he must provide. When rainfall records from which to construct a curve of the Table 10.— Percentage of Cases in which the Maximum Intensity OP Pkecipitation Occurred within Various Periods from the Be- ginning OF THE Storm Minutes after beginning of storm Per cent of cases in which maximum intensity occurred within period atj Washingtou Savannah St. Louis 5 17 10 31 10 38 31 61 15 59 52 69 20 64 65 74 25 72 72 76 30 81 82 78 35 86 87 80 40 91 88 88 45 93 92 93 50 94 97 98 60 100 100 100 rate of precipitation in the city where the sewers are to be built are not available, the formula z = 12 -4- t°-^ will give intensities unlikely to be exceeded more often than once in 10 years in New England and New York. The formula i = 15/t°-^ will give the Table 11. — Quantity in Inches op Rain Falling in the Specified Periods of Time at the Rates Indicated by Curve op Intensities, i = 15A»-5 Time, minutes Rate of precipitation, inches per hour Accumulated depth of precipi- tation, inches 5 6.71 0.56 10 4.75 0.79 15 3,88 0.97 20 3.36 1.12 30 2.75 1.38 45 2.24 1.68 60 1.94 1.94 90 1.58 2.37 120 1.37 2.74 probable maximum rate of precipitation during a period of about 15 years and Table 11 has been computed by its use. In this table the periods of time are not necessarily measured from the beginning of a storm or even from the beginning of the downpour. QUANTITY OF STORM WATER 71 In making an allowance for the effect of the rate of precipita- tion on the run-off factor, it must be kept in mind that in most cases the maximum rate of precipitation does not occur over a large area, and it may be proper in planning a large drainage system to use a lower rate of precipitation than should be used for a smaller area. In a district where the duration of rainfalls is short, the run-off coefficient may not be so high as where the rainfalls are pro- tracted, so that the depressions in the surface become filled with water, the earth becomes saturated and the conditions are such that a larger proportion of the rainfall runs off than is the case with shorter rainfalls. The duration of storms has another effect on run-off, for if the duration of a storm causing flood conditions is less than the time required for water to flow from the most distant point on the drainage area to the point for which computations are made or at which gagings are made, then the maximum discharge will come when less than the entire drainage area is contributing water. This condition need rarely be considered in preparing designs, and then only for very large drainage areas, but it must be con- sidered in studying gagings of the storm flow in sewers and drains. There is some retardation in run-off due to the time required to fill gutters and sewers and build up sufficient head to carry off the water in the drains as fast as it falls upon the tributary area. In designing sewers, it is usually best to take no account of this storage capacity, but it must be considered in studying and comparing gagings of storm flow in sewers with rainfall records. Values of Run-off Factor. — There will inevitably be a differ- ence in the influence on the run-off factor which different en- gineers attribute to each of the conditions mentioned in the last section. These differences are shown in Table 12. Other authorities do not attempt to make close estimates of the different kinds of surface in an urban district, but con- tent themselves with average values of the proportional run-off, as follows:^ For the most densely built-up portion of the district . . 70 to . 90 For the adjoining Well built-up portions . 50 to . 70 For the residential portions with detached houses ... . 25 to . 60 For the suburban portions, with few buildings 0. 10 to 0.25 1 See also Horner's assumptions tor St. Louis, Mo., shown in Table 13 following. The authors have used similar figures modified to meet special conditions and have developed in some cases a zone system. 72 SEWERAGE AND SEWAGE DISPOSAL Table 12. — Range in Estimates of Run-ofp from Different Classes OF Stthface in Proportion to the Rainfall Intensity From Bryant and Kuichling's Report on the Adequacy of the Present Sewerage System of the Back Bay District of Boston, etc., 1909 For water-tight roof surfaces . 70 to . 95 For asphalt pavements in good order . 85 to . 90 For stone, brick and wooden block pavements with tightly cemented joints 0.75 to 0.85 For same with open or uncemented joints . 50 to . 70 For inferior block pavements with open joints . 40 to . 50 For macadamized roadways . 25 to . 60 For gravel roadways and walks . 15 to . 30 For unpaved surfaces, railroad yards and vacant lots. . 10 to . 30 For parks, gardens, lawns and meadows, depending on surface slope and character of subsoil 0.05 to 0.25 In general, in the absence of suitable information from which to estimate directly the run-off factor for a given area under con- ditions assumed to exist at the end of the "economic period of design," this factor may be most satisfactorily approximated by estimating the "equivalent percentage of totally impervious area," as it is sometimes called. Thus, if it is assimied that in the future 15 per cent of the area of the district will be covered by roofs for which the coefficient would be 0.95, 30 per cent by pavements with coefficient 0.90, 40 per cent by lawns with co- efficient 0.15, 15 per cent by gardens with coefficient 0.10, the resulting coefficient for the entire district will be 0.4875, or, in round numbers, 0.50. Relative Adequacy of Laterals and Mains. — Much more damage will generally result from the flooding of a main sewer than from the inadequacy of one submain or lateral; yet the damage from overtaxing the capacity of a large number of sub- mains may be much more serious than that from overtaxing the main sewer, particularly if the latter is provided with storm relief overflows. Moreover, it is simpler and usually less expensive to reinforce a main sewer than to rebuild many small laterals. The additional cost of constructing the latter of ample size when first built will generally be inconsiderable, while the additional cost of a main sewer large enough to care -for the run-off from the most severe storms many years hence, after the ground area has been made much more impervious by pavements, sidewalks and many more buildings, may be prohibitive. It is generally advisable, therefore, to build lateral sewers large enough for the QUANTITY OF STORM WATER 73 ultimate requirements, giving the mains and submaiiis a suffi- cient capacity for the probable run-off during some years, with the expectation that relief sewers will eventually be required to care adequately for the entire run-off from the district. SOME EXAMPLES OF THE USE OF THE RATIONAL METHOD The procedure^ followed in using the rational method of design for storm -water drains at St. Louis is as follows: The problem was to design a drainage system in a residence district in a new subdivision. An average half city block there measures 172J^ ft. from the center of the street to thecenterof the alley and 860 ft. between the centers of cross-streets, giving a total area of 3.38 acres, and its population is about 40 per acre. The impervious portion of this half block will be: Streets, 20,000 sq ft., 13.7 per cent of whole area. Alleys, 6,500 sq. ft., 4.5 per cent of whole area. Sidewalks, 6,500 sq. ft., 4.5 per cent of whole area. House roofs, 27,500 sq. ft., 18.9 per cent of whole area. Shed roofs, 4,000 sq. ft., 2.7 per cent of whole area. Yard walks, 2,000 sq. ft., 1.4 per cent of whole area. Total impervious area, 1.35 acres, 45.7 per cent of whole area. The rate of precipitation for which drains were designed in St. Louis in 1910 is given in Fig. 24. More recently a curve giving higher intensities has been used. For a rain of 10 min. duration it is assumed that 60 per cent of the water falling on impervious surfaces and 20 per cent of that falling on lawns and hard ground will run off. For rains of longer duration, it is assumed that these percentages will increase as shown in Table 13. These Table 13. — Assumed Percentages op Run- off fob Illustration of Rational Method Duration t in minutes Per cent run-off from Coefficient C Impervious portion | Pervious portion 10 15 20 30 60 120 60 70 80 85 95 95 20 30 35 40 50 60 0.40 0.50 0.55 0.60 0.70 0.75 percentages of run-off are based on the assumption that the critical rainfall of the assumed duration occurs at the beginning of a storm, before the surface has been thoroughly saturated. This is not always true, particularly for the shorter periods, and a somewhat safer basis for many cases would be to make no reduction in the coelBcient for the shorter times of concen- 1 A more detailed explanation of this procedure is given in " American Sewerage Practice," Vol. 1, p. 275-287, amplified from an article by W. W. Horner in Eng. News, Sept. 29, 1910. 74 SEWERAGE AND SEWAGE DISPOSAL tration. From these data the run-off curve, Fig. 27, is plotted to give the value of a for any time of concentration. This curve is only applicable to the conditions in this particular residential district. The run-off curve is constructed by taking the values of C as found in Table 13 and multiplying each by the value of i for the corresponding dura- tion as given in Fig. 24. The results are plotted as ordinates in cubic feet per second per acre to the corresponding durations in minutes as abscissas. For example, the coefficient of run-off, C, for 30 min. duration is 0.60 and the intensity of precipitation, i, is 2.75. Then the rate of run-off in cubic feet per second per acre is Ci = 0.60 X 2.75 = 1.65. Records of the future may make possible the introduction of a factor to cover also the effect upon the intensity of rainfall and rate of run-off of the relative extent of the area. Available data are too meagre to permit this now. ju" *^2 \ \| \ (E D ^~ — • .^^ _ - 10 JO 50 40 50 60 70 80 30 100 110 120 130 140 150 160 'Time of Concen+ration (t) in Minutes. Fig. 27. — Hate of run-off for St. Louis residence districts. The remaining data for the computations must be taken from an accurate plat of the district, similar to Fig. 28. On this are entered the elevations of the proposed or established street and alley grades and, if no contour map has been made, the existing surface elevations should also be shown. The storm water inlets along the gutters of the streets and alleys are then located on the plat on the higher side of all street intersections and at all low points between streets, with no interval greater than 600 or 700 ft. between inlets. After the location of the inlets has been fixed on the map, sewers to reach them are laid out, attention being paid to the sewerage of all the private lots in the district. The most economical layout usually follows the natural surface slopes in the shortest line toward the outlet of the district and concentrates the storm flow as rapidly as possible. Some- times several preliminary layouts should be made and compared to ascertain which is the cheapest. It is now comparatively easy to calculate the area tributary to the sewer at any point. The designer must form a mental picture of the district as it will be at the end of the period of economic construction, with the grading and paving done and buildings erected. This concept is necessary in order to locate the minor ridge lines dividing the small areas draining into streets from those draining into alleys, and to fix the areas tributary to each inlet. The final step in preparing the data is to fix in a preliminary way the grades of the sewers. The start in this work of approximating grades is made at the lower end, where the elevation is fixed approximately by FiQ. 28. — Sewerage map of a portion of St. Louis. (Facing Page 74) Portion of CARLISLE BROOK DRAINAGE DISTRICT Fio. 29. — Map for problem of storm water drain design. ilnsert FoUawing Fig, 28' QUANTITY OF STORM WATER 75 outside conditions. Then in the second trial, beginning at the upper end, the final grades can be established at the same time the sizes are determined. A simple problem from the authors' work will be used to illustrate the various steps taken in the computations. Hypothesis. — The district shown in Fig. 29 is the same as that used in the problem outlined in the previous Chapter on "Separate Sewers, viz.; a portion of the Carhsle Brook drainage area in the City of Springfield, Mass. This district has been selected as a basis of study because it is part of an actual problem and is of interest on that account. Some modifica- tions, however, have been made to simplify the conditions. The location of the proposed main drain (the so-called "Carlisle Brook drain") is shown on the plan and the invert elevation is given at the point where the pro- posed branch drain is to be connected and for which provision has been made in the design of the main drain. The general scheme of drainage of this portion of the city provides for the deepening and improving of the channel of Carlisle Broook, so that for a few years it will serve to carry away the storm water run-off from this drainage area. For this reason the branch drain must be constructed at a sufficient elevation to enable it to discharge into the improved brook channel and also be capable of extension to the Carlisle Brook drain on the further side of the brook, when that drain shall have been constructed. The re- quired minimum elevation of the invert of the branch drain is, therefore, given at the proposed point of discharge into the Carlisle Brook drain. A careful study of local conditions, including the present and probable future development of the district, indicates that a coefficient of imper- viousness of 40 per cent is a reasonable allowance, and that it is reasonable to assume a coefficient of run-off of C = . 35. The time of entrance, that is, the time required for the water to flow from the most distant point of the drainage area to the upper end of each drain, has been assumed to be 20 min. The rate of rainfall is to be taken from the assumed curve of intensity of precipitation represented by the formula i = 20.4/i''-" where i is the inten- sity of precipitation in inches per hour and t is the time in minutes. The formula indicates the rate of rainfall which may be expected to be equalled or exceeded once in every 6 years. It resulted from a careful analysis of the rainfall records. While it was recognized that drains designed on this basis might be overtaxed on the average once in 5 years, in view of other financial obligations facing it, the city was not thought to be justified in going to the extent of providing for storms of greater intensity, which would have involved greater cost. During the earlier years of the life of the drains they will be able to care for much greater rates of flow than later, because the assumed coefiicient of run-off is based upon future rather than present conditions. A progressive increase in impervious surface and rate of run-off is caused by the gradual substitution of roofs and paved areas for woods, grasslands, etc., usually referred to as unimproved areas, as the city grows. In the future when the district is more densely built up and when the city has available funds, relief drains can be constructed, to serve those areas where flooding has become sufficiently serious to warrant the expenditure. 76 SEWERAGE AND SEWAGE DISPOSAL Fig. 29 shows the streets with surface contours, also the drainage area, within the dash and dot lines. The limits of this area are influenced not only by the surface contours but also markedly by the areas served by existing combined sewers and drains not shown on the plan. The drains are to be designed in general with the crown at a depth of at least 5 ft. below the surface of the street. The minimum size of drain is to be 12 in. The assumed minimum allowable velocity is 3 ft. per second when flowing full. For determining the capacity of drains, use a value of n = 0.015 in the Kutter formula for sizes of 24 in. and under and n = 0.013 for sizes over 24 in. ; it being assumed that the smaller sizes will be of vitrified pipe and s:5.o \ ■ \ \ \ \ \< t-v \ ^ ?'-> "^ id f^' 52i i^ i=,-i h-2. ^_^ — ■ ""^^ t ^ ?__ a4.5 |4.0 (J ^3.5 S. o S1.5 ^0.5 o A ^5 10 15 W Z5 30 35 40 45 50 55 60 65 70 75 80 85 Time in Minutes(t^ Fig. 30. — Rainfall and run-off curve (problem in combined sewer design). the larger sizes of concrete with smooth interior surfaces. If other materials are to be used, the value of n should be fixed accordingly. Solution of Problem. — Having determined the rainfall intensity curve, the probable future percentage of impervious surface and the corresponding run-off coefficient (see "Hypothesis"), a run-off curve is constructed as previously described from which the values of Ci can be easily read (Fig. 30). The following steps must now be taken in order to reach a rational solution. First. — Draw a Una to represent the drain in each street or alley to be drained. Place an arrow near each drain to show the direction of flow in it. The drains should, in general, slope with the street surface. However, it will usually prove to be more economical to so lay out the system that the water will reach the main drain by the most direct route. Also it will, in general, prove best to concentrate the flow from small areas as quickly as possible into one drain. QUANTITY OF STORM WATER 77 In some localities with a large percentage of pervious surface, the roof water is allowed to discharge upon the ground, thence flowing to the gutter inlets and the drains. In some such instances drains are provided only to the last gutter inlet, rather than to a point opposite the last house lot, thereby effecting some saving in cost. This practice is not recommended, however, as it does not give equal privilege to all property and is, therefore, undesirable. In the district under consideration it is intended to provide drainage facilities for all property within the area designated. Second. — ^Locate the manholes tentatively, giving to each an identification number. There should be a manhole at each bend or angle, at all junctions of drains and at all points of change in size or slope, and at intermediate points where the distance exceeds 400 ft., except where a good velocity will be available during practically all conditions of flow, in which case the interval may be increased to 700 or 800 ft. In any circumstance, sufficient man- holes should be built to allow ready access for inspection and cleaning; later when the profiles are drawn and the final slopes fixed, it may be found desirable to change the location of some manholes in order to have the drains at the most advantageous depth, particularly where the slope of the street surface is not substantially uniform. Other considerations, such as obstacles under the street, may require the installation of additional man- holes, due to change in alignment or special forms of construction involved in junctions or connections with other drains. Third. — Sketch the limits of the tributary drainage areas at each manhole. The assumed character of future development and the topography will determine the proper limits. Fourth. — Measure each individual area by planimeter or other method which will give equally satisfactory results. Fifth. — Prepare a table in which to record the data and steps in the com- putations for each section of drain between manholes (see Table 14). Column 1 identifies the lines in the table. Columns 2 and 3 contain the identification numbers of the manholes. Column 4, the street name in which the drain is located; Column 5, the length in feet of the drain between manholes. Column 6, the area increment tributary to the section of drain under consideration. Column 7, the summation of the area increments given in column 6, and is the cumulative area from which the run-off in the drain is computed. Column 8, the time in minutes elapsed to the upper end of the section of drain. It is equal to the inlet time or time-of-entrance plus the time re- quired for water to flow in the drains from the upper end to the point under consideration. Column 9, the time increment or time required for passage of water through each individual stretch of drain. It is equal to the length divided by the velocity in feet per minute and cannot be computed until after the size and slope of the drain have been determined. Column 10, the rate of run-off in cubic feet per second per acre. It is taken from the run-off curve, Fig. 30, for the duration corresponding to the elapsed time given in column 8. _ Column 11, the maximum rate of riin-off for the section of drain. It is 78 SEWERAGE AND SEWAGE DISPOSAL the product of the total tributary area, column 7, by the rate of run-off, column 10. Column 12, the slope of drain, feet of fall per foot of length. Column 13, the velocity in feet per second for the drain when flowing full. Column 14, the internal diameter of the drain, in inches. Column 15, the capacity in cubic feet per second corresponding to the slope given in colunm 12. It should at least equal the maximum rate of run-off shown in column 11. It may exceed it because of the limitation as to minimum velocity, minimum sewer diameter adopted, etc. Column 16, the elevation of ground surface at the upper end of the section. Column 17, the fall in feet between manholes. It is equal to the product of the length shown in column 6 and the slope, in column 12. Columns 18 and 19, the invert elevations of the drain at the upper and lower manholes respectively. As the velocities are low in this drain no special provision has been made for velocity head. The head required is negligible in amount and is in a measure allowed for in the design, since the required capacity of each sec- tion of drain is computed on the basis of the total area tributary at its lower end, though at its upper end the flow is less than this, and consequently there will not be required quite so steep a slope as that for which the section is designed. Table 14. — Method of Tabulation of Data and Com- Line Manhole Location, street Length, feet Area in acres Time elapsed in minutes Run-off Ci from curve cubic feet per second per acre From To Incre- ment Total To upper end of section In sec- tion 3 4 5 6 7 8 9 = col. 5 ^ (col. 13 X 60) 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Dawes Dawes 300 300 300 165 325 400 35 230 240 110 95 295 260 145 225 380 165 2 28 2.40 2.15 1.54 2.16 3.11 6.02 10.14 5.65 11.91 11.02 5.60 19.73 11.30 2.88 13.07 4.16 2.28 4.68 6.83 8.37 10.53 13.64 19.66 29.80 35.46 47.36 58.38 63.98 83.71 95.01 97.89 110.96 115.12 20.0 21.7 23.4 25.1 26.0 27.6 29.7 29.9 31.0 32.0 32.5 32.9 34.0 35.0 35.6 36.4 37.7 1.7 1.7 1.7 0.9 1.6 2.1 0.2 1.1 1.0 0.5 0.4 1.1 1.0 0.6 0.8 1.3 1.15 1.10 1.05 1.00 0.99 0.94 0.90 0.90 0.88 86 0.85 0.84 0.82 0.81 0.80 0.79 0.77 Dawes Andrew Burr Burr Burr Burr Oak Grove Ave. Oak Grove Ave. Shattuok Shattuck Shattuck Shattuck Shattuck Right-of-way . . . QUANTITY OF STORM WATER 79 The student may fill in the data of columns 2, 3, 4, 5, 6, 7 and 16, also the first line of columns 8, 10 and 11, from the principles and data given here- tofore in this chapter. Before proceeding further with the complete solution the student must study Chapter IV on "Hydraulics of Sewers." For report or preliminary purposes, a profile of the main drain should be prepared showing the street surface, the drain, the manholes and the invert elevations at each manhole given in figures, also the names of streets and ave- nues where they intersect the fine of the drain. It is advisable also to give the "station" of each manhole or the distance in feet between manholes. Each lateral should be investigated to make certain that its elevation is such that when the main drain is running full, the carrying capacity of the lateral should not be reduced. Each lateral is designed with reference to the requirements of the "hypoth- eses" and economy of its construction. Therefore, its design must be reviewed subsequently, in connection with that of the main drain, to make certain that it will, in fact, have the free discharge into the main drain presupposed in its design and contemplated in the "hypotheses" under conditions of maximum flow. If the assumed conditions of flow in the lateral cannot be realized under the designs, it will be necessary either to redesign the lateral or drop the main drain to an elevation which shall permit this realization. PUTATIONS FOR StOBM DbAINS AND COMBINED SeWERS Maximum rate of Slope of drain. Velo- Size internal Capa- city, Eleva- tion of Elevation of in- vert at run-off, decimal of a foot per foot city, feet cubic feet per second ground Fall cubic feet per second per second diameter in inches surface at upper end in feet Upper end Lower end 11 = (cols. 7 X 10) 12 13 14 15 16 17 = (cols. SX 12) 18 19 2.6 0. 0049 3.0 15 3.7 206.2 1.47 199.85 198.38 5.1 0.0037 3.0 18 5.3 206.6 1.11 198.13 197.02 7.2 0. 0028 3.0 22 7.9 206.6 0.84 196. 69 195.85 8.4 0.0031 3.2 22 8.4 207.1 0.51 195.85 195. 34 10.4 0.0029 3.3 24 10.4 207.4 0.94 195.17 194. 23 12,8 0.0017 3.2 27 12.8 206.1 0.68 193.98 193. 30 17.7 0.0018 3.6 30 17.7 203.2 0.06 193.05 192. 99 •26.8 0.0011 3.3 39 27.0 203.2 0.24 192.24 192.00 31.2 0.0014 3.8 39 31.3 201.9 0.34 192. 00 191.66 40.7 0.0011 3.7 45 40.7 201.6 0.12 191.16 191.04 49.6 0.0012 4.0 48 49.6 202.1 0.11 190. 79 190. 68 53.8 0.0014 4.4 48 55.0 202.8 0.41 190. 68 190.27 68.6 ■0.0012 4.4 54 70.0 203.2 0.31 189.77 189.46 77.0 0.0011 4.3 67 78.0 203.6 0.16 189.21 189.05 78.3 0.0012 4.5 57 78.5 203.7 0.26 189.05 188. 79 87.6 0.0014 5.0 57 88.0 202.3 0.54 188. 79 188.25 88 fi 0.0015 5.0 57 89.0 199.0 196.0 lower end 0.24 188.25 188.01 80 SEWERAGE AND SEWAGE DISPOSAL In Table 14 is indicated a concise and desirable method of recording the data and elements of design of the main drain and laterals. The compu- tations are not complete but they are carried through the line of the main drain to indicate the procedure. The difficulties encountered in harmon- izing the laterals with the main drain will appear to the student, as the solution develops. A characteristic variation in practical design is indicated in Problem 2, through the solution of which there may be developed by the student the approximate difference in cost resulting from two different assumptions as to direction of flow in a short section of one of the laterals. 10 20 30 40 50' 60 7.0 80 90 100 110 120 nO MO 150 160 170 \60 190 ZOO Values of Q - Cu.-fr. per sec. Q= Cu.iipersec.Reaching Sewers. j= Rainfall in cu.-ft.persecper Acre A = Drainage Area in Acres. Practically = Inch per Hour. C= Constant ^ = ^'"P^ ( Feet per 1.000) Fig. 31. — Comparison of run-off formulas for slope, S, of 10 ft. in 1,000 ft., and small areas. Problem 2. — With the same hypotheses as to minimum depth, size, velocity, rainfall, run-off, etc., given for the preceding problem, design the laterals in Edgewood, Sycamore, Acorn and Center St., and in College Ave. between Acorn and Shattuck Sts., with the flow in the direction shown in Fig. 29. Prepare a profile of each drain showing the street surface, the manholes and the drain, and mark in figures the elevation of the invert at each manhole. Assuming a width of trench 2 ft. greater than the inside diameter of the drain, compute the quantity of pavement and earth to be removed and replaced. With the following unit prices, estimate the cost of construction of the above drains and of the sections in Oak Grove Ave. between Center and Shattuck Sts. and in Shattuck St. between Oak Grove and College Ave. Earth excavation, ft.-8 ft. deep at $2.00 per cubic yard. QUANTITY OF STORM WATER 81 Earth excavation below 8 ft. deep at $4.00 per cubic yard. Drains (including cost of laying) at $1.50 per linear foot for each foot of diameter. Manholes $10 per foot of depth, assuming the depth to be to the invert of the drain for sizes under 48 in. in diameter and to springing-line for sizes 48 in. and over in diameter. Pavement (including removal and renewal) at |2 per square yard. Note that these assumptions have been condensed for convenience and to save time in solution. No added principles would be illustrated in more detailed figures. 50a looa 1500 2000 Z500 3000 3500 Values of Q - Cu.fi-. per sec. 4000 4500 500 ttti nm I M l»'>t« HMM III iii in t- M tt m 1000 900 800 700 »*> m t»>»f I iMi» '* 600 500 400 300 200 100 100 200 300 400 500 600 700 &00 900 Drainage Area, S & & 5/ ope / in 5/op< I ■ I I I I I I L I J I I I J — I — 1 I ' ' I I I ' ' ' I I ' 1 1 I I I » S> C3 Ci S ^ Ci !=, Cb s I I I 1 I I I §S ?> 2> & ^ 111^.^. mil I I I Ss ^ s> "=> 1^^ R ^ ^ ^ h 1 1 I I e //7 reef § ^ «vi K^ s-s^^'^^oo^JNoj^vs^QvCi I ■ I ■ I ■ I I I I I 1 I I I I I I I I ri I — I r I I I I I . I . I , I I I I I , III., i ^5 55 5 I. , , . I . , tVg I II I I I I I I •IS I ^' iJ_L. PJ in I . . I.I . I I i '<^ ^ ^ vte f-< Ci <5i Q _L .■■■.... , .7.717. r. I.I.I. I. I . I ■■. ■■■ I I.. ■ . I. 1,1 I , I , , I I l l I i.Mii . ■ . . M.I.I Si .a ^ <5i gj s,,^,,,.^,^.^,^:§,^,^,^,^.^,^|,,.|....^.^.^,^.^li^^ Trn"v;TTT] j:0 ■ I V' Q .6-^1 ^i. '0 r. <.','!" r+ 100.000 -- 500.000- ■400,000 300,000- ■zao.OQo- _I000 '-700 ■5D0 ■400 -300 -500 -400 200 '-300 ■loo.ooo- ■70.000 -[-J oa 70 -50,000- -40,000- -30,000- _^ -zo.ooo- -^o -10.000 ■7.000- -la :-7. -5.000- , 000,000- ■ ^ MOO T— 4,000- -3,000 rrtOOO -1000 -700 ■zIOQ \70 -50 4€ -30 u I -10. -^ V-4: 3. -7. -^ -5. -4. -3. f-50, dmde thousand bi/ la. Fia. 49.— Discharge of egg-shapod sewers running full depth, n = 0.015. (John H. Gregory.) {Insert Following F-i(]. 48.) HYDRAULICS OF SEWERS 103 studies are warranted and less reliance should be placed upon coefficients and the arbitrary selection of sizes from the diagrams ordinarily used. When n is taken as 0.013 in Kutter's formula, c in the Hazen- Williams formula may be taken as 120; when n is taken as 0.015, c may be taken as 100. Kutter's formula is generally used with the aid of diagrams giving not only the velocity but also the discharge. Figs. 46, 47 and 48 are examples of such diagrams prepared by the authors and Fig. 49 is one of an admirable set of hydraulic l.U ■" n =: ■s. — \ — — — — r- T- EZ" srr -p PI DOl s t ^v s ^ - g0.0 \ / ~ i _J_ \ y / tn _ __ \ > h? y ■" '.^07 _ \ / I > ( ^ * » / y y / o 06 \ / / y y i5 \ /' > // ' ■" ■Po'- £ 3 1/ / /' '1 /^ ~ 04 < y i > r - ■6 "•' /' •• / r»« P. ~ toi / / 1 »» ^ -> s y ,1 H' ^> <<' '^ ■~ 'fe oz / / / / ,^ > ^ y Nl< /, /' /■ ^', ,-- £ 1 / / ^ ^. // --- ,-9 ^ on L d g " 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 :.3 ftati'o of Hydraulic Elements of Filled Segment tothoseof Entire Section. Fig. 50. — Hydraulic elements of circular section. n = 0.015; s = 0.005; Z) = 1 ft. Area = 0.7850'; Wetted Perimeter = 3.1416Z>; Hydraulic Radius = 0.250D. diagrams prepared by Prof. John H. Gregory. ^ The discharge is equal to the product of the mean velocity by the cross-section of the flowing stream. In making the preliminary studies of a system of sewers, tables of the discharge of sewers laid on a grade of 1 per cent (such as shown in Tables 20 and 21) coupled with the use of a slide rule, keeping in mind the principle that velocities and discharges vary about as the square roots of the grades, have been used in the past, particularly in Europe, and may be devised in the field in emergency. The modern method, and the one followed by most engineers, however, is to make use of 1 Other examples of the Gregory diagrams are given in "American Sewerage Practice," vol. i, p. 94, and in the published set of the diagrams, which may be obtained from Piof. Gregory, The John Hopkins Univ., Homewood, Baltimore, Md. Table 20. — Velocity in Feet per Second (F) and Discharge in Cubic Feet per Second (Q) op Circular Sewers with Different Proportions (//) of the Diameter (D, Inches) Filled AND A Grade of 1 .100 (Based on n = 0.013 in Kutter'a Formula) D 4-iQ. e-in. 8-in. lO-in. 12-in. IS-in. 18-in. 20-iii. 22-in. 24-m. 30-in. 33-in. 36-in. 42-m. 48-m. 60-in. 1 H V Q V Q V Q Y Q V Q F Q F Q F Q F Q F Q F Q F Q F Q F Q F Q \ V Q 0.1 0.65 0.0052 0.88 0.015 1.15 0.036 1.36 0.068 1.68 0.113 1.85 0.189 1.93 0.311 2.28 0.416 2.45 0.52S 2.61 0.682 3.06 1.249 3.26 1.615 3.46 2.04 3.85 3.09 4.22 4.42 1 4.90 8.02 0.2 1.14 0.0142 1.41 0.045 1.79 0.099 2.11 0.185 2.61 0.306 2.87 0.567 3.28 0.934 3.55 1.247 3.81 1.574 4.05 2.045 4.75 3.74 5.08 4.84 5.39 6.13 6.00 9.27 6.57 13.66 7.61 24.1 0.3 1.30 0.030 1.82 0.094 2.30 0.201 2.71 0.391 3.35 0.647 3.69 1.196 4,22 1.97 4.56 2.63 4.90 3.41 5.21 4.34 6.11 7.81 6.53 11.08 6.93 12.96 7.71 19.6 8.44 18.00 9.81 50.9 0.4 1.59 0.053 2.22 0.163 2.80 0.369 3.32 0.686 4.10 1.135 4.51 2.10 5.16 3.43 5.58 4.62 5.88 5.98 6.37 7.58 7.43 13.89 7.98 17.95 8.48 22.7 9.43 34.3 10.32 39.1 11.99 89.2 0.5 1.811 0.0790 2.621 0.2475 3.168 0.553 3.770 1.028 4.335 1.702 5.131 3.148 5.872 5.190 6.343 6.92 6.797 8.97 7.237 11.365 8.481 20.81 9.067 26.92 9.63 34.04 10.71 51.50 11.73 73.70 13.62 133.75 0.6 1.95 0.106 2.72 0.332 3.44 0.741 4.07 1.378 5.03 2.28 5.54 4.23 6.32 7.96 6.85 9.28 7.32 12.02 7.82 15.22 9.16 27.9 9.80 36.1 10.40 45.6 11.56 69.0 12.67 98.7 14.71 179.2 0.7 2.06 0.133 2.87 0.411 3.64 0.966 4.30 1.728 5.31 2.86 5.82 5.29 6.68 8.62 7.23 11.63 7.75 15.07 8.25 19.10 9.57 35.0 10.34 45.2 10.98 57.2 12.21 96.4 13.37 123.8 15.52 224.5 0.8 2.10 0.155 2.92 0.485 3.70 1.083 4.37 2.03 5.41 3.33 5.95 6.17 6.79 10.18 7.35 13.58 7.86 17.58 8.40 22.4 9.84 40.7 10.52 52.8 11.17 66.7 12.42 100.9 13.61 144.4 15.80 262.2 0.9 2.08 0.170 2.90 0.530 3.67 1.188 4.34 2.21 4.99 3.65 5.90 6.75 ■?,74 11.12 7.29 14.86 7.80 19.26 8.33 24.4 9.78 44.7 10.43 57.8 11.07 73.1 12.32 110.6 13.50 158.2 15.67 287.0 1.0 1.811 0.1580 2.521 0.4950 3.168 1.106 3.770 .. J 2.057 4.335 3.405 5.131 6.296 .j;872 10.38 6.343 13.85 6.797 17.94 7.237 22.73 8.481 41.63 9.067 53.85 9.63 68.09 10.71 103.00 11.73 147.40 13.62 267.5 Table 21. — Velocity in Feet per Second (F) and Discharge in Cubic Feet sr Second (Q) of Egg-shaped Sewers op the Type Shown in Fig. 49, with Different Proportions {H) of the Height \ Filled and a Grade op 1: 100 (Based on n = 0.013 in Kutter's Formula) D 24/16-m. 27/18-in. 30/20-in. 33/22-m. 36/24-in. 39/26-m. 42/28-in. 45/30-m. 48/32-m. 51/34-m. 54/36-in. 57/38-m. 60/40-in. 63/42-in. 66/44-in. 69/46-m. H F Q V Q V Q F Q F Q F Q V Q F Q F Q 1 V Q F Q F Q F Q F Q V Q F Q 0.1 0.2 0.3 2.4 3.6 4.4 0.2 0.8 1.8 2.6 3.9 4.8 0.3 1.2 2.5 2.9 4.2 5.2 0.4 1.6 3.3 3.1 4.6 5.6 0.5 2.0 4.3 3.3 4.8 6.0 0.7 2.6 5.5 3.4 5.1 6.3 0.9 3.2 6.8 3.6 5.4 6.6 1.1 3.9 8.3 3.8 5.7 7.0 1.3 4.7 10.0 4.0 6.9 7.3 1.6 5.6 11.9 4.2 6.2 7.6 1.9 6.6 14.1 4.4 6.4 8.0 2.2 7.6 16.3 4.5 6.7 8.2 2.5 8.8 19.0 4.7 6.9 8.5 2.9 10.1 21.8 4.8 7.2 8.8 3.1 11.6 24.8 5.0 7.4 9.1 3.7 13.1 28.1 5.1 7.6 9.4 4.2 14.8 31.6 0.33 4.59 2.32 5.01 3.20 5.43 4.28 5.82 5.56 6.21 7.06 6.59 8.78 6.97 10.78 7.32 13.0 7.66 15.47 8.00 18.24 8.35 21.35 , 8.67 24.69 8.98 28.35 9.29 32.32 9.60 36.65 9.91 41.37 0.4 0.5 0.6 5.0 5.6 6.2 3.3 5.1 7.0 5.5 6.1 6.8 4.5 7.0 9.6 5.9 6.4 7.3 6.0 9.3 12.9 6.3 7.1 7.8 7.8 12.1 16.7 6.7 7.5 8.3 9.8 15.3 21.2 7.1 8.0 8.8 12.2 19.0 26.3 7.5 8.4 9.3 15.0 23.3 32.4 7.9 8.8 9.8 18.0 28.1 38.8 8.3 9.2 10.2 21.5 33.4 46.6 8.6 9.6 10.7 25.3 39.3 54.3 9.0 10.0 11.1 29.5 45.8 63.2 9.3 10.4 11.5 33.9 53.1 73.2 9.7 10.8 11.9 39.1 60.9 84.1 10.0 11.2 12.3 44.7 69.5 96.0 10.3 11.6 12.17 50.6 78.7 108.8 10.6 11.9 13.1 57.0 88.6 122.6 0.67 6.3 8.46 6.87 11.68 7.42 15.69 7.94 20.17 8.46 25.57 8.94 31.72 9.41 38.74 9.90 46.75 10.35 55.62 10.78 65.44 11.21 76.28 11.65 88.29 12.06 101.27 12.46 115.34 12.86 130.70 13.25 147.21 0.7 0.8 0.9 1.0 6.4 6.5 6.5 5.92 9.1 11.1 12.7 12.08 7.0 7.1 7.1 6.44 12.5 15.3 17.6 16.65 7.5 7.7 7.7 6.97 16.7 20.5 23.4 22.25 8.1 8.3 8.3 7.47 21.6 26.5 30.3 28.83 8.6 8.8 8.8 7.94 27.4 33.6 38.3 36.48 9.1 9.3 9.3 8.41 34.0 41.7 47.6 45.36 9.6 |9.8 i9.8 8.86 41.6 51.0 58.2 65.43 10.0 10.3 10.3 9.31 50.1 61.5 70.1 66.80 10.5 10.8 10.8 9.74 59.6 73.2 83.5 79.55 11.1 11.3 11.3 10.16 70.3 86.3 98.4 93.71 11.4 11.7 11.7 10.56 81.8 100.5 114.6 109.22 11.8 12.2 12.2 10.97 94.8 116.3 132.7 126.41 12.3 12.6 12.6 11.37 108.8 133.5 152.4 145.16 12.7 13.1 13.1 11.76 124.1 152.2 173.6 165.45 13.1 13.5 13.5 12.14 140.6 172.5 196.6 187.45 13.5 13.9 13.9 12.51 168.3 194.1 221.6 211.08 {Jnsert Facing Page 104 J HYDRAULICS OF SEWERS 105 as the square root of the head under velocities corresponding to the ordinary conditions of flow, it varies more nearly directly as the head under extremely low velocities. Within the ordinary 1.0 ^ 0.9 ■ a> E 0.8 1 0.7 ^0.6 '■P 0.5 o u- 0.4 • »*- o a. V S 0.Z o 10.1 IK 0.0.K ^^•. ^ ~~:''^^ ^ ^v L i LL^ ir >v T n '- ■ % ^\ L /t&^ " "^ X '-^A^^ '7 ^a: ^ -.V^i ]' -%^ V vft^ U ■^A^ \ 1^2 r y \ ^^ L J ^-1 -^^ r .^^ ■■ ~1 ^'^ ^ *Sv ^ X'' ! K'^ -^^upfo_^2 ^r^^s '^^^^'^ -> ^.^f'4^^ J?,/ ^^1^^^^ ^zz i^t^^ /l^X ^^J" ^ 7^^ 4 ---"L r 0.0 O.l 0,2 0.3 0,4 0,5 0.6 0.7 0.B 0.9 1.0 I.I 1.2 Ratio of Hydraulic Elemetife of Filled Segmentlothoseof Entire Section.' Fig. 53. — Hydraulic elements of horseshoe section, Wachusett type. 0.013; s = 0.003; D = 7 ft.; Horizontal diameter, H = ]_ ft. 8 in.; Area = 44.74 sq. ft. = 0.913D2; = 0.263D. Wetted perimeter = 24.26 ft. = 3.466j[); Hydraulic radius = 1.841 5: N ""■ " "■ r ■_• — ■; 'S |0.9 \ .^ >> / \ \ i? f' s J V '?(! \ ' / I 'k / t / .|0.7 '. \ V* /. i' y \ I / \>^ > „ 0.6 |0.5 \ Q / y =CJ 1 <- y ■1 y = iS' ^i I 1 / A / = 0,4 tn ! / '. ^ > : / f- T .# •«l <« 5§ ^ rt ' / !£J 61. 1 y ^0.2 ^ J< ; y ' S '', / \ t ^ io.i :;/ / / ^ 7 "I'i l»' 1 ^ £ ^ 1 \ 1 iL _ n = 0.015; s 0.0 0.1 0.2 0.3 0.4 05 0.6 0.7 0.8 0.9 1.0 I.l 1,2 Ratio of the Hydraulic Elements of the Filled Segment- to tho&e of the Entire Section. Fig. 54. — Hydraulic elements of catenary section. 1/3,000; D = 7.44; Vertical diameter = D = 1.063 diam. equiv. circle; Area = 0.70277D2; R = 0.23172D. velocity limits of from 1 to 6 ft., the formula finds its most trustworthy application. It is fairly reliable up to 10 ft. per second velocity. 106 SEWERAGE AND SEWAGE DISPOSAL For special cases which may be outside the range of the formula, such as 20 ft. per second or higher velocity, the engineer should consult the original data. n s T J \ r ^ s 1 / ]\ "< : y J <>. \ / f s / / N / / \ Q / ^ I V f1' y / V " \ Pi V ^ i^^- .-■•■ U._- . 70 fl - -J 1 ' Q 9HW- w 'y W- - C '' / -.*;i^^ % / '. / Ui ^^^S t y ^ s^ i;>:^ • / y /P '0 ■^ i^ •^^ i / / \ k. " Q T // \ -' t>' ■ s i=a g di ■- _ jk i \x 0.0 0.1 0.Z 0.3 0.4 0.5 0.6 0.7 0.9 0.9 1.0 1.1 Ratio 0? Hydraulic Elements of Filled SegraenttottioseoftntiteSection.i Fig. 56.^ — Hydraulic elements of Louisville semi-elliptic section. n = 0.013; » = 0.0003; £) = 7K ft.; Area = 0.785D2; Wetted perimeter = 3.26D; Hydraulic radius = 0.242D. Hughes and Safford have summed up^ the application of this formula in an excellent manner as follows: ■ Hydraulics, p. 343. HYDRAULICS OF SEWERS 107 "That, for hydraulic radii greater than 10 ft., or velocities higher than 10 ft. per second, or slopes flatter than 1 in 10,000, the formula should he used with great caution. For hydrauhc radii greater than 20 ft., or veloc- ities higher than 20 ft. per second, but little confidence can be placed in the results. "That, considering the variable accuracy of the data on which the formula is based, results should not be expected to be consistently accurate within less than about 5 per cent. "That, for any slope steeper than 0.001 the values of C computed for S = . 001 may be used with errors less than the probable error in the ordinary use of Kutter's formula. 1.0 — "T — X — — =— — -r- r- rs r-i r—i |0S — i— " >. s. ~ i H I N / \ "08 ) 'S s > r i j D , 1 s i/ iid, y 1 |0i7 1 1 \ \4, V T 0. 4- s / fij '>" \r ~ ^06 - -> > / > 1 _ Q> 5- '■ 1 ^^ / J 0.5 Q [ / r .1 f ij ' Cl z' \ 4 " r^OI / V I \ ^ / f 0.3 ; 1 ^. y / K^ — ^ - ^ -¥■ / /' t '' / ^ <^' ■^02 y /' / 1 y %'■" / y 1 .^ ^ i°' ^ / \ ^ ^ ^■ -^ ^ -" ■* ,0.0 u ^ _ -^ — 1 s: "^ •0.0 0.1 OZ 0.3 0.4 0.5 0.6 (XI 0.8 0.9 \0 1.1 12, Ratio of Hydraulic Elements of the Filled Segment to those of the Entire Section. Fig. 57. — Hydraulic elementa of semi-circular section, n = 0.013; s = 0.001; Z) = 9 ft. 2J^ in.; Area = 1,2697D2; Hydraulic radius = 0.2946D. "That between slopes of 0.001 and 0.004 the maximxim variation at the extreme values of n and J? in C is about 4 per cent; for such values as fall within the range of ordinary practice the maximum variation is but 2 per cent. "That between slopes of 0.0004 and 0.0002 the maximum variation is about 5 per cent, but for such values as fall within the range of ordinary practice the maximum is less than 3 per cent. " That for higher values of S the divergence in the values of C increases; but the occasions when slopes flatter than 0.0004 are to be considered in design are not common, and when they do occur they are usually for structures of such high character that they warrant special study and some basis in addi- tion to a general empirical coefficient. And considering that a degree of precision of 0.001 is rarely exceeded in leveUng for ordinary construction work, and that in picking out the value of n, a variation of 0.001 for small values of n and R may change the value of C as much as 17 per cent, and for 108 SEWERAGE AND SEWAGE DISPOSAL moderate values as much as 5 to 8 per cent, it should be obvious that hair- splitting calculations with the Kutter formula are a needless waste of time, producing merely mechanical accuracy instead of a high degree of precision." Flow in Partly Filled Sections. — In many of the problems arising in sewerage design it is necessary to know the velocity and discharge when a sewer is partly filled. This information is obtained in actual practice from diagrams Uke those of Figs. 50 to 57 inclusive. They are usually prepared by plotting the cross-section carefully and then measuring with a planimeter the area of the stream when the section is filled to various depths. These areas divided by the wetted perimeter, also carefully measured, give the mean hydraulic radius. A slope and a co- efficient of roughness, n, are then assumed and the velocities and discharges for the different depths are computed and plotted in terms of the velocity and discharge when flowing full. The curves thus obtained show the relative velocities and discharges at different depths as compared with those of the full section. These curves, although strictly correct only for the data given, are sufficiently close for other sizes, slopes and friction factors to be of general use, and as a rule the difference may be neglected. Attention is particularly called to the egg-shaped section. Fig. 51. This was introduced about 1846 by John Phillips in order to obtain adequate velocities in combined sewers when the depth of sewage in them was small. When the quantity of sewage is not more than two-tenths of the total capacity of the sewer, the velocity in an egg-shaped sewer will be somewhat greater than in a circular sewer of the same total capacity. The depth of the sewage will also be greater and there is a result- ing better flotation for solids. These differences are not of great magnitude, however. Discharge and Velocity at Different Depths.— The proportion- ate velocity and discharge at different depths are shown in Figs. 50 to 57 inclusive. For a 12-in. sewer, the capacity and velocity when running full are 2 cu. ft. per second and 2.5 ft. per second respectively, the slope being 0.005 and n being 0.015. When the depth of flow is 0.1 ft., the discharge will be but 0.06 cu. ft. per second or 3 per cent of its capacity when running full. At that depth, however, the velocity of flow will be 33 per cent of that of the full sewer, or 0.84 ft. per second. At a depth of 0.2 ft., the discharge will still be small, but the velocity will have increased substantially, to 1.38 ft., 55 per cent of the velocity HYDRAULICS OF SEWERS 109 when the sewer is full. In other words, the velocity increases much more rapidly with increasing depth than does the discharge, as shown in detail in Table 22. Table 22. — Actual and Proportionate Dischakges and Velocities FOB A 12-iN. Sewer on a Slope of 0.005 with n Taken as 0. 015 Proportionate depth Actual Proportionate Capacity, cubic feet per second Velocity, feet per second Capacity Velocity 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8' 0.9 0.932 1.0 0.04 0.16 0.36 0.64 1.00 1.34 1.70 2.00 2.16 2.18 2.00 0.84 1.38 1.85 2.23 2.50 2.70 2.85 2.90 2.85 2.77 2.50 0.02 0.08 0.18 0.32 0.50 0.67 0.85 1.00 1.08 1.09 1.00 0.33 0.55 0.74 0.89 1.00 1.08 1.14 1.16 1.12 1.12 1.00 ' Point of maximum velocity. 2 Point of maximum discharge or capacity. This proportionately greater increase in the velocity than in the discharge is an important condition of which the designing en- gineer should take advantage. He should also keep in mind that a circular sewer will have the same velocity when running half full as when full, that the maximum velocity is generally reached when the sewer is between 0.8 and 0.9 full, and that the maximum capacity is generally reached when the sewer is between 0.9 and 0.95 full. The velocity ordinarily used as a basis of design is that of the sewer when half full or full. The maximum velocity, however, may have an important influence in washing out deposits which form at times of very low flow. Selection of Sewer Sections. — The problem of the design of masonry sewers is not solved with the determination of the required carrying capacity, but incWdes a number of other features which may be of considerable importance. The most economical shape for the waterway cross-section can 110 SEWERAGE AND SEWAGE DISPOSAL only be selected after careful consideration of the special condi- tions imposed and the relative merits of one type as against another to meet these special conditions. While the circular cross-section has been used for a large number of the inasonry sewers constructed in this country, Figs. 58, 59 and 60, there has been an increasing use of other forms such as the horseshoe, Fig. 61, semi-elliptical, Fig. 62, and rectangular sections, Fig. 52. In the older combined sewerage systems constructed previous to 1890, and built for the most part of brick for sizes above 24 in. in diameter, the egg-shaped cross-section. Figs. 68 and 63, was fre- quently used, but since that time the extended use of separate systems has caused it to decrease in popularity. The old Massachusetts North Metropolitap System was a departure Cireular Sewer, Egg-5haped Sewer. Fig. 58. — Standard plain concrete section (Borough of the Bronx, New York) . from the practice of the time in that it included such types as the Gothic, catenary. Fig. 64, and basket-handle sections.' The general adoption of concrete for masonry sewers has brought about a more extended preference for the flatter types of inverts on account of their being more easily constructed than the inverts of circular or egg-shaped sections. Aside from the hydrauhc properties, such considerations as the method of construction, character of foundation, available space and stability may be instrumental in determining the best type of sewer section to adopt for a given case. The selection of the proper thickness of masonry for a given size of sewer, unless determined in the light of experience with 1 Chapter XII, vol. i, "American Sewerage Practice" contains a large number of illuB- trationa and descriptive notes of different types of sewer sections, some of which are shown in Figs. 65 and 66. HYDRAULICS OF SEWERS 111 Sand&Gravel . Backfilling . '•lA'f (aster Underdrain /4^ ■ Scresned Grv(vsl Yellow Clay -a- Onderdraln- T -b- M assacV-iuser-ts. fFhster-._iJ^ t:i Plaster^ -d- PhlladelpHia. {brick Arch: 14-r Bricks '/^. 1?!^.- ■v ^ /'^b.l ! \ ^V ;3. i'7''\ ,' \ > 'A ..^'^ 'i-./ \jib'\\ Vj ". " ^x^ r-^i.\ / ■ b". 0- hv ^ini ^t- -e- T r u r O.I Embank- '?[ ,, 6 foff£/Sfe ment^ Concrete- Rock -f- Phi ladelphia . Fig. 59.-*-Typical circular sections with brick arches. 112 SEWERAGE AND SEWAGE DISPOSAL flgnliing'' -a- '^ "Underdniitt Brooklyn, *\ K-UndelB- W°'"°%-,A Queens. Wilmington. Fig. 60. — Typical circular sections. HYDRAULICS OF SEWERS 113 ^och. -a- 'Wachusett AquedutH T^^^^^.5 Plaster B alti more. -e- -b- Hartford Aqueduct. -d- Wa+crbury, * :: ■J'? ■■.-.■> ^ f--.-— •.■/ 7 a~ v, L 5? D o & -h O M Fig. 61. — Typical horseshoe sections 114 SEWERAGE AND SEWAGE DISPOSAL -C- Cdtskill Aqueduci- • 2'J*'-- -/5'6"- - ^ -d- Phi la de 1 phio. -f- Uo u i s V i I \ e Fio. 62. — Typical semi-elliptical sections. HYDRAULICS OF SEWERS 115 P_ |-ii ladelphta. FiQ. 63. — Typical egg-shaped sections. 116 SEWERAGE AND SEWAGE DISPOSAL ■ 5'4~ -a- Wortester. :4'Plaster Q8"Dram -b- Massachuselti; ) '-Portland -c- ' Indurated Clay Concrete Mc. ^ o t^ O O) rH "o b-h-h-OMCOfNO) ^ fe 51 ■^ CO a? 00 •^•^ CO o o o »c >c _^ - lo o »o (D ira Tj 1^ P rH (N « OS fl 1 o o o o o o . OJ OOONrfUlOiO ° s CO (N W CM M o o oooooo 1^ -tJ-fJ-t^OOOOO ojincoddddd •rf< (M (N o o o -a o o o 1J <3 odd o 1=1 d ■<*< CO CD CO CM N 1 O O OOO'^OCD U M ^J-tJ-tJCOC^iHTHi-H fe§ O O 00 s Ph-" CO (N >-< 'o a o ^ C» CO ;3 ,o 00 CD t- 'oa Q) o o o > a — • • ■ O lO N W (M ° "S 000'*TjiOO(M •3| m 3 -, _ i-H T-H (N - CO o i-H 'd O d OO i-H t* to M C O (N b- t* OS O ^ W rH ^ r-( a ■M o o o o o fl 12 d d d d d 03 CO a o O d ■ -^ • •i - 1 ^ t^ h-t a ■s o bo fl £ ;=] >, o o o ^ go g :=3 1 a ilia Cond xir Creek baugh Ca Canal, on Comp; Umat Sulph Riden North gati -^ N CO ^' HYDRAULICS OF SEWERS 123 encountered. Thus, if the length of a certain section of concrete channel was composed of 25 per cent curves and 75 per cent tangent, a value of n = (0.25 X 0.018) + (0.75 X 0.013) = 0.014, was used in that section. It is to be hoped that at some time in the near future some experimenter may investigate and measure such losses in order to determine a more rational and accurate basis for safe allowance for them. Velocity Head. — The absorption of head in the creation or increase of velocity and the possibility of its partial or complete recovery by proper design, are hydraulic factors which are often negligible under low velocities but are of serious importance if the velocities are relatively high. Fall must be provided not only for losses due to friction, bends, etc., but also for "velocity head," so called. This head,expressedbythetermt;^/23increases very rapidly, by the square of the velocity, being 0.25 ft. for 4 ft. per second velocity, 1 ft. for 8 ft. velocity and 4 ft. for 16 ft. velocity. At velocities normally found in sewers in flat terri- tory, the velocity heads may be inconsiderable, but this hydraulic element should be considered in design, and rejected only if relatively unimportant in the problem under analysis, as failure to provide for it has frequently led to serious loss of capacity, unexpected choking and overflow of sewers, and other inconvenience. Some cases which require special attention in this particular may be cited. A large sewer with slow velocity changing for any reason to a smaller sewer with high velocity ; grit chambers, where sewage moving slowly through the enlarged chambers must increase its velocity to flow away through an outlet passage or conduit of smaller dimensions ; effluent conduits or channels leading away from sedimentation tanks; restrictions in conduits, such as gates or contractions necessary to pass other structures; in short, any cases where velocity must be built up to initiate flow in a channel or conduit. Recovery of Velocity Head. — When velocity is reduced by enlargement of a channel or conduit a considerable proportion of the change in kinetic energy may be recovered as potential energy, manifested by a rise in water surface, provided the change in section is smooth and gradual. The U. S. Reclamation Service has observed the extent of this recovery in numerous instances, see Table 23&, where irrigation canals have been contracted in section and then enlarged again further downstream. 124 SEWERAGE AND SEWAGE DISPOSAL The principle of this recovery of velocity head is well illustrated in an analogous problem, that of water flowing under pressure in the Venturi meter. In it the outlet cone is gradually expanded so as to recover practically the entire difference in velocity head between throat and outlet. Table 235. — Recovery of Velocity Head in Transition Sections FROM Experiments by U. S. Reclamation Service Location and name Velocity, feet per second Up- stream end Down- stream end Percentage of velocity head recovered Discharge, cubio feet per second North Platte Project < Hope Creek Flume Uncompahgre Valley Project, Happy Canyon Flume* Tieton Main Canal Tunnel Outlets: 1. Trail Creek 2. Columnar 3. Tieton 4. North Fork Okanogan Project: Experiment Number IV Experiment Number VI 5.79 1.64 5.87 1.70 5.67 1.65 6.47 2.85 12.12 8.06 11.42 7.98 11.58 7.60 10.25 2.62 5.39 2.61 3.31 2.46 54 25 28 100 nearly 84 60 68 100 50 1 62 / 160 105 99 361 271 261 286 286 1 Transition unusually favorable for recovery. Problems The following problems may be solved to test the students' grasp of the fundamental principles of this chapter before proceeding to the application of these principles in problems such as those worked out in detail in Chapters II and III. 1. A sewer 800 ft. long with a fall of 2.4 ft. must discharge 10 cu. ft. per second when flowing full. What size of circular vitrified-pipe sewer is required? What will be the velocity of flow? What coefficient n should be used in the Kutter formula? 2. In the sewer of Problem 1, a minimum rate of flow of 0.75 cu. ft. per second is expected. Find the depth and velocity under these conditions. 3. Re-design the same sewer using an egg-shaped section flowing full and compare the depth and velocity at minimum flow with the results of Problem 2. 4. In an existing sewer system, a 96-in. concrete sewer is found to have been laid on a slope of 0.00085 and its condition is found to warrant the use of ra = 0.013. What is its capacity and the corresponding velocity of discharge in it when flowing full-depth? When flowing 0.65 depth? When flowing 0.3 depth? HYDRAULICS OF SEWERS 125 6. The sewer of Problems 1 and 2 discharges into a concrete trunk sewer (re = 0.013) 48 in. in diameter on a slope of 0.002 and which carries a, minimum flow of 9 cu. ft. per second above the junction. For topographical reasons it is impracticable to build the crowns of the two sewers coincident in order to avoid surcharging the smaller sewer when the larger is flowing full; it is, however, desired to avoid backing up in the smaller sewer at times of minimum flow in both sewers. At what elevation, above the invert of the trunk sewer, must the invert of the lateral be connected? 6. By the method involving the use of tables for slope 1 : 100, find the size of circular sewer (with n = 0.013) required to carry 32 cu. ft. per second on a slope of 0.0025 when flowing full. Find also the size of egg-shaped sewer to carry the same quantity when flowing 0.75 full. 7. A natural stream is to be replaced by a rubble-hned channel (re = 0.017) for a length of 2,750 ft., including 800 ft. of curves. Assuming that con- sideration of expected velocities and the radii of the curves leads to the decision to compensate for the extra slope required on curves by the addition of 0.004 to the value of re on the curves, determine a balanced value of re to use throughout the entire length of the proposed channel. 8. A sewer discharging at 3 ft. per second, when full, comes to an abrupt pitch such that the new velocity will be 9 ft. per second when full in a sewer of properly reduced diameter or section. What is the required drop in the hydraulic gradient to give this increased velocity? 9. Show by a sketch the relative position of the two sewer sections to meet the conditions of Problem 8 without permitting surcharge of either sewer section. 10. A series of grit chambers in which the velocity is 1 ft. per second terminate in an outlet channel in which the velocity is 6 ft. per second. The flow from the chambers must pass through gates which reduce the cross- section of flow at the point of passage to 60 per cent of that available in the grit chambers. Estimate the total loss of head, including the loss of head through the gates and that necessary to produce the increase in velocity in the outlet channel but neglecting the friction loss. 11. A series of grit chambers in which the velocity is maintained at about I ft. per second converge into a single outlet channel. For what velocity shall the outlet channel be designed so that the loss of head (drop in water surface) involved in increasing the velocity shall not exceed 0.5 ft., out of which an allowance of 0.15 ft. must be made for the losses through gates, etc. 12. A conduit flowing at a velocity of 7 ft. per second passes through a transition section into a larger section in which the velocity is 4 ft. per second. The water surface in the larger section is found to be 0.3 ft. above the hydraulic gradient, as determined by the normal slopes required for friction alone. What is the percentage of recovery of velocity head in the transition from the smaller to the larger sections? The student should now return to the problems at the end of Chapters II and III and follow through the method of solution there illustrated by which the sewer sizes corresponding to the basic data relating to quantities, etc., previously worked out, were determined. CHAPTER V SEWER APPURTENANCES AND SPECIAL STRUCTURES A variety of accessory structures were defined in Chapter I which will be described in detail in the present chapter. These structures have never been standardized and each engineer designs them to fit the local conditions, relying mainly upon judgment based upon his previous experience and a knowledge of what has proved satisfactory elsewhere. STREET INLETS AND CATCH BASINS Where storm water is removed underground, street inlets are provided to take it from the gutters to the sewers. Catch basins are employed where this street wash contains so much refuse of various kinds that it is better to give it a chance to settle in an easily cleaned basin rather than to allow everything to flow with- out check into the sewer. Street inlets should be located so that they will keep the gutters in such condition, even during a heavy rainstorm, that teams may drive close to the curb and pedestrians cross the street with minimum inconvenience. Officials responsible for the mainte- nance of gutters have found that inlets increase the difficulty of keeping the gutters in repair, because the wheels of vehicles wear away the pavement about the inlets more rapidly than where there is no break in the smooth surfaces of the street and curb. Consequently they often oppose the installation of suffi- cient inlets to remove the street wash, particularly on steep grades, and oppose the lowering of the gutter grades at inlets so as to leave slight depressions in the gutters to assist in diverting water into the inlets. The best guide in locating inlets is knowledge of the quantity of water which actually flows along gutters in a given city. In the absence of such knowledge, it is customary to assume that inlets should never be more than 300 to 350 ft. apart where gutters must be kept free from more than a small quantity of water, and in no case more than 700 ft. apart. An inlet must always be placed where two grades join to form a valley. The gutters 126 SEWER APPURTENANCES AND SPECIAL STRUCTURES 127 should be so constructed, the cross-section of the street so plan- ned, and the inlets so placed that storm water will not flow across the pavement to reach an inlet. Although it is customary in many places to locate inlets at the angle of street corners, this is a poor place for them if the travel on the street is heavy, because the wheels of trucks rounding the corner close to the curb are Plan. Section A-B. Fig. 67. — Standard inlet for park and light-traffic drives, with Concord grate (Metcalf and Eddy). particularly hard on both pavement and inlet castings in such a position. An inlet at each side of the corner, just before the cross walk is reached, is a preferable arrangement. Fig. 67 shows a type of inlet which the authors have found very satisfactory in parks and streets with light travel. Dura- bility is afforded by the substantial concrete block, and other 128 SEWERAGE AND SEWAOE DISPOSAL advantages are relatively low cost, a large grate area and ease of construction. The authors' experience with inlets having risers of straight pipe is that the gratings do not have adequate openings. Therefore a flaring mouthpiece is inserted in the end of the pipe to give a larger opening. ■..e'6-:..-A -D- FlG. -E- . — Standard street inlets, Borough of the Bronx. Fig. 68 illustrates a number of types of street inlets adopted as standard in 1913 in the Borough of the Bronx, New York. Their main dimensions are: Type A B C D E Inlet opening, inches. . Box depth, inches Box section, inches . . . 7 by 32 66 41 by 42 7 by 32 64 41 by 32 5 by 36 20 36 by 18 14 by 32 18 30 circular Sticks, waste paper and leaves cause most of the clogging of the openings of the inlets, and some engineers accordingly use openings which present hardly any obstacle to the entrance of these three classes of refuse. It is questionable, however, whether SEWER APPURTENANCES AND SPECIAL STRUCTURES 129 sticks should be allowed to enter an inlet owing to the danger of their becoming fixed in the pipe leading to the sewer. As a general rule, street inlets are better adapted for busy streets with good pavements that are kept clean, particularly where there are no steep grades nor any topographical conditions tending to concentrate the storm run-off at a few points, than they are for streets furnishing large quantities of refuse rarely removed by street cleaning and liable to have the run-off con- centrated at a few points to which many of the storms are certain to wash a large part of the street litter of every sort. Catch Basins. — The catch basin was formerly considered an absolutely essential part of any American combined sewerage or drainage system. The velocity of the sewage in many sewers had proved insufficient to prevent the deposit of sludge within the sewers, and it was more expensive to remove this sludge from sewers than from catch basins. Since that time sewers laid on self-cleansing grades have become standard practice and streets are better paved and better cleaned, so that less refuse is flushed off by the street wash and consequently catch basins are now much less favored by engineers. The system of cleaning streets by heavy flushes of water, washing the refuse into catch basins, should not be adopted until an investigation has been made to ascertain which of the two general systems of cleaning is more economical in a given case, sweeping up the refuse and removing it before it gets into the catch basins, or flushing it into the basins and later removing it from them. If the latter plan is adopted the street cleaning department should be charged with a part of the cost of cleaning the catch basins. While there is not the imperative need for catch basins at frequent intervals which was formerly believed to exist, they have their uses where large quantities of grit will be washed to the gutters and, if it reaches the sewers, will obstruct them. If they are used, they must be kept clean. This work is expensive and often neglected. In some localities, where water remains in them for some time during the summer, they become breeding places for mosquitoes. The practical objections to them were summar- ized as follows in 1914 in a report by the Metropolitan Sewerage Commission of New York: "Theoretically desirable, catch basins are, in reality, among the most useless devices employed for the removal of solid material from sewage. They are generally ineilective because they are not cleaned with suflBcient g 130 SEWERAGE AND SEWAGE DISPOSAL frequency to enable them to serve as traps. It seems impracticable to keep them clean. To maintain catch basins in serviceable condition re- quires much hand work and this is costly. The work is usually carried on to the annoyance of pedestrians and householders. Some sewerage systems are without catch basins and >]<—/'/o'--^ their elimination, as a general procedure, is much to be desired." At the time this report was made, the standard catch basin in the Boroughs of Manhattan and the Bronx, New York, was that illustrated in Fig. 69. The drawing shows an inlet at the angle of a street corner, but the same construction of the basin was standard with straight inlets used between street corners. The hood which formed the trap was hung from a plate set in the masonry. The hood could be lifted off this plate and its purpose was to prevent the entrance of float- ing refuse into the pipe lead- ing to the sewer. A type of basin which long found favor, particularly in New England, is illustrated in Fig. 70. This type is put out of service automatic- ally when it becomes filled, and its special feature is its trap. As sediment collects, it reduces the space available for water above its top and below the water line established by the lip of the trap. Eventually there will be very little water capacity, and in summer, in prolonged dry weather, the water will evaporate to such an extent that odors may arise from the catch basin. If no odors arise and the cleaning gang does not reach the basin in its regular routine, the sediment will gradually Fig. 9. — Standard catch-basin, of Manhattan. Borough SEWER APPURTENANCES AND SPECIAL STRUCTURES 131 collect until it overflows the edge of the trap, blocking it. When this occurs the first heavy storm will give undeniable evidence of the necessity of cleaning. In this way the trap serves a useful purpose by preventing the escape into the sewer of large quanti- ties of silt which might form deposits. Another advantage of this basin, due to its trap, is that the water which accumulates in it can be bailed by the cleaning gang into the trap and thus delivered directly into the sewer, instead of being lifted to the top and thrown over the street. The great disadvantage of the trap is its liability to freeze in cold weather, although it should not be forgotten that the air inside the sewers, which will come up to the sewer inlet, will be somewhat warmer than the outdoor atmos- Plan without Vertical Section. Manlnole Frame. Fio. 70. — Standard catch-basin, Providence. (Bottom changed in 192 1 by insertion of a sump to facilitate cleaning by eduotor. ) phere, and the sheltered position of the trap also has some effect in reducing the danger of this nature. Where basins are con- nected to storm drains there will be much greater danger of the freezing of traps. Like all attempts to use traps on catch basins or inlets, the permanence of the water seal is very questionable. It will evaporate during prolonged dry weather and it is idle to expect that a sewer department will keep all traps filled by means of a hose during such weather. Attempts have been made in some cities to use traps formed of vitrified pipe elbows, but this practice is inadvisable because of the liability of damage to the elbows when the basins are cleaned. Cleaning is rough work done as quickly as possible and everything within a catch basin must be designed to withstand hard usage. 132 SEWERAGE AND SEWAGE DISPOSAL Castings for Inlets and Catch Basins. — It is difBcult to obtain from the practicing engineer any definite statement of his reasons for preferring one type of casting to another. The Concord grate, shown in Fig. 71, has been used for many years in New England where the gutters are not Hkely to have heavy wagons passing over them. Of all the commercial types, it probably affords the easiest means for the storm water to enter the inlet, provided it does not become clogged with leaves, which experience shows have a marked tendency to accumulate. The North Berwick catch-basin head is of much the same type except that it is heavier and there is no entrance for the water around Merrimac. Concord Fig. 71. — Types of commercial catch-basiii covers. the rim. The D-pattern frame and grate were long used in Boston, but about 1910 a rectangular frame was adopted. There seems to be a general tendency toward the use of these rectangular frames, of the general type shown by the picture of the Merrimac frame. Fig. 71. They have two decided advantages over types having curves. The first is that it is practicable to keep the pavement of the gutter in better condition with a square than a curved casting for it to rest against. The second advantage is that the grate can be made as strong as desired without much difficulty and still have a large area available for the passage of storm water into the inlet. The Borough of the Bronx adopted in 1913 a cast-iron inlet head shown in Fig. 72, which has a curb opening as well as a gutter drain. Whatever type is adopted SEWER APP URTENANCES AND SPECIAL STRUCTURES 133 should afford an opportunity for securely bedding the frame upon the masonry of the inlet or catch basin, for otherwise it will be- o UJ (1> c a O c o o in o I a c o come loosened speedily and in rocking under passing vehicles come loosenea speeaiiy anu lu iockii it will destroy the pavement about it. 134 SEWERAGE AND SEWAGE DISPOSAL The material from which the frames and covers are made is rarely specified. Even where it is specified it is unlikely that it is tested or inspected at the foundry. Certain foundries are known to furnish good castings and when they sublet such work to other foundries they hold up the quality of the product in order to protect their own reputation. There is a danger in very loose specifications and lack of proper inspection, particularly in times of business depression. A foundry in a territory not ordinarily serving the city may conclude that it can manufacture poor castings which will be just good enough to be accepted under the existing conditions and it can afford to send out such poor castings because it will probably never desire to do business with the city again, owing to the freight rates against it. This danger can be avoided by requiring the castings to meet the standard specifications for gray-iron castings of the American Society for Testing Materials and adding a clause requiring them to be painted with an asphaltum, coal-tar or graphite paint or other covering desired by the engineer. MANHOLES, DROP MANHOLES AND WELLHOLES Manholes did not come into general use on sewer systems until experience in cleaning sewers without them proved they were necessary. The chief opposition to them was based on the belief that they would permit the escape of sewer air and that this air was poisonous, an opinion not surprising when one reads about the evil odors from some old drains. In the early days of sewerage, when a sewer became so badly clogged that it must be cleaned, it was customary to dig down to the sewer, break through its walls, remove the obstruction and then rebuild the walls. Nowadays with the better design and greater care to get self- cleansing velocities in the sewage flow there is much less trouble and nuisance of this sort. Manholes have commonly been placed approximately 300 ft. apart for convenience in entry and cleaning of the sewers, but the frames and covers cause serious inconvenience and ex- pense in repair and maintenance of the highways, which has led in some cases to substantial increase of this interval; thus in a number of cases in Worcester, Eddy increased the spacing to intervals as large as 1,000 to 1,200 ft. without subsequent dis- advantage. It appears likely that the substantial increase in cost of manholes, now nearly treble the pre-war cost, will lead SEWER APPURTENANCES AND SPECIAL STRUCTURES 135 to further examples of this sort. Manholes should, however, be placed at all intersections of the sewer for convenience in access and cleaning, regardless of the interval. Fig. 73 illustrates a standard form of manhole for pipe sewers. It is generally considered desirable to have the walls of the channel rise nearly to the crown of the sewer and to have a berm on each side of the channel and sloping slightly toward it. In the Newark manhole the berms are a little higher than is often considered necessary. Fig. 73. — Standard manhole, Newark. This manhole, like the great majority of these structures, is built of brick, although under some conditions concrete may possibly be used to advantage, particularly where a number are to be built so that standard forms can be utilized, or where the man- holes are very deep and require considerable masonry. The expense of procuring forms and the delay which their preparation frequently entails, the difficulty of placing them and of placing the steps in the concrete, and the small quantity of concrete which is used, generally make it less expensive to employ brick in ordinary manhole construction. 136 SEWERAGE AND SEWAGE DISPOSAL Where the sewer is much larger than the diameter of the manhole, the outside of the latter is usually made tangent to one side of the sewer, for otherwise it will be difficult to enter the sewer and a special ladder will be required to reach the invert. Occasionally, on very large sewers, the manholes are built entirely apart from the sewer proper and have a passage leading into it, as shown in Fig. 74. The manholes of small sewers are usually made about 4 ft. in internal diameter when of circular cross-section, or about 3 by 4 ft. when oval. The same size is usually maintained for all sewers except when special conditions may require manholes of larger Section through Monhole. Longitudinal Sectiort on Sewer. Fig. 74. — Manhole on large St. Louis sewer. size, as where gaging devices must be used at the bottom of the manhole or it is desired to have a considerable storage capacity in the manhole chamber to enable it to hold enough sewage for flushing a long line of pipe on a flat grade. Brick manholes are usually built of 8-in. brickwork down to a depth of 12 to 20 ft., although for many years the manholes on Cincinnati sewers, and possibly those in some other cities, have been built of a single ring of brick. Below the depth stated, 12-in. brickwork is used, as a rule. The sides are carried up vertically to within 3 or 4 ft. of the top and the upper part is corbelled in or laid in the form of a dome or reverse curve, Fig. 76. The four manhole bottoms shown in Fig. 75 illustrate some- what different types of design. The Memphis and Seattle bottoms have flat lower surfaces while the Concord and Syracuse bottoms have lower surfaces curved to correspond with the SEWER APPURTENANCES AND SPECIAL STRUCTURES 137 channels through them. Which type is better adapted for the soil conditions at any site can be ascertained only by examination; ->; %^ ' 1 ^ J J 5f € =^ - )\ 'W.'l.lll. Memphis. Seattle. Concord. Svratose. Fig. 75. — Types of manhole inverts. the saving in material in the second type may be counterbalanced by a higher unit cost. While the base of each manhole illustrated FiQ. 76. — Types of manhole tops. was constructed of concrete, as a matter of fact a good sewer mason can lay up brickwork to form practically any channel 138 SEWERAGE AND SEWAGE DISPOSAL that may be desired, and can carry the work on very expeditiously if he is so minded. Two types of concrete manholes have been used in Syracuse on concrete intercepting sewers. In the first type the manhole has a reinforced shell 6 in. thick, running up from the sewer to within 5 ft. of the ground surface, where a funnel-shaped top Section A-A _ ., _ „ Section B-B. Fig. 77. — Double manhole for separate system. begins to corbel in. The reinforcement is of 3^-in. rods 12 in. apart when horizontal and 18 in. when vertical. The other type of concrete manhole is formed of pre-cast reinforced concrete pipe set on end. The sections are 4 ft. in diameter and 4 ft. long. SEWER APPURTENANCES AND SPECIAL STRUCTURES 139 Double manholes are sometimes used where sewers and drains are so located as to make them convenient. The structure shown in Fig. 77 was used by the authors for such a purpose on the separate sewer system of Hopedale, Mass. Each chamber is Sectional Plan. FiQ. 78. — Drop manhole, Newark, N. J. 5 by 4 ft. in plan and the dome has a depth of 4 ft. The walls are 9 in. thick. Where underdrains are employed it is sometimes desired to afford access to them, and in such cases various expedients are 140 SEWERAGE AND SEWAGE DISPOSAL , Surfa ce \ of Street 1 i ^il \ 7 Circular . Brick Sewer I Fig. 79.— Wellhole, Morgan Run sewer, Cleveland. SEWER APPURTENANCES AND SPECIAL STRUCTURES 141 employed. The most usual one is to divert the underdrain a short distance to one side of the sewer, where it passes under the manhole, and to bring up a riser from the underdrain to the floor of the manhole. Drop Manholes. — There are many types of drop manholes, of which one of the simplest is shown in Fig. 78. This particular structure is at the head of an oval sewer 51 in. high, into which two circular sewers discharge at different elevations. Where underdrains are used below sewers which are connected by drop manholes, the underdrain of the high sewer is generally Ifc — ..,.., ^s'6^ J,ongl*uolinal Section. FiQ. 80. — Flight sewer, Baltimore. Section of Manhole. connected by a pipe bedded in concrete with the underdrain of the lower sewer. Wellholes. — ^Where the sewage drops through a considerable distance in the manhole itself, the structure is usually termed a wellhole. Fig. 79 shows such a structure built in Cleveland, where wellholes have been used freely. In some cities, notably Minneapolis, the depth of the wellholes is sometimes over 100 ft. and special attention is paid to lining all parts subject to erosion with granite, granolithic block or other material which will resist the attrition of the sewage. Wellholes are rarely used except where the lower sewer is built by tunneling. Flight Sewers. — ^While a drop manhole or wellhole affords a means of changing grade sharply, other devices have been used 142 SEWERAGE AND SEWAGE DISPOSAL SEWER APPURTENANCES AND SPECIAL STRUCTURES 143 to accomplish the same or a similar purpose. The flight sewer, which gets its name from its resemblance to a flight of stairs, is occasionally used in such situations. It has a steep grade, but steps in the invert tend to check the velocity of the current; the resistance they offer probably diminishes as the depth of the sewage increases and if the descent is long great care should be taken to obtain massive, durable construction and freedom from obstruction to flow at the bottom of the flight. An example of such structures is shown in Fig. 80. Vertical Section. Elevation. ^cn 'I. <--rt~t- 7 - 'i^y////////////y -f 'X-~4-- -> XT H Sectional Plan. Fig. 82. — Cast-iron box step, Boston. Manhole Castings.^ — ^Experience indicates that the outside face of a manhole frame should be vertical from top to bottom and without projections, for a blank surface apparently enables the pavement resting against it to show a little better resistance to wear than where there are projections at the top of the frame or the latter has a broken surface. Circular frames are universally used as they are better adapted to the service than rectangular frames with which there is a tendency for the formation of ruts 144 SEWERAGE AND SEWAGE DISPOSAL Cross Section. in the pavement along the sides parallel to the direction of the traffic. The practice of making the covers rather deep, with a pocket in the top in which asphalt or wood block is placed, has given way to the use of cast-iron covers with the surface broken by a shallow pattern of some sort which will prevent horses from slipping and yet cause no jar when wheels pass over the cover. The standard design in the Borough of Manhattan, New York, shown in Fig. 81, consists of a frame weighing 475 lb. and a 135-lb. cover. The latter is raised by inserting the end of a pick or bar in the recess, C. The cover has six ventilating holes. The use of ventilating covers is considered necessary by some engineers when a system of sewers is first put into service and there are but few house connections with it to afford ventilation. After the system has been in use for some time, there seems to be a general tendency to use closed covers on a considerable portion of the manholes and employ the ventilating covers only where —5 p^ the need for them is evident. ^2 ^ The openings in the covers are often closed by oak plugs, but the authors have found that the best way is to have a black- smith plug them with rivets. Attempts to fill them with cement or an asphaltic mixture are not successful for any length of time, as a rule. In some cases the manhole covers are provided with a groove for a slide below the perforated portion; when the cover is to be closed the slide is inserted in place, closing the bottom of the holes, which are then filled with grout or some equivalent material. Steps. — In shallow manholes steps were generally provided, until about 1910, by leaving bricks projecting at points, about 15 in. apart vertically. They are slippery when wet and liable to become broken, so forgings came into use, as shown in the Newark manhole, Fig. 73. These forgings are usually placed from 12 to 18 in. apart vertically and somewhat staggered. Fig. 82 shows a cast-iron step used in Boston where the manhole must be kept free from projections from the wall. It is one of the Plan. Fig. 83. — Cast-iron manhole steps. SEWER APPURTENANCES AND SPECIAL STRUCTURES 145 best devices in use. Fig. 83 shows a cast-iron step used in the same city where there is no objection to the projection of the step from the wall. When wrought-iron steps are used they should be fashioned out of not less than ^-in. round iron. If made of steel they corrode and waste in section rapidly. JUNCTIONS Where large sewers come together with a horizontal angle between their axes less than about 30 deg., a special structure called a junction is required. For many years these junctions were of the type shown in Fig. 84 and were sometimes called "bell-mouths" or "trumpet arches" on account of their shape. Where, at the springing lines, the outside surfaces of the arches of the two sewers come together, no further attempt is made to have the upper part of the two sewers independent, but a large arch is thrown across the two. A manhole is usually built just in front of the brick wall which closes the large end of the structure. These bell mouths are usually built of brickwork but sometimes concrete has been used. So much skilled labor is required in constructing them of either material that flat-topped junctions, Fig. 85, have come into use. The particular junction illustrated is rather more expensive than most of this type because it was constructed on a large existing brick sewer which it was desired to disturb as little as possible. In designing junctions it is desirable to have the two streams of sewage join in such a way that there shall be the minimum check- ing of the current in either sewer, to prevent consequent deposit of grit and other suspended matter. This condition requires the two streams of sewage to have the same surface elevation as they approach each other and the same velocity, which is not easy to bring about when one sewer is a large trunk sewer and the other is a much smaller sewer. If the invert of the smaller sewer is placed at an elevation which will make the surface elevation of both streams identical during dry weather, the two streams will probably have very different elevations when con- siderable storm water is flowing away. Consequently a study of conditions at the junction when the two sewers carry various quantities of sewage may show that, to prevent sewage being backed up in the smaller sewer during periods of large discharge, it will be desirable to increase the grade of the smaller sewer some- what for a short distance from the junction. 10 SEWERAGE AND SEWAGE DISPOSAL SEWER APPURTENANCES AND SPECIAL STRUCTURES 147 148 SEWERAGE AND SEWAGE DISPOSAL In large cities the junctions are not always such simple affairs as those shown in Figs. 84 and 85. In Fig. 86 a complicated junction in Philadelphia is illustrated. Here there is a brick sewer 9 ft. in diameter crossing a brick sewer 8 ft. 3 in. in diameter, and the problem was to put in junction chambers and separate sewers in such a way that the course of the larger sewer, beyond this intersection, would serve as a relief for the storm water from the smaller sewer, and that the dry-weather sewage in the latter would flow into the channel which would also carry the dry- weather sewage from the former. This was accomplished by four junction chambers and two 30-in. cast-iron pipe sewers, as shown in the illustration. A very large proportion of the section of the 9-ft. sewer will be utilized before there is any discharge from it into the overflow sewer, while in the case of the S^^^-ft. sewer everything that is not dry-weather sewage will be immedi- ately discharged into the overflow outlet. INVERTED SIPHONS An admirable report upon "Inverted Siphons for Sewers" was prepared by a committee of, and presented to, the Sanitary Section of the Boston Society of Civil Engineers in 1921, from which, as well as from "American Sewerage Practice" the dis- cussion herein has been prepared. General Features of Inverted Siphon. — Any dip or sag intro- duced into a sewer to pass under structures encountered, such as conduits or subways, or under a stream or a valley, is termed an inverted siphon. While the sewage both above and below the inverted siphon normally but partly fills the sewer, it completely fills the siphon. The siphon is illustrated on a small scale by the running trap on the main house-drain, which formerly was generally used but which is now more often omitted in the interest of better ventilation. Size and Velocity. — Practical considerations, such as the increased danger of stoppage in small pipes, tend to fix the minimum diameters for inverted siphons about as for ordinary sewers, 6 or 8 in. in the separate system and say, 12 in. in the combined system, although there are rare examples of the smaller sizes. As obstructions are much more difficult to remove from an inverted siphon than from a sewer, especial care should be taken to prevent their formation. As high velocity as possible should be maintained in the siphon, say 2 to 3 ft. per second for Section C-C. Section H-H,. FiQ. 86. — Large sewer junction, Philadelphia. (Insert Facing Page 148.) SEWER APPURTENANCES AND SPECIAL STRUCTURES 149 domestic sewage and 4 to 5 ft. per second for storm sewage, although existing conditions frequently make it practically- impossible to ensure the development of such high velocities. In such cases correspondingly greater reliance must be placed on artificial flushing and other means of cleaning the siphon. In some cases catch basins or grit chambers have been built just above the siphons, but these are troublesome to clean and the material removed from them is usually very offensive. Siphons should be flushed frequently and their operation inspected regularly, in order to assure prompt removal of obstructions before they form a serious barrier. A serious experience following the clogging of the Dean St. and Fourth Ave. sewer in Brooklyn in 1911 gives point to this.^ "The siphon consisted of two 48-in. pipes for the storm flow and one 18-ia. pipe for the dry-weather flow, and replaced a portion of the 72-in. circular sewer in Dean St. Four days were consumed in pumping out the sewage and gaining access to the siphon. When it was possible to enter the siphon chamber it was found that the entrance to all three pipes was completely obstructed with a mound of debris consisting of a mattress, a lot of lumber of various sizes, a long-handled shovel, a piece of bluestone curbing, a wooden barrel, several iron pails, umbrella frames, together with rags, wire, paper and other refuse." Means of Flushing. — The flushing is accomplished in various ways, depending upon the available facilities and surrounding conditions. It may be done by temporarily speeding up the pumps, if there is a pumping station on the sewer line; or by receiving the sewage flow, at the head of the inverted siphon, in flush tanks which automatically discharge when full; or by pro- viding a permanent flushing gate in the manhole at the head of the inverted siphon or else grooves for the temporary insertion of stop planks when desired; or by opening the blow-off at the low point of the pressure pipe of the inverted siphon; or by admitting clean water at the head of the inverted siphon from a permanent con- nection with an accessible stream, a street water main or through a hose line from a neighboring hydrant; or by cleaning by hand by the use of jointed rods with suitable scrapers or other tools, after draining the siphons. Manholes or Clean-out Chambers. — Manholes or clean-out chambers should be provided at each end of a siphon, to give 1 Municipal Eng. Jour., 1917, paper by J. L. Hunt. 150 SEWERAGE AND SEWAGE DISPOSAL access for rodding, pumping and, in the case of pipes of large size, for entrance. There is objection to the introduction of inter- mediate manholes on an inverted siphon in such a manner that the sewage will be free to rise in them, since grease and other scum tends to fill up the shafts in them, with a solid plug. They are advantageous, however, if the sewage be confined within the siphon as it passes through the manhole, affording access or means of ridding the siphon of deposit, through a gated connection or similar device. Materials of Construction. — Since an inverted siphon is subjected at all points of its cross-section to an inner pressure, the walls will be in tension, although the amount of tension may- be modified if the exterior of the inverted siphon is subjected to external water pressure or to the pressure of earth. On account of these tensile stresses, inverted siphons are usually constructed of steel, iron, reinforced concrete or wood-stave pipe heavily banded, though vitrified pipe has sometimes been used suc- cessfully under a small head or if the pipe be encased in concrete. Size of Pipe for Inverted Siphons. — The computation of the sizes of pipe for inverted siphons is made in the same way as that of sewers and water mains. Their diameter depends upon the grade and the maximum quantity of water to be carried. The latter depends in the case of inverted siphons under rivers, upon the presence or absence of a storm overflow before the inverted siphon is reached, and on the degree of dilution of the sewage before the overflow outlet comes into service. The head or drop in the hydraulic gradient actually required at any time, for the existing flow, will be the difference in level in the free water surface at the two ends of the siphon. It will equal the sum of the friction head determined by the quantity of sewage to be carried and by the character and condition of the interior surface of the siphon, its actual length measured along its axis, the head lost at bends, changes in cross-section, and the loss in velocity head involved by the variation in section. Marked variation in velocity has an important effect upon the velocity head and the total head required. It is to be remembered that the losses will be relatively small for low velocities, other things being equal, but that they increase roughly as the square of the velocity. For a clean 12-in. siphon 50 ft. long and a velocity of 3 ft. per second, a total loss of 6 in. would probably be an outside figure; for a velocity of 2 ft. per SEWER APPURTENANCES AND SPECIAL STRUCTURES 151 second 3 in. ; but for 6 ft. per second 24 in. ; the friction loss alone being not over one-third of the above values. Practical experience has shown that where the siphons are built of unduly large size they soon silt up to a point where the reduced section will increase the velocity of flow through it to a point sufficient to maintain the section. Such a condition was observed by A. L. Shaw in a 36-in. siphon under the Brandywine River in Wilmington, Delaware, in which the equivalent reduced diameter of the pipe for the observed quantity and slope was estimated at 17 in. and 24 in. under two different conditions of flow, the former corresponding to an estimated velocity of 2.7 ft. per second, the latter to 4 ft. per second. To overcome this difficulty, experience in this country has dictated the advantage of using several or multiple pipe hnes, instead of one pipe line, for the siphon, arranged in such manner as to throw additional pipe lines progressively into action with increase in discharge of the sewage. See solution of siphon problem at the end of this section. English authors have emphasized the importance of ven- tilating long inverted siphons, asserting that otherwise the flow may be interfered with by accumulations of air or gas, but American sewerage practice does not seem to have developed such difficulties. Care must be taken that inverted siphons built on or under river beds, have sufficient material to prevent their flotation. Essential Principles in Design. — The essential principles in design have been well set forth in a paper by S. B. Bleich,* summarized as follows : 1. Complete and effective separation of house sewage and industrial waste from the storm run-off at all times.'' 2. Simplicity of construction. 3. As slight and easy changes of direction of flow as are practicable at the entrance and exit legs. 4. Easy curve where necessary. 5. Uniform section throughout entire length of siphon pipes. 6. Elimination of all features tending to obstruct flow. 7. Omission of all moving parts and mechanical devices. 8. Provision for easy access to all pipes, without impairing any of the essential features. 9. Entrance and discharge openings to have sufficient area within the available height not to cause any backwater; and change of section or direction not to be too abrupt. I Municipal Eng. Jour., 1917, Paper 110. ' This first statement is not a principle, nor is it justified as broadly put. 152 SEWERAGE AND SEWAGE DISPOSAL SOLUTION OF A SIPHON PROBLEM A problem encountered in the authors' practice involved the design of an inverted siphon of several pipes, to replace an existing single-pipe siphon which had given trouble from sedimentation due to low velocities; its solution is given as an illustration. The basic data are as follows: Length of siphon 440 ft. Available fall (invert to invert) 3 . 2 ft. Maximum depression of siphon 9 ft. Gravity sewers connected by siphon: Diameter 30 in. Material Concrete Slope 0.0037 Typical rates of flow: Minimum 4 cu. ft. per second Maximum dry weather 13 cu. ft. per second Ultimate maximum capacity of gravity sewer The capacity of the 30-in. concrete gravity sewer on slope 0.0037 (n = 0.013) is about 25 cu. ft. per second, and the velocity flowing full 6.1 ft. per second. Available fall 3.2 ft. Assumed loss at inlet 0.4 (see later discussion) Available for friction in siphon 2 . 8 ft. 2.8 Tjn = . 0064 approximate available slope. Local conditions indicate economy in the use of vitrified pipe. The surrounding concrete (primarily for anchorage) can be easily reinforced for the internal pressure due to the 9-ft. depression of the siphon. Standard commercial vitrified-pipe sizes will therefore be selected. Two conditions must, if possible, be satisfied; the velocity in the pipes selected should be suflicient to insure scouring (3 ft. per second, if possible) ; and the available hydraulic slope of 0.0064 must not be exceeded. Judgment suggests the selection of pipe sizes which will be particularly adapted to the minimum flow, the maximum dry weather flow and the ultimate maximum flow. Three siphon pipes are found to meet these requirements, with regulation at the inlet so that the minimum flow will be confined to one pipe; the maximum dry-weather flow to two pipes; and all three have a capacity equivalent to that of the gravity sewer in which the siphon is to be inserted. The computations follow : FoT minimum flow, 4 cu. ft. per second, from Fig. 47, (n = 0.015 for vitrified pipe) 15-in. pipe requires s = 0.006; velocity, 3.3 ft. per sec- ond. Capacity at s = 0.0064, 4.2 cu. ft. per second (beyond which second pipe begins to operate). SEWER APPURTENANCES AND SPECIAL STRUCTURES 153 For maximum dry-weather flow 13 . cu. ft. per second Capacity of 15-in. pipe 4.2 Required capacity of second pipe 8.8 cu. ft. per second From Fig. 47, 20-in. pipe, s = 0.0057; velocity, 4.1 ft. per second. Capacity of 20-in. pipe at s = 0.0064 9.3 cu. ft. per second. Capacity of 15-in. pipe 4.2 Combined 15- and 20-in 13 . 5 cu. ft. per second (beyond which third pipe begins to oper- ate). For ultimate maximum, Capacity of 30-in. sewer 25 . cu. ft. per second. Combined capacity, 15- and 20-in. siphons .... 13.5 Required capacity, of third pipe 11.5 cu. ft per second From Fig. 47, 22-in. pipe requires s = 0.0058, velocity 4.3 ft. per second Capacity of 22-in. pipe at s = 0.0064 12.0 cu. ft. per second. Combined capacity of 15- and 20-in. pipes. ... 13.5 Total available capacity 25 . 5 cu. ft. per second (25 required). Design of Inlet Chamber. — A type of inlet chamber. Fig. 87, suitable in this instance has the invert of the pipe which carries the low flows continu- ous with that of the 30-in. gravity sewer; one side of the sewer is cut down to an elevation which permits overflow to the second siphon when the capacity of the first is exceeded; and the other side of the sewer is cut down to a higher elevation, which allows the flow in excess of the capacity of the first two pipes to reach the third. Ascertain, by the method of Chapter IV, the depth of flow in the 30-in. gravity sewer when the 15-in. siphon shall have reached the limit of its capacity, (4.3 cu. ft. per second), and when the combined capacity of the 15- and 20-in. siphons (13.5 cu. ft. per second) is flowing. These depths determine the elevations of the tops of the walls which permit overflow to the 20- and 22-in. siphons and are found to be about 8 and 16 in. respectively. Under maximum conditions for either of the two larger pipes, the over- flow walls will be submerged both upstream and down. They cannot, therefore, be considered as weirs but more as obstructions causing certain loss of head in passing the desired quantities of sewage over each. The flow over these walls is at right angles to that in the approaching sewer, so that this loss may be taken as the head required to produce the necessary velocity across the top of the wall, assuming the energy of velocity of approach to be lost in the change of direction. A maximum of 9.3 cu. ft. per second must pass across the wall to the 20-in. siphon. The depth on the wall will then be at least 8 in. or the difference in elevation between the two walls, the higher wall having been figured to overflow just as the capacity of the first two pipes was exceeded. 154 SEWERAGE AND SEWAGE DISPOSAL The higher wall must pass at least 11.5 cu. ft. per second, and the depth available is the distance from the top of the wall to the crown of the 30-in. sewer, or 14 in. Assume the length of walls as 7 ft.; the inlet losses may be approximated as follows : 15-in. pipe; negligible, since invert is continuous with that of 30-in. sewer, and transition can be made to take advantage of the velocity of approach which, even at low flows in the gravity sewer, is equal to or greater than the siphon velocity. (Velocity of approach may be determined by method of Chapter IV for the partially filled sewer.) 1 jMax W.S. '^ Section A-A Inlet Chomber Section S-a Outlet Chamber Sectional Elevatloh 'O-o Fig. 87. — Inverted siphon inlet and outlet chambers (problem in inverted siphon design). 20-in. pipe; velocity over wall, 9.3/0.67 X 7 = 2.0 ft. per second; corresponding head, 0.06 ft. After passing over wall, velocity of 4.3 ft. per second in 20-in. pipe must be built up in new direction, which gives a loss of head of 0.29 ft. Total 0.35 ft. (less than assumed inlet allowance of 0.4). 22-in. pipe; velocity over wall 11.5/1.17 X 7 = 1.4; corresponding head, 0.03 ft. Velocity in 22-in. pipe, 4.3 ft. per second;head, 0.29; total, 0.32 ft. Allowance of 0.4 ft. for inlet losses appears ample, and no revision of the computation of available slope is required. Design of Outlet Chamber. — At the lower end of the siphon, the junction of the three siphon pipes with the 30-in. gravity sewer should be so designed as to reduce the opportunity for eddies to carry sediment back into those pipes which are not at the moment operating, but which are full of standing water. This is especially important in the case of the 22-in. pipe, which will not be in operation except at unusual rates of flow. It may be accom- plished by maintaining the three pipes, or the corresponding channels within the junction chamber, as nearly up to the point of intersection SEWER APP URTENA NCES AND SPECIAL STR UCTURES 155 as possible, to avoid pooling and reduction of velocity in the chamber. As a further precaution, the outlet of the 22-in. pipe (least frequently required) may be raised so that the invert of its channel has a sharp forward pitch toward the intersection. The crown of the pipe must not, however, be raised above that of the 30-in. sewer, or it will lie above the hydraulic gradient. BRIDGES Highway bridges are not often used to support sewers. If they are so employed, the grade of a sewer must generally first rise and then fall in passing over the structure, converting the sewer into a true siphon, which is objectionable to most designers of sewer systems. In some cases, however, it is desirable to — S'O;-- <- 3jj?f- >i< 3j;,.-. -^"-^Hfe--K----i-j tn i. Brick '"^Paving IZ'ihhk Boulders in Concrete Vpsfream and Downstream from Pipe Longitudinal Section. FiQ. 88.- • Cross Section Enlarged. Sewer bridge, joint outlet sewer, New Jersey. carry a sewer on a special bridge without any change in grade, as shown in Fig. 88, illustrating a structure designed by Alexander Potter. Occasionally reinforced-concrete beams are used rather than steel beams for such structures. In any case, care must be taken that the piers and abutments of the bridge do not unduly restrict the waterway of the stream crossed. Most engineers prefer to use inverted siphons rather than bridges for stream crossings, if the head necessary for the successful operation of the siphon is available. FLUSHING DEVICES The primary object of flushing devices is to permit sewers to be laid on flat grades which, while giving adequate velocity to assure the desired capacities at the depths assumed in the compu- tations, are not enough to give at other depths velocities which will cany off at all times all suspended matter. The problem of 156 SEWERAGE AND SEWAGE DISPOSAL flushing, strictly speaking, is usually merely one of keeping small sewers clean from their dead ends to the points where the flow of sewage is great enough to accomplish this without assistance from the water mains. The quantity of water used for this purpose in one flush is rarely over 350 gal. Occasionally the problem is one of furnishing a large volume of water to clean a main sewer or an inverted siphon. Flushing from Brooks. — In some cities, where a large sewer is laid on flat grades near a brook or river, a flushing manhole is built, with a connection to the stream through which the sewer can be flushed with brook water. This has proved useful during dry weather when very little sewage is flowing and the low velocity permits solid matter to settle on the invert. Flushing Manholes.^ — The flushing of small sewers is usually carried on either by hand or with the help of automatic apparatus. As a general proposition, all automatic flush tanks require some maintenance, and their cost is therefore dependent, in a measure, upon the time spent in inspecting and repairing them. The cost of this time, plus the interest and depreciation on the invest- ment in the apparatus, plus the cost of the water used by the apparatus, must be offset against the cost of labor and water where hand flushing is practiced, for the difference in the cost of the manholes used, in the two cases is negligible. The quan- tity of water to be used for flushing and the frequency of flushing depend not only upon the grade of the sewer to be kept clean but also upon the possibility of dirt finding its way into the sewer. Hand flushing is generally done by means of a hose from the nearest fire hydrant, inserted into the manhole at the end or summit of the sewer to be cleaned. Flushing manholes are also used to some extent. In this case a 1- or 13^-in. service pipe from the nearest water main is run into the manhole and the entrance to the sewer can be closed with a flap or tripping valve. The same end is accomplished in some places where valves have not been installed, by plugging the end of the sewer with a disc consisting of sheet rubber faced with canvas and held firmly between boards about 3^ in. smaller than the diameter of the sewer. When the tank is filled with water, this plug is drawn out, starting the flush. Automatic Flush Tanks. — The flushing done with automatic apparatus is generally more frequent than where hand flushing is practiced, the usual rule being to discharge the flush tank once SEWER APPURTENANCES AND SPECIAL STRUCTURES 157 every 24 hr. The water usually enters the tanks through special orifices, of which a variety are manufactured by the makers of automatic siphons, so that any desired rate of flow under any street main pressure can be obtained by screwing the proper orifice or jet into the end of the service pipe. As a rule these jets are also accompanied by a mud drum or screening device and a blowoff cock, provided to keep the jet clear. The operation of a siphon of the simplest type is as follows: In Fig. 89a, the siphon is shown just ready to discharge. There 158 SEWERAGE AND SEWAGE DISPOSAL are two volumes of water separated by the compressed air in the long leg, V, of the trap. As the pressure on every part of this confined mass of air must be equal to the hydrostatic pressure, and as there are but two surfaces where the water is in contact with the compressed air, it follows that the depth of water, C, in the tank must be the same as the depth, H, in the trap. When the depth, C, is increased, the water flows over the raised lip of the trap at D, this discharge allowing a little air to escape below the bend at E. The air pressure being released in this way, water passes up with a rush within the bell and into the long leg of the trap. The elevation of the lip of the short leg at D above the bottom of the outlet is an important detail, as upon it the first sudden discharge of the trap seems to depend. In the older types of flushing apparatus this first strong flush was accomplished by using an auxiliary siphon at the bottom of the trap casting — a ii.- ^'i'k >|!Vi|< ^ I I I I I I I I I I n "1^ £ 4- O I FiQ. 90. — Flushing chamber, SEWER APPURTENANCES AND SPECIAL STRUCTURES 159 detail retained in the Rhoads-Miller siphon, Fig. 89d, for use where shallow construction is imperative. When the water has been drawn down in the tank until its surface is below the sniff hole S, air rushes into the bell and stops Concord inverted siphon. 160 SEWERAGE AND SEWAGE DISPOSAL the siphonic action there. In consequence the water in the two legs of the trap at once forms a seal there, and the apparatus is ready for discharge when the tank is filled again. The dimensions of the Miller tank, required by designers, are given in Tables 24, 25 and 26. The diameter of the tank is the minimum which is generally considered desirable for siphons of the size hsted. The discharge is the average given by the mak- ers for that size and setting of siphon. The setting shown in Fig. 89a does not afford access to the sewer, so the late Andrew Rosewater devised the special design shown in Fig. 89c to overcome this defect. The manhole at the A'P/aifer Section C-D. Section A-B. Fig. 91. — Van Vranken flush-tank. dead end of the sewer is provided with a flush tank and siphon, and while this is more expensive than the standard type, it not only affords an opportunity to insert a cleaning rod into the end of the sewer but also, it is stated, gives a higher rate of discharge. Fig. 90 illustrates a large flushing chamber at the head of an inverted siphon at Concord, Mass. The purpose of the chamber is to accumulate the sewage until a large quantity has been stored and to then discharge it rapidly, so as to keep the inverted siphon under the river clean. The chamber discharges from 15 to 20 times in 24 hr. With this flushing chamber the Van Vranken siphon, Fig. 91, is used. The sewer to be flushed ends in a well in the floor of the tank, which has a watertight metal cover. A 5-in. siphon has its long leg carried through the plate. The bottom of the leg is SEWER APPURTENANCES AND SPECIAL STRUCTURES 161 .g " & - 2 ff i^ ■* "3 CO oc =3 o i ■» a 'E.s >^ M (N cq N M t^ 5 00 d o §^- ^ Hh fi Tt< lO CO CD o m rn T3 1 5 (M -^ TtH .-( iM M ■^ U5 p -a M •9?. ->l 1- b 4d » S -2 "^ ■* M O iO 02 o=a ■H N CO CO s O s w M M rH ■* K N) iJ iJ 0..S • w[^ Hn Hn S &i ^ 00 0> "H CO i-H i-H Eq O m s H-« Ml-* w|« H« ^ &i ^ M< (N OS CO o 1-1 -l (M 3 ^ i CO 00 o m U.5 l( ^ 0^05 ■# CO 00 IN =3 '^ „ 1 o w s-.a ■^^ .3 N.S 05 .g K^ 4^ M.g s.g is.g cj 00 o I Hlea CO O ■ (N 'fl.g 3 CO 00 f-'-S •* lO CO 00 cJ.g 05 CO T-t iO tH (M C13 CO a:.s CO CO Td 10 <=.g 00 (N >0 tH T-i tH tef.g lO CD CO t> ^".s 4£ CO CO '^ »ni ki Oi 01 I-H M s-.d Hi-* n|^ ni>tfi n|oo CO t^ tH 00 1-H tH C Trans, Am. Soo. C. E., vol. Iv., p. 341. 2 Depth of flow in the 5-ft. sewer would be about 4J^ ft.; in the 14-ft. 9-in. sewer about 2J ■r 1 r rf. «/ ,d J 1^ /. »- 7 ^ r ' H ? r4 1 7 .7 r <• -^ 7 '? 7 (> 4, 7 >2 * w 15 '- 'F» y < 7 C7 r 'P (1 fp s; r '7 r / r^ ■7. 1? p- <■ '7 r /, ^J ■7 6W ■' '■i 1 f^ , ?' 1 7 >77 *- . fr If? T (x r r ■ 7 - / H 7 7 I* =?1 7 ,' ^o 3 S.M T •« v^. . < T ,=?! H i''^ 9i ■ n r. r A ^ '^ /. H, /.- IP r < <: '■P 7 ( ?o r 4' <, (?- f V ., . r' r/ ( ■< IT r7 c « ^ r ^ *■ r. J, fi 'fl t, I. 7 T fl r ff FT 1 {, ,4 7 T"** ^ ^ iti , fl'i 1 ' 7 f* /, - rf h '■f z -7 /) ? i < 1 i 7 r" « .< -^ ff. ^ r- 'f> -Sr i'^ yi U * * 'rt. / r ,'^ 9 '7 r fl* ii 9 p f r '7 < ,? f ^f '^- .1 ■1 rf ( T ?2 T (V- *■ r* /t, ?i ^ y f ^ .-T 7. '■^ <; 7 ? r/, «( 3f r fl M / .- ,r 9. fei ff £7 <9 ,*■ f "?, 9i fl .* '.■f p J 7 7 ij •f 9 H ^ F, 9 ? fVi ,t- 7 i. o l^ > 5- •r /; I 'm f f. >^ 'f >« P , -' Fia. 101. — Form of notes for staking out sewer grades; kept on facing pages of note-book. Each of the two pages of this note-book is 4^ in. wide and 7 in, high. The ruling on the left-hand pages is blue with every fourth vertical line red, while that on the right-hand pages is blue except for a vertical red line down the center. The lines are 3^6 in. apart vertically and horizontally. The date of the work and the initials of the instrumentman, rodman and note-taker should be placed on the top of one of the pages. group figured should be independently determined by the addition to, or subtraction from, the elevation of the first station of the group, of the difference in elevation between these two extreme stations determined by the intervening rate of grade, multiplied by the distance, as an independent check on all of the work. In running any levels, incident to setting batter boards for example, the last set up should be checked on a bench mark. PRELIMINARY INVESTIGATIONS 185 This bench mark should always be a different one from that used at the beginning in order that every group of levels may be tied to at least two bench marks. This method serves not only as a check on the work but also to disclose injury, whether accidental or malicious, to any bench mark. The cut found by levehng should be roughly checked by comparison with the cut shown upon the profile. On account of the permanent value of records of sub-surface conditions, particular attention should be paid to recording the characteristics of the soils, ledges, ground water, pipes, culverts and other underground structures encountered. It is well, also, in recording the date, to enter the weather condi- tions when the survey is made. Quantity Svirveys, Location of Specials, etc. — The measure- ments which must be made for quantity surveys will be governed by the method of payment stated in the contract. If the pay- ment is made on the basis of the feet of trench, it is unnecessary to make such precise measurements as when the payment is com- puted on the unit-price basis. In the latter case, a profile along the center Hne must be made, showing the ground surface and the profile of the top of all ledge encountered. The finished bot- tom of the trench may also be recorded if desired, though this is not essential, the essential thing being the necessary removal of the ledge and the replacement of suitable foundation material. Care must be taken to make the records of all conditions affecting quantities complete in order to avoid unnecessary controversy over payments. It is customary in some engineering offices to prescribe certain limits to which the quantity of excavation shall be paid for, although the contractor may excavate beyond those limits if necessary and, if the structure can be built in a narrower excavation, he is not required to take out material to the full width shown by the "lines of excavation " on the drawings. If the contractor is paid by unit prices, however, it is advantageous that the engineer's notes should show exactly what he did, irrespective of contract agreements regarding "lines of excavation," for it is only by having accurate records of what was accomplished that the actual facts can be reported in case of controversy, and such information has real practical value to the engineer of limited experience. Difficulties often arise on account of insufficient measurements and records, but very rarely on account of too many. In the case of works covering some area, such as pumping 186 SEWERAGE AND SEWAGE DISPOSAL stations or disposal works, the surface to be excavated should be cross-sectioned, as should all areas of ledge encountered as the work progresses. All features calling for special payment under the contract should be measured in advance of the work, such as the area of pavement to be removed, the area and quantity of top soil to be removed and kept for future use, and the length of pipe taken up for relaying. The length of each size of pipe laid in each section of the work and the character and location of all branches and other specials must be recorded. The reporting of the branches by inspectors is often unsatisfactory. It is sometimes the practice, when the engineer is not present all the time the work is in progress, for a pole to be placed at each branch, with its top extending to the surface of the street, so as to enable the engineer to prepare such reference marks or records as he considers best when he next visits the work. Other engineers require their inspectors to locate the branches from marks on the sheeting of the trench or by ties to stakes driven on one side of the trench. On account of the many mistakes which arise when branches are located by persons unfamiliar with the importance of the work, it is preferable to leave them uncovered until the engineer can take the necessary measurements. One method of location is by determination of the "station" of the branch or other fitting, by measuring the distance from the preceding batter board. This should be in- dependently checked by the measured distance from the next batter board, and the sum of the two measured distances should be compared with the distance between the two batters. The dimensions of manholes, catch basins, flushing chambers and other structures should be measured as soon as they are com- pleted. The length, size, location and construction of under- drains should be measured and recorded. The dimensions of all sheeting left in place and its location should also be recorded. Structural Surveying. — The surveying done in connection with the construction of disposal works, pumping stations and like parts of a sewerage system is performed like the surveying work in building operations. The object is to transfer lines from carefully established bases and grades from permanent bench marks, and to do this work in such a way as to avoid delaying the workmen at any time. The nature of the construction operations carried on is such that the working gangs can proceed most rapidly if they are provided with means of checking many of the PRELIMINARY INVESTIGATIONS 187 lines and grades approximately without calling upon a surveyor. Such means are furnished by batter boards, generally placed at the main corners and at other places where they will be most serviceable. A batter board is a horizontal board held in place, outside the building line far enough to be undisturbed by the work, on posts set vertically and firmly in the ground. On the top edge of this board one or more nails are driven directly on the building line to which the work is being conducted; at first in many cases, this line will be that of the footings, and later it will be that of the wall. As a board will often carry a number of nails, each should be marked to indicate the line on which it lies. If a cord is stretched tightly between the nails on two batter boards on opposite sides of a building it will give the line of a footing or wall. The top edges of the batter boards should, if possible, be a whole number of feet above the bottom grade, for fractional dimensions in working from batter boards often cause mistakes. During the progress of the excavation the foreman can be provided with a rod having a crosspiece nailed to it at a dis- tance from one end equal to. the elevation of the batter board above bottom grade. By sighting across the batter boards at the top of this crosspiece while the rod is held vertically it is possible to check closely the depth of excavation at any point and thus avoid taking out more material than is necessary. In carrying out such surveying care must be taken to pro- vide one or more base lines which will not be disturbed. These lines should have transit points accurately established on them so that the main points on the batter boards can be checked from time to time. In some soils which are much affected by the action of frost and large fluctuations of temperature, the main points on the base lines will require unusually long stakes or lengths of substantial iron pipe, for ordinary stakes will not remain in place. Similarly the bench marks must be selected so that their permanence is assured and the surveyor can use them readily as the work proceeds. Care should be taken to have sufficient bench marks to make turning points unnecessary for most of the work, and to permit ready checking of the eleva- tion of the batter boards at frequent intervals as the work progresses. As the work progresses lines and grades should be given by the surveyor on all forms before concrete is poured and the correct location of all parts of the structure should be tested by 188 SEWERAGE AND SEWAGE DISPOSAL him before the work on them proceeds. It is only in this way that costly errors can be avoided. INSPECTION The object of inspection is to see that the engineer's plans and specifications are carried out faithfully. Inspectors are assigned to the resident engineer for this purpose and one of their duties is to safeguard all his reference and bench marks, batter boards and the like, and to check them at frequent intervals as well as they can, so as to detect at once any disturbance of these points which will introduce errors into the work. The inspectors should also see that all transfers of line or grade from the engi- neer's references by the contractor's forces are made accurately. Workmanship must be inspected constantly in order to ascertain that pipe inverts are laid smoothly at the proper grade, the joints made as required, the surplus jointing material removed from the pipe, water kept away from the joints until they have set, concrete is mixed properly and the forms are located correctly and built to give a satisfactory surface on the finished work. The eye must be trained to notice instantly any deviation from courses, lines, grades, or standards of work. Excavation. — Carelessness in excavation may cause the settlement of adjoining structures, with consequent financial losses. The inspector should, therefore, watch out for unneces- sary excavation, insufficient bracing of the trench and failure to drive the sheeting sufficiently ahead of the excavation to prevent materials from flowing into the trench. When it seems possible that the excavation may cause settlement or other damage to adjacent buildings, elevations of controlling points, such as steps, corners, window sills and the like, should be taken before the work nearby has begun and at later intervals as the work progresses. Where the structures are important or damage to them will result in large financial losses it is desirable to have photographs of them taken from time to time, the negatives being dated and signed by the person making them. When the excavation approaches grade in the trenches the inspector should take precautions to prevent the excavation of material from below grade except when it is of such a character that it must be replaced by better material. The backfilling of an excavation must be watched carefully to see that there is no danger of settlement of the sewer or other PRELIMINARY INVESTIGATIONS 189 structure. Where underdrains are laid it is particularly impor- tant to see that the specifications are carried out faithfully and all refilling below subgrade is done thoroughly. As the backfilling progresses the inspector should prevent the dumping of material from a greater height than 5 ft. on pipe or masonry. Where the fall of the material is greater than 5 ft., it should drop first on a timber grillage or other device to take up the shock. Sheeting which has been driven below the springing line of a sewer should not be removed ordinarily, for it is very difficult to fill the voids left by such withdrawal. This is true, in even greater degree, of sheeting driven below the bottom of the trench. When sheeting is withdrawn the work should be done preferably slowly, a little at a time, and the voids thus left should be filled with the aid of a water jet. A piece of M-in. pipe is attached to a garden hose, forced down through the backfilling and moved about while water flows through it. The exact method of work will be determined largely by the character of the ground traversed. Concrete Materials. — The inspector is not usually required to test cement, but he should be watchful of all deliveries to make sure that only cement which has passed the tests is used in the work. Where cement is obtained from a dealer's warehouse this will often require tagging acceptable cement and placing the tagged bags aside for the exclusive use of the sewerage work. Where the cement is delivered to a contractor's storehouse in carload lots, tagging is unnecessary, but no lot of cement which has not passed the tests should be stored with approved deliveries. The degree and nature of the inspection of sand, gravel, crushed stone and brick will depend upon the local conditions and specifications for such materials. The inspector must be alert to detect at once any change in the quality of these materials, even if the change is not prejudicial to the work, and to watch for the presence of loam and dirt in aggregates intended for concrete. The deliveries of reinforcing steel should be checked to make sure that they conform with the bill of materials and the cross- sections of the rods should be measured. Absolute adherence to the usual specification that rods shall be free from rust and mill scale is practically difl&cult of attainment and engineers are reluctant to compel a contractor to assume the substantial expense which strict compliance with the requirement entails, in view of the protective effect of the mortar on the steel. Serious 190 SEWERAGE AND SEWAGE DISPOSAL rusting or pitting is ground for rejection of the steel, but the inspector should get clear instructions from the resident engineer as to what "serious" shaU signify on the work in hand, and what may constitute a reasonable interpretation of the specifications relative to this subject. Cast-iron Pipe. — This should be inspected as it is unloaded from the car and again before it is lowered into the trench. Special attention should be paid to the detection of cracks in the spigot. The inside and outside are rubbed with a heavy cotton mitten or bunch of dusty waste, which forces enough dust into a crack to make it visible. The presence of a crack can also be detected by the ring of the metal when it is struck with a wooden mallet, while suspended in the unloading slings or on the rolls. Vitrified Pipe. — If vitrified pipe is furnished by a manufacturer with a reputation for good mill inspection it is generally con- sidered unnecessary to inspect the pipe at the mill, but on account of the liability of being cracked in transportation the pipe should be inspected as it is unloaded from the cars and again before it is placed in the trench. The inspection comprises ringing the pipe with a light hammer and examining it for warping, deforma- tion, defects in the bell and spigot, and poor interior condition. A light tap of the hammer on a pipe set on end will indicate the presence of a crack so small as to be invisible until the surface has been wiped with chalk and the pipe again tapped lightly with the hammer, which will cause the crack to become filled with chalk and visible. EXCAVATION The excavation is usually the part of sewer construction most important in determining costs. To handle it to advantage re- quires good management and judgment on the part of the man responsible for its execution. The materials used in sewer con- struction generally cost about as much as the labor, but the contractor makes little profit on supplying them. The chief profit is made by handling the work so well that labor charges are kept to a minimum. The time element is also important, for rapid work during the summer when costs can be kept low may avoid a large unnecessary expense if the work is prolonged into the winter. Charges for enginemen, firemen, tagmen and machinery are among the expenses which mount up rapidly when progress is slow. It is necessary to allow for the delays PRELIMINARY INVESTIGATIONS 191 due to unfavorable weather, particularly rain and frost, over which the sewer builder has no control, and on large undertakings contractors now often arrange with the local oSice of the U. S. Weather Bureau for early notice of probable unfavorable condi- tions. Classification. — There is little agreement among sewerage engineers as to the classification of excavated materials for pur- poses of payment. The prevailing opinion among Eastern engineers, whose work is with materials in which the rock is usually hard and fairly sharply distinguishable, is that a classification into rock and earth should be made. Among the engineers of the Pacific Slope it appears customary to use no classification. The authors use the following classification for rock: "Rock, wherever used as the name of an excavated material, shall mean boulders exceeding J^ cu. yd. in volume or solid ledge rock which, in the opinion of the engineer, requires for its removal drilling and blasting, or wedging, or sledging or barring. No soft or disintegrated rock which can be removed with a pick; no loose, shaken or previously blasted rock or broken stone in rock fillings or elsewhere, nor rocks exterior to the maximum limits of measurement allowed, which may have been previously loosened in excavating for water pipes or other purposes and which, by reason of such loosening, may fall into the trench, will be measured or allowed as rock." Opening the Trench. — Where the paving materials are to be used again they should be placed in neat piles outside the prob- able limits of the excavation. The crushed stone from maca- dam pavements is often piled on the opposite side of the trench from that used for storing the remainder of the excavated material. Where the excavation is deep or likely to be open long the material to be replaced is so likely to become lost that, if of good quality, it is sometimes hauled for storage on vacant lots or on one side of several cross streets. Frost enters compacted ground to a greater depth than where it is in its natural condition. Where the frost sinks to a depth of 3 ft. in a field or wood lot it may penetrate 5 or 6 ft. in a neigh- boring well-traveled road or street. Opening a trench in frozen ground by picking and barring is so slow that the surface is first thawed by building fires on it, usually at night so the ground will be ready for work the next morning. Sometimes the surface can be steamed, using a portable boiler and a number of wooden boxes 12 to 15 ft. long, about 10 in. high and approximately as 192 SEWE]iAGE AND SEWAGE DISPOSAL wide as the trench. They are laid with the open side downward and holes are bored in the top every 12 in. longitudinally and at 2-ft. intervals across the box. Each hole has a wooden plug to prevent the escape of steam. After the boxes are in place they are banked with earth. The steaming is done at night, one man being enough to look after the boiler and thaw the ground. He does this with a %-in. gas pipe about 6 ft. long provided with an iron crossbar for a handle. To the end of this pipe is attached a steam hose connected with the boiler. The operator pushes the pipe down vertically through one of the holes in the top of the box, and as the steam thaws the ground, he gradually works the iron pipe down until it has penetrated the entire frozen stratum. The pipe is then pulled up, the hole plugged and the operation repeated at the next hole. In this way a man can thaw out an area 4 to 6 ft. wide and 48 ft. long in one night if the frost is not over 4 ft. deep. Hand Excavation. — On small work it is desirable to allot a given length of trench to each laborer or pair of laborers, or on larger work to a squad of laborers, so as to furnish each man or squad with approximately the same amount of work for a half day, or for a full day in very deep trenches. This gives the foreman an opportunity to distinguish between the good and poor men, and the men themselves are put on a competitive basis. It is desirable to throw all material, not rock, excavated from the trench, on one side of it. That from the first 6 ft. of depth should be cast as far to the side as it can be thrown, so as to leave room nearer the edge of the trench for that from lower depths. Unless the material is plastic or sticky, it is usually possible for the men to throw it from the trench until a depth of about 8 ft. is reached. Then platforms are constructed about 6 ft. from the surface, supported on the cross braces if the trench is sheeted or on temporary braces otherwise. Such a platform is made long enough to receive material from an excavator at either end, and one staging man can keep the platform clear. Where a trench is wide the platform is sometimes made narrow so that material can be thrown on it at any point along its length, but in narrow trenches the platform extends across the entire trench. When the trench exceeds 14 ft. in depth another set of platforms must be used, placed about 12 ft. below the surface. The staging man on this platform casts the material to the higher one. Square- pointed shovels should be used on the platforms and round- PRELIMINARY INVESTIGATIONS 193 pointed shovels for work in the bottom of the trench. After a depth of about 5 ft. is reached it is usually necessary to provide one man on the bank for each two in the bottom of the trench, to overcast the material thrown out so that none of it shall remain closer than 2 ft. to the edge of the trench. This space is necessary to allow the workmen to walk safely and handle materials, and to reduce the risk to the men in the bottom of the trench from stones rolling back and falling into the trench. Trenching Machinery.— There are two general classes of trench machinery, one for excavating and the other solely for handhng excavated material. The former is usually termed ditcher or trencher and the latter trench machine. The ditcher first came into general use where the soil and earth were particularly favorable for such equipment and close sheeting and heavy bracing were unnecessary. Later, however, they were used to advantage in shale and hardpan. Pipe lines across trenches interfere with their use, but in some cases it may be more economical to run the risk of occasionally tearing out such pipes with the machine and replace them later than to forego its use. The impression seems to be that as a rule such machines do not give a substantial saving unless the trench is at least 6 ft. deep or labor expensive. The increased cost of labor since the war, has materially enlarged the advantageous limits of use of such machinery. The trench machine makes comparatively little difference in the amount of labor employed in picking and shoveling in the bottom of the trench. To arrive at the comparative cost of all- hand excavation and excavation with such a machine, it is only necessary to compare the cost of staging and overcasting the excavated material and shoveling it back into the trench, with the cost of operating a machine, including labor, fuel, repairs, rental and all incidentals. When these two charges balance, it may be wise to adopt some kind of trench machine as it is much more convenient than hand work, does not usually obstruct a street so much, reduces the time required to perform the work, and decreases the number of men required. In studying the practicability of machinery in sewer trenching, careful consideration should always be given to the probable quantity of work to be done. Formerly it was not uncommon to find too much plant, with resulting unnecessary costs, but the effect of the World War on the wages of labor has modified 13 194 SEWERAGE AND SEWAGE DISPOSAL the conditions that existed prior to 1915. In any case, it will probably be inadvisable to use a machine on a short piece of work, unless it is already at hand, for the cost of unloading and erecting it, taking it down when the work is done and reloading it on the car, together with freight charges, is so great that the machine may prove a source of expense instead of economy. The ditchers or trenchers are usually operated by engines mounted upon wheels. The digging is done by buckets carried by an endless chain running over sprocket wheels on the end of an arm suspended from the rear of the machine and is adjusted Fig. 102. — Austin trencher excavating a trench 36 in. wide and 15 ft. deep. to permit the excavation of the trench to a precise grade regard- less of the inequalities of the surface over which it passes. The width of the trench excavated is governed by the width of the buckets selected, each machine being supplied with buckets of several widths though some increase in width can be obtained by barring off additional material. The material excavated by the buckets may be dropped by an endless belt operating crossways of the ditch and discharging the excavated material in a windrow on either side of the trench or into wagons to be hauled to the rear of the work for the backfilling or to be wasted. The Austin trencher is illustrated by Fig. 102. Another somewhat similar type of trencher, the Buckeye traction ditcher, has the buckets PRELIMINARY INVESTIGATIONS 195 attached to a large breast wheel instead of an endless chain, Fig. 103. The trench shown in the illustration was 32 in. wide and 12 ft. deep. Steam shovels, although not particularly well adapted to ordinary sewer construction, have been used with more or less success upon some large work. For this purpose, it is usually desirable to provide an unusually high crane or a long dipper arm, or both, although a shovel of standard dimen- sions may be used for first cuts in the trenches. r ■> Fio. 103. — Buckeye traction ditcher. A common type of trench machine is the Carson, Figs. 104 and 105, designed by Howard A. Carson for sewer trenches and first used upon work in the vicinity of Boston, Mass. It is particu- larly well adapted to this class of excavation. Another type of trench machine is the cableway which is used frequently on many classes of work. A large cableway of the ordinary type is not particularly well adapted for sewer trenching. For this work, light towers which can be easily moved should be provided and the cableway should be relatively short, usually 1% SEWERAGE AND SEWAGE DISPOSAL PRELIMINARY INVESTIGATIONS 197 not over 300 ft. in length, Fig. 106. Only one bucket is handled at a time and this has a capacity varying according to the require- ments of the work from }i to l}i cu. yd. As the cableway is relatively high above the ground and is supported and steadied 108 SEWERAGE AND SEWAGE DISPOSAL only at the ends, 300 to 1,000 ft. apart, it is difficult to prevent the buckets from swinging and occasionally knocking out the timbering. Rock Excavation. — The method to be adopted for the excava- tion of rock in sewer trenches depends upon the quahty and quantity of rock to be excavated. Ledge varies greatly in character, from that which may be removed by picking and shoveUng as readily as some kinds of earth, to rock which requires drilhng and blasting for removal. Some kinds may be removed by picking, barring and wedging without the use of explosives, and some lie in nearly horizontal strata and may be removed by methods similar to those employed in the ordinary quarrying of building stone. In some cases the quantity to be removed is so small or the danger of damaging adjacent structures may be so great as to warrant its removal by picking, barring and wedging even where the nature and quality of rock are such as to warrant drilhng and blasting under other conditions. The methods commonly used for drilling ledge are hand drilhng and power drilling. Hand drilling in turn may be subdivided into three classes, single hand, double hand and churn drilling. Power drilling may be classified in accordance with the type of machine drill used, which may be a rotary or core drill, a recip- rocating or percussion drill or a hammer drill. Hand Drilling. — For smaU holes and occasionally for large ones of moderate depth, the drill may be held in one hand and struck by a hammer held in the other hand, the hammer \^eighing about 4 lb. This method is largely used in quarrying, where a number of small holes are drilled in a line along which the rock is to be split by driving small wedges into the holes between tapered strips of steel. The wedges and strips of steel are called "plugs and feathers." For larger and deeper holes it is customary to have one man hold and turn the drill, while two or three men strike the head of the drill in rotation with hammers weighing from 8 to 12 lb. In this method, the driUs should be turned about 45 deg. each time in order to insure a nearly round hole. In churn drilling the hole is drilled by the churning of a steel drill, which is raised by hand and allowed to drop of its own weight, or it may be thrust into the hole, although but httle advantage is gained thereby. Each time the drill is raised, it is turned about one-eighth of a revolution so as to prevent its PRELIMINARY INVESTIGATIONS 199 striking the bottom of the hole successively in the same position. By confining the turns to about 45 deg., the hole is maintained more nearly circular than would be the case if a larger turn were given each time. Where added weight is desirable, a ball of iron or steel is sometimes welded to the driU rod, at or slightly below its middle point, care being taken to have the center of gravity of the ball in the line of the axis of the rod. The bit ordinarily used in double hand drilling is provided with a chisel edge and is made from octagonal steel }i in. in diameter. The wedge-shaped edge varies in sharpness according to the quality of rock to be drilled. It should be made as sharp as possible and yet not break or become rapidly dulled. The angle of the wedge-shaped edge usually varies between 45 deg. for soft rocks and 90 deg. for hard rocks. Bits are usually made up in sets, depending upon the depth of holes to be drilled. The bottom of the hole must be about 1 in. in diameter to receive the ordinary dynamite cartridge. The finishing drill must therefore have a face at least 1 in. in length, and as a matter of fact, it is usually made somewhat longer, in order to allow for wear. Each bit must be slightly smaller than the bit preceding it in order to allow for wear in drilling. A set of four bits will ordinarily vary approximately between 1% in. for the starting bit to 134 in- for the finishing bit. It is necessary to re-sharpen the drills at frequent intervals, not only that they may be reasonably sharp, but that they may be of proper gage. Spacing of Drill Holes. — The spacing of drill holes and the quantity of rock loosened per foot of hole and per pound of dynamite vary greatly with the character of rock to be excavated, the strength of the dynamite used, the width of trench and the necessity of loading lightly to protect buildings and underground structures. Blasting in trenches requires much more powder than blasting on the surface, for it is possible to provide but one face, and that often a narrow one, toward which to blow. Cleaning Holes. — During the process of drilling, water is poured into the holes, which tends to hold the powdered rock in suspension and forms a sludge which must be removed from time to time. This may be done by the use of a scraper or "spoon," made by attaching a semi-cylindrical strip of steel to the end of an iron rod about H or H in. in diameter. The bottom of the cylindrical portion of the spoon should be turned 200 SEWERAGE AND SEWAGE DISPOSAL up much the same as the point of an ordinary tablespoon. Where such a spoon is not available, the broomed end of a round stick of sUghtly smaller diameter than the lower portion of the hole may be dropped into the hole and withdrawn, bringing with it a portion of the sludge, which may be removed by striking it over a nearby object. Drilling by Machinery. — When the depth of holes, hardness of rock or quantity of ledge to be excavated is such that the economical hmitations of hand drilhng are exceeded, drilling machinery should be used. Of the three different types of drilling machines in common use, the rotary or core drill is rarely if ever used for the drilling required on ordinary sewer work, although it may occasionally be used to advantage for making test borings. The rotary drill rotates the bit which cuts or chips the rock at its base. The more common form is the so-called " core drill," by which an annular hole is cut by the bit, thus leaving inside the hollow steel a core of the rock which may afterward be broken off and withdrawn, thus furnishing a sample of the material encountered. The reciprocating drill has been used upon all kinds of rock excavation for many years. It is driven by steam or compressed air and is used upon heavy work where hand or hammer drilling would not prove economical. It consists essentially of a steam cyhnder in which a piston is driven forward and backward by the steam. The bit is attached to the lower end of the piston rod, which is rotated on the upward stroke in order that the hole may be drilled round and true. The bit is made to follow the rock as fast as it is cut away, by moving the entire cylinder toward the rock by means of a feed screw and crank. This type of drill withdraws the bit from the rock at each upward stroke and forces it back against the rock at each downward stroke, whence its old name of percussion drill. There are several types of reciprocating drills, differing as to details of construction. The drills ordinarily used are often distinguished by the nature of the valve controlling the steam or air supply, as shown in Fig. 107. Hammer Drills. — The hammer drill came into general use after 1900. With it the bit remains at the bottom of the hole in contact with the rock, instead of reciprocating with the piston at each stroke, as in the case of the reciprocating drill. The PRELIMINARY INVESTIGATIONS 201 piston, or hammer, strikes the upper end of the bit, which is rotated either by turning the drill by hand or automatically by a mechanical device. Differential Valve Drill , Tappet Valve Dri.U. Butterfly Valve Drill, Fig. 107. — Four types of valves for percussion drills. The advantages of the hammer drill over the reciprocating drill are that it is light, can be readily handled by one man and requires no tripod or other mounting, as shown in Fig. 108. It 202 SEWERAGE AND SEWAGE DISPOSAL is particularly serviceable in narrow trenches where it is difficult to handle the reciprocating tripod drill. Holes may be drilled with this machine in any direction. There is considerable saving in time in changiiig bits, as with the hammer drill it is simply necessary to lift off the machine, take out the drill and replace it with a new one, replacing the machine on the new bit; while with the reciprocating drill it is necessary to crank back the feed screw, withdrawing the bit from the hole, remove the bit from the chuck, place the new bit, and finally turn down the feed screw to bring the bit to the required position. This drill occupies a sort of middle place be- tween the double-hand drill and the reciprocat- ing drill, superseding each under many con- ditions.. The double- hand drill will still be used when there is in- sufficient drilling to be done to warrant provid- ing an air compressor for operating power drills, but where such a ma- chine is available there appears to be Httle field left for the hand drills. Similarly the reciprocating drill has been replaced upon work too heavy for hand drilling and yet within the practical limits of the hammer drill. Where power is available the hammer drill may be used advantageously for block-hohng to break boulders, for trimming smaU outcroppings of ledge, and for following up reciprocating drills where shallow holes are required. Explosives and Blasting. — Dynamite is the explosive now most commonly used on sewer excavation, although gunpowder or black powder, contractor's powder and TNT are sometimes used. The disruptive and shattering effects of an explosion are - -'-a^ "» ::o'; r*- FiQ. 108. — Jaekhammer drill (IngorsoU-Rand) . PRELIMINARY INVESTIGATIONS 203 due to the instantaneous generation in a restricted chamber of large volumes of gases at high temperatures. The active and determining ingredient of dynamite is nitroglycerine. It burns quietly when ignited in the open air, explodes at a temperature of 388°F. and freezes at 41°F. When inhaled or absorbed through the pores of the body, it causes headaches and sickness. Nitroglycerine is a very unstable compound exploding very easily when subjected to detonation or percussion. Being a liquid it is readily absorbed by various substances such as sawdust, wood pulp, etc. and these are therefore used as carriers. True dynamite is a mechanical mixture of nitroglycerine and an inert absorbent. False dynamites are composed of nitroglycerine and absorbent mixtures which themselves enter into the chem- ical reaction causing the explosion and add to the strength and power which would be furnished by the nitroglycerine alone. Freezing of Dynamite. — Several kinds of dynamite freeze at a temperature of 42 to 50°F. Dynamite which is frozen or Miner's. Bradforct. Catasauqua. Fig. 109. — Commercial kettles for thawing dynamite. chilled cannot be detonated easily. The first warning of this condition generally comes when a number of holes miss fire, while others appear to detonate in the ordinary manner. If a careful investigation be made, it will generally be found that the charges which appeared to detonate as usual were far less effec- tive. Yet dynamite while chilled or frozen is exceedingly sensi- tive to friction and sticks should not be broken or cut while in this condition. Probably more blasting accidents occur in using or thawing frozen dynamite, than in any other way. The thawing of dynamite requires special precaution to prevent explosion. Thawing by immersion in water is not only dangerous 204 SEWERAGE AND SEWAGE DISPOSAL but objectionable for the reason that by diminishing the quantity of nitroglycerine, the explosive power is decreased. Thawing by holding over a hot stove, either in the hand, upon a shovel or in a container of any sort is foolhardy and dangerous as the nitro- glycerine which may thus be freed by heat is hable to premature explosion, and there is danger of the explosion of the adjacent stores of dynamite. Thawing kettles, especially designed for this work. Fig. 109, may be procured from dealers in contractors' suppUes. They are ^^^;&k^'^ fJank-M^'n Plan. :5'Z-- sszs^^ssss Plan. Cover with Section - 5ingle- PiQ. 110. — Thawing '''Section' - Double - boxes for dynamite (du Pont). designed for thawing relatively small quantities, as from one to two dozen sticks, and may be used with comparative safety provided the directions, usually accompanying the kettles, are strictly followed. Perhaps the best method of providing a supply of thawed dynamite is to place the dynamite, immediately upon delivery on the work, in a box set in a suitable pit in the ground and en- tirely surrounded with green horse manure, as shown in Fig. 110. PRELIMINARY INVESTIGATIONS 205 Such a thawing box may provide for one, two, or possibly more, cases of dynamite at one time. The sticks may be taken from this pit from time to time as required upon the work, care being taken not to take the dynamite out until it has had ample time for thawing if it was delivered in frozen condition. Dynamite should not be left in such a pit longer than necessary, as damp- ness may in time penetrate and injure it. For larger quantities, regular thaw houses should be provided. The method of heating should be indirect, and such as will under no circumstances permit the temperature to be raised above 100°F. Exploders and Fuses. — For exploding contractors' powder or dynamite, a powerful shock is required, for which purpose either electric fuses or blasting or fulminating caps are provided. The caps are made of copper shells 3^ in. in diameter and about 1 in. long, loaded with a little fulminate of mercury and carefully sealed. The fulminate of mercury occupies only about 3^ in. of the closed end, while the balance of the shell serves to hold the fuse. The electric fuse consists of a special cap containing fulminate of mercury, as in the blasting cap, and in addition, two electric wires projecting through a sulphur plug and connected by a platinum wire which becomes heated to a red heat by the current of electricity used, and thus explodes the fulminate of mercury. In selecting exploders, care should be taken to purchase only those which are amply strong for the dynamite being used. In shallow holes one fuse only will generally be used and that will be placed in the top, or next to top, stick. Where deep holes are used, fuses should be placed about 5 ft. apart. The following instructions for priming dynamite and other high explosive cartridges are given in a bulletin published by the E. I. du Pont de Nemours Co. : " To prime a dynamite or other high explosive cartridge with blasting cap and safety fuse, make a hole in the end of the cartridge after unfolding the paper shell, or in the side of the cartridge near one end, with a small pointed stick about the diameter of a lead pencil. This hole should not be much larger in diameter than the blasting cap, for an air space around it always detracts from the force of the shock that a detonating blasting cap gives the explosive surrounding it. The best results will be obtained if the blasting cap is pointed straight down into the primer cartridge. "When the blasting cap has been put in the end of the cartridge, the paper must be folded carefully about the safety fuse, and tied securely with a piece of string. When it is inserted in the side of the cartridge near the end, the 206 SEWERAGE AND SEWAGE DISPOSAL safety fuse is held in position by tying it to the cartridge with a double loop of string. "If the work is wet, cover the safety fuse where it enters the blasting cap with soap or tallow to prevent water getting into the blasting cap. Oil or thin grease should never be used for this purpose, as they may penetrate the safety fuse, and destroy the efficiency of the powder in it. " The correct way to prime a high explosive cartridge with an electric fuse is to follow the same method as when fuse and blasting cap are used. The common custom of taking one or more loops, or half hitches, around the cartridge with the wires themselves, after inserting the electr c fuse cap in a hole made diagonally in the side of the cartridge near one end is always to be condemned. The principal objection is that the looping of the wires is very likely to break the insulation, causing short circuits, or leakage of cur- rent in wet work. Sometimes the wires themselves are broken." When charging, the sticks of dynamite should be pressed firmly into place with a wooden tamper, having no metal parts. Where the holes are practically dry, it will be found advanta- geous to put a few slits in the side of the dynamite wrappers so that the cartridges may be pressed into the hole in such a way as to fill it completely. If the holes are wet, the cartridges should not be slit unless gelatin dynamite is being used. Care should be taken in pushing the top cartridge into place not to dis- turb or displace the exploder or damage the fuse or the connecting wires. Pumping. — If good joints in a pipe sewer are desired, it is absolutely necessary to have a dry trench. Cement joints must not be submerged or exposed to running water until the cement has acquired its final set. It is undesirable to allow water to come into contact with mortar until the latter has been 24 hr. in place. Diaphragm pumps are most often used in handling water in trenches. Such a pump, operated by hand, will lift from 30 to 50 gal. per minute from a depth of 16 ft. The usual size for such work has a 3-in. suction hose, which costs about as much as the pump and should be cared for attentively, in order to prevent damage by kinking or abrasion, with the attendant danger of failure when most needed. Larger sizes of diaphragm pumps are made for operation by gasoline motors. Good portable kits mounted upon wheels are now available. When the quantity of water to be handled exceeds the capacity of a diaphragm pump a centrifugal pump or steam vacuum pump is generally employed, for a reciprocating pump is likely to have the plungers and valves cut by the grit in the water and PRELIMINARY INVESTIGATIONS 207 is therefore not suitable for such use. "Where the drainage of a trench can be accomphshed through underdrains discharging into a sump which can be kept in service for a considerable time, it may be advisable to install a centrifugal pump at the sump, but the low lift and dirty water to be handled make high efficiency less important than reliability. Where the water is very dirty and the position of the pumps must be changed frequently steam vacuum pumps are now largely used when the diaphragm type has inadequate capacity. Underdrains. — Underdrains, usually constructed with socket and spigot vitrified pipe, are of two classes; construction under- drains put in to facilitate the building of a sewer and generally abandoned after its completion, and permanent underdrains provided with permanent outlets. They are seldom less than 4 in. or more than 8 in. in diameter. As they are laid below the sewer they are likely to receive any leakage from the latter and care should be taken to prevent the discharge of their contents into water which should be kept free from contamination. If a construction underdrain is laid in the bottom of a rock trench care should be taken to reduce the amount of excavation to a minimum. In gravel trenches there is little danger of fine sand entering the underdrain if the trench in which it is laid is backfilled with clean gravel or broken stone. Great care must be taken in laying the underdrains in fine water-bearing sand; here the tile should be carefully surrounded with gravel and sand, the coarser material next the tile. The joints are wrapped with one thickness of cotton cloth in strips 6 in. wide securely tied, to keep the sand out. Where careful precautions of this nature are not taken an underdrain may become completely clogged in a few hours. Even with the greatest care fine sand will sometimes enter an underdrain and where this is likely to occur it is customary to thread a manila rope through the under- drain as it is laid. This rope can be pulled back and forth to stir the sand and allow the water to carry it off. When the rope proves inadequate a chain is sometimes attached to it, which is similarly pulled back and forth until the drain is cleared. If the underdrain is properly constructed little of the sand which causes trouble enters along its length but comes in at the upper or free end. It is difficult to prevent this, as the work of excavation keeps the sand stirred and the water carries it into the underdrain. It is possible to intercept much of the sand 208 SEWERAGE AND SEWAGE DISPOSAL by digging sumps where it can settle, but these greatly delay the laying of the underdrain and it is generally considered better to keep the drain laid close to the excavation, and to keep it free from sand with a rope or chain. Where the water is removed by hand pumps only, a compara- tively short length of trench bottom should be worked at one time and the underdrain should be laid as soon as a portion of the bottom is ready for it. The upper end of the drain should be kept protected all the time by a substantial strainer to keep out the fine sand. One or more hand pumps are used to remove the sand-laden water from this short stretch of trench bottom, while other pumps take care of the water gathering in advance of this portion of the trench. Quantities of Excavation. — The cost of sheeting, bracing and pumping will be about the same whether the trench is of the 12 II ■■ ■ " ^ ■~ "" "* ~* ^ ^ L — ^ J a' ■ ^ ^ ' i ' ^ ■ > • :■ w'-Oih •A ^ )\ X ^ ^ ^ '^ X* ' ^ ^ X ^ ^ ^ -^ 1 ^ y\,' r / X ^ ■s-' ^ ' ■^ rK> X ^ > -^ ** ,^'> 'Y ,^ ^ ^ ' ^ ^ ^ ^ ■ ^ y ■' '' ^ ' ^ ^ 0- x^ ' -^ ' pp '^ c ^ ^ ■ ^ f^ — ' •^ -- ■^ ^. -^ ^ -. J , -' ■ -^ :>A ^ - ^ ^ ^ •* ^ ^ > ■'- ^^ ^ --' ' ^ ^ <>• ■— "■ y '> ^ ^ "- jo < ^ ^ -- ' ^ ,^ ■*■ - != ;;<; rj r-^ '^ ^ -^ ' ■^ -• p5 T k <'■ -- '_^ " ^ — — ' ' - — "* ■^ "" ^ >' ^ -— ^ — -^ -~ — ■ "^ = = 1 1 m It 1 -— ^ 'jr — -" _ ^ — ^2- ^ ^ ^ z s = = - - = = = = - ~ ~ 3 i3 id id ^ iJ _1 _1 J _ __ _ _ _ _ _1 _J _ 1 13-1 5 6 7 8 . 3 10 II II 15 14 15 16 n IS 13 SO" Kax WTd^hof Masonry in Feet-W. Fig. 111. — Theoretical quantity of excavation for trendies 1 to 16 ft. for sewers of 5 to 20 ft. outside diameter. deep, exact width needed or a foot or two wider. The cost of picking, shovehng, backfilhng and hauling surplus material to the dump wiU be closely proportional to the quantity of material actually excavated. Where some types of machinery will be used, the width of the trench must sometimes be wider than necessary for the structure. In hardpan and dry clay, the banks may stand without bracing, thus making it possible to dig the trench of practically the width of the structure to be built in it, while the same sewer in a trench in running sand may require heavy PRELIMINARY INVESTIGATIONS 209 timbering. In rock excavation it is very difficult to excavate exactly to the line theoretically required. To prevent controversy as to the quantity of excavation to be paid for on contract work, it is customary to prescribe certain limits, called "lines of excavation," to which the quantity to be paid for shall be calculated, but the contractor may, if necessary. tmm^ - ■■" ^~ "^ "^ "~ ~ 7 Lt '""m. "^ — ■ n / 21 \ / -A / \ I- W+|?.35--) ^"Ki / / / 20 1 y / / S CJ / <. / / > / 19 c ■^ t- W+fOJsM ^ifc- / / / d f y i / /' / J r / V 18 d / / / / y / o' ^ / / / / / / n / 1-% / / /' / / ^ > '>, / / / y / if^ ?v y / f"- ?' / y y / ,.eV. / / / y y y / / X \^ ■,\ / ^ / / / , i^- / / y J A / / / y / / ,^\ y y y 14- ,0 / / ^ ^ / y !> / ^ y y \ / ', / / y y y y y y y f> y y 13 oV ^, > / ^ y /■ r\ I' y y ^ .^^ y ^/ y < / y ,/ / y / \V y y •^v 'h ■6 y y y y y ■' \*1 y y / ^/ y y y ^ y If" y / ''/ '/ y y / y / y y y y y y r X ^ ^''d / / / / y y y y y y y y y y / '// /, y /\ y. r y y y y y y y (b ^1 0\ /y. y V ,/ ■y i> y y y y y y 9 / V' /'/ y / -> y y y y y / O /. // y y y i y y a , ^''Z ^/ '/. /,' /'; y X ^ y y / /, /// y^ 'y y- •> y' y •^ y y ^ y 7 / '/^*, '■// y, < <-' v. y y y ^ '/f w / <', > /. y^ y y X ■' n i y/. <'- -> ^ ■>' _j _ ^ 5 S 6 10 II 12 15 14 15 16 17 18 13 20 Max. Width of Masonry, Feet. I Fig. 112. — Theoretical quantity of excavation for trenches from 13 to 28 ft. deep for sewers of 5 to 20 ft. outside diameter. excavate beyond these and if the structure can be built in nar- rower excavation he is not required to excavate the full width shown upon the drawings. Some engineers provide in the contract that earth excavation shall be paid for per linear foot of sewer construction. This is an excellent method, provided the preliminary engineering has been done with sufficient care 14 210 SEWERAGE AND SEWAGE DISPOSAL and thoroughness so that there is no possibility of a change in plans. 5 6 7 8 9 10 II IZ 13 14 15 16 17 18 19 ZO Max.Wi'dth of Masonry, in Feet-W. FiQ. 113. — Theoretical quantity of excavation for trenches 25 to 40 ft. deep, for sewers of 6 to 20 ft. outside diameter. The quantity of excavation required for trenches for masonry sewers from 1 to 40 ft. in depth and up to 20 ft. in width may be obtained from Figs. Ill, 112 and 113. These diagrams are PRELIMINARY INVESTIGATIONS 211 based upon the assumption that the top set of sheeting will be 11 or 12 ft., the bottom set 16 ft., and the intermediate set 12 ft., that 2-in. sheeting will be used, and that 10 by 10-in. rangers and braces will be employed. The quantities of excavation obtained by the use of these diagrams should be regarded as approximate only, for special calculations wiU be needed for any case where exact quantities are required. The averages of actual cross- sections of about 25 sewer trenches in Louisville showed that the excess excavation for small sewers having an extreme outside width of 6 ft. was nearly 33 per cent, whereas the excess excava- tion for sewers from 12 to 16 ft. wide, outside, was 1 to 5 per cent, which indicates the decreasing importance of excess width as the width of the sewer increases. CHAPTER VII SHEETING AND BRACING OF TRENCHES Shoring Trenches.- — Trenches may be excavated in most materials to a depth of 3 to 6 ft. without much danger from caving banks. In some materials, as hardpan, excavations may- be carried down to greater depths, 12 to 20 ft., without sheeting and without much danger if the trench is not kept open long. Usually, however, it is necessary to shore the sides of trenches thoroughly for the protection of the workmen and to make progress possible. It is well to have a trench so protected as to be safe at all times, and to anticipate storms which may cut away the sides unless they are well shored, and other causes of failure. Under some conditions it is desirable to place low banks of earth along the sides and across the ends of the trench to pre- vent surface water from flowing into it and from cutting away the earth behind the sheeting. The method of shoring trenches is usually left to the judgment of a foreman in charge of the work. Good judgment based on wide experience is essential to determine when shoring is needed and what system shall be adopted. There are no rules by which an inexperienced man can determine the depth to which a trench can be excavated before sheeting should be put in place. In many cases the work is inadequately done, resulting in delays and unnecessary expense. On the other hand, it is not unusual to find an uneducated foreman or superintendent, with many years of practical experience, who is capable of rendering a better judgment upon this subject than that of an educated engineer without wide experience in this subject. This is largely because the practical application of theoretical principles of earth pres- sures involves many assumptions which can only be made cor- rectly by an experienced engineer. It is for this reason that the young engineer should cultivate the habit of observing the slopes taken by all exposed materials, the methods of bracing used in all trenches and excavations he sees, and the behavior of this bracing. Trenches are shored by placing braces across them to hold 212 SHEETING AND BRACING OF TRENCHES 213 their sides in place. This often involves sheathing the sides with planks, called "sheeting" when so used, supported by- longitudinal timbers called "rangers" or "wales," which are held by "cross-braces." Sometimes the sheeting is used ver- tically, sometimes it is placed horizontally and is then often called "box sheeting," and sometimes special forms of shoring are employed which will be described later. Behavior of Different Materials in Excavation. — ^Little or no shoring may be required in trenches in hardpan. It usually caves in large masses so that an occasional brace is enough to hold it. While new clay banks stand fairly well for a time after excava- tion, this material is Uable to swell and in many cases water will gradually seep toward the trench, which makes it such a treach- erous material that it is desirable to provide close sheeting and heavy timbers. In dry alluvium sheeting will generally be required, although the banks usually stand well for a time. Occasionally it may be possible to leave spaces between the planks, for the material is not Ukely to ravel. Close sheeting is almost always requii-ed in trenches in gravel because of the tendency of this material to ravel and roll through small openings. It is usually necessary to shore trenches in gravel when they are 3 or 4 ft. in depth. Sand behaves in different ways, depending upon its size, imiformity and the moisture in it. Close sheeting is generally necessary and should be placed as soon as the trench is down 3 to 5 ft. Where the sand is fine and contains considerable water, high pressures are hkely to be exerted against the shoring, approaching and perhaps exceeding a corresponding hydrostatic pressm-e. In such material the sheeting must be close, strongly braced and driven a considerable distance below the bottom of the trench to help hold the planks in their proper position below the lowest ranger as well as to prevent the sand in the trench from being forced up by the pressure of the soil outside the sheeting. The intelligent handhng of quicksand is a highly special problem. The following definition of quicksand was given' by Col. Charles R. Gow, a contractor of wide experience in building foundations and the handling of difficult work: 1. "Quicksand, Its Nature, Behavior and Methods of Control," Ptoc. N. E. Water Works Assoc, vol xxxiv, 171; September, 1920. Fig. 114 has been reproduced from this paper by courtesy of the New England Water Works Association, 214 SEWERAGE AND SEWAGE DISPOSAL "Any material of a granular nature may become a quicksand if there is an upward movement of ground water through it, of sufficient velocity to lift and carry the individual particles. On the other hand, no such material will become 'quick' in character unless there is such a flow of water." Another definition of quicksand was given by Allen Hazen, in a discussion of the subject before the American Society of Civil Engineers, as follows: "Sand containing for the time more water than would normally be con- tained in its voids, and, therefore, with its grains held a little distance apart so that they flow upon each other readily." Fig. 114, rearranged from diagrams in the paper by Col. Gow just quoted, illustrates methods of deaHng with quicksand. While trenches in rock may sometimes be excavated safely without shoring, it should be kept in mind that rock is treacherous material if seamy and the seams he at an angle exceeding 30 deg. with the horizontal. This is particularly true when the seams contain clay or other material on which the rock will slide easily. It is often difficult to determine how far back from the face of the trench the rock has been cracked in blasting, and cracks are likely to permit large masses to slide into an unshored trench. When the banks are very treacherous it may be wise to provide close sheeting, but usually where the trenches are of moderate depth frequent stay-braces or skeleton sheeting, described later, will be sufficient. Blasting makes shoring troublesome to maintain, and good judgment is therefore necessary in directing the blasting. The methods ordinarily used for calculating earth pressures disregard the cohesion of the particles of soil. It is this property which makes it possible to excavate a trench to a considerable depth without shoring, and the constructing engineer should study it well. A given soil will often show different cohesive qualities under different pressures. These qualities will increase with the increase in depth of the excavation, but an increase in the water content of the soil is likely to be encountered as the trench is carried down which will tend to reduce the cohesion. These assumptions should not be carried too far, however, as soils are rarely of uniform character from the top to the bottom of the trench. The adhesive qualities of soil are usually small. If a trench has at some previous time been excavated parallel and near SHEETING AND BRACING OF TRENCHES 215 to a new trench, the soil, if allowed to yield, will usually break away to the line of the first trench, showing the lack of adhesion .-. lU between the original soil and the refilled material. It is often necessary to sheet trenches solely because at some previous time the soil near them has been excavated, the cohesion broken and 216 SEWERAGE AND SEWAGE DISPOSAL no adhesion established since then. For this reason the bracing and shoring of trenches in virgin soil is often much simpler than the work needed for similar trenches in city streets that have been excavated frequently. The time element is important in shoring. Usually the sides of an excavation will be found to exert no marked lateral pressure when first cut. Gradually, sometimes quite rapidly, the soil changes in character, ravels or flakes, and cracks appear in the surface. Sudden and great changes may be caused by rains. Leakage from water mains may soften the material. Voids behihd the sheeting may cause a gradual loosening of the soil under the vibration of passing vehicles or when heavily loaded with excavated material. Movements of the earth due to any of these causes result in changed and often greatly increased pres- sures on the sheeting, and may also change the point of apphca- tion of the resultant earth pressure upon it. Earth Pressures. — Earth pressures are usually calculated by Rankine's formula, which takes no account of cohesion, generally an important property of most soils. Consequently in applying the pressures calculated by this formula to practical problems they are usually divided by 2, 3 or 4; according to the engineer's judgment. The formula is: „ wh^ „ cos — \/cos^ 6 — cos^ Notation: w = weight of a unit volume of earth. P = resultant pressure of the mass of earth against a vertical surface. ■ H = horizontal component of P. = angle of repose of a particular earth. d = angle which the surface of the ground makes with the horizontal. h = depth of any point on the vertical surface resisting the earth pressure. p = intensity of the horizontal pressure at any depth, h, below the surface. In evolving this formula it is assumed that the plane against which the lateral pressure is exerted, or the sheeting, is vertical and that the resultant earth pressiu-e, P, is parallel to the surface of the ground. SHEETING AND BRACING OF TRENCHES 217 Ordinarily in cases where the earth is sloped away and up from the vertical plane it may be assumed that the angle 9 between the surface of the ground and the horizontal plane is equal to the angle of repose cj>, in which case the formula becomes, P = —^ cos If there is no surcharge and the surface is horizontal 6 will equal zero, and the formula becomes, 2 1 + sm <^ In all cases if it is assumed that the direction of the resultant pressure P is parallel to the surface of the ground H = P cosd If the angle of repose is reduced to zero, as in the case of a liquid, the formula becomes, P = ff = ^' In all cases P varies as the square of the height of the bank, as in the case of a liquid, and it has therefore been assumed by most engineers that the resultant earth pressure, P, is apphed to the sheeting at a point one-third the height from the bottom. The intensity of pressure at any point may be found by the formula, 2P It is generally assumed in calculations of this kind that the angle of repose is 30 deg. The formula then becomes, P = H=^ and wh These last formulas are simple and for practical purposes are all that need be used to obtain a general idea of the forces which sheeting and bracing are likely to have to withstand. If an additional load is placed upon the banks, other than that due to earth cast out of the trench which, if sloped back at an angle equal to the angle of repose, or say 30 deg., can be cared 1 Values of the last factor are given in Table 27. 218 SEWERAGE AND SEWAGE DISPOSAL for by the formula, the weight of such additional material should be calculated and reduced to an equivalent depth of superimposed earth, thus correspondingly increasing H, in which case the surface should be assumed to be horizontal. Table 28 will be useful in solving problems. Due consideration should also be given to the possibility of vibrations caused by passing vehicles affecting the pressures. Table 27. — A>fGLE or Repose and Weight of Different Kinds of Eabth Kind of earth Angle of repose Slope' 1 — sin0 1 + sin0 Weight, lb. per cu. ft. Alluvium., Clay, dry. . . Clay, damp. Clay, wet. . . Gravel, coarse Gravel, graded sizes. Loam, dry Loam, moist Loam, saturated. Sand, dry Sand, moist. ... Sand, saturated. 18° 26° 45° 15° 30° 40° 40° 45° 30° 35° 40° 30° 3 to 1 2 to 1 1 to 1 3.2 to 1 1.7 to 1 1.2 to 1 1.2 to 1 Itol 1.7 to 1 1.4 to 1 1.2 to 1 1.7 to 1 0.53 0.39 0.17 0.59 0.33 0.22 0.22 0.17 0.33 0.27 0.22 0.33 90 110 120 130 110 120 80 90 110 100 110 120 * Batio of horizontal to vertical dimension. Occasionally water will accumulate back of sheeting so that it may influence the pressure, but no definite information about such pressures has been obtained. It is known from repeated observations made in trenches in very wet, fine sand, which runs freely through even minute cracks in the sheeting, that if this sheeting remains in place the running sand rarely rises more than a few feet above the bottom of the trench. This indicates that the material is not free to act hke a fluid, else the trench would become filled with it. Consequently it is generally con- sidered that the maximum pressure which sheeting will be called upon to withstand is hydrostatic pressure. The angle of repose of many different earths has been deter- mined, but the reported values are of only general applicabihty because of the absence of any method of describing the character- SHEETING AND BRACINa OF TRENCHES 219 istics of earth so definitely that exactly what material is meant by a description is clearly known. Rough average values are given in Table 27, taken from Baker's "Treatise on Masonry Construction," supplemented with a column of factors for use Table 28. — Intensity op Horizontal Prbssuee and Total Horizon- tal Pressure of Earth at Various Depths («; = 100 pounds per cubic foot; = 30°. ff and Pare per foot of trench) Horizontal surface Surcharges: => 30° Depth h p = iwh pounds H = p = Depth h p cos 9 = i wh B = P COS e ■= i wh' ft. pounds ft. pounds pounds 1 33 17 1 75 38 2 67 67 2 150 150 3 100 150 3 225 338 4 133 267 4 300 600 5 167 417 5 375 938 6 200 600 6 450 1,350 7 233 817 7 625 1,838 8 267 1,067 8 600 2,400 9 300 1,350 9 675 3,038 10 333 1,667 10 750 3,750 11 367 2,017 11 825 4,538 12 400 2,400 12 900 5,400 13 433 2,817 13 975 6,338 14 467 3,267 14 1,050 7,350 15 500 3,750 15 1,125 8,438 16 533 4,267 16 1,200 9,600 17 567 4,817 17 1,275 10,838 18 600 5,400 18 1,350 12,150 19 633 6,017 19 1,425 13,538 20 667 6,667 20 1,500 15,000 21 700 7,350 21 1,575 16,538 22 733 8,067 22 1,650 18,150 23 767 8,817 23 1,725 19,838 24 800 9,600 24 1,800 21,600 25 833 10,417 25 1,875 23,438 26 867 11,267 26 1,950 25,350 27 900 12,150 27 2,025 27,338 28 933 13,067 28 2,100 29,400 29 967 14,017 29 2,175 31,538 30 1,000 15,000 30 2,250 33,750 220 SEWERAGE AND SEWAGE DISPOSAL in the Rankine formula for pressures when the ground surface is horizontal. Vertical Sheeting. — Vertical sheeting, although usually more expensive where other methods can be adopted with equal safety, is the type most commonly used on sewer work and par- ticularly on deep and difficult work, and in trenches in dry running sands, gravels, and quicksand. Where trenches are excavated under conditions making it important to prevent settlement of pipes and other underground structures, pave- ments, and abutting buildings, it is often unwise to attempt to use other methods of shoring. It is desirable to excavate the trench as deep as possible without endangering the banks before placing the sheeting. When the trench is excavated to this depth the first set of rangers (which should usually be about 1 ft. below the surface of the ground) and, if possible, the second set, are put in position against planks placed vertically against the properly trimmed banks at each end and in the center of each ranger. The end planks should be so placed as to come flush with the ends of the rangers. Hav- ing placed the rangers in position and supported them tempo- rarily, the cross-braces are driven to hold the two opposite rangers in place. Braces should be cut so as to require reasonably hard driving and to set the planks back sKghtly into the bank, as otherwise they may become loose and ineffective. The banks should then be trimmed plumb and to a line, which will require the rest of the sheeting to be driven Hghtly, to get it down between the banks and the rangers to the bottom of the excava- tion thus far made. Some sewer builders maintain that it is better to hold the top and second sets of rangers temporarily in place until all the sheeting planks have been placed behind them against the bank, and then drive the braces, thus securing a somewhat tighter hold on the banks than where the planks are sKpped in behind the rangers after the latter have been finally braced. The former method will generally prove more con- venient and will give equally good results if care is taken to so trim the banks that each plank will have a driving fit. The second set of rangers should be placed as soon as the first set has been secured in position, or better, if conditions will warrant, at the same time that the first set is placed. The ver- tical distance between the several sets of rangers will depend to some extent upon the local conditions and the character of SHEETING AND BRACING OF TRENCHES 221 material, but generally, upon work of moderate size, they should be spaced about 4 ft. on centers. The rangers should be directly over one another, thus allowing the sheeting to be driven plumb. After the rangers are securely braced, lap blocks or cleats made of 2-in. lumber the width of the braces and about 18 in. in length, should be placed on top of and across the joints between the ends of the braces and the rangers and securely nailed to the braces to prevent them from falHng if they become loosened. In some cases it may be convenient to attach the cleats to the braces before the latter are driven, thus supporting them while they are being placed between the rangers and driven into their final position. The rangers should generally be placed level, although in side-hill work where the excavation is hght and requires the use of only one or two sets of rangers, it may be advantageous to place them parallel to the surface of the street. This should never be done upon deep excavations or where the trenches are in treacherous material. Rangers will vary in length according to the character of the work being done and particularly the class of machinery used upon the excavation. Common practice, especially in the eastern part of the country, is to make the rangers 16 ft. long, thus fixing the spacing of the braces at a trifle less than 8 ft. on centers, which allows for the convenient use of certain types of trench machines. Upon heavy work it may be advisable to use somewhat longer rangers, thus giving more room between braces for the operation of excavating machinery. It wiU generally be found advisable to use three braces upon each set of rangers, one at the center and one at each end, although some practical sewer builders use but a single brace to hold the ends of two abutting sets of rangers, a 2-in. block being used at the end of the brace to cover the joint between the two rangers. By the former method each section of trench will act independently of the adjacent sections, whereas if the end braces engage upon the abutting sets of rangers, a slipping of one set is likely to affect the next set. Where two braces are used at the joints between rangers, some saving in lumber may be effected by making them of somewhat thinner stock than the center braces, as for example, by using 4 by 6-in. braces at the ends and 6 by 6-in. braces in the center. After the sheeting has been placed in a section and tapped down to the bottom of the excavation, the driving is commenced. 222 SEWERAGE AND SEWAGE DISPOSAL One man, sometimes called the "plank lower," should always be stationed in the trench to dig out for and guide each plank as it is being driven. He casts the dirt to the center of the trench and where the material is such as to permit, digs slightly in advance of the plank. After the plank has been driven as far as possible (usually in good digging about 2 ft. below the normal depth of the trench before driving began) it is left and the next one is treated in a similar manner. As at the beginning of the driving the sheeting usually stands several feet above the ground, platforms constructed of planks loosely laid upon wooden horses are provided at a suitable height upon which the plank drivers may stand and work to advantage. These horses may be conveniently made about 6 ft. high with horizontal cleats about 18 in. apart, so that the platform may be lowered from cleat to cleat as the planks are driven. In trenches from 3 to 8 ft. in width, it will often be found advantageous to alternate the excavation and the placing and driving of the sheeting between contiguous sections of the trench, as the room in the trench in which the two classes of workmen must carry on their labors is somewhat restricted. This method of procedure involves the excavating of two sections of trench at the same time, thus increasing the length of open trench, which in some cases may be impracticable. On the other hand it often happens, especially in trenches in loose gravel and running sand, that it is necessary to drive planks short distances as the excavation progresses, in which case the driving must be done in connection with the excavation. The position of the bottom of the sheeting relative to the excavation depends largely upon the character of the niaterial being excavated. If the trench is in hardpan or alluvium, the excavation can often be carried considerably below the sheeting without endangering the stability of the banks. In sand and gravel it is necessary to drive the sheeting down at least as low as the bottom of the trench, and in many cases several inches below. In such cases it is always well to have the "plank lower " dig the earth away from the ends of the planks as they are driven, thus making it possible to keep the sheeting at all times somewhat below the general elevation of the bottom of the trench. In wide trenches, where such material is fairly stable, it may be possible to carry the excavation in the central part of the trench from 1 to 2 ft. below the sheeting, care being SHEETING AND BRACING OF TRENCHES 223 taken not to dig too close to the sheeting. This aids the "plank lower" in his work of digging the earth away from the planks as they are being driven. The general method of shoring by means of vertical sheeting, is illustrated by Figs. 115 and 116, which are companion drawings Note ■• /III Braces should be Cleated, to Prevent them from Falling. Fig. 115. — Isometric sketch of typical trench sheeting. of a narrow trench carried to such depth as to require three sets of vertical sheeting. The braces are shown cleated, which is good practice upon work of this kind. Staging platforms are shown extending half way and all the way across the trench, as both methods are commonly used. 224 SEWERAGE AND SEWAGE DISPOSAL The most common practice is to use sheeting planks sharpened with "chisel" edges, placed with the beveled side toward the inside of the trench as shown at A. In some cases, especially in difficult work where it is necessary to drive the sheeting long distances ahead of the excavation, the planks are beveled in two directions, as shown at B, the object being that the planks shall drive straight and that they shall tend to hold tightly against the planks previously driven. Cross Section, w^ Longitudinal Section FiQ. 116. — Typical trench sheeting. In practice the rangers are rarely spaced with exactness or the planks driven to a uniform depth. In many cases lighter rangers are used, as 4 by 6-in. timbers, and occasionally where the excavation is not to be deep, 2-in. planks are used for rangers. Upon extensive work, however, where the depth of excavation will exceed 10 ft. or where the material to be excavated is such as to make it necessary to shore the banks securely, it is wise to use rangers at least as heavy as 4 by 6-in. timbers, and heavier stock as may appear to be necessary from the local conditions en- countered. SHEETING AND BRACING OF TRENCHES 225 As the upper portion of the trench, supported by the first set of sheeting, is the widest, it is wise to make the first set of sheeting the shortest, thus effecting a saving in excavation. Furthermore, the use of relatively short sheeting for the first section avoids the necessity of using horses to any great extent, in connection with the driving of the planks. Stages can be placed upon the rangers in the trench, from which the second and third sets of sheeting can be conveniently driven, even though they are relatively long. In many cases the length of sheeting adopted for the several sets will be governed by the stock of lumber on hand when the work is begun. If, however, new lumber is to be purchased, care should be taken to order only those sizes which, by avoiding waste, will be economical. While very long sheeting is some- times used, it is common practice to Hmit the lengths even of the longer sets to 16-ft. plank. Braces should ordinarily be of the same depth as the rangers. After the first set of sheeting has been driven to the full depth to which it is intended to extend, guard rails made of 2 by 6-in. planks are spiked to the lowest set of braces at a distance from the rangers approximately equivalent to the thickness of the sheeting. The second set of sheeting is then passed through the space between the guard rail and the adjacent ranger and placed in its proper position ready for driving. If the sheeting is more than 10 ft. long, it may be well to place a second guard rail on the second set of braces above the first guard rail. This will hold the upper ends of the planks rigidly and prevent their springing, so that they will be more readily driven than if held simply by the guard rail near the bottom. Additional sets of sheeting should be started in the same manner. It is not possible to place and drive planks in a second set of sheeting immediately under the braces of the first set. There- fore, as the excavation is carried down, a space equivalent to the width of the braces above will be left unshored. This space, sometimes called a "window," is frequently closed by placing short horizontal planks behind the adjacent vertical planks. In cases where the material is Hkely to run, if exposed for a short vertical distance, notched planks are used. These "notches" are usually made of 12-in. planks by cutting out about 6 in. from the upper end of the plank, leaving the lower portion, about 4 ft. in length, the full width of 12 in. These notches are placed under 15 226 SEWERAGE AND SEWAGE DISPOSAL the braces and are driven like the other planks. As they are driven a "window" opens below the braces which must be closed from time to time with horizontal planks set in behind the notches or vertical poling boards fastened to them. Notched planks must be both "rights" and "lefts" so they may be fitted under the double braces from both sides. Box Sheeting. — Horizontal or "box" sheeting is sometimes used where the material excavated is such that the banks will stand untimbered for a considerable length of time without great danger of their caving. With it less care need be taken in trim- ming the banks of the trench plumb than in the case of vertical sheeting, as it is not absolutely necessary, though it is preferable, that each plank be placed directly under the one above. After the excavation has been carried as deep as practicable without bracing (usually the width of three or four planks), the planks are placed along the sides of the trench and held in position by 2-in. cleats placed vertically at each end and usually in the center of the planks. These cleats are in turn held firmly by cross- braces. As the width of the trench between the opposite planks is likely to vary, extension braces are to be preferred to the ordinary driven brace, as they are more readily adjusted and by their use the trench may be more quickly braced. The extension braces are also preferable, as there is little or no jar in connection with placing them, while if driven braces are used the jar resulting from the driving may be sufficient to endanger the banks. Box sheeting has been used in some cases where the material ex- cavated was very unstable, such as coarse gravel and dry sand. In such cases the excavation can be carried only deep enough to accommodate a single plank at a time. Each set of planks is temporarily braced until the trench has been carried deep enough to accommodate uprights supporting three or four planks, after which the uprights, with one or two braces against each pair, may be substituted for the individual plank braces. In this class of material box sheeting may be more expensive and troublesome than vertical sheeting, but it is quite useful under certain difficult conditions, as for example in trenches between car tracks or under structures which interfere with the placing of vertical sheeting. The use of the wooden horses and temporary stages required by plank drivers is also avoided by the use of box sheeting, which is often advantageous, especially in narrow streets. SHEETING AND BRACING OF TRENCHES 227 Sometimes the first few feet of excavation may be in dry clay, alluvium or other material which will stand fairly well, while beneath this stratum sand or gravel may be encountered. In such cases a combination of horizontal sheeting on top and vertical sheeting below, may be used with success. Fig. 117 shows a trench about 12 ft. in depth, the banks of which were supported by two sets of box sheeting, each three planks in depth. The planks used were 2 in. thick, 12 in. wide, TRjff-y-- ■■■fo Foling Soared 117. — Box sheeting for 45-in. sewer. FiQ. 118. — Poling boards and vertical sheet- ing for sewer 7 ft. 6 in. by 6 ft. 8 in. and 18 ft. in length. The cleats were 2 by 12 in. and 3 ft. in length. Extension braces were used,- as shown in the sketch. The material excavated was a dry alluvial clay. Poling Boards. — This method of timbering is similar to that of vertical sheeting, except that short planks 3 or 4 ft. in length called "poling boards" are used and the sheeting is placed against the banks from time to time as the excavating proceeds instead of being driven. The planks are placed vertically against the properly trimmed banks, and are supported by means of rangers and braces, as in the case of vertical sheeting. In some cases where the banks stand well it is possible to get along 228 SEWERAGE AND SEWAGE DISPOSAL 6 ' \5heet/ng ■Sheeting with a single set of rangers and braces for one set of poling boards, as illustrated in Fig. 118. This, however, is not as safe as the method shown in Fig. 119, where two sets of rangers and braces are used for each set of poUng boards. The nature of the material in which the trench illustrated by Fig. 118 was excavated was such that a space of from 6 to 12 in. could be safely left unsupported between the bottom of one set of poling boards and the top of the set next below. Where this method of %fement sheeting can be used, econ- omy in excavation will re- sult, because the trench will be of the same width from top to bottom, regardless of depth. Thus with a trench sufficiently deep to require three sets of vertical sheet- ing, a saving of about 3 ft. in width at the top and about l}^ ft. opposite the second set of sheeting, will result from the use of poling boards. Short pieces of plank which accumulate on sewer work requiring some vertical sheeting can be advantageously used for pol- ing boards, so that a com- bination of poling boards and vertical sheeting often proves economical. The use of poling boards has the same advantages as the use of box sheeting, in that they do not project above the surface of the ground at any time. The chief disadvantage in the use of poling boards lies in the fact that if there is a collapse in the timbering of any portion of the excavated section, resulting from the running out of the material behind the sheeting, serious deformation or even collapse of the entire timbering of the section may result. The danger of this is much greater with poUng boards than with vertical sheeting. dheefing Fig. 119. — Poling boards and vertical sheet- ing for 7-ft. sewer. SHEETING AND BRACING OF TRENCHES 229 Stay Bracing. — When the excavation is in hardpan, dry clay or dry loam, it is often unnecessary to provide close sheeting, although some support for the banks may be needed. In such cases stay bracing, or skeleton sheeting, will prove economical. The simplest form of stay bracing involves the placing of pairs of vertical planks braced tightly against the banks of the trench at intervals of 6 to 10 ft. The braces may conveniently serve as supports for staging platforms in trenches which are sufficiently deep to require that the material be staged out. If a trench must be left open for several days, the use of stay bracing is not to be Fig. 120. — Trench in alluvium; stay bracing. recommended unless conditions are especially favorable, as heavy rains are likely to soften the banks and cause expensive and dangerous caving. Fig. 120 clearly shows the method of supporting the banks of a trench by means of stay braces. In this case, the pairs of planks were supported by single braces, some of which were 6 by 6-in. timbers, and others consisted of two 2 by 6-in. planks. Skeleton Sheeting. — In some cases where it is difficult to determine prior to the completion of the excavation, whether or not it will be necessary to use vertical sheeting, skeleton sheet- ing is used. This consists of the usual rangers and braces with planks between the rangers and the banks of the trench opposite 230 SEWERAGE AND SEWAGE DISPOSAL the braces. By putting in these frames, it is possible to slip planks in later without delay wherever found necessary. It is often desirable to place such planks at intervals of 2 or 3 ft. where it is unnecessary to provide close sheeting. Bracing Curves. — When a sewer is to be constructed around a curve, the trench is generally laid out on chords of a length de- pendent upon the degree of curvature. The ends of the rangers are cut on radial lines, thus making the outer rangers longer than the inner rangers. It is desirable to have the braces nearly at right angles to the rangers but where the curve is very sharp this is not always practicable. Whenever the braces are not placed substantially at right angles to the rangers, cleats or short planks should be securely spiked against the rangers on both sides of each brace to prevent it from sHpping. In some cases where the banks of the trench are not very heavy, it maj be possible to excavate the trench on the line of the curve and, instead of heavy timbers, one or two thicknesses of 2-in, planks bent to the curve may be used for rangers. In such cases it will usually be helpful to scarf with a saw the inner side of each plank at intervals of 4 to 8 in. to a depth of from Ji to 3^ in. to aid it in taking the curve without cracking or breaking, the inner side referred to being the side of shorter radius of curvature. Jetting. — Where sheeting must be driven much ahead of excavation, it is occasionally found impossible to make sub- stantial progress in the driving of sheeting by ordinary methods. In such cases a jet of water may be forced into the soil just below the bottom of the plank so that as the soil becomes softened the plank can be driven with comparative ease. In some cases the bottom of each plank is fitted with a special appliance by means of which the jet of water is introduced into the soil immediately below the plank. In other cases the water may be conveyed to the lower end of the sheeting by a ^- or 1-in. pipe, independent of the sheeting itself, and pushed down by hand. Where the planks are fitted with special jetting shoes, it is economical to drive them a considerable distance below the bottom of the trench at one operation. Practical Suggestions on Timbering. — Good workmanship is essential to the successful timbering of trenches. If it is desired to practice strict economy, it is wiser to adopt a less expensive method and to do the work in a first class manner than to adopt a more expensive method and resort to careless workmanship, SHEETING AND BRACINO OF TRENCHES 231 to a 1 o a J 9 ■3 a 1 sjuuq JO . A M ^ • CO iH -«J< • N CO Tt( Jiu^q JO 3.12 3.34 4.68 ? a §1 S Iz; 3 • CD CO ^ • «' OJ rt ■ CR 01 •* ■ rH iB S ° S O »0 CO O t- I-H lO CO b- (N CO OC S i| g «= i Si K« III c, ^ O & g i- 1 = < g 5 s O £^ W li, m ■ l> t^ X ; CO CO K: • ei rfl h." • 00 O) U3 • -"l^ lO W ■3 £-3 O t* CO O O CD d d Tt o " m .a °° X X s >,X .5 S -9 2 ^ .9 ^=1 :SS8 : ^' ?S ^ ; o o o •si 1^ 0, 88S §§§ Tf -* « 2S|x-gi2x =0 ^E . O O CD ■ ■* ■^ M • 00 06 00 ■ CO CO CD -^ a 2 S8S gg's? •* Tjt 00 ^ a ^ .a .S § -o .s gSg;oE-'':goo . W t^ T-. • CO CO rH • IV 00 5 O O CO N M CO •-I Ul CO ^ -l Tf N t> !N -* OS CO CO Tt* Em 3[n^q JO •»} -be 18,1 O C OS ^ c^ t> lO lO w N N « « d CO ■* N CO CO W 2 OS (^ O r-l i-H CD ,-t w r-l ,-1 O b- CD O O CO M (N CD 00 o w w ^- t* b- a-S "^ c9 =^ 0, odd a-S « 0. 888 Hi eo CO CD 05 3 §888 g§§2§2 rt (N (N lO a S a" M S8S? -*■«*< 00 (M bO 888 "3 »0 O 0> A ■* co" d CO IQ lO TlH ,-. "5 S J3 N (N CO rH lO CO .-H « t* d M CO 3gg p H 1^ 3[nBq JO ■%} -bs JOJ CO 00 -H Tj< M N CO ■^ o CD 3JUBq JO •?J-bBJ8J 2 2 S iH ^ t- CO CO Tj* JO -^j -bB JOJ S5_8 CO CO •<** lo CO CO m C4 CO 00 U3 N CO oi d ■^(1 ■*■<*< N S S ° a 00 00 X o o 2 (j 14-. -^ t- rt o 2 ID lO M t^ (N W d rn' U3 T|< lO rH —I rH (N O CO V H " 43 O CO CO CO O CO CO M CD ^* r^ 00 3^1 O CD « H -^ ^ OS w d « ■* CO 1^1 8SS ID M U5 CSl ■# ■ f; 2 i 1 CO CO lO CO CO lO CO* U5 lO j3 2 2 w O O CO d 00 CO '■Jt O OS rH CO « a Sdqoai q JO 9ZJS CO ^ ^ 00 X X X X -* CD eo 00 >4 M CO CO CO 00 XXX fD <& (Xi CD CO W XXX CD CO 00 S o opis^no) ^nijeoqs P q^PTM ■qf Tj* w N Tft « M ^ « N rH 232 SEWERAGE AND SEWAGE DISPOSAL The banks should be carefully trimmed to a plane surface before placing the sheeting. As the trench is deepened the a ij o 4: ^1 5^ ■O'S ■ah. Is e material of the bank should not be permitted to run out or ravel or be trimmed back to allow the sheeting, when driven down, to slip over it without friction, inasmuch as this practice, particu- SHEETING AND BRACING dP TRENCHES 233 larly in running sand or gravel, is likely to lead the workmen to carelessness in allowing the formation of pockets behind the sheeting. Such pockets are undesirable not only because they may result in serious settlement of the ground and increase in pressure upon the bracing, but because they are likely to result in deformation of the finished sewer by spreading of the arch when the trench is backfilled. In excavating a trench in loose material, hay or meadow grass will be found useful to prevent the material from running through the cracks in the sheeting. The banks of a trench should never be allowed to move, if it is possible to prevent movement. The sheeting and bracing should be placed in the trench as soon as practicable, in order to prevent any movement of the top layers of the soil. The sheeting should be driven as fast as the excavation proceeds and in quicksand it is, of course, necessary to drive the sheeting in advance of the excavation. Great care should be taken to prevent the banks from raveling where notches or "windows" are required. The pressures exerted upon timbering increase greatly as the soil back of the sheeting becomes cracked, loosened, or in any way shifts its position. Such movements of the banks are likely to cause the breakage of adjacent water pipes and sewers, resulting in the partial or complete saturation of the soil with the water leaking from them, and producing pressures approaching hydro- static pressure. Braces should always be tight. Where they are short and of wood they may be cut to correct length, and driven, but where they are long they should be tightened by hard wood wedges driven to refusal. Upon the lighter work extensible screw braces will be found more convenient than wooden braces. Except upon very heavy and expensive work it will generally be found more convenient to use rangers and braces of the same size from the top to the bottom of the excavation, than to attempt to vary their size in accordance with theory. If, as is frequently the .case, it seems desirable to increase the strength of the timber- ing toward the bottom of the excavation, or at any other point, this can be done by decreasing the space between the rangers, and adding braces. Trenches should always be braced in such a manner as to give as much room as is possible for working between the rangers and between the braces. For this reason, as well as to reduce the 234 SEWERAGE AND SEWAGE DISPOSAL quantity of excavation, it is common practice to use rectangular rangers on edge or with the long side vertical. While this is not the most economical method of utilizing the strength of the tim- ber, it does save room within the trench and makes it possible to construct the trench sHghtly narrower than if the rangers were turned so that the long side would be horizontal. More- over, this method gives greater bearing area for the braces, which may be required as the transverse crushing strength may control the section. It is generally desirable to use two braces at the contiguous ends of two sets of rangers, so that each set may act independ- ently of the other and a settlement of one section of bracing will not affect the adjacent set. Center braces should generally be of the same size as the rangers. End braces should be of the same depth as the rangers but may generally be somewhat nar- rower, because the loads put upon them are smaller than those put upon the center braces. All braces should be securely cleated to the adjacent rangers so that they will not fall to the bottom of the trench in case the banks yield and they become loosened. Lumber and Tools. — Ordinarily the first cost of timber is an important if not the chief consideration in its selection. If sheeting requires hard driving, it is advisable to get either a soft, tough wood or a hard, strong wood. Spruce is very satis- factory and oak stands driving fairly well. Hemlock is usually unsatisfactory because it spUnters. Oak is heavy and likely to warp. Sheeting planks should not be more than 10 in. wide, ex- cept notched planks, nor less than 6 in. It requires about as much labor to drive either a 4-in. or an 8-in. plank; two 8-in. planks can generally be driven in about the same time as one 12-in. plank. Many sewer builders prefer planks 10 in. wide for sheeting. Sheeting is generally square edged, sawed stock. Rangers and cross-braces are generally similar stock, although round tim- bers may be used. Planks for poling boards and box sheeting require no dressing. Rangers should be of uniform length, generally 16 ft., so that the cross-braces may be set directly under each other at each place where they are used. Braces should be beveled along their vertical edges for about 1 in. each way, to faciUtate driving and reduce splitting at the ends. The bevel at the bottom of a sheeting plank should be about 6 in. long, with the edge about 3^ in. thick. The tops of the sheeting planks SHEETING AND BRACING OF TRENCHES 235 should be dressed to receive plank caps, placed over them to protect the ends against brooming under the driving maul. The cast-iron caps should fit loosely on the plank ends or they will be cracked in driving. ^ On account of the rapidly rising price of lumber, steel sheet piling has come into extensive use for heavy engineering founda- tion work. It is not as easily handled, driven or pulled as plank, but its useful hfe is much longer, and hence its economic value has been constantly rising as the price of lumber has increased. A number of special tools are used in trench work. The sheeting is driven with wooden mauls or beetles, made of elm or ,yi ^- Edge Slightly - - — Rounded" \ ^?Ti >o .-ds Co. Akon.Ohio. - B- Canne/lvn SewerFipeCo. Canneltonjnd, ' C = Portland Stoneware Co. Portland, Me. ■ D= American Sewer Pipe Co, Akron, Ohio. ' E-EvensiHoivardFireBrickCo.—StLouis,Mo. ■ F'TheStreaiorClayMfci.Ca. Streator, III. '- G= H. Stevens'Sons Co. Mac,on,6a. y^^ >t'' J ,i' -^ .>^'" . Jy :> / > / ^ ^ A y^ ^} V y v^ fe f / -/ 4 ^ F f' Z ^t V '.< ^F ^ :'o.f ' ' 1 ^ f.^^ , Listed Thickness ■ OF ■ Standard and Double Strength - Vitrified Sewer Pipe. y —p ^ \ ^ 'M- 9' t Standard Shown thus ~ . 3'" 1 1 1 Proc. A. S. T. M., vol. xix, 1919, pp. 543-568. s Proc. A. S. T. M., vol. xx, 1920, pp. 264-265. PIPE, BRICK AND CONCRETE SEWERS 243 Rejection. — Pipes shall be subject to rejection on account of the following: " (a) Variations in any dimension exceeding the permissible variations given in Table 33. " (6) Fracture or cracks passing through the shell or socket, except that a single crack at either end of a pipe not exceeding 2 in. in length or a single fracture in the socket not exceeding 3 in. in width nor 2 in. in length will not be deemed cause for rejection unless these defects exist in more than 5 per cent of the entire shipment or delivery. Table 33. — Permissible Variations in Dimensions OP Clat Sewer Pipe Lengths Internal diameter. Depth of socket, inches (-) Thick- Normal Length, of two inches ness of size, inch per opposite barrel. inches foot (-) sides, Spigot Socket inches inches (±) ( + ) (-) 6 H yi He M M He 8 Vi H Vi He y^ He 10 H H y4. He K He 12 H H He M M He 15 H H He H M %2 18 M He H He Ya, %2 21 ' V4 He He H M M 24 % H yi He H M 27 H K % iHe M H 30 % K H ^He y, H 33 % H H ^He H He 36 % % Vi ^He K He 39 % % Hi "■He y^ He 42 % % M ^He K He Note. — The minus sign ( — ) alone indicates that the plus variation is not limited; the plus and minus sign (±) indicates variation in both excess and deficiency in dimension. " (c) Blisters where the glazing is broken or which exceed 3 in. in diameter, or which project more than H in- above the surface. "(d) Laminations which indicate extended voids in the pipe material. "(e) Fire cracks or hair cracks sufficient to impair the strength, dura- bility or serviceability of the pipe. " (f) Variation of more than H in- per hnear foot in alignment of a pipe intended to be straight. " (g) Glaze which does not fully cover and protect all parts of the shell and ends except those exempted in Section 31; also glaze which is not equal to best salt glaze. " (h) Failure to give a clear ringing sound when placed on end and dry- tapped with a light hammer. " (i) Insecure attachment of branches or spurs. 244 SEWERAGE AND SEWAGE DISPOSAL "Salt Glaze. Section 31. — The glaze shall consist of a continuous layer of bright or semi-bright glass substantially free from coarse blisters and pimples. If present none of these shaU project more than }4 in. above the surrounding surface. Not more than 10 per cent of the inner surface of any pipe barrel shall be bare of glaze except the socket, where it may be entirely absent. Glazing wiU not be required on the outer surface of the barrel at the spigot end for a distance from the end equal to two-thirds the specified depth of socket for the corresponding size of pipe. Where glazing is required there shall be absence of any well-defined net work of crazing lines or hair cracks. All glazing shall be equal to that produced by the best salt-glazing process.'' The specifications of the American Society for Testing Mate- rials Hmit the clay sewer pipe lengths to 2 ft. for pipes 6 in. in diameter; 2, 2}/^ and 3 ft. for 8- to 24-in. diameters, and 3 ft. for 27-in. and larger diameters. But one type of socket and one standard thickness for pipe of each nominal diameter is pre- scribed. The tendency is in the direction of the use of these single standards rather than to the multipHcation of types. The 3-ft. lengths cost a Uttle more than the 2-ft. lengths, but are preferable because of the small number of joints to be made with them. The difference in thickness between the. standard and the double-strength sewer pipe, as made heretofore by differ- ent manufacturers, is shown in Fig. 125, with which the require- ments of the standard specifications of the American Society for Testing Materials (Table 34) can readily be compared. Table 34. — Dimensions of Clay Sewer Pipe A. S. T. M. Inside Internal diam- eter, inches Laying length, feet diameter at mouth of socket, inches! Depth of socket, inches Minimum taper of socket Thickness of barrel, inches Thickness of socket 6 2 8H 2 20 V The thick- 8 2, 2K, 3 lOH 2M 20 K ness of the 10 2, 2i4. 3 13 2H 20 % socket yi 12 2, 2H, 3 15>i 2J^ 20 1 in. from 15 2, 2H, 3 18Ji 2H 20 IK its outer 18 2, 2J^, 3 22K 3 20 IK end shall 21 2, 2M, 3 26 3 20 IH not be 24 2, 2H. 3 29K 3 20 2 less than 27 3 33K 3M 20 2K three- 30 3 37 3H 20 2y2 fourths of 33 3 40".; 4 20 2% the thick- 36 3 44 4 20 2H ness of the 39 3 47K 4 20 2% barrel of 42 3 51 4 20 3 the pipe. 1 When pipes are furnished having an increase in thickness over that given in last column the diameter of socket shall be increased by an amount equal to twice the increase of thi^.k- ness of barrel. PIPE, BRICK AND CONCRETE SEWERS 245 The deep- and wide-socket pipe are favored by some engineers because they give more space for jointing, but the general belief is that the standard sockets, if made true to form so as to give the specified space for joints, are preferable because tighter joints can be made with them. Difficulties with the usual socket are largely due to imperfections in manufacture and inspection which result in jointing spaces being too small at times to allow the gasket and mortar of the joint to be put into place properly. >'' ■* ^v IB94-I903. 1903-I9lt. Fia. 126. — Cradle and joint of Washington pipe sewers. Concrete 'f ^. irV"-.*'^-, ^> Chii'Z^ Concrete I BoarS) Philadelphia. Bronx.. Fig. 127. — Types of cradles. Medford. Vitrified pipe without any socket is used in the sewerage system of the District of Columbia, but not to any extent else- where in the United States. The joints are made by a ring of mortar over the top while the sides and bottom are made tight by the concrete cradle which supports the pipe. Fig. 126 shows two types of cradle and joint used in the District of Columbia, and Fig. 127 shows three types of cradles used with bell and spig(5t pipe. 246 SEWERAGE AND SEWAGE DISPOSAL Concrete Sewer Pipe. — The sand and gravel or broken stone used in cement pipe are generally proportioned to give a mixture of about 1:4 for sizes up to 20 in. to be used under heads exceeding 15 ft., while a 1 :3 mixture is perhaps more used for pipe over 20 in. in diameter. Pipe 10 in. or less in diameter are preferably made without any coarse aggregate, on account of the difficulty of producing a dense, uniform material with coarse aggregate in a thin shell. The coarse aggregate used for larger pipe is from J^ in. to half the thickness of the wall of the pipe. Dieckmann recommends that the fine aggregate should be graded so that not more than 10 per cent will remain on a 10-mesh sieve and not more than 30 per cent pass a 50-mesh sieve. Two methods, dry and slush, have been used in making cement pipe. The mixture used in the dry method contains only enough water to leave web-like markings on the concrete when the forms are removed and to ball up when pressed in the hand. This mixture is tamped into molds and the quality of the pipe depends on the thoroughness of the tamping as well as character of the mixture and the care taken in curing. With the slush method, used for large pipe, the materials are so wet that they will flow into every part of the mold with little tamping, and as a result the pipe must remain in the molds longer than with dry mixtures. Small sizes of pipe are generally made by machines, which are sometimes used for larger sizes. Cement pipe are sometimes made without sockets and are jointed like the straight vitrified clay sewer pipe used in the District of Columbia or by placing pre-cast cement bands and mortar over the joints. If they are without sockets there is a decided tendency to give one end an inside bevel and the other end an outside bevel, in order to make a closer joint than is possible with square ends. Reinforced-concrete sewer pipe have the advantage over rein- forced concrete sewers built in place of doing away with forms in the trench and permitting the manufacture of the pipe under the most favorable conditions to produce a dense, uniform material in which the reinforcement is placed accurately. Laying Sewer Pipe. — Great care must be taken in handling sewer pipe to prevent injury to it. Pipe up to 24 in. in diameter are usually taken from the cars by hand, but where larger sizes must be handled in quantities a guy derrick will probably save time and breakage, although these large pipe can be handled PIPE, BRICK AND CONCRETE SEWERS 247 with the aid of skids by hand if care is taken. If the pipe is placed in storage it should be laid in piles, the sockets of one row alternating with the spigots of the next lower row, and all pipe securely held by blocking, to prevent a run of the pipe in the lower rows. If pipe has to be carried through the winter, which is undesirable, it should be supported so that it cannot become frozen to the ground and that water will not collect in it. The pipe should not be taken from the yard to the street, as a rule, until a day or so before they are needed, owing to the greater danger of breakage at the site of the work. Pipe are generally lowered into the trench by hand but when they are 24 in. in diameter or larger a three-leg derrick which can be set up across the trench is useful in handling them. Where a sHng is desirable in lowering them what is known as a Fig. 128. — Pipe buggy and lifting hook. "hook rope" is often used. This is a rope with a hook spliced in one end and a chain and hook spHced in the other end. The chain is passed through the length of pipe and the hook fastened over the chain at the spigot end of the pipe, thus forming a loop in which the pipe hangs as it is lowered into the trench. After being lowered into the trench the pipe is usually shoved into place by the pipe layer, his helper lifting the spigot end by means of a wire or strap passed under the pipe. If the pipe is so large as to be hard to move in this way a lifting hook or pipe buggy, Fig. 128, may be used. If, when the pipe is in place and tested for line and grade, it is found to be sUghtly too high, it is sometimes possible to force it down to grade by lightly tapping it, but otherwise it must be lifted and just enough earth scraped out to give it a uniform and firm support. The bottom of the trench in earth should be shaped as nearly as possible to receive the bottom 248 SEWERAGE AND SEWAGE DISPOSAL quadrant of the pipe, with a deeper hole under each socket to enable the joint to be made. If the trench is in ledge the rock is generally taken out 6 in. below the invert of the sewer and the space refilled with good earth or gravel. If the pipe is found to be too low when tested it must be raised and earth carefully spread below it so as to bring it to grade when it is again lowered. In most sewer pipe laying three or four men constitute a working gang. One handles the pipe on the bank and lowers it into the trench, another receives it and places it in position, another makes the joint and, with the fourth, carefully tamps the earth under and around the sides and fills the socket hole com- pletely. The fiUing and tamping until the backfill is a foot above the pipe must be done very carefully to avoid disturbing or cracking the pipe. It must proceed equally on each side of the pipe and the tamper should have a face not exceeding 5 by 13^ in. and should not weigh over 7 lb. For work about the sides of the pipe a tamper of even smaller size is useful; it has a face about 4 in. long and ^ in. wide and a handle of ^-in. iron set at an angle so that the filhng under the pipe can be reached easily. Cement Joints. — Most sewer pipe are laid with cement joints. The cement may be used neat or in a 1:1 or 1:2 mortar with sand. The more cement there is the more easily can the joint be filled but the more liable the joint will be to crack. More time is required for a mortar joint to harden than one of neat cement. A 1:2 mixture is most commonly used, except when rapid setting is desired. It is usual to begin a joint by placing a gasket formed of a piece of jute or oakum soaked in cement grout around the end of the spigot of the pipe to be laid. Its purpose is to keep the pipe centered in the socket of the pipe aheady in place and pre- vent the squeezing out of the mortar in the bottom of the joint by the weight of the pipe. If no gasket is used, mortar is spread in the lower part of the socket of the pipe already in place in order to hold the new pipe in place, but this mortar is liable to be forced out to some extent and joints made in this way are not so good as those made with a gasket. After the pipe is in position, a man with a trowel and rubber glove fills the joint carefully with mortar, working it into place with a wooden calking tool. The mortar is usually left with a slope of about 45 deg. from the outer edge of the bell to the pipe with which the joint is made. This is termed "overfiUing" the PIPE, BRICK AND CONCRETE SEWERS 249 joint and is generally believed to give a tighter joint than finish- ing off the mortar flush with the end of the bell. Where unu- sually good joints are required, they are sometimes only partly filled with mortar, then another gasket is rammed into the joint fiush with the end of the socket, and the joint is finally overfilled with mortar. After a joint has been made it is desirable to wrap about it a piece of cheese cloth, securely tied at the top, to hold the mortar in place, for there is a tendency for the mortar to sag away, particularly if it is rather wet, and thus open the joint somewhat. The use of the cheese cloth also permits the use of a rather moist mortar with which a better joint can be made than with a stiff, dry mortar. Except where a gasket is used there is a tendency for mortar to work through the joints into the interior of the pipe, where it will harden into an obstruction to the flow of sewage if not removed. It is therefore desirable to scrape the inner surface of a pipe sewer as it is laid, for which purpose a "go-devil" can be used. This consists of two wooden discs sUghtly smaller than the bore of the pipe. Each disc is made of two thin pieces of wood clamped together with a piece of cloth-insertion rubber packing between them, cut to fit tightly in the bore of the pipe. The discs are fastened together about 2 ft. apart by three M-ii- iron rods. A ^^-in. manila rope passing through and fastened to the discs is threaded through the pipe as laid, after which the go-devil is pulled along, effectually cleaning the sewer pipe. A poor substitute for this is the swab or "half moon" made from a piece of board cut to the curve of the interior of the pipe but not filling it. This board, fastened to a stick a httle longer than pipe, is drawn along first on top and then on the bottom, to clean the pipe. Sulphur and Sand Joints. — A composition of sulphur, tar and sand has been used for many years in England in making the Stanford pipe joint, which has found little or no favor in the United States. Joints of sulphur and sand have been used to some extent in the United States for many years, however, particularly in connecting several lengths of pipe on the ground before lowering them into the trench. The purpose of the sand is to keep the jointing material at a constant volume while it is coohng and prevent the shrinkage which occurs if little or no sand is used 250 SEWERAGE AND SEWAGE DISPOSAL with the sulphur. The sand must be very fine and without a gritty feeUng when rubbed between the fingers. Equal parts of sulphiu- and sand are generally used. The temperature of the material must not be raised above that necessary to keep it liquid, for too high a temperature will change the liquid to a plastic mass, which cannot be used until it cools again to the correct degree. The material is heated and poured Hke lead in jointing water mains, an asbestos jointer being useful in retaining it in the joint, except that the funnel should be higher, 3^ in. at the throat and 3 in. high. The joint is absolutely rigid when finished and for this reason some engineers prefer a more plastic joint. Asphalt and Tar Joints. — These are generally made with asphaltic preparations of the general character used in pavement work for joint fillers. The joint is first calked with oakum and then the hot filler is poured into it as lead is poured into the joint of a cast-iron water main. When the joint is filled it is overfilled with cement mortar. The material has a low melting point which prevents its successful use in places where large quantities of hot water or steam find their way into the sewer, as is the case in some large cities. In a large mileage of sewers in very wet soil where it was particularly desirable to prevent the infiltration of ground water into the sewers through the joints, Alexander Potter used the following methods: The inside of the bell and the outside of the spigot were swabbed with tar and a gasket of jute soaked in tar was forced into the joint. Into a bucket full of portland cement tar was slowly poured and kneaded by hand until the mixture was of the consistency of rather stiff dough. This was then rolled on a board into a roll of proper size to be forced into the joint, where it was calked until the space was filled to within }4 in. of the end of the bell, after which an overfilled cement finish was given to the work. In this work the spigot had to be supported while the joint was hardening or the weight of the pipe would cause the filling material to exude from the joint, and some difficulty was experienced at first in conducting the work so as to avoid such troubles. Linseed Oil Jointing Preparations. — Pipe joints have been made of several preparations in which linseed oil is the character- istic material. The G-K compound is described by its makers as "a homogeneous plastic mass composed largely of vulcanized PIPE, BRICK AND CONCRETE SEWERS 251 linseed oil and a binder of anhydrous clay." Marbleoid is com- posed of linseed oil, sulphur and marble dust. Jointite and Filtite are other plastic jointing compounds, the latter being made in several grades for poured joints and also in a grade so stiff that it must be molded and pressed into the joints, a property which is claimed to make it of value for joints that must be filled under water. The G-K compound is furnished in barrels of about 400 lb. The staves must be knocked away and the material broken up with a hammer or hatchet into pieces small enough to melt in a pail or kettle. The melting must be done carefully, for if the material is overheated it becomes carbonized and worthless. It bubbles and foams when first heated but after it has been kept in a melted condition for 15 to 30 min., the foaming disappears to a great extent and the material is ready for pouring. The joint is first calked with jute and the filler is then poured like melted lead in a cast-iron pipe joint, using an asbestos jointer to keep it in place. On work done under the authors' direction, a number of joints were poured in this way while water was stand- ing in the trench half way to the crown of the pipe, without any imperfections resulting in the joints, but they do not recommend this practice under ordinary conditions. Platforms and Cradles. — The necessity, of supporting sewer pipe securely is met by the use of timber platforms or concrete cradles. Figs. 126 and 127. Where a single plank laid directly under the pipe is used as a platform, the water in the bottom of the trench can flow along either side without undermining the sewer itself but there is some danger of its washing the sides of the trench. To meet this objection, longitudinal timbers are laid a few inches apart, each length of pipe being supported by a transverse timber near each end, cut to conform with the shape of the pipe. (Wedge blocks must not be used as they produce a concentrated load on the pipe which is likely to cause rupture.) With such a platform the water flows through the spaces between the longitudinal timbers. The objection to it is most evident where there is an underdrain below the pipe, for the water flowing under the sewer and over the underdrain may wash earth from below the sewer into the underdrain, causing settlement or cracking of the sewer. It is for this reason that concrete cradles are generally employed, even when a platform must be constructed to receive the concrete. 252 SEWERAGE AND SEWAGE DISPOSAL In very wet trenches in fine sand it is necessary in many cases to hold the bottom down until the pipe or concrete cradle can be placed. Frequently gravel and coarse sand spread over the bottom will be of some service; brickbats, straw, broken stone and cinders have been used for this purpose with more or less success. The best results are undoubtedly obtained when con- crete is used, but even this material cannot be relied upon at times to hold down the bottom unless a platform is first con- structed to receive it. The planks should be worked down through the sand and water until they are 6 to 8 in. below the bottom of the sewer, and the material above them should then be cleaned off as well as possible before placing the concrete. In quicksand and similar material, it is necessary to carry down the bottom of the trench by special expedients, and in carrjdng them out it is highly important to do the work rapidly and reduce the walking about on the bottom of the trench to a minimum, for the more such material is stirred up the more difiicult it is to control. In order to put in a cradle under such conditions, two planks are placed on edge far enough apart to include the cradle between them, thus forming a kind of bottom- less box. The ends of the planks are fastened together so as to hold them vertical and the planks are then pushed down into the quicksand. When they are in place they can be securely held there by braces between them and the sheeting of the trench. Planks cut to form a bottom for this box are then put in place one by one, by men standing on each plank and working it down to its final position, where it is held by cleats. Brief reference has already been made to methods of handling quicksand in Chapter VII under "Excavation." Branches. — A branch must be set in the sewer at every house, well down the line of the sewer from the point where the main drain passes through the wall of the building. Branches must also be provided for all vacant lots, usually every 30 to 40 ft., depending upon the character of the building operations in that vicinity. The branches have the side outlets closed with stoppers until the house connection between the sewer and soil pipe of the building is put in. The stoppers should be put in place before the branch is lowered into the trench and should be made as nearly water tight as possible. The work can be done con- veniently by placing a number of Y's on a horse constructed to PIPE, BRICK AND CONCRETE SEWERS 253 hold the branch openings in a horizontal plane. The stoppers are then placed in position, jute is calked around them, Fig. 129, and considerable mortar is placed around and over the jute and stopper. Then a piece of wood, cut in the form of a truncated cone, is twisted around on top of the stopper. This forces the mortar to the outside, forms a circular ring of cement in and over the joint, and leaves the central part of the stopper clean. Afterward, when it is desired to make a connection with one of the branches, it is necessary only to break the center of the stopper with a hammer and chip out the remainder of the stopper and mortar with a chisel. ^^^'^',';^,^'^^^'^V^''^^^^ '"'^x^^,;Vj Stopper Fig. 120. — Stopper in branch and former for sealing it. Backfilling. — Backfilling the trench is one of the most im- portant steps in sewer construction. The work of this nature about the pipe, up to the level of its crown, must be done with selected material from which hard lumps and large stones have been removed. This is usually done by the pipe-laying gang. The backfilling of the trench above this point is of much im- portance, because if not thoroughly done settlement is likely to take place. Only loose dirt should be placed around the pipe until the backfill is carried to a height of 6 to 12 in. above the crown of the pipe, or higher if filling is dumped from a height in large quantities, as from a bucket. Above this filling the coarser material may be used, but care should be taken to have rocks and lumps of hard material completely surrounded with fine m_aterial, so that there shall be no large voids. Usually one rammer to one or two men shoveling the earth into the trench, will accomplish satisfactory work. With some kinds of material the backfilhng may be con- sohdated by the use of water, the process being known as pud- dhng. Water should usually not be put into the trench until the backfilling has been carried up a short distance above the top of the pipe; then the trench should be partly filled with water and the earth thrown into it. The use of water is particularly well 254 SEWERAGE AND SEWAGE DISPOSAL adapted to materials through which the water can be readily drained away. Water should not be used in clayey materials which will hold it for a long period of time and which tend to shrink as the water dries out. By thorough ramming it may be possible to put more earth into a trench than by puddling with water. With some classes of material, such as loam, it is possible to put more material back into the trench than came out of it, and with sizes of pipe up to 15 in. usually all such material taken out can be put back. With other material, such as gravel, clay and hardpan, there will be a surplus left after backfilling, in excess of the cubic space taken up by the pipe and manholes. It is often desirable to fill the trench completely full of earth or even overfill it and then to roll it with a steam roller. If the trench is narrow this may best be done by so operating the roller that one of the rear wheels will pass over the trench and follow the backfilling down below the surface of the street. If sufficient material is put into and over the trench to completely fill it after rolHng, it will be necessary to remove a portion of the surface to make room for the macadam or paving to be put back later. When a roller is used it is important to take care not to hit the manholes, as the frames may be displaced and the brickwork injured. In very deep and wide trenches it may sometimes be less expensive to repair the street after settlement than to go to the expense of very thorough tamping during construction. The engineer in drawing his specifications for the consolidation of backfilling must use judgment as to the expense to which he cares to go to have the backfilling tamped or puddled, basing his judgment upon the character of the work and the nature of the locality through which the sewer is to be built. Where it is not desired to spend sufiicient money to thoroughly tamp the backfilling much can be accomplished with little labor by spread- ing the dirt to make sure that the voids around the stones and hard lumps of earth are filled. The use of water in puddhng is generally less expensive than tamping. Upon some large work it is not practicable to fill the trench with water and throw the material into it. If this is impracticable considerable good can be done by the intelligent use of a stream of water from a hose. The cost of backfiUing may in part be offset by the saving in the amount of material to be hauled away. Where it is necessary to PIPE, BRICK AND CONCRETE SEWERS 255 haul the surplus material a long distance, the cost of hauling may be fully as great as the cost of the tamping necessary to put the material back into the trench. BackfiUing should always be done as soon as possible after the sewer is built, in order to clean up the street and prevent incon- venience to the public due to large piles of dirt left in the street unnecessarily long. Where there is a surplus of material, the finer material and that which can be consolidated to best advan- tage should be used for backfilling, the coarser part being carried to the dump. In freezing weather it is sometimes difficult to get enough fine, unfrozen earth for backfilling immediately around the sewer. This should be done, however, even if necessary to thaw the earth thrown out upon the bank for the purpose. Filhng the trench with frozen earth is always unsatisfactory and it is never possible to secure a well-compacted fill under such conditions. Settlement follows thawing in the spring, when further filling becomes necessary. Moreover, the integrity of the sewer pipe may be endangered by such practice. Where it is not necessary to reopen the street to travel, as when the sewer is built through open fields, the amount of labor devoted to tamping may sometimes be reduced or wholly elimi- nated and the fill allowed to settle and become compacted by rains. Tamping is a portion of the work which is very likely to be neglected. It is not excessively expensive and is usually well worth the cost, especially in city streets. The cost of tamping varies greatly, but in ordinary city trenches where the sewers are small, it may involve from one tamper per shoveler, to one tamper to three shovelers. CONCRETE AND BRICK SEWERS Circular Sections. — The circular section encloses a given area with the least perimeter and, therefore, gives the greatest velocity when flowing full or half full. Under ordinary con- ditions, such sections are economical in the quantity of masonry required, although in flat-bottom trenches or under conditions requiring pile foundations or timber platforms, additional masonry is required to support the arch. In combined systems, where the dry-weather flow of sewage is very small in comparison with the storm-water flow, the velocity for the low flows is greatly 256 SEWERAGE AND SEWAGE DISPOSAL seduced in the circular section and on that account this section may not be so advantageous as the egg-shaped section. The standard circular concrete section built at Louisville is shown in Fig. 130 and Table 35. The dimensions were based on what experience had shown to be a safe thickness of masonry under the conditions there. The minimum thickness at the crown and at the invert was fixed at 5 in. because of the practical difficulty of ob- taining with certainty a first-class wall of monolithic concrete of less thickness. The soft-ground section was used where the earth would not stand when trimmed to the shape of the firm-ground section. Other circular sewers are illustrated in Figs. 58, 59 and 60. The circular and egg-shaped section? are usually preferred for sewers under 5 ft. in diameter. Egg-shaped Sections.— In combined sewers where the dry-weather flow of sewage is small compared with the ca- pacity of the sewer required for storm water, or in sanitary rewers for a district where the present population is but a small Table 35. — Dimensions of Plain Cikculak Conceete Sewers, Louisville 6/" -^ biamei-erofSewei', '<^'% Fig. 130. — Louisville standard concrete section. Quantity of Dimensions of the sections concrete cu. yd. per lin. ft. sewer Diam- eter Hi fl-j B. B^ ifi Ki fli iJ2 iJs Firm ground Soft ground 24" 6" 6" y 7i" V 81" 6" I'lOJ" I'O" 1' 6" 1' 5" 0.13 0.15 27" 5" S" 1' w I'lO}" 6i" 2' IJV" 1' U" 1' 8" 1' 6J" 0.15 0.18 30" S" 5" 2' 0" 2' \Y 7i" 2' 4i" 1' 3" 1' 10" 1'8" 0.18 0.21 33" 5" 5" 2' 28" 2' 4i" 8i" 2' 6W" 1' 4J" 2' O" V 9J" 0.19 0.23 36" 5" 5" 2' 4J" 2' 6J" 9 " 2' 9i " 1'6" 2' 2" I'll" 0.22 0.26 39" 5" 5" 2' 7i" 2' 9J" 9}" 3' OA" 1' Ih" 2' 4" 2' Oi" 0.26 0.29 42" 6" 6" 2' 9i" 3' " 105" 3' 3| " V 9" 2' 6" 2' 3" 0.29 0.35 45" 6" 6" 3' 0" 3' 2Y Hi" 3' 6iV" I'lOj" 2' 8" 2' 45" 0.33 0.40 48" 6" 6" 3' 2S" 3' 5i" 1' 0" 3' 9" 2' 0" 2' 10" 2' 6" 0.38 0.45 51" 6" 6" 3' 45" 3' 8 " 1' OJ" 3'llH" 2' IJ" 3' 0" 2' 7i" 0.41 0.49 54" 6" 6" 3' 7i" 3'lOi" V w 4' 2J" 2' 3" 3' 2" 2' 9" 0.43 0.53 57" 6" 6" 3' 95" 4' U" V 2i» 4' 5,^" 2' 4§" 3' 4" 2'105" 0.47 0.57 60" 6" 7" 4' 0" 4' 4 " 1' 3" 4' 85 " 2' 6" 3' 6" 3' O" 0.63 0.65 63" 6" 7" 4' 2J" 4' ej" 1' 3}" 4'llA" 2' Ih" 3' 8" 3' U" 0.57 0.71 66" 6" 7" 4' 4J" 4' 9J" 1' 4i" 5' IJ" 2' 9" 3' 10" 3' 3" 0.61 0.77 69" 6" 8" 4' 7i" 5' " 1' 5i" 5' 4H" 2'IOJ" 4' 0" 3' 45" 0.66 0.84 72" 6" 8" 4' 9f" 5' 21" 1' 6" 5' 7i" 3'0" 4' 2" 3' 6" 0.70 0.88 PIPE, BRICK AND CONCRETE SEWERS 257 proportion of the ultimate development, the ideal sewer section is one in which the hydraulic radius remains constant as the depth of flow decreases. It is impracticable to obtain the ideal, but the egg-shaped or oval section. Figs. 51 and 63, comes nearer to it than any other thus far devised. As explained in Chapter IV, the advantage of the egg-shaped section is that for small flows the depth is greater and the velocity somewhat higher than in a circular sewer of equivalent capacity, i The depth of flow in the egg-shaped sewer is always greater than in the equivalent circular sewer for equal discharges, and for small flows this increase in depth produces better flotation for the soHd matter. In deep trenches there will be a saving in total excavation by using the egg-shaped sewer, due to the decrease in width of trench, which may more than offset the small increase in depth. The disadvantages of the egg-shaped section are that it is less stable, more liable to crack, requires more masonry, and in general is more difficult to construct than the circular section. In very stiff earth or in rock it is sometimes possible to excavate the bottom of the trench to conform to the shape of the invert of the sewer, but in general, in yielding earth or where founda- tions are poor and piles or platforms are needed, the egg-shaped section requires more masonry backing under the haunches to support the arch than does the circular sewer. Hence the egg- shaped section may be more expensive than the circular section in many cases, and much more expensive than some of the other sections. Semi-elliptic Sections. — The arch of this section, Fig. 131 and Table 36, is either a true semi-elUpse or is made up of three circular arcs approximating a semi-ellipse. As the center of gravity of the wetted area is lower than in a circular sewer, the normal flow line will be much lower, which may be of some advantage in locating lateral connections lower or in raising the invert of the main sewer. This, of course, contemplates the possible operation of the lateral sewers under a head at times when the main sewer is running full. Although this is of mate- rial advantage where the allowable difference in water level is small, it should generally be avoided. The chief advantage of this section is that the shape of the arch more nearly coincides with the Une of resistance of the arch under 1 This is shown in detail in "American Sewerage Practice," vol. i, page 384. 17 258 SEWERAOE AND SEWAGE DISPOSAL actual working conditions than is the case with other sections. Because of this the arch can be made relatively thin and still keep the stresses in the masonry within allowable limits. The section is dependent to only a small extent on the lateral pressure of the earth to prevent failure, and does not depend upon the resistance of the earth filling on the sides, as is very often the case with a horse-shoe section. Table 36. — Dimensions of Authors' Semi-elliptical Sewbb Section (See Fig. 131) 1 2 3 4 5 6 7 8 9 10 11 ^-[T Inside vertical di- ameter "D" Area of water- way Hy- draulic mean radius Thickness of concrete Interior radii Exterior radii Area of con- crete Quan- tity of con- crete ou. yd. Crown Center of in- vert and spring Crown intra- doa and side Side intra- dos Invert and side extia- dos Crown extra- dos Invert line wall per ft. in. sq. ft. ft. ft. in. ft. in. ft. in. ft. in. ft. in. ft. in. ft. in. sq. ft. Un.ft 6 28.2 1.442 6 9 2 6 3 7 6 2 6 8 3 14.12 0.523 6 6 33.1 1.562 61 9f 2 2 6 9i 8 U 2 81 8 11} 16.58 0.614 7 38.4 1.683 7 lOJ 2 4 7 3i 8 9 2 11 9 71 19.21 0.712 7 6 44.05 1.803 7i Hi 2 6 7 9i 9 45 3 1} 10 3} 22.08 0.817 8 50.1 1.923 8 1 2 8 8 4 10 3 4 U 25.10 0.930 8 6 S6.6 2.043 Si 1 Oi 2 10 8 10} 10 71 3 61 11 8} 28.44 1.054 9 63.4 2.163 9 1 li 3 9 4i 11 3 3 9 12 41 31.80 1.177 9 6 70.7 2.284 9i 1 21 3 2 9 10} 11 lOJ 3 111 13 0} 35.41 1.311 10 78.3 2.404 10 1 3 3 4 10 5 12 6 4 2 13 9 39.24 X.453 10 6 79.3 2.625 lOi 1 3i 3 6 10 11} 13 1} 4 41 14 5} 43.26 1.60S 11 94.75 2.646 11 1 ii 3 8 11 5i 13 9 4 7 15 11 47.48 1.757 11 6 103.5 2.764 llj 1 5i 3 10 11 11} 14 4J 4 91 15 9} 61.89 1.921 12 112.75 2.884 1 1 6 4 12 6 15 5 16 6 66.51 2.092 13 6 122.4 3.005 1 Oi 16J 4 2 13 0} 15 7} 5 2} 17 2} 61.31 2.270 13 132.4 3.125 1 1 1 7i 4 4 13 6} 16 3 5 5 17 10} 66.32 2.456 13 6 142.7 3.246 1 1} 1 8i 4 6 14 0} 16 101 S 7} 18 6} 71.51 2.649 14 153.5 3.365 1 2 1 9 4 8 14 7 17 6 5 10 19 3 76.91 2.849 Area of waterway = 0.78311)2. Area of concrete section = 0.3924^2. The semi-eUiptical section depends to a larger extent upon the stabiUty of the invert than do the circular, egg-shaped, cate- nary, gothic, basket-handle and horseshoe sections. Where the semi-elliptical section is constructed in compressible earth and the structure is built monolithic, with reinforcing bars running continuously from the center of the invert to the crown of the sewer, there will be a large bending moment at the center of the invert. Under such foundation conditions, the invert should be made as thick as the arch at the springing Hne and should be heavily reinforced to withstand the stresses. Unless this is done, cracks are likely to occur in the center of the invert. PIPE, BRICK AND CONCRETE SEWERS 259 As in the horseshoe type, the invert of the semi-elliptical section readily conforms to the bottom of the trench excavation and for that reason the quantity of masonry below the springing line is not excessive. This section is not as advantageous for low flows as the circular, because of the wide and shallow invert in which there is a very low velocity. However, for sewers where the quantity to be carried is not subject to wide variations and the normal flow is as much as one-third of the total capacity of the sewer, this Fig. 131. — Authors' standard semi-elliptioal section. disadvantage may be neglected. The hydrauUc properties of the semi-eUiptical section are very good in general, which, with the very desirable structural features, make this type one of the best for sewers over 6 ft. in width. Other Sections. — The catenary section. Figs. 54 and 64, was used extensively on the Massachusetts North MetropoUtan sewerage system under the direction of Howard A. Carson. Its principal advantage is that it conforms so nearly in shape to the available space inside the wooden timbering of earth tunnels. The line of resistance keeps fairly well within the arch section. 260 SEWERAGE AND SEWAGE DISPOSAL the hydraulic properties are fairly good, and the center of gravity of the wetted section is low with respect to the height. The gothic section, as developed on the Massachusetts North Metropohtan system, has a horizontal diameter about 17 per cent less and a vertical diameter about 8 per cent more than the diameter of the equivalent circle. The horseshoe and basket-handle sections are much alike. The latter has a horizontal diameter about 6 per cent less than the vertical. The arch is slightly pointed and the invert is flatter than a semicircle. The horseshoe section, Fig. 53, has been widely used. Above the springing Une it has a semi-circular arch, while the side walls below the springing line are vertical or incline inward, sometimes with a plain and sometimes with a curved surface. The invert varies in section from a horizontal line to a circular or paraboUc arc, or other design calculated to concentrate the low flows near the center of the invert. The invert conforms to the trench bottom, saving masonry. For a given width, this section will have less height than the equivalent circular sewer. Its main disadvantage is that it depends upon the ability of the back-filling to resist the lateral thrust of the arch transmitted to the side walls near the springing line. The parabolic or delta section, Fig. 55, has a somewhat larger carrying capacity than a circular section of the same height. It is both economical and strong, the normal flow line is lower than in the circular section, and the sloping invert is well adapted for low flows. It requires a wider trench than the semi-eUiptical section for equal capacity- and height. The U-shape section, Fig. 66, has fairly good hydraulic prop- erties until it becomes filled, when the width of the roof causes a large increase in the wetted perimeter. The invert adapts it for low flows and is easily constructed. It requires considerable masonry in proportion to its area, and its main field is for sewers about 3 ft. wide and considerably over 3 ft. in depth. The rectangular section. Fig. 52, has been steadily increasing in favor because of its excellent hydraulic properties until it is filled, simplified form work, economy of masonry and space in the trench, and ease of construction. The V-shape invert is frequently used with the rectangular section on account of its suitability for low flows. The semi-circular section, Figs. 57 and 65, was formerly usde for large sewers on low land, where the natural ground surface PIPE, BRICK AND CONCRETE SEWERS 261 was below the top of the sewer. The rectangular section has largely displaced it as being less expensive and possessing better hydrauUc properties. Analysis of Masonry Arches. — The complete analysis of the stresses in masonry arches is a highly complex problem which few engineers are ever called upon to solve. A considerable proportion of the large masonry sewers in the country were designed without anything more than a very elementary investigation of their arches. The increasing use of reinforced concrete sewers is responsible for a more general effort by de- signers to analyze the stresses in these structures, and now that a complete, accurate method^ of doing this has been developed, the design of large sewers can be made to depend upon both theory and experience instead of on the latter mainly. There are three general methods of analysis^ now in use, as follows: The first method, called the "voussoir arch method," based on the so-called "hypothesis of least crown thrust," is appUcable only to that portion of the sewer section above the springing Hne of the arch. Either the sewer must have very heavy side walls or the thrust of the arch must be carried by the sides of a rock trench, in order to make this method strictly appHcable. The second method, based on the elastic theory of the arch and following the method described by Turneaure and Maurer in their "Principles of Reinforced Concrete Construction," is appUcable to all sewer sections and can be used to cover all conditions. It has some mechanical disadvantages when appUed to the analysis of the entire sewer structure, invert included. The third method, also based on the elastic theory but using the so-called method for indeterminate structures, is of special advantage in the analysis of the entire sewer section as it permits a more suitable division of the axis in the side wall and invert. It does, however, require some additional labor over the second method v/hen applied to an arch with fixed ends. For large sewers constructed in compressible soil and built of monoUthic reinforced concrete, the third method is the most desirable. Influence of Construction Conditions on Section. — The method of construction of a brick or concrete sewer, whether in open cut or in tunnel, may have an important influence on the selection of the cross-section, as well as considerations of strength ' "American Sewerage Practice," vol. i, p. 4S8. ' "American Sewerage Practice," vol. i, p. 471. 262 SEWERAGE AND SEWAGE DISPOSAL and of the hydraulic elements of the different sections. In a tunnel it is desirable to use a section which will utilize to the best advantage all of the space inside the tunnel bracing. In earth tunnels with the usual timbering, the catenary or semi- elliptical sections conform readily to the available space, while in rock tunnels the circular or horseshoe sections are likely to be advantageous. When the sewer is built in an open cut, the choice of the section will be influenced by the bearing capacity of the trench bottom and the abihty of the arch to carry the load of the backfilling. In rock or firm soil, it is possible to shape the bottom of the trench to the profile of the bottom of the sewer and thereby save considerable masonry when using circular or egg-shaped sections. If the excavation is in soft material and the bottom of the sewer must be flat or the sewer is to be carried on piles or a timber platform, considerable additional masonry will be required for the circular or egg-shaped sections, which is not needed if other sections are employed. Sometimes the cross-section is influenced by the approach of the crown of the sewer to the surface of the street, sometimes the side room is limited by adjacent subsurface structures, and the available depth may be limited by tidal Influence on the ground water or other conditions controlling the permissible depth of the hydraulic grade line. The rectangular section has proved one of the most useful for such conditions, although the horseshoe section, with the horizontal or vertical diameters adjusted to meet the conditions, has been largely used. In a few cases the full elliptical section has also been used in restricted places. Where the hydraulic grade line depth is limited, it is desirable to use a sewer section which will carry the maximum and minimum flows with the least variation in depth of flow. The catenary, parabolic, semi-elliptical and rectangular sections are especially suitable for this purpose, as the center of gravity of the wetted area is comparatively low down from the crown in contrast to the circular section. The semi-circular section has also proved useful in this connection, although the rectangular section is being used instead in the more recent work of this character. Cost of Excavation and Materials. — The cost of excavation required by one type as compared with another should be care- fully considered, for if the excavation is in earth in a deep trench, PIPE, BRICK AND CONCRETE SEWERS 263 it will probably be cheaper to use a narrower deeper section and thereby save considerable width of excavation, even though the depth of excavation be slightly increased. This will be especially true in a deep rock trench where it may be found of advantage to use a narrow rectangular section having a height 1}4 to 2 times the width. For a sewer built in very shallow cut, or practically on the surface of the ground, a wider section will be advantageous, because little additional cost is incurred by in- creasing width whereas greater depth may increase materially the cost of excavation. Furthermore the cost of an embankment over a wide section will generally be less, because of reduced height and narrower side slopes. The parabolic or delta section is especially useful for crossing low lands where the sewer is largely out of the ground and must be covered by an embank- ment. The semi-circular section has also been much used for this same purpose, but has been superseded more recently by the rectangular section, having a width about IJ^ times its height. In former years a great many sewers were constructed of quarry stone or large cobbles, but in recent years other materials have proved less expensive and better adapted for the work. The cost of brick varies greatly in different localities and this may influence to a large extent the type of construction selected. In general, concrete is more desirable than brick, but where brick masonry can be had much cheaper than concrete it may be advisable to build the sewer of brick. The object in designing a sewer section should be to obtain one in which the quantity of the masonry and other materials is a minimum consistent with the requisite stability, hydraulic properties and other considerations. For sewers in which the normal flow is at least one-third of the maximum flow, it has been found that the semi-elliptical section is very economical in masonry, and at the same time provides for the other requirements. Stability. — Where a sewer is constructed in open trench, the structure must be designed to carry the earth or trench load as well as any superimposed load. The circular arch is not as strong as either the gothic, the parabolic, or the semi-elliptical arch. The semi-circular arch depends to a great extent upon the lateral pressure of the sides of the trench and also to a certain extent on the lateral resistance or passive pressure of the earth backfilling, although this can be obviated by increasing, the 264 SEWERAGE AND SEWAGE DISPOSAL thickness of the side walls or abutments. The semi-circular sections obviate part of this difficulty by omitting the side walls and resting the springing fine of the arch directly on the invert or foundation. In a rock trench the resistance of the sides of the trench is so great that the side walls of the sewer can be greatly reduced in thickness, the thrust of the arch being carried directly into the walls of the trench. In this latter case a very flat arch can be used to advantage. Imperviousness. — Where a sewer is to be constructed under a river bed or below the water table, it may be of particular importance for the walls of the sewer to be impervious. To this end, if the sewer is built of concrete, it is desirable to insert longitudinal reinforcing bars in the concrete, with a total area of 0.2 to 0.4 per cent of the sectional area of the concrete, in order to distribute the stress throughout the length of the sewer barrel and thereby prevent the formation of large cracks which would permit leakage. Unless the cracks are very small there may be some danger of corrosion due to the water passing through them and coming in contact with the reinforcement. This might in time weaken the structure. While the possibility of leakage or infiltration does not ordi- narily determine the shape of the waterway of a sewer, it is worthy of consideration when the selection is to be made. For example, if a sewer is to be built below the water table it may be well to adopt a section which is least likely to crack, whereas under other conditions the advantages of a different section might be sufficiently great to warrant its use even though small arch cracks are to be expected. The stability of the horseshoe section depends to a certain extent on the lateral pressure of the earth backfilling, and on that account, the semi-circular arch is apt to crack and may produce unsatisfactory conditions, not only because of leakage into the sewer, but especially on account of the rusting of the steel reinforcement. Materials for Arches. — -In constructing brick arches, three general types of bonding have been used. In the first, "rowlock bond," the arch is built of concentric rings of brick, all laid as stretchers. In the second type, part of the brick are laid as stretchers and part as headers, with radial joints in which the outer end of the joint is widened by increasing the thickness of the mortar or by inserting thin pieces of slate. In the third method. Fig.. 59/, the masonry is divided into blocks or sections. PIPE, BRICK AND CONCRETE SEWERS 265 Plain concrete arches have been useji to a considerable extent in recent years, and have an advantage over the stone or brick mas- onry arches in that the structure is soniewhat more elastic and may withstand tensile stresses to a slight degree although they should not be designed with this in view. In the design of such arches, as well as those of stone and brick, the line of pressure should fall within the middle third of the section, in order that no tensile stresses may be developed. If all the loads acting on the sewer were known exactly, it would be possible to design the section so that at no time would the line of pressure lie outside the middle third, but practically this is impossible, as our knowl- edge of the action of earth pressure is a matter of approximation only. On that account, under special conditions the stresses in the arch section may not be entirely due to direct compression, but in addition bending stresses may be developed. Arches of reinforced concrete are not subject to the limitations just mentioned, but can be made to withstand heavy bending moments by reinforcing the section with steel bars to carry tensile stresses. In arches in which the line of pressure lies within the middle third, the stresses in the arch are mainly due to compression and the concrete must of necessity carry the principal part of the load, so that the steel cannot be stressed to the allowable limit. On the other hand, the presence of the steel reinforcement furnishes a sort of insurance to the structure, to care for tensile stresses which may occur on account of unequal settlement of the foundations, temperature changes and many other conditions which cannot be foreseen. The steel is also an additional factor of safety against careless and defective construc- tion. On account of its presence, it is possible to increase slightly the allowable working stresses in the concrete over those which should be used for plain concrete masonry. Because of these considerations the authors beheve that for large sewer arches reinforced concrete offers greater advantages than plain concrete, even though an analysis of the section shows that the line of resistance for the conditions considered will remain within the middle third of the masonry section. Lining for Conctete Construction. — From the observations made by the authors of the condition of sewers after some years of service it appears that on all slopes in which the estimated velocity of the sewage will be 8 ft. per second or greater, the invert may advantageously be paved with hard-burned or prefer- 266 SEWERAGE AND SEWAOE DISPOSAL ably vitrified paving brick with square edges, laid with the edges projecting as little as possible and with full portland cement mortar joints. This invert paving should extend well up on the sides of the sewer, on straight sewers covering, in general, the bottom quadrant of a circular sewer. The use of brick paving is preferable to concrete on account of the greater ease of making repairs and further on account of the probability that vitrified or even hard-burned brick will withstand the wear better than concrete of average quality. It is desirable when sewers are to be built of concrete to use hard aggregates, especially for inverts, and a first-class granolithic finish where the surface is subject to greatest wear is better than the ordinary concrete finish. Brick Laying. — The brick should be laid by shoving them into a full bed of mortar, forming full joints on all sides at one operation. The longitudinal face joints should be as narrow as practicable, usually }/i to ^ in. The joints between the rings should not exceed 3^ in. unless it is desired to put in a thicker joint to prevent infiltration. The exterior of the arch is often covered with a plaster of cement mortar J^ in. thick. After the centers have been removed the interior surface of the sewer should be cleaned and pointed as smooth as possible. In some cases one or two coats of cement wash have been applied to the inner surface for the purpose of reducing roughness. Where the trench is composed of suitable material, the bottom is trimmed to conform to the desired outside surface of the invert masonry, and each brick laid directly on a bed of mortar placed on the bottom. Templates or "profiles" are placed in position and a mason's line is tightly drawn between them. Bricks are then laid in a single row in the center of the bottom of the invert for the full length of the section, about 12 ft. The several rows of bricks are then laid on either side, care being taken to lay each brick accurately to line at the time it is slid into place. A skillful mason will rarely touch the brick after the joint is made, and hammering a portion of the wall to bring it to line should be discouraged, as the adhesion of mortar to brick may be destroyed in this way. Care should be taken to see that the outside ring of brick is well supported by the bank. It frequently happens that the trench cannot be excavated exactly to the line of the proposed brickwork, in which case masons will often fill the space between the brickwork and the bank with loose material insufficiently rammed. In fact it is very difficult to ram the PIPE, BRICK AND CONCRETE SEWERS 267 material filling this space without injuring the green masonry. It seems probable that many of the cracks which have been found in brick sewers may be attributed to loose and improper filling between masonry and the solid bank of the trench. Care should be taken in laying the arch rings to see that all joints are properly filled and not unnecessarily thick. The key should be laid by an experienced mason who realizes the importance of tight joints and a well-fitting key properly forced into place, that there may be no injury to the arch when the centers are struck. After the first few courses in the invert are laid, it is wise to provide a few loose boards or planks upon which the masons and tenders may walk, thus protecting the green brickwork from injury, due to movement of the brick resulting in a lack of adhesion of mortar to the surface of the brick. Upon large sewers the invert may become so high that it will be necessary to provide a working platform for masons and tenders. In building such platforms care should be taken to protect the green masonry from injury. In trenches in wet or loose material when the soil will not main- tain any given shape, the brickwork may be laid in a cradle. Such a cradle may be constructed of ribs cut to the shape of the outside of the invert, making an allowance for the thickness of the plank which are to form the cradle proper. These ribs are set in the bottom of the trench at the proper elevation and 2-in. planks are nailed to them, earth being carefully tamped behind each one as it is put in place. The brick are laid, as already described, on this cradle. Profiles and Centers for Brick Sewers. — A profile should be made as light as possible and at the same time strong and rigid enough to always hold its true shape. As shown in Fig. 132, nails are driven in the profile, properly spaced for the several courses. Upon large work and where skilled masons are employed, two or sometimes three or more courses of brick may be laid to one stretching of the line. This is especially true upon relatively flat inverts, as those of horseshoe-shaped sewers. Profiles should extend high enough to enable the mason to carry the invert one or two courses above the springing line to allow room for the centers to fall after being "struck." Two profiles are usually spaced 16 ft. apart when a new invert is begun, and thereafter one profile only is used and it is generally placed about 16 ft. ahead of the finished work, a longer spacing 268 SEWERAGE AND SEWAGE DISPOSAL f^gjg^t L— 4Zf?g.'A Joint- Fig. 132. — Profile for 36-in. brick aewer. being unwise because of the danger that the line may sag between the nails. The dimensions and method of constructing a profile suitable for use in building a two-ring brick sewer 36 in. in diameter, are shown in Fig. 132. Such a profile should be made preferably of clear soft stock, such as white pine, planed on both sides and of uniform thickness. Provision should be made for the mortar joint between the outer and inner rings and the nails should be so spaced as to allow for joints of proper thickness between courses. It is gen- erally well to send the profile from the shop without having the line naUs driven, so that the mason may space them to conform to the thick- ness of the brick actually being used. The profile represented in Fig. 132 is designed for brick 4 in. in width and 2 in. in thickness. A profile may be secured in place by two vertical boards nailed to its sides and to a trench brace or other support above, and should be accurately set to line and grade and carefully leveled before any brick are laid. Arch forms or centers, generally built of wood, are usually 8 or 10 ft. in length. It is common practice to build them in one piece, although some prefer to have them in two or more segments to facilitate handling, especially inside the completed sewer. They are set on chairs or legs and can be adjusted to the proper height by means of wedges. After the arch has been completed and the forms have been in place the required length of time, the centers are struck by loosening the wedges. It will be found convenient to have e;iough centers for a full day's work, the length required varjdng greatly, but commonly ranging from 24 to 48 ft. Arches turned one day may generally have centering removed the next morning; in fact, centers are sometimes struck without apparent injury to the work as soon as the masonry is completed and backfilled to a depth of 1 or 2 ft. above the crown. Centers are usually set as soon as the inverts are completed. The size of the lagging and thickness and spacing of the ribs are PIPE, BRICK AND CONCRETE SEWERS 269 governed by the size of the sewer. Centers should be made as light as possible to make them easy to handle, but they must be strong and rigid enough to receive and support the masonry without deformation. When sewers are constructed in tunnel, block centers are more convenient than full centers, although the latter may be used if very short, say 3 or 4 ft. in length. A block center is shown in detail in Fig. 133, and derives its name from the key block. The two large segments are first set in position and through the space at the top between them the brick and mortar are passed to the masons, after which, as the work progresses, the short key blocks are placed and the arch is completed. i ^J Supplemenlary ^X ,...'}% ,. Key Block. «! H-CCBJ" -r"//' ../D'O' tll'-^- I'll" 3'' fi End Elevation. Fig. 133.- Sprlng Line-' Section A-B. Details of block center. Placing Concrete. — Upon small work, where excavations are shallow, it will usually be most satisfactory to shovel the concrete into the forms from the mixing bed. Where the work is deep and it is not convenient to place the mixing bed over it, the concrete can be conveyed from the mixing bed to the forms through chutes, which may be made of wood lined with sheet iron, or preferably entirely of sheet iron. Such troughs are frequently U-shaped, although, and especially where they must be placed in a nearly vertical position, they may be made nearly cylindrical in form. In either case they should be so constructed as to telescope, the lower end of the upper trough fitting into the upper end of the trough below. It will usually be found con- venient to provide a metal hopper at the top, into which the con- crete may be thrown or dumped and from which it will flow into the troughs. Concrete should be made of such a consistency that it will flow 270 SEWERAGE AND SEWAGE DISPOSAL readily through a trough inclined 1 ft. in 3 ft., and at least this inclination should be provided, as concrete will not flow readily through troughs laid upon a flatter slope. It is necessary many times to hang the troughs or pipes upon a much steeper slope and difficulty is often experienced when the angle of inclination is at or somewhat more than 45 deg. by the tendency of the stone, and particularly of rounded gravel, to separate from the mortar in passing through the trough or pipe. Where the angle of inchnation is nearly vertical, this tendency is again reduced. Where there is a decided tendency for the stone to separate from the mortar, it may be necessary to catch the concrete in a box and re-mix it before finally placing it in the forms. If this is not done, care must be taken to prevent "pockets" of cobbles or broken stone in which there is insufficient mortar and fine aggregate. Concrete is now usually mixed wet and does not require tamp- ing, but does need more or less spading and slicing with thin metal blades. These are of assistance in working the concrete through reinforcement and for scraping and prying the larger stones away from the forms, so as to leave a good mortar finish when the forms are removed. It is to be remembered, however, that excess of water produces a weaker concrete than a well filled thoroughly tamped drier mixture. Two or more men, depending on the size of the work, are required for placing and cutting the concrete, and one man, when wheelbarrows are employed, to assist in emptying them into the chutes. Small circiilar sewers, perhaps up to a diameter of 4 ft., can in many cases be advantageously built in one operation, full cylindrical forms being placed before any concrete is put into the trench. Upon larger sewers there is a great variety of forms available for use, some being made of wood and others of steel. Much ingenuity may be exercised in the design of large forms to reduce the cost of labor in removing and carrying them forward and resetting them. Sewers more than 4 ft. in diameter are generally built in two, three, or more operations. The invert is first placed for a con- siderable distance. Upon this the side walls are built and after these become set the arch is turned. This method of progressive construction requires that a considerable length of trench be kept open at aU times for concreting, and as time is required for the concrete to become thoroughly set and acquire sufficient PIPE, BRICK AND CONCRETE SEWERS 271 strength so that the forms may be removed, it is a source of considerable delay which should be taken into consideration in planning the rate of progress to be attained. Inverts are generally laid and screeded to templates. In some cases the concrete is placed in alternate blocks, from 8 to 10 ft. in length, between bulkheads. Where the concrete is placed continuously it is necessary to provide templates at frequent intervals, which may conveniently be constructed of M by IJ^-in. steel on edge and hung from the bracing overhead, Fig . 134. As the concrete is screeded these templates are moved forward. Fig. 134. — Templates to which invert was screeded, Louisville. Where the concrete of which the invert is constructed is very wet and contains a large quantity of mortar, it may be possible to so work it as to bring the mortar to the surface and provide a satisfactory and smooth invert surface by screeding. ■ If the concrete is lean or dry it may be necessary to use a small quantity of mortar to secure a smooth finish, usually not more than J^ in. This, of course, should be applied before the concrete has acquired its initial set. If good concrete work is to be secured it is absolutely essential to have watertight, well-made and accurate forms, and to mix the concrete of the proper consistency. If these two points are given careful attention, little difficulty should be experienced in providing a satisfactory concrete structure. 272 SEWERAGE AND SEWAGE DISPOSAL Placing Reinforcement. — In the construction of reinforced concrete sewers, placing and holding the reinforcement in its proper position is a troublesome as well as an important matter. This is particularly the case where expanded metal or woven fabrics of fairly coarse mesh are used. The fabric acts as a screen, preventing the free flow of concrete around the forms. In such cases, as with larger circular sewers, it is desirable to lay part of the invert under the more inaccessible part of the form, by means of templates and screeds. Part of the reinforcing material, bent to shape and extending above the springing line, is held in proper position by fastening it to longitudinal timbers on either side of the trench, and, if necessary, it is also supported by blocks underneath it, removed as the concrete reaches them. When the invert is more or less flat, the concrete may be first placed up to the position of the steel, which is then put in position and the concreting continued, or the whole system of reinforce- ment may be first placed in position and held by blocking and fastening to the trench braces, before any concrete is placed. After the sewer is completed to the springing line and the centers placed, the arch steel is set and fastened by wiring it to the pro- jecting ends of the invert steel, a proper space between the centers and the steel being maintained, if necessary, by blocks which are removed as the concrete reaches them. Metal and concrete chairs or supports may also be fashioned or used advantageously in some kinds of work. Forms in General. — ^The forms and centers used for concrete sewers differ from those used for brick sewers in that they must have smooth surfaces and be watertight. In selecting the type to be used and the material of which it is to be built, the chief con- siderations are lightness and economy, for substantially as good workmanship can be obtained with the several types and materials provided the forms are properly built and maintained. The first cost, cost of upkeep, and cost of placing, striking or collapsing, and pulling ahead, are some of the essential items to be considered. There are no arbitrary rules by which to determine whether wooden forms^ wooden forms covered with metal or steel forms are most economical for a given case, for this depends largely upon local conditions. The design of forms which can economi- cally and conveniently be erected, struck and moved ahead under the varying conditions encountered in the construction of sewers. PIPE, BRICK AND CONCRETE SEWERS 273 especially in deep trenches, has taxed the ingenuity of many experienced sewer builders. The hfe of wooden forms is comparatively short. In many instances after being used 8 or 10 times they have to be practi- cally rebuilt, although often the ribs can be used again, the lag- ging only requiring renewing. When new and well built, wooden forms give a smooth and satisfactory finish, but after repeated use, patching and repatching, they become so rough that the interior surface of the concrete is likely to be unsatisfactory. Building Concrete Sewers. — Where concrete sewers are built in two or more operations there is difficulty sometimes in securing good joints between the horizontal sections, and there is usually some finishing required because of the rough concrete where the centers do not fit the concrete of the invert. It has therefore been found advantageous in constructing sewers from 2 to 4 ft. in diameter to pour the concrete of both invert and arch at one operation, thus reducing to a minimum the cost of chipping, patching and finishing the inner concrete surface. On small sewers where the quantity of concrete per linear foot is only a small fraction of a cubic yard, the cost of preparing for mixing and pouring concrete, such as placing mixing boards, hoppers and chutes, which is a considerable item, is the same whether half or all of the sewer is built at one time. The cost of preparation per cubic yard is therefore half as much when all the sewer is built at once as when it is built in two operations. By completing the sewer in one operation the length of open trench required is less than when the invert is first laid and allowed to set before the arch is poured. The cost of handling the forms for a sewer built in one operation is also considerably less than the cost of handling the half-round forms used when the work is done in two operations. It is usually necessary to construct sewers having flat bottoms in at least two operations, because of the difiiculty encountered in securing the desired smooth finish of the invert if the concrete for both invert and arch is poured at one time. The sewer shown in Fig. 138 was built in this way. The invert and side walls were run first, the concrete extending back to the sheeting so that outside forms were not needed. The lowest set of trench timbers, which came 4 in. above these walls, was removed 48 hr. later, the concrete thereafter serving to brace the sheeting. Four days after the arch was poured the centers were collapsed IS 274 SEWERAGE AND SEWAGE DISPOSAL and moved ahead, through other forms in place, and set up ready for another run. FuU Round Wooden Forms.— Figs. 135, 136 and 137 illustrate the wooden forms used in constructing a 3-ft. sewer in one operation. Fig. 135 shows the cylindrical forms and out- side jackets in place. The sketch also shows how these forms Isometric View Fig. 135. — Wooden forms for building circular sewer in one or two operations Louisville, Ky. can be used for building the sewer in two operations, if that method of procedure is desirable, the concrete of the invert being shown as completed for a portion of the distance and the bulk- head being shown in place to receive the wet concrete of the arch. The inner forms are built in sections of four panels, shown in detail in Fig. 137. A skeleton support consisting of rings held PIPE, BRICK AND CONCRETE SEWERS 275 together by two stringers, one at the top and one at the bottom, is first placed in position in the trench and the panels are placed upon and attached to this support as shown in Fig. 136. When the concrete has acquired suflBcient hardness, the stringers holding the skeleton ribs in place are removed, allowing the ribs and panels to be collapsed and carried out through other forms supporting fresh concrete. There is no internal bracing required by these forms, so that the space inside the skeleton ribs is entirely free. ( Upptr Quarter of Lagging Ferhnved It) Shtnv Interior.) Gross Secf ion B-B. Fig. 136. — Full round wooden forms, Louisville, Ky. When full round wooden forms are used they should be thor- oughly soaked with water before they are placed in the trench, otherwise the water in the concrete will cause them to swell and crack the sewer the whole length of the run. Wooden forms, particularly when new, must be protected from the sun or they will dry out and the joints will open so that when concrete is poured the thin mortar will run through the joints and leave the concrete with a rough surface which must be chipped, pointed and smoothed, involving considerable expense and leaving a sewer 276 SEWERAGE AND SEWAGE DISPOSAL not so satisfactory as that built with tight forms. After being used a few times the forms will give no further trouble from swell- ing. They should be thoroughly cleaned and oiled before each use. Forms for Sewer Built in Two Operations. — The forms used in the construction of a 12 by 12-ft. semi-elliptical reinforced concrete sewer are illustrated in Fig. 138. It was first intended /In^/e Iron |x/ "xS'i'5" Note: ' i' Holes Space i"& 4-" from Angle. Fig. F. Details of Circles 7" ■>,.^ |<- gg .- Ficf.H. Detail of Panel Ribs. \ 3k'- No.l Common T&e. Flooring, N.C.P\ -TIOV- Fig.I. Plan of Panels, Detatls of Panels. Fig. 137.- — Details of full round forms, Louisville, Ky. to build this sewer in three operations — ■ invert, side walls and arch — and the forms were built with that method in prospect. It was later decided to build the invert and bench walls to the height shown on the drawing in one operation, after which the arch was to be poured. The supplementary bench wall form, A, Fig. 138, was therefore provided, but, as the lower portion of the arch form, B, had already been constructed, it was used PIPE, BRICK AND CONCRETE SEWERS 277 1^ 1 ■^fN — t**! ' 1 s ~ 1 t ^ ■, J lII jMIIIIII w f — TT' SlA *!• ^ hi- ^^ \ 278 SEWERAGE AND SEWAGE DISPOSAL throughout the work although had the original intention been to pour the concrete in two operations, this portion of the form would not have extended more than 2 to 4 in. below the top of the bench wall. The sewer was built in sections 48 ft. long. The bench wall forms with the angle iron templates attached were first placed in correct position as to line and grade, the forms being hung. Manholes. — Where 8-in. work is used, the brick may be laid as headers with the long sides forming radial joints, or in common J 1 1 1 1 3 Section ,f^^ ^^?*v ' e ff/is -' '^ X '>) '^''M^fA u^ Section C~C. utslde Form r- •?»■ — 1 Elevation of Inside Forms for Base of Manhole. Cross Section. Fig. 139. — Form for manhole base, Louisville, Ky. bond with several courses forming double rows of stretchers to be followed by a course of headers laid with the long sides making radial joints. In manholes requiring 12 in. of brickwork, the bond is complicated by the fact that if the joints are made radial the outer portion will become very wide. This may be obviated by laying the outer rings with ordinary % or J-^-in. joints and introducing stretchers or headers, as the case may be, when the opportunity offers, filling in the gaps with broken or half brick. PIPE, BRICK AND CONCRETE SEWERS 279 A good mason can lay the brick by eye so as to conform with accuracy to the prescribed hne. On small sewers the pipe are usually left out where manholes are to be built. A foundation of concrete or brick is first laid up to the invert of the sewer, which is then formed either with brick, concrete or a split tile of the same diameter as the sewer. Great care should be taken in the construction of manholes to have all joints in the brickwork well filled with mortar. A fruitful source of leakage is the joints around the pipes entering the manhole. These should be filled and where possible the hub of the pipe should be laid in the masonry, forming a cut-off. Fig. 139 illustrates the forms used for building a concrete base for a brick manhole on the 36-in. concrete sewer built with the forms shown in Fig. 135. The section in that illustration shows the inside manhole form in place, attached to the sewer forms, after pouring the concrete of the sewer and the manhole CHAPTER IX CHEMICAL AND BIOLOGICAL CHARACTERISTICS OF SEWAGE The successful disposal of sewage depends upon the extent to which the changes that take place in its organic matter and bac- terial flora are controlled, so as to prevent the development of undesirable conditions. These changes are chemical and bac- teriological, closely interwoven in their nature, and affect the water-borne solids which constitute less than one thousandth part (0.1 of one per cent) of the total weight of the sewage, in most cases. Many sanitary investigations, the operation of sewage treat- ment plants, and sometimes the intelligent control of the disposal of sewage in large bodies of water, must depend to a considerable extent upon the information furnished by chemical analyses. Among the living things which play a part in the disposal of sewage are rats, the scavengers of sewers, in which many live; gulls and other birds which feed upon floating organic matter discharged from sewers; fish, which often congregate about sewer outlets for food; certain plants which thrive on some nitrogen compounds of sewage, and organisms too small to be seen without the aid of a magnifying glass or microscope. The last class of living things appears to be most important, and is separated into two groups, bacterial organisms and plankton. Bacteria are plants, studied with dif&ciflty with a microscope and investi- gated mainly by their behavior when cultivated under different, carefully standardized conditions. Plankton comprises both plants and animals, is easily studied with the microscope and does not require special culture. If sewage causes offense, certain bacteria are responsible; if it is oxidized to odorless substances possessing fertilizing value, certain other bacteria are the cause. Chemical analyses show what changes have been wrought in the sewage by the treatment, but they do not reveal the living cause of these changes, upon which successful sewage disposal depends. Chemical analysis of 280 CHEMICAL AND BIOLOGICAL CHARACTERISTICS 281 polluted water discloses its general character but not the presence or absence of disease germs. Where sewage has contaminated water not subsequently purified, the resulting ill effects upon the health of the drinkers of the water have been due, so far as is now known, to disease-producing bacteria. For this reason a study of bacteria is essential to a correct understanding of the sewage disposal problem. The changes taking place in sewage in filter beds are probably due in a very large degree to bacteria. When the sewage is discharged into rivers, ponds or the sea, the changes in it may be largely due to the plankton of these waters. This plankton lives to some extent upon the organic matter and certain parts of the inorganic matter of the sewage. The extent of the work accomplished by the plankton is still undetermined, for little is known of the fundamental principles upon which depend the development and control of these organisms. That their mission is to convert substances like those in sewage into living tissue is clear, but how to take advantage of this action, to accomplish the greatest good under the many conditions encountered, is not evident from present knowledge. Sampling. — The general character of sewage is best ascertained by mixing samples taken every half hour for 24 hr. and analyzing the composite sample. For very accurate determinations, the quantities of half-hourly samples mixed together should be pro- portionate to the flow of sewage at the time of sampling. As some of the suspended matter may be quite coarse, it is desirable to take 1 qt. or more of sewage every half hour and pour it into a clean tub or barrel. The contents of the barrel are thoroughly mixed at the end of 24 hr., and from 2 to 4 qt. of the composite are taken for analysis. Care must also be exercised that the small samples taken for laboratory test are fairly representative of the large sample sent in. PoUuted water should ordinarily be analyzed within 12 hr. after sampling, and raw sewage and sewage effluents within 6 hr. Certain determinations, like that for dissolved oxygen, should be made on the spot. If the samples are kept at a low temperature, as in an ice box, the extent of the changes taking place in them wfll be lessened materially. If samples are shipped to a distant laboratory or are combined in a weekly composite, they must be sterilized by adding chloroform, formaldehyde, mercuric chloride or some other germicide. Certain sterilizing agents 282 SEWERAGE AND SEWAGE DISPOSAL interfere with some analytical determinations, so it is sometimes desirable to sterilize two different portions of the same sample, each with a different germicide, as formaldehyde and sulphuric acid. For strong sewage it may be necessary to use the equiva- lent of 1 to 2 c.c. of the concentrated chemical solution to each 500 c.c. of the sample, while one-half this quantity will suffice for good effluents and for polluted waters. MEANING OF CHEMICAL ANALYSES General Characteristics of Sewage. — "Concentration" is a term commonly used to designate the proportion of sewage matter to water in a given sewage. A strong or concentrated sewage contains a relatively larger proportion and a weak or dilute sewage a smaller proportion of sewage matter. There is no recognized standard by which a sewage may be classed as strong or weak. In sewage disposal problems, the organic mat- ter is more important than the mineral matter, and is therefore the usual basis for the classification, which takes into account the organic nitrogen, albuminoid ammonia, oxygen consumed, oxygen demand and volatile solids, the significance of which will be explained later in this chapter. Where any of these determinations is regularly made, it may be taken as a standard of strength although it is much wiser to base judgment upon a consideration of all of them, giving to each the proper weight under the circumstances. "Composition" is a term used to designate the different in- gredients in a given sewage and their amounts. If these amounts in the case of certain of the ingredients are high or low as com- pared with certain assumed standards, then this sewage can be termed strong or weak. Composition must be determined before concentration can be judged. The "condition" of a sewage depends upon the changes which have taken place in it since it entered the sewerage system. Although no new ingredients are added to it as it flows toward the treatment or disposal site, changes take place in its composi- tion, due largely to biological action. The changes are mainly in the organic matter, which is made up of complex compounds of carbon, hydrogen, oxygen, nitrogen, sulphur and other elements, more or less readily broken up into other and usually simpler compounds by chemical and biological action. Methods of CHEMICAL AND BIOLOGICAL CHARACTERISTICS 283 chemical analysis are largely dependent upon such chemical action, while it is believed that some sewage treatment processes are mainly dependent upon biological action. The objectionable characteristics of sewage must be overcome by removing the matter causing them and disposing of it apart from the rest of the sewage; by transforming it into unobjectionable matter; or by killing directly or indirectly the objectionable living or- ganisms. Frequently all three methods are utilized in a single treatment plant. Solids or Residue on Evaporation. — This residue is determined by evaporating down a known volume of sewage, and drying the residue at 212°F. In the drying, all gases and small quantities of very volatile substances escape, but their relative quantity is usually very small. The weight of this dry residue is compared with the weight of the original sample of sewage to determine the solids in parts per 1,000,000, the usual method of recording the results of such analyses. Suspended and Dissolved Solids. — The total residue on evapo- ration contains the solids which were dissolved in the sewage and those which were in suspension or carried along by it without being dissolved. In many sewage treatment problems, the solids in suspension are more important than those in solution. Their quantity is determined by filtering the sewage, evaporating the filtrate to dryness and determining the dissolved solids by weighing the residue thus obtained, and finally subtracting the quantity of dissolved solids thus determined from the total solids as determined in the manner described in the previous paragraph. The suspended solids are often determined directly by filtering a known volume of sewage through a Gooch crucible, which consists of a crucible with a perforated bottom provided with an asbestos filter. The residue collecting on the asbestos mat is dried at 212°r. to constant weight and the suspended solids are calculated. This method usually gives higher results than those obtained by the difference of the total sohds and dissolved solids found by filtering through ordinary filter paper described above. Extreme care must be taken in all the steps from the beginning of sampling to the final weighing in order that the proper pro- portion of all the suspended matter may be included in the final residue and that changes shall not take place in the colloidal 284 SEWERAGE AND SEWAGE DISPOSAL matter during the period of sampling and analysis. (See follow- ing paragraph on "Colloidal Matter.") Settling Solids. — The determination of the suspended solids in sewage and in the effluent from the plant where the sewage is treated shows the effect of the treatment on this important part of the matter which the sewage contains. If the sewage is passed through a sand filter practically all the suspended solids are removed, but if the sewage is merely allowed to settle for a few hours in a sedimentation basin smaller proportions of the suspended solids will be removed. The efficiency of sedimenta- tion is sometimes expressed by noting the percentage of the suspended solids removed, but this may be misleading because it is impossible to remove all the suspended solids by any practi- cal method of sedimentation. It may be preferable, therefore, in some cases to state what percentage of the solids removed in the laboratory under standard conditions by sedimentation are actually removed in the settling basins, since this figure will show the practical efficiency of the process. In the operation of the large sewage treatment works at Providence, R. I., this determination is made in the following manner: The sample of the sewage which represents its character during each 24-hr. period is well shaken and a portion of this is analyzed to deter- mine the suspended solids in the raw sewage as it arrives at the works. The 24-hr. sample is allowed to stand quiet for 4 hr., the time taken by the sewage to pass through the sedimentation basins, and another portion of the sample is then decanted and its suspended solids determined. The difference between the two determinations is assumed to be the quantity of suspended matter in the Providence sewage which will settle in 4 hr., and the actual performance of the settling basins is referred to this quantity in order to ascertain how closely the sedimentation in the basins approached the practicable maximum. Another method of determining the quantity of settling solids is to place a measured volume of sewage in a conical glass, 4 in. in diameter at the top, the size of a thimble at the bottom, 17 in. tall and graduated at the lower part to show the number of cubic centimeters below each mark. A liter sample of sewage is placed in one of these cones, called Imhoff glasses, and allowed to stand for the number of hours which the sewage is assumed to remain in the sedimentation basins. At the end of this time there will be a certain amount of sludge in the bottom of the glass CHEMICAL AND BIOLOGICAL CHARACTERISTICS 285 and the quantity is measured by the graduations. This method of determination is liable to serious errors if the suspended matter is coarse, for the larger pieces tend to clog the narrow tapering bottom of the glass and thus prevent the solids from filling the space below them. For this reason such glasses are more useful in determining the settling solids in effluents than in raw sewage. Colloidal Matter. — ^As a practical matter, a substantial por- tion of the solids usually reported as being in suspension when the determination is made by filtration, evaporating and weigh- ing, may be considered as being in a colloidal state. Although there is no standard method at present of determining the quan- tity of colloidal matter, it is certain that such substances are of much importance in sewage treatment, especially where industrial wastes are discharged into the sewers. Colloids are substances, it should be kept in mind, which while soluble as judged by ordinary physical tests will not pass through a parchment mem- brane. Substances which, while in solution, will pass through such a membrane are termed crystalloids. Practically speaking, the term "colloids" is often used in discussing sewage treatment problems to include the very finely divided suspended matter which will not readily settle in ordinary sedimentation basins, and true colloid substances, which can hardly be classed as finely divided suspended matter in the sense that they are visible to the naked eye, but consist of substances which, under certain conditions, may be thrown out of their state of pseudo-solution and retained mechanically upon the surfaces of tank walls or filtering materials. Fixed Residue, Fixed Solids or Mineral Matter. — A definite quantity of sewage is evaporated to dryness, the residue weighed to give the total solids, and this residue is then ignited at a rela- tively low temperature to burn off the organic matter. The weight of the remaining matter gives approximately the amount of mineral matter in the original sample. Loss on Ignition, or Organic Matter.^ — The difference between the weights of the total residue and the fixed residue represents approximately the quantity of organic matter in the original sample. The test merely gives an idea of the quantity but not of the character of the organic matter present in the sample, and thus does not show how this matter will behave under arti- ficial or natural conditions. This behavior is very important 286 SEWERAGE AND SEWAGE DISPOSAL in sewage treatment and can be best illustrated by explaining the nitrogen cycle of change in organic matter. Nitrogen Cycle. — The nitrogen cycle is shown diagrammati- cally in Fig. 140. Dead organic matter is decomposed through the process of decay, sometimes accompanied by putrefaction, and the nitrogen it contained appears first as free ammonia. Later oxidation proceeds and the ammonia is first changed to nitrites and then to nitrates. The nitrates are good food for plants, which convert them into living complex nitrogenous organic matter or proteins. ' ° f"" ' Eventually the plants die and decay or they are eaten by ^''"^''"^ ji^iasiiop I ^^\^mmonh animals, a part being con- verted into animal tissue which will later die and decay. Another part is converted into ftateim \^' || ^ Nifriies waste substances like urea, ^^ which is easily oxidized, and '^NifraiZ Js thus chaugod into ammonia Fig. 14 0. — Nitrogen cycle (after without going through the ^'''^T^- ,,^Tr'""'^J,«'^'''°'°^^'"^'^'*^'^ complete cycle. There is an- by Marshall, page 126.) .... other deviation from the com- plete cycle due to the fact that certain bacteria reduce nitrates to nitrogen gas, a process termed "denitrification." If there were no compensation for this the amount of nitrogen in the soil available for vegetation would gradually become exhausted, but fortunately there are other bacteria which have the power to use the atmospheric nitrogen for the formation of their own sub- stance, a process termed "nitrogen fixation." As discharged into the sewers, some of the organic nitrogenous substances are relatively stable and others only loosely bound together. Some of the nitrogen is even present in inorganic form, as ammonia and ammonium compounds. The organic substances are easily broken up directly or indirectly by bacterial action, into other substances. Some of the nitrogen which at first was loosely bound is liberated from its organic combina- tion and may escape as gaseous nitrogen or it may be changed into some inorganic form such as ammonia. The nitrogen of other nitrogenous substances may first become part of less complex and less stable matters only to be changed later into gaseous nitrogen or into inorganic matter if the process is allowed CHEMICAL AND BIOLOGICAL CHARACTERISTICS 287 to proceed far enough. However, some nitrogenous organic compounds will persist for a very long time. In the treatment of sewage by bacterial action, much depends upon the conditions under which the processes are carried on. Sometimes the action is conducted in the absence of air, as in airtight or air-trapped tanks, and the changes are confined to an interchange of the elementary chemical substances present in the sewage. This process is often termed "hydrolysis," because it is generally accompanied by a breaking up of the molecules of water and the combination of the hydrogen and oxygen of the water with carbon, nitrogen and other substances from the organic matter. In other cases, the bacterial action goes on in the presence of an abundant supply of oxygen, furnished either directly from the atmosphere or from that dissolved in relatively pure water. .Under such circumstances the action is one of oxidation, the organic substances being split up and com- bining with the oxygen, the nitrogen finally appearing as nitrites or nitrates of some base or alkali, as sodium nitrite (NaN02) and sodium nitrate (NaNOa). There are accurate, convenient methods for determining the quantity of nitrogen present in sewage and of ascertaining whether it is in relatively stable or unstable combination as organic matter or in any one of the three combinations just mentioned as indicating progressive change from the organic substance to the highly oxidized mineral compound, nitrates. It is this characteristic of the determination of nitrogen com- pounds which has given them so much importance in sewage analysis, rather than any particular significance in the quantity of nitrogen present. If specific information is desired as to the quantity of trouble-making matter present, the determination of organic sulphur may throw more light on the problem than the nitrogen determination. Organic or Kjeldahl Nitrogen.— This is the nitrogen combined with carbon and other elements in the form of organic matter; it does not include nitrogen as free ammonia, nitrites and nitrates. The total quantity of nitrogen is determined by prolonged boiling with strong acid and finally distilling the ammonia thus formed. This treatment is severe enough to break up the more stable organic nitrogenous substances and to liberate the nitrogen in the form of free ammonia. By determining the total organic matter in a filtered sample information is obtained concerning 288 SEWERAGE AND SEWAGE DISPOSAL the -quantity of dissolved nitrogenous organic matter, and, by diiference, the quantity of such matter in suspension in the sewage. Albuminoid Ammonia. — Organic nitrogenous substances may be divided into two classes, those readily jdelding ammonia by boiling with alkaline permanganate of potash for a short time, and those which are not so decomposed. The nitrogen from the latter class is sometimes called "residual organic nitrogen." The ammonia formed from the former class by the boiling process mentioned is distilled off and accurately measured, the result being usually reported as "nitrogen as albuminoid ammonia" because the nitrogen of albumin may be driven off in this form by similar treatment. Albuminoid ammonia does not exist as such a compound in sewage, but the test results are a measure of the quantity of the less stable nitrogenous organic substances present. If the results are reported as ammonia, as they some- times are, they maybe expressed in terms ofnitrogen by multiply- ing the amount of albuminoid ammonia by the factor ^fij =0.82. A knowledge of the quantity of nitrogen as albuminoid am- monia in sewage and effluents may be of considerable value to the operator of treatment plants, and in many cases it is well worth determining and may be more instructive than the de- termination of total organic nitrogen. In studying sewages from different cities or the condition of sewage from the same city at different times, it is often desirable to determine both the nitrogen as albuminoid ammonia and the total organic nitro- gen. By deducting the former from the latter some idea may be formed of the relative quantity of the more stable forms. Free Ammonia. — For many years free ammonia has been determined by distillation, but the method is uncertain because some of the unstable substances in sewage, such as urea, give off their nitrogen when heated to the degree required for distillation. It is practicable in many cases to determine the free ammonia directly in the sample, without distillation, and where the results are reliable this method is preferable. As free ammonia is a decomposition product, the quantity present is a valuable indication of the freshness of the sewage. The fluctuations in the quantity of free ammonia in the effluents from some treatment processes are a helpful guide to the efficiency of the biological action upon which the treatment depends. While free ammonia may or may not indicate an unfavorable CHEMICAL AND BIOLOGICAL CHARACTERISTICS 289 condition of the treatment plant, it does not of itself constitute a source of offensive odor and it does not ordinarily represent organic matter which will decompose and create offensive conditions. Nitrites and Nitrates. — The determination of nitrogen as nitrites and nitrates is instructive as regards the efficiency of the last step in the complete biological purification of sewage. The quantity of nitrites is determined by adding to the sample quantities of acetic acid solutions of sulphanilic acid and a- naphthylamine, which produce a pink color if nitrites are present. The depth of color thus produced is compared with a set of standards containing progressively increasing quantities of sodium nitrite and treated in the same way. The sample con- tains the same quantity of nitrites as the standard it matches. Nitrates may be determined by a colorimetric method similar to that used for determining nitrites. The sample after evapora- tion is treated with phenolsulphonic acid and made slightly alkaline. If nitrates are present a yellow color is produced, the depth of which, when compared with similarly treated stand- ard solutions of potassium nitrate of known strength, indicates the quantity of nitrates in the sample. This method is not applicable to water containing more than 30 parts per million of chlorine, as is sometimes the case with sewage and often with industrial v/astes. Such samples should be treated by the reduction method, by which the nitrates are reduced to ammonia by nascent hydrogen generated from strips of aluminum foil placed in the samples after they have been made alkaline with potassium hydrate. After reduction the ammonia is distiUed off and measured as in the determination of free ammonia. Oxygen Consumed. — Determinations of nitrogen do not give any information about the quantity of organic matter in which nitrogen is absent. As a measure of the carbonaceous as well as nitrogenous organic matter, the so-called determination of "oxygen consumed," "oxygen absorbed" or "oxygen required" has long been used. The sample for this test is accurately measured, acidulated and treated with a known quantity of a solution of potassium permanganate of standard strength. It is then placed in a bath of boihng water and digested for exactly 30 min., in the standard method of the American Public Health Association, although the practice in some laboratories differs at this stage of the test. The quantity of the oxidizing agent 19 290 SEWERAGE AND SEWAGE DISPOSAL remaining after digesting is measured and the quantity of oxygen used up or absorbed by the sample is computed. This determination is, in a way, Kke the determination of the loss on ignition, except that it is wet instead of dry combustion, and is carried out in such a manner that the quantity of oxygen ab- sorbed is under control and subject to accurate measurement. Unfortunately the combustion is never complete, so that the quantity of oxygen absorbed is only that used up by the more easily attacked substances and the quantity consumed depends largely upon the length of time the digestion is continued. Thus many different results may be obtained from the treatment of the same sample for different periods of time, and different sew- ages may yield widely different results with the same treatment, depending upon the proportion of the readily oxidizable sub- stances contained in them. This determination is, therefore, of most service when used for comparing the quality of sewages from the same community from day to day, or comparing a sewage with the corresponding effluent from a plant in which it has been treated. It must not be forgotten, however, that the organic matter indicated by the oxygen-consumed test of an effluent may not be in the same form as that in the original sewage. Putrescibility Tests. — The consumption of oxygen in the oxygen consumed test, due to chemical action, is paralleled in nature when the original organic substances are broken up, largely by the action of living organisms, and the carbon and oxygen are united in a stable combination, substantially un- attended by disagreeable odors. The oxygen required to oxidize the dissolved and suspended matters is derived from the supply dissolved in the water, including the oxygen combined with nitrogen in the form of nitrites and nitrates. If there is an insufficient supply of this oxygen, an entirely different biolog- ical process will set in, due like that of oxidation to the work of living organisms, but of a different kind. The organic matter wiU be broken up as in the other process, but the products of this process will contain among them offensive smelling matter such as hydrogen sulphide, not formed in the presence of an abundant supply of oxygen. As oxygen is required to prevent putrefaction and as water can dissolve only a limited quantity of oxygen, an effort has been made to determine the stability or putrescibiUty of samples CHEMICAL AND BIOLOGICAL CHARACTERISTICS 291 by determining the quantity of oxygen dissolved in them after standing for different periods of time, adding thereto the oxygen available from nitrites and nitrates, although the quantity which can be derived from the latter is often negligible. By observing the loss in oxygen due to the changes going on in the sample, a knowledge of its putrescibility may be obtained. If the oxygen is not all used up the sample will prove stable. The standard putrescibiUty test of the American Public Health Association is made by placing a sample of the sewage in a 150- to 200-c.c. glass-stoppered bottle to which a small quantity of methylene blue is added. The sample is then incubated, preferably at 20°C. during 4 days, observations of the color being made at least once a day. If the sample is stable the blue color of the chemical will persist. If the sample is putrescible, the color will disappear on account of the reducing action of the sulphur compounds, provided the sewage or effluent contains no trace of hypochlorites which may have been used in disin- fecting it. This test is based on the action of the organisms in the sample. If there is sufficient organic matter of the right kind to maintain their existence in adequate numbers, they will ex- haust the supply of oxygen and the biological action taking place will be changed from oxidation to putrefaction. If the quantity and nature of the organic matter are unsuited for the rapid growth of the bacteria, there will be an ample supply of oxygen to meet the requirements of such changes as go on in the organic matter, and the process will be one of oxidation through- out and will not be attended by the production of offensive odors. Table 37 gives the relative stability numbers corresponding to the time of incubation at both 20°C. (68°F.) and 37°C. (98.6°F. blood temperature). The relative stability number is assumed to indicate the ratio of available oxygen in the sample to that required for the complete oxidation of the organic matter. For example, if a sample of effluent when incubated at 20°C. retains its color only until the end of the third day, it is said to have a stability number of 50, which means that it contained, as disolved oxygen and oxygen in nitrites and nitrates, one-half of the oxygen required for complete oxidation of the organic matter. The American Public Health Association's committee which recommended this test stated that if a sample retains its color for 4 days it may be considered as practically stable, except 292 SEWERAGE AND SEWAGE DISPOSAL where great accuracy is desired, but the authors believe that the additional information obtained by continuing the incubation for 14 days at 20°C. is well worth the labor involved. Table 37. — Relative Stability Numbers ( "Standard Methods of Water Analysis," American Public Healtli Association) (20 hi s £20 (37 S (20 taj S <20 «37 S 0.5 11 3.0 1.5 50 8.0 4.0 84 13.0 6.5 95 1.0 0.5 21 4.0 2.0 60 9.0 4.5 87 14.0 7.0 96 1.5 30 5.0 2.5 68 10.0 5.0 90 16.0 8.0 97 2.0 1.0 37 6.0 3.0 75 11.0 5 5 92 18.0 9.0 98 2.5 44 7.0 3.5 80 12.0 6.0 94 20.0 10.0 99 S = Relative stability or ratio of available oxygen to oxygen required for equilibrium ; expressed in percentages. /20 = Time in days to decolorize methylene blue at 20°C. (37 == Time to decolorize at 37°C. This methylene-blue test requires only simple, inexpensive apparatus and can be carried out by persons without a knowledge of chemistry. It has been quite generally adopted as a routine test for the control of sewage treatment and dilution. Where merely treatment to insure freedom from offensive conditions is needed, it will be found a useful test and perhaps, in some cases, the only one required. Biochemical Oxygen Demand. — The putrescibility test shows the relative amount of oxygen required to prevent putrefactive conditions. It is often important, however, to measure the actual oxygen requirements of a sewage, effluent or trade waste. The amount of oxygen used up by a sewage or waste, in a stated period of time at a definite temperature, usually 20°C., is called the "biochemical oxygen demand." It has been defined by the U. S. Public Health Service as follows: "The biochemical oxygen demand of a polluted water, a sewage or an industrial waste, is defined as the milligrams of oxygen per liter required for the stabilization of its organic matter by aerobic bacterial action." This may be determined by the following methods: The relative stability method consists in adding to the sample to be examined a proper volume of aerated tap water of known dissolved oxygen content, a small amount of methylene blue, and incubating the mixture at 20°C. until the blue color disap- CHEMICAL AND BIOLOGICAL CHARACTERISTICS 293 pears. The biochemical oxygen demand is then calculated from the initial dissolved oxygen content, the relative stability num- ber, the ratio of available oxygen to oxygen required for equi- librium or stability, and the dilution. The method described above is often modified by omitting the methylene blue and adding to the sample sufficient aerated tap water to ensure residual dissolved oxygen (of about 30 per cent saturation) at the end of a 10-day period of incubation. The difference between the oxygen in the initial diluted sample, and in the sample after incubation, gives the amount of oxygen re- quired by the amount of the original sample that was used for dilution. The nitrate method is based on the biochemical consumption of oxygen from sodium nitrate. In this case, a measured amount of sodium nitrate solution is added to the sewage in tightly stoppered bottles and incubated at 20''C. for 10 days. At the end of the period of incubation, the residual nitrite and nitrate is determined in the sample. The biochemical oxygen demand is the difference between the available oxygen added as sodium nitrate and that found at the end of the period of incubation as nitrates and nitrites. In the determination of the biochemical oxygen demand of sewage or of certain trade wastes, it is usually not necessary to include in the initial oxygen content, the amount of oxygen present in the sample in the form of dissolved oxygen, nitrites or nitrates, as the amount present is relatively small as compared with the amount required. However, in the determination of the biochemical oxygen demand of effluents, polluted streams, or relatively weak trade wastes that have a comparatively low demand for oxygen, it becomes necessary to determine the amount present in the form of dissolved oxygen, nitrites and nitrates and include it in the initial oxygen content. It should be noted here, that in making this determination a sterilized sample cannot be used and if the sample is acid or possesses hydrate alkalinity, bacterial activity will be inhibited. In order to obtain comparable results by any one method, care must be taken that the temperature of incubation remains practically constant, that no oxygen is Uberated during incu- bation and that the time of incubation is the same in all cases. It is not possible to obtain results that check closely on all samples when determined by the different methods, due possibly 294 SEWERAGE AND SEWAGE DISPOSAL to the selective action of the bacteria as well as to the variation in the speed of the reaction as the supply of available oxygen is diminished. Dissolved Oxygen. — It is important not to confuse the deter- mination of dissolved oxygen with that reported as oxygen "consumed," "required" or "absorbed." The results reported as dissolved oxygen are obtained by the determination of the quantity of atmospheric oxygen which is dissolved in a given sample of water or sewage. Pure water constantly in con- tact with the atmosphere is capable of dissolving only a definite quantity of oxygen at any temperature and atmospheric pressure. The quantity varies inversely with the temperature. Pure water dissolves more oxygen than does impure water. Soft surface water wUl dissolve more than hard ground water of sea water. Water is said to be saturated with dissolved oxygen when it contains all the quantity it is capable of dis- solving when intimately mixed with air at the specified tem- perature and pressure. Water usually contains less than this amount and the results of dissolved oxygen tests are reported in percentages of saturation, as 60 per cent saturation. This method of stating the results may cause confusion unless the facts just stated are kept in mind. For example, if the oxygen has been used up by biological processes in two waters at tem- peratures of 15 and 27°C. respectively, until the percentage of saturation has fallen to 50 in each case, it is not true that the same quantity of oxygen has been used up in the two cases nor that the same quantity remains. The dissolved oxygen remain- ing is 5.07 and 4.03 parts per million respectively. The water at the lower temperature contains about 25 per cent more oxygen than the other in spite of the fact that its loss of oxygen has been 25 per cent greater. Chlorine. — Common salt, sodium chloride, is present in all water supplies and is also an ingredient of food. It occurs in relatively large quantities in kitchen wastes, wash waters, urine and feces. Salt is therefore found in greater quantities in sewage than in most natural waters, and its presence in the latter in excess of that normal to them indicates their probable con- tamination by sewage. As the chlorine which is combined with the sodium in salt can be easily and accurately determined, it is customary to analyze the water or sewage for this only, disregard- ing the sodium and reporting the result in terms of chlorine. CHEMICAL AND BIOLOGICAL CHARACTERISTICS 295 The quantity of chlorine per inhabitant reaching the sewers in a given unit of time is practically uniform; therefore the chlorine content of the sewage may be used to measure its con- centration. To do this it is necessary to know the quantity of chlorine in the water supply and also any possible source of an unusually large quantity of chlorine discharged into the sewers. Similarly the quantity of water entering the sewers may some- times be ascertained by the degree of dilution afforded the sew- age, the chlorine content of the ground water being known. The proportion of ground water in the effluents of filters and irrigation fields can be ascertained in the same way. Appearance. — The turbidity, sediment and color of the efflu- ents from sewage treatment works and of water into which sewage and effluents are discharged are frequently observed and recorded along with the results of chemical analyses. It is generally sufficient to designate the turbidity as "slight," "distinct," "decided," "milky" and the hke. Sediment is more rarely reported, and then it is sufficient to record it as "shght," "distinct" or "decided." Where industrial wastes enter the sewage it may be important to report the color in general terms, but this is not usually desirable. Odor.— The odor of an effiuent or of water containing sewage or effluent may furnish valuable information, but this is rarely the case with sewage. The odor is best determined by half filling a bottle holding 2 to 4 qt. with the water or effluent, cork- ing it, shaking it vigorously for about a minute, and then apply- ing the nose quickly to the mouth of the bottle. Where the odors are due to gases dissolved in the water they are sometimes so completely liberated by the shaking that a second shaking will produce no further odor, and for this reason a second sample should be used in case of doubt. Some samples which yield only a slight odor when cold will give off a decided odor when heated. While it is difficult to convey to others the exact meaning of arbitrary terms adopted for designating the character and strength of odors the authors have found the observation of odors helpful, particularly to the operator of filters of the types commonly used. Many years ago they adopted in connection with the operation of intermittent sand filters at Worcester, Mass., a scale of odors of the effluent to indicate the efficiency and condition of the ffiters. This scale was: 0, no odor; 1, 296 SEWERAGE AND SEWAGE DISPOSAL slightly musty; 2, distinctly musty; 3, decidedly musty; 4, offensive; 5, very offensive. With these observations of odor there were parallel tests of the putrescibility of the effluent. Such samples as had no odor or but a slight musty odor were always non-putrescible. Samples which were 3 or higher on the scale of odor were putrescible almost without exception, while samples observed as 2 were generally non-putrescible. Such a scale of odors can be worked out by which the quality of the effluent from other treatment plants can be judged fairly well without waiting for the results of chemical analyses. CHANGES IN SEWAGE CAUSED BY BACTERIA Comparatively little study has been devoted to the isolation, identification and life processes of the species of sewage bacteria. The natural biological treatment of sewage appears to be through such control of conditions as wiU cause the vigorous predominant growth of species capable of bringing about desired changes. Such treatment is governed by more or less accurate knowledge of the work accomplished by masses of bacteria and the results to be accomplished by providing different working conditions for them, rather than by a knowledge of bacteria themselves and of their individual characteristics. For such purposes, the grouping of bacteria into three classes originally proposed by Pasteur is more helpful than some of the later classifications made by bacteriologists. Pasteur called those bacteria "aero- bic" which can exercise their functions only when air is present; others, called "anaerobic," can function only when all oxygen is absent; the third class, called "facultative," can function under either aerobic or anaerobic conditions although not always with equal vigor. Number of Bacteria in Sewage. — The organic and mineral matter in sewage usually affords suitable food for bacteria, which multiply in it with extraordinary rapidity at a favorable temperature, such as 20°C. Experiments made at Lawrence' with a sample of sewage placed in a bottle and kept at a tem- perature of 16° to 20°C. showed an increase in the number of bacteria from 1,190,000 per cubic centimeter after a lapse of 2 hr. to 23,475,000 after 253^ hr. Then the number began to decrease and at the end of 192 hr. it was 2,341,000. This decrease was probably due in part to the diminution in the food ■ Report Mass. State Board of Health, 1894, p. 461. CHEMICAL AND BIOLOGICAL CHARACTERISTICS 297 supply and the dissolved oxygen, which disappeared entirely in less than 24 hr. after the sample was taken, and in part to the increase in the quantity of those products of bacterial action which check such action. Such figures show that the number of bacteria to be expected in sewage will vary with the age of the sewage at the place where samples are taken. Manifestly it will also be affected by the dilution of the sewage by water. It was found at the Lawrence Experiment Station that fresh sewage from toilet rooms contained an average of 2,850,000 to 3,160,000 bacteria per cubic centimeter, while sewage sampled after flowing from the city to the Station averaged 1,136,000 to 2,110,000 bacteria. At Columbus, Ohio, George A. Johnson found the number was 320,000 to 27,000,000, averaging 3,600,000. In Philadelphia, George S. Webster found from 1,100,000 to 5,800,000, the average being 3,000,000. The more concentrated sewage of English cities contains from 10,000,000 to 100,000,000 bacteria, according to the Royal Commission on Sewage Disposal. The number varies also with the hour at which the samples are taken. The number of bacteria may also be affected by industrial wastes discharged into the sewers. Wastes from packing houses and some tanneries contain materials stimulating the increase of bacteria, whereas piclding liquids from galvanizing works, being germicides, tend to reduce the number. Paper-mill wastes are practically free from bacteria in some cases, and if organisms are introduced into them, the bacteria will not multiply sufficiently in a reasonable length of time to be effective in causing biological changes, unless the wastes are diluted with water or sewage or treated chemically to overcome their germicidal nature. Breaking Down of Complex Substances. — The bacterial de- composition of sewage is often carried out by several species of organisms working in relays, so to speak. Each utilizes in turn some of the decomposition products of the life processes of the preceding organisms. If it were possible to control these proc- esses, sewage might be passed from tank to tank or filter to filter, in each of which only one process would take place. The multiplication of bacteria in sewage is attended by the formation of ammonia, carbon dioxide, and sometimes hydrogen, nitrogen and methane. Some organisms produce hydrogen sulphide. In all these changes, it is important to remember, there is no loss of matter, only a change in its form and character- 298 SEWERAGE AND SEWAGE DISPOSAL Carbon Carbohydrafes Fcrfs, FroMns 141. — Carbon cycle (after Rahn). (From "Microbiology," edited by Marshall, page 125.) FlQ. istics. The changes in nitrogenous matter were shown in Fig. 140, page 286. A somewhat similar change occurs in carbonaceous matter, illustrated in Fig. 141. At the death of the organisms the first stage in the cycle of change is that of decay, when putre- faction may perform an important function. This stage ends Dzad Or^ani^ms ^1*^ "^^^ Oxidation of the car- bon into carbon dioxide gas or carbonates. In the form of carbon dioxide the carbon is food for green-plant life and is converted by it into such complex organic substances as carbohydrates, fats and pro- teins. The mature plants die and the elements of which they are composed may at once recommence the cycle or the plants may serve as food for animals, by which they are converted into animal tissue such as fats and proteins. Another process of evolution important in sewage disposal problems is the sulphur cycle, Fig. 142. Sulphur is always present in sewage ready to enter into any biological process, although varying greatly in quantity. The oxygen dissolved in sewage is soon exhausted by the bacterial processes going on and the organisms then appear to appropriate the oxygen of DeadOr3amims some of the mineral and organic substances in the sew- age. If sulphates are present they will yield their oxygen, and the resulting sulphides may be decomposed with the formation of hydrogen sul- phide and the production of offensive odors. These odors may be due not only to hydro- gen sulphide but also to volatile substances such as indol, skatol, cadaverin and mercaptan, about which little is known. If sewage can be supplied with dissolved oxygen, its decomposition will not be attended by the offensive odors so evident when it Frofeins Proteins Hijdrogeri 5ulphide Sulphur Sulphates Fig. 142. — Sulphur cycle Rahn). (From " Microbiology,' by Marshall, page 127.) (after edited CHEMICAL AND BIOLOGICAL CHARACTERISTICS 299 is undergoing anaerobic changes, and therefore an effort is usually made to bring about the desired changes in sewage under aerobic conditions. After the organic sulphur-bearing compounds become mineral- ized, with the formation of hydrogen sulphide, which does not appear suitable for the nutrition of higher plants, another trans- formation is brought about by the so-called sulphur bacteria, capable of oxidizing the hydrogen sulphide, provided oxygen is present for this process. These organisms produce sulphates, which are taken up by plants, and the cycle thereafter continues like those already explained. Decomposition of Sewage with Diminution of Oxygen. — There are several terms much used in connection with the treatment and disposal of sewage whose significance is based upon the bacterial changes described in this chapter. The sewage which reaches the public sewers contains free oxygen so that aerobic bacteria thrive, and in this condition is termed "fresh." Unless sewage is vigorously agitated it absorbs oxygen from the air very slowly, and in the course of several hours the oxygen is greatly depleted or exhausted, if any aerobic bacteria are present their vitality is waning, and the sewage is termed "stale." The facultative bacteria have been thriving while these changes were taking place. After the last traces of oxygen disappear the sewage becomes very dark in color and various sulphides, including hydrogen sulphides, are formed. Sewage in this stage is termed "septic." When sewage stands for a long time or flows very slowly through tanks, the anaerobic organisms gain the ascendancy, and ammonia, marsh gas, carbon dioxide, hydrogen, nitrogen and other gases may be formed as the complex organic sub- stances are broken down into simpler ones. This is a reduction or septic process. Oxidation Processes. — The process of oxidation of free ammonia to nitrites and nitrates, already explained in its chemical features, is called "nitrification." That it is due to bacterial action is proved by the fact that it is checked or stopped entirely by low temperatures, by heating to the temperature of steriliza- tion, and by germicides. Furthermore, temperatures known to be favorable to bacterial activity cause nitrification to proceed rapidlj'. In treating sewage by irrigation and intermittent filtration 300 SEWERAGE AND SEWAGE DISPOSAL the production of high nitrates in the effluent is evidence of practi- cally complete oxidation. Where sewage is applied to beds of very coarse material and conditions do not closely simulate those of natural soil, nitrates are not always found in the effluent in such large quantities and consequently organisms producing them are not always regarded as so essential as formerly. A number of species of bacteria called nitrite-formers have been isolated, but apparently only one species is present in one locality. Like the sulphur-oxidizing bacteria, they thrive on inorganic matter and are unfavorably affected by organic matter. Their power to convert ammonia into nitrite is apparently dependent upon the presence of carbon dioxide. The nitrate-formers are also capable of growing upon inorganic media. They can obtain their carbon from carbon dioxide or from loosely bound carbonates. While not so sensitive to organic matter as are the nitrite-formers, its presence interferes with their development although not to such a degree with their ability to oxidize nitrites. A small quantity of ammonia will arrest their development. Under certain conditions a reduction or deoxidation process may become established in some types of sewage fflters. The nitrates and nitrites appear to lose oxygen, and ammonia or nitrogen or both may be liberated. These changes may be due to strictly chemical reactions or may be due to the action of some species of bacteria; they are termed "denitrification" and the organisms producing them grow under both aerobic and anaero- bic conditions provided nitrates or nitrites are present. Detection of Contamination by Test Bacteria. — Bacterial analysis is capable of disclosing the presence of much smaller proportions of sewage in water than can be detected by chemical analysis. The contamination is traced by species of bacteria originating in the human intestinal tract. Of the three impor- tant groups of sewage bacteria, B. coli, B. enteritidis sporogenes and sewage streptococci, the first is universally found in human urine and feces and is the commonly accepted test organism. There are many varieties of coli bacteria, none of which is considered harmful to man. Some of these varieties are much more significant of the contamination of water by sewage than others, and in stating the number of B. coli present in a sample of water, only those varieties which presumably indicate sewage contamination are reported. The importance of these tests is CHEMICAL AND BIOLOGICAL CHARACTERISTICS 301 due to the fact that sewage contamination may mean that the germs of water-borne diseases are finding their way into water used for drinlcing purposes. The relative degree of contamina- tion is indicated by the numbers of B. coh and of all species of bacteria in the different samples of water. A rough classification of different degrees of contamination, made for the International Joint Commission! by A. J. McLaughlin, John W. F. Mc- Cullough, John A. Amyot and Frederick A. Dallyn is as follows : 1. Relatively pure waters; under 2 B. coli per 100 c.c, under. 10 bacteria of all kinds per cubic centimeter on agar at 37°C. 2. SHght pollution of relatively pure water; 2 to 10 B. coli per 100 c.c, 10 to 25 bacteria of all kinds per cubic centimeter. 3. Considerable pollution; 10 to 20 B. coli per 100 c.c; 25 to 50 bacteria of all kinds per cubic centimeter. 4. Serious pollution; 20 to 50 B. coh per 100 c.c; 50 to 100 bacteria of aU kinds per cubic centimeter. 5. Gross pollution; over 50 B. coh per 100 c.c; over 100 bacteria of all kinds per cubic centimeter. Tjrphoid Fever. — B. typhosus, the baciUus of typhoid fever, is present in enormous numbers in the feces and urine of persons having that disease. Some of these persons continue to discharge the bacilli long after they have apparently returned to normal health and are called "typhoid carriers" in consequence. When the bacilli are discharged into water some of them may hve for weeks and months, and if this water is drunk the drinker is in danger of becoming infected. In a general way, B. coh and B. typhosus show much the same life characteristics, the former being somewhat more resistant to unfavorable conditions. While the germs of the disease are transmitted by milk, shellfish, flies, new vegetables and fruits, by contact, and by various means now unknown, the chief mode of transmittal in the past has been by water. It is certain that the contamination of municipal water supplies by sewage is hkely to result in ty- phoid epidemics^ because of the large number of persons exposed to the danger of infection. Sewage should be disposed of and water supplies procured in such a manner as to reduce so far as ■ "Progress Report," Jan. 14, 1914, p. 20. 3 A summary of epidemics traced to this cause down to 1908 is given in "American Sewer- age Practice," vol. iii, p. 122. The typhoid fever death rate in 51 American cities in 1900 to 1912 inclusive is given on pages 118 and 119 of that volume. The death rate in all leading American cities is published annually in the Jour. Am. Med. Assoc. 302 SEWERAGE AND SEWAGE DISPOSAL practicable the danger of spreading this justly dreaded disease and others, such as cholera and dysentery, similarly communicated. PLANKTON The present condition of our knowledge of the life processes of plankton is very unsatisfactory. It is known that at least some kinds of these organisms thrive best in waters rich in nitro- gen, and thus are found in great numbers in sewage-poUuted streams and ponds where there is also enough oxygen to support them, but vigorous growths of them are also found in ponds far removed from any possible contamination by sewage. Free ammonia has been repeatedly found high in waters containing large numbers of such organisms and low where they were few, but this ammonia may represent the products of decomposition of the organisms instead of their food. For the present, therefore, the chief practical value of our knowledge of plankton is the information afforded by observing the growths in streams and ponds receiving sewage and effluents, for these growths indicate certain desirable conditions. The algse and diatoms are the forms of plankton which apparently perform the most important part played by micro- organisms in changing inorganic into organic matter. There are some species which appear as thick mats or strands of light- or dark-green color in shallow streams, ditches and ponds; there is a blue-green group which occurs in rivers and lakes, and other groups which are found in heavily polluted pools and channels and leave black or red scum on the ground when the water dries up. These plants break up the carbon dioxide in the water, liberate the oxygen and convert the carbon into organic matter, and this in turn is converted within their cells into other organic substances with the help of nitrogen from the ammonia, nitrites and nitrates in the water. Some of these plant forms are also apparently able to absorb dissolved organic matter directly. An example of what such plankton can accomplish is afforded by the authors' observations of a small stream at Cincinnati. A sewerage system serving a population of about 1,700 persons discharged sewage into the upper reaches of this stream, which has a sluggish flow for about a mile below the sewer outlet and then spreads out into a shaUow creek with occasional pools. The odors from the water just below the sewer outlet were very offen- sive at times, causing much complaint. At the time of the inspec- CHEMICAL AND BIOLOGICAL CHARACTERISTICS 303 tion, the water near the outlet contained about 70 per cent of its saturation value of dissolved oxygen. Farther down stream green algse appeared and this growth increased in luxuriance as the distance from the outlet increased. Where it was heaviest enormous numbers of gas bubbles were thrown off by the green masses, which, when collected in a bottle and tested by ignition, appeared to be pure oxygen. At ' a distance of IM niiles below the outlet, the water was 130 per cent saturated and free from objectionable appearance. While, therefore, the green growth was indicative of pollution it also showed that active purification of the stream was taking place. An entirely different group of organisms, the euglenas, also produces a green scum on polluted water. It is sometimes classed as animal because it absorbs organic matter directly and some- times as vegetable because it gives off oxygen in the sunlight. It appears on pools of sewage and in stagnant parts of polluted streams. The animal micro-organisms live on bacteria and other organic matter, and if they are abundant in water it may be inferred, not only that many bacteria are present but also that the organic matter which is food for bacteria is present. An abundance of such organisms is, therefore, strong evidence of the contamina- tion of water. The larger animal organisms feed to a consider- able extent on the smaller forms, and young fish feed on these larger animalculae. If, however, the quantity of sewage dis- charged into the water steadily increases, a time comes when there is so much nitrogenous matter and so little available oxygen in the water that the micro-organisms decrease in num- bers. The work is too hard for them. With the disappearance of these minute oxygen-producing plants, the capacity of the water to absorb sewage is still further decreased and offensive anaerobic conditions are imminent. This was well shown by the way in which the progressive pollution of the head of Narragan- sett Bay, in Rhode Island, shifted seaward the frontier line of the region where algse and diatoms lived. COMPOSITION OF SEWAGE Sources of Total Solids. — Part of the total solids in sewage come from the water which is the principal constituent of the sewage (upwards of 99.9 per cent). The whole of the water supplied to consumers does not reach the sewers and the portion 304 SEWERAGE AND SEWAGE DISPOSAL that does is supplemented by water furnished by private water supphes and by the ground water which enters the sewers. If the combined yields from all these sources form a very hard water it will contribute a considerable proportion of the solids in the sewage, while soft water will furnish an insignificant pro- portion. Solids are also contributed from kitchen sinks, laundries, baths and other wash waters. Feces, urine and paper furnish further solid matter, and industrial wastes contribute still more. The actual amounts of the contributions from some of these sources depend rather closely upon the population and those from other sources depend rather upon the character of the water in the sewers and the industries of the city, so that there is naturally a considerable variation in the solids in sewage of dif- \Organic Sizff-ling Solids I00'\ c./-j^ \ Settling SoIld3(2hK) ISO ytlnrral 5e#ling Solid;, ^^^^^OS^ing is Tohll Solids BOO •200 Colloidal Solids SiJspended \ t ,^ Solids 300 \5uspended Colloidal Wrgcmic Suspended Colloidal (Bi/riHerl^pcr) [ ' ^^jy^ ,gQ < Solids 100 yilneral Suspended Colloidal \ Solids SO {Organic Dissolved Colloidal [Dissolved Colloidal i „. , ^.^°'''f'^ 11. „ . , , Solids SO Wineral Dissolved Colloidal Dissolved _ } •5"''^^ '" [pissolvedJr^Moidal Hl-^i^ltdaltlid, 160 Y^'^^y^HS""' Solids SOO 800 Total Solldi \tiineral Dissolved Cruslalloidal Solids zaa Fig. 143. — Physical condition of principal constituents of sewage of medium strength. (Numbers are parts per million.) ferent cities of about the same population. If a city has a combined system of sewerage the storm water wiU also contribute a certain amount of solids. Table 38 gives an estimate of the average amount of solids to be expected from all these sources, and Fig. 143 gives an outline of the physical condition of the principal constituents of sewage of medium strength, and the average quantities of each. The data given in Table 39 are averages of analyses of the sew- age of numerous cities, grouped according to the characteristics of the cities. Their chief value is illustrative and as guides in estimating the probable quality of the sewage from a community of known size and character. Grams per Capita. — The different amounts of water in sewages result in the average analysis of the sewage of one community being sometimes of little help in predicting the composition of CHEMICAL AND BIOLOGICAL CHARACTERISTICS 305 another sewage. If, in addition to accurate analyses, however, the quantity of sewage and contributory population are known, it is possible to calculate the weight per capita of each chemical constituent, and from such results reasonably accurate estimates of the composition of sewage of other similar places can be made. Even by this method, the prediction is uncertain be- cause of the lack of accurate data; greater accuracy is possible in the case of a residential city than in that of an industrial community, but even then, where a water supply is very hard, it may have a marked effect on the solids in the sewage. Table 38. — Rough Estimate of Total Solids in the Sevebal Con- stituents OP Sewage Constituents Grams per capita per day Items Total Water supplies and ground water, assumed to be soft Feces Urine Toilet and news paper (suspended) Sinks, baths, laundries and other sources of domestic wash waters Total for residential sewage from separate sewer- age system Industrial wastes Total from industrial city with separate sewerage system Storm water Total from industrial city with combined sewerage system 12.7 20.5 43.3 20.0 86.5 183.0 200.0 383.0 25.0 408.0 Suspended Matter. — In chemical analyses all sohds floating or suspended in sewage are reported as suspended soHds, which average about one-third of the total sohds in sewage. As a rule, the older the sewage the more finely divided the suspended matter. When considering the results of analyses in which the quantities of suspended matter are given it should be borne in mind that it is impossible to take samples which represent fairly the coarser particles. Where suspended matter is removed from 20 306 SEWERAGE AND SEWAGE DISPOSAL s < m < o o s ■ft g I s 2 o German commu- nities of the Ems- cher dis- trict 7.5 3.6 11.1' 550. 151.0 399.0 153.0 82.0 71.0 397.0 69.0 328.0 English mfg. citiea Ol OS to (N • o o • o ■ IN to n • o ■ CO • o CO Ameri- can res- idential and rural commu- nities O ■* lO t^ ci ui : 05 tp ' .H CO : (N --I 183.0 120.0 63.0 105.0 82.0 24.0 77.0 39.0 39.0 Small Ameri- can mfg. cities lO OO 00 oi N «3 : ° ° 185.0 112.0 73.0 58.0 49.0 9.0 126.0 63.0 63.0 ft CO t* O 00 C4 00 • O OS ' 00 -* 263.0 147.0 116.0 131.0 88.0 43.0 167.0 74.0 93.0 Large Ameri- can cities 00 to iH ^ONtooo ooo OOO oooo COCOtO (DOOM (Nt--^ -^ Oi -H C^ rH lO tH CO iH •* CO § 1 SJ ft German commu- nities of the Ems- cher dis- trict o g 2044.0 623.0 1421.0 601.0 332.0 269.0 1443.0 291.0 1152.0 Large Enghsh mfg. cities 49.0 36.5 14.8 O O to N O CD o d o §' o 00 N Ameri- can resi- dential and rural commu- nities 80.0 27.2 7.8 18.0 o o 603.0 393.0 210.0 342.0 260.0 82.0 261.0 133.0 128.0 Small Ameri- can mfg. cities 69.0 38.9 11.3 23.8 o o b- eo O 00 730.0 448.0 282.0 242.0 203.0 39.0 488.0 246.0 243.0 Ifl 95.0 26.5 11.9 24.1 0.26 1.19 133.0 109.0 129.0 1058.0 635.0 423.0 384.0 288.0 96.0 608.0 270.0 338.0 37.0 Large Ameri- can cities (com- bined sewers) O tDOOi-iTtiOO«3 OXOO (NCOM UJiOOCO GO dt-'ooddosodco iciwoi "ti ^ ai cq-n-nio 1 1 Sewage flow, gal. per cap. per day. Nitrogen as: c £ o! *C i 1 ■a J J 1 E c: g o c 1 o ■| < CO a O > a) s 3 'o a c( CO 1 I CO « *> — 2 c a > 1 E i 5 CHEMICAL AND BIOLOGICAL CHARACTERISTICS 307 the sewage by grit chambers or sedimentation tanks, the quantity and character of the material so removed may in some cases be more accurately determined by careful measurements and Table 40. — Relations Existing Between Grains per Gallon, Parts PER 100,000, AND Parts per 1,000,000 (From "Standard Methods of Water Analysis," 1912, p. 14) Grains per U.S. gal. Grains per Imperial gal. Parts per 100,000 Parts 1,000,000 1 grain per U. S. gallon 1.000 0.835 0.585 0.058 1.20 1.00 0.70 0.07 1.71 1.43 1.00 0.10 17.1 14.3 10.0 1.0 1 grain per Imperial gallon 1 part per 100,000 1 part per 1,000,000 Table 41. — Comparative Analyses of Water and Sewage (Parts per million) Place Brookton, Mass. Water supply (1905-09) Sewage (1905-09) Worcester, Mass. Water supply (1905-09) Sewage (1905-09).: Providence, R. I. Water supply, filtered (1910). Sewage (1909) Chicago, 111. Water supply (1908) . . Sewage (1909-10) Mansfield, Ohio. Water Sewage 32.7 2210.0 31.4 871.0 56.0 171S.01 156.0 471.0 391.0 876.0 Nitrogen as 0.013 47.8 0.019 17.7 0.009 15.40 0.034 8.8 0.081 13.3 e 9 I s e ««« o a a Eh-" 0.103 24.5 0.120 7.2 0.078 7.00 0.104 7.6" 0.012 31.72 ■O &la 0.019 19.3 0.021 4.3 0.002 3.50 6.2 144.3 2.1 113.4 6.3 496.5 5.0 40.0 8.6 108.7 Nitrogen as 0.052 0.29 0.35 1.5 0.2 0.001 0.00 0.11 0.0 0.0 2.2 433.7 3.0 122.0 3.5 89.3 2.6 38± 0.05 51.6 8.7 285. 0> Brookton sewage from Brockton city reports. Worcester sewage from Worcester city re- ports. Brockton and Worcester water analyses from Mass. State Bd. Health Rept., 1909, page 19S. Chicago, Langdon Pearse and Report of Lake Michigan Water Commission, 1909, page 45. Providence water analysis, Pratt. Providence sewage analysis, Bugbee. Mans- field, Ohio, Dittoe. ^ Estimated. ' Organic nitrogen (about twice the nitrogen aa albuminoid ammonia). ' Average of Hedges spring supply and supply from main pumping station wells. Al- kalinity plus incrustants. 308 SEWERAGE AND SEWAGE DISPOSAL analyses of the sludge produced than by analyses, however carefully made, of the sewage and effluent. Table 42. — Typical Massachusetts Strong, Mediitm and Weak Sewages (Parts per million) Character of sewage Strong Medium Weak Place Brockton Worcester Stookbridge Nitrogen as Free ammonia 61.2 22.1 50.2 247.0 1584.0 1074.0 892.0 166.0 20.3 7.1 18.0 193.0 882.0 429.0 276.0 114.9 12.4 3.7 7.1 26.8 298.0 177.0 99.0 18.0 Albuminoid ammonia Oxygen consumed Total solids Volatile solids Suspended solids . . Chlorine Table 43. — Percentage of Suspended Matter in Several Sewages Settling in Short Periods Location Suspended matter in sewage, p.p.m. Settling period, hours Per cent settled Authority Chicago, 111. At stockyards At stockyards At 39th St. Station. At 39th St. Station. At 39th St. Station. At 39th St. Station. 1,775 560 156 72 156 72 2 2 2 2 4 4 89 66 54 31 63 43 Wisner and Pearse Wisner and Pearse Wisner and Pearse Wisner and Pearse Wisner and Pearse Wisner and Pearse Providence, R. I 314 4 55 Bugbee Columbus, 210 4.2 60 Johnson Atlanta, Ga 320 114 3 3 80 50 Hommon In the treatment of sewage, the quantity of suspended matter which will settle during the period the sewage is retained in tanks CHEMICAL AND BIOLOGICAL CHARACTERISTICS 309 is important. The proportion of the suspended matter in a sewage which may be classed as settling depends upon the strength of the sewage, the character of the suspended matter, the detention period in the tanks, and possibly other factors. The effect of the strength is shown by the results of the investi- gations summarized in Table 43. CHAPTER X SEWAGE DISPOSAL BY DILUTION Methods of Sewage Treatment. — The most natural method of disposing of sewage is to discharge it into the nearest body of water. This is called disposal by dilution. It is not a make- shift, but just as thoroughly a scientific process of purification as any other, if it is intelligently practiced within the safe limits of the digestive capacity of the water. When the processes of nature by which the water is kept inoffensive are overworked, the sewage-laden waters become offensive and it is necessary to treat the sewage in some way to assist the forces of nature which had previously prevented any nuisance — the primary object in study- ing methods of sewage treatment being to find ways of changing the character of the sewage so that the effluents from the treat- ment works wiU neither cause offensive conditions nor be a menace to health when they are discharged into some body of water. If the floating matter alone is objectionable, it is necessary to remove that only. If deposits of suspended matter are the source of complaint, the removal of the settling solids may be sufficient. Where dilution fails because of lack of oxygen, the treatment of the sewage may be carried far enough to reduce its oxygen demand to the point where the natural water is able to provide an adequate supply of oxygen. It is essential, therefore, before determining the exact type of treatment to be adopted, to ascertain to what extent it is necessary to remove the objection- able constituents from the sewage. Having ascertained the extent to which nature must be assisted in order that disposal of the effluent by dilution shall be permissible, to accomplish his object the engineer selects one or more of the following: 1. Screening, (a) Coarse. (6) Fine. , 2. Sedimentation. (a) Rapid in grit chambers. 310 SEWAGE DISPOSAL BY DILUTION 311 y (6) Slow in settling basms. Plain. With chemical precipitation. With septic action in single-story tanks. With sludge digestion in two-story tanks, (c) With aeration. 3. Contact bed treatment. 4. Trickling filter treatment. 5. Intermittent filtration. 6. Broad irrigation. 7. Disinfection, y These methods of treatment are explained in the following chapters. Prior to 1890, little attention was given to objectionable con- ditions produced by the discharge of sewage into American rivers. Since that date, however, there has been a tendency toward pro- gressively greater restriction upon the discharge of untreated or raw sewage into bodies of water and toward constantly raising the degree of improvement required of treatment plants. While there can be little doubt that in certain cases progress in these directions has exceeded logical requirements, it is equally true that over the greater part of the country the requirements are still quite lax and more care in disposing of sewage will be neces- sary in the future than has been exercised in the past. It is desirable, therefore, when adopting a plan for the disposal of sewage, to select for present use such process or processes as may meet the needs of the present and immediate future, but at the same time lend themselves to later development of more effective means of treatment when they shall be required. Nature of Disposal by Dilution. — Disposal by dilution is the discharge of sewage or the effluent of sewage treatment plants into natural waters where it will be dispersed through the waters in such a manner as to be carried away so rapidly, or be so changed in composition and character, that it will not prove offensive or a menace to health. In 1905 FuUer estimated that 28,000,000 persons in the United States were served by sewerage systems and 96 per cent of the sewage was disposed of by dilution. Estimates made in 1915 by the authors indicated that about 41,800,000 persons were- then served by sewerage systems and that the sewage of about 84 per cent of this population was dis- posed of by dilution. The conditions favorable to successful disposal of sewage by 312 SEWERAGE AND SEWAOE DISPOSAL dilution are: (a) freshness of the sewage; (6) freedom of sewage from floating matter and sohds capable of settling, a condition attained by treatment; (c) thorough diffusion through the diluting water; (d) diluting water of high oxygen content; (e) swift currents to carry the sewage to points of unlimited dilution; (/) biological equilibrium; (g) absence of slips and coves tending to facihtate sedimentation accompanied by sludge deposits. The United States Government has charge of navigation in New York Bay, and before it would permit sewage from a number of cities in New Jersey to be discharged into the Bay it required the Passaic Valley Sewerage Commission, which was building the joint outfall sewer to the Bay, to agree that the sewage should not produce visible suspended particles in the Bay, deposits objectionable to the Secretary of War, odors, grease or color at the point of dispersion, injury to the public health, public or private nuisance, or reduction in the dissolved oxygen of the waters of the Bay to an extent to interfere with major fish life. The Commission agreed to pass the sewage through coarse screens and grit chambers, then through screens with openings not larger than 0.4 in., and finally through sedimenta- tion basins where the mean velocity of flow is not over 0.5 in. per second for average conditions and 0.75 in. for maximum flow conditions. The miminum period of detention of the sewage in the tanks is fixed at 1 hr. and the average at not less than 1}^ hr. The principal physical change which takes place in the sewage discharged into a body of water is due to sedimentation and dilution. It is particularly important, therefore, to pay attention to the suspended solids, for while they are similar in composition to the dissolved substances the changes they undergo are subject to somewhat different laws. Much of this suspended matter is inorganic and not food for living organisms. The organic por- tion, which is such a food, is usually greater than the birds, fish and other food consumers can remove immediately and a part of it is carried along by the current or deposited about the sewer outlet. The current of the body of water receiving the sewage may therefore be important in disposal by dilution, for if the current is so small that organic suspended matter is deposited on the bottom there is the possibility that offensive conditions will result from the decomposition of such deposits, and if the deposits are large objectionable shoaling may occur. The decomposition of sludge deposits is relatively slow. In SEWAGE DISPOSAL BY DILUTION 313 some rivers, sedimentation will remove the settling solids from the waters during normal conditions, throwing upon the rivers principally the burden of the dissolved and colloidal organic matter. But in time of freshet, when there is an ample volume of water and a swift current, such deposits may be scoured out and the rivers relieved of the burden of changing this settled organic matter also into more stable substances. Under such conditions sedimentation in the river performs a function similar to that in settling tanks, through which the sewage may be passed before its discharge. Decomposition of Organic Matter. — If the ratio of diluting water to sewage is inadequate, bacteria may so thrive that they will exhaust the available supply of oxygen and anaerobic condi- tions likely to cause offensive odors will become established. In any event such processes are likely to go on in the sludge banks formed by the precipitation of organic sewage matter. On the other hand, if there is enough oxygen to meet the demands of the bacteria, the aerobic organisms will predominate and the organic matter will be oxidized as explained in Chapter IX. The oxidation of sewage under conditions favorable to the growth of plants is certain to be followed by the growth of plankton. The oxygen given off by some of the plankton is an important factor in maintaining the supply of dissolved oxygen in water under such conditions. In the case of ponds and lakes the saturation of the water is iisually confined to the upper por- tion of the water, because the plankton producing it thrive where the Hght is strong. At considerable depths bacteria thrive and make large demands on the dissolved oxygen there, which is still further depleted by contact with the products of the anaerobic decomposition of the mud at the bottom. In addition to the results of chemical and bacterial studies, there are records of the effect of the discharge of known amounts of sewage into waters, prepared mainly by the investigations mentioned on page 314. Steams' Investigation. — After an investigation for the Mas- sachusetts State Board of Health, Frederic P. Stearns, then its Chief Engineer, concluded ("Special Report on Water," 1890, p. 791) that if the flow is less than 2}4 cu. ft. per second per 1,000 inhabitants an offense would be almost sure to arise.i 1 Stearns estimated that each person contributed to sewage daily an average of O.OIS lb. free ammonia, 0.003 lb. albuminoid ammonia, 0.218 lb. dissolved solids and 0.042 lb. chlorine. 314 SEWERAGE AND SEWAGE DISPOSAL With larger volumes than 7 cu. ft. per second per 1,000 inhabi- tants, the pollution would be too smaU to cause any nuisance. Where the water is to be used for manufacturing purposes, the amount of dilution should be greater, he stated, and in the case of a stream used for domestic water supply he was unwilling to say that any degree of dilution would make the water entirely safe to use. Goodnough's Investigations. — In 1902 another investigation was made for the Massachusetts State Board of Health by Goodnough, who succeeded Stearns as Chief Engineer of the Board. He narrowed the range of dilution fixed by Stearns and summarized the results ("Eeport State Board Health," 1902, p. 452) as showing that where the quantity of water available for the dilution of the sewage in a steam exceeds about 6 cu. ft. per second per 1,000 persons contributing sewage, objectionable conditions are unlikely to result. Where sewage was discharged at many outlets into a large body of water he thought that objec- tionable conditions might not result from somewhat less dilution than that named. In every case where the flow was less than 3.5 cu. ft. per second per 1,000 persons objectionable conditions had resulted. In 1902, in a letter to the Committee on Charles River Dam, Goodnough made the following statement : "The degree of dilution which has been found necessary to prevent un- sanitary conditions where sewage is discharged into a stream, assuming 75 gal. of sewage per person, ranges between 20 to 1 and 60 to 1. In esti- mating the degree of dilution of the sewage, no account has been taken of the purifying effect of the water of the basin itself." (Evidence, etc., before Committee, page 108.) J. Herbert Shedd testified in 1902 before the Committee on Charles River Dam that about 5 cu. ft. per second of the ordinary flow of the river, per 1,000 persons contributing sewage, would render the presence of the sewage unobjectionable (Evidence, etc., before Committee, p. 365). In 1908 and 1913 Goodnough made for the Board investigations of the condition of the Merrimack River, the results appearing in special reports. The condition of other rivers was examined to obtain corroborative evidence regarding pollution, and it was found that wherever the dilution was 3.4 cu. ft. per second or less per 1,000 persons, a nuisance followed. Where serious pollution was observed with higher rates of dilution the nuisance SEWAGE DISPOSAL BY DILUTION 315 was usually due to the discharge of large quantities of industrial wastes into the stream. No case of dilution was found between 5.8 cu. ft., where industrial wastes and sewage combined to cause offense at Webster, and 9.2 cu. ft. at Ware, where the effect of the sewage was noticeable for a considerable distance, except at cities along the Merrimack. The dilution in the latter river was 8.7 cu. ft. below Lawrence and 7.6 cu. ft. below Haver- hill. Below Lawrence the surface of the river was reported to be covered often for a long distance with froth, scum, and oily and greasy matters, particularly in summer, when there was a noticeably disagreeable odor at times along the south bank. Except at the latter place, the pollution had not rendered the river offensive except near some of the large sewer outlets and where banks of sludge were exposed at low water, the report stated, and these conditions could be ameliorated by improving the methods of discharging sewage into the stream, so as to provide uniform diffusion through the river water. Poliution of Ohio Rivers. — In 1897 the Ohio State Board of Health had an investigation of the condition of certain Ohio rivers made under the direction of Allen Hazen. In discussing the results he stated that in the case of sluggish streams, or of streams the waters of which are already somewhat polluted, the quantity required for proper dilution may become 6, 8 or even 10 cu. ft. per second per 1,000 population. (" Preliminary Report of an Investigation of Rivers," p. 32.) New York Harbor. — The MetropoUtan Sewerage Commission of New York estimated in its 1910 report that the sewage dis- charged into New York Harbor was diluted with 32 parts of water, and that this ratio would become 1 to 13 about 1940. The flow of diluting water in the harbor was estimated at 4.7 cu. ft. per second per 1,000 population, which would be reduced to 2.65 cu. ft. in 1940. The permissible limit of pollution in sea water depends upon whether the object in setting the limit is the prevention of nui- sances or the protection of fish. This has been emphasized in disputes over the permissible poUution of New York Bay, where the dissolved oxygen should not be reduced below 70 per cent saturation, according to Black and Phelps, in order that food fishes might continue to live in the waters. Fuller stated* on this topic : 1 Trans. Am. Inst. Chom. Engrs., vol. iii, p. 392. 316 SEWERAGE AND SEWAGE DISPOSAL "With respect to guarding against objectionable odors I think it is clearly necessary for the chemists and bacteriologists to keep in mind that putrefaction does not exist so long as oxygen remains at all. In fact you can go further than that, and say that so long as oxygen is available from nitrates, nitrites, or other oxidized salts there is substantially no putrefaction. I am aware that that does not provide for one feature that may be of some importance, and that is the protection of major fish life. I believe, however, that the European custom in many places is sound in indicating that 30 per cent of the dissolved oxygen necessary for saturation provides a reasonable margin in the case of a majority of species of fish life of the larger kinds. Perhaps some may call for more, but so long as there is 30 per cent remaining at all places at all times, it is a matter of deduc- tion from our well-established laws of biology and chemistry that there can be no putrefaction. The larger number of the principal rivers in this country serving as public water supplies do not contain as much as 70 per cent of the oxygen necessary for saturation. Among the rivers with which I have been personally familiar through analysis, I may mention the Merrimack River at Lawrence, Mass. Twenty years ago it had as low as 50 per cent and sometimes but 30 per cent of dissolved oxygen. In those days it served as the water supply for Lawrence without being filtered, and in the last 17 years, since filtering, it has been regarded as one of the good water supplies of the world. I believe that if this 70 per cent margin suggested by Dr. Soper were applied to Lawrence, it would show that the Merrimack River at that place was not providing a proper disposal for the sewage at Lowell and the cities above, notwithstanding the fact that it provides an excellent water supply at that point." Temperature. — The effect of temperature upon disposal by dilution is important. A river may receive in winter, without causing objectionable conditions, a quantity of sewage which, in summer, would make it a nuisance. Bacterial action at low temperatures is relatively slow, and sewage may be carried by the river receiving it to a point where ample dilution is afforded before the bacteria have caused any offensive results. This has an important economic aspect, for, in summer, sewage treatment may be carried to a degree insuring satisfactory conditions at an expense which, if continued through the year, would be prohibi- tive. Advantage may be taken of the winter conditions by providing a less expensive treatment, care being taken to avoid sludge deposits which may prove objectionable during the succeeding warm season. Fresh and Salt Water. — When sewage is discharged into salt water there is a greater tendency for the settling solids to form sludge banks than when it is discharged into fresh water. This is due in part to chemical action of the salt water on the sewage and SEWAGE DISPOSAL BY DILUTION 317 in part to the inability of the salt water to carry as much matter in suspension as the fresh water can. Sea water normally con- tains about 20 per cent less dissolved oxygen than distilled water so that, other things being equal, fresh water is able to dispose of more sewage than sea water. Tides. — If a river discharges into a tidal estuary, a mixture of fresh and salt water is produced, changing in character from hour to hour. If the estuary is short and steep, it clears itself with each ebb tide; if long and with complex entries, the water may oscillate backward and forward with a varying degree of salinity. Inasmuch as the tidal currents must be utilized to remove sewage, it is necessary to have a comprehensive knowledge of them before locating the outlet of a sewerage system, in order to be sure that offensive conditions will not arise at any stage of the tide. At the turn of the tide, the incoming salt water follows the bottom, owing to its greater density, while the overlying brack- ish water is still traveling seaward. Gradually, as the depth of the incoming wave increases, the motion of the entire section changes, flowing inward with increasing velocity until the maxi- mum velocity, or "strength of the tide," is reached. The veloc- ity then falls off until ebb tide begins, when the whole mass of water flows outward with increasing velocity until the maximum run is reached. The velocity then decreases until the turn of the tide is reached and the cycle begins again. The velocity of the ebb tide is greater than that of the flood tide because the flow of fresh water is in the same direction as that of the outgoing salt water, instead of in the opposite direction, as on the flood tide. The "tidal prism" of a tidal basin is the volume of water con- tained within it between the elevations of high and low water. It depends upon the mean range of the tidal rise and fall, which decreases in going from the northern to the southern limits of the United States, as follows : East Coast Range, West Coast Ranoe, City Feet City Feet Boston 9.6 Seattle 11.6 New York 4.4 Astoria 6.4 Baltimore 1.2 San Francisco 4.0 Charleston 5.2 San Diego 4.0 Savannah 6.5 Key West 1.2 Galveston 1.0 318 SEWERAGE AND SEWAGE DISPOSAL Floats. — Observations of the movements of floats are made to ascertain the direction of currents. Fig. 144 illustrates two types employed by the New York Sewerage Commission in its investigations of the currents in New York Harbor and neighbor- ing waters. The small float had the advantage of low cost, ease of handhng, and small area exposed to the wind. On the other hand it was destroyed by the paddles and propellors of steamers Small Float. Large '" Float. Fig. 144. — Types of floats used by Metropolitan Sewerage Commisaion. and, as it was unable to support a lantern, it was unsuited for night work. The upper cylinder was 6% in. in diameter and 5 in. long. The lower cylinder was 63^ in. in diameter and 14 in. long, and was weighted with sand until the top of the upper can was nearly level with the surface of the water. The large float was made in two sizes, 5 and 6 ft. long. The illustration shows the smaUer size, which had an upright of 2 by 2-in. timber, buoyed by a cork float at the top and provided at the bottom with four 12 by 24-in. vanes. SEWAGE DISPOSAL BY DILUTION 319 In making use of floats to determine the direction and velocity of tidal currents it must be kept in mind that sewage discharged into tidal water will move over the surface more rapidly than floats. For example, the early investigations of the tidal currents in the vicinity of the Moon Island outlet of the Boston sewerage system were made with floats 8 ft. long. These floats showed the direction of the currents, but they moved with the mean velocity of the water and not with the surface velocity during the ebb tide, during which alone it was proposed to dis- charge the sewage. After this outlet was placed in service it was found that the sewage moved on the surface much more rapidly than was anticipated from the results of the float observations, and in later—efeservations of this nature the Massachusetts State Board of Health used floats about 8 in. long for most observations and 24 in. long for the remainder, as it was found that the sewage tended to form in a shallow sheet on top of the water, and hence the surface and shallow depth velocities were of greater importance than the mean velocities of deeper sections from the surface down. Depth of Outlet in Tidal Waters. — Experience with the dis- charge of sewage in tidal waters has shown that the outlet should be at a considerable depth, in order to disperse the sewage through as large a volume of water as possible before it rises to the surface. Not only do the large solids in sewage present an offensive appearance if they float on the surface, but the greasy substances cover the surface with a film or "sleek" which is not necessarily an indication of putrescible organic matter yet is unsightly and likely to be the cause of complaint against the dis- posal of the sewage in this way. These undesirable conditions are much reduced when the sewage is discharged in depths of 25 ft. or more. For example, the Peddocks Island outlet in Boston Harbor, the point of discharge of the sewage of a population estimated at 382,000 in 1912, is in 30 ft. of water at low tide, and the sewage is so thoroughly dispersed that the Massachusetts State Board of Health reported in 1910 that it would be diflicult to detect its position but for the fact that it is the feeding ground for large numbers of gulls. The authors showed, however, that considerable quantities of sewage rise to the surface at this outlet, although not visible, by dissolving 2 lb. of eosine in water and discharging this dye into the shore end of the outfall sewer. On proceeding to the vicinity of the outlet the colored sewage 320 SEWERAGE AND SEWAGE DISPOSAL could be seen rising with considerable velocity and spreading over an area of many acres, although the experiment was per- formed during a very heavy wind and rain storm. Salinity. — Although the normal specific gravity of sea water, which depends upon its salinity, varies from about 1.022 to 1.028, it is practically constant near any one place in the ocean. This fact has been utilized in studies of the dilution of sea water by fresh water in tidal basins, and affords an easy method of determining certain tidal movements which it is difficult to investigate in other ways. For instance, off New York the nor- mal specific gravity of the sea water is 1.025, corresponding to a chlorine content of 18,000 p. p.m. under the assumption that 88.6 per cent of the total excess specific gravity above unity is due to the chlorides. If a sample of water from New York Harbor is found to have a specific gravity of 1.015, it must result from a mixture of about 60 per cent sea water and 40 per cent fresh water. By conducting a series of observations of this nature in the East River at New York, the Metropolitan Sewerage Com- mission showed that there was a distinct flow southward through this tidal river, although float observations and studies of tidal movements, made mathematically, pointed to the absence of any such flow. The investigations of this nature are made by a salinometer,* a form of hydrometer about 12 in. long, with a stem ^2 in- in diameter, carrying a 4-in. scale reading specific gravities from 1.00 to 1.03 by intervals of 0.0005. A sample of water is placed in a tall cylindrical glass and the salinometer lowered into it. Both the thermometer and hydrometer scales are read and the hydrometer reading is corrected by an amount ranging with the temperature from -0.0011 at 35°F. to +0.0028 at 82°F., which was determined by experiment. Effect of Wind. — The wind blowing over the surface of a body of water tends to set up currents in it, owing to the fric- tional resistance of the water to the currents of air over its surface. On-shore and off-shore winds produce complementary currents in opposite directions at greater depths. Judson found^ that on Lake Michigan the rate of travel of the surface currents was about 5 per cent of the velocity of the wind. The U. S. ^''tTse of the Salinometer in Studies of Sewage Disposal by Dilution," Kenneth Allen, Jour. Assoc. Eng. Socs., April, 1911. 2 Report Lake Michigan Water Commission, 1909. SEWAGE DISPOSAL BY DILUTION 321 Hydrographic Service reported' that a study of many observa- tions in the North Atlantic indicated surface currents of about 3.2 per cent of the velocity of the wind. Whipple found^ in Lake Erie that the surface currents were from 3 to 6 per cent of the velocity of the wind. It is believed that the effect of these surface currents may extend to depths of 30 to 40 ft. Knowledge of the currents produced by the wind has been gained largely by studies of temperature changes in the water at different depths. The readings of temperatures below the Indicator Fig. 145. — The thermophone. (From Whipple's "Microscopy of Drinking Water.") surface are most easily made with the thermophone,' invented by Whipple and Warren. It is an electrical thermometer, Fig. 145, consisting of a sensitive electrical coil which may be lowered to any desired depth and an indicator connected with the coil by lead wires. An electric current is supplied by dry batteries and a telephone receiver is connected to the indicator by which the operator can detect, by a buzzing sound, when the movable 1 Monthly Weather Review, 1902. ' Report on Cleveland Water Supply, 1905. ' " Microscopy of Drinking Water," Whipple, p. 88. 21 322 SEWERAGE AND SEWAGE DISPOSAL pointer is made to approach the correct temperature reading, at which the buzzing ceases. Thus by reading the dial at this point, the operator can ascertain the temperature of the distant coil. Thermophones adjusted for a range from 40 to 250°F. will give readings correct to within 0.2°. Milwaukee Investigation of Lake Currents. — In 1909 to 1911 an investigation was made by Alvord, Whipple and Eddy of the methods of disposing of sewage and protecting the water supply of Milwaukee. This involved a study of the currents in Lake Michigan, which exemplifies the scope of such investigations. On-Shore mnd Inwa rd Sur face Current "Region of Deep '" '"'"'Wafer Fig. 146. — Effect of wind on lake currents at Milwaukee. As the capacity of the lake is about equal to its discharge during 100 years, the lacustrine current toward its outlet at the Straits of Mackinac is inappreciable, and temperature and winds are responsible for both surface and deep movements of the water. Fig. 146 illustrates the manner in which winds were found to cause sudden large fluctuations in the temperature of the water drawn in at the intake of the Milwaukee water works. Thermo- phone readings of the water every 5 ft. in depth taken at the in- take and at points 1, 2 and 3 miles farther out in the lake, showed that 2 or 3 miles from the shore, where the lake becomes SEWAGE DISPOSAL BY DILUTION 323 deep, the water below a depth of about 65 ft. is practically stag- nant in the summer. Nearer the surface, however, the wind has a marked effect in circulating the water. The temperature observations were supplemented by investiga- tions of the number of bacteria in samples of water collected at ModerateN.E. followed by S. E. Winds. Sept?2.l910. & ^ Wff;^ ^m ^ ^0i ^^^^ % ^ i^ West- Winds. Seph 2ft. 1910. / / zo 40 eo 60 WM. M » H ^^^ ^^^ ^!^^^^ "Wf, 5. followed by N.E. a- S,W. Winds. Oct. 5. 1910. S. W, followed^ by West Winds. Oct.lt. 1910. N,E. followed by N. Winds. Oct. 17.1910. 5.W. followed by W. Br N.E. Winds. Oct.24,1910. ^Position of Wafer-Works Intake. Fio. 147. — Effect of winds on vertical distribution of bacteria in a plane passing east and west through the intake of the Milwaukee water-works. numerous points, at both the surface and the bottom, and by determinations of the chlorine in the water at the same points. The results obtained on 6 days are shown in Figs. 147 and 148, and illustrate how the wind affected the distribution of the sewage through the lake in the vicinity of the city. 324 SEWERAGE AND SEWAGE DISPOSAL On September 22 there was an on-shore wind. The bottom water had practically the same temperature as that at the surface and the turbidity of the water was the lowest observed at any time during the investigation. The surface water containing the largest amount of chlorine was driven to the northward and kept near the shore and the largest number of bacteria was found Break' wafer Dumping 6mund Sept.22.1910. V^ind. Sept20.N.E.Sr5.W. Light " 21. N.E.Modera-fe " 22. 5. E. Light 5ept.28.1910. Wind. Sept.Se.N.E. Brisk '• 27. W. Light " 28. N.W. Light Oct.5.1910. Wind. Oct 3. S. Brisk " 4. N.E. Light •I 5. 5. W. Brisk Ochll.1910, Wfnd. Q£f. 9. N. Lighf «. iO.S.W.Liqhf It ' 16. N.B.lighf « }7. N. Light Oct. 24. 1910. Wind. Cct.tt.W. Light » 23. 5.W. Light « 24. W.&N.E.Liqhf FiQ. 148. — ^Effect of winds on distribution of chlorine in surface waters of Milwaukee Bay. near the shore. The sewage of the city is discharged into the lake by the rivers shown in Fig. 148. On September 28, there was a hght off-shore wind, and little difference was found in the temperatures of the surface and bottom waters. The turbidity was high on this day. The chlorine extended more directly from the mouth of the river, SEWAGE DISPOSAL BY DILUTION 325 whence the pollution came, and was more widely distributed in the bay. The surface water containing the largest number of bacteria was now spread out from the shore. On October 5, the wind was off-shore and colder water was brought in at the bottom, the difference between the average surface temperature and the lowest bottom temperature being Wind ^Dumping Ground "I Sep.22.1910. *^ Average Surface Temperature 53.$^ Wind Sep.75.1910, Average Surface Temperature ■SS.5* mnd Ooh5.19IO. Average Surface Temperature £$.5° f %* ■«* 0C+.II.I910. OC+.17.1910. Oc+.24.I9IO. Average Surface Average Surface Average Surface Temperature S6J^ Temperature 56.2" Temperature Siti" Fio. 149. — Effect of winds on temperatures of bottom waters in Milwaukee Bay. 13.3°, as is evident from Fig. 149. The surface water containing the largest amount of chlorine had been driven to the north and there was a well-marked surface drift of water with large numbers of bacteria toward the water works intake. On October 11, with off-shore winds, the temperature condi- tions, chlorine and bacteria were about the same as on October 5. On October 17, with a north wind, the difference between the average surface temperature and the minimum bottom tempera- 326 SEWERAGE AND SEWAGE DISPOSAL ture was 3.3°. The surface water with the largest amount of chlorine and numbers of bacteria had been driven southward. On October 24, when there was an on-shore wind, the turbidity was high, the largest proportion of chlorine was near the shore, but the distribution of bacteria was unusual, probably due to heavy pollution of the river caused by the refuse washed from streets by a rainfall of 0.5 in. on October 21 after a long period of dry weather. Temperature Changes in Deep Ponds. — Temperature changes in the water of deep ponds cause vertical currents due to differ- ences in the specific gravity of the water. As the surface water cools in the autumn it becomes heavier and sinks, thus causing currents reaching to increasing depths until all the water ac- quires the temperature of maximum density, 39.2°F. If it were not for the wind, circulation would then cease, but the slight differences of specific gravity near this temperature facili- tate movement of the water and it is possible for the entire contents of the pond to reach a slightly lower temperature than 39.2°. In the spring, when the icy surface water becomes heavier as it is warmed, vertical currents are again established and continue until the temperature of the lower strata has reached a point somewhat higher than 39.2°, due to wind action. There is thus an overturning and mixing of the water in the spring and autumn, while in summer and winter there is stagnation below the depths affected by winds. These conditions do not exist in shallow ponds because the winds mix the water through- out its depth. It is interesting to consider what would happen if the maximum density of water were below its freezing point. Quiet and Running Water. — The absorption of oxygen by quiet water is very gradual, but when the surface is agitated the absorption proceeds more rapidly. Therefore a running stream absorbs oxygen more rapidly than a pond, and as oxygen' is required for the aerobic change of organic matter into mineral matter there was a prevailing opinion for some years that running water was preferable to quiet water for receiving sewage, and that a polluted river purified itself in a distance of 20 miles. This theory has been exploded. It is now known that it does not do so and that there are other factors which influence the changes taking place in the organic matter, such as sedimentation, winds, sunlight, bacteria and plankton, which are more active in quiet waters than in running streams. SEWAGE DISPOSAL BY DILUTION 327 The sewage discharged into running water is broken up and altered so that the stream may appear to have been unaffected by the sewage discharged into it above the place of observation, whereas the organic matter in it may have undergone only part of the changes necessary to render it innocuous. The improvement has been in appearance rather than in composition. In 1899 Sedgwick stated^ "that sedimentation and the destruction of micro-organisms by various agencies are more completely effected in standing than in moving water, so that modern sanitary science has reversed the tenet of 30 years ago and now unhesitatingly affirms that it is quiet water rather than running water that' purifies itself.' " Notwithstanding the importance of sedimentation and of the action of organisms in relatively quiet water, the great value of aeration of water by wave action, agitation by propellors, and falling over dams and riffies, must not be ignored. It is easily conceivable that a stream may be so seriously polluted as to become putrid if practically quiescent, whereas, if caused to fall over dams or pass down rough rapids at frequent intervals along its course, it will absorb sufficient oxygen to maintain the processes of oxidation and thus avoid putrefaction. Basic Information for Planning Dilution Projects. — From what has been stated in this chapter it will be seen that in order to pass intelligently upon the disposal of large quantities of sewage in water, (a) hydrographic surveys may be required; (6) a study of the quality and temperature of the lake water and of the currents and winds affecting them may be needed, with chemical and bacterial surveys of the water at different points and depths, from which can be plotted zones of pollution of different degrees; (c) a study of possible locations for and forms of the sewage outlet may be desirable to ascertain any danger of pollution of a water supply, of nuisance by washing of the sewage back to the shores of the lake or other body of water, or of objectionable sleek on the surface of the water; and, finally, (d) the engineer will have to weigh carefully the character and extent of the preliminary treatment required to fit the sewage for dilution, the desirability of disinfecting the sewage or effluent before discharg- ing it into the water, and the relative desirabiUty of treating the sewage or purifying a water supply drawn from the lake or river. I Report Pittsburgh Filtration Commission, p. 17. 328 SEWERAOE AND SEWAGE DISPOSAL Sewage Dilution at Boston. — The first large outfall sewer discharging into Boston Harbor had its outlet near Moon Island, where float observations showed that the average velocity of the ebb tide was 0.74 mile per hour and that the floats traveled about 4 miles seaward during the whole ebb tide. The sewage is stored on the island and discharged during ebb tide. Experi- ments showed that when 22,000,000 gal. of sewage are discharged in 45 minutes, about 750 acres of harbor surface are plainly dis- colored on a comparatively calm day, and an objectionable appearance is presented by two-thirds of this area, although offensive odors arise from but a relatively small portion of it. Suspended matter is sometimes seen 1}4 miles from the outlet, and sleek is occasionally observed at still greater dis- tances. Generally speaking, the upper 2 to 3 in. of the sewage- covered area contain much the greater percentage of the sewage. The conditions described last for 2 to 3 hr., depending largely on the wave action. The second large outfall sewer discharging into the harbor originally had its outlet near Deer Island Light but a few feet below the surface near the edge of the main ship channel. The sewage was discharged continuously and discolored the water over an area of 350 acres on ebb tide and for a distance of IJ^ miles from the outlet, although sleek was sometimes observed farther out on calm days. In 1916 this outfall was extended 315 feet further out toward the ship channel, discharging through 14 openings the deepest being at a considerably greater depth than the discharge end of the original outfall. Of the multiple openings the outermost is 48 in. in diameter and the rest ellipti- cal-shaped varying from 25 by 44 to 13 by 23 inches. The third large outfall sewer has its outlet near Peddocks Island at a depth of 30 ft. at low tide; the sewage is discharged continuously. The sewage is diluted so quickly by the sea water that the percentage of sewage in the surface water directly above the outlet is very small. Dilution at Cleveland. — Whipple found at Cleveland in 1904 that the general current in Lake Erie near that city was a mile in 6 days, which is negligible in comparison with currents due to winds. During 11 months of the year the prevailing winds are down the lake; the winds across the lake are practically equal in both directions. The strongly polluted water of the Cuyahoga River, see Fig. 3, was found to pollute the lake more than the SEWAGE DISPOSAL BY DILUTION 329 outlets along the lake front. The turbidity of the water did not have an appreciable relation to the amount of sewage pollution, although the numbers of bacteria varied with the turbidity, decreasing rapidly off-shore. The chlorine determinations indicated a general outward sweep of the polluted water eastward away from the shore. Winds were found to have the same effects that have been described in connection with the Milwaukee investigation. The odor of the lake water near the shore was frequently moldy and oily, but 2 or 3 miles off-shore little odor was perceptible. Chicago Drainage Canal. — The sewage of Chicago is carried to the Desplaines and Illinois Rivers by the Chicago Drainage Canal, and is discharged eventually into the Mississippi. The canal was designed to carry a minimum of 3.3 cu. ft. per second of lake water per 1,000 persons contributing sewage. The Cana- dian Government has not been favorably disposed to the diver- sion of water through the canal, and since it was opened there has been an enormous and unforeseen increase in the trade wastes discharged into it. The effect of the dilution of the sewage was investigated thoroughly in connection with an unsuccessful suit by the State of Missouri to obtain an injunction against the discharge of Chicago sewage into tributaries of the Mississippi. It was claimed that the sewage contaminated the water supply of St. Louis, while the defendants contended that the processes of self-purification in the canal and rivers removed all danger of such contamination. In 1911, Wisner reported that the dilution in summer was inadequate to prevent the reduction of the dis- solved oxygen throughout the canal to less than 2}/^ p. p.m. In the lower 15 miles of the canal it averaged from to 10 per cent saturation, the latter figure being less than 1 p.p.m. under summer conditions. CHAPTER XI GRIT CHAMBERS, RACKS AND SCREENS Sewage, particularly sewage from combined systems while they are carrying storm water, may contain a large amount of sus- pended matter. Part of it is heavy mineral matter like sand, gravel, bits of coal and cinders, which is called grit. It can be removed, if necessary, in small settling basins called grit cham- bers. Sewage also contains cloth, paper, lumps of meat, bits of vegetables and pieces of wood ranging in size from matches to railroad ties, which can be removed by screens. Object of Grit Chambers. — Among the objects which may make the removal of coarse mineral matter desirable are: (1) to prevent injury to pumping machinery by grit; (2) to keep mineral matter from the sludge of settling basins, which flows less readily if coarse grit enters it : (3) to prevent inverted siphons from becoming clogged with material often difficult to remove, (4) to prevent deposits of inorganic matter in bodies of water into which the sewage is discharged; (5) to keep heavy grit from settling on the sludge in septic tanks and tending to seal it and to make the ebullitions of the gases of decomposition, when they occur, very violent. Object of Racks. — Racks are used for the removal from the sewage of the coarser materials, such as pieces of wood, cork, bottles, orange skins, cloth, cotton waste, wool, hair, fiber, and similar matter, which find their way into sewers, and which are likely to cause obstruction or injury. They may be placed upon the upstream or downstream side of the chamber and are in the nature of coarse screens. Object of Screens. — Among the objects accomplished by screening the following are prominent: (1) the protection from injuries and clogging of appliances for conveying, pumping and treating sewage and sludge; (2) the prevention of unsightly matters floating on tanks at treatment works or on bodies of water into which raw sewage is discharged; (3) the reduction of the amount of sludge settling on the bottom of slow-moving bodies of water and liable to cause offense by decomposing; (4) 330 GRIT CHAMBERS, RACKS AND SCREENS 331 the prevention of heavy and extremely tough floating scum on the surface of septic and other tanks; (5) the removal, so far as possible, of the fine suspended matter in addition to the coarser matters already mentioned, either as a final treatment to prepare sewage for disposal by dilution or as a preliminary one to assist other methods of treatment. As a temporary expedient, for service while developing more complete methods of treatment and building the plant for applying them, screening may offer the advantages of low cost and considerable efficiency. Theory of Grit Chambers. — The object sought in the grit chamber is t o remove the grit, in general th einorganjc naaterial, without depositing any of the organi_e_m.atter. Grit settles from flowing sewage according to its size and specific gravity, and to the velocity of the stream. A velocity which will just permit pebbles }i in. in diameter to settle will carry along sand. To provide a uniform separation of grit from other settling solids, a uniform^ velocity must be maintained, and this is difficult on account of the constant variation in the quantity of sewage and the large capacity of the chambers in proportion to the flow through them during the early years of their use. Yet the veloc- ity of flow must not be too low as this will result in the deposit of the organic material. Another element entering into the problem of removing grit is the depth of sewage through which it must settle. Upon this will depend the length of the grit chamber, for a definite period of time is required for a particle of given size and weight to fall a specified distance through a flowing stream. The grit chamber, therefore, must be long enough to provide time for the particle to reach a depth below which it wiU be prevented from passing out of the basin. The difficulties of operation will be reduced by retaining in the grit chamber only such matters as settle at velocities not lower than 1 ft. per second, if it is practicable to handle the remaining heavy materials in other portions of the treatment plant. The organic matter likely to be precipitated in grit chambers may become extremely offensive if the velocity of flow is reduced below 1 ft. per second, and wherever practicable it is more satisfactory to handle it with the sludge from settling tanks than to attempt its removal by grit chambers. Even with a velocity of 1 ft. per second, there is danger of intercepting with the grit such organic matter as coffee and tea grounds, raisin and fruit seeds, rags, paper, bits of meat and vegetables, and some 332 SEWERAGE AND SEWAOE DISPOSAL feces. For this reason the real value of the removal of the small amount of grit which can be safely effected in such chambers should always be determined carefully before adopting them, for they may be found to have no useful purpose, particularly with a separate sewer system. They should not be expected to remove more than from 10 to 40 p. p.m. of the suspended matter. Quantity and Character of Grit. — The quantity and composi- tion of the grit collected in a grit chamber depend so largely on the composition of the sewage, the specific gravity of the grit, the dimensions of the grit chamber and the rate of flow through it that reports of the grit collected in different cities rarely show close agreement. In addition to these factors, part of the storm water, which carries much grit, escapes through storm-water overflows in many combined systems of sewerage, so that not all the grit reaches the grit chambers. In the Emscher District in Germany Imhoff found that with a velocity of 1 ft. per second very little organic matter settled with the grit and the latter averaged about 0.24 cu. yd. per 1,000 persons annually. This is the quantity estimated by Fruhlingi as a fair average of many German reports on grit; but it is to be remembered that the dilution of the sewage is much less and hence its concentration is substantially greater in Germany than in America. At Worcester, Mass., with velocities of 0.2 to 0.6 ft. per second, averaging 0.5 ft., 3.2 cu. yd. of grit were collected annually per 1,000 persons, but most of the storm water did not reach the grit chamber in this case. The grit which collects in the Worcester chamber when re- moved after 4 to 8 weeks is extremely offensive and must be covered immediately after removal to prevent the escape of such odors. It averages about 2,014 lb. per cubic yard. The dry solids run from 53.3 to 75.4 per cent averaging 64.1; the loss on ignition from 18.9 to 28.1 per cent; averaging 22.7; the organic nitrogen in dried samples from 0.515 to 0.871, averaging 0.675 per cent. The grit collected in grit chambers at Hamburg was found to consist of about 33 per cent of water, 3 per cent of organic matter and 64 per cent of mineral matter. Design of Grit Chambers. — The quantity of sewage reaching a grit chamber fluctuates so greatly that in order to keep the veloc- ity of flow within the desired range it is generally considered desirable to provide at least two independent basins in such a i"Entwasserung der Stadte." GRIT CHAMBERS, RACKS AND SCREENS 333 chamber, as shown in Fig. 150. One of these will suffice for the dry-weather flow and moderate rainfalls and will afford an opportunity for cleaning the other, while both may be used when large quantities of sewage reach the chamber. The stand- ard design of the Emscher District provides three narrow, paral- lel basins, Fig. 151. There is some uncertainty in estimating the velocity of flow through grit basins introduced by the pit in which the grit is caught, which is below the invert level of the inlet and outlet conduits. It is sometimes assumed that the gSi fffc* „. ' L Section A- B Section C'D Fig. 150. — Grit chamber at Worcester, Mass. liquid below this level is without motion, and experiments at several places have indicated the existence of a plane below which stagnation exists. Its position depends upon the shape and length of the chamber and is apparently below the invert level of the inlet and outlet in most cases, so that in estimating veloci- ties some movement below this level should be provided for if the actual velocity of flow through the chamber is to be great enough to prevent the settling of organic solids. A cross-section of the general shape shown in Fig. 52 is desirable for a grit chamber on account of its hydraulic properties. Where 334 SEWERAGE AND SEWAGE DISPOSAL the grit is to be removed by shovel, the bottom slopes should be gentle, but if the grit is removed mechanically it is desirable to have the bottom form a long trough or succession of hoppers in the bottom of which the material will collect. There is some evidence to indicate that in order for the grit to slide to the bottom of such pits the sides must have a slope of at least 45 deg. Where mechanical cleaning is employed, it is desirable to round off all angles in order to facilitate the sliding of the grit to the bottom of the pits. The longitudinal section of American grit chambers has gen- erally had vertical end walls and a horizontal bottom. Experi- na'in Screen Bypass ' li 6cree/r Drainage Oufhf Drainage Pipe effluerri--^^'"''' ^ ^,|...| -f.,.}.,-^..^^.,^^^ ^^^^ Kaised Inyert /lorfPiifribirfor- /(Yerf/ca/eans) p. ffainScrfen • |^.t...fH-+4-H-l"-l--4-t--t-^ 6afe H^.4.>4..t.4.4-4^...|.„}„.^..^... ^g Lower Dmln Sais A ' Screen Bar Section MdihSeretli Section C- D Section A"B. Fia. 151. — Imhoff screen and grit chamber. ments have shown that there are vertical currents at these end walls. At Rochester, N. Y., for instance. Brown' found that the vertical wall at the outlet of the chamber caused such a strong eddy that submerged floats were raised by it into the currents moving downstream. Unless it is particularly desired to have these eddy effects, the outlet end of the channels or basins may be flared out so as to give a gradual change in the velocity of flow, although it is rarely possible in such short basins to use the most desirable angle of flare, 13)^ deg. There is a great difference of opinion regarding the depth of the pit in which the grit is collected. Fig. 150 shows a deep 1 Report on Sewage Disposal System of Rochester, 1913. GRIT CHAMBERS, RACKS AND SCREENS 335 pi£, such as is used when it is desired to have a few inches of stagnant sewage over the largest quantity of grit which it is proposed to collect before cleaning. Fig. 151 shows a shal- low form of construction used when it is desired to have sufficient velocity at all times in all parts of the chamber to prevent the deposit of organic solids. The length of the grit chamber should be governed mainly by the rate at which the grit settles in sewage flowing at the assumed minimum rate, a subject discussed in Chapter XII. Cleaning Grit Chambers. — Provisions must be made for cleaning a grit chamber as soon as the deposits in it reach such an elevation that fluctuations in the velocity of the current above them disturb the surface of the grit. If there is but one basin in the chamber, there should be a bypass around it to carry the sewage during cleaning, unless use is made of some mechanical method of doing the work without interrupting the flow through the chamber. It is probable that mechanical cleaning can be employed on quite small basins, as buckets are now made in small sizes easily handled by portable derricks or gantry cranes. If a basin must be put out of service for cleaning, some provision must be made for unwatering it in order to get at the grit. Where this cannot be done by drains, as in Fig. 151, portable diaphragm or centrifugal pumps may be used. Where hand cleaning is done, provision must be made for shoveling the grit into buckets which can be dumped into the carts which remove it. Inasmuch as rapidity in the removal of grit is desirable, even when the material is handled by laborers, the permanent installation of tracks and a pillar crane as shown in Fig. 151 has much to commend it. These chambers are most useful during the first part of a heavy storm, when their capacity may be taxed to the utmost, and the ability to clean a basin quicldy may save considerable damage to a pump or the filling of valuable space in an Imhoff tank with mineral grit which should never be found there if practicable measures to prevent it can be provided. Where the basins are cleaned by clamshell buckets, it is de- sirable to protect their floors with some hard covering, such as a granite block pavement or steel plates, to prevent the buckets from breaking them. Theory of Screening. — There are two distinct sei-vices ren- dered by screening. The first is the removal of material likely 336 SEWERAGE AND SEWAGE DISPOSAL to obstruct or injure machinery, to render dUuting waters un- sightly, and to cause deposits of coarse material in siphons and on the bottoms of bodies of diluting water. For this purpose coarse screens are used, generally made of parallel bars which are seldom less than 3^ in. apart. Fine screening is carried on to remove a much larger proportion of these substances and is essentially a treatment process, whereas coarse screening is rather a precautionary process to insure the satisfactory opera- tion of the sewage disposal or sewage-treatment system. Fine screens are commonly made of perforated metal or wire cloth, the perforations rarely exceeding 3^^ in. in size. In using them care must be taken not only to keep them clean but also to prevent sewage passing through them with velocities high enough to macerate the organic matter which it is desired to intercept. The claim that fine screening is the equivalent of sedimentation, which is sometimes made, is not justified. While most of the matter taken from sewage by fine screens can be removed by sedimentation, not all of the material which will be collected in sedimentation tanks is removed by fine screens. In sedimenta- tion, weight determines the separation of solids from the sewage, whereas in screening the separation depends primarily upon size. Screening tends to break up the suspended solids whereas sedi- mentation leaves them unchanged. From such data as are now available it appears that coarse screens will remove from less than 5 to perhaps 10 p.p.m. of coarse material and that fine screens may be expected to remove from 20 to 70 p.p.m., when the sewage is strong and fresh and where conditions are otherwise favorable for screening. The disintegration of suspended matter during its passage through sewers and pumping machinery has an important bearing upon the efficiency to be expected from a screening plant. In this way much of the matter is converted into a colloidal condi- tion, which enables it to pass the screen. It is important, there- fore, when fixing the size of screen openings, to give careful consideration to the nature of the sewage which must be screened and the location of the screens. Under certain conditions it may be wise to provide for screening the sewage before it has time and opportunity to become disintegrated and colloidal. Racks or Parallel-bar Screens. — Racks are often used in con- nection with grit chambers, and in such cases the best position for them depends to some extent upon the type of grit chamber QRIT CHAMBERS, RACKS AND SCREENS 337 and the kind of machinery, if any, used to remove the grit. Where the grit chamber is large and intercepts much coarse or- ganic matter, the work of cleaning the racks will be less if they are placed at the outlet of the chamber instead of at the inlet. An objection, sometimes serious, to placing racks above grit cham- bers is that when partially clogged the racks cause the sewage to back up in the sewers, forming deposits there rather than in the grit chambers. As the sewage passes the racks it is given an irregular eddying motion likely to reduce the sedimentation in a chamber immediately behind the racks, which is another reason for placing them in the outlets of such chambers. Racks are of two general types, those which are stationary or "fixed," and those which are in motion while in service and are known as "movable." The same classification applies to fine screens. Fixed Racks. — Racks and screens are generally placed at right_angles to the axis of the channel they cross. Their length should depend upon the quantity of sewage to be screened, the size of the openings and the permissible loss of head due to them. If only a narrow channel is available, the desired screen- ing area may be obtained by curving the rack in plan, or giving it an angular shape in plan, or placing it on a very flat slope. Boston experience indicates that with combined sewers the rack should have an area of at least 150 per cent of the channel leading to it. The size of the openings depends upon the purpose of the screening. Where the rack screens sewage passing to centrifugal pumps with closed impellers, it is necessary to take out material of rather fine size in case the pumps are small, this size being ascertained from tables supplied by the makers of the pumps. Where open impellers are used, the openings of the racks need not be so small, and a clear distance between bars of ^ in. has been found to work well. The size of the openings in racks guarding treatment works is gradually becoming standardized at ^i to l}-^ in. A 1-in. opening has become standard in the Boston district for the racks screening sewage before its disposal by dilution. Where the screening devices have an important function which must be performed regularly, it is necessary to provide dupUcate apparatus so that one unit may be in service while the other is being repaired or cleaned. The duplication is provided either 338 SEWERAGE AND SEWAGE DISPOSAL by using two channels, each containing one or more racks, or by placing one rack behind another in the same channel. The preference is generally given to two channels because where there is but one an accident to a screen may render it diflScult for the other to serve its purpose. There is considerable variation in the degree of importance attributed by managers of sewerage works to keeping the racks clean. In some small separate systems, where no storm water enters the sewers, hardly any attention is given to the removal of screenings. In some large stations, like those of the Boston Metropolitan District, the removal of the screenings is occasion- mm Fig. 152. — Waste-weir at rack of an Emschergenossenachaft plant. ally an arduous task. There is apparently no way to foresee exactly what the conditions will be, and it is the designer's duty, therefore, to make the racks and their supports strong enough so that in case they become clogged and the sewage is dammed back by them, they will resist this extra pressure until their condition is relieved. If there is any likelihood of the development of such a head upon the rack, it may be desirable to install a float- operated alarm system which will automatically give notice when the sewage above the rack reaches a dangerous height, or this may be avoided by providing a bypass with a screened waste weir of sufficient length, as indicated in Fig. 152, to prevent the sewage from rising above a certain height. Inclined racks are more easily cleaned than vertical racks ORIT CHAMBERS, RACKS AND SCREENS 339 and give a larger screening area, and for these reasons are most used. The indination is from 30 to 45 deg. from the vertical, in most cases. In detailing the construction it is desirable to curve the top of the bars backward, so that the rake used in cleaning can be pulled back freely and carry the screenings into a trough or to a platform at the top of the rack without any difficulty. Where the bars forming the rack are of considerable length and must be supported at intermediate points as well as at the ends, they should have lugs or ears by which their rear edges can be attached to the supports in such a way that these supports will offer no obstacle to the tines of the rakes used in cleaning. The details of the sides and bottom of the channel where the rack lies should be so worked out that there are no openings larger than the clear openings of the rack, particularly where it is desired to take out material less than 1 in. in size. In all but very small installations grooves are left in the walls of the channel for the insertion of stop planks when it is desired to repair the racks. If stop planks or their equivalent are not provided for, it will be necessary to close the channel with sand bags or some other temporary expedient in order to repair the racks. Cleaning Fixed Racks. — No matter whether it is expected that cleaning will be infrequent or frequent, provision must be made for it and for the expeditious removal of the screenings, which have usually a very offensive odor. When the racks are cleaned by hand, it is desirable to have some kind of trough or platform to receive the screenings as they are raked over the top of the racks. At a few European plants the screenings are raked over the top into a cart or small truck which is placed below the over- hanging upper part of the rack, but such elaborate detailing is not common in the United States. Mechanical cleaning is employed with some large installations of fixed racks. This is the case at Toronto, Ont., where there are six racks, two in chambers 33 ft. deep and four in chambers 14 ft. 3 in. deep. Each rack is made up of bars 103^ ft. long, }4 in- thick and spaced to give 3^-in. openings. The width of a rack is 5 ft. 8}i in. They are cleaned by means of rakes attached horizontally to endless-chain belts driven by shafting and gearing at the top of the racks. The material clinging to the tines of the rakes is detached by a cleaning bar with four rows of teeth, The bar is placed horizontally in the headframe of the mechanism SEWERAGE AND SEWAGE DISPOSAL GRIT CHAMBERS, RACKS AND SCREENS 341 for moving the chain belts and revolved so that the teeth pass between the tines of the rakes. All the refuse brought up in this way drops into a sloping tray or trough which discharges it into a screw conveyor. This conveyor also removes grit dropped into it by a bucket elevator used in cleaning the grit chamber in front of the screens. Fig. 153 shows this installation while it was being set up. On the right is the head frame for operating one of the endless-belt devices, with one of the rakes attached to the chains and provided with a part of the malleable-iron teeth which are used to drag the screenings to the surface. The Yenf fo ChimiKi/ ■4ii^m^//li^&'///A^^//////^^ Fig. 154. — Screen chamber at Ward St. station, Boston. revolving bar for cleaning the rakes is also shown in place. On the left are the head frames of some of the bucket elevators used in raising grit. Cage Racks. — The only type of movable racks which has been used to any extent in the United States is shaped like a cage, and is raised vertically to be cleaned. It has a bottom of steel plates or a grid of iron bars and three sides formed of upright bars, the fourth side being the opening which is placed aci-oss the channel through which the sewage passes so that the rack resembles a basket. Fig. 154 is a plan of a typical screen chamber at Boston and Fig. 155 gives a general idea of the machinery by which a cage is operated. Each cage is about 9 ft. high and wide 342 SEWERAGE AND SEWAGE DISPOSAL VI c o Af-S Rj ivS' > s u3 d ■-. lO o '^- TJ s GRIT CHAMBERS, RACKS AND SCREENS 343 and 3J^ ft. across from front to back. The bottom is of steel plates perforated with holes for drainage, and the sides are verti- cal ^-in. bars with clear openings of 1 in. left between them. The cages are counterweighted and operated by small reversing engines. Similar cage racks more recently installed at Boston, Washington and Milwaukee are equipped with electric driving motors. During heavy storms the screens need constant atten- dance and cleaning. The screenings are pressed in a hydraulic press to drive out a large part of the moisture and are then burned in a boiler plant in the immediate vicinity. Wing Racks. — The wing or Frankfort rack is a type of movable rack developed for the purpose of keeping all parts subject to frictional wear above the sewage. This is accomplished by plac- ing three to six racks on a revolving horizontal shaft, as shown in Fig. 156. As it is necessary to have at least one rack intercepting the entire channel at all times, for otherwise part of the sewage will not be screened, a depression must be made across the bottom of the channel and at least three racks used. The larger the number of racks, the smaller the depression need be. The racks are cleaned mechanic- ally. A pendulum arm hinged at the top carries a brush at the bot- tom, which pushes the screenings from the inner edge of each rack toward the outer edge, finally de- livering them upon a belt conveyor. The face of the brush is protected by a light rake. The rack revolves in a direction opposite to the cur- rent, and the pressure tending the rack is somewhat greater rent itself. Fig. 156. — Frankfort wing rack. to force material through than that due to the cur- If the rack is built with small openings, the col- lection of screenings against its face may increase the pressure so as to make considerable power requisite for driving the apparatus. A modification of this form of rack, in which the racks are curved instead of flat and a somewhat different method of clean- ing them mechanically is used, has been introduced in a few European cities. It is called the Geiger rack. 344 SEWERAGE AND SEWAGE DISPOSAL Link -bar Racks. — Very broad endless link belts are used as mov- able screens in several German cities, the most notable installation being at Hamburg. The latest of these plants there has belts made of links 14.2 in. long, held in angle-iron frames about 9.8 ft. long and 15 in. wide, so as to leave 0.4-in. openings between the links. Each . rack consists of 46 of a a these frames, hinged S together in such a way I that the openings be- '1 tween the frames are I very small. The links ^ were first made of wood, ^ then of hard rubber and I finally of an aluminum "5 alloy. The rack moves tj around sheaves at the ^ top and bottom and is cleaned by a rake or stripper having a long row of rubber teeth passing across the entire width of the rack. Drum Screens. — Drum screens are hori- zontal cylinders covered with perforated metal or wire cloth and re- volved slowly while the sewage flows into them. The screenings are col- lected inside the cyl- GRIT CHAMBERS, RACKS AND SCREENS 345 inders and must generally be loosened from the screens by jets of water or air directed against the outside of the cylinders near their top. In Europe the cylinders are usually short and of large diameter, while the few used in the United States are longer and of smaller diameter. This type was first developed in the United States by O. M. Weand for use at Reading, Pa. The first installation had a 40- mesh screen on a cylinder 6 ft. diameter and 12 ft. long, making about 8 r.p.m.. Fig. 157. It was used for 43^ years screening an average daily flow of 5,500,000 gal. from an estimated contribu- tory population of 40,000. From 5,000 to 10,000 gal. of water were needed daily to clean it and about 5 hp. was required for its operation. It removed about 1,500 lb. of wet screenings per million gallons of sewage, and these screenings were partially dried to a weight of 950 lb. in a centrifugal machine. About 20 per cent of the suspended matter in the sewage was removed in this way, at a total cost, including maintenance and repairs, of about $2 per million gallons of sewage (under pre-war prices). Another screen of this type was placed in service in Brockton, Mass., in 1911, but has since been removed. It was 12 ft. long, 6 ft. in diameter and covered at first with 38-mesh wire cloth, later replaced by 30-mesh cloth. It was driven by a 20-hp. oil engine and about 20,000 gal. of water were used daily in washing it. From 1,500 to 2,000 lb. of screenings were removed from each million gallons of sewage, and were partly dried in a centrifugal machine. They were used for fiUing at first, being covered at once with ashes or soil, but later they were mixed with coal and burned under boilers, as was done at Reading. A drum screen llj^ ft. in diameter and 93^^ ft. long is used to screen settled sewage at Baltimore. A 26-mesh wire cloth is used. The introduction of the screen made it possible to dispense with about 50 man-hours of labor daily for cleaning the nozzles of triclding filters and kept the filters in better operating condition than was practicable when that amount of labor was used. The Dorrco screen, Fig. 158, was first developed for screening industrial wastes and was first employed in screening municipal sewage at the Dorr Company's experimental activated sludge plant at Mt. Vernon, N. Y., where it removed 12.2 per cent of the suspended solids and delivered screenings with 80 per cent average moisture content. The drum is a plate punched with Hi X 3^-in. slots parallel with the shaft on which the drum 346 SEWERAGE AND SEWAGE DISPOSAL rotates. The drum is partly submerged in the sewage, which passes through the slots and is discharged through an open- ing in one end of the drum. Except for this opening, both ends are closed. At one point on the circumference of the screen a 1 by 1 by ^^-in. angle extends the whole length of the drum. The drum is driven so as to have a peripheral speed of Screen Drum, Slofi-ed Punched Plate .■%ix'4Sh1-s. .Screened Sewage Discharue ffrimg Sprocket .-$piJlwaij 'Screened Sewacfe Discharge ^j~\Seyrag9 Inlef ^-■f -Serfage Inlet Screenings Seitling ^-' Plan Fia. 158. — Dorroo sewage screen. 300 ft. per minute. About Yi hp. is required to drive a drum 4 ft. 10 in. in diameter and 4 ft. long. The makers rate the screen at 50,000 gal. per 24 hr. per square foot of screen area. The raw sewage coming through the inlet strikes the screen as it rotates and the coarse particles are held against the outside of the screen by the pressure of the sewage. The rotation of the screen causes a depression in the water level at the point of sub- GRIT CHAMBERS, RACKS AND SCREENS 347 mergence and an elevation of the water level at the point of emergence, thus causing an outward flow of the liquid at the latter point. The coarse particles are carried on the surface of the screen until they reach the point of emergence, where they are forced off by the outward flow of the liquid. By this action, the space at the point where the screen emerges from the liquid becomes filled with coarse solids, and as the angle on the circum- ference of the screen reaches this point it carries a portion of the solids up and throws them over the baffle into the screenings Ball Bearing General arrangement. Segmen-h iTIa] 'C"ScreeningD!s» I '. II I' ' un.tL. iii M. i.g v.m\\ I Screening plate. Detail of seal. Fig. 159. — Rienach-Wurl screen. settling chamber. Here the screenings settle and are drawn off by a pump. Any liquid carried with the coarse solids into the screenings settling chamber flows under the baffle and over the side wall into the shallow trough which returns it to the screen chamber. Disc Screens. — The disc, separator, or Riensch-Wurl screen, provides an adequate screening area by placing the screening surface on a very flat slope, as shown in Fig. 159. It was devel- oped in Germany and has been used to some extent in the United States. It consists of a disc made of sheets of perforated metal, 348 SEWERAOE AND SEWAGE DISPOSAL with ol- without a frustum of a cone attached to its center, the whole mounted on a shaft whose inclination from the vertical determines the tilting of the disc. One-fourth to one-third of the area of the discs is cut out by the perforations through which the sewage passes. The weight of the discs in screens up to 16 ft. diameter is carried by a ball bearing at the top of the shaft. In larger sizes the shaft is stationary and carries an annular ball-bearing support on which the frame moves. In each case the bearing is above the level of the sewage. As the screenings are raised above the surface of the sewage, they are swept into a circular gutter by brushes on the ends of the arms of a large spider. The brushes revolve and the arms Be/t Tension Adjusting Device—.^ Trough for Screenings. .^ fWcrferWfieef Sfreave Pig. 160. — A typical English belt screen. also revolve, the combined motion of the several parts being such that every portion of the screening surface of the disc is passed over at least twice. This cannot be accomplished, however, with the conical screen in the center of the disc, which is cleaned by a vertical brush of conical form. Where there is much fat in the sewage it has been found desirable to blow steam or hot water over the disc once or twice a day, and the condition of the brushes is improved by the same treatment, and by washing with kerosene oil. Screens of this type have been constructed with discs from 4 to 26 ft. in diameter and with perforations from 0.03 to 0.2 in. wide. The largest plants installed down to the time of the world war were at Dresden and Bremen. Belt Screens. — Endless-belt screens are used quite extensively in Great Britain. They are generally supplied by a manufac- turer at Carshalton, in Surrey, whence the common British name GRIT CHAMBERS, RACKS AND SCREENS 349 of "Carshalton" screen used for them. The belt consists of twisted steel wires and plates perforated with J-i to J^-in. holes, and is operated as shown in Fig. 160. It is not considered desir- able to attempt to screen more than about 3,500,000 gal. per day with a single belt; two belts should be used for a larger quantity of sewage. A belt screen used in screening sewage from the Union Stock Yards at Chicago consists of two parallel chain belts carrying removable frames to which 40-mesh wire cloth is attached. The cloth is cleaned by jets of compressed air, jets of water and a cylindrical brush. About 1,000 lb. of refuse per milhon gallons of sewage are removed. Screenings. — Reliable information regarding the quantity of screenings is meager and somewhat incomplete. Table 44, however, gives typical data obtained with various kinds of screens at certain American cities. The offensive character of screenings makes their prompt removal very desirable. Coarse screening to remove large floating matter in sewage will continue to be practiced extensively. In this connection however the experience with screens at Worcester, Mass., is instructive. When the chemical precipitation works were put in operation in 1890, about 3,000,000 gal. of sewage was screened through racks with openings of about ^ in. Cleaning them called for about all the time of a day laborer during dry weather and had to be carried on continuously through the 24 hr. in wet weather. The expense was such that the racks were taken out and the whole of the suspended matter in the sewage went to the sedimentation basins, until the grit chamber. Fig. 150, was built. At this time the sludge was pumped from the sedi- mentation basins by a centrifugal pump and it was found necessary to screen the sludge before it entered the pump. A laborer had to be stationed at the rack to keep it clean while the pumping was being done. Eventually a Shone ejector replaced the centrifugal pump, and as it was able to handle very coarse material screening was unnecessary. Thus, by practical experiment, the entire cost of screening was eliminated at Worcester, except for screen- ing the portion of the sludge pumped to filter presses. This experience corresponds with that in Birmingham, England, where it has been found more desirable to screen sludge than sewage. 350 SEWERAGE AND SEWAGE DISPOSAL c3 fac 11 um and) eshes o o >c o CO o ."f oJ t^ o CO c^ '2 « "J^ ">> 1 '"' '^- tf ■* >^ •_ „ ^ ^5 s 1 2q N « '^ . .00 • -* • >, >i ^ : In ! CD '■ o o SSX 0^ • — ^ — ' u a Ol ^ "^ a: V S > o o Tjf i> r> o g >> A ss-^ ■" Csi O CD i-H CD 00 H 3- CO i » j,c~.-=o-.g O o a p J ^ ■< >< M o> o » >. N O O CO lO t^ o s S § g ^ i^ CD O < to 3 X S aH to ^ Tf 6 CO oj w' »« ■ rQ Ol ^ 00 -H OJ 5 3 M d ^3 :s^ a Ph -S 3 o ft c : : : :18 : o >< & o : ! ; ! CO ■ Fh a" o 2 ^ Q ■♦J CO P 1 -c "o . . . . t^ . cq k< ft c . . . . to . C o o '.'.'.'. c^ '. Til O s S S • : ^ ft ! ft ) -^ H H^ g ft a; ^ .^ < " a s „- o" S O .a s £ 1 g ° a> en 4 J o» in tllll II „ C s" S _ q < ^ a *> a ~ CO / 1 1 ,S 7 i o j to X u o il \ 1 'A ^1 1 Di Kg m H^^ — A • ■ ■^ ff Cr^ o: u: M* i-t- »J> vfl r^o a^ =^V S y / ■* / t/ / 5^ f / --• 5^ y ^ / /§ / s. < S f< •o > S ^ *• ", a-^ / £2 '^i / s g S ' ^ i \ 3 " c 3 3 J 3 O •JOjiiiH jsd s+jBj.Bi 4U5I1UU3. *ij sp!|09 p5pus+ot that the particles are of about the same character. Practically, the sedimentation of sewage is greatly comphcated by the changes constantly occurring in its composition, by the evolution of gas. SEDIMENTATION AND SEPTIC TANKS 353 by the decomposition of sludge, by the presence of extremely fine suspended matter and colloidal matter, and by temperature changes. Currents exist in basins at times to such an extent that the actual velocity of a particle of suspended matter may be several time the average velocity with which the whole volume of sewage passes through the tank. As a matter of fact the velocity of Wooden Channel t \ — \ ♦ — t t — t t — f \ — t t -JJT Cui Oiif for Cleaning -^r^- t i — t t ^ -5 Ci/f-Oi/f-fisr CJeantr?a P\aTi'f wH-h Additions of 1B93 Origlncal Plan+«rf FiQ. 162. — Sedimentation tanks at Worcester, Mass. flow within practical limits appears to have very little if any effect upon the efficiency of sedimentation. The period of detention, however, is very important essentially governing the proportion of the suspended matter removed from a particu- lar sewage. To attempt to estimate the sedimentation in a large settling basin on theoretical principles is very likely to prove 23 354 SEWERAGE AND SEWAGE DISPOSAL unsatisfactory, and the authors prefer to employ as a general guide in such estimates the actual results obtained by settling basins in a considerable number of cities which furnish sewages of different chacteristics. Fig. 161 gives such data. Types of Sedimentation Basins. — The older type of settling basins is a shallow, rectangular structure through which the sewage flows horizontally. Recognizing the fact that fine and light particles of suspended matter require a longer time in which to settle than do heavier particles, many of the older sedimenta- tion plants were designed to provide a long route through which the sewage was obliged to flow. This was frequently accom- pUshed by arranging the tanks in series, like the original chemical precipitation tanks at Worcester, Mass., Fig. 162. Here the sewage was admitted to the first tank at a, allowed to flow through the tank and out at b, then in the same way through each of the other tanks, unless one or more were cut out for cleaning. Relatively high velocities caused considerable agita- tion at the inlet and outlet weirs. Additional tanks were subsequently provided and the old tanks were used for roughing basins to collect the heaviest part of the suspended matter. The sewage passed through them in series but was admitted to the new tanks in parallel, thus providing for a minimum rate of flow in the latter. This method of operation proved much more satisfactory than the old method, especially because it facilitated the handling of the sludge, a large portion of which settled in the roughing tanks. These were cleaned at frequent intervals, at times as often as once in 2 or 3 days. The secondary or finishing tanks received relatively little sludge and did not require cleaning more often than once in 2 or 3 weeks in summer, to prevent active putrefaction of the sludge, while in winter they were sometimes used for 5 or 6 weeks without being cleaned. Some settling tanks are cylindrical, sewage entering somewhat above the bottom and overflowing at the top. Fig. 163. Sludge is drawn from the conical bottom without requiring the emptying of the tanks, and may be delivered by gravity at a considerable elevation above the bottom of the tank. In such tanks, the upward flow must be less rapid than the downward velocity of most of the settling particles, some of which tend to collect in a stratum between the influent orifice and the overflow weirs. Particles moving upward may be caught in this stratum, which SEDIMENTATION AND SEPTIC TANKS 355 acts as a sort of filter and is probably some aid to the sedimenta- tion in such tanks. Vertical-flow tanks are particularly adapted to the clarifica- tion of the effluent from trickling filters, because the suspended matter in such effluents, if allowed to accumulate on the sides and bottom of very shallow tanks, will reduce the quantity of dis- ■ Collector Trough Removed Showing Slwl frame Work Embedsled Jn Concrete ''W Sectional Elevation A-^ Sectional Elevation 5*6 Fio. 163. — Vertical-flow settling tanks, Gloversville, N. Y. solved oxygen in the liquid passing through them. This makes it desirable to reduce as much as possible the time of contact of the liquid and sludge, which is best accomplished in this type of tank. Detention Period. — When sewage stands quiescent or moves with a very low velocity the suspended matter settles out gradu- ally, the heaviest portions first. After prolonged standing, a part of the suspended colloidal matter may be thrown down gradually through coagulation due to physical contact or changes in the 356 SEWERAGE AND SEWAGE DISPOSAL composition of the sewage, and eventually a part of the dissolved colloidal matter may be precipitated similarly. Such prolonged sedimentation is not practical in sewage treatment, not only because of the expense of providing sufficient settling basins but also because the changes which take place in sewage and sludge would generally render such a process inadvisable. The quantity of suspended matter precipitated in various periods of time from the sewage to be treated should be ascer- tained if possible, and that period of sedimentation chosen which will cause the removal of the largest quantity required by the conditions. This fixes the detention period, or the length of time assumed to be required for the influent to displace a tankful of sewage. If the rate of sewage flow is 2,400,000 gal. a day and sedimentation for 1 hr. is considered sufficient, then the detention period will be 1 hr. and a tank capacity of 100,000 gal. must be provided. Such a period may be very short, perhaps not over 30 min. in some cases, and it is unlikely to exceed 4 to 6 hr. In the case of horizontal-flow tanks, it was formerly the practice at some works to fill the tanks with sewage, allow it to stand for the desired period of time, and then draw off the super- natant liquid. This method of operation, called the fiU-and-draw method, is very rarely so desirable as the continuous-flow method of operation. If the sewage passes through sedimentation tanks sufficiently slowly to provide a suitable period of detention the sedimentation efficiency is substantially as good as in the fiU- and-draw method where a similar detention period is afforded and the operation of the tanks is much more convenient and economical. Design of Horizontal-flow Tanks. — It is customary to build tanks from 6 to 10 ft. in depth and with a ratio of width to length of 1:2 to 1:6. It is desirable even though more expensive to keep the width small in comparison with the length, in order to promote uniformity of distribution of the sewage on entrance and uniformity in its flow through the basin. In large basins this uniformity of flow is affected by winds, oscillations of the sewage due to gases escaping from the sludge, and by differences of temperature of the sewage in different parts of the basin. The length is governed by the detention period, the quantity of sewage to be handled, and the area available for the basins. The number of basins should be sufficient to afford flexibility of operation by cutting in or out one or more basins. Even small SEDIMENTATION AND SEPTIC TANKS 357 works should be provided with at least two units, to allow clean- ing and repairs without interrupting service. The heavier and greater part of the settling solids subside quickly near the inlet end if velocities are not excessively high. Therefore it is convenient to provide a somewhat greater depth at this end, so that the sludge may be retained as close as possible to the sludge outlet sluices, which should be placed at the inlet end of the tank. This wUl result in saving considerable labor where tanks are cleaned by scraping and squeegeeing the sludge. The gates for drawing off the supernatant hquid when the basins are to be cleaned may well be placed at the outlet end of the tank, K-'?* Fig. 164. — Baffle plate in front of inlet pipe of settling basin. where the sludge is usually not so deep as at the inlet end and it is frequently possible to draw the water lower than at the other end. The most obvious way to effect uniform distribution of the sewage entering and leaving a basin is by means of inlet and outlet weirs extending entirely across the ends of the basin. But, while such weirs are satisfactory for the effluent, they are not satisfactory for introducing the sewage into the tank. This is chiefly because of the tendency of suspended matter to settle in the influent channel and to collect on the edge of the weir. Attempts have been made to solve this difficulty by cutting orifices in the side of the influent channel, but they are generally unsatisfactory. The best method appears to be to provide a few relatively large openings through which the sewage will flow with 358 SEWERAGE AND SEWAGE DISPOSAL moderately high velocity, which is checked almost immediately by small baffles placed in front of and close to the influent open- ings. Fig. 164 illustrates such a baffle used by the authors in a covered sedimentation tank. Baffles and scum boards, used quite generally to assist in maintaining uniformity of flow, must be carefully arranged, for increasing the rate of flow by too much baffling will result in mix- ing and stirring up the sludge. Scum boards extending a few inches below the surface are usually placed near the outlets, and sometimes near the inlets also, to prevent movement of the scum. Detdit rf I Fig. 165. — Swiveling outlet pipe tor settling basins. ' Slip Joint. Sometimes transverse baffles extending to a considerable distance below the surface are placed at equidistant points throughout the length of the tank, these in turn alternating with sludge dams 1 ft. or more high, built across the bottom of the tank. Experiments made at Philadelphia show a marked superiority in baffled tanks over similar tanks without baffles. In many small and medium-size plants the sedimentation basins have been roofed, sometimes on account of severe winter weather and sometimes to conceal the tanks and their contents from public view. There is a small but indeterminate increase in the efficiency of the tanks caused by roofing them but not enough to justify the expense of the covering. Open tanks SEDIMENTATION AND SEPTIC TANKS 359 should have their walls extend at least 6 in. above the surface of the liquid in order to prevent the latter from being blown over the coping by winds. They should also be surrounded by fences if there is any danger of children falling into them. When a tank is to be cleaned, the supernatant sewage is drawn off by means of a pipe extending from the bottom to the top of the basin and provided at the bottom with a swivel joint, Fig. 165, which permits it to revolve through an angle of 90 deg. To the top of this pipe is attached a float of sufficient size to prevent the pipe from falling to the bottom in which case it would draw sludge. The float is so adjusted as to hold the pipe a few inches below the surface of the sewage as it is drawn down, and a stop is provided so that when the pipe has fallen to a predetermined point it can go no farther. A gate should be provided on the pipe line with which the swivel pipe is connected so that the up- right pipe can be kept partly filled, because its buoyancy will throw it up out of the water if empty. The upper end of the pipe should be enlarged and cut back at an oblique angle so that as it falls the opening wiU gradually approach a horizontal position and the sewage will be drawn from the surface. Sedimentation in Vertical Tanks. — In these tanks the influent pipe extends to a considerable depth below the surface, where the sewage is distributed at a relatively low velocity throughout the horizontal cross-section of the tank. The pipe ends several feet above the elevation of the sludge deposits, to avoid any agita- tion of the sludge by the incoming sewage, unless it is desired to have the sewage as it enters come in contact with the sludge, to take advantage of any influence the latter may have in pro- moting precipitation by attraction and coagulation, which is apparently of some value with certain industrial wastes. After leaving the inlet orifice, the sewage spreads out as it rises in the tank, and its velocity is gradually reduced to a rate at which the particles of suspended matter are just held in equi- librium, neither rising nor falling. As more sewage passes this zone the suspended matter is mechanically filtered out by the floating mass, which increases in density, and coagulation occurs by the aggregation of particles. When this mass becomes suffi- ciently dense, portions drop out of the stratum and settle down to the sludge at the bottom. The progress of sedimentation in such tanks was studied experimentally by the authors at Gloversville, N. Y., where the 360 SEWERAGE AND SEWAGE DISPOSAL sewage contains large quantities of tannery wastes which often cause a fairly good chemical precipitation and thus aid materially in the efficiency of the sedimentation process. Sewage entered the test tank 16 ft. below the elevation at which it flowed out and samples of the sewage were drawn from the tank through faucets at different elevations. Table 45 gives the results of these tests, which indicate that the deeper the tank the more efficient the sedimentation at a given rate of flow. It will be seen that greater efficiency can be obtained by increasing the vertical height through which the sewage must fiow than by increasing the area of the tank and decreasing the velocity and the height through which the sewage must rise. Table 45. — Percentage of Sttspended Matter Removed at Different Depths in Vertical Tank at Gloveksville, 1909 (Averages of 4 experiments except as otherwise stated) Upward velocity, ft. per hour Depths of sedimentation 4 ft. 7 ft. 10 ft. 13 ft. 16 ft. 6 46 53 59 62 67 8 37 46 55 59 65 10 S6 63 67 70 80 12 52 62 68 70 73 14 49 64 68 73 79 16' 37 51 63 64 65 16 18 55' 59 62" 63 66^ 68 32« 5P 20' 22 24' 27 33" 43 36 57 49 54 64 54 67 74 62 1 One experiment only. * Average of 2 experiments. 'Average of 3 experiments. Design of Vertical-flow Tanks. — The form of construction of these tanks lends itself particularly to the removal of sludge with- out drawing off the supernatant liquid, and they are invariably built with this method of sludge removal in view. The most extensive use of these tanks in North America is in Toronto, Ont., where the installation was built from the plans of John D. Watson, who built and operated a number of them in the sewage works at Birmingham, England. The Toronto tanks, Fig. 166, were built to handle 240 gal. of sewage per capita from a SEDIMENTATION AND SEPTIC TANKS 361 population of 550,000. The sewage is screened before entering these tanks, which are 24 in number. Each is 25 ft. wide and » J3UUDIJ0 100 ft. long with four hoppers in the bottom. The flow Une is about 16 ft. above the lowest point of the hopper and the capacity 362 SEWERAGE AND SEWAGE DISPOSAL of a tank filled to this elevation is about 18,000 gal. The tanks remove about half of the suspended solids, which range from 170 to 400 p.p.m. in the screened sewage. The cost of tanks of the vertical type is generally greater than that of the horizontal-flow type and less than that of the Imhoif type, described in Chapter XIII. Studies made by Wisner at Chicago in 1911 of the cost of the three types, designed for a sewage flow of 200 gal. per capita, indicated that if the cost of a horizontal-flow tank with a detention period of 6 hr. is taken as l.OD, that of a horizontal-flow tank with a detention period of 8 hr. wiU be about 1.33, that of a vertical-flow tank with a deten- tion period of 4 hr. about 1.45 and that of an Imhoff tank with a detention period of 3 hr. about 2.49. Septic Action. — The bacterial action in septic tanks is similar to, and a continuation of, that going on in sewers hitherto explained. It differs from the action in the sedimentation tank owing to the longer detention period, the degree to which bacterial action is carried, and the type of bacteria cultivated. When a septic tank is first filled, the bacteria in the sewage increase rapidly in numbers, any dissolved oxygen or nitrates which are present in the entering sewage become exhausted quickly, the sewage becomes staler and staler and anaerobic decomposition begins. The suspended matter which accumulates as sludge in the bottom of the tank is also attacked by organisms and part of the solids are disintegrated, coarse matter becoming finely divided. In these changes a large amount of gas is produced by the decomposition of the sludge and is retained in it until parts of the sludge become so Kght that they break away from the remainder and rise to the surface. The gases are occasionally given off in such volume that the liquid appears to boil. At temperatures of 16° to 20°C., such conditions may become estab- lished in 3 to 6 weeks. If, however, a new tank is seeded with a small quantity of sludge from an active septic tank the period required to develop a septic condition can be materially reduced The eflSuent from septic tanks generally has a greater avidity for oxygen than fresh or stale sewage and it is probable that under certain conditions over-septicization causes conditions which are inimical to the bacterial life upon which further bacterial treatment of the sewage depends. The danger can be overcome to some extent by thorough aeration of the septic tank effluent. The effluent is likely to be more offensive than fresh SEDIMENTATION AND SEPTIC TANKS 363 sewage, and its aeration is frequently accompanied by offensive odors. Septic Sludge. — -The organic matter in the sludge in septic tanks undergoes such changes that a reduction in its volume is inevitable, but this reduction is by no means so large as was claimed by early advocates of septic treatment. It amounts to 10 to 40 per cent, the average being about 30 per cent. A con- siderable part of the reduction in volume is due to an increase in its density, due to its disintegration and the compacting, resulting from prolonged stay in the tank. The sludge removed from a septic tank does not exceed 20 to 25 per cent of the volume re- moved from sedimentation tanks operated so as to prevent septic action. There is a great variation in the character of the sludge, due to variations in the quality of the sewage, temperature, construction of the tanks and methods of operation. Some sludge from septic tanks is reported to have very little odor but usually it is decidedly offensive. Scum. — Scum forms on some septic tanks. It consists of the coarse suspended matter which floats and its quantity depends mainly on the character of this suspended matter. If the sewage is fresh and the suspended matter not much disintegrated, large quantities of scum are probable. The suspended matter brought to the surface of the sewage forms such a compact mass that the entrained gases can be liberated but slowly. Meantime the formation of more gas in the remaining sludge carries more suspended matter to the surface, increasing the thickness of the scum perhaps to 2 ft. or more in extreme cases, and it often projects 2 to 6 in. or more above the surface of the sewage. Under such conditions, especially in open tanks, the surface of the scum is likely to become dry and leathery, thus forming a fairly tight roof, sometimes cracked by the gas pressure below it. Molds and fungi often develop in this mass, binding it together, and eventually weeds may grow over its surface. Much trouble has been experienced under some conditions, particularly in the south, with the breeding of flies in this mat, necessitating screen- ing and treatment with oil or other insecticide. Volume of Sludge for Which Provision Should Be Made. — In computing the operating capacity of septic tanks, the space occupied by the sludge must be estimated by the same methods followed in the case of a sedimentation basin and the result modified by the reduction of volume already explained. Inas- 364 SEWERAGE AND SEWAGE DISPOSAL much as the dissolution of a part of the soHds takes place only in the organic matter, it may be desirable where combined sewage must be treated by anaerobic methods to pass the sewage through grit chambers before it reaches the septic tanks, but there may be some cases where combined sewage will not readily part with the suspended mineral matter in a grit chamber unless the velocity is checked so that organic matter also settles out, forming an offen- sive mass troublesome to dispose of. If the grit is first removed, the solids settling from combined sewage during dry weather in septic tanks will probably not differ materially in character and amount from those obtained from separate sewage. The additional organic matter washed into combined sewers by storm run-off can be estimated only from local conditions, among which the efHciency of the street cleaning service is prominent. Provision should be made for storing about 60 to 70 per cent of the suspended matter not removed by coarse screens and grit chambers, plus its accompanying water. The total storage space to be provided will be fixed by the assumed lapse of time between tank cleanings, which depends in turn upon the condition of the sewage reaching the works and the method of disposal of the tank effluent. In some cases during very warm weather sewage reaches such tanks in such a septic condition that they can be operated only as sedimentation basins, in order to prevent over-septicization and the production of an efHuent which is not satisfactory for further treatment. There is always the probability that septic tanks built for small American towns will not receive proper attention, and the small basins are usually given large sludge storage capacity in consequence. With the fresh, weak sewage likely to be received at such plants, there is more danger of large amounts of suspended matter in the efHuent, due to high velocities of flow, than of over-septicization, and consequently it is well to provide storage for at least 9 months' accumulation of sludge consisting of 60 per cent of the total suspended matter and enough water to con- stitute about 90 per cent of the total volume. Liquefaction of the soKds may be expected to keep tanks designed on this basis in fair working condition for a somewhat longer period. With plants for large cities, where fairly competent supervision is Kkely to be given, the provision for sludge may be reduced by designing the tanks so that a small amount of the oldest part of the sludge is drawn off at frequent intervals. This procedure is SEDIMENTATION AND SEPTIC TANKS 365 much favored in England, and the bottom slopes, sludge channels and sludge gates are designed to facilitate it. With frequent removal of the sludge, the percentage of water in it will be higher than when a long period of digestion is permitted. The designer ^ & y-_ IS Sewer InleisA^ " No.l OuHsfs- - Ji .. T a No.2 n k s No.3 -//a'- 1 No.4 5top Flank ' Oallerij for 14" Pipe l8"Bypass <-IO- -> <.~wj>o^-. Control House Ouilei- Chqnner Washington, Pennsylyania Inlefs.. Discharge ■f=F ^— (< /i-'— •>(<— VP'— -> Shear , Siphon -I2---^ 'Shear (bate Channel T \ I Dosing TanK Fairmont,' Jlinnesota Oktribufing Conduif in Y(all\. ?• I- i I <~SS' DisfribuHn j Ufall^ S Baffle Colkcfing Con iuif in Wall-^ a se'" ^■S3- yfalh^ Sturgls, Michigan Fig. 167. — Different arrangements of septic tanks. K- — 118' — ^K—II8' — ^ Columbuft , OMo. must keep in mind the fact that sludge with 95 per cent of water occupies twice the space required to store sludge consisting of the same amount of solids but with only 90 per cent of its volume composed of water. 366 SEWERAGE AND SEWAGE DISPOSAL Number of Septic Tanks. — The size of septic tanks depends upon the desired detention period, character of sewage, space desired for sludge accumulation in the tanks, and local topog- raphy. The best size having been determined, the number of tanks depends upon the quantity of sewage to be treated. Subdivision of tank capacity not only gives to an intelligent operator the means of adjusting liis plant to the fluctuations in flow, but also permits him to dilute trade wastes, avoid over-septicization, clean out his tanks in rotation so as to avoid great variations in the quality of the effluent, and carry on experiments with different operating conditions. Fig. 168. — Inlet of septic tank, Wash- ington, Pa. She/ y/eir Drain f>:: 1 yVashing+on, Ttx, Hhaca , N .Y, Floaftng 0n'f;'c9 Cliamier » to Fi'UerS^ Columbus, Ohio. Mansfield.Ohlo Fig. 169. — Outlets of septic tanks. Fig. 167 illustrates four methods of fulfilling these con- ditions. At Washington, Pa., Pratt used four 100 by 25-ft, SEDIMENTATION AND SEPTIC TANKS 367 tanks with an average depth of sewage of 8 to 10.5 ft. The inlet detail is shown in Fig. 168. The outlet detail is shown in Fig. 169. The height of the sewage in the outlet channel is controlled by automatic apparatus. If it is desired to employ series opera- tion, stop planks are placed in the outlet channel between tanks 3 and 4. Then tank 1 or tank 2, or both, will discharge sewage through the outlet channel into tank 3. Here the direction of flow is reversed from that for parallel operation. On reaching the further end, the sewage enters three outlets in an 18-in. pipe bypass, which conveys it to the same number of outlets in tank 4, where the normal direction of flow is resumed. The two tanks at Fairmont, Minn., designed by Marston, have a combined capacity of 80,000 gal. and their effluent is delivered to two inter- mittent sand filters. The five tanlcs at Sturgis, Mich., designed by Pierson, are 8.5 ft. deep and can be employed, by various settings of the valves and gates, to give capacities of 31,600, 37,700, 46,600, 52,700, 58,800, 72,100 or 78,200 gal. The large installation of tanks^ at Columbus, Ohio, designed by Gregory, consisted of four primary tanks, numbered 1 to 4, and two secondary tanks, numbered 4 and 5. The sewage entered each primary tank through four 24-in. sluice gates. In the distributing waU at the lower end of these tanks was an overflow channel at the top, a 5-ft. collecting and distributing conduit in the center and a 5 by 3.3-ft. blow off conduit at the bottom. The sewage passed from each tank into the collecting and distributing conduit through four 24-in. sluice gates, and passed from it through eight 24-in. sluice gates into either or both of the secondary basins. The sewage was drawn from each of the latter basins through eight 24-in. sluice gates into a collect- ing channel in the end wall. Fig. 169. One secondary and one or two of the primary basins were used continuously, except in winter, until septic action became violent enough to interfere seriously with the subsidence of sus- pended solids. Then these tanks were put out of service and the other secondary and primary basins were put in service. The tanks shut off remained full until the sludge had become fairly well digested, which was indicated by the comparatively httle boiling and a change in the color of the supernatant hquid from jet black to a clearer greenish color. When the tanks were shifted the clear liquid in those which had been standing fuU 1 Later converted into Imhoff tanks. 368 SEWERAGE AND SEWAGE DISPOSAL was drained into the river, the sludge renaaining in the tanks. The Kquid was non-putrescible, contained less than 10,000 bac- teria per cubic centimeter and was very low in suspended matter. Roofs. — It was claimed by some of the early advocates of septic tanks that they should be roofed in order to ensure proper anaerobic changes in the sewage and sludge, but this opinion is no longer general. The main reason for roofing them at present, particularly small tanks, is to reduce the odor often arising from them. If the wind is prevented by a roof from agitating the surface of the sewage, the odor is rarely so strong as to cause annoyance beyond the immediate vicinity of the basin. Occa- sionally a covering is desirable to keep the sewage warm or to keep off children. The roofs are sometimes wood but more often concrete. When the latter is used the surfaces exposed to damp air should be finished as dense and smooth as possible in order to prevent deterioration of the concrete by the action of the hydrogen sul- phide given off in the gases from the decomposing sewage and sludge. Vent pipes must be provided so that the sudden filling or emptying of a tank will not cause any large pressure above or below the roof. As the gases are sometimes explosive, a notice to this effect should be posted near the entrance to any roofed tank. Inlets and Outlets. — Action in septic tanks, in liquefjang organic matter, is apparently less active at mid-depth than at the level of the sludge or just below the scum, and it is therefore desirable to draw off and admit the sewage near mid-depth. This also avoids disturbing the scum by currents. Where weirs are used they are usually guarded by scum-boards from 2 to 4 in. in front of them, extending to a depth of at least 2 ft. The nar- row openings between the boards and weirs are easily kept free from scum, and the sewage is compelled by the depth of the boards to take the same course it would follow with submerged openings. Fig. 168 is an inlet used at Washington, Pa. In some cases of small basins, the sewage enters and leaves the basins through pipes running horizontally across the basins a few inches from the walls. These pipes are provided with tees at intervals and from each tee a pipe is carried down to the desired depth, where it ends in a second tee from which short branches run horizontally and end in quarter-bends, opening toward the wall. SEDIMENTATION AND SEPTIC TANKS 369 Spindle fo Standard- - Lifting Nuti--- Top Joint Opened. Four types of outlets used with septic tanks are shown in Fig. 169. At Washington, Pa., there are three of the outlets illustrated used in each tank 25 ft. wide. In the tanks at Mans- field, Ohio, Barbour used two horizontal rows of 2-in pipe running through the wall, 6 in. apart vertically and 12 in. apart horizon- tally. The upper row is 2 ft. below the surface of the sewage. These pipes discharge into a concrete conduit running along the outside of the end wall, 4 ft. deep, 2 ft. wide at the top and 18 in. wide at the bottom. Oppo- site the center of each tank, this conduit passes through a cham- ber where these wage is drawn off through a floating orifice. Each orifice is 2 by 12 in. and sup- ported by an iron frame held in place by two 21 by 36 by 8-in. floats. The orifices can be adjusted for submersion to any depth down to 1.5 ft. The other two outlets shown are self-ex- planatory. In drawing off the contents of a septic tank by means of a swiveling arm of the type shown in Fig. 165, the hp must be fitted with a scum board, and if the scum is tough or thick this form of draw-off pipe may not be so serviceable as that shown in Fig. 170. This has a vertical standpipe consisting of sections about 1 ft. long. Inside each section is a cross-arm with a hole in the center, through which passes a spindle raised and lowered by a gate standard of the rising stem ts^pe. Lifting nuts are fixed on the spindle at such points that the top joint must be opened before the one next it, and the second joint must be opened before the third. By using this detail there is a minimum disturbance of the scum and sludge. In case it is desired to operate with very thick scum, the first and second lifting nuts can be adjusted so that the top joint will not open. Bottom Slopes. — The slopes of the bottoms of septic tanks do not ordinarily differ from those employed with sedimentation basins. Septic sludge free from grit can be moved with scrapers Fig. Sludge Drain 170. — Draw-off pipe for septic tanks. 370 SEWERAGE AND SEWAGE DISPOSAL on slopes of 1.5 to 3 per cent without difficulty, but thick scum, which settles on top of the sludge when tanks are drained, must be broken up before it can be handled easily on such slopes. The general arrangement of the bottom will usually be influ- enced considerably by the method of disposing of the sludge. Wherever practicable it should be discharged by gravity into a well from which it can be pumped or into a sludge drain. There is Httle information about the flow of sludge through pipes, but it is significant that at the septic tanks at Mt. Vernon, N. Y., no difficulty was experienced for several years in draining off the sludge through a 16-in. pipe on a 1 per cent grade. Here the tanks are of the half -hopper type, with the sludge outlet in one of the side walls, so that the floor slopes toward it at the rate of 1:12 to 1: 24 from both end walls and the other side wall. In Great Britain a slope of about 1:15 is considered desirable for the discharge of septic sludge by gravity. The sludge outlets are usually closed by shear gates unless they are so large that sluice gates are necessary for convenient operation; the latter are generally more satisfactory as they are less likely to leak. All conduits for sludge should be run in straight lines between manholes, if possible, in order that any stoppage can be removed without serious difficulty. Small septic tanks which collect only a small quantity of sludge can best be drained into sludge beds, trenches in earth, or compost- ing beds. If on account of local conditions, they can be drained by gravity only by the construction of expensive accessory works, they can often best be cleaned by hand. The supernatant liquid which cannot be drained off by gravity and sludge of ordi- nary density can be pumped by diaphragm pumps into carts, as nightsoil is handled. This causes odor in the vicinity, but is preferable in the case of some small installations to expensive excavation in rock or to far distant sludge beds. The floors of such small tanks may be left flat if the cost of excavation to give easy grades to one point would be heavy. Chemical Precipitation. — Chemical precipitation is a method of increasing the sedimentation of suspended and colloidal matter by the addition of chemicals which form floe in the liquid, the floe drawing to itself the substances it is desired to remove. Calcium oxide or lime, CaO, sulphate of alumina, Al2(S04)3, ferrous sulphate, FeS04, and ferric sulphate, Fe2(S04)3, are used to produce chemical reactions; clay and soil have been added to SEDIMENTATION AND SEPTIC TANKS 371 weight the precipitate, and charcoal has been added as a deodorizer. The theories upon which the chemical reagents act differ somewhat, but are fairly represented by the action of copperas and lime. The lime is added to the sewage to give it the alkalinity necessary for the reaction and then the copperas or ferrous sul- phate is added. The chemical reaction is: FeSOi + Ca(0H)2 = Fe(0H)2 + CaSOi The soluble ferrous sulphate is converted into the very bulky flocculent ferrous hydrate, which strains out the suspended matter as it precipitates. The degree of clarification depends largely upon the quantity of chemicals used. If enough is employed in an intelligent way, the effluent will generally be clear, substantially free from sus- pended and colloidal matter and without pronounced color. In many cases the quantity of chemicals required to produce this result is so great as to entail serious expense, and consequently the process is not carried far enough to give a clear effluent. The handling and disposal of the sludge from chemical precipitation tanks is also a costly matter, for the volume of the sludge often reaches 0.5 per cent of the volume of the sewage treated. Fur- thermore, although the effluent may have a moderately satisfac- tory appearance it is ordinarily putrescible and not comparable with good filtration effluents. These drawbacks to the use of the method have prevented its more general adoption and even led to its abandonment in some cases. The greatest field for usefulness of chemical precipitation at present seems to be in the treatment of industrial wastes. Quantity of Lime Required. — Normal domestic sewage is slightly alkaline. Where lime alone is used as the precipitant, it is necessary to add a sufficient quantity to combine with all the free carbonic acid and the carbonic acid of the bicarbonates, producing normal calcium carbonate, which acts as the coagulant. Much more lime is generally required when it is used alone than when sulphate of iron is also employed. Where industrial wastes introduce mineral acids or acid salts into the sewage, these must be neutralized before precipitation can take place and may in- crease the quantity of lime required. The required degree of alkalinity where lime is the only precipitant may be determined by titrating one portion of the 372 SEWERAOE AND SEWAGE DISPOSAL sewage with erythrosine and another portion with phenol- phthalein. The erythrosine alkahnity must be more than twice that shown by the phenolphthalein test. In case cop- peras and lime are the precipitants, it is necessary to add enough hme to provide a normal carbonate alkahnity. If enough lime is used, a pink color will appear in a glass of the treated sewage to which a few drops of an alcoholic solution of phenolphthalein is added. Only enough lime should be used to produce the pink color. If sulphate of alumina or ferric sulphate is used, the sewage may not need lime, because bicarbonate alkalinity is capable of precipitating these salts. Bicarbonate alkalinity is indicated in this case by the pink color produced when erythro- sine is added to a sample of sewage in the presence of chloroform; no color is produced by erythrosine with neutral or acid solutions. If too much lime is used in the treatment of sewage, some of the suspended organic matter will be dissolved and the effluent may be worse than the raw sewage. If not enough lime is used, the effluent will not be well clarified, and if an acid sewage is treated the effluent may remain acid and be poorly prepared for treatment in contact beds or trickling filters. In Worcester, Mass., the quantity of lime used varies greatly, the dose being determined by the phenolphthalein test. In 1893, 1,233 lb. per 1,000,000 gal. of sewage were used, the amount gradually being decreased to 905 lb. in 1912. In 1908, when the general business depression caused a great reduction of out- put in the local foundries and wire works, but 871 lb. of lime per 1,000,000 gal. of sewage were used. The pre-war practice in Providence, R. I., is indicated in Table 46. In London the sewage was treated with 475 lb. of lime and 119 lb. of copperas per 1,000,000 gal. When a basin is closed off for cleaning, its contents are aUowed to stand for 2 hr. The top 4 ft. of supernatant liquid is then drawn off through telescopic weirs and the remaining contents of the basin are then drawn into a sump and allowed to stand for 24 hr., when the top liquid is decanted, treated with 2,370 lb. of lime and 1,190 lb. of copperas per 1,000,000 gal., and returned to the basins, the sludge stored being carried to sea by sludge steamers. Slaking Lime.^ — ^Lime does not dissolve readily in water and on this account it is best applied to sewage in the form of milk of lime. This is prepared by slaking quicklime in a small amount of SEDIMENTATION AND SEPTIC TANKS 373 Table 46. — Cost of Chemical Precipitation at Providence, R. I.' Year Pounds per 1,000,000 gal. Water in sludge, per cent. Cost per 1,000,000 gal. Lime Copperas Chemical precipitation Sludge disposal Total 1903 1904 1905 1906 1907 1908 1909 1910' 1911 1912 1913 606 683 559 638 654 726 700 486 438 65.0 58.0 51.5 72.1 83.1 94.63 92.46 92.11 92.58 92.15 91.77 91.29 92.07 92.57 90.03 89.60 $3.31 3.42 3.13 3.50 3.54 3.42 3.36 3.11 $2.44 2.57 2.66 3.10 3.07 3.43 4.22 4.06 $5.75 5.99 5.79 6.60 6.61 6.85 7.58 7.17 2.85^ 2.503 2.49 2.30 5.34 4.80 1 Since this chapter was prepared, lime treatment has been resumed. 3 In 1910, the city began to experiment with the disinfection of its sewage, and the following year these experiments entailed so many modifications of operation that all the records could not be abridged into the usual summarized form. In 19 12 chemical precipitation was aban- doned and plain sedimentation followed by disinfection of the effluent was adopted. During the period when chemical precipitation was practiced, the removal of suspended organic mat- ter, as determined by albuminoid ammonia, was from 80.5 to 86.4 per cent, while during 1912 and 1913, it was 41.0 and 40.2 per cent respectively. Lime treatment was resumed for a time about the year 1915 and subsequently discontinued. It is now (1921) being used again. 8 These figures include the cost of disinfecting the sewage. water and allowing it to stand over night, and longer if possible. This is then usually mixed with about 20 times its weight of water, although a thinner mixture is sometimes used. Lime of fair quality will take up about one-third of its weight of water in slaking, but poor grades will not require so much The amount necessary for the lime to be used should be deter- mined by experiment. A quick-slaking lime will require more water than hard, close-grain grades. In general the lime will increase from 50 to 100 per cent in bulk in slaking. The temperature during slaking is an important index of the character of the operation. It should be as near 200°F., or just below the boiling point, as possible. If the temperature stays below 200°, too much water has been added and the slaking is less vigorous and thorough. If not enough water is added, the tem- perature will become too high and the lime will be in danger of becoming "burnt." 374 SEWERAGE AND SEWAGE DISPOSAL Pipes carrying milk of lime may become choked with precipi- tated carbonate of calcium, and it is desirable to have facilities for flushing them with very weak acid and water. It is some- times helpful to blow steam or air through the feed pipes. Mixing Chemicals and Sewage. — The precipitants must be mixed with the sewage before the latter enters the tanks, in order to obtain the best results from this method of treatment. It is inadvisable to add precipitants before pumping sewage, because the floe formed by the precipitants may become more or less broken up and charged with air, which interferes with their Fig. 171. mixing channel or "fish ladder" leading to precipitation basins, Worcester, Mass. sedimentation. One of the most satisfactory means of mixing the dose of chemicals with the sewage is the fish ladder, Fig. 171. Detention Period. — The best detention period depends upon the amount and character of the suspended matter in the sewage, the amount and nature of the trade wastes which may be dis- charged into the sewage, the method of operating the tanks, and the treatment of the effluent after leaving the tanks. British experience favors a detention period of 8 hr. for tanks operated continuously and 2 hr. for tanks operated on the fill-and-draw method, in the case of domestic sewage. At Worcester, Mass., SEDIMENTATION AND SEPTIC TANKS 375 where chemical precipitation has been employed longer than elsewhere in the United States, the detention period is about 7 hr. Design of Precipitation Tanks. — After many years' experience in supervising the operation of chemical precipitation works at Worcester, Mass., and Providence, R. I., Bugbee has summarized his opinions regarding the design of tanks for chemical treatment as follows : The depth of the tank should be between 6 and 9 ft. Since the tanks should be cleaned very frequently, it is unnecessary to provide space for the storage of sludge at the bottom of the tanks, as was done in some of the older precipitation plants. The length of British tanks was formerly made about two and one-half times the width, but it has been found that better results were obtained with a greater proportional length. The bottom of the tank should rise toward the outlet and also slope from each side wall toward a drain running longi- tudinally through the tank. If the bottom could be given a grade of at least 10 per cent, the sludge would all flow to the drain. This is obviously impossible with large tanks, in which it is necessary to force the sludge adhering to the bottom into the drain by means of wooden scrapers. Hence narrow tanks, involving the least travel for the men employed in this work, are most economical in operation. The inlets for the sewage should be as nearly the full width of the tank as possible, in order to reduce the entrance velocity to the minimum and to avoid stagnant corners where septic action is Ukely to occur in a short time. For the same reason no dead ends should exist on the inlet and effluent channels, and the bottoms of these channels should be constructed to retard the sedimentation of suspended matters, just as is done in sewers carrying variable quantities of sewage. The sluice gates through which the sludge is drawn off should be at the inlet ends of the tanks where most of the sludge will be deposited. There should be a floating arm and valve for drawing off the supernatant Kquid before removing sludge from a tank. The valve should be at the extreme end of the tank in order to remove the greatest possible quantity of water from the sludge. Provision should be made for supplementing the work of the scrapers used in sludge removal by means of gates in the side 376 SEWERAGE AND SEWAGE DISPOSAL walls, through which supernatant hquid may be introduced from adjoining full tanks for flushing. The size of the individual tanks should be so proportioned to the flow as to give at least four units, in order that the tanks may- be cleaned as often as once a week in hot weather, without reduc- ing the desirable working capacity of the plant. Frequency of cleaning is an important operating duty, because of its value in preventing septic conditions in the effluent and also because of its effect on the work of sludge pressing. CHAPTER XIII THE IMHOFF OR EMSCHER TANK If the settling solids can be removed from sewage and their digestion take place under anaerobic conditions which will not affect the sewage, the e3di.au|j4mi,Qf the dissolved oxygen and nitrates, if any, in the sewage will be retarded. By keeping the sewage as fresh as possible in this way, odors are reduced to a minimum and the subsequent treatment of the sewage or its disposal by dilution is generally aided. Dr. William Owen Travis developed a type of two-story sedimentation tank. Fig. 172, in which this separation of the processes of sedimentation and sludge digestion is largely accomplished, only one-sixth to one- eighth of the sewage being sent through the lower or sludge- Sewage flows through allfhree Chambers, Sewage -flaws fhiHughSedintenfatj'oriChcimbefsonl/. Travis Tank Imhoff Tank Fio. 172. — Types of two-story tanks. liquefying chamber, for the purpose of seeding the sludge with the bacteria required for its digestion and liquefaction. This development was carried farther by Dr. Karl Imhoff, who believed that it was unnecessary and undesirable to pass any portion of the sewage through the liquefying chamber, as the solids were already sufficiently seeded. He therefore developed the tank known as the Imhoff or Emscher tank. Fig. 172, the latter name coming from the district in Germany in which the tanks were first installed on a comparatively large scale. Sewage flows through the sedimentation chambers only, and the function of these chambers is solely the removal of the settling solids. These solids drop to the inclined surfaces of the bot- 377 378 SEWERAOE AND SEWAGE DISPOSAL torn of a sedimentation chamber and slip through a slot into a sludge chamber. The slot is trapped so that no gases from the sludge chamber can rise into the sedimentation chamber. In the sludge chamber the solids undergo septic decomposition as in a septic tank. The gases given off by this decomposition rise through the portions of the sludge chamber extended to the top of the tank, called gas vents, and scum collects on the sur- face of the sewage in these vents just as it does on the surface of septic tanks, and often to a greater depth. On account of the depth of the sludge chamber and the long period of time the sludge is allowed to digest, the sludge from tanks of this type which are operating normally is more dense than that from plain sedimentation tanks or septic tanks. Theoretically, therefore, the tank carries on simultaneously and independently by means of its two-story construction, the functions of both plain sedimen- tation tanks and septic tanks. One of the most important advantages of this form of tank is its production, when operating properly, of sludge which is easily disposed of. As the sludge is withdrawn from the tank, it contains large quantities of gas held mechanically in it. When relieved from the pressure of the considerable depth of superna- tant liquid in the tank (when the latter are of adequate depth) and spread out upon sludge-drying beds, the solid matter will be buoyed up by the entrained gas, thus permitting the relatively clear liquid below to pass quickly into and through the beds. As the water disappears the sludge gradually settles down on the surface of the bed, and the gas is liberated and replaced by air, leaving the mass quite porous and in a condition favoring rapid drying. Good sludge from such tanks is dark colored, not sticky or pasty but rather granular, free from offensive odor, somewhat frothy and resembling dark garden soil mixed with water. A peculiar feature of the operation of some Imhoff tanks is the production of so much gas in the sludge chamber that the sewage in the gas vents seems to foam or boil. Occasionally the gases have brought up with them so much sludge that the scum has overflowed the tops of the gas vents. The early tanks of this type in the United States were designed on the basis of experience gained in operating tanks with the strong sewage of German industrial towns, and it was soon learned that the areas of the gas vents which were adequate for German THE IMHOFF AND EM8CHER TANK 379 conditions were not large enough for American conditions, par- ticularly where the climate is warm. Apparently the sludge chamber requires alkaline conditions for satisfactory operation. When the Imhoff tank does not operate satisfactorily the sludge is relatively light colored, has an offensive odor, is somewhat sticky or pasty, and does not dry- well. In some cases the addition of hme to the liquid in the sludge chamber has been tried as a cure for the troubles but its efficacy is doubtful. In other cases the troubles may have been due to filling the tanks with fresh sewage late in the fall, when the weather is not adapted to develop bacterial activity, resulting in a long period of very moderate bacterial activity in the sludge chamber. It is considered desirable by some to fill the tanks with hard water before any sewage is turned into them, so that operation will begin with assured alkaline conditions, and if possible to seed the lower compartment with bacteria, through sludge from an operating tank. ' Diffusion Between Chambers. — The sludge chamber of Imhoff tanks is intended to retain its contents in a stagnant condition, without any circulation of hquid through the slots opening into it from the sedimentation chambers above. It is therefore desir- able to design and operate the tanks so that no material fluctua- tions of sewage level will occur in the tanks, for such fluctuations, particularly if sudden, will cause differences of hydrostatic pressure and, consequently, surges of septic sewage up through the slots. Furthermore, if currents down through the slot exist to such a degree that considerable fresh sewage enters the sludge chamber, there is danger of the evolution of hydrogen sulphide. It was a desire to prevent any septic sewage rising from the sludge chamber into the sedimentation chambers, where nothing but fresh sewage should be present in order to assure the most satisfactory operating conditions, that led the engineers under Dr. Imhoff to pay unusually careful attention to grades and other features controlling the elevation of the sewage during its flow through the tanks. The danger of diffusion increases with the number of sedimen- tation chambers over a single sludge chamber. Care must be taken to have the inlets and outlets designed to give the proper proportion of flow to each sedimentation chamber. If the operating conditions are such as to require careful attention from the superintendent, it may be advisable to connect the sedimenta- 380 SEWERAGE AND SEWAGE DISPOSAL tion chambers at each end of the tank so as to maintain the same surface elevation in all of them. Diffusion, temperature changes, and the gradual increase in the volume of sludge in the sludge chamber must result in some septic liquid passing from it into the sedimentation chambers. Investigations by Bach show that, under satisfactory operating conditions, this diffusion is very slight and without substantial effect on the sewage in the sedimentation chambers. The volume of the sludge is decreased from 25 to 50 per cent in passing through the slots, according to these experiments; this is true not only of sewage sludge but, to a somewhat smaller degree, of sediment formed of fresh, washed iron hydroxide and very finely powdered hard coal. Furthermore, as the sludge collects above the slot before passing down through it, its weight and volume hinder the formation of any currents contrary to the motion of the subsiding sludge. In addition to these influences retarding diffusion, the specific gravity of the hquid in the sludge chamber is greater than that of the sewage passing through the sedimentation chambers. Quantity of Sludge. — The size of the sludge chamber is based primarily on the amount of solid matter which is expected to settle out of the sewage during the assumed maximum period between removals of sludge. On account of the relatively high cost of Imhoff tanks compared with plain sedimentation tanks, it is desirable to form as close estimates as possible of the volume of sludge to be expected. In the Emscher District this was not a serious matter, because it was customary there to estimate sewage treatment requirements only 5 years in advance of con- struction. The works were built and operated under skilled supervision and additions made to the plants as needed. In this way the capital investment was kept to a minimum and the utmost service obtained from the installation. In the United States, where it is customary to construct treatment works capable of serving without additions for a much longer future period than 5 years, this greater investment of capital in extra space for future requirements makes it necessary to be as accurate as possible in forecasting conditions. Imhoff assumed in his designs that the average amount of sludge where combined sewers were used would be 0.007 cu. ft. per capita daily and 0.0035 cu. ft. where separate sewers were used. For reasons just explained, great accuracy in these estimates was THE IMHOFF AND EMSCHER TANK 381 not attempted, for the original plans were prepared so that addi- tional tank capacity could be added whenever needed without reconstruction of the older works. Imhoff preferred to estimate the quantity of sludge on the basis of persons connected with the sewerage system rather than on the basis of quantity of sewage on account of the variation in dilution of sewage in different cities. In designing his tanks he felt the most important thing! ^Q \jQ considered was the detention period which he varied to meet local conditions. Where the sewage was fresh and strictly domestic, a detention period of 2 hr. was considered most satisfactory, about IJ^ hr. was the average for all his - .£ 80 o 5 iisnpnded Matter in Inf!uent,400 ParfsperMHIioh-\ ^^o r^ rv 1 —■ " " X £ 60 -" L-J^ y -- "' J 50 / / V 100 |40 , ~ ~ ' ■ / / ^ §30 / 1 / / l2 20 4- • 1 ^ / 1 / / S 10 ^/ / C^ f _ _ _ _ _ _ _ _ _ _ _ __ _ I Z 3 4~'5 6 7 a 9 Period of Sediman+atlon in Hours. FiQ. 173. — Removal of suspended matter in large settling tanks. 10 plants, and where a large amount of iron salts was present in the sewage 1 hr. was considered the longest period that should be adopted. Imhoff opposed a longer period than 2 hr. for German conditions on account of the danger of septic conditions being set up in the sedimentation chambers. The rate of flow through the sedimentation chambers adopted by Imhoff was 108 ft. per hour for dry-weather conditions and 324 ft. when the sewage was diluted threefold with storm water. American conditions are so different from those in the Emscher district that it was felt by designing engineers to be unsafe to employ detention periods as low as those used by Imhoff, in > The relative importance of the factors influencing sedimentation was rated by ImhoS as follows; Detention period, 100; change in temperature, 30; depth, 30; construction of inlets and outlets, 30; character of sewage, 30; velocity through tank, if less than 350 to 600 ft. per hour, 0. 382 SEWERAGE AND SEWAGE DISPOSAL laying out the early American tanks of this type, or to use rates of flow as high as he employed. These early American tanks had detention periods about 50 per cent greater than the Emscher district practice and the rates of flow in the sedimentation Table 47. — Volume op Sludge of Different Specific Gravities and Percentages op Moisture Obtained from 100 lb. of Settled Dry Solids Cubic yards of sludge, with 90 per cent moisture, having Percentage of speciiic gravity, given in column head moisture 1.02 1.03 1.04 1.05 1.06 1,07 99 5.922 5.917 5.912 5.906 5.901 5.896 98 2.955 2.950 2.945 2.939 2.934 2.929 97 1.966 1.961 1.956 1.950 1.945 1.940 96 1.467 1.462 1.457 1.451 1.446 1.441 95 1.175 1.170 1.165 1.159 1,154 1.149 94 0.977 0.972 0.967 0.961 0.956 0.951 93 0.827 0.822 0.817 0.811 0.806 0.801 92 0.729 0.724 0.719 0.713 0.708 0.703 91 0.647 0.642 0.637 0.631 0.626 0.621 90 0.581 0.576 0.571 0.565 0.560 0.555 89 0.527 0.522 0.517 0.511 0.506 0.501 88 0.482 0.477 0.472 0,466 0.461 0.456 87 0.444 0.439 0.434 0.428 0.423 0.418 86 0.403 0.398 0.393 0.387 0.382 0.377 85 0.383 0.378 0,373 0.367 0.362 0,357 84 0.359 0.354 0.349 0.343 0.338 0,333 83 0.337 0.332 0.327 0.321 0.316 0.311 82 0.318 0.313 0.308 0.302 0.297 0.292 81 0.300 0.295 0.290 0.284 0.279 0,274 80 0.284 0.279 0.274 0.268 0,263 0,258 Note. — The sludge in the column headed 1.02 is of the character to be expected from small strictly domestic separate sewerage systems, where, through infiltration or high water consumption, the sewage is very weak. The sludge in the column headed 1.07 is from very strong sewage delivered by a combined system, with but little of the mineral matter removed by grit chambers. A specific gravity of 1.03 may be assumed for average sewage from strictly separate systems and 1.05 for average sewage from combined systems. THE IMHOFF AND EMSCHER TANK 383 chamber ranged from about 7 to about 87 ft. per hour. Recent American practice has shown a tendency to shorten the detention period. Fig. 173 has been found useful by the authors in esti- mating the removal of suspended matter during different detention periods. The quantity of sludge to be expected will depend very largely on its density, as is evident from Table 47. The Imhoff tanks in the Emscher district jdelded sludge of unusual density, but how much this was influenced by the great depth of the tanks, which averaged 30 ft., and how much by prolonged storage is uncertain. What Imhoff desired was sludge which would dry readily without causing offensive conditions, and this he believed could be obtained most economically by deep tanks. The dense sludge from these tanks held more entrained gases, in his opinion, than sludge from tanks of less depth, and he believed that the greater the amount of entrained gases in the sludge the more rapidly it would dry. Experiments by Clark and Gage with the sludge collected in sedimentation tanks 17 and 30 ft. deep partly confirmed this opinion. The total reduction of suspended solids was 19 per cent in the shallow tank and 30 per cent in the deep tank. The sludge contained 91.3 per cent of moisture in each case, but that from the shallow tank contained 76.6 per cent of moisture after draining for 10 days while that from the deep tank contained only 65.9 per cent. The sludge which is collected from sewage differs in the weight per unit volume of the solid matter as well as in the percentage of the suspended matter which has settled in a given period of time. There is a tendency among engineers to compare sludges by expressing their characteristics as those of similar sludges with 90 per cent moisture, which is readily done by means of Fig. 174, from Webster's report on the Philadelphia experiments. The specific gravities of most sludges of which analj^tical records have been examined by the authors lie within 1.02 and 1.07, with 90 per cent moisture. The exceptions to such a relation were too erratic to make their consideration necessary. Sludge of 1.02 specific gravity at 90 per cent moisture contains very light sus- pended matter, such as might be expected from weak domestic sewage. A specific gravity of 1.07 with 90 per cent moisture is indicative of very heavy solids, such as those entering combined sewers from macadam streets in times of storm. Probably 1.03 will prove a fairly 3.ccurate figure to assume for average sewage 384 SEWERAGE AND SEWAGE DISPOSAL from strictly separate systems, and 1.05 for average sewage from combined systems. An examination of American records of septic, sedimentation and Imhoff tank sludges indicates that the probable percentage of moisture is highest in sludges with solids of light weight, rang- ing from 91 to 98 per cent; with average weak sewage the sludge, of a specific gravity of 1.03 at 90 per cent moisture, wUl have Weight in Pounds of One CubicYard of Wet Sludge. 1700 1720 1740 1760 1760 1800 IWO 1840 100 too- 300 1.00 Fig. 174. — Relation between specific gravity, percentage of moisture and weight of dry solids in 1 cu. yd. of wet sludge. Example of Use. — Assume that the sludge is found to contain 85.5 per cent of water and has a specific gravity of 1.057. From the 1.057 point on the scale of specific gravity, at the left, follow a horizontal line to the diagonal. Directly above the intersection, on the top line, will be found the weight of 1 cu. yd. of such sludge, 1,783 lb. Drop a vertical line from the same intersection point and extend to meet it a horizontal from 85.5 on the scale of per- centage of moisture. From their intersection, follow the sloping lines to the right to 258 on the right-hand scale, giving the pounds df dry residue in 1 cu. yd. of sludge. from 90 to 96 per cent of moisture; with average strong sewage, the sludge, of a specific gravity of 1 .05 at 90 per cent moisture, will have from 87 to 93 per cent moisture. The following figures of the percentage of moisture in sludges from American Imho^ tanks indicate that the dense sludges of the Emscher district are THE IMHOFF AND EMSCHER TANK 385 not likely to be obtained here: Stock Yards Testing Station, Chicago, tank 17 ft. deep, sludge with 91.5 per cent of moisture and specific gravity of 1.02. Experimental Imhoff tank at Worcester, 22 ft. deep, sludge with average specific gravity of 1.016 and 91.86 per cent of moisture. Akron experimental Imhoff tank, 16 ft. deep, sludge with specific gravity of 1.12 and 74 per cent moisture. Proctor Creek tanks, Atlanta, sludge with specific gravity of 1.02 and 87.05 per cent of moisture; Peachtree Creek tanks, Atlanta, sludge with specific gravity of 1.02 and 90.2 per cent of moisture. Sludge Storage Capacity Required. — This is estimated in the following manner: From analyses of the sewage or by a study of the sources of the sewage when analytical records are lacking, the amount of suspended matter is first determined. For example, take this as 175 p. p.m., which is probable for domestic sewage col- lected by small separate sewerage systems. This quantity is equivalent to 1,460 lb. of suspended matter per 1,000,000 gal. of sewage. It will be assumed that a detention period of 3 hr. is selected and that in this time 60 per cent of the suspended matter passes from the sedimentation chamber to the sludge chamber, a decidedly high assumption, as reference to Fig. 173 will show. On this basis 876 lb. of solids will be collected in the sludge chamber from each million gallons of sewage. Some of this solid matter will be digested by the anaerobic processes going on in the sludge chamber, and this reduction of the solids will be a'ssumed as 25 per cent, leaving 657 lb. of solids to go into the sludge. The sludge is allowed to remain for 7 months in a deep tank, it wiU be assumed, so the water content of the sludge will be taken as 85 per cent. Table 47 shows that sludge of this water content measures 0.378 cu. yd. per 100 lb. of suspended matter. Hence 6.57 X 0.378 = 2.48 cu. yd. of sludge storage capacity must be provided for every million gallons of sewage passing through the tank during the sludge digestion and storage period. This period may be taken in this case as 7 months, with the result that about 528 cu. yd. of space is required for a daily flow of 1,000,000 gal. In providing this amount of storage for the sludge, care must be taken that the shape of the tank will permit all the room allotted to the sludge to be actually available for it and that the sludge does not approach within 18 in. of the slot measured vertically. With some sewage the ebullition in the sludge 25 386 SEWERAGE AND SEWAGE DISPOSAL chamber and the formation of scum are very active at times, particularly in protracted hot weather. Another method of estimating the volume of sludge space has been developed' by Kenneth Allen, who has proposed the follow- ing f ormiilas : Storage (combined sewage) in cubic feet = 10.5 PD Storage (domestic sewage) in cubic feet = 5.25 PD where P is the population in thousands and D is the detention period in days. The formulas are based on sludge with 80 per cent moisture. It is to be noted that in certain industrial wastes, such as those from packing houses, the total amount of suspended solids may be so much greater than in the sewages ordinarily dealt with, as to require abnormally large sludge compartments, scum spaces and gas vents which will necessitate radical change in their design and in the usual proportions of Imhoff tanks. Radial-flow Tanks. — There are two general types of Imhoff tanks, one in which the sewage flows radially from an inlet at the center of a circular basin and the other in which the sewage flows longitudinally through a basin rectangular in general plan. The Imhoff tanks designed under the direction of Calvin W. Hendrick by Leslie C. Frank for Baltimore are examples of radial tanks. As shown in Fig. 175, the sewage passes from the main distributing channel through a trough 12 in. wide and 12 in. deep to a central distributing ring of the same cross-section: There are eight 12 by 12-in. openings in its bottom through which the sewage passes down into the sedimentation chamber; provision was made for partially closing the openings near the inlet trough with flat plates if this seems necessary to equalize the flow of sewage. One purpose in admitting the sewage to the sedimenta- tion chamber in this way was to give it a downward velocity so that short-path surface currents would not exist. Deep baflEle plates have also been used in radial flow tanks to attain the same end. The settled sewage is drawn off through 50 V-shaped weirs on the inner side of a circumferential effluent collector. The weirs are made of galvanized-iron plates, slotted so that they can be accurately adjusted on the side of the channel by means of thumb screws inserted through the slots. From the ' "Sewage Sludge," p. 228. THE IMHOFF AND EMSCHER TANK 387 Fig. 176. — Radial-flow type of Imhoff tank, Baltimore. 388 SEWERAGE AND SEWAGE DISPOSAL collector the sewage passes into an effluent channel leading to the control house of triclding filters which give the next stage of the treatment. The tanks were designed for a 2-hr. detention period, each tank being designed to care for the sewage from a population of 4,000 persons. *i- 14; To Doiing Tank FiQ. 176. — Partial plan of Imhoff tanks, Fitchburg, Mass. The solids settle through the sewage in the sedimentation chamber and collect in the annular V-shaped space at the bottom, eventually shpping through the slot there into the sludge cham- ber below. It will be noticed that the slot is guarded by a pro- jecting ring on the wall of the sludge chamber, so that bubbles of gas cannot pass through it into the sedimentation chamber. THE IMHOFF AND EMSCHER TANK 389 The sludge chamber was designed to hold 1 cu. ft. of accumulated soHds per capita of tributary population. The depth of the lowest part of the chamber below the surface of the sewage in the sedimentation chamber is 25,5 ft. The gases given off during the digestion of the sludge escape through the central gas vent. The conical upper part of the sludge chamber, or the part above the slot in rectangular tanks, is sometimes called the scum cham- ber because of the large quantities of scum often formed there in some tanks of this type. The sludge is removed through a pipe WeirFlafe Scumboard L Sec+icrn C-,C Fig. 177. — Sections of Imhoff tanks, Fitchburg, Mass. (The lines on which the sections are taken are given in Fig. 176.) by gravity when the outlet gate is opened. The outlet is 4.85 ft. below the surface of the sewage in the sedimentation chamber. Parallel-flow Tanks. — An example of the parallel-flow typo of tank is given in Figs. 176 and 177, showing an installation built at Fitchburg, Mass., from the plans of D. A. Hartwell and H. P. Eddy. This plant consists of five 90 by 30-ft. tanks placed side by side, with an inlet conduit and effluent channel at each end, so that the direction of flow through the tanks can be reversed. The sewage enters each tank through three 12-in. sluicegates, 390 SEWERAGE AND SEWAGE DISPOSAL one for each trough in the sedimentation chamber. In most American Imhoff tanks only one or two troughs have been used over a sludge chamber. The sedimentation chambers were designed to detain the sewage for 3 hr. when flowing at the rate of 125 gal. per capita daily for a population of 55,000. The velocity in the tanks under such a condition is about 30 ft. per hour. The actual detention period in 1918 was from 5.4 hr. in March to 11.4 hr. in July, averaging 8 hr. for the year. The tanks removed in that year 64.6 per cent of the suspended matter in the sewage delivered to them, and about 95 per cent of the setthng solids were removed. At the foot of the tank the settled sewage falls over adjustable metal weir plates into the effluent channel. The solids slip through the three slots into the sludge chamber, which was built large enough to hold the sludge accumulating for 6 months at the rate of 0.007 cu. ft. per capita daily. The gas vents are arranged in four lines along each tank and their area is 17 per cent of the total area of the tank. The bottom of the sludge chamber consists of three hoppers, from which the sludge is raised by air lifts because the sludge beds are too high to be reached by gravity discharge and pumping with centrifugal pumps would tend to remove the gases which seem to be essential for the advantageous characteristic drying of the material. Inlet and Outlet Channels. — On account of the desirability of keeping the surface of the sewage in the sedimentation cham- bers at a constant elevation, in order to prevent diffusion through the slots, the inlet and outlet channels must be designed carefully. In the case of the radial-flow tanks of Baltimore, Fig. 175, thin longitudinal walls were so placed in the inlet channel that one of the sub-channels does not have to serve more than four tanks. This subdivision was made because it was considered easier to distribute a given flow properly among a few units than among many, and experience had shown elsewhere that four tanks could be served by one channel. The following notes on this feature of the design were written by Frank:' "In subdividing a given channel, the total difference in water level in its length is increased, but this increase can be neutralized by increasing the total width of channel. This was done in the present instance. The whole distributor system has been so computed hydraulically that the elevation of the water surface at the initial point of subdivision is practically the same ' Bng. Bee, July 4, 1914. fHE IMHOFF AND EMSCHER TANK 391 in all the subchannels. This has resulted, of course, in subchannels of different widths. The distributor channels have, in addition, been provided throughout with adjustable distributor wings so that when the plant is taken into operation the whole system may be regulated to the desired nicety." In the case of parallel-flow tanks another factor must be taken into account. Space in the sludge chamber of an ImhofE tank is expensive and it is desirable to utilize it fully. If sewage flows through the tank in one direction, there will be a greater accu- mulation of sludge near the inlet than the outlet, and conse- quently provision has sometimes been made for reversing the flow from time to time. In German practice this reversal was made at least once a month; at Fitchburg, Mass., in 1918, only three reversals were made during the year, but the usual fre- quency is about once each month. The width and depth of the inlet channels must be such that sedimentation will not take place in them. It is advisable that none of the outlet weirs of the tanks be submerged by the effluent in the lower part of the effluent channels during periods of heavy discharge. Scum boards in front of all weir outlets are desirable as in septic tanks. The practice in the Emscher district was to sub- merge them 12 to 1& in. and let them rise 12 in. above the surface. Special attention should be paid in placing them to avoid causing any increase in the currents in the lower part of the sedimenta- tion chamber. Sedimentation Chamber. — The length of the sedimentation chamber, or its radius if a radial-flow tank is used, is determined by the detention period adopted. When that is fixed the length of the tank or the radius, according to the type chosen, will depend upon the rate of flow which is considered proper for the given local conditions. After the length has been adopted, the other dimensions are obtained by selecting a depth and width which will best suit all local conditions. In this connection it should be kept in mind that it is desirable to have the horizontal velocity gradually decrease toward the lower part of the sedimentation chamber, in order that there may be no currents to prevent the setthng soUds from passing through the slot into the sludge chamber. This con- sideration also affects the depth and position of the baffles if used in long Imhoff tanks, for deep baffles will cause a disturbance 392 SEWERAGE AND SEWAGE DISPOSAl in the currents in the parts of the chamber where it is most desira- ble to have steady flow. As it is very desirable for the settling solids to move steadily to and through the slots, the inclined sides of the bottom of a sedimentation chamber must have a sharp inclination and a very hard, smooth surface. Even with great care in construction, it is necessary to clean them from time to time with rubber squeegees, for soft, sticky material accumulates on them which will cause undesirable conditions in the chamber unless it is removed. For the same reason, it is desirable to have the top details of the tank such that this cleaning can be done readily, expeditiously, and thoroughly, either from permanently fixed platforms or temporary plank walks. The slot at the bottom of the sedimentation chamber should be from 6 to 8 in. wide, depending upon the size of the chamber and the character of the sewage. The horizontal overlap to prevent gases passing through the slot should be at least 4 in. and preferably more. Solids which collect on the surface of the sedimentation cham- bers are skimmed off by the attendants and thrown into the gas vents. Gas Vents. — The gas vents should be so designed that work- men can enter the sludge chamber through them when the tank has been emptied. In the Emscher district the vents were from one-fourth to one-third of the area of the tanks, but as no diffi- culty had been experienced there in a few cases with vents amounting to only one-tenth of the tank area, the early American tanks of this type generally had a small ratio of vent to tank area. This has been found to be a mistake. Where the vents are small, the scum forms rapidly, and if the attendants do not break it up as soon as it collects in large quantities, it may rise over the edge of the vents. Imhoff recorded an instance of scum rising 6 ft. above the level of the sewage because the gases in it could not escape. The vents must be easy of access to enable the attendants, if necessary, to break up the scum and liberate the gases in it, which hold up the solids of which it is composed. These solids should be in the sludge chamber undergoing digestion. It is, however, very difficult to break up and force the scum down in many cases, and after this has been accomplished the solids are usually again carried to the surface within a very short time. It THE IMHOFF AND EMSCHER TANK 393 has been found by some engineers that excessive scum forma- tion may be prevented by stirring the sludge with jets of water from perforated pipes provided for the purpose at the bottom of the tanlc, thus liberating entrained gases. Imhoff considered such stirring was also helpful in assisting the decomposition of the organic matter by removing the toxins caused by previous energetic bacterial action. If the scum becomes too voluminous, even after stirring the sludge with jets of water, it may be removed to the sludge beds, although some operators prefer to break it up and push it down into the sludge chamber. At Fitchburg, where the scum is thick and heavy, openings have sometimes been made through it to afford a free outlet for the gas. Foaming has very seldom occurred at this plant. It has been promptly stopped after a small quantity of sludge in the hopper below the vent had been pumped out. Sludge Removal. — The sludge chamber should be designed so that the oldest portion of the sludge can be drawn off first, which necessitates inclined bottoms sloping to sumps. In the Emscher district it was found that a slope of 1 on 2 was practicable if provision was made for flushing the sludge with water if it failed to slip down. The water was admitted throiigh a horizontal 13^- to 2-in. lead pipe at or near the top of the slopes, and occasionally through another ring or single jet at the sump at the bottom of the hopper. The perforations were about ^2 in. in diameter and about 20 in. apart. The sludge is usually removed through a pipe of about 8 in. diameter. In the Emscher district tanlcs this is always carried up straight from the sump until the top is above the surface of the sewage, so that in case it becomes clogged with sludge, which occasionally forms a mass within it, the material can be stirred by a jet of water through a cock tapped into the cap on the end of the pipe. In America a straight pipe, although desirable, has not been regarded as essential. Where a straight pipe is used, the sludge is not discharged from the end but through a curved branch with a valve, which is located, if possible, so that the outlet of the branch is from 4 to 6 ft. below the level of the sewage. This head of sewage is enough to force the sludge, after it is loosened by jets of water from the perforated pipes before mentioned, up through the sludge pipe and its branch to a drain to sludge-drying beds. Where a gravity discharge is impracticable, the sludge can be raised vertically by an air lift 394 SEWERAGE AND SEWAGE DISPOSAL to a sufficient elevation to ensure a gravity flow to the drying beds. In the Emscher district, after the sludge chamber had de- veloped normal operating conditions, a portion of the sludge was drawn off about every 4 weeks, except when winter weather made it undesirable to place sludge on the drying beds. American tanks have been designed as a rule so that no sludge need be removed from them during the winter, provided they are operated properly. Experience shows that aU the sludge should never be withdrawn, for if this is done it is necessary to repeat the tedious and some- times difficult initial ripening process. In drawing off sludge, the rate of withdrawal should be slow in order that the whole mass of sludge may settle and no coni- cal depression exist in the center. In the latter case a consider- able quantity of relatively fresh sludge will be withdrawn and some of the older well-rotted material left, whereas all the sludge which is still undergoing active decomposition should remain in the chamber to keep up, without interruption, the changes taking place there. CHAPTER XIV ACTIVATED SLUDGE The idea from which the activated sludge process for the treatment of sewage has been developed appears to have origi- nated from a series of experiments started at the Lawrence Experiment Station, in 1912, to determine the effect of aeration upon sewage. Sewage had been aerated experimentally and in at least one case as a method of regular treatment,^ before this, but not in the same way. The results of the work at Lawrence, which was continued during subsequent years, were so unusual that a knowledge of them led to other experiments along similar lines at Manchester, England. This process of aerating sewage in the presence of sludge of a special character continues to produce similar sludge, which the English investigators called "activated sludge." The treatment of sewage by aeration in the presence of such sludge is now generally called activated sludge treatment. In plain sedimentation an opportunity is given for a part of the settling solids to subside naturally. An attempt is usually made to keep the sewage in practically the same condition as when it enters the sedimentation tanks, except for this partial clarifica- tion, and therefore the slud ge is rem oved before act ive ana erobic de^OEjposition takes place. In chemical precipitation the removal of solids is greatly increased by the formation of small particles of floc^ of a somewhat gelatinous nature in the sewage, which attract the minute particles of suspended matter and some of the colloids to themselves. This sludge is much more volumi- nous than that from plain sedimentation and is removed before active anaerobic decomposition takes place. In the septic process sedimentation takes place as in the process first men- tioned, but the sludge is allowed to remain in the tanks for a long period of time, during which it is in a state of active anaerobic decomposition which soon makes septic the sewage entering the tank. In the Imhoff tank plain sedimentation goes on in the ' S. R. Loweock, Proc. Inst. C. E., vol. clxiv, Part II, 1906. ' The word "floe" as used herein, signifies a single sponge-like mass or an aggregation of such masses. 395 396 SEWERAGE AND SEWAGE DISPOSAL upper story and septic decomposition in the lower story, the two processes being kept separate by preventing diffusion between the liquids in the two stories and the escape of gases and sludge from the lower to the upper story. In activated sludge treatment a wholly different principle is utiHzed, the removal of suspended matter coupled with the aerobic decomposition of some of the organic matter, and in some cases the subsequent nitrification of the ammoniacal com- pounds. This is similar to the process which nature follows in changing the composition of sewage while it passes through a successfully operated intermittent sand filter. The septic tank treatment is merely an exaggerated and accelerated example of the processes by which Nature changes the composition of sewage discharged into bodies of water relatively so small as not to provide sufficient oxygen to carry on successful aerobic decomposition of the organic matter. Description of Process. — When sewage containing a suitable supply of dissolved oxygen is agitated for a suflJcient length of time, the fine suspended matter and the colloidal substances gather into floe. When the sewage thus coagulated is permitted to stand quiescent, this floe quickly settles and forms at the bottom of the tank a very thin mud or sludge. This sludge is then withdrawn and normally contains about 2 per cent of solids. If such sludge be introduced into raw sewage the rapidity with which the suspended and colloidal matters are coagulated may be greatly increased. The process, as ordinarily carried out, consists of adding sludge to the sewage in proper proportion — roughly, 1 to 5 — introducing sufficient air to provide enough dissolved oxygen to maintain aerobic conditions, and agitating the mixture until practically all of the suspended and colloidal matter has been flocculated or absorbed by the floe introduced into the sewage. The mixture is then conducted to tanks where the floe is removed by sedimentation and the clear supernatant water passes away as effluent. That portion of the sludge which is not required for the treatment of the incoming sewage is diverted and disposed of as described later. Although the idea from which the early development of this process followed appears to have originated in this country, the process was carried through the first stages of development in England and certain patent rights are claimed by some of these ACTIVATED SLUDGE 397 early investigators. Much of the recent development and the perfection of the process has been in this country. Theory of Purification.' — The fisst and perhaps most notice- able function of the process is that of coagulating or flocculating the suspended and colloidal matters in the sewage. This action is similar in effect to the well-known chemical coagulation with sulphate of alumina or sulphate of iron and lime, and the floe resembles the chemical coagulum, particularly the ferric hydrate from the ferrous iron and lime treatment. The floe is a sponge-like mass, or, as expressed by Stein^ "an open-mesh network" which, in the process of formation, may envelop, entrap or entrain colloidal matter and bacteria. The sponge-like structure of the floe offers a very large surface area for contact and this floe appe'ars to be able to absorb colloidal matter, gases and coloring compounds. When the floe is driven about in the liquid it has a sweeping action^ by which the colloidal substances may be said to be swept out of the water, or as stated by Parker* the "process may be regarded as passing a filter through the water in place of passing the water through a filter." Thus far the process appears to be primarily of a physical nature. It has been demonstrated, however,^ that it cannot be carried out under sterile conditions. Just what the action of bacteria and other organisms may be, is not fully understood. One plausible theory is that the bacteria which are contained in the cell-like structure of the floe feed upon the very finely divided matter and thus relieve the floe of its burden of such substances and restore its faculty of absorption to such an extent that, when introduced into the incoming sewage, the floe efiiciently performs its function of absorbing the col- loidal matter, which will again be consumed by the living organ- isms which thus cause its regeneration. It is because of these properties that the sludge has come to be called "activated sludge," a term suggested by Ardern and Lockett. One of the cardinal principles of this method of treatment is that there must be dissolved in the sewage an ample supply of 1 Adapted from article by Harrison P. Eddy in Jour. West. Soc. Engrs. vol. xxvi, p. 259; 1921. « Stein "Water Purification Plants and their Operation," p. 143. s Wm. R. Copeland, Second Annual Report, Milwaulcee Sewage Commission, Dec. 31, 1915, p. 115. < "The Control of Water," p. 556. » Ardern and Lockett, Jour. Soo. Chem. Ind., vol. xxxiii, p. 535, "Oxidation of Sewage without Filters." 398 SEWERAGE AND SEWAGE DISPOSAL oxygen to maintain aerobic conditions. The action of the organ- isms in consuming the colloidal matter is one of digestion or oxidation, sometimes referred to as "moist combustion." Through this action the actual weight of suspended and colloidal matter is reduced, the products of the combustion being carried off in the form of dissolved or gaseous matter. This process, under favorable conditions, may extend to nitrification, by which substantial quantities of nitrates and nitrites are formed. Nitri- fication, however, is not necessary to the maintenance of sludge activity. 40 35 30 r o = 25 z fe 20 a. in t 15 a. 10 5 \ WITHOUT 5LUDSE 1 < 1 \ -n i\ \> 1 \ / 1 1 1 \ / \ \ \ \ / A ^ *f \ \ \ \ / / / \ \ ] \ \ .1 :^ \^alb.ammonia\ 10 12 14- 16 Oaus Fia. 178. — Effect of aeration with and without sludge. The importance of the presence of activated sludge in the aerating tank is shown in Fig. 178 giving the results of parallel experiments made by Bartow and Mohlman of the University of Illinois. In one case sewage was aerated without sludge and in the other with sludge. In both cases nitrification follows the nitrogen cycle, but without sludge the completion of the cycle is a matter of days and the stages of change are distinct, while with activated sludge the cycle is completed in a few hours and the nitrite nitrogen is oxidized at once to nitrate nitrogen. Establishment and Maintenance of Sludge Activity. — In order that the treatment of sewage by this process may proceed prop- erly and as rapidly as possible it is necessary for the sludge to become "activated" with great numbers of aerobic and facuita- ACTIVATED SLUDGE 399 tive bacteria, which are able to bring about the decomposition of the organic matter and the conversion of the nitrogen into nitrites and nitrates. The production of activated sludge from the suspended matter in raw sewage was formerly a tedious operation. The most rapid method developed at Manchester, England, was to fill a tank with sewage, aerate it for 21 hr., let it settle for 3 hr., then draw off the supernatant sewage retaining the sludge, and repeat this process during a period of 12 days. Then the period of aeration was reduced to 9 hr. and the tank was filled twice a day. A few days later the aeration period was reduced so that the tank could be filled three times daily. After 48 fillings requiring 26 days, the accumulation of partially activated sludge amounted to 18.6 per cent of the contents of the tank measured after 2-hr. sedimentation. The sewage in the tank was then replaced with fresh sewage and aeration carried on continuously for 15 days, in order to nitrify the sewage completely. The clear liquid was then replaced by fresh sewage and aerated for three days until nitrification was complete. This process was continued until, at the end of 8 weeks from the beginning of the work, activated sludge was obtained which produced a nitrified effluent from raw sewage after 8-hr. aeration. In the original work at Manchester at least 6 months were needed to obtain satisfactory sludge. The accumulation of an initial supply of well-activated sludge can be completed much more quickly when starting with the sludge from sedimentation tanks receiving effluent from trickling filters, for such sludge is similar in character to activated sludge and can be thoroughly activated more quickly than the sludge which accumulates from plain sedimentation or from the aera- tion of fresh sewage. For example, in starting some of his activated sludge experiments at the Milwaukee testing station, Copeland produced good sludge in this manner in a little more than a week. Under favorable conditions, with daily accretion and periodic withdrawal of the excess, the sludge will retain its activity, and treatment by this process can be continued indefi- nitely. It is necessary, however, not to overwork the sludge, for in that case it will graduaUy lose its activity and become ineffective. Bartow and Mohlman secured a supply of activated sludge from fresh wholly domestic sewage at the University of Illinois experiment station, by drawing off the sewage and adding fresh 400 SEWERAGE AND SEWAGE DISPOSAL sewage at intervals of 24 hr. for 15 days. The sewage was changed every 12 hr. during the next 8 days and then every 6 hr. for 4 days. The difference between this process and the original English method of obtaining activated sludge, is that no attempt is made to aerate each charge of sewage until the ammonia nitro- gen has been removed. Later experience showed that it is best to add the fresh sewage every 6 hr. from the beginning of the run. With the kind of sewage used in the test sufficient sludge could be obtained in this way, in a week, to produce a fair effluent, and at the end of 20 days there was enough sludge to produce in 6 hr. an effluent with less than 1 p. p.m. of ammonia nitrogen. In a city having several independent activated sludge plants, it may be desirable to start up new units that may be built, with activated sludge brought from the older operating plants, in order to reduce the period of time required to produce activated sludge. Care must be taken that none of the floe collects in the channels between tanks, and in other places where it might become septic, for if septic decomposition takes place to a material extent, the proper operation of the activated sludge process may be seriously disturbed owing to, or at least accompanied by, reduction of the dissolved oxygen in the liquid. For the same reason it is im- portant to avoid storing the sludge for long periods of time, unless it is being continuously supplied with air and moderately agitated. Physical Properties of Activated Sludge. — Thoroughly acti- vated sludge floe generally has a golden-brown color and is relatively compact. Under-aerated floe is usually light-brown in color, fluffy and relatively light in weight. Over-aerated floe may be of a muddy-brown color, probably due to unabsorbed colloidal substances, and may be considerably broken up and disintegrated into relatively fine material which settles slowly and leaves a turbid supernatent liquid. Well-activated floe settles more readily than under-aerated floe and forms a denser sludge, leaving the supernatant liquor very clear, xmder favorable conditions. Under-aerated floe is so light that it forms a very thin and bulky sludge and a considerable proportion of it is apt to be carried out of sedimentation tanks. Activated sludge contains a very small proportion of sus- pended solids, as compared with sludges obtained from the treatment of sewage by other processes. The activated sludge, ACTIVATED SLUDGE 401 when well-aerated and accumulated under favorable conditions, generally contains about 2 per cent solids (98 per cent water). Particularly well-activated sludge may contain as much as 3 per cent sohds, or perhaps slightly more, while under-aerated sludge may contain only 0.5 per cent suspended solids, or even less. This variation in density has a very important bearing upon the design of the plant and the handling of the sludge, for a sludge containing 0.5 per cent soUds will have a volume four times as great as that containing 2 per cent solids, as ex- plained in Chapter XV. Hatton reported that sedimentation of the Milwaukee sludge containing 2 per cent solids in a tank 26 ft. deep increased the solids to 6 per cent. In such work it is necessary to keep in mind that if the depth of sedimentation tanks is increased, the time required for the concentration of the sludge from one proportion of moisture to another is also increased. Pratt and Gascoigne found that the rate of increase and time required for concentra- tion was about as the cube root of the depth. Copeland found that sludge acidified with sulphuric acid and heated for 2 hr. was concentrated from approximately 2 to 9 per cent solids. Another property of activated sludge, of importance to the designer, is the manner in which it clings to sloping surfaces. It does not slide readily on slopes much flatter than 60 deg. with the horizontal. This has been observed many times and the deterioration of sludge which has collected on slopes of 45 deg., instead of sliding down them freely to the sludge removal pipes as was expected, has been reported as the cause of inferior results in some work with the activated sludge process. Reactivating Sludge. — If the period of aeration of the sewage, containing the proper proportion of activated sludge, and all other conditions be properly regulated, the sludge can be kept continuously in good active condition. On the other hand, if the sewage be under- or over-aerated, or if the activated sludge be overworked, it will lose a portion or the whole of its activity and must be reactivated in some manner, in order to restore it to a condition in which it can exercise its proper functions efficiently. Such reactivation may be carried out in the mixture of sewage and sludge, or the sludge, after separation from the major portion of the treated sewage in sedimentation tanks, can be removed to separate tanks and there be aerated for a sufficient period of time to restore its normal activity. 26 402 SEWERAGE AND SEWAGE DISPOSAL In some cases it has appeared that the reactivation of the sludge was advantageous as a regular part of the process, per- mitting the use of a smaller quantity of air and volume of return sludge than where the process was so conducted as to maintain the sludge in prime condition throughout. The balance between the work required of the sludge and its activity, in at least one case, was so closely adjusted that the sludge showed a gradual decrease in activity from Monday to Saturday, and a rapid recovery of activity during the period of lighter load from Saturday noon until Monday morning. During the latter time the sludge regained the activity lost during the preceding days of heavy load. In this case the sludge was reactivated in tanks set apart for this purpose, but the general reactivation process during the period of light load was required in addition to the regular re-aeration. In other cases there has appeared to be little if any advantage in re-aeration of the sludge in separate tanks. In tests at Milwaukee, where sludge containing from 2 to 3 per cent of solids was aerated, the air appeared to accumulate in large bubbles and escape from the surface at irregular intervals. This tendency appeared to render the air less effective as a means of agitation, and possibly also to interfere to some extent with the uniform distribution of the dissolved oxygen throughout the sludge. Bacterial Digestion of Sludge. — More or less bacterial digestion of the organic matter of the sludge goes on continuously in aeration and sedimentation tanks. This action is more marked at moderate temperatures when bacterial activity is high, than at the low temperatures of winter. In certain tests made by the authors it was found that aeration can be carried so far that the sludge will become disintegrated and lose its clarifying power. At such times the increase in the proportion of sludge accumulation in the tank becomes very small, and there may even be a decrease. This condition can be remedied by removing a portion of the sludge, by reducing the period of aeration, or both; this will result in the rapid building up of a large quantity of fresh activated sludge and, almost immediately, in a well-clarified effluent. Winter Operation. — Very high or very low temperature of the sewage is detrimental to treatment. The unfavorable effect of low temperature has been repeatedly demonstrated. ACTIVATED SLUDGE 403 In very cold weather at Milwaukee, Copeland found that the free ammonia was not reduced and nitrates were not formed to a material extent, but the removal of organic matter was so great and the low temperature of the sewage permitted the liquid to absorb so much dissolved oxygen that the effluent was stable in the absence of nitrates. It was also found that melted snow from the streets and the oold muddy storm water in the early spring caused a decrease in the size of the floe and in its ability to absorb colloidal matter. This condition could be limited, in part at least, by increasing the air supply about 25 per cent. Cold weather was found by Pratt and Gascoigne to have no appreciable effect on the process at the Cleveland Testing Station, where clarification instead of stability was the operating aim. Lederer has suggested that in cold weather the turbidity of the effluent is a good index of its quality. At such times he found the stability (explained in Chapter IX on "Chemical and Biological Characteristics of Sewage") was 100 per cent with turbidities of 10 p.p.m. or less; it varied between 50 and 100 per cent with a turbidity of 15; and fell off rapidly when the turbidity exceeded 15. Extent of Purification Secured. — An important characteristic of this method of treatment is the high degree of purification which may be accomplished. So large a proportion of the suspended solids and colloids may be removed that under ordinary conditions the effluent will be clear and contain but very little suspended matter and comparatively few bacteria. The reduction in bacteria, however, is accomplished by physically removing them, rather than by destroying them as in the case of disinfection. In many cases color may be removed, probably by action simi- lar to that of mordants. In the case of odors, some may be oxidized directly and some absorbed, the resulting effluent being comparatively free from odors noticeable to the ordinary observer. Oxidation of organic matter in sludge occurs coincidentally with the flocculation, and there may be nitrification where the treatment is fairly complete. In the latter case, the treated sewage may be charged with a substantial quantity of available oxygen in the form of nitrites or nitrates, which will serve as a factor of safety against future putrefaction. An interesting characteristic of this process is its adaptability 404 SEWERAGE AND SEWAGE DISPOSAL to partial treatment. If the removal of colloids and bacteria and the production of a moderate degree of stability be sufficient, the process can be stopped short of material oxidation of organic matter and nitrification. On the other hand, if complete stability and oxidation of a substantial quantity of organic matter are required, the process can be so operated as to accomplish this. Table 48. — Pubification of Sewage Compared with Period op Aeration (T. Chalkley Hatfcon, Engineeririg News, vol. Ixxv, 1916, page 307.) Aeration, hours Cubic feet of air per minute . Cubic feet of air per gallon . . Appearance of settled liquor . Stability, hours Bacteria removed, per cent . . Free ammonia, parts per million Nitrites, parts per million . . Nitrates, parts per million. . Dissolved oxygen, parts per million Turbid 22.00 0.08 0.08 0.00 1 160.00 0.67 Clear 2.00 52.00 17.00 0.00 0.04 0.30 2 160.00 1.32 Clear 33.00 81 + 15.00 0.95 0.70 1.90 3 160.00 1.9? Clear 120 + 92 + 11.00 1.75 2.80 4.30 4 160.00 2.64 Clear 120 + 95 + 7.00 2.20 6.60 5.90 5 160.00 3.31 Clear 120+ 98 + 5.00 2.50 8.20 6.70 The progressive nature of the changes that take place is shown in Table 48, giving the results of a continuous supply of 0.67 cu. ft. of air per minute to sewage which originally had 225 p.p.m. of suspended matter. These tests were made in 1915 at the Milwaukee Testing Station, by Copeland. The data in this table indicate that all that it is necessary to do to obtain an effluent of the desired quality is to select the proper detention period for the operating conditions. This, however, is not as simple as it appears, for these conditions, such as the design of the tanks, the rate of application of the air, the character of the sewage, and the quality of the activated sludge, will vary greatly under different conditions and some of them wiU vary con- tinuously in the same plant. As early as 1917 Hatton determined upon the following stand- ard as a measure and limit of purification to be effected by the activated sludge process at Milwaukee: Reduction of bacteria, per cent 90 Reduction of suspended solids, per cent 95 Stability by methylene blue, hours 72 ACTIVATED SLUDGE 405 Prolonged tests with the strong sewage containing a substan- tial proportion of industrial wastes have demonstrated that it is entirely practicable to meet this standard at Milwaukee. While it may be proper, at the present state of development, to assume that, in general, purification may be carried to almost any extent desired, this assumption must be made with great caution. It is important to consider carefully the character of the sewage or industrial wastes to be treated, for there may be disturbing influences similar to those which appear to have been important at Worcester, where it is stated that it was practically impossible to operate the activated sludge plant so as to obtain, continuously, uniformly good results, and, probably because of the acid iron wastes from the wire mills, that the extent of puri- fication was decidedly limited, notwithstanding the use of very large quantities of air, a goodly period of aeration and the re- aeration of the sludge. Preparatory Treatment of Sewage Required. — In the treat- ment of some sewages, particularly where industrial wastes such as those from tanneries and packing houses are present, it may be necessary, or at least advisable, to subject the sewage to preparatory treatment for the removal of those solids which are likely to form deposits upon the porous plates, to be described later, through which the air is introduced, otherwise the plates may be rendered practically ineffective. It has been found at Milwaukee that suflicient solids are removed by first passing the sewage through grit chambers and fine screens. In some cases it may be found that subjecting the sewage to a brief period of sedimentation may remove sufficient suspended matter to permit subsequent treatment with a very much smaller quantity of air than would otherwise be required. Proportion of Return Sludge. — There is an intimate relation between the period of aeration, the volume of air required, and the proportion of activated sludge introduced into the sewage. The variation in one of these factors may be offset within hmits by proper adjustment of the others. It is important to seciu-e the most effective and economical regulation of these quantities. The activated sludge is the workshop of the bacteria. Here they receive the raw materials of the sewage, work them over producing remarkable bio-chemical changes, and finally dis- charge some of the products of their processes, which pass away in the effluent. But the sludge performs another very important 406 SEWERAGE AND SEWAGE DISPOSAL function, that of serving as a vehicle for the removal, from the water in which they came to the tanks, of much of the unchanged and partially changed matters, including the bacteria themselves. It is necessary to have an ample supply of this sludge, the pro- portion or quantity depending in a measure upon the character of the sewage and the degree of the purification desired. In general, the quantity of sludge (containing 2 per cent sohds), required for the treatment of moderately strong municipal sewage will be equivalent to about 20 per cent of the volume of sewage to be treated; this proportion, measured as it is introduced into the raw sewage, will generally correspond approximately with the volume of sludge remaining in a test sample taken from the aera- tion tank, after standing for }^ hr. At Milwaukee, Copeland found that there was practically no difference in the quality of effluent obtained with 18, 22, 26 and 30 per cent of activated sludge in the aeration tanks, measured in this manner. With very strong sewage, particularly that containing large quantities of colloidal matter, a greater proportion of sludge, even as high as 50 per cent of the volume of the sewage to be treated, may be required for satisfactory results. Pearse found, with a very strong sewage from the packing houses in Chicago, that the quantity of activated sludge in the aerating tanks should be much greater than that required at Milwaukee. The character and condition of the sewage or industrial wastes which necessitate the use of a very large proportion of activated sludge are also hkely to require long periods of aeration, such as ten hours, and large volumes of air, such as from 5 to 10 cu. ft. per gallon of sewage, in order to maintain satisfactory working conditions. The use of an unnecessarily large volume of sludge is likely to be disadvantageous because of its demand for oxygen. The volume of sludge required for successful treatment will depend in large measure upon the density of the sludge. If for any reason the sludge is particularly voluminous as drawn from the sedimentation tanks, say such as to have a density of but 0.5 per cent solids, the volume required to produce satisfactory results will be from two to four times as great as if it contained 2 per cent solids. Period of Aeration. — A substantial length of time is required for the fine suspended and colloidal substances to become coagu- ACTIVATED SLUDGE 407 lated and for the absorption of colloidal matter by the return sludge and the newly formed floe. As indicated by Table 48, the greater the degree of oxidation of the organic matter required, the longer must be the period of aeration. The character of the sewage has a marked influence upon the required period of contact. A dilute fresh domestic sewage may become well coagulated in from 3 or 4 hr., whereas a strong .sewage containing industrial wastes may require 6 or even 8 hr., and it may be necessary to aerate for 8 or 10 hr. or even longer industrial wastes high in colloidal matter, to secure the same relative extent of purification. For American municipal sewages the economic period of aeration appears to lie between 4 and 8 hr. Quantity of Air Required. — Air may be utilized for two pur- poses, for maintaining aerobic biological conditions and for agitation. Agitation, however, may be accomplished by purely mechanical means, in which case the introduction of air would be necessary simply to maintain suflBcient dissolved oxygen for aerobic conditions. In some cases the supply of oxygen has been derived from contact with the open air and no artificial supply has been required. The quantity of oxygen necessary for the maintenance of aerobic conditions is relatively small. From a comparison of analyses of the air entering and the gases escaping from aeration tanks, Crawford and Bartow estimated that only about 5 per cent of the oxygen of the air introduced during the aeration process was utilized in maintaining aerobic conditions. In the case of strong tannery wastes the authors found that only 10 per cent of the oxygen in the air introduced disappeared during the process. Experimenting to determine the actual amount of oxygen necessary to stabilize sewage of "average strength," Clark found the average requirement to be 42.6 p.p.m., equivalent to 350 lb. of oxygen per 1,000,000 gal. This is somewhat less than 1 per cent of the quantity which would be used in the aeration process if the rate were 2 cu. ft. of air per gallon of sewage treated, or about 2 per cent in case 1 cu. ft. of air were sufficient for sewage of this character. Where agitation is effected by blowing air into the sewage, the quantity of air required is very large. It is necessary to provide sufficient agitation to prevent the heavy suspended solids 408 SEWERAGE AND SEWAGE DISPOSAL from settling down on to the air diffuser plates and thus clogging them. As already pointed out, removal of suspended matter by fine screening may be a substantial aid in preventing such clogging. Even sewage which has passed a fine screen having openings }i by 2 in. may contain sufficient solids of this character to cause clogging of the plates, if the air supply be permitted to fall much below 1 cu. ft. per minute per square foot of plate area. Finer screening might make it possible to reduce this quantity slightly. The volume of air required in any given case will depend mainly upon the following general conditions, previously discussed: 1. Quantity of sewage to be treated, 2. Quality of effluent desired, 3. Period of aeration, 4. Proportion of sludge to be returned to the sewage, 5. Temperature of aeration. As a general statement, where the aeration is effected by blow- ing air into the sewage, passing continuously through tanks from 10 to 15 ft. deep, with the area of diffuser plates in the proportion of 1:6 of tank area, and a period of aeration of about 6 hr., the quantity of air may be about as follows: Cubic Feet of Free Aih Chahacteb of Sewage Peh Gallon of Sewage 1. For rather weak domestic sewage 1.0 2. For strong municipal sewage containing some industrial wastes not particularly detrimental to bacterial life .... 1.5 3. For weak municipal sewage containing considerable indus- trial wastes some of which are detrimental to bacterial life, as for example acid-iron wastes 2.0 4. For strong municipal sewage containing considerable industrial wastes some of which are detrimental to bac- terial life, as for example, acid-iron wastes 4.0 5. For municipal sewage containing sufficient industrial wastes, not specifically inhibitive of bacterial life, to de- cidedly influence the composition of the sewage, e.g., packinghouse, tannery, etc 3.0 6. For strong packinghouse wastes 4.0 7. For strong tannery wastes containing a very large pro- portion of colloids, but settled before aeration 6.0 As the cost incident to the use of a large quantity of air is burdensome in operating expense, as well as in the fixed charges ACTIVATED SLUDGE 409 (interest and depreciation) upon the installation, it is necessary to reduce it to the minimum practical limit. Some English experiments indicate that 0.5 cu. ft. of air per gallon will give excellent purification, while at Milwaukee 1.5 cu. ft. is as small a quantity as gives assurance of satisfactory results. This difference may be due in part to a difference in the character of sewages, and in part to the fact that the English tests were made on a small laboratory scale while some of those at Milwaukee were made on a relatively large scale. Many experiments have been performed with a view to discovering means for reducing the quantity of air required. In some cases, instead of agitation by blowing air into the sewage, the sewage has been pumped over and over, and only a small quantity of air introduced. In other cases the sewage has been agitated for a short period, allowed to stand for a time during which the biologi- cal process was supported by the accumulated dissolved oxygen in the sewage, this period being followed by another short period during which the sewage was aerated. By continuing this inter- mittent aeration, it is stated that satisfactory results have been obtained with a relatively small volume of air. From experiments with paddle agitators at Sheffield, England, Haworth concluded that given suitable agitation to obtain thorough intermixing of the activated sludge with the sewage and to prevent settlement, the necessary air required to maintain the biological activity might be obtained by surface contact. To accomplish this it was found necessary to produce constant change of surface of the liquid. Consequently agitators con- sisting of vertical pistons of the Root blower type. Fig. 181, were installed at the ends of relatively long channels. By rotating the agitators the sewage was circulated at a velocity sufficient to prevent settlement of the sludge. A velocity of IJ^ to 2 ft. per second was found to be ample to accomplish this result. The power expended by this process was calculated to be about 20 hp. per million gallons, but there appeared to be no doubt that the agitators employed could handle a much larger volume with only a slight additional power consumption, thus reducing the power consumption per million gallons. Measurement of Sludge. — The volume and character of sludge is such an important matter in this process of treatment that its measurement is very essential. The volume of sludge con- tained in the aeration tank liquor may be determined by allowing 410 SEWERAGE AND SEWAGE DISPOSAL a measured portion of the liquor to settle for a definite period of time and then reading the volume of the settled sludge. If comparisons are to be made of the quantities of sludge produced and used, it is necessary that measurements be made under reasonably comparable conditions. One of the most convenient means of measurement is a 1,000-c.c. laboratory glass measuring cylinder about 23^ by 16 in. inside dimensions, graduated in 10 c.c. units, from which the percentage of sludge can be read directly. For determining and regulating the proportion of activated sludge present in the aeration tanks and entering the sedimentation tanks, it is probable that a period of sedimentation in the measuring glass of 3^ hr. will generally prove most satisfac- tory, particularly as the results of tests can be obtained with relative promptness. Tests for a shorter period are subject to too great error and variation. Iso 80 in I 60 s ^° g.40 10 of 20 ^ 30 2 10^ 1 \ 50 ;§ \ k. 60 t3 70 J 80 "^ :::^ — ■^ — ' — 90 5 '/ 2 Ho urs , 6 ! f £ iec (irr en > t-a-t ior 2 ^2 ■7 Fio. 179. — Rate of sedimentation (Milwaukee). It has been found at Milwaukee that a test of this kind, made at the same hour each morning, serves as a valuable means of regulating and controlling the process, the adjustment of the sludge being accomplished by fixing the rate of drawing off sludge for disposal, according to the indications of the test. An occasional series of tests for several different sedimentation periods may furnish valuable information and enable the operator to compare his operating conditions with those of other places where different test periods are adopted. The rate of sedimenta- tion, or the volume of sludge after different periods of test at Milwaukee, is shown by Fig. 179. ACTIVATED SLUDGE 411 Importance of Measurements of Sewage, Sludge and Air. — In order that the plant may be most advantageously operated, it is important that convenient means be provided for obtaining accurate measurements of the rate of sewage flow, the volume of sludge returned to the sewage, the volume of sludge drawn off to be disposed of, and the volume of air used in the aeration process. For these purposes the Venturi meter with the register- indicator-recorder equipment has proved satisfactory. Such measurements should serve as a guide in proportioning the return sludge and air to the volume of sewage being treated. The results of all such measurements should be entered in the records. Aeration Tanks. — The earlier experiments were performed with fill-and-draw tanks which permitted a uniform period of detention and practically uniform contact of the sludge floe and air with all portions of the sewage. It was subsequently found, however, that the continuous flow method of operation was not only entirely practicable but much to be preferred. While, for practical purposes, it is customary to speak of sewage being aerated in a continuous-flow tank for a period of time corresponding to the assumed detention period, as a matter of fact some of it will be aerated a much shorter time and some a much longer time. In such a tank sewage does not flow forward steadily and uniformly from inlet to outlet. The experiments by Clark at the Lawrence Experiment Station in 1917 showed that in a tank 8 ft. deep above its conical bottom, operated with a detention period of 10.5 hr., with air applied at the rate of 0.15 cu. ft. per gallon per hour, some of the sewage passed through the tank in 1 hr., half passed through in 7}4 hr., and all of it did not pass out until the 35th hour. The tests were made by dosing the entering sewage with salt or ammonium chloride and determining the proportion of these substances which the effluent contained during the period of experiment. In experiments with horizontal-flow tanks at Cleveland, Pratt and Gascoigne found that instead of securing a displacement of 100 per cent of the content of the tank only 40 per cent was obtained, until under-and-over vertical bafiles were installed, and thereafter only 75 per cent. Thus the detention period under the two conditions actually averaged only 40 and 75 per cent, respectively, of its computed length. The depth of aeration tanks will depend in part upon local conditions, although it is important to remember that the amount 412 SEWERAGE AND SEWAGE DISPOSAL of power required for compressing the air increases with the increase in depth of liquor into which it is introduced, as dis- cussed later. At Milwaukee, where a depth of 15 ft. was adopted for the tanks, the controlling consideration was the imprac- ticability of obtaining sufficient land for shallower tanks. Such tanks are usually rectangular in shape and from 6 to 15 ft. in depth. One tjrpe of rectangular tank is illustrated by Fig. 180. Where air is introduced through porous plates in the bottom of the tank, the plates are generally arranged crosswise of the ACTIVATED SLUDGE 413 jlow of sewage, thus causing the sewage to be subjected to suc- cessive impacts of air, sometimes referred to as "air baffles." Theoretically there appears to be advantage in securing as great a number of such impacts as possible, which favors the use of long tanks. The length of the tank is also influenced by the desirable width. The desirable width of tank may be influenced by the arrange- ment of the air piping and diffuser plates. Thus far it has seemed wise to limit the number of diffuser plates to from eight to ten 12-in. square porous plates per container to be fed by a single air line. By installing the air main along the center Hne of the tank and supplying plate containers on either side, a tank width of about 20 ft. results. Much study has been given to the shape of aeration tank floors. The most common is the alternate ridge and valley type, some- times called the "sawtooth" type. The porous plates are placed in the valleys. The ridges between the plates are intended to deflect the settling solids to a position over the plates, whence they will be again carried up by the upward currents induced by the air escaping from the plates. The aeration tank dimensions are controlled by the period of aeration selected as most economical and suitable for the extent of purification required, under the conditions of the particular problem being considered. At Milwaukee, the aeration period is to be 6 hr. at the average rate of flow, resulting in tanks about 22 ft. wide, 237 ft. long and 15 ft. deep. The volume of the tanks below the level of the surface of the sewage is determined arbitrarily by dividing the average 24-hr. flow of sewage by four, and providing a tank capacity equivalent to this volume. The most suitable proportion of the area of air diffusers to the area of the tank appears to lie between 1 to 4 and 1 to 6, although smaller ratios have been adopted in some cases. Where perforated pipe diffusers are used, the type of tank bottom may be similar to that adopted in the case of the porous plate diffuser. Air Diffusers. — Much study has been given to the subject of air diffusion, with a view to securing that type of diffuser which shall deliver the air to the sewage in a state of very fine division and which shall not become clogged, and which shall not offer excessive frictional resistance to the passage of the air. 414 SEWERAGE AND SEWAGE DISPOSAL The simplest method is to dehver the air to the sewage in the tank through holes in pipes laid in the bottom of the aerating tank. There has been some difficulty with this system, however, because of clogging of the orifices with rust, and with sludge after the air supply has been shut off. Monel-metal cloth has been used experimentally for diffusing air and is apparently efficient, giving small, well-distributed bubbles; but the cloth becomes clogged in continuous operation, causing a great increase in frictional resistance and for this reason has not been used practically. In a number of English installations, the diffusers are flat porous tile laid as covers of concrete troughs containing per- forated pipes through which the compressed air is conveyed to the space below the porous tile. In the course of the investigations at Milwaukee, tests were made of diffusers of basswood 3^ in. thick and from 2}^ to 4 in. square. They were developed by Nordell, who designed a concrete container for holding them in place and delivering com- pressed air to their lower surface. They gave good distribution of air, but were found to have objectionable features which will be mentioned later. Satisfactory results have been obtained with diffusers made of filtros plates. These are usually 12 in. square and about IJ^ in. thick, made of ground quartz sand and a silicious binding material baked until hard. There are several grades of plates having the same degree of porosity but varying in frictional resistance. Tests made at Milwaukee of one of these grades showed that the initial loss of pressure for passing 2 cu. ft. of air per minute per square foot, under 5 lb. air pressure, was % lb. and for every additional cubic foot of air there was a further loss of J^ lb. Attempts have been made to set filtros plates in cast-iron holders but there is danger that they will become choked with iron rust if this method of support is used. In 2}/^ years service in aeration tanks at Houston, Tex., rust from such frames reduced the capacity of filtros plates from 7.1 cu. ft. of air per minute under a pressure of 2 in. of water to 0.7 cu. ft. The plates were set in frames with sulphur joints, and about one-tenth of them were broken in removal. Those that were uninjured were immersed in hot hydrochloric acid and after being cleaned in this way were able to pass 6.2 cu. ft. of air per minute. ACTIVATED SLUDGE 415 The best setting for plates seems to be portland cement mortar. Of a number of grades of bituminous mortar used by the authors for this purpose none has been found to be satisfac- tory in service. In a series of tests by Bartow and Mickle at the University of Ilhnois in 1918, with perforated pipes, wood blocks and coarse and fine filtros plates, it was concluded that the filtros plates were most efficient, giving sludge that settled readily, leaving clear, stable eflBuents; there was little difference in the effluents from the tanks with fine and coarse filtros plates. The wood block diffusers were difficult to install, deteriorated rapidly, and gave unstable effluents in a majority of samples, and these effluents were inferior chemically to those from filtros plate aeration. The effluents from tanks with perforated pipes were the least satisfactory. Filtros plates become clogged if the compressed air contains oil and dust. For this reason, it is important to provide clean air and to have the air compressed by a mechanism that does not aUow oil to enter the air even in very minute quantities. This trouble was experienced at Milwaukee in 1916. As a result of the clogging of the plates, the air was unevenly distributed and sand, pieces of water-logged wood and other coarse material gathered over the spots where the air was shut off, and finally covered parts of the plates through which a decreased amount of air rose. Through the plates which were porous, the air streamed in excessive volume, forming large bubbles, less effective than those }i in. or less in diameter. In later experiments at Mil- waukee, the air was filtered through 10- or 12-oz. duck, then through canton flannel and was finally washed by the hydro- turbine blower used in compressing it. The air is best supplied to the diffusers through pipes protected against rust, for if rust forms in these pipes the plates are liable to become clogged by it. After the activated sludge has been introduced into the raw sewage, it is important to prevent it from settling in channels leading to aeration tanks and from aeration tanks to sedimenta- tion tanks. In order to avoid loss of head within the plant, it may be advantageous in some cases to construct such channels with relatively large dimensions, either level or on very slight slopes. The velocity of flow will then be low. In such cases it may be advantageous to provide for aerating the mixture of 416 SEWERAGE AND SEWAGE DISPOSAL sewage and activated sludge while flowing through the channels, by equipping them with air diffusers in a similar manner to that adopted for the aeration tanks. Power Required for Compressing Air. — The pressure at which the air must be supplied depends chiefly upon the depth of the sewage in the aeration tanks, but is influenced by the frictional resistance of the distribution system, and particularly that of the diffusers in case porous plates are used for this purpose. Practical limitations will be such in most cases as to require an air pressure of between 2.5 and 10 lb. per square inch. In selecting the type of blower, there should be borne in mind the probability that impurities in the air will gradually accumulate on the surface or in the pores of plate diffusers, resulting in partial clogging and a gradual increase in the frictional resistance. The initial operating pressure may have to be increased consider- ably from time to time on this account, and the blower should be selected with this in view. It is also important to provide a flexible blower plant capable of adjustment in air supply to the volume and quality of sewage and other conditions subject to frequent change. Variation in volume of air can be secured by throwing in or out of service one or more compressors. Greater fle.xibility, however, can be secured by the use of centrifugal compressors. Fig. 181, with which the volume of air furnished can be varied by opening or closing the blast gate in the suc- tion pipe. The power required to drive this type of compressor varies approximately with the volume of air used, when operat- ing at a constant speed. The possibility of varying the volume of air compressed, without wasting power, is one of the advantages of this type of machine. On the other hand, this type of compressor furnishes air at a uniform pressure, which can only be changed by changing the speed of the impeller, which with certain types of driving apparatus is not practicable. In such cases, the pressure must be high enough to meet all emergencies. Therefore at all other times there is a waste of power in compress- ing the air to a higher pressure than is necessary. At a given speed a practically constant volume of air is fur- nished by positive pressure blowers and rotary blowers alike, Fig. 181, the pressure depending upon the resistance offered, as by the head of water against which the air is introduced into the aeration tanks, and by the friction in pipe lines and diffuser plates. One of the advantages of this type of blower is that it will build ACTIVATED SLUDGE 417 up pressure sufficient to overcome increased frictional resistance in the filtros plates and that it is not necessary to waste power Posi-|-ive Pressure Blower (Connersvi lie and Root Type) Centrifugal Compressor (General Elec+nc Company) WATER OUTLET (S^ INLET PORT'' "OUTLET PORT ■WATER Hytor Compressor (Nash) Fig. 181. — Types of air compressora. by raising the pressure unnecessarily high, as in the case of the centrifugal blower. On the other hand, if great flexibility be required in the volume of air, there may be a waste of power due 27 418 SEWERAGE AND SEWAGE DISPOSAL to compressing an unnecessarily large volume. Probably in most activated sludge plants suificient flexibility can be obtained by providing a proper number of units of appropriate size and varying the number of units in use. The amount of power required and its cost can be estimated from data in Table 49, by applying the assumed efficiency and other data peculiar to the case under consideration. The efiiciency which may be expected from a high-grade electric-motor-driven blower plant will vary according to the type and size of machine^ but in general may be assumed to be from 60 to 70 per cent. Table 49. — Power Required fob Compbessing Aib Final pressure of air (lb. per sq. in.) Theoretical work to compress mil. cu. ft. of free air (horsepower-hours) Theoretical power to compress 100 cu. ft. free air per minute (horsepowers) Theoretical cost of compressing 1 mil. cu. ft. free air with elec. power at 1 ct. per kw.-hr., dollara 1 72.3 0.43 0.539 2 144.0 0.86 1.074 3 200.8 1.20 1.498 4 265.2 1.59 1.978 5 325.8 1.95 2.430 6 384.7 2.31 2.869 7 442.5 2.66 3.301 8 490.2 2.94 3.656 9 543.4 3.26 4.053 10 596.5 3.58 4.449 12 697.1 4.18 5.199 14 785.4 4.71 5.858 16 875.9 5.26 6.533 Note. — This table is based on the assumption that the air is compressed under adiabatio conditions (without cooling) as is the practice in nearly all blower work, from atmospheric pressure (14.7 lb. per square inch) and an initial temperature of GO"!?. Weight of 1 cu. ft. free air = 0.0764 lb. Slightly lower power consumption may be obtained with good water- jacketed reciprocating compressors. A convenient approximate- unit of power to bear in mind is that 30 hp. per hour will be required for furnishing 1.5 cu. ft. of free air per gallon of sewage for the treatment of 1,000,000 gal. of ordinary municipal sewage in 24 hr., in tanks 10 ft. deep Sedimentation Tanks. — The mixed liquor^ from the aeration tanks, containing from 0.2 to 0.6 per cent solids, is run into settling tanks for clarification. The extent of treatment in the aeration tanks, therefore, has a very important effect upon the > Aeration tank effluent; aerated sewage including floo. ACTIVATED SLUDGE 419 process of sedimentation. It has been proved in spme cases that aeration sufficient to produce a satisfactory efHuent, if the floe could be removed, was insufficient to produce a floe which would readily settle in sedimentation tanks, or a reasonably dense sludge. In these cases, therefore, the extent of aeration may be controlled by the requirements of the sedimentation process or by the required density of the sludge. In some of the earlier experimental work, sedimentation was secured in the same tank in which the aeration process was conducted, that is, the tank was filled with sewage and activated sludge, the mixture aerated for a sufficient length of time for the formation of floe, and thereafter the mixed liquor was allowed to stand until a clear supernatant effluent could be drawn off and the excess sludge removed. The process was then repeated. This process is too complicated for practical use in a large plant, and has been succeeded by aeration and sedimentation in separate continuous-flow tanks. Such sedimentation tanks may be constructed either according to the vertical flow or the horizontal flow principle; Pearse used the former successfully at the Paclcing- town (Chicago) experimental plant, and the latter has been used at Houston, Milwaukee and other places. Copeland found that in order to secure complete removal of the floe the velocity of flow of liquor through the sedimentation tank must not be allowed to exceed 3 ft. per minute. This velocity should be computed upon the approximate cross-section of flow to which the stream of liquor may be expected to be confined. As the area of the sedimentation tank with respect to the volume of mixed liquor which it can successfully handle is an important consideration, it has been found convenient to use the unit of horizontal area in computations relating to sedimentation tanks. As a result of experiments at Milwaukee, the proposed sedimentation tanks have been designed to provide 1 sq. ft. of tank area for each 1,600 gal. of sewage to be treated, measured at the time of assumed maximum flow of sewage, such rate being expected to occur only in times of storm. This is equivalent to 1 sq. ft. of tank for each 850 gal. more or less, based upon the average daily flow. While sedimentation can be efficiently carried out in very shallow tanks, it is important that the tanks should have a sufficient depth to provide for temporary storage for the volume 420 SEWERAGE AND SEWAGE DISPOSAL of sludge produced during peak flows, or at other times when sludge is produced at an unusually high rate or when the sludge is particularly voluminous as a result of under-aeration. Sedi- mentation tanks may be either rectangular or round, and may be provided with hopper bottoms or with bottoms sloping slightly toward the center, provided the tanks are equipped with mechan- ical squeegees. In the case of hopper-bottom tanks, the slope of the bottom should be at least 60 deg. with the horizontal, in order to assure the sliding along it to the outlet pipe situated at the apex of the hopper, of that portion of the accumu- lated sludge which comes in contact with the floor. The expense and difficulty of providing hoppers with such steep slopes for very large tanks has led to the adoption, in some cases, of tanks in which the floors have only a very slight slope toward the center, as, for example, % in. per foot. Such tanks, equipped with mechanical squeegees or plows for scraping toward the center of the tank any solids which may actually reach the floor, should be round at the bottom. They may be relatively large, those proposed for Milwaukee being about 98 ft. in diameter. Most of the sludge in such tanks flows toward the center without coming into contact with the bottom of the tank. The plows are necessary only to push along that portion which does lodge upon the bottom. Volume and Character of Sludge Produced. — The first stage of successful sewage aeration with activated sludge produces a clear effluent after sedimentation, and the quantity of sludge yielded by this stage of the treatment is probably as great as, or greater than, that yielded by carrying the treatment along to the production of a thoroughly nitrified efiiuent. This quantity depends primarily upon the quantity of suspended and colloidal matter in the sewage, and as there is a great variation in the content of such matter there is inevitably a great variation in the quantity of sludge produced, particularly when the sewage contains certain classes of industrial wastes, such as iron salts and organic matter from packing houses and tanneries. Experience indicates that a short period of aeration with good activated sludge will reduce the suspended matter carried by the influent about 90 per cent. A portion of the suspended matter thus removed is converted into gaseous and dissolved substances by digestion, the remainder forming sludge. The investigations of Pratt and Gascoigne, at the Cleveland Testing Station, appear ACTIVATED SLUDGE 421 to indicate that sometimes as much as 50 per cent of the suspended matter might be digested, the average being about 35 per cent. Bartow and Mohlman found that the hquefaction was about 4.5 per cent in their experiments at the University of Ilhnois. This change in the suspended matter makes it difficult to estimate the probable quantity of such matter in the sludge produced by this treatment. The tests of Bartow and Mohlman indicate that with strictly domestic sewage there will be from 750 to 1,200 lb. of dry sludge per million gallons of sewage treated. ^ At Milwaukee, with sewage containing nearly 300 p.p.m. of ' suspended solids, the quantity of dry solids in the sludge may be as high as 2,000 to 2,500 lb. per million gallons. If industrial wastes and storm water are added to domestic sewage, corresponding increases in suspended solids must be provided for. The volume of activated sludge depends upon its density, which may be expected to vary according to many conditions from 0.5 to 3.0 per cent solids (99.5 to 97 per cent water). In many cases, for economical handling and dewatering, it is probable that the aeration and sedimentation processes should be so controlled as to provide a sludge having 2 per cent or more of solids. The sludge drawn from the sedimentation tanks consists of the activated sludge introduced into the raw sewage plus that formed from the solids in the sewage thus treated. The relation of the volume of sludge to the volume of sewage treated, and the volume of waste sludge to that returned to the sewage, are illustrated by the following assumed proportions, based upon a sludge containing 2 per cent solids. Volume of sewage treated 1,000,000 gal. Volume of sludge returned to sewage, 20 per cent of the sewage 200,000 gal. Excess sludge, 12 per cent of the sewage 120,000 gal. Total sludge from sedimentation tanks, 32 per cent of the sewage • 320,000 gal. Return sludge = (320^00) " ^^'^ ^^^ °^°* °^ *°*'^^ sludge. Excess sludge = ( 320000 ) " ^'^'^ ^^"^ ^^"^^ °^ *'°**^ ^^^'^^^^ » The quantity per million gallons will vary greatly, according as the sewage is strong or weak. 422 SEWERAGE AND SEWAGE DISPOSAL There has been a tendency to underestimate the volume of sludge produced by this treatment. The early investigation at Milwaukee indicated that 3,000 to 6,000 gal. of excess sludge with 2 per cent solids would be produced per miUion gallons of sewage treated, in addition to the quantity returned to the acti- vating tanks. Later investigation showed that 15,000 gal. might be produced and the treatment works were finally designed on the assumption of 15,000 gal. of excess sludge containing 2 per cent solids. Sludge from the sedimentation tanks contains so much water that it will flow readily in pipes. In fact in computing the size of pipe Knes and conduits to transport sludge, but slightly lower coefficients of discharge are used than for water. Where sludge is aerated during its flow through conduits, an additional allowance may be advisable on account of interference with flow caused by cross-currents. In some computations the authors have used a coefficient as low as C = 80 in the Hazen and Williams formula, to allow for both of these influences. The velocity in sludge pipes and conduits should be not less than 2 ft. per second, if deposition of solids is to be avoided, which is essential. Where the velocity must be less than this, air should be utiKzed to secure agitation and to maintain the sludge activity. Activated sludge may be raised by centrifugal and other types of pumps, by the air lift, and by the compressed air ejector. It is important to avoid violent agitation which will break up the floe and reconvert it into colloidal condition. There is no serious danger of this in slow or moderate speed centrifugal pumps of suitable design, and even less in the air hft and the ejector. Where the ejector is used, as may be desirable in sludge pressing plants, provision should be made to prevent as far as possible forming a layer of sludge on the bottom and a stratum of com- paratively clear supernatant liquor. Disposal of Sludge. — Three methods of disposing of activated sludge have been studied, as follows : 1. Drying on sand beds and using the dried sludge for filling waste land; 2. Pumping the sludge into lagoons in which eventaaUy it will dry out and become suitable for raising crops, either in situ or by removal to appro- priate areas; 3. Conversion into commercial fertilizer. At Worcester Lanphear dried the sludge from the experimental plant upon sand beds, but found that this method would be ACTIVATED SLUDGE 423 impracticable on a large scale because of the small depth of sludge, 4 or 5 in., which could be applied to the beds at one time, the usual dose being 2.0 to 2.5 gal. per square foot. If more than this depth was applied, the sludge settled rapidly and water remained on top, which had to evaporate before actual drying of the sludge commenced. The dose had to be of such size that most of the water would drain away during application to the bed. In good weather the partially dried sludge began to crack in about 24 hr., and it was possible to remove the sludge from the bed in about 48 hr. from the time when it was applied. At Houston, Tex., sludge from two plants has been pumped into lagoons for a considerable period of time. When these lagoons were examined by the authors, during hot weather, no offensive odor was observed. There was, however, the charac- teristic odor of algae, which appeared to be growing in considerable quantities. The sxipernatant water was clear. Many gas bubbles were rising to the surface, carrying with them some sludge which, however, promptly resubsided with the escape of the gas. There was very little floating sludge upon the lagoons. Activated sludge usually contains from two to four times as much nitrogen as sludge from other processes of sewage treat- ment. The sources of this increase in nitrogen content are not clearly defined. Clark showed in 1916 that the colloidal matter in the Lawrence sewage contained from two to three times as much organic matter as the coarser suspended matter, and that the increase in nitrogen in the sludge, due to these colloids, is about 33 per cent, but this accounts for only about three-fourths of the total increase. It has been suggested that part of this unexplained increase may be due to the fixation of nitrogen, although Clark showed, by experiments in 1917, that part was 'apparently due to the more rapid oxidation of the carbonaceous matter than of the nitrogenous matter, which thus increased the proportion of the latter. The proportion of nitrogen in some activated sludges has been sufficiently high to justify the hope that its conversion into commercial fertilizer may be economically practicable. Preparation of Sludge for Market. — Preparation of sludge for use as a commercial fertilizer involves two major processes, dewatering and drying, aside from such minor treatments as grinding, bolting and bagging. 424 SEWERAGE AND SEWAGE DISPOSAL The most difficult step in the whole activated sludge process, including manufacture of fertilizer, has been the dewatering of the sludge. It has been extensively studied at Milwaukee, Packingtown (Chicago), Houston and Worcester. It has been demonstrated repeatedly that well-activated sludge can be dewatered more advantageously than under-aerated sludge, or stale septic sludge. It is therefore necessary to operate the plant, insofar as practicable, to avoid under-aeration, prolonged storage which tends to cause septic action, and, probably also, over- aeration. As previously stated, it is probable that the aeration process must be controlled and Hmited with a view to producing the sludge in the best condition for dewatering, unless a method be found for conditioning a poor sludge so that it may be suc- cessfully dewatered. For dewatering activated sludge, machines of three types are available, the plate or standard filter press, the squeeze press commonly known as the Worthington or Berrigan press, and the Besco-ter-Meer centrifuge which is a modification of the older Shaefer-ter-Meer centrifuge. These machines are described in Chapter XV. Experiments upon a relatively large scale with each of these machines have been carried out at Milwaukee. From investigations there and elsewhere, it appears that activated sludge is much more difficult to dewater than sludge from chemical precipitation, which has been dewatered for many yesLTS by means of filter presses. It has been found in the plate press and in the squeeze press that the sludge is much more easily dewatered under summer conditions when the biochemical agencies are probably most active in the aeration tanks, than at other seasons of the year, particularly the winter and spring. This fact suggests that the sludge may be conditioned by subjecting it to biological action at summer temperature. The treatment of the liquid sludge with a small proportion of sulphuric acid improves its physical condition in such a manner as to greatly facilitate dewatering by pressing. This treatment also fixes some of the nitrogen compounds so that they are not driven off during the subsequent process of drying. The corresponding increase in nitrogen in the dried sludge in some cases has been sufficient to offset the cost of the acid treatment. For a sludge similar to that produced at Milwaukee, amounting to 1.5 per cent of the volume of sewage treated or 15,000 gal. per 1,000,000 gal. of sewage, the quantity of sulphuric acid ACTIVATED SLUDGE 425 required will be about 2.5 gal. per 1,000 gal., of sludge, equiva- lent to 270 lb. of commercial acid per 1,000,000 gal. of sewage. At 1 ct. per pound, the cost of acid will be about $2.70 per 1,000,000 gal. of sewage treated. The nitrogen or ammonia, ^ the latter term being generally used in the trade, will vary considerably in different sludges. At Worcester, the ammonia content is very low, probably because of the comparatively high proportion of mineral matter in the sewage due to the iron wastes. In dried acidified press cake from municipal sewage not affected in this way by inorganic material, it may be assumed very roughly at from 5 to 6.5 per cent of the dry weight of the sludge. Sludge drying in revolving machine dryers is similar to the dry- ing of many other materials. The process has been thoroughly developed in the arts and there does not seem to be any unusual problem peculiar to the drying of activated sludge. There is, of course, the danger of burning the sludge, of creating objection- able odors and of the escape of considerable dust. The dryers should be designed with a view to reducing these dangers as much as possible and should be carefully operated with the same end in view. If the gases from the dryers are passed through suitable dust chambers and are thoroughly washed, it is probable that dust can be entirely eliminated and there does not seem to be danger of dissemination of disagreeable odors. The gases, however, should preferably be discharged through a relatively high stack. Management. — One of the most serious inherent dangers in the application of this process lies in its management. The process is complicated by a large number of steps, each more or less depend- ent upon the preceding ones. Skillful, conscientious, technical management will be required to keep all features operating to best advantage with the marked variations in volume, character and temperature of the sewage and the multitude of ordinary operating difficulties. At this time (1921), it remains to be demonstrated whether such a process can be successfully operated as a commercial project under municipal conditions generally prevailing in the United States. To assure success, it will be necessary to enter the market and dispose of the commodity produced as advantageously as the law of supply and demand will permit. Prices are subject to radical fluctuations but the ^Ammonia = ^K* or 1.215 nitrogen. 426 SEWERAGE AND SEWAGE DISPOSAL costs under municipal control will probably not fluctuate in a commensurate manner. The one thing that is assured from the outset is the very substantial cost, but it may be hoped that in some cases enough revenue will be derived to so reduce operat- ing expenses as to prevent them from becoming an unreasonable burden. The process is in its infancy and marked improvements are to be expected which will greatly reduce the cost. The advantages over some other processes are so marked and there is such promise of improvements that further investigation is warranted. CHAPTER XV SLUDGE The term "sl udge" is gene rallyjiised^o. designate the deposit which accumulates during o.ne_^of_the_tank treatments of sewage and also the deposit from the final sedimentation, of the effluent from triclding filters. Grit-chamber sediment and screenings are not usually classed as sludge, but when no preliminary separation of these solids is made the sludge contains both these elements. When treating similar sewages under similar conditions the volumes of sludge produced by the different processes are in the following order of magnitude : activated sludge process, chemical precipitation and sedimentation in plain, septic and Imhoff tanks. Volume of Sludge. — The volume of sludge depends mainly on the percentage of water it contains, and it is sometimes desirable in comparing sludges to express their characteristics as those of similar sludges with 90 per cent of water, which is readily done by either of two formulas: 7, _ 100 - Pi , , V\ ~ 100 - Pa ^"'' where 72 is the volume desired, 7i is the volume under the actual conditions. Pi is the percentage of water in the original sludge, and Pi is the percentage in the sludge under the assumed modi- fied conditions; 7i = P. ^^^ Where Vi is the volume desired, 7i is the volume under the actual conditions. Pi is the percentage of solids in the original sludge, and Pi is the percentage of solids in the sludge under the assumed modified conditions. These formulas disregard the effect of the dissolved solids, which are generally small in proportion to the total solids. While it has hitherto been customary to consider sludges in terms of their water content, the student may find it more direct and helpful to consider them in terms of their content of solids. 427 428 SEWERAGE AND SEWAGE DISPOSAL Table 50 will be found helpful in making such comparisons and shows clearly the great effect of a small increase in water content upon the volume of sludge already containing a large proportion of water. Table 50. — Weight and Specific Gravity of 1 cu. yd. of Sludge with Different Percentages of Moisture and Containing Solids of Different Weights per Cubic Foot 1 .a o a Weight of dry residue in pounds per cubic foot 68 78 92 110 134 3 5S 3 ss P O H 3 °% 3 °g 9 « 0, >> O a !■. u p. !- u P< >. u P. o 1 t!| 1 a (A s p. 1 ti, P. ''1 i U tu «_^ 03 bO •*-.+^ ^ bo bu H-i-S ^ bO •".ti S OS bD o 1 ■a r •13 OS fl 1 ■3 -a lis o s ■a" •5S-0 i s ■H |5S ■a", 98 1688 5.04 1.002 1692 6.08 1.004 1696 5.13 1.007 1699 5.18 1.008 1702 5.23 1.010 97 1689 3.35 1.003 1695 3.38 1.007 1702 3.41 1.010 1707 3,44 1.013 1711 3.47 1.016 96 1691 2.51 1.004 1699 2.53 1.009 1708 2.55 1.013 1715 2,57 1.018 1721 2.59 1.022 95 1692 2.01 1.004 1703 2.02 1.010 1713 2.03 1.017 1723 2,04 1.023 1731 2,06 1.027 94 1694 1.67 1.005 1706 1.68 1.012 1718 1.69 1.019 1730 1,70 1.027 1740 1,71 1.033 93 1695 1.43 1.007 1710 1.44 1.015 1724 1.44 1.023 1738 1,45 1.032 1750 1.45 1.039 92 1697 1.25 1.007 1713 1.25 1.017 1730 1.26 1.027 1745 1,26 1.035 1760 1.26 1.044 91 1698 1.11 1.008 1717 1.11 1.019 1735 1.11 1.030 1752 1,11 1.040 1770 1,11 1.050 90 1700 1.00 1.009 1720 1.00 1.020 1740 1.00 1.032 1760 1.00 1.044 1780 1,00 1.056 89 1701 0.91 1.010 1724 0.91 1.023 1748 0.91 1.037 1768 0.91 1.047 1790 0,91 1,062 88 1703 0.83 1.010 1727 0.83 1.025 1752 0.83 1.040 1776 0.83 1.054 1800 0,83 1,068 87 1704 0.77 l.OU 1730 0.77 1.027 1758 0.77 1.043 1785 0.76 1.059 1810 0.76 1.074 86 1706 0.71 1.012 1734 0.71 1.029 1764 0.71 1.046 1793 0,70 1.064 1820 0.70 1,080 85 1707 0.66 1.013 1737 0.66 1.031 1770 0.66 1.050 1802 0,65 1.069 1831 0,66 1.086 84 1709 0.62 1.014 1741 0.62 1.033 1776 0.62 1.054 1810 0,61 1.074 1841 0,61 1.093 83 1710 0.59 1.016 1744 0.58 1.035 1782 0.58 1.058 1818 0,57 1.079 1852 0,56 1.099 82 1711 0.56 1.016 1748 0.55 1.037 1788 0.55 1.062 1827 0,64 1.084 1863 0,53 1.106 81 1713 0.53 1.017 1761 0.52 1.039 1794 0.52 1.065 1835 0,61 1.089 1874 0,50 1.112 80 1714 0.50 1,018 1755 0.49 1.041 1800 0.49 1.068 1844 0,48 1.094 1885 0,47 1.118 79 1715 0.47 1.019 1758 0.46 1.043 1807 0.46 1.072 1853 0,45 1.100 1896 0,44 1,125 78 1717 0.45 1.019 1762 0.44 1.045 1813 0,44 1.076 1863 0,43 1.106 1908 0,42 1,132 A satisfactory method of studying sludge problems of different plants is by means of the dry solids in the sludge, reduced to the number of pounds daily per 1,000 population. In this way the influence of differences in the water consumption and the ground water leakage is eliminated, but a chance for serious error is introduced if the population of a city is only partially served by sewers and there are large storm overflows. At Providence, R. I., chemical precipitation produced sludge containing from 202 to 266 lb. of dry solids per 1,000 population daily and plain sedi- mentation produced sludge with 136 to 140 lb. The chemical precipitation of weak night sewage at Worcester, Mass., gave sludge containing from 140 to 180 lb. of dry solids per 1,000 SLUDGE 429 population daily, while plain sedimentation of strong sewage for 7 hr. gave from 145 to 180 lb. Chemical precipitation at London gave about 236 lb. Another method of reporting the quantity of sludge, which can be used advantageously at times, is in terms of cubic feet per capita daily. For example, the slow sedimentation of the domestic sewage of a small Massachusetts city was found by the authors to give from 0.0146 to 0.0165 cu. ft. per capita daily, the average being 0.0162 cu. ft. While these two methods of reporting the quantity of sludge are the best for comparing sludges of different cities, the common practice of operators of treatment plants is to report the gallons of sludge per million gallons of sewage, and its water content. The weight of the dry solids in sludge has little influence on the weight per cubic yard of the sludge produced from most types of treatment, its influence only becoming marked in sludges with the lower percentages of moisture given in Table 50. Measuring Sludge. — The elevation of the top of the sludge in a basin can be measured by means of a glass tube, through which a wire is strung, a hollow ball being attached to the end of the wire. The tube is lowered slowly into the tank, allowed to remain there long enough for all disturbances to quiet down, and then the bottom is closed by pulling the rubber ball against it. The elevation of the surface of the sewage is marked on the tube. When the latter is carefully raised, the elevation of the top of the sludge can be determined from its distance below the sewage level mark, and the volume of the sludge in the tank can then be computed from the dimensions of the basin. Kinnicutt used a 2j'^-in. glass tube 7 ft. long at Worcester. Johnson used at Columbus a K-in. glass tube 2}'i ft. long, fastened to one end of a wood rod 12 ft. long and closed at the bottom by a rubber stopper which could be pulled into place by a fine wire strung along the rod. This tube had too small a diameter to admit coarse matter into the bore, it was found. The elevation of the upper surface of the sludge can sometimes be measured by slowly lowering through the sewage a weighted i flat board or disk of sheet steel. The motion of the plate is/ noticeably checked when it reaches the sludge, unless the latteij is very fluid. Characteristics of Sludge. — The most voluminous sludge that produced by the activated sludge method of treatment, 430 SEWERAGE AND SEWAGE DISPOSAL amounting to 6,000 to 12,000 gal. or more of excess sludge at 973^^ to 98 per cent moisture per million gallons of sewage treated, and even greater volumes from some classes of industrial wastes. It is generally of a brown flocculent appearance. When the color is very dark the sludge may be approaching septic condi- tions and if the color is very light there may be over-aeration with a tendency for the solids to settle slowly. If the sludge is flowed over a sand bed to a greater depth than 4 or 5 in., the suspended matter settles rapidly and clogs the pores of the sand before the water can pass through them, thus making the drying process a slow one even under the most favorable conditions, because a large part of the water can escape only through evaporation. This sludge has an inoffensive earthy odor when in good condition, but sometimes it has a tendency to become septic rather rapidly and when this occurs it has a dis- agreeable odor of decay. It is so voluminous that apparently the only practicable way of preparing it for final disposal is by mechanical dewatering. Its nitrogen content is generally so high as to be a noticeable characteristic. The sludge from plain sedimentation is gray in color and pos- sesses a very offensive odor in most cases; the sludge from dis- infected sewage at Providence was not particularly offensive, being an exception to the usual condition. The sludge from plain sedimentation is slimy, usually quite dense and very diffi- cult to dewater in filter presses. It can be dried by spreading it on porous beds, but must be spread very thin in order to enable the water to percolate away rapidly. The large volume of sludge produced by chemical precipitation is due to the chemicals introduced into the sewage. If it is left in the tank it undergoes decomposition like the sludge from plain sedimentation, but at a slower rate. It gives off gas in substan- tial quantities and its density is increased by standing. If it contains considerable iron its surface may be red but below the surface the color is generally black. The odors from it may be objectionable but not so bad as those from plain sedimentation sludge. While it is somewhat slimy the hydrate of iron or alumi- num in it makes it gelatinous. When spread on porous draining beds the water will gradually drain away or evaporate, leaving after several weeks a stratum of slimy sludge of about the con- sistence of lard, containing 70 to 80 per cent of water. Septic tank sludge is black and usually very offensive on SLUDGE 431 account of the hydrogen sulphide and other gases it gives oflf. That produced at Birmingham, England, is said to have rela- tively httle odor because of the copper compounds in it. Other septic sludges are reported to possess relatively little odor, due in most cases to long digestion in the septic tanks. The coarser particles settling in the tanks are largely disintegrated by the putrefactive action going on in the sludge, so that it contains relatively small quantities of coarse material. The sludge can be dried on porous beds, if spread out in thin layers, but objection- able odors are to be expected while it is doing so. The sludge from the Imhoff tank is characterized by an excep- tionally large amount of gas contained in it. Although usually containing a larger proportion of sohd matter than other sludges, it flows readily because of the fluidity imparted by the entrained gases. It is less offensive than sedimentation or septic sludge, its odor being like that of hot tar or sealing wax. When drawn off on to porous beds in layers 6 to 10 in. deep, the solids are first carried to the surface by the entrained gases, leaving a sheet of comparatively clear water below them which drains off rapidly and allows the solids to sink down slowly on the bed. As the sludge dries the gases escape, leaving it more or less spongy and with an odor resembling that of garden loam. The sedimentation of trickling filter efiiuents gives a sludge quite different from any produced by the primary tank processes. It is brownish, relatively inoffensive when fresh, flocculent and usually contains a high percentage of water. It undergoes decomposition more slowly than other undigested sludges, although when it contains many worms it may become offensive quicldy. In warm weather gases are liberated from it and large masses of it may at times be carried to the surface by entrained gas. It is like activated sludge in being very difficult to dry on porous beds, and at Worcester, where it contained large quantities of hydrate of iron, it could not be dried readily unless applied in a layer not more than a few inches deep. In some cases, as at Fitchburg, Mass., the secondary tank sludge is pumped to the sewage Imhoff tanks, to take advantage of the fibrous matter and entrained gas in the Imhoff tank sludge and thus secure a sludge mixture which is much more easily drained and dried than the secondary tank sludge alone. By this method of operation, unstable organic matter in the sec- ondary-tank sludge, such as worms, will be put through the 432 SEWERAGE AND SEWAGE DISPOSAL rotting process and reduced to a stable condition not likely to cause offense when the sludge is dried on porous beds. Table 51a gives analyses of dried sludges produced by all tank treatments at Worcester, Mass. Additional analyses of activated sludge produced at four periods during the investiga- tions gave the following average percentages: Total nitrogen, 4.06; available nitrogen, 1.39; total phosphoric acid (P2O6), 2.26; citrate insoluble acid, 0.50; citrate soluble acid (available), 1.76.1 Table 51a. — Percentage Composition of Dbied Shjdges from Difpirent Methods of Sewage Treatment at Worcester, Mass. Chem- ical pre- cipi- tation Plain sedimen- tation Septic tank ImhofE tank Seconda- ry sedi- menta- tion tank Acti- vated sludge treat- ment Volatile solids Fixed solids Silica (SiOz) Iron stilphide (FeS) Iron, not as sulphide Total iron Sulphur, not as sulphide Aluminum oxide CAI2O3) Calcium oxide (CaO) Magnesium oxide (MgO) Phosphorus pentoxide (P2O6) Carbon (C) Hydrogen (H> Nitrogen (N) Fats 47.26 52.74 25.46 0.02 5.80 0.44 0.57 2.88 0.74 0.47 28.60 4.21 2.77 51.04 48.96 28.69 0.57 2.45 0.60 1.94 0.61 0.29 1.71 31.26 4.46 3.05 43.94 56.06 20.41 16.58 2.98 0.64 7.29 1.14 0.97 1.85 23.95 3.64 3.01 49.12 50.88 4.78 51.05 48.95 18.57 63.75 36.25 12.72 2.24 4.29 2 to 9 ^Treating the effluent from a trickling filter. Removing Sludge from Tanks after Draining off the Sewage. — There are two distinct methods of removing sludge, depending upon whether the tank is thrown out of commission and the sewage withdrawn preparatory to removing the sludge or the latter is removed while the tank is in operation without with- drawing the supernatant liquid. The bottom of the tank must be designed to conform with the method of sludge removal adopted. In case the supernatant liquid is first drawn off it is desirable to give the floor a slope which will permit it to be cleaned readily by squeegees and will cause the thin sludge to flow towards the 1 R. S. Lanphear, Eng. News-Bee., April 22, 1920. SLUDGE 433 drains without much assistance. It is often impracticable to provide such slopes in large tanks, however, for laborers are unable to move about readily on a sludge-covered slope steeper than 1 in 20. In such tanks a series of sumps with steep sides is necessary for the complete removal of sludge by gravity, and as such construction is expensive it is customary to give the bottom a slope as great as the conditions of construction and operation permit and rely upon hand cleaning for the complete removal of the sludge. In order to facilitate the work, pipes with hose connections may be run between the tanks, so that the sludge may be flushed along with the aid of water, or sewage from one tank can be used in cleaning adjoining tanks. The cost of hand cleaning is low, amounting under pre-war conditions to 4.9 to 7.3 cts. per 1,000,000 gal. of sewage at Wor- cester, with labor at $2 per 8-hr. day. The lower figure is for chemical precipitation tanks and the higher for plain sedimenta- tion tanks. A more important element than labor cost in hand cleaning is the length of time during which it is necessary to keep the basins out of service. For several hours before the cleaning is begun they must be cut out to permit the sedimenta- tion of the solids in the supernatant sewage. After this, con- siderable time is required for drawing off the supernatant liquid before sludge removal is possible. At Worcester, Mass., about 10 per cent of the basin capacity is out of service all the time for cleaning. The offset to this disadvantage of intermittent re- moval is that the sludge is sometimes more dense than that ob- tained by continuous removal, and its subsequent handling may therefore be easier and less expensive under some conditions. Labor-saving mechanical devices for removing sludge from basins after the sewage has been drained from them are used in England and on the Continent. The Ashton-Booth cleaner is a car running on a longitudinal track on the bottom of the tank and provided with a steel frame for holding a bulkhead of 1-in. planks with a rubber sealing piece along the bottom. After the tank has had its sewage drawn off, sewage from an adjoining tank is admitted behind the car until it attains sufficient height to shove the car along the track toward the lower end of the tank, pushing the sludge along in front of it. When the car reaches the lower end, it is moved back to its original position by admitting sewage against the lower face of the bulkhead. In Bremen a somewhat similar system of cars is used except that they are 23 434 SEWERAGE AND SEWAGE DISPOSAL pulled along a track by rope and windlass at the lower end of the tank. Sludge Removal without Interruption of Tank Service. — Where the topographic conditions permit, it is often desirable to draw off the sludge by gravity without interrupting the opera- tion of the tanks. Fig. 182 illustrates settling tanks at North Attleboro, Mass., designed by Barbour for operation in this manner. Most vertical-flow sedimentation tanks are similarly Plan'iwith Half of Roof Removed Baff/e 5ec+iona! El evasion A^B Fig. 182.— Settling tanki at North Attleboro, Mass. operated. The Gloversville tanks illustrated in Fig. 163, for example, which yield a very heavy sludge on account of the large quantity of tannery wastes in the sewage, are operated continuously. Some difficulty has been experienced, due to the gradual accumulation of solids in the tanks in spite of the frequent withdrawal of all sludge which would flow to the outlet. In one of these tanks the slope of the conical bottom was increased to about 60 deg. and the wooden distributing cone removed. These SLUDGE 435 changes proved a material aid in getting the sludge out, but in spite of these changes it has been found necessary to draw the sludge once or twice a day, to avoid the accumulation of solids in this tank. In the other tank hosing and other means of brealcing down the accumulation of heavy sludge are necessary. Experience at many plants has shown that where gravity is the only force relied upon to make the sludge slip down slopes, the latter should be at least 2 on 1. Sludge which contains gela- tinous floe, often called colloidal matter by plant operators, has a Fig. 183. — Fidler sludge remover, Bolton, England. tendency at times to adhere to very smooth surfaces as well as those of a rougher character, and wherever two surfaces form an angle the sludge tends to remain. Slopes sufficient to cause sludge to flow freely after the supernatant sewage has been drawn off are not necessarily suflBcient to cause it to flow when the sludge is submerged, for the effective weight of the latter in causing motion is reduced by the weight of the displaced sewage. As a result, in drawing off sludge from tanks operated continuously, there is a danger of forming a cone-shaped opening through the mass of sludge in the bottom of a tank and removing sewage as well as sludge, unless the bottom of the hopper is so formed by steep sides that the sludge is inevitably concentrated in it. 436 SEWERAGE AND SEWAGE DISPOSAL The construction of hopper-bottom tanks is often expensive and in order to avoid this expense, reduce labor charges and at the same time operate the tanks continuously, mechanical appliances are used to remove the sludge. The Fidler apparatus has been employed in a number of British cities for this purpose. It con- sists of a spiral band of sheet steel revolved by means of gears Inf/uenf Sewen Sludge Pipe' liernaHve Method of Sludge Removal which mau be subafifufect for Centra I Wei I. Spider FiQ. 184. — Dorr thickener or sewage clarifier. which gradually draws the sludge toward an outlet under the center of the band, as shown in Fig. 183. Experience with this apparatus shows that it is highly desirable to construct it 9f materials which will resist the corrosive action of sewage. The same problem has arisen in removing sludge formed in the cyanide metallurgical process and the Dorr thickener, employed in that process, has recently been adapted for remov- SLUDGE 437 ing the sludge from the settling tanks employed in the activated sludge plant at Milwaukee and at some industrial wastes treat- ment plants. The apparatus, Fig. 184, consists of radial arms at the bottom of a vertical shaft and carrying a number of plows The shaft is at the center of the tank and driven very slowly, so that the plows force the sludge gradually and without dis- turbance toward a sludge outlet at the center of the bottom of the tank, whence it may flow by gravity through a suitable outlet pipe or be pumped from a central cylinder carried by the revolv- ing mechanism. In case it is necessary to pump sludge the work can be done with various types of equipment. If the quantity is small a (rubber) diaphragm pump such as is used for removing water when excavating trenches in wet earth is often employed. Centrifugal pumps are often used in permanent installations, but care must be taken that their suctions are screened against the entrance of large objects. The air lift is employed successfully in pumping sludge from the Imhoif tanks at Fitchburg, Mass., and the Shone ejector has been used successfully for pumping sludge from sedimentation basins to sludge basins at Worcester, Mass. If sludge must be forced into filter presses for dewatering, stuff pumps, such as are used in handling pulp in paper mills, and ejectors give satisfactory service. Disposal of Sludge. — The disposal of sludge is one of the most troublesome problems connected with the tank treatment of sewage. A method followed in some cities is to carry it to sea and dump it in deep water, Table 51f>, and at London six steamers built for the purpose were regularly engaged in this method of disposal in 1903. Most sludge is discharged from the tanks in which it settles to beds where it dries in the open air. Part of the water sinks into the soil and the rest evaporates If the soil does not drain readily and the sludge is dense and greasy, the drying process may take some weeks. The beds are preferably porous and well under-drained with 4- to 6-in. open-joint tile. The effluent from the beds must be kept under observation, for it may be necessary to return it to the works for treatment as raw sewage if there is no opportunity for filtering or otherwise treating it inde- pendently of the sewage. The area required for the beds depends upon the character of the sludge to be treated and the climate in which they are located, a greater area being required where the 438 SEWERAGE AND SEWAGE DISPOSAL Table 516. — Cost of Dumping Sludge Containing 90 Feb Cent op Water at Sea Under Conditions Existing at Several Cities Practicing such Disposal City and year Tons per year Mois- ture, per cent Cost per ton Dry solids Sludge with 90 per cent moisture Miles Dj dump Dublin, 1906-7 Glasgow, 1 06-7 London, 190S-6 Manchester, 1903-5, 7 . . Salford, 1902-6 Southampton, 1906-7 . . . Providence, R. I., 1910. 128,300 341,600 2,838,100 188,700 152,300 95,600 90.0 86.8 92.0 86.0 79.0 90.0 72.4 $0.91 0.75 1.02 1.25 0.82 3.04 180.09 0.07 0.10 0.13 0.08 0.30 0.0591 10 40 50 61 60 25 14 ("American Sewerage Practice," vol. iii, p. 486) 1 Cost per ton for disposal of pressed sludge cake, using a bottom-dump hopper barge. rainfall is 48 in. annually than where it is 24 in. As a general rule, in the Northern part of this country, the range is from about 0.5 sq. ft. per person where quick-drying sludge from Imhoff tanks is to be treated to 7 sq. ft. where the voluminous sludge from chemical precipitation must be dried in this way. It is questionable if sludge from chemical precipitation or acti- vated sludge tanks can be dried economically on open beds except in small plants or in very dry climates. Both classes must be spread in thin layers in order to dry and their volume is so great that a relativelj^ large area of drying beds must be provided and the cost of their operation will be considerable. Owing to the uncertaintj'- about the effect of rainy weather on the drying of sludge in thin layers on porous beds, the method of lagooning has been employed in disposing of it in some cities. A lagoon is a tract of some size surrounded by a dike of earth, which is filled with sludge to a depth of 1 to 5 ft. The sludge is left to dry by evaporation and such small losses of water through the soil as may take place naturally. This process of drying may take from 2 to 6 months and is usually accompanied by offensive odors. When it is finished the sludge is wet and its disposal in this condition is not so easy as is that of a drier sludge. A modification of the lagoon method of treatment is to run the sludge into a trench about 3 ft. wide and 18 in. deep and then cover it with earth. This method has been used in a number of English disposal works where it has been found to require from ^ SLUDGE 439 to 1 acre of land per 1,000 long tons of sludge. As the land once used in this way cannot be employed again for the purpose for a long period, a very large tract must be available for trenching when sludge is disposed of in trenches. Sludge from drying beds can be burned if mixed with coal or other fuel, but generally an attempt is made to get rid of it by dumping it on low land as a waste material without value. In a few places it is hauled away by farmers as a low-grade fertilizer, and probably a greater use of the material in this way would be advantageous if persistent endeavor were made to interest farmers in its value. Dewatering Sludge Mechanically. — The first method of de- watering sewage sludge which was employed practically was filter pressing, used extensively in various industries. A filter press. Fig. 185, is a device by which the liquid sludge is forced under about 80 lb. per square inch pressure into cloth lined chambers in which the solids are retained between the cloths and the water is forced through the cloth and drainage ducts in the plates and escapes. When the chambers have become filled with solids the press is opened and the cakes of compressed material are readily removed because the cloth prevents the material from adhering to the plates. In order to press sewage sludge it is often necessary to add lime to it; from 20 to 30 lb. per 1000 gal. are added in treating the sludge from the chemical precipitation process at. Worcester, and 100 lb. were found necessary in preparing septic sludge for pressing. The sludge press shown in Fig. 185 comprises a fixed head stock to which is attached the piping through which the sludge is pumped into the press, a fixed tail stock provided with the hand or power-driven apparatus for opening and closing the press, from 50 to 75 cast-iron plates, and a follower which is heavier than the plates and receives the pressure exerted by the operating apparatus carried by the tail stock. The plates and follower move along parallel steel side rods which also tie the head and tail stocks together. The plates used in pressing sludge are generally about 36 in. in diameter and 3 in. thick, with a 6-in. hole at the center. Tests at Providence show that the best cloth used with the plates, called "clothing," is 11-oz. army duck. The clothing is made by cutting strips a little more than twice the vertical length of a plate and then cutting two 6-in. 440 SEWERAGE AND SEWAGE DISPOSAL holes in each strip to correspond with the central hole in the plate. The cloth is hung over the top edge of the plate, and a clamp is put through the holes in the cloth and the opening in the plate and drawn tight, thus covering both faces of the plate with canvas. The surfaces of the plates are corrugated and the liquid which gathers in the corrugations in the pressing process runs eventually Fig 185 — Filter press for sewage sludge (Bushnell). into a drainage duct, usually a J^^-in. hole in the bottom of the plate, through which it escapes to a drainage channel below the press. After the plates have been closed, they are forced to- gether by the closing apparatus and the sludge is pumped into the press. The pressure under which it is pumped forces out a part of the water. After the presses are filled, the inlet valve is closed, the follower is pulled back, and the cakes of sludge are allowed to drop through hoppers into cars which remove the SLUDGE 441 material from the press house. At Worcester the pressing process expels sufficient water to leave a sludge cake of 68 to 74 per cent moisture. The sludge cake is used for filling and a subtantial amount is hauled away by farmers for use as a fertilizer. Dewatering may also be accomplished by centrifugal machines. The first practical sludge centrifuge was the Schaefer-ter-Meer machine, Fig. 186, used in several German cities. The sludge is admitted to the top from an elevated tank and passes down through a hollow shaft to the runner of the centrifuge. This runner has six radial cells, one radial side of each cell being formed of a sieve which permits the water to escape from the sludge while it is being whirled rapidly. The sludge is admitted to these cells from the hollow central shaft through openings which may be closed by gates, and it is thrown out of the cells through openings in the outer ends of the cells, which can also be closed by gates. While the cells are filling with sludge, the outer gates are shut and the inner set open. When the cells are full, the inner gates are closed and the machine kept whirling until the operator judges the dewatering has reached the desired point. The outer gates are then opened and the sludge is thrown out of the cells by centrifugal force against the walls of the chamber. The impact breaks it into pieces which fall through a hopper to a conveyor by which they are removed. The water is driven through the solids accumulated in each cell and is collected by a plate behind each sieve and guided to a vertical drain delivering it to an efHuent trough. The gates are moved by oil- pressure and it is unnecessary to stop the runner in order to charge and unload the cells. The apparatus is reported by German engineers to furnish sludge with its moisture content reduced to 55 to 75 per cent. A centrifuge developed from and said to be an improvement upon this Schaefer-ter-Meer machine has been used to some ex- tent and is now (1921) being tried upon sludge from the activated sludge demonstration plant at Milwaukee. It has a drum or basket similar to that of the usual laundry centrifuge, but sub- divided around its perimeter by radial vertical vanes projecting about 6 in. in from the drum wall. As sludge flows into the revolving drum, the solids are driven into the compartments, where a cake is gradually built up from the wall inward. In this case, none of the water is forced through the solids, but, on 442 SEWERAGE AND SEWAGE DISPOSAL SLUDGE 443 the contrary, the solids are driven by centrifugal force outward against the drum wall leaving the water in the center to escape over the top of the drum. Like its predecessor, this centrifuge may be run continuously, the opening and closing of the valves being done automatically while the drum is revolving at full speed. Tests thus far made with the activated sludge show a tendency for the coarser and heavier solids to be formed into cake quicldy and for the finer solids to be carried over with the efHuent, The cake generally contains upwards of 80 per cent moisture, although a drier cake can be produced by prolonging the cycle and admitting the sludge slowly. While numerous experiments in dewatering sludge have been conducted with centrifugal machines made in the United States for treating slimes and other industrial purposes, they have not been encouraging by comparison with the results obtainable with filter presses. Drying Sludge by Heat. — If sludge is to be dried to serve as a fertilizer base or other purpose requiring a practically dry material, a rotary dryer such as is used in many industrial processes may be employed. At Milwaukee, where the dewat- ered sludge cake must be converted into fertilizer base, the pre- liminary experiments showed that in producing bone-dry material in such an apparatus from pressed acidified sludge containing 82 per cent moisture, no nitrogen was lost; in reducing pressed sludge containing 78 per cent moisture to material with 5.6 per cent moisture, the volume of the dried mater'al was diminished to 18 per cent of the volume of the sludge put into the dryer. Where sludge cake is treated in this way provision must be made for preventing odors. At Milwaukee the gases from the experimental dryer passed through a dust chamber and were then washed with water by a spraying apparatus, after which they were discharged through a stack. Fertilizing Value of Sludge. — The ingredients most needed in fertilizers are potash, phosphoric acid and nitrogen. Potash is usually obtained as muriates and sulphates, and all forms used in fertilizers are freely soluble. Phosphoric acid comes mainly from phosphates, in which it exists in combination with lime, iron or aluminum. It occurs in three forms: (1) soluble in water and readily taken up by plants; (2) slightly soluble but never- theless readily assimilated, in which condition it is termed 'reverted;" (3) very slightly soluble and assimilated but slowly. 444 SEWERAGE AND SEWAGE DISPOSAL The first and second forms are collectively termed "available" phosphoric acid. The term "superphosphate" is applied commercially to any material containing soluble phosphoric acid as a chief constituent. Nitrogen is usually supplied as sulphate of ammonia or nitrate of soda. These ingredients are dissolved so easily in the case of commer- cial fertilizers that they are not applied until just before they are needed. On the other hand, in the case of barnyard manure a portion of those elements useful to stimulate plant growth are so combined with other substances as to require breaking down into simpler and more easily soluble matter through bacterial action in order to render them available for plant food. Hence farmers sometimes apply manure to their fields in the autumn in order that its valuable ingredients may be made available and leached into the soil by the time they are needed in the spring. Sewage sludge is similar to barnyard manure in these respects. Many practical tests of the fertilizing value of the sludge from plain and chemical precipitation tanks have been made in Great Britain. In a general way they indicate that such sludge is of little value on grass land or as a fertilizer for root crops of rapid growth, but it has increased the yield of wheat 10 to 12 per cent in some cases. Investigations made by Lipman and Burgess at the University of California show that different soils have markedly different capacities for utilizing the nitrogen of sludge. Their experi- ments were made with nine sludges having the fertilizing values given in Table 52. The "availability" of the nitrogen, by which they consider such materials should be judged, was determined by a method originated by them. One gram of thoroughly dried, finely ground sludge was mixed with 100 g. of soil and enough water added to produce the best moisture conditions. The tumbler containing this sample was covered with a Petri dish cover and incubated for a month at 28 to 30°C. At the end of this period nitrate determinations were made on these soil cultures. Experiments were also made to determine the average "availability" of the nitrogen in several commercial organic fertilizers. These investigators were much encouraged by the results of their tests. In the dry climate of parts of California it may be practicable to produce air-dried sludge more easily and certainly than where there is a greater rainfall, and in this way obtain at SLUDGE 445 low cost a useful material for adding to the humus-poor soils of that State. An important fact connected with the results given in Table 52 is that the absolute amounts of nitrate pro- duced from sludge nitrogen were often 50 to 75 per cent as high as those produced from similiar weights of dried blood and high- grade tankage. Table 62. — Partial Composition or Air-dried Sludges and their Available Nitboobn Compared with that of Commercial Organic Fertilizers (Experiments by C. B, Lipman and P. S. Burgess, University of California) Material Vola- tile mat- ter, per cent Ash, per cent Total N., per cent Nitrate N., per cent Phos- phoric acid, per cent Percentage of N. in sludges and fertilizers avail- able when inoculated with 03 O a" " S Sludge: Orange, Imhoff tank (city) . . . Fullerton Imhoff tank Anaheim Imhoff tank Lindsay septic tank Pasadena Imhoff tank Orange Imhoff tank (county) , Worcester exp. Imhoff tank . . Cleveland exp. Imhoff tank . . Chicago Stock Yards exp. Imhoff tank. Fertilizers: Dried blood High-grade tankage Low-grade tankage Fish guano Cottonseed meal Goat manure 49.68 25.31 33.09 42.92 29.34 38.41 43.86 36,37 50.46 50.32 74.69 76.91 57.08 70.76 61.59 56.14 63.03 49.54 2.66 1.23 1.54 1.83 1.68 2.38 2.10 1.44 1.73 0.012 0.045 0.115 0.090 0.135 0.060 0.010 0.000 0.400 1.11 0.86 0.99 0.89 1.46 0.77 1.82 1.28 1.46 32.50 43.90 32.40 28.70 38.00 25.70 26.90 32.90 24.50 12.79 16.21 27.39 15.11 14.18 4.89 32.30 43.50 36.00 18.00 28.20 21.40 12.40 8.30 9.80 0.00 0.00 22.70 trace 2.00 3.50 27.20 40.60 40.20 18.80 35.70 15.70 34.50 44.10 10.10 4.05 3.95 43.89 4.65 21.45 10.39 A possible explanation of the failure attending the use of sewage sludge as a fertilizer in some places may be that it was applied too late for its useful constituents to have time to leach out, or the soil may not have been adapted to receive such sludge. Among sewage sludges that from tanks used to settle trickling filter effluent, after drying, appears to be a relatively good fertilizer. Activated sludge has given some surprisingly good results, as at Toronto,' where root crops gave yields with ^ Engineering <& Contracting, July 10, 1918. 446 SEWERAGE AND SEWAGE DISPOSAL it from 40 to 55 per cent larger than those obtained with barn- yard manure. Many attempts have been made to sell dried sludge as a fer- tilizer in Europe, probably the most successful being that of Glasgow, Scotland, where all the sludge produced at the Dalmar- nock works is sold in the form of pressed cake or dried fertilizer. This commercial success was ascribed to careful business organi- zation and judicious advertising by the Royal Commission on Sewage Disposal, which estimated that the sales of sludge decreased the net cost of sewage treatment about 80 cts. per 1,000,000 gal. Investigations of the practicability of drying sludge to form a fertilizer base have generally been discouraging in the United States. Since the advent of the activated sludge process there has been a somewhat brighter prospect for sludge utilization at large treatment plants. The advisability of preparing sludge for the market will ordinarily depend upon whether or not the fertilizer and grease which can be recovered have sufficient value to render this com- plicated method of sludge disposal materially less expensive than other simple methods which may be available and suitable for the local conditions. Before the sludge can be put on the market its water content must be reduced to about 10 per cent, which is considered commercially dry. This dewatering can be done by combinations of the methods already explained. The authors have estimated^ that activated sludge can be pressed and drum- dried to 10 per cent moisture for about $4.90 per 1,000,000 gal. of sewage treated, assuming that each 1,000,000 gal. of sewage will produce 4500 gal. of activated sludge containing 97.5 per cent water. To this must be added the cost of marketing, handling and transporting the final product. The theoretical fertilizer value of commercially dried activated sludge containing 3 per cent available nitrogen and 3.5 per cent phosphoric acid (P2O6) was estimated at $10.30 per ton, or $5.47 per 1,000,000 gal. of sewage treated, assuming that each 1,000,000 gal. of sewage would produce 1,063 lb. of fertilizer. The volume of activated sludge produced and weight of dried material per unit volume of sewage will vary at different places according to the composition of the sewage. The composition of the dried material will also vary at different places. If grease is present in such proportion as to be seriously detri- 1 "American Sewerage Practice,'' vol. iii, 2ad ed., p. 825 et seq, (1916). SLUDGE 447 mental to the fertilizer, it will be necessary to degrease the dried sludge as provided for in the Miles acid process of sewage treat- ment. The best way to degrease the dried sludge appears to be by naphtha percolation, a process somewhat dangerous and not likely to prove advisable unless the sludge contains a relatively large proportion of grease. The cost of degreasing Milwaukee dried activated sludge was estimated to be about 90 cts. per 1,000,000 gal. of sewage treated, and after degreasing this material will probably have to be redried at an estimated cost of $1.70 per 1,000,000 gal. of sewage. This makes an additional cost of $2.60 per 1,000,000 gal., for the recovery of grease. It was estimated that the value of the grease in activated sludge containing 5 per cent of grease in the dry material, would be about $4.50 per ton of fertilizer, or about $2.39 per 1,000,000 gal. of sewage. This proportion of grease would not be sufficient to warrant its recovery. It is not so high as to seriously affect the value of the sludge for fertilizer. From what has been said it will be seen that the preparation of fertilizer from activated sludge is a manufacturing process having a number of stages which can probably be conducted profitably only on a large scale. In addition to this production aspect of the work, there is a business side involving the sale of the product at the highest possible price and the management of the entire enterprise in an efficient yet economical manner. Whether it is desirable for a municipality to embark upon such a business is a debatable question and wiU depend upon the local conditions. Unless the enterprise is well handled there are possibilities of grave financial troubles. The overhead expense is likely in any case to be large in comparison with many muni- cipal enterprises, and the proportionate cost of operating a small sludge fertilizer plant will probably be much greater than that with large plants. CHAPTER XVI CONTACT BEDS Prior to the development of the activated sludge process, tank treatment of sewage did not remove fine suspended and colloidal matter and it was necessary to treat tank effluents in order to oxidize their organic matter before their discharge into some waters. The purpose of these treatments was essentially to transform the organic matter in the tank effluents into substances possessing so little demand for oxygen that they would not exhaust the supply of that element in the waters receiving the effluents and produce putrefactive conditions. Perhaps the Automatic Dosing A ppara tus Automatic Draining Apparatus Aufomatic Draining Apparatm The horizontal dimensiona are shown greatly contracted in comparison with the vertical dimensions. Fig. 187. — Arrangement of double contact beds. most fogical step forward from the types of tank treatment in use prior to about 1913 was treatment in the contact or bacteria bed. The contact bed is a watertight tank, generally uncovered, filled with broken stone, cinders or other inert material, com- monly }/2 to 13^ in. in size and 4 to 5 ft. deep. This material is often termed "ballast." When new the bed will have from 40 to 50 per cent of voids, but these gradually become filled with sewage solids and the contact material has to be removed, cleaned and replaced after a service of perhaps 5 years. Such beds are often built in series, as shown in Fig. 187; the first bed to receive the sewage is termed the "primary" bed, and its effluent goes to the "secondary" bed; sometimes the effluent from the secondary bed is treated in a "tertiary" bed. 448 CONTACT BEDS 449 The contact bed is operated by filling it with sewage, allowing it to stand full for a time, then drawing off the sewage and finally- allowing it to rest empty. After resting, the cycle is repeated. The changes that take place in the sewage during the cycle are physical, biological and chemical. Although the sewage applied to a contact bed has generally been settled, it retains a part of its settling solids and much of its suspended and colloidal matter. The settling solids are largely retained in the interstices of the ballast and the colloids are withdrawn from the sewage by the physical attraction of the contact material. Certain bacteria develop in the organic matter adhering to the ballast and the pieces of stone or clinker become covered with a gelatinous film. The organisms in this film play a large part in converting the organic matter of the applied sewage into more stable organic and mineral substances by oxidation processes. In the contact bed operating successfully, the organic nitrogen compounds are transformed, with the production of free ammonia, the free ammonia is transformed into nitrites and nitrates, and nitrogen gas is evolved. There is much uncertainty about the processes which take place, but apparently the nitrates are formed mainly while the bed is standing empty and are partly or wholly reduced during the standing-full period, so that the effluent has a relatively low nitrate content. While there is probably some mechanical straining out of suspended and colloidal matter, this mechanical action accounts for but a part of the removal effected in a contact bed, for a matured unclogged bed will with- draw more suspended matter from sewage than a bed not matured but otherwise the same. The contact bed requires relatively less head for operation than the trickling filter does, which is a decided advantage where the available head is sufficient for a contact bed, but pumping is necessary for a triclding filter. Further, with contact beds there is a comparative absence of the objection- able odors and the moth flies more or less prevalent about trickling filters. F illin g and Draining. — As these two operations have no biolog- ical significance, it is generally held that the more quickly they take place the better, so long as the movement of the liquid is not sufficiently rapid to disturb the materials in the bed or the film adhering to them. 29 450 SEWERAGE AND SEWAGE DISPOSAL Standing Full. — This period should not be so long that an- aerobic decomposition takes places to an appreciable extent. In some cases Clark and Gage found^ no marked difference when the period of contact varied from to 5 hr., but, on the whole, periods exceeding 2 hr. gave inferior effluents. Johnson found that at Columbus even 1 hr. was too long a period to give the best results. The Royal Commission on Sewage Disposal stated in its Fifth Report that 2 hr. contact and 4 hr. rest gener- ally gave the best results in the practical operation of contact beds, but added that no rule of general appUcability could be laid down. Standing Empty. — It is generally considered that the most important biological purification of the organic matter takes place while the bed is standing empty, and that this period should be made as long as possible, taking into account the necessary time for filling, standing full, and draining. It is diSicult to determine the proper number of fillings per day in advance of actual experience with a plant, for this depends upon the quality and strength of the sewage, the quantity and character of its suspended matter, the size and character of the ballast, the extent to which the ballast is clogged, and the character of the effluent desired. The stronger the sewage and the greater the quantity of suspended matter in it, the smaller will be the quantity which the contact bed can convert into an effluent of good quality. A fine contact material cannot treat as great a quantity as can material of medium or coarse size, although for a time the quality of the effluent produced by the first may be decidedly better. Clogging the interstices of the ballast results in a loss of capacity, for it becomes necessary to operate the bed with small quantities of sewage and allow long periods of rest. As before stated, British experience has indicated that 4 hr. rest is enough for average conditions, but this statement applies only to matured beds and does not take into account any periods of complete rest for a day or more when the ballast has become unusually clogged. Maturing of Beds. — In starting contact beds with either new or washed ballast, it is necessary to allow the sewage a long period of contact in order to obtain a non-putrefactive effluent. At Manchester, England, where the largest installation of contact beds has been made, it was found advisable to employ a 12-hr. I Report Mass. State Board Health, 1908, p. 445. CONTACT BEDS 451 contact period with beds having clean ballast. This period was gradually reduced as the beds matured until, after a service of several years, the total time for filling, standing full and draining did not exceed 2 hr. This slow maturing of the bed seems to be necessary in order that the pieces of the ballast may become coated with the gelatinous film which plays such an important role in the action of a contact bed and to afford an opportunity for the proper biological conditions to become established before the beds are worked to their highest capacity. Loss of Capacity. — As soon as a contact bed is put in operation it begins to lose capacity. This may continue until the voids are completely clogged and the bed is inoperative, or it may stop after reaching a matured condition, depending upon circum- stances of design and operation. The position of the clogging material within a bed sometimes varies from year to year; in some beds it will be near the bottom and in others near the top. The Royal Commission on Sewage Disposal found that there were eight causes of clogging, as follows: 1. When beds are constructed of friable material, it may disin- tegrate, so that the voids will become partly filled with small pieces of the ballast. 2. The ballast may become more consolidated than when it was placed in the bed, particularly if the material disintegrates to any extent. Consolidation apparently is more serious in beds of fine material than in those of material of medium or coarse size. 3. Deposition of colloidal matter. 4. The growth of organisms is at once the cause of increased efficiency in the bed and of loss of capacity, according to Dr. Fowler. By working beds at high speed, without long periods . of rest, the effluent may remain good, but the bacterial growth increases so rapidly that the bed becomes too spongy and will not allow the liquid to drain away freely. When this condition is observed a rest of 1 to 2 weeks should be given the bed, Fowler stated. 5. The volume of liquid passed on to the bed. 6. Insufficient rest. 7. Inefficient drainage. 8. Where the greater part of the suspended matter in the incoming sewage finds its way into the body of the ballast, as is the case with coarse material, the Royal Commission on Sewage Disposal found that there was a tendency for the loss of capacity 452 SEWERAGE AND SEWAGE DISPOSAL to vary in direct proportion to the quantity of suspended solids in the applied sewage. But if the bed is composed of very fine material or a bed of material of medium size is covered with very fine material, a large proportion of the suspended solids arc caught on the surface. In a few American contact beds, clog- ging of the ballast by suspended solids has been reduced by replacing from 6 to 12 in. depth of ballast at one corner of the bed with fine cinders and constructing a bank of the same cinders about 1 ft. high around this small area. The sewage is applied at this corner and must filter through the cinders before entering the mass of the ballast. The clogged surface cinders may be raked off from time to time. Multiple Contact. — Under some conditions a single contact has not furnished a satisfactory effluent and another contact has been given it in order to obtain an effluent of the desired character. Investigations have generally shown that double contact wiU permit a somewhat greater quantity of sewage to be treated per acre of the entire area with satisfactory results than is possible with single contact beds of the same area. Triple contact generally requires so much head for operation that it is financially impracticable. Kinds of Contact Material. — Beds have been constructed of various kinds of contact material, cinders, slag, coke, broken stone, pebbles, broken brick, slate and saggars or waste from potteries. Experience has shown that the contact material should be strong enough to withstand the strains developing within a bed without disintegrating. Cinders are generally easily broken into fine pieces. Clinker is usually hard enough to withstand such action. Coke is so light that the rising of the sewage in a bed during the filling period and the reverse action during this drainage period tend to displace it and this movement aids disintegration. The rough surface of all such materials is helpful in holding the gelatinous film which is the seat of biologi- cal activity in a bed of this type. On the other hand, friable materials may prove so much cheaper than those more durable that economy will dictate their use, with enlarged area to offset reduced unit capacity and with provision for frequent wash- ing and replenishing of the quantity of material lost at each washing. EngUsh experience with the various kinds of contact material has been summarized by the Royal Commission as follows: CONTACT BEDS 453 "The experience at Manchester, with regard to contact beds, has been that hard furnace clinker is the best material, being vesicular and yet per- manent. Dr. Fowler, however, expressed the opinion that broken Stafford- shire bricks, saggers and gravel, if properly sized, should do equally well. "It will be seen from the evidence that many witnesses have stated that substances like gravel, flints, or broken granite, are suitable as a filtering medium, but that, in most cases, they have expressed a decided preference for the use of those materials which present the roughest and most irregular surfaces, such as clinker, coke and saggers. "So far as our own experience goes, we should prefer, as a general rule, to use coke or good clinker in a contact bed in preference to other materials.' It is true that there is some evidence to the effect that rather better effluents can be obtained from contact beds of coke than from contact beds of clinker, but we think that this advantage is balanced by the tendency of the lighter material, coke, to shift slightly every time the bed is filled, and therefore to be more liable to disintegration." (Fifth Report, p. 66.) Johnson found at Columbus, Ohio, in 1905 but httle choice between broken stone and coke. There is some evidence that hard, smooth material offers less resistance to the passage of suspended solids into the lower portions of the bed, where it is undesirable that there should be an undue accumulation of organic matter. Size of Contact Material.^ — ^The Royal Commision on Sewage Disposal has summed up^ the British experience as follows: "It is important that the material in a contact bed should not be too small, especially if the liquid to be treated contains an appreciable quantity of suspended matter, as some of this suspended matter will undoubtedly find its way into the interstitial spaces and prevent proper drainage of the bed. "In deciding upon the size of the material to be used, the amount of suspended matter in the liquid to be treated must be considered. As a general rule, the greater the amount of suspended matter in the liquid the larger the material should be. "With a crude sewage containing 40 parts per 100,000 of suspended matter, the material will probably have to be 3 in. and upward in diameter, and even then sludge wUl accumulate on the top. "With septic tank liquor containing 8 to 10 parts per 100,000 of suspended matter, material of a diameter from H to % in. may probably be used effectively; while with a good precipitation liquor containing from 1 to 3 parts of suspended matter the best results will probably be obtained with material as fime as J4 in. diameter. "It is, however, impossible to make any but the most general statement as to the most suitable size of material for contact beds, as, in some cases, ' "Broken 'saggers' might be equally good, but we have had no experience of this material in contact beds." ' Fifth Report, p. 67. 454 SEWERAGE AND SEWAGE DISPOSAL there may be special circumstances which affect the question, such as the character of the suspended matter or the smoothness of the filtering material." Another point to be considered in connection with the size of the contact material is the effect of the size on the clogging of the bed. Beds constructed of materials of the sizes recommended by the Royal Commission have often become so clogged that these materials had to be removed, washed and replaced after 3 to 5 years' use. This is an expensive operation. There is some evidence that if contact material not smaller than 1 in. in size is used and the management of the beds is good that serious clogging of the beds can be avoided, but it is questionable if the average results obtained with such coarse beds will be as good as those obtained with the finer beds, unless a much greater quantity of contact material is used so as to obtain adequate area of gelatinous surface to effect the desired changes in the applied liquid. Furthermore, with such coarse beds, the under- drainage must receive greater attention than is necessary with beds of finer material and may require more expensive construc- tion, and the effluent may contain such a large quantity of sus- pended organic matter at times that settling basins should be provided for its removal, as is the case with the effluents from trickling filters. Depth and Drainage of Contact Beds. — The investigations of British contact beds by the Royal Commission on Sewage Dis- posal and the experiments of the Lawrence Experiment Station have shown that the depth of contact beds has very little effect on their activity when they are in good operating condition. Other conditions governing the depth are the drainage and the maintenance of beds in good condition. Ordinarily these beds have pipe underdrains laid in large stones, making an underdrainage system at least 6 in. deep. The sewage which remains in this underdrainage system is not so well treated as that in the main body of the bed, where the smaller size of the stones affords a much greater area of gelatinous film per cubic yard of material. There is little to be said in favor of such underdrainage, and better drainage can be undoubtedly afforded by a floor system similar to those now used with trickling filters. Experience with British contact beds, particularly those at Manchester, has shown that loss of capacity was often due more to imperfect drainage than to an accumulation of solids. CONTACT BEDS 455 The objectionable feature of such a floor is its absence of purify- ing influence on the sewage in it, but this can be remedied by filling the spaces in the floor with effluent from an adjacent bed before dosing the bed. In any case, it seems undesirable to have this least efficient part of the bed more than one-fifth of the whole depth, and for this reason the Royal Commission suggested 2J^ ft. as the minimum allowable depth. Deep beds place a considerable weight on the material in the lower part, which may lead to its disintegration. Furthermore, when washing becomes necessary, it is much more difficult to remove and handle the material from deep than from shallow beds and the danger of breaking it up is increased. For this reason the Royal Commission suggested that 6 ft. was probably the maximum limit of depth. Size of Beds. — The size of the units is more important with contact beds than with most other methods of treatment. The unit should be small enough so that under normal conditions of sewage flow the time required for filling shall not be unduly long. Large units are also inadvisable on account of the long time required for drainage. At Manchester, England, with a sewage flow of about 36,000,000 gal. daily, the individual beds were made 3^ acre in area. At Plainfield, N. J., and Mansfield, Ohio, the individual units were made 0.22 and 0.25 acre respectively. Constructioti. — Some of the first British contact beds were constructed merely in excavations in practically impervious earth, the bottoms and sides being unlined. Experience with them showed a general tendency of earth to work into the contact material and for the underdrains to become distorted by unequal settlement. Where earth embankments were used to separate beds the leakage through them was occasionally so great that adjoining beds had to be filled simultaneously. On the other hand, contact beds at Mansfield, Ohio, constructed with an earth bottom into which hard cinders were rolled to secure as great density as possible and with earth embankments, which went into service in 1902 were found in 1914 to have a total leakage of 5 per cent, mostly through a revolving valve controlling the discharge of the underdrains. In the United States concrete has, however, generally been employed for the walls and bottom of contact beds. Fig. 188 shows the general arrangement of the underdrainage and distribution systems of a typical bed used at Manchester, England. The bottoms of the beds are concrete, in which the 456 SEWERAGE AND SEWAGE DISPOSAL CONTACT BEDS 457 underdrain channels are formed. These channels are covered with perforated stoneware slabs laid flush with the surface of the bottom. The bottom was graded in ridges with flat slopes to the underdrains. The sewage from the supply channel flows into a small distributing reservoir, from which it passes over a semi- circular weir into channels in the surface of the bed. The chan- nels are lined with fine material which tends to arrest suspended matter and keep it from entering the bed. The material in the beds is washed and screened at intervals of 4 to 6.4 years; the portion from J-| to J^ in. in size is used for surfacing secondary contact beds and the portion from 3^ to 2 in. in size is used in primary beds. The beds receiving sewage are 3 ft. 4 in. deep and those receiving storm water are 2}^ ft. deep. The beds at Mansfield, Ohio, designed by Barbour, are laid out in the form of a circle, each of the 3'^-acre beds forming a sector separated from the next by earth embankments. The beds were underdrained by open-joint tile laid in depressions about 6 in. deep and covered to a depth of 4 to 9 in. with cinders from J^ to 1)4 in- in size. The contact material was hard clinker crushed and screened at the plant to J-^ to % in. in size; the total depth of the contact material was 4 ft. 9 in. Septic sewage was discharged by automatic dosing apparatus into an 18-in. iron pipe ter- minating in a concrete culvert at the apex of each filter. The liquid was distributed over the surface by wooden carriers 6 in. deep and from 3 ft. 4 in. wide where the sewage was received to 6 in. wide at the ends of the branches. The largest American installation of contact beds was at Plainfield, N. J., where four pairs of primary and secondary beds were constructed in 1901 and four more in 1905. This plant was superseded in 1916 by a trickling filter installation. The underdrainage was provided by 5-in. horseshoe tile radiating from the gate chambers in the center of each set of four filters, and coarse stone spread between and over them to a depth of 6 in. This drainage system proved inadequate according to Lanphear,' who operated the plant for some time. It was found when beds were opened in 1909-1910 that the lower portion of the material had become a wet, black mass resembling mud. The contact material was 4 ft. deep. That originally used in the first set of primary beds was broken trap rock }i to 13^ in. in size and somewhat smaller material in the secondary beds, slag ' Eng. Bee, Aug. 10, 1912. 458 SEWERAGE AND SEWAGE DISPOSAL being used in some and cinders in the others. In the beds con- structed in 1905 cinders }4 to l}i in. in size were used in both the primary and secondary beds. Lanphear recommended the use of trap rock 1 to 2 in. in size for primary beds under the local conditions, for finer material became too clogged. While the secondary filters were considerably clogged after a service of 7 and 12 years,' they held one-half to two-thirds of their original capacity and about two and one-half times as much as the pri- mary filters. Until 1909 all beds were filled from the top; in 1910 the primary filters were filled from below in order to reduce the odor from the exposed septic tank effluent which was applied. Experience showed that where bottom filling is practiced larger stone and better underdrainage are necessary than characterized the plant at that time. Automatic Control.^ — The four stages of an operation cycle in the contact treatment can be controlled manually or by means of automatic devices specially designed for the purpose. In con- sidering the advisability of automatic control, it must be remem- bered that both the quantity and quality of sewage vary greatly from hour to hour and from day to day, and that the capacity and efficiency of the beds change from time to time: In a small plant, with from three to five beds, it is possible that the same bed may receive day after day the strong sewage of the day time, while the other beds, receiving the weaker sewage, are called upon to do less work. In most cases, however, the fluctuations in flow from day to day will cause more or less change in the hourly cycle, and under such conditions the use of automatic apparatus for dosing will probably insure better work on the part of the beds and less expense for caretakers than the operation of the controlling gates by manual labor. In large plants the complication of the apparatus is so great and the beds vary so much in their capabilities that it is probable that better results can be obtained by the intelligent operation of gates by manual labor than by automatic dosing apparatus. For automatic control, two sets of apparatus are required, one for filling the bed with sewage and the other for emptying the bed after a definite period of contact. Various types of apparatus have been used for the purpose^ and among these the air-lock feed has been found particularly satisfactory where only a limited head is available. For discharging sewage from the bed after contact, timed siphons have been extensively used. ^ "American Sewerage Practice, vol. iii, p. 714." CONTACT BEDS 459 Air-lock Feeds.— Fig. 189 shows in outline the general features of an installation of air-lock feeds for four contact beds. to 1 i ; 1 I ^ p1| ^ <. is T r r' — ^.^-^ 1 ^ "^ML- ^.- ^1 1 'J ,.. (ii P — — . "^■^v 111 1 i o> t; 1 « hCQ 1 ■s tr '- •^■■ J f 05 ■5 1 ^ 1; 1 m^ cmr Ii. ^ ^^Y\ \ c .s cli! iT^ ^ ) i — .. .Q 1 0V J a. ! r 1 L— U^.g- ■""X. •-- ."=: « 'H- ■i: ^° 1, ?^ 1 ,n H1^ '^^ / <5 , v<^ ! aC3t 3 i W" ( V 1 1 f"S ^i _ -1- \^ "C! 1 vj 1 i ^ h) i u '' I u i At the beginning of a cycle of operation the wells in front of the air-lock feeds are filled with water, except that which is to operate fi rst. All of the blowoff traps are filled with water. 460 SEWERAGE AND SEWAGE DISPOSAL Sewage entering the channel from the settling tanks flows through the feed (the one not sealed with water) into the bed until it is filled, or until the sewage level reaches the top of the withdraw siphon. A slight additional rise in water level causes the withdraw siphon to come into operation, and the compression chamber is filled through the withdraw siphon until the sewage level inside is the same as in the bed. As the compression cham- ber fills, the air in the compression dome is put under pressure and forced into the upper part of the feed, gradually displacing the sewage flowing through the feed until the sewage level is forced down below the inside crest of the feed, when the flow through the feed ceases and the feed is air-locked. The same rise of sewage in the compression chamber also produces a pressure in the starting bell, which is transmitted to the blowoff trap of the feed next to operate. Just before feed (1) is air-locked the seal in blowoff trap (2) is forced, thus releasing the air confined in feed (2), and allowing the sewage to discharge into contact bed (2). This prevents any backing up in the distributing channel or settling tanks. After standing for the required time, the sewage in the bed is discharged by a timed siphon, described later. As the sewage level in the bed falls, siphonic action is started in the withdraw siphon and the compression chamber is drained. By this means the compression dome and starting bell are vented, the blowoff trap is filled and the chamber is ready for the next cycle of operation. Timed Siphons. — These siphons, one for each contact bed, are for the purpose of automatically controlling the time the sewage stands in the bed. The general details of the apparatus are shown in Fig. 190. At the start, the main trap, the blowoff trap and tile well in the timing chamber are filled with water. The size of the timing chamber and the size of the opening in the timing valve determine the period of contact and require trial and adjustment to obtain the specified period of contact. The timed siphon is controlled by the starting bell in the timing chamber, and until there is sufficient pressure in the starting bell to force the seal of the blowoff trap, the siphon will not discharge and the bed will stand full. The timing valve is located below the full water level in the contact bed, so that when the bed is full there is a continuous discharge through the timing valve into the timing CONTACT BEDS 461 chamber. The siphon receives air through the vent when the sewage has been drawn down to the low level and the discharge then ceases. When the timed siphon is operating, the draining siphon is discharging the sewage in the timing chamber, and at the end of the discharge the starting bell is vented, the timing chamber emptied and the apparatus is ready for the next filling of the bed. , OpTn'mg info Se7 Mam iififfon Chamber Timing Chamber f ) Plan !©i 5ed Screenj, Tile »en o ^ Eleva+iorj FiQ. 190. — A timed siphon (Miller). The total head required to operate contact beds equipped with air-lock feed and timed siphon ranges from about 12 in. for the smallest installations to 21 in. for the largest. Operating Results. — If a 1-acre bed 5 ft. in depth is composed of contact material having 30 per cent of voids, this percentage being usually substantially exceeded when the bed is new, one dose would consist of 488,700 gal. of sewage, and the rate of operation would depend upon the number of fillings per day. It is difficult to determine the proper number of fillings per day, as 462 SEWERAGE AND SEWAGE DISPOSAL this depends upon the quality and strength of the sewage, the quantity and character of the applied suspended matter, the size and character of the contact material, the extent to which the material is clogged, and the character of the efHuent to be produced. The smaller sewage treatment plants in the United States have rarely been looked after as their designers intended; they have been continued in service after their capacity was greatly exceeded, and the information to be obtained from their operating history is usually incomplete in consequence. For example, the five 0.25-acre contact beds, 4 ft. 9 in. deep, at Mans- field, Ohio, were put in service in 1902 when the sewage was weak and amounted to 765,000 gal. per day. It was found by the designer in 1914, 12 years later, that the plant was receiving 1,899,000 gal. of much stronger sewage daily. The automatic apparatus governing the flow from the septic tanks had not been looked after properly, at least during the later years, and the filling period of the beds was found to range from 3}4 to 83^ hr. But in spite of these conditions the beds were not clogged with mud at the bottom, a frequentlj' observed condition in overworked contact beds, and the air space in the contact material was about 30 per cent, so that Barbour recommended after the examination the continuation in service of the beds, but at a rate not exceeding 600,000 gal. per acre daily. The Plainfield contact beds were under the control of trained operators from the outset of their service in 1901. After 1 year operation, under excessive loads during a part of this period, Lanphear reported' that the effluent of the secondary beds was non-putrescible from October to May inclusive, and putrescible from June to September inclusive. The beds at Manchester, England, were dosed at the rate of about 480,000 gal. per acre per day during the first few years of their operation. Before the secondary beds were finished, the effluents from the primary beds were reported to be putrescible on account of the suspended matter in them. The rate of opera- tion had been increased to about 649,000 gal. per acre daily, which is greater than that contemplated by the designers and was reduced a little later to 574,000. gal. By that time it was found that the beds needed washing after about 2,700 fillings. In 1909-1 1 a large number of secondary beds were put in operation and the rate of application of the sewage to the primary beds was 1 Eng. Bee, July 1, 19H. CONTACT BEDS 463 increased to 713,000 gal.; later rates of these beds were 661,000, 576,000, 567,600 and 612,000 gal. By 1914 many of the beds had been renewed a second time, and it was reported that the number of fillings between the first and second renewals had been from 4,786 to 6,404. The water capacity of a i.^-acre bed 3 ft. 4 in. deep ranged from 216,000 gal. when new to 108,000 gal. after 2,250 fillings. At that time part of the effluent from the primary beds was treated on secondary beds at'the rate of 18 finings of the latter per week. Although the plant had been under the operating charge of specialists from the outset, the results were officially reported in 1914 to be unsatisfactory; part of the difficulty was apparently due to insufficient available head for operation. The effect of different operating conditions on similar con- tact beds was studied at the Lawrence Experiment Station by means of eight parallel experiments with beds 33 in. deep, filled with cinders }i to 13^^ in. in size and dosed with settled sewage. The general conditions and results of the tests, started in 1911 and ended in 1915, are given in Table 53. Clark's conclusions' from the results are as follows: " The differences in the degree of purification due to the different methods of operation are relatively small. The longer period of contact tends to produce a more stable effluent but with greater loss of open space. The number of fillings daily has little effect on stability. With three fillings daily there is a greater proportional decrease of open space than with two fillings daily. Filling from the bottom instead of the top decreases the stability of the effluent but gives the least loss of open space. Briefly sum- med up, these results indicated that for conditions similar to those in these experiments, the best results will be obtained by filling twice daily from the top with as much aeration as practicable and a period of contact somewhat longer than an hour." The results obtained with double-contact beds depend upon all those factors affecting single-contact beds and also upon the relative loads the primary and secondary beds are called upon to carry. Some data are furnished by a contact bed. No. 443, placed in operation at the Lawrence Experiment Station in 1912. It was dosed twice daily with the effluent from bed 425, see Table 53, and allowed to stand full 1 hr. The bed had 21 in. of broken stone, all of which passed a 3'^-in. screen, 43 per cent passed a )-^-in. screen and practically none passed a ^ "Experiments upon the Purification of Sewage and Water at the Lawrence Experiment Station," 1915, p. 31, 464 SEWERAGE AND SEWAGE DISPOSAL O Z m OS m K o O b O Z O H "5 lO 00 C ^ O O OJ ^ O U3 W 00 O t* 0> CO rt< O O CD iTD CD CD CO t^ |> (O (^ CO GO O) Ol O to i-H i-H iH W i-H 03 iH rt cq N. rH (N W Cq rH O CT) fH t^ CO .-I d d .-I cq TPCOt»(OiO 00«JrHC0O i-I Cq CO* « IN >H N Tjl IC 00 rH O) ITD CTl t^ M eq lo d d ssssg 00 CO 00 lO o ■* CO to CO CO lO M 4-- O O lO C3S OS t> OS O CO O CD O b- (N o OS m (?1 >0 Tt< tH iO OS b- to CO t* .H eq OS OS CO ■* »o ^ i-H OS CJl ^ (N ^^ '^ ID Qi OS OS OS OS rH (M CO Tj( Oi Oi Oi Oi OS ^ IN CO -^f »0 OS Os OS OS OS .3.3 .3 a a _n a a a a ^fl _g _fl fl .2 .5 .5 .S .S .2 o o o o o g o o o o 00 »-< CO O ID CO t^ OS d »d" 00 b- ^- CO ira CO CO (N r ">;(*" Csf b. |> Is. CD IN CO CO CNl N W o o o o p o o o o o o I-H CO o ^» r-T ^ -( CD ■* O ■^fflXiOO b-r*TtfU5iO CO t- 00 CO M •* CO © O CO iH co' ui t*' 00 •H i-( rf< CO ts! d rH cj ■^* -^ M ua O) »0 w i-t M CO "* lO 1-t rt iH i-H r-t O Ol O ffl 0> i-H 1-t 1-t iH iH .9 .2 .9 .S .3 go o o o _ o o o o 00 o t- o o i-T ifl 00 b-" ■^ f» CO Tt< N to to t^ ui m ■^ .-H N CO -^ lO iH 1-H rH iH tH 030)0 i-t iH iH rH i-t d d d d d iH N CO ■^ lO rH »-H 1-H i-H iH o o o o o -S .s .s .s .s o o o o o o o o o o ■^ -H N O ^ CD CD ■-jl' ^ «" O N M i-H 1-t CO ■* CO CO CO rH W CO Tti »0 iH IH ?H i-H iH o o o o o »H »H rH .-I r-l .S .S .S .S .9 O Q O O O o o o o o CO CD .-I O O i>r (sT ■^ o ^ 00 o o o o CO CO CO M CO a j= 30 466 SEWERAGE AND SEWAGE DISPOSAL Table 54. — Operating Rate and Charactee of Effluent of a Secon- dary Bed Tested at the Lawrence Experiment Station Year Rate, gallons per acre daily Ammonia Nitrates, parts per million Oxygen consumed, parts per million Samples putres- cible, per cent Loss of open space at end of year, per cent Free, parts per million Albuminoid, parts per million 1912 686,400 14.70 2.31 11.6 15.6 0.0 27 1913 548,700 12.58 3.04 15.2 20.0 0.0 25 1914 527,000 10.26 2.40 17.1 17.8 0.0 36 1915 455,700 4.60 1.50 16.1 13.1 0.0 32 1916 522,000 3.41 1.36 12.9 12.1 0.0 36 1917 433,000 2.96 2.38 10.6 21.1 11.0 40 3'^-in. screen. The operating results with this secondary bed are given in Table 54. During 1916, 96 per cent of the samples of the effluent applied to this bed were stable, but during 1917 only 63 per cent of the samples of the applied effluent were stable; the oxygen consumed of the applied effluent was 14.5 p. p.m. in 1916 while in 1917 it averaged 24.8 p. p.m. It is evident from these data and others given by Clark that the applied liquid in 1917 threw a heavier burden on the bed than in previous years. CHAPTER XVII TRICKLING FILTERS The limitations of the capacity of contact beds encouraged efforts to devise beds which would accomplish similar work without being subject to such restrictions. The result was the trickhng filter/ similar in construction to the contact bed, except that it is not necessarily built within a water-tight tank The filter material is placed upon a draining floor, through which the water can flow freely. The sewage is applied to the surface of the filter as uniformly as possible by sprinklers or other devices. Fig. 191 is a view of the trickling filter at Fitchburg, Mass., i :y' -^^' •'^•. ..-gfl ' 1 WSS^^-^ 'i:C'--''j,-.c, •■. ., , ■ - ,.,■.■' r--:..- ■■, ■■. "^ ^^^ m Fig. 191. — View of the trickling filter at Fitchburg, Mass. built from the plans of D. A. Hartwell and the authors. It is usual to apply the sewage for a few minutes and then shut it off for a short time, as 5-min. dosing and 10-min. resting. This construction offers as large an absorption surface as the contact bed. The sewage tricldes very slowly over the stones and has ample opportunity to give up colloidal and dissolved sub- stances to the bacterial jelly. The method of dosing is such that the bed may absorb oxygen continuously instead of during a brief period, as in the contact bed, and the danger of anaerobic conditions is greatly reduced. Experience has demonstrated 'Also called "percolating," "aerating," "sprinkling" or 'sprinkler" filter. 467 468 SEWERAGE AND SEWAGE DISPOSAL that usually trickling filters are self-cleansing, while the material of contact beds must be removed and cleaned occasionally. Theory.— The settling solids not removed by preliminary sedimentation are retained mechanically in the pores of the filter. The colloidal matter adheres to the surfaces through attraction and absorption, and the dissolved organic matter is absorbed by the bacterial jelly or acted upon directly, probably both processes occurring. The bacterial jelly coats the stones from top to bottom of a bed in mature condition and appears to per- form its functions in relays, that at the top acting upon so much of the organic matter as its capacity will permit, that of the next stratum below attacking the substances passing by the upper one or perhaps emitted from it only partly oxidized, and so on to the bottom, where the oxidized sewage falls through the floor grating into the underdrainage system. In the process of oxidation the organic matter is converted into humus-like material or resolved into carbon dioxide and other soluble and gaseous products, and the nitrogenous substances are oxidized with the production of ammonia, nitrites and nitrates. The processes are chiefly aerobic, for failure of the air supply is quickly reflected in the inferior quality of the effluent. The presence of vast numbers of higher forms of life, such as worms varying in size from those almost microscopic to earth worms 2 or 3 in. long, leads to the conviction that bacteria are not the only organisms laboring to transform the putrescible organic matter into substances less likely to cause offense. Unloading of Trickling Filters. — The periodic storage and disgorging of solids by the trickling filter is one of its most important functions. Being a community of innumerable living organisms, it is not surprising that it responds quickly to changes of temperature and other conditions. In summer, in the north- eastern states, the quantity of solids in the effluent is about the same .as in the applied sewage. Oxidation is then more active and nitrification will take place with doses not too great to permit efflcient oxidation. A trickling filter designed and operated for that purpose will convert a large portion of the organic nitrogen into nitrates. In the fall, the efficiency of oxidation drops progressively with the lowering of the temperature, and the proportion of solids in the effluent to those in the sewage is also reduced. During the winter oxidation is low and much organic matter applied to the filter is stored in it. But with the first TRICKLING FILTERS 469 warm weather of spring, the filter emits great quantities of solids, far in excess of those in the sewage applied at the time. Thus the stored matter, including the bacterial jelly which has served its purpose, is ejected from the filter, which thus recovers its capacity. Likewise, the renewed activity of the organisms produces more complete oxidation, and the quality of the effluent gradually improves until it equals that of previous summers. Results. — The trickling filter is capable of converting putres- cible settled sewage into stable effluent not subject to putrefaction under the most exacting conditions. The effluent contains much settling matter, especially during the spring unloading period. This should be removed from the effluent by sedimentation and the filter should be judged by the effluent thus settled. After sedimentation the effluent is usually more or less turbid and may be somewhat colored. There may be places where its quality would not be as good as the conditions require, and further purification or a substitute for trickling filters must be provided. A practical operating difficulty not yet overcome is irregular distribution of the sewage over the surface. The sewage passes more quickly through those portions of the filter which are overdosed than through the underdosed portion. On this account, the effluent is composed of portions of the applied sewage which have received varying degrees of oxidation rather than of sewage all of which has been oxidized to the degree apparently represented by the analysis of the effluent. The removal of bacteria at the Lawrence Experiment Station is probably considerably greater than that accomplished in practical sewage treatment plants. The following figures' give the relative efficiencies of the methods of treatment at Lawrence: Removal of Bacteria, Per Cent Single contact beds dosed with septic tank effluent 53-50 Single contact beds dosed with settled sewage 58-62 Double contact beds dosed with settled sewage 77-78 Trickling filter'' dosed with settled sewage 78-79 Trickling filter^ dosed with settled sewage 96-97 Trickling filters dosed with settled sewage; effluent settled and refiltered through intermittent sand filter 99 Intermittent sand filters 99 ' BepoH Mass. State Board of Health, 1908, p. 513. 2 This filter had 6 ft. of M- to 1-in. stone. 3 This filter had 10 ft. of J4- to 1-ln. atone. 470 SEWERAGE AND SEWAGE DISPOSAL Neither contact beds nor trickling filters alone will remove all bacteria, although the trickling filters appear more efficient than contact beds. In spite of the turbidity and color of the effluent from trickling filters, its large bacterial content and considerable residual organic matter, this type of filter is now the most generally practicable means of artificial oxidation of sewage. Though incapable of producing as pure an effluent as some other methods, its efficiency is such as to render it suitable for adoption in many municipalities requiring artificial sewage oxidation plants. Character of Filtering Medium. — The greater the surface afforded, the more efficient the filtering medium will be. Rough materials have the disadvantage that they tend to retard the unloading of transformed solid matters and are, therefore, more liable to become clogged than smoother materials. Clogging interferes with the natural passage of air into the filter and an abundant air supply is of the utmost importance. It may mean that a portion or all of the filtering material must be removed and washed, which adds greatly to the cost of operation. Under ordinary conditions, therefore, it will be well to avoid extremely rough materials like slag as well as very smooth materials like gravel. Certain kinds of materials disintegrate, so that the bed as a whole may in time come to be entirely different in character. In some climates the upper portion of the bed is liable to be alternately frozen and thawed during the winter, and if the material is porous there is a decided tendency toward disintegra- tion. Trap rock is a satisfactory material in this respect. Coke and cinders are undesirable. There may be places where it will be more economical to use cheap materials like cinders even though they must occasionally be renewed, than to build of more permanent but considerably more expensive material like broken stone. Size of Filtering Medium. — The smaller the filtering medium the greater will be its bacterial surface and the longer will be the time afforded the sewage for contact with this surface. The time of contact or passage through filters of different depths and sizes of ballast which was observed at the Lawrence Experiment Station is given in Table 55, but this time, it should be under- stood, relates only to the conditions of those particular experi- TRICKLING FILTERS 471 ments. Dunbar found* that if a quantity of sewage equal to the water-retaining capacity of a filter 3 ft. deep is applied to such a bed, 20 to 30 min. will elapse before that sewage is discharged from the bottom. Table 55. — Time op Passage op Sewage thbough Experimental Trickling Filters, Lawrence, Mass. (Report of the Massachusetts State Board of Health, 1908, page 386) Material Size of material, inches Depth Ave. daily rate' Time of passage of sewage, hours Nitrates, parts per million ft. in. Gal. per acre Gal. per cu. yd. Broken stone Broken stone Broken stone Broken stone Coarse clinker. . . . Coarse clinker. . . . Fine clinker Fine clinker M-1 10 10 8 5 5 9 3 10 5 9 3 10 1,150,000 1,938,000 963,400 925,500 885,000 949,000 896,500 908,900 71.3 120.1 74.6 114.7 95.4 153.5 96.6 147.1 3 2 2 2 6 1 14 6 29.5 20.5 21.2 5.2 12.4 7.3 22.0 5.0 ^ six days a week. Numerous experiments with trickling filters of different sizes of coal, gravel and clinker from refuse destructors were carried out by William CHfford, who concluded^ from the results he obtained that the time of percolation through clean filtering material varies inversely as the rate of sprinkling and directly as the amount of water taking part in the general water movement through the bed, the amount of water in motion being generally represented by the interstitial water. Assuming that a given sewage will require 100 min. of contact to become changed in character to the desired extent, he found that this contact time could be obtained by using any of the combinations of depth and size of filtering medium given in Table 56. His experiments were made with clean material instead of that which has been coated with bacterial jelly, as in practice, and for this reason some of the variations in depth given in the table are probably too great. American practice shows a tendency to use stone from 1 to 21^ in. in size. ^ "Sewage Treatment," p. 140. The water-retaining capacity of a soil is defined by Dunbar as the volume of water which remains in the soil after first drying the soil, filling it with water and allowing the excess water to drain away. 'Proe. Inst. C. E., vol. olxxii, part i, p. 289. 472 SEWERAGE AND SEWAGE DISPOSAL As small material should be used as is consistent with abundant air supply and freedom to unload from time to time the solids which have accumulated in the filter. Unloading is apparently facihtated by having the filtering material fairly uniform in size, for if the voids between the large stones are filled with finer stones the sewage solids may be retained indefinitely in the bed and cause clogging. In practice difficulty is often experienced in securing stone within the specified sizes, and unless fine material is definitely excluded at the crusher, the stone should be rescreened before it is placed in the filter. The character of the applied sewage must be considered in selecting the size of filtering medium. The more thoroughly the suspended matter is removed by preliminary treatment, the finer the filtering medium it is safe to choose. Generally sewage is fairly well settled before it is applied to trickling filters in order to reduce the danger of clogging to a minimum and to avoid placing upon the filter a burden of work which can be performed by less expensive structures like tanks. Table 56.- — Percolating Filters of Equivalent Capacity Based on A Time op Percolation of 100 Minutes (Clifford, Ptoc. Inst. C. E., volume clxxii, part i, page 291) Gravel Coal Slag Clinker Grade in. in. Depth ft. in. Grade in. in. Depth ft. in. Grade in. in. Depth ft. in. Grade in. in. Depth ft. in. 1-% yz-y^ 8 3 7 3 6 7 4 4 yi-% y^-Vs H-y^ 7 2 5 2 4 5 4 y^-% %-% Ys-y^ y^-y% 6 1 5 4 4 9 4 3 10 1-M 6 1 3 9 Depth of Trickling Filters. — The depth of a trickUng filter should be limited by the requirements of proper ventilation, for if the oxygen of the air passing through the filter is used up before the bottom is reached the lower part of the filter will be ineffective. For treating unusually strong sewages or industrial wastes having a marked avidity for oxygen, such filters should not be so deep as when ordinary domestic sewage is to be treated. Further- more, there will be a greater tendency toward surface clogging TRICKLING FILTERS 473 and consequent obstruction of the circulation of air in deep beds on account of the greater quantity of sewage applied per unit of superficial area. This condition is governed largely by the character and size of the filtering medium and the efficiency of preliminary clarification. Organic growths have appeared at times on some filters, more or less clogging the surface layer. Uneven distribution will more quickly result in clogging over- dosed areas in the case of deep beds because of the higher rate of application per unit of area. If portions of the filter area must be rested from time to time on account of surface clogging, a larger proportion of the filter will necessarily be out of use in the case of deep beds. In most cases where sufficient head is available, the cost of trickling filters per unit of volume decreases with the increase in depth, owing principally to the reduced cost of floor, under- drainage and distribution with deeper beds. If the quantity of sewage which can be satisfactorily purified by trickling filters is proportional to the depth, a subject which will be discussed later, the deeper filters will be the more economical. Estimates made by one of the authors, based on the contract prices for the tiickling filters at Fitchburg, Mass., showed that the cost per effective cubic yard of filtering medium was $4.54 in the case of 6 ft. depth, $4.20 with 7 ft. depth, $3.96 with 8 ft. and $3.60 with 10 ft. depth. It is impossible to ascertain the most economical depth of filter from the varying degrees of purification effected by differ- ent depths of the same filtering medium dosed at a uniform rate per acre. On the other hand, if similar filters of different depths are dosed with the same quantities of sewage per cubic yard of filtering medium, it is easy to determine the relation of depth to filter efficiency; or, still better, if similar filters of different depths are dosed with portions of the same sewage in such quantities as will permit of effluents of practically the same quality in all cases the respective capacities are evident. The relation of the rate of application per acre to the rate per cubic yard of filtering material is shown in Table 57. The danger of carrjfing too far the theory of equal loading per cubic yard lies in part in the greatly increased dose per unit of surface area, which may lead to surface clogging. This is illustrated by Table 68, from which it will be seen that very large doses are required for the deeper filters. 474 SEWERAGE AND SEWAGE DISPOSAL Table 57. — Equivalent Rates of Filtration pee Acre Expressed IN Gallons per Cubic Yard Gallons per acre Gallons per cubic yard for depths given in column headings 5 ft. 6 ft. 7 ft. Sft. 9 ft. 10 ft. 500,000 1,000,000 1,500,000 2,000,000 2,500,000 3,000,000 62 124 186 248 310 372 52 103 1.55 207 259 310 44 89 133 177 221 266 39 78 116 155 194 233 34 69 103 138 172 207 31 62 93 124 155 186 Table 58. — Equivalent Rates op Filtration in Gallons per Acre PER Day for Depths Given in First Column Depth, feet Dose, gallons per cubic yard 100 200 300 400 5 6 7 8 9 10 806,600 968,000 1,129,300 1,290,600 1,452,000 1,613,300 1,613,200 1,936,000 2,258,600 2,581,200 2,904,000 3,226,600 2,419,800 2,904,000 3,387,900 3,871,800 4,356,000 4,839,900 3,226,400 3,872,000 4,517,200 5,162,400 5,808,000 6,453,200 Gal. per acre- foot per day . . 161,330 322,660 483,990 645,320 Load, persons per acre-foot at 100 gal. per capita daily. . . 1,613 3,227 4,840 6,453 British Opinions on Depths. — The Royal Commission on Sewage Disposal reached the conclusion that a distinction should be made between trickling filters of coarse material and those using fine material. By coarse material the Commission meant material 3 in. in size and upward, medium-size material includes that from 3^ in. to 1 in. in size, and fine material is }/i in. in size. The Commission's conclusions'- read: ' Fifth Report, 1908, p. 87. TRICKLING FILTERS 475 1. "The deeper the filter the better the eflRuent. This holds both for fine and coarse material, assuming good distribution and aeration. 2. " For practical purposes and assuming good distribution, the same purifi- cation will be obtained from a given quantity of coarse material, whether it is arranged in the form of a deep or of a shallow percolating filter, if the volume of sewage liquor treated per cubic yard be the same in each case. Thus a filter 3 ft. deep, worked at the rate of x gal. per square yard per day, would give similar results to a filter 6 ft. deep, worked at the rate of 2x gal. per square yard. "In the absence of clogging, the balance would be slightly in favor of the deep filter, because the greater the depth the more the errors of distribution would be neutralized. There must always be a limit of shallowness (say 3 ft.), beyond which it would be unwise to go, for however good the distri- bution might be, if a coarse filter were constructed very shallow, a consider- able proportion of the sewage would pass through by 'short cuts.' 3. " With regard to filters of fine material, if the liquid to be purified were absolutely free from suspended and colloidal solids, and if thorough aeration could be maintained, the statement just made for filters of coarse material might possibly hold good for filters of fine material also. In practice, how- ever, these conditions can scarcely be maintained with large rates of flow and we think that the greatest efficiency can be got out of a given quantity of fine material by arranging it in the form of a shallow filter rather than a deep filter. But we are not in a position to make an exact quantitative statement as to the difference in efliciency of the two forms. "It is probable that for the treatment of a strong sewage liquor the filter should not be quite so shallow as for a weak liquor, but the point has not been fully worked out." Investigations at Lawrence Experiment Station. — After study- ing 16 years' operating records of broken stone trickling filters at the Lawrence Experiment Station, Clark reported^ that the maximum results in efficiency of purification and volume of sew- age purified were obtained when the filter is of considerable depth and constructed of comparatively fine material. In order to obtain more definite information four new filters, Nos. 452 to 455 inclusive, were started in 1913. They were filled with broken stone between M and 1}^ in. in size, what the Royal Commission termed medium-size stone, and were 4, 6, 8 and 10 ft. deep respectively. In 1915 four more trickling filters, Nos. 472 to 475 inclusive, were started, also 4, 6, 8 and 10 ft. in depth, but filled with stone having an average volume about six times (roughly 3 in. in size) that of the stone used in the first group of filters. These 1915 filters were of the Royal Commissions' "coarse" class. ' "Experiments upon the Purification of Sewage and Water, " 1914, p. 17. 476 SEWERAGE AND SEWAGE DISPOSAL The filters were operated at first at rates giving effluents containing 15 p. p.m. nitrates, which previous experience at the Station showed was necessary to insure stable effluents from trickling filters. Later changes in the character of the applied sewage resulted in a dropping of the nitrates to about 10 p.p.m., and still later to 7.5 to 10 p.p.m. Speaking generally, the shallow filters proved more troublesome to mature and the process took more time than was the case with the deep filters. In estimating the relative operating performance of these filters, Clark assumed that the amount of nitrification is at least an approximate measure for this purpose. The amount of oxygen absorbed on incubating samples of effluent for 24 hr. at 70°F. in closed bottles was determined at least once each month. On incubation part of the oxygen absorbed by the oxidation of the organic matter present is taken from the dis- solved oxygen and part from the nitrites and nitrates. Clark assumed that three-fourths of the nitrite, oxygen and five-sixths of the nitrate oxygen were available for this purpose. The results of these tests are given in Table 59. Clark pointed out' that the greater efficiency of deep filters is some- what more pronounced with material of the size used in filters 452 to 455 than with coarse filters. If it were practicable to have every trickling filter operated under the constant skilled supervision maintained at the Lawrence Station, then for the maintenance of high rates with a good degree of purification the material used in the first series of filters would be more desirable than that used in the coarse series. Clark reported the conclusion in 1917 that even with the coarse material used in filters 472 to 475 inclusive, which averaged 25 to 29J^ cu. cm. in size as against an average size of 3.7 to 5.3 cu. cm. in filters 452 to 455, a filter 10 ft. in depth can be operated at a rate per acre per foot depth approximately twice that of a filter 4 ft. deep. The practical importance of avoiding the use of such fine stone that a filter will become unserviceable through clogging is shown by the record of filter 135 at the Lawrence Experi- ment Station. This was 10 ft. deep and filled with broken stone from 34 to 1 in. in size. It was placed in service in 1899 and in spite of unusually careful management it soon showed a tendency to become clogged. During early years of operation, the clogging * "Experiments upon the Purification of Sewage and Water," 1917» p. 25. TRICKLING FILTERS 477 Table 59. — Opeeation of Thicklinq Filters of Medium-size and CoABSB Stone of Diffebent Depths at the Lawrence Experi- ment Station Gallons daily Total oxygen Filter number Depth, feet Year Samples stable, per cent ab- sorbed, parts Per acre Per acre per foot depth per million 452 4 1914 309,000 77,000 88 11.5 1915 466,400 117,000 94 11.5 1916 410,000 103,000 78 16.1 1917 287,000 72,000 84 31.5 453 6 1914 684,000 98,000 93 11.3 1915 805,500 134,000 100 11.7 1916 931,000 155,000 97 16.4 1917 780,000 130,000 74 34.3 454 8 1914 2,047,000 256,000 90 9.7 1915 1,767,250 221,000 100 9.9 1916 1,471,500 184,000 97 10.9 1917 1,400,000 175,000 89 27.3 455 10 1914 3,971,000 397,000 81 13.2 1915 3,501,000 350,000 88 10.9 1916 3,837,000 384,000 87 12.8 1917 3,250,000 325,000 68 29.6 472 4 1915 464,200 116,000 14 1916 339,300 85,000 37 16.1 1917 190,000 48 000 68 26.7 473 6 1915 616,200 103,000 71 1916 714,000 119,000 91 17.7 1917 720,000 120,000 48 26.7 474 8 1915 674,600 84,000 86 1916 1,004,000 125,000 80 15.7 1917 710,000 89,000 74 26.0 475 10 1915 738,100 74,000 100 1916 1,814,000 181,000 81 17.2 1917 1,755,000 176,000 79 24,2 478 SEWERAGE AND SEWAGE DISPOSAL could be removed by spading the surface and allowing it to rest for a short time. In 1917, however, its operation was very- unsatisfactory, although dosed at the low rate of 173,000 gal. per acre per foot depth. It was given a week's rest and on two occasions the top 3 in. of stone were removed, washed and replaced. The quahty of the effluent continued to deteriorate during that year and the filter had evidently become clogged throughout its depth. This experience with material of the Royal Commission's "medium-size" class shows the danger which attends attempts to secure high rates of operation by using rather fine material, and teaches the advisability of using a filtering medium which will unload freely, even if it is so coarse that a rather low rate of operation per acre per foot depth becomes necessary to maintain a stable effluent. An explanation of the greater efflciency of deep filters has been given ^ by Clark. It is based on the fact that in filters in operation there is a certain quantity of water retained in the interstices of the filtering material and held by the organic matter on the stones. He found that the deeper the trickling filter up to a certain point, the greater the tendency of the applied water to mingle with the retained water instead of merely passing by it. In other words, there was greater contact in deep filters than in shallow filters. Experiments were made with filters 452 to 455 inclusive, which gave the results recorded in Table 60. The experiments were made by adding a definite quantity of salt to the applied sewage and then determining the additional quantity of chlorine above the normal in the effluent every few Table 60. — Effect of Depth op Filter on Time op Contact op Sewage Passing theough the Filter Time in minutes for 50 per cent of the applied sewage Rate of appli- cation, gallons to reach the underdrains per acre Number 452, Number 453, Number 454, Number 455, 4 ft. deep 6 ft. deep 8 ft. deep 10 ft. deep 500,000 50 800,000 55 1,000,000 12 18 48 105 1,500,000 73 2,600,000 60 1 "Experiments upon the Purification of Sewage and Water," 1915, p. 22. TRICKLING FILTERS 479 minutes. It will be seen that the 10-ft. filter can be operated at a rate per acre per foot depth about twice that of the 4-ft. filter and maintain a contact period in each case of about an hour. Motion of Sewage in Bed. — Experiments by Taylor with filtering material of different sizes and depths showed that the lateral diffusion of percolating liquid was greater with fine than coarse material, but was not large in any case. With 2-in. crushed gneiss in a 6-ft. bed, 90 per cent of the liquid applied at the surface in drops was not diffused beyond the circumference of a circle of 7 in. radius with its center below the point of applica- tion. With 3'^-in. stone the diffusion in a 6-ft. bed did not exceed 12 in. from the center. Most of the lateral diffusion took place in the top foot of depth. Experiments with several points of application of the sewage to the same bed showed that where the cones of diffusion became joined, the increase in the quantity of liquid produced a streaming effect, which was much more marked with high than low rates of application. Other inves- tigations have confirmed these results. In practice accumulations of solids in filters cause some lateral movements of the sewage and may actually improve the distribution by their retarding effect upon the passage of the sewage through the bed. On the other hand, if the clogging causes pooling at different points, the sewage will be diverted to the cleaner areas and streaming will occur, seriously interfering with uniformity of distribution. Since little may be expected in the way of lateral distribution below the surface, the importance of uniformity of surface distribution is apparent. The ideal desired may be illustrated by a rainfall of sufficient intensity to produce the desired rate of application. A rate of 1,000,000 gal. per acre daily would be equivalent to a rainfall of about 1.5 in. per hour, or 36 in. per day. Revolving Distributors. — Traveling distributors are used extensively in Great Britain and at many of the Canadian plants to dose trickling filters. There are two general types, one con- sisting of two or four horizontal arms revolving about a central post and used to dose circular filters, and the other consisting of a low truss spanning a rectangular bed and moving back and forth on its side walls while distributing liquid over the bed in various ways, according to the make of the machine. The revolving distributor used for dosing two filters 30 ft. in diameter, shown in Fig. 192, is typical of small British appa- 480 SEWERAGE AND SEWAGE DISPOSAL ratus. There is a revolving distributor 130 ft. in diameter at Malvern and others from 70 to 120 ft. in diameter have been constructed for different places, but most installations have probably been under 70 ft. In the apparatus shown in Fig. 192 the weight is carried by a ball bearing at the top of the vertical Fig. 192. — Revolving distributors oi small English trickling filters. Fig. 193. — Traveling distributor of water-wheel type (Fiddian). standard. In some other makes the weights of the moving parts are carried by floats. The head lost in these revolving dis- tributors is very small. Traveling Distributors. — An entirely different principle is employed in the waterwheel type of distributor, which seems to be TRICKLING FILTERS 48r favored in Great Britain for dosing filters of fine material where particularly uniform distribution is desired. These distributors can be constructed in a revolving form or to travel back and forth as shown in Fig. 193. Either type is driven by the fall of liquid r* 6 — >^ 1 ecHonal ElevaHon. Fig. 195. — Taylor square-spray nozzle. filters at Springfield, Mo., built from the plans of Potter. They were adopted on account of the low-operating head, 12 in., which they required Dash Plates. — The gravity distributor used at the experiment station of the Massachusetts Institute of Technology discharged sewage from openings in the bottoms of troughs or from pipes on to concave metal discs placed beneath, which caused the sewage to splash upward and outward in the form of spray. These TRICKLING FILTERS 483 dash plates are used on 1.2 acres of filters at Mt. Vernon, N. Y. Fig. 194 is an example of this type of distributor designed by- Fountain for use at Calvert, Tex. This method of distribution is better suited for warm climates than for cold on account of the exposed distribution system. Spraying Nozzles. — Another method of distribution, originated at Salford, England, and employed almost exclusivelj'^ for dosing trickling filters in the United States, is to spray the liquid from a number of fixed nozzles. These nozzles are now made in several forms, some throwing a spray which covers an approximately square area and others dosing a circular area. A typical square spray nozzle, designed by Taylor, is shown in Fig. 195. It has an orifice 1 in. in diameter, through which a spindle passes, carry- ing a four-lobed, spreading cone, which is designed to throw the desired square-covering spray. The distributing cones of such nozzles must be kept in a definite position in order to spray con- tiguous square areas. These noz- zles are designed to concentrate the spray in a narrow zone and their success in use depends upon rapidly varying the head of the applied liquid. Taylor accom- plishes this by using a pressure- undulating valve on the main supplying the sewage to the bed. Taylor nozzles are also made to spray hexagonal and circular areas. A typical circular spraying nozzle is shown in Fig. 196, illustrating the nozzle developed at Worcester, Mass. It has at the base of the spraying cone a rim or lip, the function of which is to break up the sewage into a finer spray and spread it over a wider area. The spindle has a lock- ing device so that it can be quickly removed when cleaning the orifice. Efficiency of Fixed Nozzles. — The design of a nozzle should be such as to minimize clogging and permit rapid cleaning. Nozzles having small orifices generally produce a relatively fine 'Don:; Pipe Thread Supporting : ^-^— Locking Devki Arm-' Sectional ElevaHort, FiQ. 196. — Worcester type of nozzle. 484 SEWERAGE AND SEWAGE DISPOSAL spraj'-, somewhat greater absorption by the sewage of atmospheric oxygen, and a more uniform distribution, but the cost of keeping them clear may be a large item. If the orifice is so large that the sewage is delivered to the bed in a sheet, the streaming effect may prevent efficient purification. With a fine spray more odor and a greater reduction in the temperature of the sewage during cold weather are to be expected than with a coarse spray. Other conditions being equal, the nozzle which sprays the greatest area under a given head is the most economical. If this result is secured by a concentration of spray in a narrow ring, satisfactory distribution will depend upon a rapid variation in head during the discharge. The efficiency of practically all fixed nozzles may be greatly improved by a variation in head, because a comparatively large proportion of the discharge is concentrated upon a relatively narrow ring of the filter area if the head is constant. Methods of causing such variations are explained later in this chapter. A long series of careful tests was made at Fitchburg, Mass., under the direction of the authors, to determine the uniformity of distribution with various types of nozzle working under a proper variation in head. A battery of nozzles was used in order to determine the actual effect of overlapping sprays. Spacings of nozzles of 13 and 15 ft. and heads between 2 and 10 ft. were studied. The spray was collected in small compartments of pans; the pans had one-eighth of the area of the sprinkled surface in the tests of square-spray nozzles and one-twelfth of the area in the case of circular-spraying nozzles. Each test was con- ducted for an exact period of time at a definite working head and the quantities of liquid collected in each compartment were measured. The discharge of each nozzle under the different heads was metered and the effective head was determined in each case. The relative period of operation required at each head to produce the most even distribution was computed, and the distribution that could be effected by the nozzle with different ranges in head was determined on that basis. It was assumed that on all areas receiving between 873^ and 1123^ per cent of the mean rate, the distribution was good. For the underdosed areas receiving 75 to 87J^ per cent of the mean rate, the dis- tribution was considered fair, and also for overdosed areas receiving 112J^ to 125 per cent of the mean rate. The dis- TRICKLING FILTERS 485 tribution on the remaining areas was called poor, whether over- dosed or underdosed. A typical example of the distribution effected by the Taylor square spray nozzle is shown in Fig. 197. The marked' variation in distribution along the different radial lines from the nozzle is striking. An almost ideal distribution was given by the Wor- Vnshaded Atea Lies within the Limifs Assumed for fhe AilowabiS Variaiior? in fhe Iniensify of Dosing, from fhe Average Intensify on the Area of the Square Wefted; viz.(S7f 7o to liei% Of fheAyeraqe) and is 27.6% of. Total Area. Area of Portion iai/o fo B5% Overdosed =8.1% of Total Area l2i%*-> 25%Underdosed=l6.5%" " " •• ■"more than 2S% Overdosed =17.9%" " ' " V ZS%Underdosed=B9.9%" » Period of action of each head. 2,04 £t = 1.200 min. 3.12 " = 0.334 " 4.21 " = 1.100 " 5.29 " = 1.100 " 6.37 " = 1.160 '• 7.43 " = 1.100 " Total, 5.894 " Gallons desired per Cycle per se/.ff of Square Wefted. 0.659 FiQ. 197. — Estimated distribution in gallons per square foot per theoretical dos- ing cycle by Taylor aquare-spray nozzle, with %-in. orifice and 13-ft. spacing. The average operating heads range from 2.04 to 7.43 ft. and the period of action of each head is varied to produce an approximately uniform distribution in one cycle. A?eaTsq'tt ..'.■.■;:::;::::::::.:. 3.0 12.0 20.0 28.0 sa.o 44.0 25.0 les.o cester circular spray nozzle. An experimental dosing tank was constructed and further tests were made with it, with results shown in Fig. 198. While this distribution was not so good as that indicated by the theoretical dosing cycle, there was no area badly overdosed and the only area badly underdosed was adja- 486 SEWERAGE AND SEWAGE DISPOSAL cent to the nozzle. The latter defect would be remedied largely by a lower minimum head. In practice, there is likely to be a surplus of sewage in the distribution system before the siphon goes out of action, which will find its way on to this area. It is probable, also, that on account of the streaming effect produced Unshaded Area lie; within the Limits Mssumed fortheAllowable ^ Variation in the Intemity of Doling, from the Average Intensify g^; on the Area of the Hexagon Wetted = viz.(S7% to IIZitofAvenjae) = ri- and is 76.SS%of Total Area. ^' %£ Area of Portion IZi%toZ5%Overdosect'IIM%of Total Area- E^ n „ » I2^U'' Z9%Underdosecl'l.ll% n „ » ■fe i; V » » rnorethan2^/iiOverdosed= -% » » ; i> " i> A Z5%Underdoied=ll.'3Cf/ii» » « Kwwm^ pJ *6b 0SI9 '4b 0.SI4 *5a 0.5S4 *3i OSSl ^Z 0.071 A <*i 0.000 6a lions Desired ' psr Cycle per sq.ff. of Hexagon Wetted 0.414 Fig. 198. — Distribution accomplished by Fitchburg experimental dosing tank using Worcester nozzle with ijf g-in. orifice and 15-ft. spacing. The operating heads range from 2.00 to 7.96 ft. The quantities are in gallons per cycle per square foot. The time required to empty the tank was 5 min. 29 sec. Percentage of desired dose 75 87H 100 112M 125 Gallons per square foot 0.371 0.432 0.494 0.657 0.618 by a fixed nozzle under very low head, it would be unwise to apply as great a quantity of sewage to the area close to the nozzle as to the area more remote. TRICKLING FILTERS 487 Spacing of Nozzles. — Circular-spray nozzles which do not give over lapping sprays cover only 713^ per cent of a rectangular area when placed at the centers of contiguous square areas and only 90.1 per cent of the bed when arranged at the apices of equilateral triangles. Such a waste of filter area is impracticable. Accordingly the usual arrangement is to place the nozzles at the apices of equilateral triangles and provide for some overlapping of the sprays, as shown in Fig. 199. With all fixed nozzles, the maximum quantity of Uquid along a radial line under a given head does not fall upon the perimeter of the wetted area. Uniform distribution along any radial line necessitates overlapping the sprays, the extent varying with different nozzles and with different heads on the same nozzle. For example, in Fig. 199, Worcester nozzles having ^3d^6-iii- orifice are operating under a maximum head of 8 ft. They are 14 ft. apart at A, B, and C, the point o being 8 ft. from the nozzles. The spray is carried out 10 ft. from each nozzle, as shown by the arcs described about A, B and C. The extreme overlap beyond the point E, half way between nozzles, is 3 ft. The figure abc is the area of overlapping from all three nozzles. There is no overlapping of sprays around the edges of a filter dosed by fixed nozzles, and a smaller quantity of sewage per unit area reaches this portion. Assuming that the capacity of a triclding filter is approximately proportional to the depth, a smaller volume of filtering material may be provided around the edge of a filter, or, in other words, the bottom of the filter may be made smaller in area than the top, thus effecting a considerable saving in filtering material. For example, at Fitchburg the length and width of the filters at the bottom are 5 ft, less than at the top. Relative Advantages of Intermittent and Continuous Opera- tion. — It is necessary to vary the head on fixed nozzles in order to secure even distribution over the filter area. It foUows, there- fore, that no part of the area is in continuous operation, and all Fig. 199.- -Method of providing for over- lapping of sprays. 488 SEWERAGE AND SEWAGE DISPOSAL parts must be dosed at relatively high rates per unit of area for short intervals of time. The period of operation of fixed nozzles, from the time they go into operation at maximum head until they are again in operation at the same head, is called the dosing cycle. When a single-dosing tank is used, the nozzles are discharging nothing during the filling of the tank and this interval is known as the rest period of the cycle. At a given rate of operation, the rest period will depend upon the size of the dosing tank. The smaller the dosing tank the shorter will be the rest period and the greater 1.0 Z.0 3.0 4.0 SO' 6,0 Flowin Millions of 5allo7is psr Acre psr Odij FiQ. 200. — Curve of variations in length of rest period and number of cycles per day for various rates of flow of sewage. Based on performance of Worcester nozzle with iHe-in. orifice and 15-ft. spacing, work- ing under heads of 2.04 to 7.43 ft. on a work cycle of 6 minutes. the number of dosing cycles. A typical example of the relation of the rest period to the number of cycles at different rates of flow under specific conditions is shown in Fig. 200. Under these conditions there is no rest period above 4,800,000 gal. per acre per day, the nozzles going into continuous operation at that rate. The length of the rest period is hmited in cold climates by the danger of freezing the sewage in the exposed portion of the dis- tribution system. If the sewage is allowed to flow into the dosing tank while it is discharging, the sewage flow must be limited to the rate of discharge of the nozzles under minimum head. In many cases this quantity is less than the filter is capable of satis- TRICKLING FILTERS 489 factorily purifying. Provision should accordingly be made in such cases for storage of the flow during the discharge of the dosing tank or for an additional dosing tank. By the use of double dosing tanks, as at Fitchburg, one tank is filling while the other is emptying, thus making possible a much higher rate of operation of the filter under varying head and a shortening of the rest period. Other conditions being equal, the nearer the application of sewage approaches continuous distribution at the mean rate, the more efficient will be the purification. There appears to be some difference of opinion as to the value of periods of recuperation of longer duration than the rest periods occasioned by the dosing cycle. The experience of the authors has been that a recuperation period of 24 hr. or more is of great assistance at times in aiding trickling filters to unload accumu- lated solids. Similar resting in rotation of portions of filters affected by organic growths has sometimes proved of marked benefit. The length of the recuperation period will be governed by the extent of the clogging and the results obtained by actual operating tests. Distribution Systems. — Mains' for distributing sewage to fixed nozzles may be of concrete or vitrified pipe encased in concrete but the smaller sizes are preferably cast-iron pipe with lead joints. The pipes should be large enough to prevent undue friction and consequent loss of head at the nozzles. Economy of construction requires the lateral distributors to be relatively short to avoid the use of unnecessarily large pipe. The distributing laterals may be laid just above the filter floor, as at Columbus; nearer the surface, as at Baltimore; or at the surface, as at Fitchburg. They may rest on the filtering material with no other foundation, but if it should become nec- essary to remove the filtering material, the pipes would then have to be supported or temporarily removed. Where the dash plate or gravity nozzle is used, the distributing troughs or pipes are necessarily supported at an elevation above the surface of the bed. It is advisable in all cases to provide ample facilities for flushing and draining the distribution system. In certain cases fractures have occurred in lateral distributing pipes owing to excesssive pressures. If these pipes are laid at the bottom of the filter, repairs upon them will entail excavation of filtering material, which will add greatly to the cost of the repairs. Placing the laterals at or near the surface makes them 490 SEWERAGE AND SEWAGE DISPOSAL more accessible, but large pipe placed at the surface are likely to affect the uniformity of distribution somewhat and clogging is more likely to take place about the pipes than on the remainder of the area. In this case also, the sewage in the distributing pipes is exposed to low temperatures of a cold climate to a greater extent than when these pipes are below the surface of the bed. If the lateral distributors are at the surface of the filter, as at Fitchburg, the nozzles may be attached directly to the distribu- ting pipes. If the lateral distributors are placed beneath the surface, riser pipes will be necessary. The most common form of riser is cast-iron pipe jointed with bitumen or some other flexi- ble-joint material. The riser should be large enough to mini- mize friction, due allowance being made for a considerable deposit or growth on the inside of the pipe. Long risers are more liable to become bent or broken during construction than are shorter ones. The elevation of the top of the riser above the filter should ordinarily not exceed 6 in., although certain types of nozzle afford better distribution when the nozzle is placed 12 in. or more above the filter surface. The exposed portion of the riser is liable to cause trouble from freezing during the rest periods of the dosing cycle in cold climates. Underdrainage Systems. — Trickling filters, built in excavation without masonry floors, are objectionable because the soil may become displaced by water flowing over it, thus causing settle- ment. Furthermore, the soil may become intermingled with the filtering material and impair the drainage of the filter. Such filters should be built upon floors of concrete or other masonry. One of the earliest drainage systems consists of a concrete floor upon which is a layer of material composed of pieces pref- erably not less than 6 in. in diameter, the floor being sloped toward the main collectors. This coarse material affords better drainage than would the finer material of the bed, but clogging is liable to result from the solids unloaded by the filter. The coarse material accomplishes little in the purification of the sewage because of the relatively small bacterial surface per unit of filter volume. In order that the solids may be more readily carried away with the effluent or flushed through the drains, and the gases of decomposition may have a ready means of escape, false floor systems are generally adopted. The floor system used at Columbus and some other places. Fig. 201, is the least expensive type under some conditions. The TRICKLING FILTERS 491 492 SEWERAGE AND SEWAGE DISPOSAL concrete slopes toward the main drains and the lateral drains of inverted half-round, slotted tile practically cover the entire floor. They are bedded in cement to afford an even bearing, thus enabling them to carry the load of stone placed upon them. A layer of coarse filtering material about 4 to 6 in. in diameter is usually spread over the pipes to prevent the finer material from passing through the openings into the drains. This floor system offers a fairly good opportunity for the pas- sage of sohds from the bed into the drains. If the bed is com- posed of coke, cinders, or other material liable to disintegrate, the valleys between the pipes will probably become gradually flUed with fine material which may clog the drains. A disad- vantage of this floor is that the flow of effluent is distributed over the floor beneath the tile so that a relatively low velocity is afforded for carrying away the solids. Furthermore, it is not readily flushed because the water spreads over the bottom, rendering the stream ineffective. Fig. 201 also shows the floor system designed for Waterbury, Conn. The concrete is flat, sloping toward the main drains. A special block of vitrified clay or reinforced concrete is set on the floor, furnishing a false bottom over the entire area. The drain- age area is much greater than in the case of the Columbus type and the effluent flow is conflned to the drains. The effluent is allowed to spread out over a very large part of the floor, however, so that the self-cleansing qualities and facilities for flushing are not ideal. The Baltimore floor system Fig. 201, consists of a series of grooves and ridges in the concrete bottom. The grooves are covered with slotted vitrified clay or concrete slabs bedded in cement, concentrating the flow of effluent in narrow channels. The floor of the fllter slopes toward the main drains and the in- verts of the lateral drains are parallel with the floor, which ne- cessitates a varying depth of stone in accordance with the slope of the floor. While the practical objections to such a variation in depth of filter may be questionable, it is theoretically an ad- vantage to build the bed of uniform depth. With this type of floor system it is feasible, although possibly somewhat more ex- pensive, to lay the floor level and give the necessary grade to the drains. In this case the filtering medium will be of uniform depth. A type of floor designed by the authors and used at Fitchburg, Mass., is shown in Fig. 201. The lateral drains in the concrete TRICKLING FILTERS 493 are covered with narrow cement beams, affording a very large percentage of drainage area as well as a concentrated flow of efiiuent. Cobble stones, which were abundant at the site, were placed by hand over the openings before placing the crushed stone on the bed. It is evident that the thickness of the floor has an impor- tant bearing on the cost of the filter. The percentage of open area in the false floor for drainage and the proportion of free space for carrying away the effluent and gases of decomposition should be as large as can be provided at reasonable cost. The strength of the false floor should be ample for the load coming upon it. The possibihty that the drains may become clogged with soHds unloaded by the filter or by organic growths in them necessitates ample provision for flushing. In some cases the upper ends of the laterals are carried through the filter wall, affording an oppor- tunity for flushing, and in other cases flushing galleries have been provided. The size of the lateral drains should be suflicient to carry away the efl[luent promptly, to make flushing possible and to afford space for the circulation of air. The main drains may be rectangular or semicircular channels covered with slabs, or they may be circular. An open channel around the outside of the filter serving as a main drain. Fig. 192, affords an excellent means of ventilation. The grades of all parts of the drainage system should ensure freedom from backing- up of effluent in the ducts and channels. Ventilation. — Inasmuch as the proper action of the filter depends on an abundance of oxygen within the bed throughout its depth at all times, adequate ventilation is essential. Down- ward ventilation is secured by the natural flow of sewage through the bed. Any interference with the free passage of water and air, due to the accumulation of clogging matters or to insufficient drainage capacity, causes a decrease in the efficiency of the filter. Some filters have walls of large stones with openings for air to enter the beds laterally. In other filters, openings are left in the outer brick walls for the same purpose. While this may prove beneficial for small filters, like that shown in Fig. 192, its value for large units is doubtful. Any advantage gained during warm weather may be more than offset by decreased bacterial activity in cold weather, owing to lower temperatures produced in the air. At Gloversville, manholes were provided at the ends of the flushing galleries to aid in ventilating the filters. In several instances, vent pipes connected with the floor system and 494 SEWERAGE AND SEWAGE DISPOSAL extending above the filter surface have been provided at regular intervals. In a few cases ventilating cowls have been placed at the tops of these pipes. Filter Units. — The shape of the filters will be controlled in part by the topography of their site and to a large extent by the distribution scheme used. The size of circular filters with mechan- ical distributors is limited by the economical size of the dis- tributing device, and is influenced by the fact that it is less expen- sive to build a few large beds than many small ones. For filters using mechanical distributors more than one unit should be provided, so that in case repairs become necessary the entire plant need not be shut down. The Royal Commission on Sewage Disposal recommended not less than three units for each plant. There is a decided advantage in having small units, for they per- mit resting portions of the filter at intervals. A filter fed by fixed nozzles may be divided into as small units as desired. At Columbus, the 10 acres of trickling filters are divided into six units, each an equilateral triangle and all arranged about a central point. This arrangement is advantageous for dis- tribution but not for some forms of under drainage. It has proved unsatisfactory to have the administration building in the center of the plant, owing to obnoxious odors and flies. Preliminary Treatment. — Some form of preliminary treatment will ordinarily be required to prevent filter clogging. How far such treatment should be carried will depend upon the size of the filtering medium and the economy of construction and operation of the treatment plant. Washing clogged filtering material should be avoided if possible. Grit chambers and coarse' screens alone will ordinarily be insufficient. Fine screens have been advocated in certain instances, although it is doubtful if the solids which would pass such screens but would be removed by sedimentation-tanks can be as economically handled by filters as by preliminary tank treatment. If tanks are provided they should be designed and operated to take out all the settling solids it is practicable to remove. A disadvantage of the septic tank for preliminary treatment is that foul odors are often given off by the tank effluent. Spray- ing the effluent on trickling filters is particularly favorable to disseminating odors. Further there is a general belief that sewage can be purified by bacterial agencies more advantageously when fresh than when in a septic state. TRICKLING FILTERS 495 The Royal Commission on Sewage Disposal reached the following conclusion:^ "In the absence of special circumstances favoring a particular plan, it would appear that there is very httle difference in annual cost between the various methods of tank treatment followed by filtration through percolating filters, assuming that the kind of filter adopted in each case is that which is best adapted to the particular tank treatment provided." Sedimentation of Effluent. — Notwithstanding losses due to digestion as much suspended matter may be expected in the effluent as in the influent in some cases; indeed, colloidal matter is often precipitated within the filter so that the effluent averages higher in suspended matter than the influent in such cases. This difference may be further augmented by bits of organic growths, worms, and the Uke carried through the filter. In appearance, the suspended matter of trickling filter effluents is quite different from that in sewage. It is of a granular and gelatinous nature instead of the slimy, mucilaginous character of sewage sediment. The increase in stability of the suspended matter in trickling filter effluent over that in the apphed sewage was shown by an experiment at the Lawrence Experiment Station in which 0.2 g. of each sediment was mixed with 4,000 c.c. of river water saturated with oxygen (1908 Report, Mass. State Board Health, p. 376). The results are given as follows: Dissolved Oxygen at End of 5 Days; Pbbcbntage op Saturation River water 900 TrickKng filter 135 sediment and river water 100 Trickling filter 136 sediment and river water 76 . Sewage sediment and river water 1-5 It seems to be clearly estabUshed that the deposit from trickling filters is much more stable than that from sewage; yet it appears that the eflluent may carry, at times of unloading, such large quantities of this matter that the water is rendered putrescible thereby. Sludge resulting from the sedimentation of trickling filter effluents has usually been found to be putrescible. In some cases it has been rendered offensive by the presence of many decaying worms or by particles of organic growths. In the great majority of trickling filter installations it will be » Fifth Report, p. 43. 496 SEWERAGE AND SEWAGE DISPOSAL necessary to remove suspended matter from the effluent by tanks, strainers or filters. Much of the suspended matter lends itself readily to sedimentation, but there is considerable fine fiaky matter which does not settle readily. Quiescent sedimentation is required to remove a large percentage of this matter, and for complete removal, strainers or filters must be used. In the latter case it will probably be economical to pass the trickling filter effluent through sedimentation tanks first. Intermittent sand filters will not only remove this suspended matter but will afford further bacterial purification and may be operated at comparatively high rates, as at Gloversville, N. Y. Siphonic -dosing Apparatus. — Automatic air-lock siphons used in conjunction with specially designed dosing tanks afford a satisfactory means of regulating the head on pressure nozzles. The siphons for the purpose are similar to those used in dosing intermittent sand filters. The dosing tanks are used in order to obtain the necessary fiuctuation in head on the nozzles, so as to distribute the sewage as uniformly as possible over the bed. This head ranges from a maximum of 5 to 10 ft. down to a minimum of 1 or 2 ft. The usual form of dosing tank is hopper shaped or is tapered by either sloping side walls or steps, so that the capacity of the upper portion of the tank is much greater than that of the lower portion. This causes the nozzles to discharge longer under the higher heads and throws a larger proportion of the spray to the large outer areas. The size of the siphon should be governed by the maximum rate of discharge of the nozzles and the allowable loss in head by friction in the siphon at that rate. In general the loss of head at the maximum rate of discharge should be definitely limited and for average conditions 0.5 ft. may be all that can be allowed. In making such computations, allowance should be made for losses from the following sources: (1) friction in equivalent length of straight pipe due to roughness of surface; (2) head lost at en- trance; (3) head lost in changes of direction or bends; (4) head lost in sudden enlargements of the section; (6) head lost in sudden contraction of the section; (6) head lost at obstructions such as valves; (7) excessive surging due to impact of sudden discharge. The head lost in entrance may be materially reduced by using rounded orifices and bellmouth openings. Sudden changes in sectional area should be avoided as far as possible and where a change in section is necessary it should be a gradual one. TRICKLING FILTERS 497 An example of such a dosing installation is given in Fig. 202. The treatment works at Fitchburg, where this installation was made, were built to care not only for the normal daily sewage but also for excessive rates of flow during storms when considerable surface water is admitted to the sewers. The dosing apparatus is designed to shut off the inflow to one of the two dosing tanks while the siphon is discharging and allow the other tank to be filled, thus assuring the same size of dose and the same distribu- tion of the sewage on the filter at each dose, regardless of the rate P I a o FiQ. 202. — Dosing tanks and siphonio apparatus, Fitchburg^ Mass. of sewage flow, until a maximum storm-water rate is reached at which one of the tanks will go into continuous operation. By using two dosing tanks which operate alternately the point at which continuous operation begins is kept higher. The exact shape of the dosing tank to accompUsh uniform distribution with a given type of nozzle is largely a matter to be determined by experiment. There are a number of features entering into the design of a large tank which may not affect a small experimental tank. The following points should be considered in designing a large tank and proper allowances rnade for each: 32 498 SEWERAGE AND SEWAGE DISPOSAL 1. The capacity for each foot in depth required to give uniform distribu- tion of spray. 2. While the siphon is getting under way, the nozzles discharge under heads from zero to maximum head. The quantity thus discharged should be deducted from the respective volumes required at each foot of depth. 3. While the siphon is getting under way, the nozzles discharge under heads from zero to the maximum, thereby reducing the capacity of the dosing tank, drawing down the high-water level and resulting in a lower maximum head on the nozzles than would have been the case if the nozzle discharge did not lag behind the siphon discharge. This is largely due to the inertia of the great mass of water, which requires time to attain the maximum velocity. 4. Provide additional capacity above the maximum flow line for the capac- ity lost in filling the air vent or other spaces below the hydraulic grade line. The entire distribution system, so far as possible, should be below the mini- mum hydraulic grade line. 5. After the siphon sniffs, the nozzles will continue to discharge for a short period. This will affect the distribution at low heads. 6. Set the high-water line at the proper elevation to compensate for fric- tion losses in the siphon, distribution system and appurtenances. 7. The maximum high-water level has only slight effect, as it is imme- diately reduced by the draft from the tank. 8. When the siphon is discharging at low heads, the friction losses are greatly reduced and consequently more of the head is available at the nozzles. 9. If sewage is allowed to enter the tank continuously during discharge, the capacity of the tank at different depths and the distribution are affected according to the rate of sewage flow. Head-varying Valves. — In 1906 Stearns suggested using a mechanically operated butterfly valye in the delivery pipe to trickling filters as a means of varying the head on the nozzles, and about the same time Weand designed float-operated butterfly valves for use with the trickling filters at Reading, Pa. 6h&airGaf^ Rushing Gallen/ \ DisMbuHo/i j Kozzte- X'^'i"'^ / Butterfly Valye =il Itain-fo ^Ul lilteri Trickling Filters Fig. 203. — Sectional elevation of dosing apparatus, Baltimore. One of the largest filter plants employing this means of regu- lating head is that at Baltimore, where the head on the nozzles runs from almost zero to 9 or 10 ft. The installation is shown diagrammatically in Figs. 203 and 204. There are ten 18 by TRICKLING FILTERS 499 27-in. rectangular conduit castings, one to each filter bed, leading from the constant head chamber to the distribution system of the ,:-McHon Wheel , //Fricfion Disk, t'^ Fhor^fand // FloorSfand J Shaft connected fhrouq/i Worm and Wheel fo DC notor(mi!.!SO Kpm) iulomobile Chain to Operate Loner.. Voire yfeight to kerr> Friction conshnr y'Butterfly Voire Casting through which '^ fioivs -butterflq Valve Casting -through / nhii;h Menage porfs FiQ. 204. — Operating device for butterfly valve, Baltimore. beds. The opening to each of these castings is controlled by an electrically operated sluice gate. Each motor is connected to an auto-starter actuated by a trip on a float rod extending upward from a float in the outlet channel. The trips on these rods are set at four different levels so as to bring sections of the trickhng filters into operation one after the other by opening the motor- operated valves. For example, if two units are in operation and sewage begins to back up in the outlet channel faster than it can be taken care of by the area of trickling filters then in service, the float under the third auto-starter is raised sufficiently to trip the latter and open up a sluice gate on the inlet line to section 3 of the trickling filters. Inside of each rectangular casting above referred to is a flat iron plate swung on its horizontal axis by a shaft extending through a stuflSng box to the outside of the casting. This plate fits the inside dimensions of the casting with a clearance of about J^ in. The end of the shaft is geared to a 1-hp. direct-current motor. The plate is turned through 1 revolution (2 cycles) approxi- mately every 7 min., thus alternately damming and releasing the sewage flow to the nozzles. The spray on the filters under this action falls first on the stone immediately adjacent to the nozzle, gradually increasing its breadth to a distance of 7^ ft. away from the nozzle, at which moment the flat plate in the casting has turned till it is in a horizontal position and the maximum amount of sewage is flowing. Organic Growths and Insects. — Organic growths which ap- peared on trickhng filters at Dorking, England, were removed 500 SEWERAGE AND SEWAGE DISPOSAL without impairing the efficiency of the filters by applying a 20 per cent solution of caustic soda at the rate of 1 Uter per 1.6 sq. yd. A pinkish, slimy organic growth on the surface of several American trickling filters has been eliminated by applying a strong solution of copper sulphate or by breaking up the surface with a pick and then washing it with a hose. At Philadelphia, the growth was suppressed by a continuous disinfection of the influent, which aided in the purification and in no way interfered with the bacterial efficiency of the filter. In some cases the growths have disappeared after resting the beds for a few days and loosening their surface with a pick. Great quantities of worms develop in trickling filters, and are discharged with the accumulated organic matter when the filters "unload" these solids into the effluent. The abundance of these worms indicates that they must perform some function in the general transformation of organic matter into more stable matter which goes on inside the filters. A small, gray moth-fly is sometimes very troublesome about trickling filters, getting into the eyes, nostrils and ears of the filter attendants. At Gloversville, N. Y., they caused trouble at a tannery, 200 ft. from the filters, by lighting on skins while being tanned. Fly larvae, which clogged the upper part of a trickling filter at Andover, Mass., were removed by applying caustic soda at the rate of 1,000 lb. per acre. Neither copper sulphate nor bleach was wholly successful in killing these larvae. After the insects were killed, the clogging was removed by digging over the surface and allowing it to rest. At Plainfield, N. J., the fly maggots have been killed by closing the filter outlets and flooding the beds every seven days during the flybreeding season. The abundance of flies about these filters often leads to the development of large numbers of spiders, and many other kinds of insects have been observed to flourish at some of these plants. At Fitchburg an equilibrium appears to have been established between the spiders and the flies, and neither gains the ascendency to such an extent as to be troublesome. Comparison of Trickling Filters and Contact Beds. — The Royal Commission on Sewage Disposal made the following statements^ as a result of its investigations of these two methods of treatment: 1 Fifth Report, p. 231. TRICKLING FILTERS 501 "The amount of sewage which can be purified per cubic yard of contact bed or of percolating filter varies, within practical limits, nearly inversely as the strength of the liquor treated. This statement is based on the assumption that the size of the material of which the filter is composed is, in each case, suitable to the character of the Hquor treated, and that the material is arranged at the proper depth to secure maximum efficiency. Taking into account the gradual loss of capacity of contact beds, a cubic yard of material arranged in the form of a percolating filter will generally treat about twice as much tank liquor as a cubic yard of material in a contact bed. In the case of sewage containing substances which have an inhibitory effect upon the activity of micro-organisms, the working power per cubic yard of filter of either type may be more nearly equal. This point, however, is not clearly established. "Percolating filters are better adapted to variations of flow than contact beds. Effluents from percolating filters are usually much better aerated than effluents from contact beds, and, apart from suspended solids, are of a more uniform character. On emptying a contact bed, the first flush is usually much more impure than the average effluent from the bed. The risk of nuisance from smell is greater with percolating filters than with contact beds." The authors believe that with trickling filters of coarse material not subject to disintegration, the evidence seems to indicate that they wiU be self-cleansing if properly operated, whereas contact beds usually clog periodically. Hence the cost of treatment by trickling filters is usually much less than that by contact beds. The effluent from the trickling filters is ordinarily more highly nitrified than the efiluent from contact beds and after secondary sedimentation is more uniform in quality than contact bed effluent. The trickling filter is better adapted for variations in rates of flow than is the contact bed. The chief advantages in the use of contact beds rather than trickling filters are the rela- tively low head required, the somewhat simpler method of dosing, minimizing foul odors and avoiding a fly nuisance. CHAPTER XVIII INTERMITTENT SAND FILTERS If sewage is applied intermittently and uniformly over a bed of well-drained sand for a considerable period, two distinct changes in its character take place. The first change is the removal of the suspended matter by the straining action of the sand, and the second change is the oxidation of the organic matter, dissolved as well as suspended, by the bacteria Hving in the bed. These changes are so complete in a well-built and oper- ated intermittent sand filter that it furnishes, under the most favorable conditions, an effluent which may be said to be literally purified. It is not safe, however, for bacteriological reasons, to regard the effluent from any form of sewage treatment as suitable for drinking and generally such an effluent should be chlorinated or again filtered if it is to be discharged directly into water which may be used for domestic purposes. Mechanical Straining. — If the suspended matter in the applied sewage contains a considerable proportion of coarse particles they will soon form a thin compact mat on the sand, which intercepts the smaller suspended matter and retards the passage of the liquids into the sand. This mat must be removed to enable the sand filter to function properly. To avoid reduction in filtering capacity due to these mats and to the delays neces- sary for their removal, it is customary to pass the sewage through sedimentation basins before delivering it to these filters. The settled sewage still contains the fine suspended matter, however, which penetrates the sand to a depth of 1 in. or more. Therefore, while sedimentation reduces the frequency with which the surface mats must be removed it increases somewhat the frequency with which the dirty surface sand must be removed. The advantages of using settled sewage outweigh the disadvantage in most cases. The straining action of the sand leaves the suspended solids of the applied sewage in the upper part of the filter, where the organic matter is subject to biological action with its accompany- ing oxidation and liquefaction. The accumulation of solids is 502 INTERMITTENT SAND FILTERS 503 more rapid than the oxidation and liquefaction, and consequently the interstices of the sand gradually become filled with very small particles, some of which swell and become gelatinous when wet. This condition not only checks the passage of the sewage into the sand but also hinders the free circulation of air, a serious matter because ample oxygen must be supplied throughout the bed after each dose of sewage, in order that the desirable aerobic bacterial changes may take place. Oxidation. — In filters of this type in good operating condition, every particle of sewage comes into contact with the sand in such a way that part of the colloidal and dissolved matters may be thrown out of solution by attraction or absorption. These substances with the bacterial growths form a gelatinous film over each grain, which is apparently able to absorb oxygen at a rapid rate. When the bacterial conditions in this film have been established properly and the filter is not overdosed, the efiluent is much more mineralized and nitrified than the average efHuent from a trickling filter. To bring a sand filter to this condition will take from a week in hot weather to several months in cold weather. The formation of nitrates indicates the biological activity of sand filters. Nitrogenous matters are decomposed in the filters by the processes of either putrefaction or decay, the former occur- ring only when the filters are prevented by overdosing or other- wise from receiving an adequate supply of oxygen. These processes convert part of the organic nitrogen into free ammonia and part into nitrogen gas. If a filter has been jdelding a well- nitrified effluent and suddenly gives one with high free ammonia, it is probably being overdosed or the supply of air is checked by clogging of its pores. The increase in free ammonia may be due to reduced nitrification, or occasionally, to a reduction of the nitrates. Nitrification increases and decreases with the rise and fall of the temperature, while the inverse change takes place in the free ammonia content of the effluent. Filter Sand. — Where suitable sand or gravel is available in place, a filter bed can be constructed by grading its surface to receive sewage. Loam and silt have a tendency to hold water by capillarity, which prevents aerating a filter of such materials unless it is operated at a very low rate. Clay, cementitious sand and other relatively impervious materials are useless for inter- mittent filters. 504 SEWERAGE AND SEWAGE DISPOSAL Loam and subsoil must be removed if a filter is to be operated at more than a low rate. The extent to which subsoil is removed is usually determined by the cost of the work. Roots of trees and the fine earth about them must be removed, and the same is true of large stones. The aim should be to obtain a filter bed of uniform porosity, so that differences in the porosity will not cause overloading at places where the sewage passes through the sand most freely, or undesirable conditions in the bed due to lack of uniformity in the sand throughout its depth. An inter- mittent sand filter is not an inert mass, like a machine, but is the home of living organisms which do the work for which the bed is built. The better the home and the better the treatment these organisms receive, the better results they will give, as is the case with any other working colony. Effective Size and Uniformity Coefficient. — As a ready means of determining the comparative value of sands, the standard developed by Hazen when in the employ of the Massachusetts State Board of Health (1892 Report, p. 541), defining the "effec- tive size" and the "uniformity coefficient," has been found most useful. "As a result of experiments made at the Lawrence Experiment Station ■we have a standard by which we can definitely compare various sands. The size of a sand grain is uniformly taken as the diameter of a sphere of equal volume, regardless of its shape. As a result of nuinerous measure- ments of grains of Lawrence sands, it is found that when the diameter, as given above, is 1, the 3 axes of the grain, selecting the longest possible and taking the other two at right angles to it, are, on an average, 1.38, 1.05, and 0.69, respectively, and the mean diameter is equal to the cube root of their product. "It was also found that in mixed materials containing particles of various sizes, the water is forced to go around the larger particles and through the finer portions which occupy the intervening spaces, so that it is the finest portion which mainly determines the character of the sand for filtration. As a provisional basis which best accounts for the known facts, the size of grain such that 10 per cent by weight of the particles are smaller and 90 per cent larger than itself is considered to be the effective size. The size so calculated is uniformly referred to in speaking of the size of grain in this work. "Another important point in regard to a material is its degree of uni- formity, whether the particles are mainly of the same size or whether there is a great range in their diameters. This is shown by the uniformity coeffi- cient, a term used to designate the ratio of the size of grain which has 60 per cent of the sample finer than itself to the size which has 10 per cent finer than itself." INTERMITTENT SAND FILTERS 505 Mechanical Analysis of Sand. — Sets of sieves having several sizes of screens, as commercially obtainable, are usually made to nest one over the other, the coarsest on top. The nest is de- Mechanical Analysis of Sand for Total Per Cents by Weight Passing No,l2jl50-.4,'jfll_, ?0t§«Jun?i9iJ9|Q_ Total Per Cents bij Weight Passing a) o A) C *% e o 1 s ■fe c i c 1 1 to 2 o g U) g ^ s tTi' CJ o E E % c s +■ V c u- s E E s N u i> c 1 ■fe o E 1 « g «£> in o o «5 i ^ ^ 00 s fO rO s ^ o t M IT o o o 1 OO C> c3 i o 2 E ? >s « 0} > .s.3.a r ift^Saa" ■I _a ■a »' a S 3 ■o> -3 -•3 5 S OT IfN »^ w *^ 3 ( Eh woOOm .a§ '5§a 34j . ■fesS i^^Jt"S.a.a a:9§i|:1l^15 gi|1-oissa|o|gag •^Tf"g§a-?b-fp-ss^ ■aa-o'"T)SSMi§a'3?eg g S i « g g "'I'S.S g.S S g i.9 .-^ o ■3 a' So. J.3 ■a.S-a = la o^Sl I is .S^^ «^"s " 0) > w ti ^ a g CO d Si EbSjagu a g-oSaaa » CQ fl'^ n c3 a S 9^ a 9 fl o ^ A P ^ oj g] og CO n> d CT 10 CQ CQ O O aj-y O O O oo ICO m°^ : o7Y cq(N Mec ^^ I -00 S" f 0CO"(NO »ob-ai(NO MWOTjHt* OOO OOO OOO OOO O^OO OOO •OOO OOO r-(tO.-t M T-i OOO OOO OOO OO OO OO 0-* (DM OO OOO OO OOO OO OOO O COO 0U5 O OO »OMO iOOO«3r-lOO 0100O'<4^_0> (d'ocoo'c^ OJ 00 COt^O o> 00 -^coco O iH t^NTtf 0)10(0 t-iOCO »COM oo O rH U3 t* >* 0> 00 N« «3">0 2 II o j^ g (J a ■*> t,:3 o jh «^ OJ -oil f "I f^-^ ? .2 2'sS 5a « i^g 1?^ Is bO III III II I 11 jjl 6Sa tio..S l'°-9 E o .^ ■♦a ■♦a >i go's 6JS2 ir. _^ 0) 4) re) o fl 3 uto , a ^ +» a^l g-S« ■SS a 3 9^ Co S 3-■ '"~' ^r^l^^m 11^9 ^^^ ' .'^ -'J^Hf 1 ^9 V ^'^ ^^^'dV' ^ m. ^^^k '^■A mS^^^^^^^B ,' l^^^l ^^^HK^ti^^ l^njl ^9 Fig. 208. — Intermittent filter with sewage distributor, Marlboro, Mass. stand the pressure. As the doses of sewage must be applied quickly, 1 cu. ft. per second for each 5,000 sq. ft. of area, the pipe sizes must be liberal. The sewage is distributed over the beds by the following methods: 1. Graduated troughs running nearly the full length of the bed, as shown in Figs. 207 and 208. 2. Fan-shaped or arterial troughs, . used particularly for irregular-shaped beds. 33 514 SEWERAGE AND SEWAGE DISPOSAL 3. Quarter-point distribution, which consists of the discharge of sewage at the two quarter points on the long sides of the bed. 4. Corner distribution, in which the sewage is discharged on to the bed at or near its four corners. 5. One or two middle distributors, located upon the long side of the bed. This method, which was used for a time at Clinton, Mass., has been abandoned there as unsatisfactory. The discharge of sewage at these distribution points should be controlled in a manhole, from which the distributor branches, by means of a shear or sluice gate. Of these the simple shear gate is the cheaper and has been found on the whole to give reasonably satisfactory service. It is desirable to provide headwalls and a paved area at the point of discharge of the sewage to prevent erosion of the surface of the bed. Automatic Dosing Apparatus. — There are three classes of such apparatus used with intermittent sand filters : air-controlled siphonic apparatus, mechanically-controlled siphonic apparatus, and mechanical devices. Air-controlled Siphonic Apparatus. — The main feature is the siphon. Its principal parts. Fig. 209, are the main trap, a pipe casting with the long leg extending above the bottom of the dosing tank and the short leg connected to the discharge pipe to the filter; the bell, a cylindrical casting set over the long leg of the main trap and supported on legs or piers above the tank floor; the vent pipe, and the blbwoff trap, made up of small galvanized wrought-iron pipe. The main trap, immediately after the siphon has ceased dis- charging, stands full of water to the level Bi. The blowoff trap is also full to the level Di. The vent pipe is empty. Sewage flows into the dosing tank and the water level rises until the open end of the vent pipe is reached at A. The vent pipe then becomes full of water and the siphon is sealed against the escape of air confined in the bell and upper part of the long leg of the main trap. As the water in the dosing tank continues to rise, it exerts a pressure upon the air confined in the siphon and forces the water in the long leg of the main trap down toward the lower bend. The water in that portion of the blowoff trap under the bell is likewise forced down. At the same time the water level inside the bell rises. Just before the discharge level in the dosing tank is reached, the water level in the blowoff trap is at D^, in the INTERMITTENT SAND FILTERS 515 main trap at B2, and in the bell at C. A slight further rise of the water in the dosing tank forces the seal in the blowoff trap, thus releasing the air confined in the bell and causing a sudden inrush of water from the dosing tank into the bell, which sets the siphon into full operation. The sewage in the dosing tank is discharged through the siphon until the level is at the low-water line at the lower bend of the vent pipe, when air is drawn into the bell through the vent pipe, the siphonic action is broken, the bell is filled with air, the discharge ceases, and the main trap and blowoff trap are refilled with water. The dosing tank then fills again and the siphon is ready for another discharge. Fig. 209. — Siphon used in dosing intermittent filters (Miller). The air vent in the discharge pipe line, although not necessary for the working of the siphon, allows the escape of air previously confined in the bell and prevents trouble from air in the pipe line. The small siphons, 3 to 8 in. in diameter, in some cases do not require blowoff traps to ensure their working. Where it is desirable to dose two or more filter beds in rotation, this can be done by instalUng several siphons, each connected to a filter bed and arranged to discharge in rotation automatically. Two siphons of the type illustrated by Fig. 209, set side by side in a dosing tank, will operate alternately without special piping. For three or more alternating siphons, a special system of piping with starting bells or other controUing devices, is required. E. G. Bradbury gives^ the following formula as furnishing 1 Revort Ohio, Eng. Soc, 1910, p. 79. 516 SEWERAGE AND SEWAGE DISPOSAL a close estimate of the discharge of siphons under varjdng heads : Q = OAA\/%h where Q = discharge in cubic feet per second. A = area of discharge pipe in square feet. h = average head in feet or half the vertical height from high water in tank to center of outlet. Mechanically-controlled Siphonic Apparatus. — A variety of mechanical devices are used for directing the discharge of ./ \. ' ■'■ \ V^. \OuHet 7 ^v \ — ''' \ 12 3 I 1 1 1— s e ^f<— 1915 213. — Effect of temperature on results of intermittent filtration. be clear, substantially free from suspended matter and practically odorless. It will remain stable indefinitely, even when allowed to stand in a tightly stoppered bottle in a warm room. The eflSciency of various sand filters in Massachusetts is shown in Table 62. These results are for the most part yearly averages of monthly samples. It will be seen that there is 520 SEWERAGE AND SEWAGE DISPOSAL S m - EJ ft I- "s a I Eh •s 0) O »0 CC b- M O N CO O O CO lO CD OS b- n o OS OS to O O W CO a O 00 O (N rt< •* CO h- o o O CO 00 t- »0 CO 00 OS to OS to b* 00 X 00 O W lO M 00 CD CD lO u: .9 1 H "^ N ■a o o •s? 1-. M M iH 00 0> to 00 CO 00 « CO CO l> o N 1> w rt N CO CO OS Tjl p.? CO to N U3 CO O) CO t^ lO CO lO CO OS OS 00 »0 CO CO US' CO -< b- N ^^ M cq en OS CO cj> I> o to C<1 O >-l tH M lO N OS 0> «H 00 CD -^ « lO -^ iz; 2 1-1 tH g-S "S 00 00 00 00 a OS OS t- t- OC 00 00 c» OS o; OS 00 w w i ii< ►• 1 cq -* CO o o T-H O M « OS us OS 00 ^ OS to 00 O W CO ^ CO o c^ u e O O »« « tH to CO b- 00 1 H •a „ '3 oj C<5 00 OS iCi M CO iH N O OS b- 00 »0 « 00 ,H t> 00 »0 »0 CO OS b- 00 §'» ag lO CD (M CD Tt lO W -^ U5 10 << a »-l tH iH y-i ""* .5 ^■2 g 1 Ph •- b- « i-< 00 0] 03 00 O OS O) OS OS t> 00 00 00 00 OS OS OS 1 lO iH Ol (O (M O iH N to N « t- <0 OS l> i> to « ■* l> O t- r-J ■* 00 00 Tji M O 5 lO ^ rH CD i-l OS CO CO CO U3 CO CO t>- to "3 m O --< tH o O O tH iH W t-t o o o o o tH O O O O O O O --i ■ s H u » 3 o .S M 00 '^ CO o» tc C? «0 -* tH ,- iH « fr- iH IC lO ^ 00 tH ^ II M 00 CO CO to IC b- l> t* C ^> CO lo CO (X <0 00 rH to t" eq « 00 eq h g'^ *7. 1-H Tt< CO CO « N CO »H CO p; 1-H Tti o M a CO O lO T-( CO ^ t* t^ o I> CO lO 00 00 00 00 OS 00 CD 00 t- OS 00 to CM ■- s ^ CO Ol 00 CO lO -<}< CO CO ^ 1-H 1> O O OS «3 to CO CD OS i-H -* t> rt r- y !>. CI ^ CD M b- -# rH CO (N O « iO CO rt «3 TfH tH tH o 00 o w »0 M OO (N b. !>. r- lO ■* cq lO lO t^ 00 .H iH -^ so OS »0 W3 (M tH OS CO t^ O ^ O N CD CO 1-1 CD IN CO CO CO O CD >0 ■* ■* -fl CO rH IQ CO C^ CO ■* tH pq It; ■o! S g £ 3 B Si, ^ ^ 3 & 3 i < t c < c c 1 1 't C 1 •St -3 p a X £ « a c t 1 C ] 1 1 £ c -c T o! § 1 2 1 o c ^1 Is i ■g o s ■fill 8 S S INTERMITTENT SAND FILTERS 521 considerable difference in the quality of the sewage applied as well as in the effluents. It is probable that a majority of the analyses of applied sewages represent more accurately the stronger day sewage. Filtration does not affect the quantity of chlorine, so that the chlorine results may be used to determine whether the effluent probably corresponds to the sewage analyzed. It will be seen that there are wide discrepancies, and these should be taken into account when studying the results. In some cases the effluent has doubtless been diluted with ground water. These results are not given with a view to showing the relative amount of work done at different plants, as other conditions must be taken into consideration, particularly the character and quantity of sewage applied. These results do show, however, in a general way, the purification accomplished in practice. The efficiency of sewage filters in a northern cUmate will not be so great during the winter months as during the remainder of the year. Fig. 213. Bacterial activity is reduced by fall in temperature. The large and irregular doses applied to the fil- ters in order to keep them thawed out and to obtain the best result during the cold weather are not productive of a high degree of purification. Odors. — Opinions differ widely as to the odor coming from intermittent sand filters. A characteristic sewage odor can be detected in their vicinity, particularly on wet days. The distance to which this odor is carried varies with conditions of wind and weather, as well as of the filter beds themselves. In some cases no objection has been raised from residents within 600 ft., more or less, of the filtering area, but it is probably desirable to keep the beds upward of a quarter of a mile away from any substantial settlement, and the larger the area the more desirable it is that the beds should lie at a considerable distance. The odor is not believed to have any serious effect upon health, for the men employed at the various sewage filter and broad irrigation plants are usually strong and healthy. OPERATION OF INTERMITTENT SAND FILTERS Clogging of Beds. — On beginning the use of a sewage filter bed the surface should be smooth and level. For the first few days of use uniform distribution of the sewage over the entire bed will not be obtained, for so much sewage will penetrate the sand near the distributor outlets that the dose will be exhausted before it has 522 SEWERAGE AND SEWAGE DISPOSAL an opportunity to cover the bed. Gradually the surface will become partially clogged, with marked improvement in uni- formity of distribution. The degree and nature of the clogging vary with the character of the suspended matter in the sewage applied.' If it contains a large quantity of coarse material, such substances will settle and accumulate on the surface of the sand, the coarsest in the immedi- ate vicinity of the distributor outlet and the finest at the farthest points from such outlet. With the lapse of time a surface mat of fibrous material will form, if raw sewage is applied. When allowed to dry out between doses, the mat separates from the surface of the sand and usually cracks and curls up, the rapidity of drying depending upon wind and weather conditions. Such a mat is beneficial in preventing the entrance of very fine substances into the body of the bed. On the other hand, it must be removed at frequent intervals to enable the bed to receive its proper quantity of sewage and the requisite air. The suspended matter in settled sewage, although somewhat fibrous, is so finely divided that it penetrates the sand more deeply than the coarser matters present in raw sewage. The mat formed by the finer particles is not strong and heavy and is not effective in preventing more finely divided matter from enter- ing the bed. The suspended matter in the efl[luents from the septic tank and from chemical precipitation is still more finely divided and less fibrous than that from plain sedimentation. Consequently, such matters penetrate the sand to an even greater depth than the suspended matter from sedimentation. The sus- pended matter in these effluents does not form a mat and, therefore, it is necessary, in order to remove it, to scrape off a substantial portion of the surface sand. The very finely divided silt washed from street surfaces in times of storm, if applied to sand filters, will cause serious surface clogging. Filters which are in good condition may be so clogged by the storm water from a single brief shower as to require scraping before they can be put into operation again. Rain has the effect of beating down the surface of the filters, which may require raking before they will resume their normal capacity. On the other hand, frost has a tendency to open the pores of the bed, enabling it to receive sewage much more freely than it did just prior to freezing. Beds which are considerably 1 Eddy and Fales, Jour. Aeaoc. Eng. Soc, vol. xxxvii, p. 67. INTERMITTENT SAND FILTERS 523 clogged may be made so porous by freezing that their capacity will be temporarily much increased, providing the frost does not go deeper than an inch or two. Objections to Disturbing Surface of Filters. — Harrowing and plowing may, under some circumstances, be absolutely necessary. There are, however, serious objections to this treatment of the filter, for the disturbing of the upper layers of the sand mixes the organic matter with the cleaner sand below. This organic matter is decomposed more or less, but it does not entirely dis- appear; on the contrary it changes to a sort of humus, which remains in the sand and has the effect of reducing its capacity. It is obvious, therefore, that a time will come in the life of the filter when this material will reach such an amount as to practi- cally clog the upper layer of sand and require its removal. It is undesirable to cause the organic matter to penetrate the filter any deeper than is absolutely necessary, because this necessi- tates the removal of a greater depth of sand. It is perhaps impossible to fix any definite rule for the manipu- lation of the surface of the filter, as there are doubtless cases where it would be as unwise to disturb the sand as it would be impossible to avoid it in others. The general principle, however, seems to be that care should be taken to avoid mixing the solids deposited by the sewage with the sand and, therefore, the beds should be thoroughly cleaned before harrowing or plowing. Removal of Dirty Sand. — The surface mat should be removed at frequent intervals and as the proportion of organic matter to sand in the upper layers increases, it is also necessary to remove a certain amount of this layer of sand, humus and other clogging material. The amount of this material to be removed depends upon the character of sewage which has been applied to the filter, and probably also upon the manipulation of the surface of the filter. But the chief factor appears to be the rate at which the filter is operated. If the load is light the natural biological processes will oxidize the organic matter nearly as fast as it is applied, but if the filter is forced such substances, more or less changed, will accumulate rapidly. The mat which is peeled from the surface of the filter and the material which is scraped in a moist condition from the sand contain a large amount of organic matter. The amount of nitrogen runs, as already shown, as high as 1.3 per cent of the dried sample. This material has some value as a fertilizer, and 524 SEWERAGE AND SEWAGE DISPOSAL may in some cases be entirely removed by local farmers, and if the quantity is not too great it is possible that a small revenue may be obtained from this source. The dirty sand which is removed contains a comparatively small proportion of nitrogenous matter, and seems to have no practical use. It must, therefore, be wasted at the least possible expense. Under some conditions it may be wise to wash this material, and replace it on the surface of the filters. This, however, is open to the objection that the finer particles will be removed, leaving a coarser-grain sand for the surface, which will allow the finely divided matter to pass through it, producing the effect of stratification. Resting. — When the bed is overtaxed or the aeration becomes inadequate, fouling begins at the bottom of the bed, on account of the lack of air. Such a condition should be avoided by careful attention to dosing and care of surface, but if it exists the bed must be thrown out of commission, and allowed to dry out so as to overcome the effect of capillarity and become charged with air. Bacterial activity will still continue for a considerable period of time under such conditions, so that the period of resting may vary within wide limits without injury. If the bed is in very bad condition a period of 4 to 8 weeks may be required for its recu- peration, while 1 or 2 weeks may suffice to cure less serious cases. Replacing Sand. — So far as can be learned but few intermittent sewage filtration plants have been in operation a sufl&cient length of time to make necessary the replacing of material scraped off during the process of cleaning, although some industrial plants have required such renewal. It is quite probable that filters which are used well up toward their maximum capacity must have so much material removed that ultimate replacement will be absolutely necessary. Thus at the end of an 11-year period of use, the removal of 6 in. of material proved necessary at Brock- ton, Mass., at a cost of $4,314 for 19 acres of beds. At Worcester it was found that on the beds taking effluent from chemical precipitation tanks, clogging due to change in the character of the surface material, caused by the depositing of lime, iron, and organic matter, necessitated the removal of about 4 in. of material 3 years after putting these filters into service. After that time an average of about 3.3 cu. yd. of material per 1,000,000 gal. of chemical effluent filtered were removed. In the case of the efflu- ent from the septic tanks, the quantity of material removed from INTERMITTENT SAND FILTERS 525 the beds was substantially the same as when the chemical effluent was applied, but the clogging material had to be removed at more frequent intervals. With the effluent from settling tanks the amount of material requiring removal was somewhat less, though the total amount, including the mat, was about the same. The quantity of mat and dirty sand removed from Wor- cester beds receiving raw sewage has been as high as 14 cu. yd. per 1,000,000 gal. sewage filtered. At Concord, Mass., where the sand is relatively fine and the sewage dilute, 9 in. of sand were removed from a 1-acre bed, out of a total of 3 , 3 acres of beds, at the end of the 20th year of service. Maintenance in Winter. — The warm, dry weather of the sum- mer is by far the best period of the year for the operation of intermittent filters. The spring is wet, and the filters are then generally overcoming the ill effects of the previous cold weather, so that the best work is not done by them until the late spring or early summer. The fall is apt to be cold and rainy, and such weather is not advantageous. The winter months are the most difficult in which to get good results, as shown by Fig. 213. The success of the filter during winter weather depends some- what upon chance. If the filters are in good condition, and a heavy snow storm occurs while the filter is flooded, the snow will be melted, and the water will readily find its way into the sand. On the other hand, if the filter is seriously clogged with suspended matter from sewage previously applied, and is very slow in filter- ing, the water will be chilled by the snow which falls into it, and if there should be a sharp decline in temperature it is quite prob- able that the snow and water will be frozen to a solid field of ice, which adheres tenaciously to the sand on to which it is frozen. If a considerable depth of snow falls when the filter is not in use, it has been found that when water is turned on it is immediately chilled, and in extreme weather it is quite hkely to freeze and make the filter substantially impermeable. On the other hand, if the filter is in fairly good condition, and paths are shoveled through the snow giving the warm sewage a chance to get to the center or extreme points of the bed without becoming too cold, it has been found possible to thaw a large fall of snow and still keep the filter in good condition. More care is required in dosing filters during the winter than at other seasons. Large doses of sewage must be applied to thaw 526 SEWERAGE AND SEWAGE DISPOSAL the frost in the sand or to melt the snow on the surface of the filter. On the other hand, it is essential that the dose be not large enough to freeze to the sand before it can find its way into the filter. Various methods have been devised for preventing the ice from freezing to the sand. One of the most successful of these is to furrow the filters for use in winter. The furrows may consist of ridges and depressions about 3 ft. on centers and about 10 in. deep. When ice has formed on the sewage, the ridges hold it up as the sewage recedes, leaving a space between the bottom of the furrow and the ice, through which future applications of sewage can run. This covering of ice also protects the surface of the bed from the extreme cold, preventing serious freezing, and serves to keep succeeding snowfalls out of the sewage. Winter Cleaning. — When sewage is applied to furrowed filters the suspended matter tends to settle in the bottom of the fur- rows, which afford a much smaller area for the distribution of such matter than the level surface of the bed at other seasons. Where beds are operated at nearly their maximum capacity, especially with sewage containing large quantities of suspended matter, the winter clogging may be serious. It is important, therefore, to take advantage of any opportunity afforded for winter cleaning. In the course of nearly every winter in Central Massachusetts there is at least one period lasting from one to sev- eral days, during which the temperature conditions are favorable for cleaning. If at such time the accumulation of ice does not prevent, it is desirable to remove as much of the clogging accumu- lation as possible. The work must be done expeditiously, however, as such periods are usually of short duration. It has been found that the mat can be readily raised from the sand if it is barely frozen. If, however, it is frozen hard, the crust will be so thick and the quantity of sand clinging to the mat so great, that cleaning will be impracticable. CHAPTER XIX DISINFECTION OF SEWAGE AND SEWAGE EFFLUENTS The term "disinfection " is used here to designate the treatment of sewage or contaminated water so as to reduce greatly the number of bacteria in it. Incidentally this treatment may deodorize the liquid more or less and prevent or retard putrefaction. By "sterihzation" is meant the destruction of all organisms. In the methods of treating sewage previously described, removal of suspended matter and change in the character of the organic matter originally present are the main objects sought, whereas in disinfection and sterilization the main object is to kill the bac- teria in the Hquids treated. It is unusual for a water receiving sewage or sewage efHuents to be used for domestic consumption without previous treatment, and with modern facilities for purifying and disinfecting a water supply, it is generally wiser to apply the corrective measures to the water supply than to the sewage. There are conditions, however, under which it is desirable to disinfect the sewage even though the water supply be effectively treated. The danger of contracting typhoid fever on the bathing beaches through polluted water and by eating sewage- contaminated oysters, also justify the disinfection of the sewage of some communities. Disinfection by Acids. — Most acids act as germicides under certain conditions. Where sewage contains spent acids from industries, the number of bacteria are usually relatively small. Mine drainage often contains large quantities of acid and iron, which keep down the bacteria as shown in the following figures of conditions in the Monongahela River, furnished by Hazen and Whipple : Acidity, parts per miUion 35 30 25 20 15 10 5 Alkalinity, parts per million 5 Bacteria per centimeter 150 280 360 500 750 1,400 2,800 7,500 30,000 The cost of disinfection by acids has prevented the considera- tion of this treatment except for emergency use and where it is incidental to treatment with acid for recovery of grease. Disinfection by Copper Sulphate. — This treatment is employed mainly in killing algse in water suppHes, but has been used occa- 527 528 SEWERAGE AND SEWAGE DISPOSAL sionally in killing troublesome growths of organisms in sewage filters. The quantity of copper sulphate needed ranges from 0.07 to 10 p.p.m., according to the species to be killed, and it is accordingly necessary to determine the organism which must be removed^ before the dose can be chosen with certainty. It is important to keep in mind that this chemical affects fish, some species being more resistant than others. Kellerman reports that the following amounts of copper sulphate, in parts per million, must not be exceeded if it is desired to save the species of fish mentioned: black bass, 2.1; carp, 0.3; catfish, 0.4; goldfish 0.5; perch, 0.75; pickerel, 0.4; sucker 0.3; sunfish, 1.2; trout, 0.14. The copper sulphate is placed in a gunny sack which is dragged slowly by a rowboat back and forth over the pond, keeping the boat on each trip within 10 to 20 ft. of the previous path. In this way about 100 lb. of the chemical can be distributed in an hour. Good results have been obtained with copper sulphate treat- ment of settled sewage and the efHuents from septic tanks and intermittent sand filters, but the cost has been greater than that of disinfection by chlorine and its action is not so rapid as chlorination and more affected by temperature. Disinfection by Chlorine and its Compounds. — Investigations by Phelps^ indicate that gaseous chlorine and hypochlorites are good germicides, but chlorates and perchlorates have little value as disinfectants. These substances, like ozone, are disinfectants because of their oxidizing effect. They release oxygen in a nas- cent state, which instantly combines with the most susceptible substances at hand, among which are the organic structures of the bacteria. The reactions taking place result in the production of hypochlorous acid, HCIO, which is very unstable and is at once dissociated into hydrochloric acid and oxygen. Each molecule of hypochlorous acid contains one atom of chlorine and one of oxygen, and as an atom of oxygen has twice the combining power of the chlorine atom, it follows that the oxidizing power of the compound is twice as great as its chlorine content. Thus, if a sample of bleaching powder actually contains 20 per cent of chlorine in the form of calcium hypochlorite, its potency is double this and it is said to contain 40 per cent "avail- able chlorine," though in reality it does not. The words "avail- able chlorine" simply constitute a term by which is expressed 1 "American Sewerage Practice," voLiii, p.745- ^. s Water Supply Paper 229, U. S. Geological Survey, p. 60. DISINFECTION OF SEWAGE AND SEWAGE EFFLUENTS 529 the power of the compound to oxidize or disinfect. The quantity of chlorine used is expressed in three different ways, the equiva- lents being shown in Table 63. Table 63. — Parts of Available Chlorine, Equivalbnt Pounds op Liquid Chlorine and Equivalent Pounds op Calcium Hypochlorite ( Calcium hypochlorite assumed to contain 33 J^ per cent available chlorine) Parts available chlor- Pounds liquid chlorine Pounds calcium hypo- ine per 1,000,000 gal. per 1,000,000 gal. chlorite per 1,000,000 gal. 0.04 0.33 1.0 0.12 1.0 3.3 0.5 4.2 12.5 1.0 8.3 25.0 5.0 41.5 125.0 10.0 83.0 250.0 25.0 207.5 625.0 The effect of a given quantity of these disinfectants upon sewage depends to some extent upon the size of particles and condition of the organic matter, for apparently the time of con- tact necessary for the full effect of the disinfectant to be developed varies with the character and quantity of the organic matter. The germicidal effect of the disinfectants is apparently the same where an opportunity for complete action is afforded. The quantity of disinfectant required to treat sewage and the effluents from sewage treatment works is many times greater than that required to disinfect water supplies. The effluent from a well-designed, built and operated intermittent sand filter, not overloaded, requires the least quantity of disinfectant and septic tank effluents generally require the largest quantity. Table 64 gives the results of tests by Clark and Gage to ascertain the quantities of effective chlorine required to reduce the bacterial content of sewage and effluents to different standards. Water suitable for drinking in Massachusetts usually contains less than 100 bacteria per cubic centimeter, determined at room tempera- ture. The total bacteria developing on litmus lactose agar at body temperature is usually less than 10 per cubic centimeter, and the red colonies usually less than 5 per cubic centimeter. This they called the 100-10-5 or drinking water standard. Two 34 530 SEWERAGE AND SEWAGE DISPOSAL s d J3 « H r ^, p. P3 « R o> P *"* Q tj W y M rt g -o J3 n - M '^ .H CO CO l> CO CO w CO " CO Tjl CO " CO o il ■«o >0 00 t- MOO 00 ID 00 CO CO 00 CO >D 00 ^ 00 00 il W CO -^ T-( CO to t^ CO CD rn CO ID b- CO ■^ CO "■ CO hS 1 ^ i -^ 1 1 1 I iH pa. M 00 to »D 00 OJ CO 00 M .-1 M CD -* 1 lO CO t* ^ 1 b- CO CO 1 t^ CO ■* 1 CO CO CO 1 •go M 00 CO CO 00 CO CO 00 >D CO 00 o »D 00 ^ M 00 T-H « t- tD CO w tH CO t» '^ 1 i-H 1 "J-o S3 o & Is CO M "O <-I CO l> CO 00 o CD CO "O CO »D OC rH t^ 00 »D M M (N CO l> 00 00 1> 00 CO ID « 00 CO "■ CO Eh'^ « ^ §m ug CO M CO O 00 CO ID 00 CO CO 00 t> ID 00 ■* 00 00 gs .H CO CO ID CO 00 - 1 b- CO CD 1 .H CO »D '^ 1 t* CO -* 1 CO '^ CO 1 lO 00 CO U5 ID O »D >D ID ID CO ID 00 00 Ttl 00 Si "S.o ts. CO r-l t^ l> tH r- t- !> b- .H ID CO CO o 00 ^ 00 rt CO CO CO CO .-H W in -a + + + + + + 1?- VO lO »C ID O CO ID ID ID ID O 00 ID 00 O) 00 & ^1 O-H CO w CO CO c: CO N ^l + + + + + + + do 00 00 ■* W 00 o n 00 00 CO CO w 00 CO OS 00 CO §3 00 M Oi 00 CO o> rt CO 00 OO CO CO CO CO CD CO "^ CO r-l ""* CO 1 (y »0 CO I^ ID ID ID »D ID ID ID ID ID ID »D ID ID 00 M 'C t^ b- t> h- CO r- n C CQ m DQ O H m DISINFECTION OF SEWAGE AND SEWAGE EFFLUENTS 531 other standards containing respectively 10 and 100 times as many bacteria, called 1,000-100-50 and 10,000-1000-500 standards, were also assumed to correspond approximately to the upper and lower limits of bacterial counts on river water receiving more or less pollution. The tabulated results are all based on a period of contact of 2 to 4 hr. Table 65. — Amount of Effective Chlorine Required to Produce Stated Reductions in Numbers of Bacteria (Parts per 1,000,000 ; 1 part per 1,000,000 = 8.3 lb. per 1,000,000 gal. Abridged from Report of the Massachusetts State Board of Health, 1911, pages 345-350) Sample 99 per cent re- duction 90 per cent re- duction 75 per cent re- duction 20° C. 40<'C. 200 0. 40°C. 20°C. 40''C. Total Red Total Red Total Red Sewage: Maximum Minimum Average 11.3 3.8 6.6 11. S 3.8 6.8 7.5 -3.8 6.3 11.3 3.8 5.3 7.5 3.8 4.1 37.5 -3.8 8.6 7.5 -3.8 4.7 11.3 3.8 5.3 7.5 3.8 6.3 11.3 3.8 4.9 7.5 3.8 4.7 11.3 -3.8 7.0 7.5 3.8 4.7 11.3 3.8 6.8 7.5 3.8 6.3 7.5 -3.8 4.5 7.5 -3.8 4.1 7.5 -3.8 4.9 7.5 -3.8 4.7 7.6 -3.8 4.5 -7.5 -3.8 5.0 7.5 -3.8 4.5 -3.8 -3.8 -3.8 11.3 -3.8 4.9 7.5 -3.8 4.7 7.5 -3.8 4.5 7.5 -3.8 6.3 7.5 -3.8 4.2 3.8 -3.8 3.8 7.5 -3.8 4.9 7.6 -3.8 4.7 7.5 -3.8 4.5 7.5 -3.8 6.3 7.6 -3.8 4.2 3.8 -3.8 3.8 3.8 -3.8 3.8 7.5 -3.8 4.7 7.6 -3.8 4.5 3.8 -3.8 3.8 7.5 -3.8 4.5 -3.8 -3.8 -3.8 -3.8 -3.8 -3.8 7.6 -3.8 4.7 7.5 -3.8 4.5 7.5 -3.8 S.O 3.8 -3.8 3.8 -3.8 -3.8 -3.8 7.5 -3.8 4.9 7.5 -3.8 4.7 7.6 -3.8 4.6 7.5 -3.8 5.0 7.6 -3.8 4.2 -3.8 -3.8 -3.8 -3.8 -3.8 -3.8 Settled sewage: Maximum Septio tank effluent: Minimum Contact ater effluent: Minimum Average Trickling filter effluent: Maximum Minimum Average Sand filter effluent: Minimum — Less than stated amount. A general average of all the test results given out by Clark and Gage on the quantity of chlorine required, indicated the effect of time of contact to be about as follows : Hours of contact 1 2 4 6 24 Relative amounts of hypochlorite required. 100 84 82 77 61 At Providence, R. I., the use of calcium hypochlorite, without previous treatment of the sewage by lime, did not prove so effi- 532 SEWERAGE AND SEWAGE DISPOSAL cient as the application of bleach to the sewage after it had been treated with lime and had passed through the roughing tanks. Less than 4 p. p.m. of chlorine gave erratic results, but there was a remarkable uniformity in efficiency with from 4 to 6 p. p.m. The period of stability of the chemical precipitation effiuent was materially increased by the addition of bleach, as shown by methylene blue tests. An increase in the bacterial count after disinfection with hypochlorites is sometimes observed. Clark and Gage explain this by the fact that sewages and effluents from contact beds and trickling filters contain substances which combine with some of the oxygen liberated by hypochlorites, so that not all of it is available for destroying the bacteria. If the oxygen is used up before all the bacteria are killed, those that remain alive can multiply without restraint. In fact, by eliminating a large pro- portion of the bacteria, including probably all of certain species, the conditions become peculiarly favorable for rapid multiplica- tion of the remaining bacteria, since bacterial equilibrium has been destroyed. Sewage dosed with hypochlorites can be filtered successfully if the dose is not too large. Clark dosed raw sewage in this way with 25 p.p.m. of available chlorine without causing any decrease in the nitrification of this sewage when passed through a sand filter for 3 months. Similar experiments were made by dosing the sewage applied to a trickling filter, at first at the rate of 5 p.p.m. available chlorine which was gradually increased to 50 p.p.m. The last-named rate was continued for 3 months, decreasing the nitrates in the effluent from 20 to 1.5 p.p.m. but not checking nitrification wholly. The addition of disinfectants to sewage before filtration may be desirable during epidemics, and to prevent odors and kill insect pests about trickling filters, but where disinfection is practiced, it is usually the final effluent of the treatment plant which is dosed. The size of the dose of germicide required to disinfect sewage properly is much larger than that needed to disinfect water supplies, and the disinfection is generally needed only in cases where some other form of treatment precedes it, requiring the continuous presence of labor at the plant. These conditions favor the choice of hypochlorite of lime as a disinfectant, as being cheaper than liquid chlorine. If reliable labor is not constantly available to see that the hypochlorite apparatus DISINFECTION OF SEWAGE AND SEWAGE EFFLUENTS 533 is in working condition or the relative price advantage of hypo- chlorite over liquid chlorine materially decreases, then hquid chlorine will probably be the most advantageous disinfectant. Comparative estimates based upon local conditions and current prices are necessary to an accurate decision as to the relative merits of the two chemicals in each case coming up for consideration. Disinfection with Hypochlorites.— The quantity of hypo- chlorite to be used in a given case is best determined practically by observing the effect of different doses of disinfectant upon the bacterial content of the hquid to be disinfected. The sample of disinfected liquid should not be taken until the hypochlorite has had time to attack and be used up by the organic matter. Ordinarily it is practicable to select a definite period of contact, after which the test should be made, and it is to be expected that a definite quantity of free chlorine present at the end of this period will be a reasonably accurate indication of the effect of the disinfectant upon the sewage bacteria. The test period need not be as long as the working period of contact, which should be sufficiently long to allow the utihzation of all dis- infectant added to the sewage. Where the sewage occasionally receives large quantities of industrial wastes inimical to bacterial life, there may be times when there will be few bacteria and other times when the numbers may be large. Except in the rare cases when these wastes are discharged in accordance with a time-schedule, the operator of the disinfecting plant has no guide to help him adjust his dose of hypochlorite to the fluctuating requirements, and the best he can do is to test the efficiency of disinfection frequently and to regulate the quantity of applied disinfectant by the tests on previous days. Under this plan it will always be necessary to apply an excess of hypochlorite, because of the danger that the sewage may need a larger dose than it did the previous day. The bacterial count is subject to hourly variations and there are also perceptible variations on Sundays and holidays from the normal daily curve of bacterial content. Consequently some economy in the use of chemicals can be introduced in a large plant by varying the dose at intervals during the day in accord- ance with the curves of average bacterial content determined by tests covering a sufficient period to make their average a reason- ably reliable guide for disinfection. 534 SEWERAGE AND SEWAGE DISPOSAL In using hypochlorite of Ume, there are two steps in the process which must be taken properly or there will be failure. In the first place the hypochlorite solution must be made up correctly, irrespective of variations in the quality of the hypochlorite of lime. Some of the serious failures in the chlorination of water supplies have been due to the use of inferior hypochlorite, result- ing in the preparation of a solution so weak that its disinfecting value was very small and practically worthless. The second essential step in this method of disinfection is to feed the solution into the sewage at a definite rate. This is done in most cases by using a constant-level box from which the solution flows under a constant head through an adjustable orifice. Experience has shown that if this orifice is not inspected frequently and cleaned when clogged the disinfection will not be thorough. The bleaching powder is dissolved in a mixing tank, large enough to hold about 4 gal. of water for every pound of the powder to be dissolved in one batch. Where the mixing is done by hand, fairly strong solutions require somewhat less labor and involve fewer difficulties in mixing than the more dilute solutions. The solution should be allowed to stand several hours, preferably 12 to 24 hr., before it is diluted, to permit the lumps of bleaching powder to become soft. The fine material dissolves readily, but the lumps must be broken up in order to utilize the full value of the powder. If not enough water is used, a gelatinizing action takes place which retards the settling of the undissolved sludge of hydrated lime and silica. Clarification is important because suspended matter tends to clog the orifice of the feeding apparatus. Cast-iron tanks are made for mixing the bleaching powder in small installations, but in plants for preparing large quantities of solution the mixing is usually done in tanks made of large vitrified pipe with wood covers and concrete bottoms, wood tanks with or without concrete linings, and concrete tanks. Care must be taken on opening the packages of bleaching powder to prevent dust and fumes escaping into the room, as they affect the nose and throat. In some cases the package, as soon as it is opened, is placed in water and the powder turned out without coming into contact with the air. The mixing tank should be provided with a drain for drawing off the sludge. From the mixing tank the strong solution flows by gravity or is pumped to a solution tank or tanks where it is diluted to a strength of 1 to 2 per cent. The solution must be stirred to DISINFECTION OF SEWAGE AND SEWAGE EFFLUENTS 535 maintain a uniform strength. A 1 or 2 per cent solution is better than a stronger one because a greater quantity can be applied to the sewage, making it easier to adjust the measuring device for the feed and to obtain thorough mixing of the solution with the sewage. The solution tanks are made of the same materials as the mixing tanks. If the strong solution is discharged into them at the bottom under considerable head, the mixing of the solution with the diluting water is assisted. fn/e^ Valve Fig. 214.— Orifice feed tank (Pittsburgh). From the solution tank the solution passes to a chemical feed tank, by which the dose is discharged at a predetermined uniform rate into the pipe leading to the sewer or tank. This feed tank has a ball cock on the inlet pipe, which maintains the solution at a uniform height, and an adjustable orifice through which the solution passes to the outlet pipe. The orifice is adjusted by hand to discharge the solution at the desired rate. An apparatus of this kind, made by the Pittsburgh Filter & Engineering Co., is shown in Fig. 214. The orifice must be inspected frequently, as it is liable to become clogged with suspended matter carried over from the mixing tank. 536 SEWERAGE AND SEWAGE DISPOSAL Providence Hypochlorite Plant. — One of the largest installa- tions of hypochlorite apparatus for treating sewage was made at Providence, R. I., under the direction of Bugbee. After it had been in service for some time, a lime-bleach solution plant was built near the sewage treatment works. On account of the unusual facilities for obtaining this form of chemical from this plant its use was adopted, although the experience with hypo- chlorite had been satisfactory. The chemical plant makes lime- bleach solution by passing chlorine into milk of lime. This is allowed to settle, and the clear liquid is decanted into tank trucks and delivered to bleacheries. The lower portion of each tank SOO lb. Hupocfihrlfi , ■„,„,„ i,,,,,,,,:,^^^^,,^ '/-,-'''- \i ;^i h,,„,,„»,j,^u>J(iii(y""">"»"»"^ i'tfaicriuppty Concmte 4 CI. Chan-ouf Fips Fig. 215. — Apparatus for preparing disinfectant at Providence. contains lime sludge, which is not suitable for bleachery purposes but can be used for sewage treatment, and it is this portion of the product which is furnished to the city at a considerably lower rate than that for the same amount of chlorine as liquid chlorine or dry bleach. Deliveries are made directly into the tanks in the chemical house at the works, and each delivery is tested and diluted to the proper strength for discharge into the sewage. The dissolving tank used at Providence, Fig. 215, was cylin- drical with a conical hopper bottom. It was built of brick and covered by the floor above, which was placed directly on top of the tank, with a tight joint between the two. From the ex- treme bottom a suction pipe led to a 3-in. horizontal centrifugal DISINFECTION OF SEWAGE AND SEWAGE EFFLUENTS 537 pump driven by a 5-hp. motor. This pump had a cast-iron shell and brass shaft and runner. The discharge pipe connected tangentially with the mixing tank at a point near the surface of the solution and with the dosing tanks. By closing the valve leading to the dosing tanks and opening the valve leading to the mixing tank, it was possible to pump the solution over and over through the tank, thus thoroughly mixing the bleaching powder with the water. When the mixing was completed; the valve on the pipe leading to the mixing tank was closed and the valve on the pipe leading to the dosing tank was opened, after which the solution was pumped to the dosing tank. There was an opening 18 in. in diameter in the floor, directly over the center of the mixing tank. Pointing upward into the center of this opening was a 2-in. wrought-iron water pipe. The bleaching powder was furnished in metal drums holding about 800 lb. each. When it was desired to make up a tank of solution, a hole about 8 in. in diameter was cut in the bottom of a drum, thus avoiding the escape of chlorine gas which always collects at the top end. The drum was then inverted over the hole in the floor and so placed that the 2-in. water pipe pointed into the opening in the center of the drum. When the drum was in the proper position, water was turned on and the bleaching powder washed into the tank below, without the escape of any dust into the room and consequent discomfort to the operatives. The tank, which had a capacity of 2,300 gal., was then filled to the usual water level and the pump started. The liquid was kept in circulation for 1 hr. after which it was pumped to a dosing tank, where it was diluted to a strength of about 25 lb. of bleach per 100 gal. of water, approximately equivalent to a 3 per cent solution of bleaching powder or a 1 per cent solution of available chlorine. One mixing tank only was provided, the bleaching powder solution being pumped from it to two dosing tanks, 8 ft. wide by 20 ft. long by S}4 ft. in depth, holding about 100 gal. per inch in depth, equivalent to 25 lb. of bleaching powder. This made it easy for the operatives to gage the rate at which the solution was introduced into the sewage, by observing the drop in the level of the liquid in the tank in a given time. Emergency Plant. — An example of a very simple apparatus for temporary, small-scale use is shown in Fig. 216.* It was con- 1 Engineering and Contracting, July 17, 1912. 538 SEWERAGE AND SEWAOE DISPOSAL structed by the Indiana State Board of Health from three barrels, a commercial constant-level regulating box, a small amount of piping and a geared mixing contrivance much like that used on some types of ice-cream freezers. Fio. 216. — Emergency diainfeeting apparatus, Indiana State Board of Health. Liquid Chlorine. — In practice, the preparation of a disinfecting solution from bleaching powder is disagreeable work on account of the dust and fumes, the irregularity in the composition of the powder and consequent care needed to maintain a solution of proper strength, and the liability of the measuring orifice to become clogged. Liquid chlorine, which can now be obtained in steel tanks or cylinders from numerous producers, has come into extensive use for disinfecting purposes because these objections do not arise when it is used. A manually controlled apparatus made by the Wallace & Tiernan Co. for treating sewage with chlorine gas or a so-called solution of chlorine made from liquid chlorine is shown in Fig. DISINFECTION OF SEWAGE AND SEWAGE EFFLUENTS 539 217. The tanks of liquid chlorine, A, A, are generally placed on scales so that the variation in their weight as their contents are removed can be measured and in this way a check obtained on the rate of flow indicated by the chlorinating apparatus. Each tank has a tank valve, B, B, to which the operator attaches an auxiliary tank valve, C, C, which is the main shut-off valve Fig. 217. — Manual control direct feed chlorinator (Wallace & Tiernan Co.). during operation. When the valves are opened the liquid chlorine in the tank gradually becomes a gas and passes from the tank to a compensator, D, which maintains a constant flow of the gas irrespective of the pressure in the tank, so long as the setting of the control valve, V, forming part of the compensator 540 SEWERAGE AND SEWAGE DISPOSAL is not changed. The rate of flow of the chlorine is changed by altering the setting of this control valve. From the compensator, the gas passes to a meter, E, which is a manometer having a scale calibrated to show the flow of chlorine gas in pounds per 24 hr. or other unit desired by the purchaser. From the meter the gas flows to the place of application through a copper or galvanized iron pipe which should not be over 500 ft. long. The gas is deKvered through a chlorine check valve, F, which is necessary to prevent moisture from the sewage passing back to the chlorinator and interfering with its operation. From the check valve, the gas passes through a silver tube to the diffuser, G, submerged at least 4 ft. in the sewage. This diffuser is composed of a composition sponge of fine porosity held in a non-corrodible casing. The sponge is kept saturated with water by capillary action of the material of which it is composed and the chlorine in passing through the sponge becomes thoroughly mingled with the water and is discharged in this condition into the sewage. The longer the period of contact of the solution with the sewage before the latter is discharged into a water course, the more complete will be the disinfection. Fifteen minutes is the minimum contact time advised by the makers of the apparatus. This chlorinator is the simplest form used. It is known as a direct-feed apparatus because the gas is delivered directly to the sewage. In many cases it is desirable to use a solution of chlorine instead of the dry gas and for such cases the apparatus is provided with a solution jar, to which the gas passes from the manometer and in which the chlorine is mixed with an auxiliary water supply. The mixture passes from this jar to the sewage. This is known as a solution-feed apparatus. It is also very desirable when the flow of sewage fluctuates widely to alter the rate of application of the chlorine as the flow varies, and this is done by an additional regulating device converting the apparatus into an automatic control type. The liquid chlorinating apparatus should be placed in a well- ventilated room, so that in case there is any escape of gas into the room it can be removed immediately. This room should open out of doors and not into another room. The treatment of a person affected by the gas is merely to give fresh air and ether, inhaled to relieve the pain. CHAPTER XX IRRIGATION WITH SEWAGE Sewage contains chemical constituents which, when in suitable form, possess fertilizing value and for this reason there has been a popular belief that the agricultural utilization of sewage will be profitable. The most important plant food is water, which is the main constituent of sewage (99.9 per cent). Probably the chief agricultural value derived from the sewage when used for irriga- tion in most cases is due to its water content. Only a portion of the nitrogen, phosphorus and potassium compounds in the sewage is in a form suitable for direct utilization by plants. The nitrogen should be present as nitrates to be of most value, whereas usually only a very small part of it, or none at all, has been brought to that state and the remainder must be nitrified by biological agencies which require oxygen, a mild temperature and the presence of lime or some other base. The phosphates and potash, especially the latter, are present in such small quantities that their fertilizing value is usually small compared with that of the nitrogen compounds when converted into available form. The fertilizing constituents of sewage are associated with fat and soap, which are injurious to land. The fat and soap clog the pores of the soil and interfere with the absorption of the sewage by the soil and the subsequent aeration of the pores. Land which has become clogged and rank in this way is called "sewage sick," and one of the most important duties of the manager of a sewage farm is to prevent such a condition from becoming chronic at any part of his fields. A precautionary measure against it is the removal of the coarse suspended matter of the raw sewage by screening and sedimentation. The sewage is applied to the land in various ways. If it is flowed over sloping fields from an upper level to a lower one, the process is called surface or broad irrigation. In this method the sewage does not penetrate more than a few inches into the soil, except where the latter is unusually porous. When the sewage is applied by any method to soil of such a nature that the sewage percolates downward in considerable quantities it is called filtra- 541 542 SEWERAGE AND SEWAGE DISPOSAL tion. Filtration must be carefully distinguished from inter- mittent filtration, which is practiced on specially prepared areas without any agricultural use of the land, when this method of treatment is carried on most efficiently. Ridge irrigation is practiced by flooding sewage over land from channels on ridges between long, gentle slopes: the sewage collecting in the low places between the ridges is received by a cross ditch at the foot of the field and conducted to ridges in a field lying somewhat lower than the first. Where the soil is very heavy a third field may be laid out. In Europe, bed irrigation and land filtration are terms used to designate the distribution of the sewage in numerous ditches cutting up the land into beds, which are kept moist by horizontal seepage from the ditches. This method is called ridge-and-f urrow irrigation in England. In flood irrigation a plot of land surrounded by a bank is periodically covered with sewage to a depth of 1 or 2 ft. In sub-surface irrigation the sewage is distributed through open-jointed pipes laid about 1 ft. below the surface. In agriculture, fertilizers should be applied at certain stages in the rotation and growth of crops, and the proper fertilizers to use depend upon the nature of the soil, the climate, the crops to be grown and the rotation of crops. In sewage disposal, all these considerations must be waived in favor of the production of an effluent of suitable character. The crops should be regarded merely as a by-product. All evidence furnished by many years experience in many countries under many conditions does not reveal, however, that it is practicable to obtain much fertilizing value from city sewage as it must be used to make this method of treatment successful, but indicates that where irrigation has been successful agriculturally the same results would have been pro- duced with water. English Experience.— Irrigation with sewage has been carried on more carefully and extensively in England than elsewhere. This is not because the municipal authorities of that country have been particularly pleased with this method of treatment but because the Local Government Board has required some sort of land treatment of sewage and sewage effluents as a condition for granting its sanction to raising money for sewage treatment works. The area of land suitable for irrigation in England is limited, and consequently the Royal Commission on Sewage Disposal gave detailed attention to sewage farming. Eight IRRIGATION WITH SEWAGE 543 farms typical of different conditions were kept under close observation for two years, and observations were made less regularly at other farms. At these eight farms from 12 to 23 per cent of the area was not irrigable, either because it was unsuited for the purpose, or had not been prepared to receive sewage, or was required for roads, buildings, sludge beds and other purposes. The Royal Commission's conclusions from, its investigations were stated substantially as follows in its fifth report: 1. There is no substantial distinction between effluents from land and effluents from artificially constructed filters. Effluents from those soils which are particularly well adapted for the purification of sewage contain only a very small quantity of unoxidized organic matter, and are usually of a higher class than effluents from artificial filters, as constructed and used at the date of the report. Effluents from soils not well adapted to purify sewage may often be very impure. 2. Generally speaking the evidence points to a maximum rate of 36,000 U. S. gal. per acre, or 1,000 persons per acre, with the best land, after preliminary treatment of the sewage, although some witnesses who testified before the Commission put the rate as high as 72,000 gal., or 2,000 persons, per acre, under similar conditions. With unsuitable land, such as clay, not more than 3,600 gal. per acre can be efficiently treated, even after settlement of the sewage. 3. The total acreage of a farm must be relatively much greater when the sewage is purified by surface irrigation than when the method of filtration is employed, and a larger percentage of surplus area is also desirable in the former case. The larger the total irrigable area the greater is likely to be the working cost. On the other hand, with good management, the larger the surplus irrigable area, the better is the purification likely to be and, within certain limits, the prospect of profit. 4. In the case of sewage which is to be treated upon artificial filters, it is generally desirable to settle out as much as possible of the suspended solids before filtration. This is of less moment in the case of sand, but where the soil is heavy the sewage should first be eflaciently screened and settled. Porous sandy soil, worked as filtration farms, may be able to treat crude unsettled sewage without detriment, but, even in those cases, there is the possibility of nuisance arising from the decomposition of sewage 544 SEWERAGE AND' SEWAGE DISPOSAL solids on the surface of the soil, and such solids may cause damage to crops. 5. It is impossible to lay too much stress on the importance of sewage farms being well managed. Farm managers have a most difficult part to play, and no amount of care and attention will ever enable land of any kind to deal with a volume of sewage in excess of the effective purifying capacity of the soil. As a rule, sewage farms should not be let, 6. English soils and subsoils may be divided into the following three broad classes: Class I, all kinds of good soil and subsoil, such as sandy loam overlying gravel and sand; Class II, heavy soil overlying clay subsoil; Class III, stiff clayey soil overlying dense clay. Since variations exist in practice, both as regards the method of purification employed and the extent of cropping, the first of these three classes may be divided into three subclasses, Table 66. — Average Amounts op Land Rbquibbd for Treating a Dry-weather Flow op 1,000,000 U. S. Gallons (Royal Commission on Sewage Disposal. Fifth Report, page 153) Total area of , Volume of settled sew- land required Class of soil Method of working age ■ which can be treated per acre per 24 hours to treat a dry- weather flow of 1,000,000 U.S. gal. 1 I. Good soil and Filtration with crop- 14,400 U. S. gal. 70 acres and subsoil ping I. Good soil and Filtration with little 30,000 U. S. gal. 33 acres subsoil cropping I. Good soil and Surface irrigation with 8,400 U. S. gal. 121 acres subsoil cropping II. Heavy soil on Surface irrigation with 6,000 U. S. gal. 167 acres clay cropping III. Stiff clayey Surface irrigation with 3,600 U. S. gal. 278 acres soil on dense cropping clay ^ These areas are suifioient for the treatment in times of storm of three times the mean dry- weather flow. as follows; (1) filtration with cropping; (2) filtration with little cropping; (3) surface irrigation with cropping. The method of purification assumed to be suitable for the other classes of soil is surface irrigation with cropping. With a heavy soil and clay subsoil, by far the greater part of the purification is effected by surface irrigation, although in exceptional circumstances a good deal is also done by filtration. IRRIGATION WITH SEWAGE 545 Having regard to the volume of sewage dealt with and the purification effected at the farms kept under the Commission's observation, and the testimony given by various expert witnesses called by the Commission, it estimated that the several classes of soil and subsoil could deal effectively with the volumes of settled sewage given in Table 66, under the climatic conditions of England. French Sewage Fanning. — The utilization of sewage in one way or another has long been practiced in France. The native thrift of the French people led them many years ago to attempt quite extensive use of the wastes collected in cesspools and Victor Hugo's fervid plea^ for employing the sewage of Paris in irrigating farms about that city was long ago realized by works described later in this chapter. In some cities the night soil was collected and delivered to farmers in more or less unsatisfactory ways. An attempt was made at Paris to render it less offensive by evaporating it and mixing it with gypsum, charcoal and other materials into a fertilizing material called "poudrette." There were 24 plants about Paris at one time where poudrette was made and they became so offensive that stringent regulations were prepared to govern their design and operation. Experience with the agricultural utilization of night-soil on a large scale finally proved that it was expensive and unsatisfactory, both as a method of disposal and as a means of fertilizing fields. It is practiced locally on a small scale in some places where sewerage systems have not been built and the local conditions are favorable. Irrigation with sewage was carried on in about 30 French cities in 1910. The land irrigated was generally used for pastur- 1 "Paris casts twenty-five millions of francs annually into the sea; and we assert this with- out any metaphor. . . Twenty-five millions are the most moderate of the approximate amounts given by the estimates of modern science. Science, after groping for a long time, knows now that the most fertilizing and effective of manures is human manure . . . There is no guano comparable in fertiUty to the detritus of a capital, and a large city is the strongest of stercoraries. To employ the town in manuring the plain would be certain success; for if gold be dung, on the other hand our dung is gold ... Do you know what those piles of ordure are, collected at the corners of streets, those carts of mud carried off at night from the streets, the frightful barrels of the night-man, and the fetid streams of subterranean mud which the pavement conceals from you? All this is a flowering field, it is green grass, it is mint and thyme and sage, it is game, it is cattle, it is the satisfied lowing of heavy kine, at night it is perfumed hay, it is gilded wheat, it is bread on your table, it is warm blood in your veins, it is health, it is joy, it is life." (" Les Miserables," Book II, Chap. 1, Jean Valjean.) While the scientists of Hugo's acquaintance may have been assured of the finanial figures he gives, it is an unfortunate fact that neither science, engineering nor agriculture has yet been able to obtain these millions represented by sewage, because the expense of building works to utilize sewage and the cost of operating the works exceeds the value of the sewage, and because the requirements of public health are paramount to those of sewage utilization. 35 546 SEWERAGE AND SEWAGE DISPOSAL age or growing grass and little attention was paid to the quality of the effluent. At that date Rheims was delivering about 12,000,- 000 gal. of sewage daily to a sewage farm of about 1,472 acres, which is equivalent to a rate of about 8,000 gal. per acre daily. The sewage farms about Paris are the chief French examples of sewage irrigation. In 1910, the volume of the city's sewage, from a population of 2,800,000 in the districts connected with the farms, was 160,000,000 to 185,000,000 gal. daily, and 160,000,000 gal. was used for irrigation. The area of the farms is given in Table 67. Table 67. — Sewage Farms op Paeis in 1910 Area of farms, acres Privately owned Owned by city Total GennevilKers AchSres M^ry-Pierrelaye 1,996 386 3,731 2,137 15 2,965 1,235 210 2,011 3,361 4,966 2,347 Carri§res-Triel 8,250 4,425 12,675 The farms owned by the city had cost $7,220,000 in 1910 and the annual cost of operating them and distributing the sewage was about $1,000,000. Financial reports of the farming opera- tions are not obtainable so it is impossible to ascertain the net cost of this method of treatment at Paris. The city's land is leased to tenants at rates ranging from about $5 per acre, where the tenant agrees to apply sewage to his land only as directed by the city's supervisors, to about $40 per acre where he uses the sewage when and how he desires. The owners of land within the irrigation districts take the sewage as they desire it. The sewage is screened and settled. It is distributed through the farms in reinforced concrete conduits 1 to 4 ft. in diameter, lined with sheet steel where the pressures are heavy. These conduits have risers 1 ft. in diameter with outlets for the sewage into open carriers which distribute it over the fields. The land is partially underdrained at a depth of 13 ft., by drains of plain or reinforced concrete pipe. They discharge into open ditches with concrete lining. IRRIGATION WITH SEWAGE 547 German Sewage Farming.— In 1910 there were about 50 German cities having sewage farms. As a rule the sewage was distributed through the fields in furrows, separating the land into beds about 3 ft. wide and up to about 125 ft. long. Before sandy- soil is laid out in this way it is sometimes flooded with sewage to a depth of 10 to 20 in. during the winter, as often as seems possible. This results in a deposit of sludge on the top of the soil, and this sludge is subsequently worked into the soil with a view to improv- ing its fitness for farming. At Magdeburg the sewage was sprayed over grass meadows by hose- connected with portable 3-in. cast-iron pressure pipes laid on the surface of the ground. In a few cases surface flooding by the ridge method was followed. The general preference was to keep the sewage away from plants except as it could reach their roots by horizontal percolation through the soil. The sewage farms of Berlin afford the largest opportunity for this method of treatment yet (1921) undertaken in any country. At the close of 1910 the area of these farms was as given in Table 68 and they were treating an average of 77,000,000 gal. daily, coming from a population estimated at 2,064,000. Of the prepared land 7,994 acres were used for broad irrigation, 12,250 acres for filtration, 502 acres for settling basins, and 2,105 acres for roads and buildings. The rate of filtration was about 3,700 gal. per acre of prepared land per day. The principal crops were rye-grass, turnips, cabbages, potatoes and grain. About one- fourth of the area was used for pasturage and there were about 40 acres of fish ponds which yielded fish worth about $3,200 annually. Table 68. — Berlin Sewage Farms in 1910 Farmed by city Leased to farmers Unpro- ductive Total Area irrigated and farmed, acres . . Area farmed without irrigation, acres 16,657 10,647 3,966 2,486 395 8,868 21,008 22,001 27,304 6,442 9,263 43,009 The cost of the farms to March 31, 1910, was about $17,- 470,000. The expenses for the year ending on that date were 548 SEWERAGE AND SEWAGE DISPOSAL $1,300,000 for maintenance and $742,000 for interest and loans. The receipts were $1,241,000 and there was an estimated increase of $122,600 in the value of live stock and other property. According to Hazen^ the soil is a light-brown sand with an effective size of 0.13 mm. and a uniformity coefficient of 2.5. The top soil is thin and differs so little from the lower material that it has usually been considered unnecessary to separate the two in grading operations. The thin growth of low pine which covers most of the land was cut off the tract to be prepared for farming, the ground was usually graded to a level surface with the help of light portable railways, and 2- to 3-in. underdrains were put in at depths of 4 to 6 ft. These drains were on grades as low as 1:250 in some places and were not more than 30 ft. apart. They discharged into main drains 4 to 7 in. in diameter with minimum grades of 1:500. When the effluent reached a quan- tity beyond the capacity of a 7-in. drain, it was carried away in open ditches. These had 1 : 13^ side slopes protected with brush and longitudinal rows of small stakes until willows planted along the edges of the banks developed enough to hold them. The sewage reaches the farms through force mains with which small standpipes are connected. These are merely a few lengths of pipe set vertically to contain floats. Each float has a rod rising from it, to the top of which a large target is attached during the day and a lantern at night. The standpipe has an overflow which discharges on to the beds in the vicinity, but is not expected to come into action under ordinary conditions. The watermen are expected to watch the target or light and when it rises, to open enough gates on the pressure pipes connected with the force mains to keep the sewage from reaching the over- flow. These gates discharge the sewage into small brush-lined basins with a screen of brush to intercept large floating material. These basins give off more odor than other parts of the farm, but are considered as satisfactory as any substitute that has been proposed besides being very cheap to construct and main- tain. From these basins the sewage flows through open carriers formed in embankments around the beds into which the land is subdivided. The bottoms of the carriers are above the beds, so that the channels can be drained when not in use. A thick growth of grass soon covers them and holds their shape satis- factorily. Wooden boxes through the sides of the carriers, and » Bng. News, Sept. 16, 1S97. IRRIGATION WITH SEWAGE 549 wooden gates for closing the boxes and carriers, enable the watermen to control the discharge of the sewage over the beds. Although sewage irrigation has had strong advocates among German engineers, largely because of the considerable areas of light sandy soil available for carrying it on, it also has its critics who predict its gradual abandonment. Dr. Dunbar states^ the opinion of these critics as follows: "I am convinced that it would be cheaper for many towns to abandon irrigation and replace it by artificial biological processes. It appears fairly certain that this will be the course of affairs as soon as the growth of the towns exceeds a certain limit. I do not doubt, for example, that many of us will live to see the day when Berlin will sell its irrigation farms for building purposes and construct artificial biological works in their place." American Sewage Farming. — Sewage irrigation has not been tried in the United States on a large scale but it has been practised in a small way at a number of places, particularly in Southern California, for many years. The cities of Danbury, Conn., Pasadena, Fresno and Pomona, Cal., and San Antonio, Tex., are among the places where it has been given long trials with more or less success. Health oflBcers report from time to time that sewage is treated by irrigation when what is done is to raise crops on intermittent sand filters, thereby running the risk of clogging the filters by roots and necessitating an early and expensive reconstruction of the beds. At one time irrigation was practiced extensively by market gardeners near Ivos Angeles, but it was unsatisfactory both to the health officers and to the farmers Irrigation has been carried on in a desultory way for many years at Hastings, Neb. At Salt Lake City the sewage is sold to a farmer for $1, but he makes very little use of it. Cheyenne, frequently mentioned as a city using sewage for irrigation, has not done so since 1890. Pullman, 111., now a part of Chicago, the first large American city to irrigate farming land with sewage, gave up the practice many years ago. At the State Farm at Howard, R. I., where an irrigation system was constructed in 1885 from the plans of Samuel M. Gray, the history has been different from that of most early American sewage farms, for the fields laid out then have been in continuous use down to the present time. One field has an area of about 2.5 acres and the other an area of 5.7 acres. Both had originally > "Principles of Sewage Treatment," p. 266. 550 SEWERAGE AND SEWAGE DISPOSAL about 10 in. of fine light loam on a subsoil of sand and fine gravel, growing coarser as the depth increases. The surface was leveled and 3- and 4-in. tile drains were laid about 40 ft. apart at a depth of 5 to 6 ft. The sewage was screened and carried in channel pipe bedded in concrete on the top of embankments running around the sides of the land. This channel pipe had gates at intervals of about 100 ft. through which the sewage was discharged on the sloping fields. Dr. F. B. Jewett, superintendent, informed the authors in 1914 that it was the custom to turn the day sewage on the large lot and the night sewage on the other. Very little effluent appears at the outlets of the drains. Up to 1912 the land was cropped, vegetables being raised in the earlier years and corn later. The sewage was distributed in furrows. Since 1912 no crops have been raised and intermittent filtration has been practiced. The more extended use of irrigation in Southern California than elsewhere is probably due to the suitable character of the soil for sewage farming and to the scarcity of water. The former condi- tion materially reduces the cost of this method of treatment, while the latter condition greatly helps using the sewage in a fairly satisfactory agricultural way which does not have decided sanitary defects. The sewage of San Antonio, Tex., is conveyed 12 miles from the city to a 6.700-acre privately-owned tract, where it was used in irrigating 1,500 acres in 1911, when the farms were visited by one of the authors. Ultimately about 4,000 acres will be irrigated, it is expected. The sewage amounted to about 12,000,000 gal. daily, so that the land was dosed at the rate of 8,000 gal. per acre daily. The irrigating season was from the middle of Feb- ruary to the middle of November and during the remaining 3 months of the year the sewage was stored in a lake having an area of about 1,000 acres. CHAPTER XXI COST ESTIMATING The study of comparative costs of different types of structures and methods of building them is as important to the engineer as are considerations of strength, utihty and beauty. The esti- mation of cost or value is essentially a problem in equitable comparison. It must be founded upon past experience. The most valuable data to the engineer will be found those acquired by himself upon work of the construction of which he has intimate knowledge. The ability to distinguish and differentiate between the funda- mental conditions materially effecting cost, to discard the unes- sential and study closely the important elements, must be cultivated. Every man must develop his own yardsticks. They will of necessity be founded largely upon his personal experience; the wider his experience and the simpler and better his methods, the sounder will be his ultimate judgment of costs. Clearness of conception and accuracy of information are essential. There- fore, it is important to the young engineer to give some study to this subject, to post himself as to methods which others have found from experience to be most useful, and to keep track of fluctua- tions in market prices of labor, machinery, materials and supplies, and of the influences affecting eflSciency and output. Bookkeeping vs. Cost-keeping. — There is an essential differ- ence between bookkeeping and cost-keeping. The bookkeeper is interested in a correct financial accounting of the total cost of the final work and of such portions of it as are of interest to his principals. The engineer, on the other hand, is interested not only in final cost, but unit costs of different classes of work. Therefore, his accounting must serve to connect efficiency and output with cost. Purposes of Cost -keeping. — The methods of cost-keeping will be governed very largely by the purposes for which it is undertaken. Thus the contractor desires not only a correct financial statement, such as would be furnished by a competent bookkeeper, which will enable him to keep track of his costs, month by month, to the completion of the work, but, what is far more important, an approximate daily record which will enable 551 552 SEWERAGE AND SEWAGE DISPOSAL him to keep close track of efRciency and to do those things which may serve to lessen cost and increase output as well as excellence of work. The engineer is interested chiefly in developing final costs of works of different character and unit costs of the major items entering into them, which he may use thereafter in the making of approximate estimates or comparative studies of cost of different works and methods of building them. His estimates are in general of three kinds; approximate or preliminary estimates which must be sufficiently accurate to guide him in design, and to aid his clients in making decisions as to extent, character and method of undertaking the execution of their work; monthly or progress estimates to govern periodic payments to the contractors; and final estimates, which must be sufiiciently accurate to be equitable to the parties in interest in the case of active construc- tion work, or which may serve as evidence of cost, or fair value, of engineering works for use in valuation or rate determination proceedings. They must be able to stand the test of rigid proof, cross-examination and court review. The approximate current estimates of cost will also serve to disclose evidence of inefficiency and waste due to bad coordina- tion or methods of work, dishonesty, and expensive delays in getting materials, supplies, equipment and labor. Practical Advantages of Cost-keeping as Developed by a Successful Building Concern. — An admirable paper upon con- struction costs was presented before the Boston Society of Civil Engineers^ by W. N. Connor of the Aberthaw Construction Co., with forms illustrating its up-to-date practice, which illustrates the practical necessity of cost-keeping. The student will find this article helpful. Connor classifies the functions of any cost accounting system in the order of their importance as follows: "1. To aid in insuring the economical handling of the work while it is in progress; "2. To obtain figures to compare with the estimate; "3. To supply data for use in estimating future work." He calls attention to the fact that "costs must be available at once to be effective in aiding economical handling of the work. Delayed reports of costs are practically useless in effecting any savings while the work is in progress. Proper cost reports have been compared to a fire-alarm system in that they should not alone be able to give notice of some unusual condition, but equally lead the investigation to the seat of the trouble." ijour. Boston Soc. C. E., 1921. COST ESTIMATING 553 "Each morning the labor costs of the previous day should be known. This is not necessary or feasible on all items of work, but on such as concrete, excavation, brickwork, etc., it is possible and extremely important that the superintendent of a job should know the daily labor-unit costs. K he has them, he can take up with the foreman at once any variations between esti- mates and actual performance and endeavor to equalize disproportionate costs of any unit before it is too late." The president of the company, Leonard C. Wason, in commenting upon this paper, referred to the system as being the outgrowth of experience, and as being designed "to check the work when you can act on what you find instead of when it has become ancient history." He brought out the significant point that "8 to 12 items, out of the 50 to 70 total into which our records are subdivided, include nearly 80 per cent of the total cost of all operations. These are analyzed and on the superintendent's desk at 10 o'clock on the morning following the execution and he has all the rest of the day to act upon them, if necessary. So our costs enable him to keep the work under absolute control and thereby savings are made." The Methods of Construction should be recorded, as these and the size of the work have a marked bearing upon cost. The work may be well balanced or it may be overmanned or over- equipped. Day Work vs. Contract. — The method, of executing the work and paying for it is also an important factor, as the psychological conditions have a marked bearing upon cost. So does the extent to which the element of competition can be introduced into the work. Various methods of payment for contract work have been used, the most common being the so-called lump-sum and unit-price methods. Other methods have also been used such as cost plus percentage profit, cost plus fixed profit, cost plus profit with maximum and minimum limits of total cost or with a variable division of the difference between the actual cost and the nominal cost fixed by the contract. Simplicity. — It is important that the cost-keeping system should be as simple,, inexpensive and flexible as possible. The greater the simplicity within the limits of adaptability, the more useful will the system prove. Complexity in detail is likely to lead to incomparability. While it may be desirable and often is to subdivide units in some detail on specific classes of work, it is important that the total cost should also be kept divided into unit items of simple character, such as excavation, concrete masonry, etc. 554 SEWERAGE AND SEWAGE DISPOSAL Local conditions must be studied; the transportation facilities, the nationality, character and condition of available labor; legal limitations; the general business conditions as to supply and demand, and standards of efficiency. Two Methods of Estimating. — Cost-estimating methods may be grouped under two general classes; the first, based upon the comparative final costs of completed works with judicial estimate of the influence upon cost of the difference in existing conditions and environment; the second, the determination of costs by a synthetic building up of the items entering into or making up the unit structures. Under the first method, for instance, the total costs of intermit- tent sand-filter plants may be expressed per 1,000,000 gal. of rated capacity or per acre of area; the cost of pumping stations equipped with pumps of certain types, such for instance as high duty fly-wheel, or duplex, or centrifugal, per 1,000,000 gal. of daily capacity or per water horsepower: the cost of trickling filters may be based upon capacity in millions of gallons daily, acre-feet of area, or number of persons served. It is essential that structures of the same general type only be compared and that the estimator familiarize himself as far as possible with the character of the individual plants from which the cost figures were adduced. Under the first method, it is also possible to segregate the entire costs of the structure in question, such as a trickling filter, under a few important items, such as crushed stone in place, per cubic yard; underdrainage system in place, per square foot or per acre; distributing system, including nozzles and appurtenances, per acre of bed; engineering and overhead costs; and interest during construction. The average unit costs of each of these groups may then be found, to aid the memory, and all of the figures covering different records available should be classified and tabulated under these groups and the engineer should post himself as far as possible as to the wages, conditions and character of the work and the locality in which it was done. In using the results thereafter for estimating the cost of his own project, with these individual figures and their average before him, the engineer would select the unit cost which seemed most reasonable to him in the light of his data and knowledge for each of the similar items of his job, and thus obtain an estimate of final cost. Under the second or synthetic method, the records of various COST ESTIMATING 565 works would be tabulated according to the items of the contracts under which they were paid for with perhaps some consolidation of items, but the number of items which would be carried might well be from three to five times as many as under the method first mentioned. These records, as in the first case, would be classified, tabulated and perhaps averaged under different kinds of work, such as cement, sand, stone, mixing concrete, forms, placing, etc. From the facts thus marshalled, the engineer would make estimate of the cost of his own project by applying to the items, into which he had divided it, figures which appeared reasonable in the light of his assembled data. If the synthetic method is based upon unit costs developed in actual works fairly interpreted, the difference between these two methods becomes one of degree of segregation of items rather than of principle. It is important that in assembling the facts, the elements of cost, resulting from hazard, extra work or litigation, as well as engineering, administration, and overhead items should be included, either as separate items or be distributed amongst the other items into which the entire cost is subdivided, so that the figures may represent full cost and not partial cost. Both methods are serviceable. They are based upon past records of actual cost, found in the construction of plants more or less similar in character. The latter involves the attempt to segregate the elements of cost in somewhat greater detail. The former is open to the danger that some of the influences upon cost may be overlooked in the broad review of these past costs, but it has the advantage that with the small number of classified items, the figures and conditions are more easily visualized and that the final full cost of the other works is always kept clearly in mind. Under the synthetic method, there is danger that in the more detailed consideration and segregation of the elements and items entering into cost, the mind may be centered upon the details of the segregation, rather than upon a comparison between the resulting final estimate and the past records of cost, with the usual result that the synthetic estimate will be too low, because of the omission of unforeseen and unforeseeable items, or, in other words, of the hazard element. For the latter reason, it is desir- able, after completion of the estimate by the synthetic method, to compare the final results with the records of cost based upon 656 SEWERAGE AND SEWAGE DISPOSAL experience interpreted by simple overall unit costs or by the ratio method hereafter referred to. While it rarely happens that the hazard elements, or the causes of unexpectedly high items of cost, are the same in differ- ent engineering works, they are always present in greater or less degree. It has come to be common practice, therefore, to intro- duce an item of "contingencies" to cover them, in the form of a percentage added to the general estimate of cost. Contract Prices to Be Used with Caution. — In a brief but admirable article in Engineering and Contracting^ entitled, "The Art of Making Rapid and Reliable Estimates of Cost," Hazen has pointed out some of the dangers in the use of contract prices and outlined two desirable methods of estimate, "The Weighted Price Method" and "The Ratio Method." Contract prices or quotations for work with which the engi- neer is not personally familiar, should be used with caution if taken as a guide in making estimates of projected work. Usually such available prices or quotations do not apply exactly to the conditions under which the projected work is to be done and may not cover just the same items; the items of extra work, growing out of conditions that either were not anticipated or were inten- tionally excluded from the contract, may be overlooked or ignored ; different items or classes of items may readily be omitted or over- looked in a comparison of contract prices; the schedule of quanti- ties prepared in advance of any work may fail to agree closely with the final estimate unless the engineering work from which the contract was drawn was thoroughly done; usually contract quotations do not cover certain items, such as cost of rights-of- way and engineering expense; finally there are often wide differ- ences among the quotations upon the same price of work and sometimes the lowest bidders underestimate the difficulties of the work, suffer serious loss in its execution and augment the cost to the owner by litigation. These are elements of cost not directly disclosed by the contract prices. For these and other reasons, estimates predicated wholly upon contract prices are likely to be materially lower than the final cost of any work. Weighted-price Method of Estimate. — Under this method, the effort is made to segregate into a very simple classification (such as that used by Hazen in his filter-construction practice, to wit, excavation and earthwork, masonry, filtering materials, piping 1 March 18, 1913. COST E8TIMA TING 557 and auxiliaries, the available final costs, records of which are available and which should be as comparable as possible with the class of structure the estimated cost of which is to be determined. All of the items of cost making up the final cost of the works are then segregated into these collected groups. In this way, unit figures are finally derived from which may be selected those on the whole most closely resembling the anticipated conditions, and these costs may be applied with less liability of overlooking the contingent or hazard element always present in one form or another in engineering construction. This is an excellent method where the engineer has accumulated a sufficient quantity of reliable data. It is necessary to use such data with caution because of changes in prices of labor and materials and in methods and machinery which may have taken place from time to time as the data have been accumulated. The Ratio Method. — Under the application of this method, which is generally used where subdivision or segregation of individual classes of items cannot readily be made for want of full information, a schedule is made of a limited number of items of work which may reasonably cover the greater part of the con- struction in the several cases, the records of which are available, and a simple schedule of unit prices is then formed corresponding to these units. Then one fixed price is finally assumed for each kind of work, and it becomes the standard of comparison for such units of work. In assuming this fixed price, the young engineer will do well to benefit by the experience of older prac- tising engineers. The price assumed should, of course, be a reasonable one and as nearly as known an average one, but pre- cision is not to be expected and round figures are always used. The amount of work under each item in each job is then ascer- tained and the assumed prices are applied to them. The sum of the amounts for each job represents what that job would have cost at the assumed -price and comparison between the actual cost and the cost at the assumed price gives a coefficient or an idea of the relative economy or difficulty of the work on a ratio or per- centage basis. When the records of a number of known pieces of work are compared in this way, the assumed fixed-unit prices and coefficients found on different jobs, furnish a criterion for making preliminary estimates of cost of work in the same class. If the young engineer will get into the habit of comparing the costs adduced in the works with which he may be connected 558 SEWERAGE AND SEWAGE DISPOSAL in some such simple manner as this, he will much more rapidly acquire a sense of proportion than he could possibly do by simply comparing, in memory or by actual record, the final costs of the works or the contract prices paid for them. In using this yardstick (ratio method) in making new estimates, the fixed prices of the basic schedule will be applied to the various classes of items involved and the final cost thus built up will then be modified by the coefficient or the ratio or percentage item which judgment may dictate as the most probable under the conditions prevailing, as compared with all of the past records known to the estimator which have been analyzed theretofore. "In arriving at the ratio to be used, the engineer will compare, perhaps in his mind and without written schedules, the ease or difficulty of the proposed work, as compared with the ease or difficulty of the various works which served as a base; will take into account differences in labor conditions, in freights, in deliveries; will take into account the known or assumed effi- ciency, or lack of efficiency, in the execution of the several pieces of work from which his basic data were derived; and will reach an estimate of the addition or subtraction to be made to or from the base price in each case" (Hazen). Effect upon Cost Estimates of Length of Construction Period. — The longer the period of time involved in building the works, the greater the hazard of estimate, particularly under such conditions as those of the present time (1921), following the Great War. This will be clear from the diagram in the Preface to this book. Even under more stabilized conditions, such as the pre-war conditions, the element of chance is an important one. This was interestingly illustrated in a large contract for public utility works, built in New York City shortly before the war, at a cost of approximately $12,000,000 and requiring a construction period of about 8 years. As to the estimate of cost upon which the bidder based his proposal for the work and upon which contract was awarded, one of the successful contractors stated after the completion of his work, that one of his greatest difficul- ties in bidding had been in attempting to correctly forecast the average price fluctuation in labor, materials, supplies, and money, which might prevail during the construction period of 8 years. He gave point to his remarks by stating that during the execution of his contract, two distinct minimum and maxi- mum scales of prices were encountered which chanced to average up fairly closely with his forecast. Had either one of the mini- COST ESTIMATING 559 mum periods been eliminated he and his associates would have lost heavily: whereas if either of the maximum-cost periods had been eliminated they would have enjoyed a very substantial increment in profit. Extra Cost Involved by Complexity and Novelty of Design. — An interesting example of the effect upon the contractor of the difficult labor conditions induced by the war, was shown in the comparison of bids upon two different types of sewer con- struction at Lynn, Mass., in 1920, the one involving simple mass concrete construction, the other substantially lighter reinforced concrete structure. The heavier type of structure commanded the lower bid, although there can be little doubt that under normal conditions with reasonably skilled contractors bidding, the reinforced type of structure would have proved the cheaper:^ Similarly in the letting of a contract for a 10,000,000-gal. covered concrete masonry reservoir at Dayton, Ohio, in 1917- 1918, a more attractive bid was obtained upon a reinforced, flat-slab roof supported by cylindrical columns than upon a groined-arch type of structure which would normally have yielded the lower bid. The reason for this condition was found in the fact that the local contractors had not had experience in groined- arch construction and were adversely influenced by the failure of such a type of structure built in Cleveland sometime prior thereto, whereas they were skilled in building reinforced flat-slab concrete floors and had available forms adapted to the proposed structure, which had been used on other work. In each of the above cases, therefore, abnormal local conditions governed the type of structure chosen and the minimum price. Classification of Items. — In general, costs may be segregated under the items of labor, materials, supplies, plant, general expense and overhead costs, and profit. Under general expense are usually included engineering and legal costs, bonding and liability insurance, bookkeeping, timekeeping and storekeeping, rentals, office and miscellaneous expense (telephone and telegraph, stationery, supplies, etc.), interest and discount. The cost of the major items may for convenience be further subdivided into different items. Thus, the item of excavation may be subdivided into excavation in trench, hoisting excavated material out of trench, hauling or transportation, wasting and I Eng. News-Rec. vol. Ixxxv, No. 4; July 22, 1920, p. 148. 560 SEWERAGE AND SEWAGE DISPOSAL 4.OOJ iWSun -ISCi J3M9S J.0 (SJO||0QUl)4.S00 o o o c> ==" ^ ^ ^ ^ ^v ^ ^ E^ ^ ^ *!^ ■ '-^ >: ;^ =^-^ ^ ^ ^ ^ =^ r^ Sn ^-^ 5>^ 5^ ■^ ;:^ ^ T7 ^ v'^ "^ 3^ ^ >^^ ^ 5 ^ v^ v^ s— c ^ ^ ^ ^ — s ^ V^ ^ ^ ^ ;:^ ^ ^ v^ ^^ V^ V- h ^ A^ ri •0 >- D ir. ^ S 3= ^ c^ to n. Mi r ^ oj jr -6 1 U-l =: ■9- vo s C) _j *% £ t; ■t «t= *■■ I' c^ t S s •it g s a? m -t ro oJ' -■ C ■ — ^ ^^N». ■^ '^ ^^. ■^ 0^ "■^ — , ^ 'v ^^ ^ "^ .^ '^ ^ 1 ^s^ ^^% :\ ^■: N," ^* N ' N> ^ X^ , ^ ,N ^■^5% v\ N s N.^ igv^ N ^. s> s ^^ ^ S'^ s s s}* 1^ r-S SV s v» --^ V s s \H ^ E 3 sS ss s- ^Is --rP SS s S*!-' ^$$5. I , V ^^' \ > 2 VV Nv ^A -oN s ■> ■^v S'^ ,s \\N vs \\ \\ vs s^ \^ s, A \\ \ \ ' \ \\ A W w V \v \\ ft \\ \ i\\ \ \ L \. \\\ a a -P 0. CfeS •a u zH O >} u A cJh .ei ^- § " f. IT) 6> i o ID ^ t«-> M - c N' ^> \ "^ sS N N s ^ \ s sS N s^ SS S s ' ^ fl' S S ' sS s ' ^sV c! SS \' ^. N wi , ■- S' s ^ N s. \ !^ ?i ,s \S s '^\^ >! ^^ S S, * >\\ s S. ' ,s. \ Vi SSSi '^ s\ ss S ' 5? ss s , s lo > S -.JAV 2 \pi r.\ r\y S[-, rs \ vr \=V" A\ Wen ■^ YY AX \^ \vi t '■i\ ' ^ \ \ \^V^ ? cA \ N WvV \ \ > \ > \\v A A l\\ \\ \\ \ \\ \\ \\ \ \ ' \ \\ \\ \ \ \ \\ y \ \ \\ \' \ \ ' \\ \ \ \ \u •4JOOJ lOSUHJSd J3M39^,{sJD|10au!)4.S03 +-2 J) a COST ESTIMATING 561 spoiling, backfilling, compacting, ramming or puddling, shoring, sheeting, etc. It may involve the use of machinery, and fair allowances for its operation, repair, rental and depreciation. The operation of the machinery may in turn involve supphes, coal, oil, water, miscellaneous items, etc. While the keeping of these subdivided elements of cost may be very helpful to the young engineer, the ascertainment of the larger elements of cost is far more important, because of their comparability with the records of others. Methods of Applying Data on Cost.— Special methods of apply- ing data on cost must be developed for different pieces of work. Different men may accomphsh the purpose in a thoroughly satisfactory but totally different manner. Fig. 218 and Table 69 will serve to indicate methods for preparing estimates which have been used in certain cases, but it is to be clearly borne in mind that they are not generally appli- cable and apply directly only to similar conditions. In Chapter VI, Figs. Ill, 112 and 113 give diagramatic methods of showing the theoretical quantity of excavation for trenches of different depths and widths and in Chapter VII, Table 29, methods for showing quantities of lumber required for shoring sewer trenches in the manner assumed. In Table 69 is indicated a method of assembling data relating to the cost per linear foot of laying sewer pipe of different diame- ters and in Fig. 218 are shown curves giving the cost of pipe sewers under the different specific conditions indicated. In Figs. 219 and 220 are shown diagrams of cost of circular and semi- elliptical sewers and of vitrified pipe sewers made by the authors for use in estimating cost of sewer construction for a report in May, 1919 to the City of Springfield, Mass. These diagrams, too, are applicable only to such conditions as were assumed to prevail in Springfield at that time. They are shown merely for the purpose of indicating methods and not for the submission of specific data on cost. Progress Reports. — Progress reports, recording the number of men employed, the wages paid and the amount of work accom- plished along simple lines, are also desirable. They should be prepared to meet the specific character of work, method of con- struction and form of contract, and be so simple in form as to have wide applicability and convenience in use. Table 70 is an example of such a simple form prepared for insertion in the 33 562 SEWERAGE AND SEWAGE DISPOSAL M '■^ 00 • IM lO O O CO cc N ■ 1> o t^ Ci CO 00 IN r- Oi ■ ^H o rH o o o T-. O C ^ : WN d 6 6 d d d c T-l m ■ O o CO M l> lO CO IT 00 ■ Ol o CO 1^ (N CO IN ■- lO ■ IN o> f-l o o o rt O C •* ■ d d o d d d d c 1— 1 «© 99 ■ CO -— - i> O 05 IN O 00 C IN • ^ t^ lO »0 (N m ^ T- CO ■ O t^ o o o o ,-j o c iH d d d d d 0:= iH s «@ «© • 3" ^ 00 •* g^ i^ lO lO CS o • u :; 9 i; t^ CC CO CO IN IN T-I C Oi r-l C CO lO ■s (N IN (M o o o o o c O es c Ttl ■* T— t d o d d d d d c d d c 2 ° H S «@ ^ to oc t^ »0 00 00 OS Tf lo o t^ O lO iH C C^ IN i-H tH O C t^ —1 c 00 CO t^ o IM C< o o o o o c o o c CO CO o tH o d c d o d d d c d d c s ° p^ «& ^ 00 o: O C» (N IN t~ -* O 03 cc tH IC « '^ -^ C^ 1—1 1—1 T-I o c CD o c t^ CD SI 00 T-l T— o o o o o c o o c IN IN 8 = d d d d d c d d c g° 1-3 00 Tt CO »0 05 03 lO IN lO O: ^ uo t^ o c ■-1 .-1 o o o c 2 °^ to »-H 1- o o o o o o o o c IN >-( d c ^-DoubleSef-ofSheeting>| Fig. 219. — Cost of vitrified pipe sewers. Final Estimate. — An example of a form of final estimate, upon a piece of sewer construction done by contract, is appended. The unit prices involved were abnormally high, as the work had to be let at the peak of war prices : therefore they are of no signifi- cance and are not of general applicability. COST ESTIMATING 665 SSl^OUJ Ul SJ3M39 JD|n3JQ ^0 J3^»UJDIQ ■j3M3g|D3i4dr||g-juJ3g 'yj 566 SEWERAGE AND SEWAGE DISPOSAL cS o3 oooo i >. >i >> >. >> « 03 2 Jggoowoo ta-ii r-iH -coiNcc - CO o wcc C»^C»OiO(Mi-Hi-tr-IOOO(N'^'- =ooggggggo 0+3-*^ -t^ 03 0303 03 ooooNCQ Oi-H o Oioeo omco M -l tOCNW O = MlNO O S S S "-« « 1 fHo>^. am o«Moo o S m 9 o *° O'-' o op a COM Aqq ri fl(^ a c a ) ID v m .03 -*J Q, a " c8 J. !? P.. Qj lu HI 0* m • ij"-''s'~^ 'S.S'.S'.S'.&b; oS'^'rt .HrHrH « 3 S fl -i^. a a ii M bi EH J Paoort a.* e-o e-o oT3 N^ ca ^-^ "_— i* O^,H(M(Me0C0C0e0CCP5M'*»0OI>WW00OO'-<(N ,— l^,-|,-tr-ll-li-4i— li-HiHi-HTHi-lTHTHi-lrHi-Hr-li-HCQCvIN g 'So a g 'So c b CO Q --I o3 f -a o °^ -i § ::- .« O ■•e:s INDEX Acids, added in sludge pressing, 424, 439 disinfection, 527 in mine drainage, 527 in Monongahela River, 527 Activated sludge process, activating the sludge, 398, 399, 400 active sludge necessary, 398 adaptability of process, 403, 404 aeration periods, 404-406, 407, 413, 420 tanks, 411, 415 aerobic conditions necessary, 396, 407 agitation, 396, 407, 409 air compressors, 416, 417, 418 diffusion, 408, 413, 414, 415 requirements, 407, 408, 409 supply pipe, 413, 415 at Chicago Stockyards, 406, 424 at Cleveland, 403, 411 at Houston, 412, 424 at Lawrence Experiment Station, 395, 423 at Manchester, Eng., 395, 399 at Milwaukee, 401, 402, 403, 404, 405, 409, 410, 412, 413, 414, 419, 420, 421, 422, 424 at Sheffield, Eng., 409 at University of 111., 398, 399, 407 at Worcester, Mass., 405, 422, 424, 425 blowers, 416 character of sewage influences period of contact, 407 coagulation, 396, 397, 398 color removal, 403 defined, 15 described, 396 diffusers, 408, 413, 414, 415 pipe, 413, 415 plate, 408, 413, 414, 415 early experiments, 395, 396 effect of temperature, 402, 403 effluent, 403, 404 floe, 396, 397, 400, 401 impractical at Worcester, Mass., 405 management, 425, 426 measurements, 409, 410, 411 odors removed, 403 origin, 395 Activated sludge process, oxygen require- ments, 407 preparatory treatment required, 405 progressive changes during, 404 purification secured, 403, 404 sedimentation, 396, 410, 418, 419 septic decomposition, 400 sludge, addition of acid, 424, 443 beds, 422, 423 character, 407, 420, 421, 422, 423, ■ 429, 430 concentration, 401, 421 conditioning 424 dewatering, 424, 425, 441, 442 digestion, 402 disposal, 422, 423, 424, 425 drying, 422, 423, 425, 443 fertilizer value, 423, 424, 425, 443, 445, 446, 447 floe, 396, 397, 400, 401 marketing, 423, 424, 425, 446, 447 measurements, 409,-411 nitrogen content, 423 physical properties, 400, 420 pumping, 422 reactivating, 401, 402 return, 405, 406 sohds in, 396, 400, 401, 421 velocity in pipes, 421, 422 volume, 421, 422 standard of purification, 404 theory, 396 venturi meter, 411 winter operation, 402 Adams, Col. Julius W., run-off formula, 80, 81, 82 Aeration, blowers for compressing air, 416 periods in activated sludge process, 404. 406, 407, 413, 420 predecessor of activated sludge process, 395 requirements of various sewages in activated sludge process, 408 reqmres power plant, 21 tanks, 411 diffuser plates, 408, 413, 414, 415 pipe diffusers, 413, 415 with and without activated sludge, 398 Air compressors, centrifugal, 416, 417 Connersville, 417 efficiency, 418 569 570 INDEX Air compressors, General Electric, 417 Hytor, 417 Nash, 417 positive pressure blowers, 416, 417 power requirements, 418 Root, 409, 417 rotary blowers, 416, 417 lift, for pumping sludge, 393, 422, 437 Air-lock feeds for contact beds, 459 Akron, Ohio, exceptional growth, 24 Imhoff tank sludge moisture, 3S5 Albany, N. Y., density of population, 28 AlgsB, see Plankton. Alkalinity of sewage of American, English and German cities, 306 Allen, Kenneth, sludge capacity of ImhofE tanks, 386 Alliance, Ohio, leakage into sewers, 39 Altoona, Pa., leakage into sewers, 39 American Public Health Association, defini- tions, sewage disposal and treat- ment, 13-17 sewerage features, 1-3, 5, 6 oxygen consumed method, 289 putrescibiiity test, 291, 292 American Society for Testing Materials, specifications for castings, 134 for clay sewer pipe, 242, 243 Amherst, Mass., efficiency of sand filters, 520 intermittent filter operation, 508 Ammonia, album noid, 288, 306 free, 288 i nitrogen cycle, 286 Anaheim, fertilizer value of air-dried Imhoff tank sludge, 445 Analyses, bacteriological, bacteria standard for waters, 301, 531 B. coli determination, 301, 302 B. coli standards, 301, 302 classification of waters by, 301 of effluents from filters, 469 of sewage, 296, 297 of sewage after disinfection, 530, 531 significance of, 300, 301 chemical appearance of sample, 295 care necessary, 293 comparative, water and sewage, 307 determinations, alkalinity, 371, 372 ammonia, free and albuminoid, 288 biochemical oxygen demand, 292, 293 nitrate method, 293 relative stability method, 292 chlorine, 294 colloidal matter, 285 fixed residue, 285 fixed solids, 285 loss on ignition, 285 mineral matter, 285 nitrates, 289 nitrites, 289 Analyses, chemical, determinations, nitro- gen, 286, 287 odor, 295 organic matter, 285 organic sulphur, 287, 298 oxygen consumed, 289 oxygen dissolved, 281, 289, 294 putrescibiUty, 290 solids, 283, 284, 304, 305 Gooch crucible, 283 grams per capita calculation, 304, 305, 307 Imhoff glasses, 284 importance of, 280 limitations of, 280 of sewage of American, English and German cities, 306, 307, 308 of filter effluents in Massachusetts, 520 of Massachusetts water supplies, 307 Milwaukee sewage after treatment, 404 putrescibiUty tests, 290, 291, 292 sampling, 281 significance of, 282-296 sludges at Worcester, Mass., 432 solids in sewage, 305, 306 sterilizing samples, 281, 282 suspended matter, 305, 308 titrating, 371, 372 of sewage, see Analyses, bacteriological, chemical, of water, dissolved oxygen determina- tion, 281 limitations of chemical, 280, 281 of Massachusetts supplies, 307 sampling, 281 significance of bacteriological, 300, 301 solids contributed by water, 303, 304, 305 sterilizing samples, 281, 282 Andover, Meiss., efficiency of sand filters, 520 flies on filter, 500 intermittent filter operation, 508 sewage flow, 40 Angle of repose, 217, 218 Arches, analysis of masonry, 119, 261 brick. 111. 264, 267, 268 concrete, 265 elastic theory, 261 forms, 268 least crown thrust, 261 materials for, 264 reinforced concrete, 265 voussoir method, 261 Ardern and Lockett, named activated sludge process, 397 Area, increase of, in cities, 27 of American cities, 29 INDEX 571 Ashton Booth cleaner, 433 Asphalt jointa, 250 Astoria, Ore., tidal range, 317 Atlanta, Ga., Imhoff tank sludge moisture- experimental tank, 385 Peachtree Cr. tanks, 385 Proctor Cr. tanks, 385 suspended matter in settled sewage, 308 Attleboro, Mass., intermittent filter opera- tion, 508 sewage flow, 40 B B. coli, see also Analyses, bacteriological, significance in waters, 301 standards, 300-301 B. typhosus, 301 Bach, Dr. D. W., diffusion between cham- bers in ImhofiF tanks, 380 Backfilling, economio considerations, 254, 255 hosing, 254 inspection, 188 nature of material, 253 puddling, 253 ramming, 254 rolling, 254 settlement, 254, 255 should be done promptly, 255 surplus material, 255 tamping, 254, 255 Bacteria, act on organic matter, 18, 287, 297, 299, 362, 397, 403, 503 action within floe, 397, 398 B. coli determinations, 300, 301 carbon cycle, 298 classification of waters by, 301 denitrifying, 286, 300 effect of temperature, 402 of trade wastes, 297 of winds on distribution in waters, 323 oxidation, 287, 299 increase after hypochlorite disinfection, 532 in sewage, 18 influence filter bed action, 281 nitrate formers, 300 nitrite formers, 300 nitrogen cycle, 286 fixation, 286 number in sewage, 296, 297 oxidize sludge deposits on banks and shores, 313 oxygen requirements, 287 Pasteur's grouping, 296 produce nitrification, 299 reduction by application of liquid chlorine, 530, 531 removal by various methods, 469 in septic tanks, 362 Bacteria, significance in waters, 301 standard for Massachusetts drinking water, 529 standards for waters, 301, 531 sulphur cycle, 298, 299 Baffles, Imhoff tanks, 391 inlet of settling basin, 357 Baltimore, Md., butterfly valve, 499 complete system designed, 10 density of population, 28 discharges sewage into Patapsco, 11 dosing apparatus, 498 drum screen, 345 flight sewer, 141 house sewage pumped, 11 Imhoff tanks, 386, 387, 390 pumping sewage at, 11, 171 rainfall curve, 66 sewer section, 113 sewerage plan, 11 tidal range, 317 trickhng filter, 489, 491, 492, 498 Barbour, F. A., dosing siphon, 516, 517 Mansfield contact beds, 457 septic tanks, 369 North Attleboro settling tanks, 434 Bartow and Crawford, air requirements of activated sludge process, 407 Bartow and Mickle, diffuser tests, 415 Bartow and Mohlman, aerated sewage, 398 method of activating sludge, 399 sludge produced by activated sludge process, 421 Batavia, N. Y., infiltration, 36 Batter boards, defined, 187 in structural surveying, 187 to mark sewer line, 182 Bellmouth junction, 146 Berlin, radial system of sewers, 10 Birmingham, Eng., screening sludge, 349 sludge almost odorless, 431 vertical flow sedimentation, 360 Black, Col. W. M., New York Harbor dilu- tion requirement, 315 Blasting, charging holes, 206 cleaning holes, 199 drilling, by hand, 198 by machine, 200 exploders, 205 explosive, amount varies with character of rock, 199 dynamite, 203 freezing, 203 kinds used, 202 thawing, 203, 204 fuses, 205 good judgment necessary, 214 seamy rock, 214 spacing holes, 199 Bleaching powder, see Disinfection, bleach. Bleich, S. B., design of inverted siphons, 151 572 INDEX Blowers, see Air coinpresaors. Bolton, Eng., Fidler sludge remover, 435 Boston, Mass., cage racks, 341-343 density of population, 28 disposal by dilution at, 328 float measurements in harbor, 319 manhole step, 143, 144 rainfall curves, 58, 66 regulator, 163 screen, 350 screenings, 350 sewer sections, 113, 114, 117 Summer St. pumping station, 174, 175, 176 tidal range, 317 tide gate, 169, 170 water consumption, 32 Boston Society of Civil Engineers, report on inverted siphons, 148 Braces, cleating, 221, 223, 226, 227 on curves, 230 economy in lumber, 221 extension, 226, 233, 236 hammers for driving, 235 placing on rangers, 221, 234 should be tight, 233 size, 233, 234 stay, 229 with vertical sheeting, 212, 220 Bradbury, E. G., siphon discharge formula, 515, 516 Branch sewers, construction, 252, 253 deflned, 2 placing, 252 Bremen, disc screen, 348 sludge car, 433 Brewer, Bertram, designer of pumping station at Waltham, 176 Brick arches, bonding, 264 centers or forms, 267, 268 rowlock bond, 264 typical, 111 inspecting, 189 invert lining, 119 laid as headers, 278 laying, 266, 279 manhole base, 278, 279 sewers, arch forms, 268 centers for, 267 construction influences section, 261 erosion of, 99, 3aid on cradles, 267 laying brick, 266 profiles for, 267 sections, 255-261 stabihty, 119 Bridges, to carry sewers, 155 Bristol Company, diaphragm pressure gage, 88 Brockton, Mass., analysis of raw sewage, 307, 308 Brockton, Mass., analysis of water, 307 efficiency of sand filters, 520 infiltration, 36, 39 intermittent filter operation, 508, 524 ratio of water supply to sewage, 30 sewage flow, 40 Weand screen, 345 Bronx, Borough of, N. Y., catch basin, 130 cover, 132 inlets, 128, 132 aewer section, 110, 117 Brooklyn, N. Y., abnormal growth, 26 experience with siphon clogging, 149 sewer sections, 115, 117 typical circular sewer section, 112 use of separate and combined systems, 13 Brown, N. A., experimental grit chamber, 334 Brussels, sewer section, 118 BuflFalo, N. Y., density of population, 28 Bugbee, Julius W., hypochlorite apparatus, 536 precipitation tanks. Providence, 375 Buggy for laying pipe, 247 Builders Iron Foundry water level recorder, 87 Bureau of Surveys, IT. S., rainfall formula, 66 Biirkli-Ziegler, run-off formula, 80, 81, 82, opp. 85 Calvert, Texas, splash plate and discharge tube, 481, 483 Canton, Ohio, infiltration into sewers, 36, 39 Carbon cycle, 298 Carriers of typhoid, 301 Carson, Howard A., catenary section, 259 Castings, box step, 143 catch basin, 130, 131, 132 drop manholes, 141 inlets, 127, 128, 132. 133 manholes, 134-142. 143, 144 defined, 6 wellholes, 141 Catch basins, castings for, 132 Concord grate cover, 127, 132 covers, 127, 132 defined, 5 experience with in New York, 129 Manhattan, 130 Providence, 131 utility, 129 Catskill Aqueduct section, 114 Caution, restorm water allowance in sepa- rate sewers, 48 Cement, inspection, 189 joints, 248 Centers, see Forms. INDEX 573 Cesspool,, defined, 1 Chairs, concrete, 272 metal, 272 on aewer construction, 268 Charleston, S. C, tidal range, 317 Chart, for sewage gages, 88, 89, 91 from FitzGerald gage, 58 from Friez gage, 67 Chemical precipitation, addition of clay, charcoal or soil, 370 alkalinity determinations, 371, 372 with erythrosine, 372 with phenolphthalein, 372 with calcium oxide, 370 copperas, 371 ferric sulphate, 370, 372 ferrous sulphate, 370 lime, 370, 371, 372, 373 slaked lime, 372, 373 sulphate of alumina, 370, 372 sulphate of iron, 371 defined, 15 degree of clarification, 371 described, 370 design of tanks, 375, 376 detention period, 374 effluent, 371 formation of floe, 395 at London, 372 at Providence, 372, 373 for industrial wastes, 371 theory, 371 at Worcester, 372 Chesapeake Bay, receives Baltimore sewage, 11 Chestnut Hill, precipitation records, 60-61 Chicago Drainage Canal, 329 Chicago, 111., analysis of raw sewage, 307 of water, 307 density, 28 drainage areas, 41 dry weather run-oflf, 41 exceptional population growth, 26 rainfall curve, 66 sewer section, 117 suspended matter in settled sewage, 308 Chicago Stockyards, belt screen, 349 fertilizer value of air-dried Imhoff tank sludge, 445 Imho£f tank sludge moisture, 385 proportion of activated sludge in aerating tanks, 406 sedimentation tanks for activated sludge process, 419 Chimney, defined, 6 Chlorine, as disinfectant, 528 determination, 294 effect of winds on distribution in Milwaukee Bay, 324 In sewage of American, English and German cities, 306 Cincinnati, Ohio, area iacreasea, 27 gaging chamber, 90 hourly rates of sewage flow, 46 manholes, 136 plankton cleanse stream, 302 sewage flow from industrial districts, 47 gage, 89 Cities, comparative sewages of, 306 densities, 28 growth of American, 25 in Massachusetts, daily sewage flow, 40 policy re extension, 28 population and areas of American, 29 water consumption, 30, 31, 32 Clariflcation, defined, 14 Clark and Gage, aeration tanks, 411 colloidal matter in sewage, 423 contact beds, 463, 464, 465, 466 increase in bacteria after application of hypochlorites, 532 oxygen requirement to stabilize sewage, 407 sludge densities, 383 studies on liquid chlorine, 529 trickling filters, 475, 476, 477, 478 Clean-out chambers on inverted siphons, 149 Cleats, 221, 223, 226, 227, 230 Cleveland, Ohio, aeration tanks, 411 currents in lake, 321, 328 density of population, 28 fertilizer value of air-dried Imhoff tank sludge, 445 intercepting system plan, 19-20 sludge produced by activated sludge process, 420 storm overflow, 167, 168 water consumption, 32 wellhole, 140 winter operation of activated sludge process, 403 Clifford, Wra., experience with trickling fil- ters, 471, 472 Clinton, Mass., distributor, 514 efficiency of sand filters, 520 intermittent filter operation, 508 leakage into sewers, 39 sewage flow, 40 Clogging, see Deposits. Coagulation, in activated sludge process, 396, 397 in chemical precipitation, 370, 371, 372, 373, 374, 375 Coefficients, c in Hazen and Williams form- ula, 101, 103, 422 ci in run-off formulse, 86 n in Kutter's formula, 101, 103, 121, 122 n in Markmann formula, 120 n various values, 102 Coffin regulator, 162 Cohesion of soil, 214 574 INDEX Colloidal matter, defined, 14 Columbus, Ohio, bacteria in sewage, 297 contact bed, 450, 453 density of population, 28, 29 septic tank, 365, 366, 367 outlet, 366 sludge measurements, 429 suspended matter in settled sewage, 308 trickling filter, 489, 490 Combined sewer, defined, 2 system, allow for maximum velocities, 99 compared with separate, 8 defined, 3 in Brooklyn, 13 in Manhattan, 12 method of tabulating data for, 78, 79 relief outlet increases hydraulic gradient in, 95 Commercially dry sludge, defined, 16 Common sewer, defined, 2 Concentration, time of, 68 Concord, Mass., cleaning sand filters, 525 efficiency of sand filters, 520 flushing chamber, 158, 159, 160 intermittent filter operation, 508 leakage into sewers, 39 manhole invert, 137 sewage flow, 40 Concord grate, catch basin cover, 127, 132 Concrete, aggregates for sewer pipe, 246 inspection of, 189 arches, 261, 265 base, for brick manhole, 278, 279 consistency, 269, 270, 271 forms, 270, 272, 273 inspection of materials, 189 inverts, 270, 271 mixing, 269, 270, 273 placing for sewers, 269 pouring, 273 reinforcement, gives added resistance, 119 in arches, 265 in sewer pipe, 246 inspection, 189 placing, 272 sewers, 255 analysis of arches, 261, 265 built in one operation, 270, 273, 274 in more than one operation, 270, 271, 273, 274, 276, 278 construction influences section, 261 erosion of, 99, 119, 266 flatter inverts preferable in, 110 forms for, 268, 270, 272, 273, 274 lining, 99, 265, 266 manufacture of pipe, 246 monoliths, 119 placing concrete, 269 reinforcement, 272 Concrete sewers, pipe, 246 pouring, 273 reinforced, 119, 246, 265, 272 sections, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 255, i;56, 257, 258, 259, 260, 261, 262 stability, 119 subject to compressive stress, 119 Connections, house, defined, 2, 3 incorrectly made to separate sewers, 9 minimum size of house, 3 Connor, W. N., construction costs, 552 Consumption of water, 30-34 see also Water supply, consumption. Contact beds, absence of flies and moths, 449 air lock feeds, 459 arrangement of double, 448, 452 automatic control, 458-460 at Columbus, 450, 453 at Lawrence Experiment Station, 450, 454, 463, 464, 466, 466 at Manchester, Eng., 450, 451, 453, 454, 455, 456, 462, 463 at Mansfield, Ohio, 455, 457 at Plainfield, 455, 457, 462 bacterial action, 449 character of influent, 449 clogging, 451, 454 comparison with trickling filters, 500 defined, 16 depth, 454, 455 described, 448 drainage, 454 filling and draining, 449 floor, 455 foundations, 455 head, 449 loss of capacity, 451, 454 materials of construction, 452, 453 maturing, 450 multiple contact, 452 operating, 449 results, 461^66 removal of bacteria, 469, 470 size, 455 standing empty period, 450 full period, 450 theory, 449 timed siphons, 468, 460, 461 underdrainage, 454 Contamination, defined, 13 Copeland, Wm, R., activated sludge produc- tion at Milwaukee, 399 analysis of eflfluent from activated sludge process, 404 concentration of activated sludge, 401 effect of varying aeration periods with activated sludge process, 404 proportion of return sludge at Mil- waukee, 406 INDEX 575 Copeland, Wm. K., sweeping action of floe, 397 velocity through sedimentation tank, Milwaukee, 419 Copper sulphate, disinfection, 527 method of treating alga, 528 Cost data, cleaning out sludge, 433 Connor's studies of construction costs, 552 degreasing sludge, 447 dumping sludge at sea, 438 German sewage farms, 547, 548 Paris sewage farms, 546 reasons for excluding, vi replacing sand on intermittent sand filters, 524 trickling filter, 473 see also Prices estimating, advantages of cost keeping, 552 application of cost data, 560, 561, 562, 564, 565 bookkeeping versus cost-estimating, 551 caution on contract prices, 556 classification of cost items, 559 complexity of design influences cost, 559 construction period influences cost, 558 cost curves, 560, 564, 565, vii final estimate, form of, 564, 566, 567 functions of cost accounting classi- fied, 552 hazard elements, 556 importance of preliminary studies, 551 kinds, preliminary, progress, final, 552 local conditions influence, 554 methods described — final construc- tion costs, 554, 555 ratio to standard, 557 synthetic up-building, 554, 555 weighted price, 556, 557 of doing work — day labor, contract, 553 of payment, cost — plus, 553 lump sum, 553 unit price, 553 progress reports, 561, 563 purposes govern methods, 551 recording construction methods, 553 simplicity, 553 Cradles, construction and use, 251, 252 for brick sewers, 267 defined, 5 typical, 245 in wet work, 252 Crown, defined, 5 Crystalloids, defined, 285 Currents, effect of winds, 320 in Imhoff tanks, 391, 392 in Lake Erie, 321 in Lake Michigan, 320, 322 in North Atlantic, 321 lake, 20, 320-325 use of thermophone, 321 Curvature, compensation for, 120 effect on n in Kutter's formula, 121, T22 losses due to, 120 Curves in sewers, bracing, 230 compensations for, 120 loss of head due to, 120, 121, 122 D Damages, account inadequate sewers, 21 lack of repairs, 22 Dash plates, see Distributors, splash plates. Dead end, defined, 4 Definitions, miscellaneous, batter boards, 187 box sheeting, 213 cross braces, 213 ditcher. 193 dynamite, true and false, 203 poling boards, 227 quicksand, 213. 214 rangers, 213 rock excavation, 191 sewage disposal and treatment, activated sludge process, 15 aerobic bacteria, 296 anaerobic bacteria, 296 available chlorine, 528 ballast, 448 band screen, 14 bar screen, 14 biochemical oxygen demand, 292 cage screen, 14 chemical precipitation, 15 clarification, 14 coagulation, 397 coarse screen, 14 colloidal matter, 14, 285 commercially dry sludge, 16 composition, 282 concentration, 282 condition, 282 contact bed, 16 contamination, 13 crystalloids, 285 denitrification, 286 disc screen, 14 disinfection, 17, 527 disposal by dilution, 310 distributor, 17 Dortmund tank, 15 dosing apparatus, dosing tank, 17 576 INDEX Definitions, sewage disposal and treatment, drum screen, 14 effluent, 14 facultative bacteria, 296 filtration, 541 fine screen, 14 floe, 397 fresh sewage, 299 grating, 14 grit, 16, 330 grit chamber, 14 hydrolytic tank, 15 hydrolysis, 287 Imhoflf tank, 15 influent, 14 intermittent filter, 16 irrigation, 16 medium screen, 14 mesh screen, 14 moist combustion, 398 nitrogen fixation, 286 pollution, 14 primary bed, 448 putrescibility, 14 relative stability, 14 screenings, 16 screens, 14 scum, 16 secondary bed, 448 sedimentation tank, 15 separate sludge digestion, 16 septic sewage, 299 tank, 15 settling solids, 14 sewage disposal, 13 oxidation, 16 purification, 13 -sick land, 541 treatment, 13 slate bed, 17 sleek, 16 sludge, 15, 427 cake, 16 concentration, 16 dewatering, 16 digestion, 16 drying, 16 drjang bed, 16 pressing, 16 sprinkler nozzle, 17 stabiUty, 14 stale sewage, 299 sterilization, 17 sub-surface irrigation, 16 surface irrigation, 16 suspended solids, 14 tank treatment, 14 tertiary bed, 448 Travis tank, 15 trickling filter, 17 wing screen, 14 Definitions, sewerage features, basin, 5 branch sewer, 2 catch basin, 5 cesspool, 1 chimney, 6 combined sewer, 2 system, 3 common sewer, 2 cradle, 5 crown, 5 dead end, 4 domestic sewage, 2 drain, 2 drop manhole, 5 fittings, 6 flush tank, 5 ground water, 2 house connection, 2, 3 drain, 3 hydraulic gradient, 93 industrial wastes, 2 infiltration, 5 inlet, 5 intercepting sewer, 2 invert, 5 inverted siphon, 4 lamp hole, 5 land drain, 3 lateral sewer, 2 leakage, 5 main drain, 3 sewer, 2 trap, 3 manhole, 5 casting, 6 outfall sewer, 2 outlet, 6 pipe sewers, 5 plumbing, 3 regulator, 6 relief sewer, 2 roof water, 4 run-off, 4 separate sewer, 2 system, 3 sewage, 2, 23 sewer, 2 system, 3 sewerage works, 3 slant, 6 soil pipe, 4 specials, 6 stacks, 4 storm drain, 2 overflow, 6 water, 2 water overflow sewer, 2 street wash, 4 sub-main, 2 surcharged, 93 surface water, 2 INDEX 577 Definitiona, sewage features, trunk sewer, 2 under drains, 5 waste pipe, 4 wellhole, 5 sheeting, 213 trencher, 193 underdrains — "construction" and "permanent," 207 wales, 213 window, in sheeting, 226 Denitrification described, 286 Density of population, 27 in certain American cities, 29 Denver, Col. density, 29 intensity of rainfall, 64 rainfall curve, 66 Deposits, in catch basins, 129. 131 in contact beds, 451 in grit chambers, 331, 332 in inlets, 128 in inverted siphons, 148, 149 minimum velocities to prevent, 97 in pipes carrying milk of lime, 374 in tanks, 434 in underdrains, 207 in vitrified pipe sewers, 100 removal from, grit chambers, 335 in grit chambers, 330 in racks, 330, 336 on screens, 330, 336 from sewers, 97 from tanks, 359, 433-437 Des Moines sewer section, 112 Detroit, density of population, 28 exception growth, 24 Diamond versus North Attleboro, 22 Diatoms, see Plankton. Dieckmann, Geo. P.j cement pipe mixtures, 246 DifFuser plates and pipes, see Aeration. Dilution, by discharge into lakes, 20 into streams, 20 into tidal waters, 18 defined, 310 described, 310-329 see Sewage Disposal by dilution Disinfection, acids, 527 application of chemicals, constant level box, 534 orifice feed tanks, 535 Wallace and Tiernan direct feed chlorinator, 539 available chlorine, 528, 529 bacteria, removal of, 530, 531, 533 bleach, 528, 529, 531, 534-537 calcium hypochlorite, 529, 531, 534- 537 chlorates, 528 chlorinator, 539, 540 chlorine, 528-533 copper sulphate, 527 37 Disinfection, defined, 17, 527 efficiency, 529, 530, 531 emergency plant, 537, 538 filtration after, 532 hypochlorites, 528, 529, 532, 533 industrial wastes influence success, 533 at Lawrence experiment station, 530, liquid chlorine, 529, 530, 531, 532, 533, 538, 539, 540 mixing bleach solution, 534, 536 more effective after sedimentation, 351 ozone, 528 perchlorates, 628 previous to disposal by dilution, 311 at Providence, 531, 532, 536 quantity required, 529, 530, 532, 533 solution tank, 534 time of contact, 531, 533 variation in bacterial counts, 533 Distributors, arterial troughs, 513 defined, 17 fan shaped troughs, 513 graduated troughs, 513 revolving, 479, 480 splash plate, 481, 482 traveling, 479, 480, 481, 482 Dodd, C. H., .designer of Summer St. pumping station, Boston, 176 patentee of tide gate chamber, 170 Domestic sewage, defined, 2 Dorking, Eng., growths on trickling filters, 499 Dorr, "E. S., directed design of Summer St. pumping station Boston, 176 rainfall formula, 66 Dorr thickener, 436 Dortmund tank, defined, 15 Dosing apparatus, air controlled siphon, 497, 514, 515, automatic, 514, 515 defined, 17 mechanical, 516, 517, 518 mechanically controlled siphon, 516 Dosing tanks, 485, 486, 489, 497, 498 Ansonia at Newton, N. J., 518 defined, 17 Drain, defined, 2 see Storm drains. Underdrains. Drainage area, size in relation to run-off, opp. 85 Dresden, disc screen, 348 Drilling, cleaning holes for, 199 hand, 198 machine, 200 spacing holes, 199 Drills, auxiliary valve, 201 butterfly valve drill, 201 churn, 198 differential valve, 201 double hand, 202 578 INDEX Drills, hammer, 200, 202 percussion, 201 reciprocating, 200, 202 rotary, 200 spacing holes, 199 tapper valve, 201 Drop manhole, defined, 5 on connections, 7 to equalize flow on steep slopes, 100 Dryers for sludge, see Sludge drying, du Pont de Nemours Company, priming cartridges, 205 Dublin, sludge disposal at sea, 438 Dunbar, Dr. W. P., sewage farming, 549 Dynamite, composition, 203 exploding, 205 freezing, 203 method of charging with, 206 priming cartridges with, 205 thawing, 203, 204 boxes, 204 kettles, 203 in wet holes, 206 E Earth, adhesion, 214 angle of repose, 217, 218 cohesion, 214 excavation, see Excavation, pressures, 216 horizontal, 219 Rankine's formula, 216 at various depths, 219 weight, 218 Earth excavation, see Excavation, East Orange, N. J., leakage into sewers, 39 East River, receives Manhattan sewage, 19 Eddy, H. P. and Hartwell Fitchbury Imhoff tanks, 389 trickling filters, 487 Effiuents, defined, 14 from activated sludge process, 403, 404 from broad irrigation, 543 from chemical precipitation, 371 from Mass. sand filters, 520 disinfection, 527 methods of treatment, 18, 19 odors from septic tank, 494 point of discharge into streams, 20 into lakes, 20 into tidal waters, 20 requiring treatment, 18 sedimentation of trickling filter, 495 from septic tanks, 362, 364 stability in cold weather, 403 Ejector, for pumping activated sludge, 422 Shone at Worcester, 349, 437 Elastic theory for analysis of stresses in sewer arches, 119, 261 Emscher District, grit chamber design, 333, 334 grit in grit chambers, 332 ImhofE tank, 377. 393, 394 sludge chamber design, 393 sludge estimates, 380 submerged scutnboards, 391 waste weir, 338 Erosion of inverts, 99 Euglenas, see Plankton. Excavation, backfilling, 253, 254, 255 bracing, 212 classification of materials, 191 cost, general discussion, 208, 262 cross-sectioning surfaces, 186 dangers encountered in, 188 earth pressures, 216 for various sewer sections, 263 frost penetration, 191 general discussion, 190, 208 handwork, 192 in various materials, 213 inspection of, 188 lines of, 185 machine work, 193 opening the trench, 191 pumping in wet trench work, 206, 208 quantities, 208, 209, 210, 211 replacing old pipe lines, 193 rock, classified, 198 underdraining 207 thawing ground, 192 trenching machines, 193 underdraining wet work, 207 Factor of safety in sewer design, 92 Fairmont, Minn., septic tank, 365, 366, 367 Fall, minimum in treatment works, 19 Fats in sludge detrimental to fertilizer value, 446, 541 in typical sewages, 307 Fertilizers, ingredients necessary, 443 poudrette, 545 preparation of sludge for, 423, 424, 425 sewage, applied directly, 641-550 value of city sewage, 542 of cottonseed meal, 445 of dried blood, 445 of goat manure, 445 of guano, 445 of Imhoff tank sludge, 445 of septic tank sludge, 445 of sludge, 423, 424, 425, 432, 443, 444, 446 of tankage, 445 Filtering through sand, previous to disposal by dilution, 311 removes suspended solids, 284 see Intermittent filters, Trickling filters. INDEX 57S Fish, dissolved oxygen requirements of, 316 effect of disinfectants on, 528 feed on sewage, 280 Fitchburg, Mass., dosing tank, 485, 486, 497 flies on filters, 500 Imhoff tanks, parallel flow, 388, 389, 391 receive secondary tank sludge, 431 nozzles, 484 pumping sludge, 437 sewage flow, 40 sewage flow from industrial districts, 47 trickling filter, 467, 473, 484, 489, 490, 491, 492, 500 Fittings, defined, 6 FitzGerald, automatic rain gage, 57 Flies, absence on contact filters, 449 breed in cesspools, 1 in septic tank scum, 363 on trickling filter, 500 Flight sewers to equalize flow on steep slopes, 100 Floats for current measurements, 318, 319 on swivel draw-off pipes, 358, 359 FIoc, bacterial action within, 397 defined, 396 description in activated sludge process, 400 formation for chemical precipitation, 395 in activated sludge process, 396, 397, 407, 408 Flood flows from large areas, 86 Flush tank, defined, 5 Flushing devices, 155 automatic flush tanks, 156 fire hose, 156 flush tank defined, 5 flushing chamber at Concord, 158, 159 from brooks. 156 inverted siphons, 149 manholes, 156 Miller flush tanks, 157, 161 Rosewater improved flush tank, 160 Van Vranken tank, 393 Foaming, Imhoff tank, 393 Forms, centers, 267, 268, 269, 273, 277 collapsing, 268, 273, 275 for arches, 268 for building in two operations, 276 for inverts, 167, 268, 271 for manhole base, 278 for semi-elliptical sewer, 276, 277 full round wooden, 274, 275 general discussion, 272, 273 ingenuity of design, 270 lagging, 268 life of wooden, 273 moving, 273 profiles, 267, 268 protecting wooden, 275 Forms, ribs, 269 setting, 268 skeleton, support for, 274, 275 striking, 268, 273, 275 Formulas, Adams, run-off, 80, 81, 82 Allen, sludge storage capacity, 386 Bradbury, siphon discharge, 516 Btirkli-Ziegler, 80, 81, 82, opp. 85 Dorr, rainfall, 66 earth pressures, 216, 237 empirical not dependable, 119 Fuller, run-off, opp. 85 Ganguillet and Kutter, velocity of flow, 100 Gregory, C. E., rainfall, 66 run-off, 80, 81, 82 Gregory, Jn. H., run-off, 84 Grunsky, rainfall, 66 Hawksley, run-off, 80, 81, 82 Hazen and Williams, velocity, 101 Hendrick, C. W., rainfall, 66 Hering, Rudolph, run-off, 80, 81, 82 Hill, C. D., rainfall, 66 Horner, W. W., rainfall, 64 Knauff, rainfall, 66 Kops, de Bruyn, rainfall, 66 Kuichling, rainfall, 66 Kutter, velocity, 101, 103, 104-107, 121, 122 McMath, run-off, 80, 81, 83, 84, 85, opp. 85 Markmann, loss of bead, 120 Metcalf and Eddy, rainfall, 64, 66 Murphy, run-off, opp. 85 Parmley, run-off, 80, 81, 82 Rankine, earth pressures, 216 resistance to pressure, 240 sand diameters, 506 Sherman, rainfall, 66 siphon discharge, 516 sludge volume, 427 Talbot, rainfall, 63, 66 U. S. Bureau of Surveys, rainfall, 66 Williams (Hazen and), velocity, 101 Fort, E. J., use of separate and combined systems in Brooklyn, 13 Foundations, of sewers, 262 Framingham, Mass., efiBciency of sand filters, 520 intermittent filter operation, 508 leakage into sewers, 39 sewage flow, 40 Frank, Leslie C., Imhoff tank design, 386, 387, 390 Franklin, Mass., sewage flow, 40 French, A. W., analysis of elastic ring, 120 Friez, automatic rain gage, 56 rain gage chart, 57 register for rain gage, 56 Fuertes, James H,, water consumption studies, 31 580 INDEX Fuller, Geo. W., disaolved oxygen require- ments of waters, 316 run-off formula, opp. 85 Fullerton, fertilizer value of air-dried Imhoff tank sludge, 445 G Gagea, automatic rain, 55 diaphragni pressure, 88 float, 87, 91 maximum flow, 89 pneumatic pressure gage, 88 setting rain, 58 sewage, 90 sewer, 87 staff, 89 Gaging chambers, 90, 91 sewers, storm water flow in, 87-89 Galveston, Texas, precipitation record, 62 tidal range, 317 Ganguillet, E., and Kutter, formula for velocity of flow, 100 Gardner, Mass., efficiency of sand filter, 520 intermittent filter operation, 508 leakage into sewers, 39 sewage flow, 40 Gary, Ind., exceptional population growth, 24 Gas in Imhoff tanks, 378, 388, 389 in sedimentation tanks, 356 in septic tanks, 362, 363 vents of Imhoff tanks, 378, 392 Gascoigne, Geo. B., aeration tanks, 411 concentration of activated sludge, 401 sludge produced by activated sludge experiments, 421 winter operation of activated sludge plant, Cleveland, 403 Gasket, 248, 249 Glasgow Scotland, sale of sludge as fertilizer, 446 sludge disposal at sea, 438 Gloversville, N. Y., flies in filters, 499 intermittent filters, 496 manholes in flushing galleries, 493 suspended matter removed by sedi- mentation, ,360 trickling filter floor, 491 vertical sedimentation tanks, 355, 359, 360, 434 Gooch crucible for determining suspended solids, 283 Goodnough, X. H., dilution requirements, 314 Gow, Col. Chas. R., defines quicksand, 213 Grades of sewers, affected by method of dis- posal, 19 general discussion, 7 Grating, defined, 14 see Castings. Gravel, inspection, 189 placing around underdrains, 511 Great Britain, treats storm water, 18 GreatLakes, currents, 20, 320, 321, 322, 323, 324, 325 Lake Erie currents, 321, 328 Lake Michigan currents, 320, 322 points of discharge into, 20 Gregory, C. E., rainfall formula, 66 run-off formula, 80, 81, 82 Gregory, Jn. H., discharge of egg-shaped sewers, opp. 102 run-off formula, 84 septic tanks, Columbus, 367 defined, 16 Grit cham-bers, American design, 334 before septic tank treatment, 364 cleaning, 335 defined, 14 design, 332, 333, 334 object of, 330 quantity and character of deposits, 332 theory of, 331 Ground water, defined, 2 general discussion, 34 infiltration, 36, 37, 39 Growth in population, arithmetical rate, 25 curvilinear rate, 24 decreased rate, 26 exceptional, 24, 25 of Akron, 24 of American cities, 25 of Brooklyn, 26 of Chicago, 26 of Detroit, 24 of Gary, 24 of Manhattan, 26 of New Bedford, 26 of Washington, 26 percentage rate of growth, 24 Growths, killing with copper sulphate, 528 on trickling filters, 499 Grunsky, C. E., rainfall formula, 66 H Harrisburg, density, 28 water consumption, 32 Hart versus City of Neillsville, 22 Hartford, Conn., aqueduct section, 113 Hartwell, D. A., and Eddy, Fitchburg Im- hoff tanks, 389 trickling filter, 467 Ilatton, T. C, analysis of efiluent from activated sludge process, 404 progressive changes in activated sludge at Milwaukee, 404 solids in Milwaukee activated sludge, 401 standard of purification by activated sludge process, Milwaukee, 404 INDEX 581 Hawkaley, run-ofE formula, SO, 81, 82 Haworth, agitation at Sheffield, England, 409 Hazen, Allen, dangers of contract prices, 556 defines quicksand, 214 formula for velocity of flow, 101 German sewage farms, 548 Ohio streams pollution, 315 sand standard, 504 Hazen and Whipple, acidity of Mononga- hela River, 527 Head, in inverted siphons, 150 losses due to curvature, 120 recovery of velocity, 123, 124 refers to hydraulic gradient, 93 velocity, 123 Hendrick, C. W., Imhoff tank design, 386 rainfall formula, 66 Henry, Prof. A. J., rainfall study, 69, 70 Hering, Dr. Rudolph, advantages of sepa- rate system, 9, 10 grades on steep hills, 12 run-off formula, 80, 81, 82 Hill, C. D., rainfall formula, 66 Hoboken, N. J,, density of population, 28 Hobrecht, radial sewerage at Berlin, 10 Holyoke, Mass., water consumption fluc- tuations, 33 Hopedale, Mass., double manhole, 138, 139 efficiency of sand filters, 520 intermittent filter operation, 508 sewage flow, 40 Hopson, E. G., data on losses due to curves, 120 Horner, W. "W., allowance for time of con- centration, 68 describes use of rational method, 73 rainfall formula, 64 House drain, defined, 3 connections, defined, 2, 3 depth, 7 minimum size, 3 Houston, Texas, activated sludge plant, plan and sections, 412 filtros plates, 414 horizontal sedimentation tanks for activated sludge plant, 419 sludge drying, 423 Hudson, Mass., efficiency of sand filters, 520 intermittent filter operation, 508 sewage flow, 40 Hudson River, receives Manhattan sewage, 19 Hughes and Safford, Kutter formula, 107 Hugo, Victor, plea for sewage utilization, 545 Hydraulic elements of sewer sections, caten- ary, 105 circular, 103, opp. 102, 107, 109 delta, 106 egg-shaped, 102, opp. 102, 104, opp. 107, 108 Hydraulic elements of sewer sections, gradient, 93, 95, 145, 150 horseshoe, 105 parabolic, 106 rectangular, 104 eemi-elliptic, 106 see also Sewer sections. Hydraulics of sewers, 92 problems relating to, 124, 125 Hydrolysis, defined, 287 Hydrolytic tank, defined, 15 Imhoff, Dr. Karl, grit chambers, 332, 334 screen, 334 sludge estimates, 380, 381 tank design, 377 Imhoff glasses, use of, 284 tanks, addition of lime to, 379 air lift, 394 at Baltimore, 386, 387, 390 character of sludge, 378, 379, 383- 385, 431 at Chicago Stockyards, 385, 445 cleaning, 393, 394, 433 connecting sedimentation compart- ments, 379 defined, 15 density of sludge, 383 described, 377, 378 detention period, 381-383, 391 diffusion through, 379, 380 at Fitchburg, 388, 389 foaming, 393 gas formation, 378, 391, 393 gas vents, 378, 392 inlets, 379, 390, 391 longitudinal flow, 389 maintain sewage levels in, 379 odors 377, 379 outlets, 390, 392 parallel-flow, 389, 390, 391 quantity of sludge, 380 radial-flow, 386, 390 scum, 386, 392 skimming solids, 392 slots, 392 sludge chamber, 379, 380, 385, 391, 393 moisture, 383, 385 removal, 393, 394 storage capacity, 385, 386, 391 period, 385 for treating industrial wastes, 385, 386, 445 value of sludge as fertilizer, 445 Indiana State Board of Health, emergency disinfection apparatus, 538 Indianapolis, Ind., density of population, 29 582 INDEX Industrial Instrument Company, diaphragm pressure gage, 88 wastes, affect bacteria, 297 defined, 2 influence disinfection process, 533 preparatory treatment before acti- vated sludge process, 405 treatment by chemical precipitation, 371 treatment by Imhoff tanks, 386 Infiltration, defined, 5 of ground water, 35. 36, 37, 38, 39 possibility influences section, 264 quantity depends on length of sewer, 38 Influent, defined, 14 Inlet chamber design- problem, 153 time, see Time of concentration. Inlets, authors* standard street, 127 Bronx, 128, 130 castings for, 132 clogging, 128 covers, 127, 132, 133 defined, 5 Imhoff tank, 390, 391 location, 126 precipitation tank, 375 septic tank, 366, 368 street, 126 Inspection, duties of inspectors, 188 of backfilling, 188 of cast-iron pipe, 190 of concrete materials, 189 of excavation, 188 of sheeting, 189 of underdrains, 189 of vitrified pipe, 190 Intercepting sewers, Baltimore, 10-12 defined, 2 first flushings from, 48 general description, 8 provision for storm water, 47 use of regulators on, 162 Intermittent filters, accomplishment, 502 aerobic condition necessary, 503 bacterial action in, 503, 504, 506 banks, 510 capillarity, 607 cleaning, 524, 526, 532 Clinton, 514 clogging, 521, 522 defined, 16, 507 depth, 506, 507, 508 distribution, 609, 512, 513, 514, 521 dosing. 508, 525 air controlled siphonic device, 514, 515 automatic, 514 mechanically controlled siphonic device, 516, 517 mechanical devices, 518, 619 Intermittent filters, effect oxidation of organic matter in sewage, 300 efficiency, demonstrated by analyses of eflSuent at 23 Mass. plants, 519, 620 effluent, 603, 519, 520 establishing bacterial action, 503 filtering material, 503 freezing, 522, 523, 525 grading surface, 310 grouping, 509 Lawrence Experiment Station, 507, 508 Marlborough, Mass., 513 mats, formation and removal, 502, 512, 522, 623 Newton, N. J., 518 North Attleboro, 612, 516 odors, 521 operating conditions in Massachu- setts, 508 operation of, 621 overdosing, 503 preliminary studies necessary to design, 510 preparation of foundation, 504 rate of filtration, 507, 608, 509 removal of bacteria, 469 require large aresis of sand, 21 resting period, 624 sand, 503, 504, 505, 507, 523, 524 size and shape, 508, 509 straining action, 502 temperature affects results, 521 underdrainage, 609, 610, 511 winter operation, 521, 525, 526 International Joint Commission, B. coli classification, 301 Inverted siphons, 148 clean-out chambers on, 149 clogging, 149 defined, 4 effect of undue capacity, 151 essential principles of design, 151 experience at Wilmington, 151 flushing, 149 general features, 14S manholes on, 149 materials of construction, 150 multiple pipes, 151 problem in design, 162, 153 report of Boston Society of Civil Engineers, 148 size, 148, 150 as trap in plumbing, 3 velocities in, 97, 148 Inverts, brick, 267 concrete, 270, 271 defined, 5 design with reference to hydraulic gradient, 94 INDEX 583 Inverts, erosion, 99 flat, 273, 278 flatter for masonry sewers, 110 lining with vitrified brick, 119 of concrete sewers, 266 of long sewers, 94 of manholes, 137 resistance to abrasion, 99 screeding to templates, 271 stability influences section, 258, 259, 260 velocities In, 97 Iowa, specifies bearing strength of pipe, 241 Irrigation (Broad). acreage, 543, 544, 546, 547, 550 American experience, 549, 550 application of sewage, 541, 541 at Achdres, 546 at Carridres-Triel, 546 cost, 546, 547, 548 cropping, 544, 547, 549 crops secondary consideration, 542 at Danbury, Conn., 549 defined, 16 described, 541, 542 distribution, 546, 547, 548, 550 effects of oxidation of organic matter in sewage, 300 eflfluent primary consideration, 542 eflluents, 543 English experience, 542, 543, 544 French experience, 545, 546 at Fresno, Cal., 549 at GennevilUers, 546 German experience, 547 at Hastings, Neb., 549 at Howard, R. I., 549 at Los Angeles, 549 at Magdeburg, 547 management, 541 at Mery-Pierrelaye, 546 at Pasadena, 549 pasturage, 545 at Pomona, Cal., 549 previous to disposal by dilution, 311 previous treatment, 543, 546, 550 at Pullman, 549 rate, 543, 547 at Salt Lake City, 549 at San Antonio, 549, 550 sewage-sick land, 541 soils, 544, 548, 549, 550 supplemented by storage, 550 theory, 541 underdralnage, 546 Ithaca, N. Y., septic tank outlet, 366 Jersey City, N. J., density of population, 29 Jetting sheeting, 230 Johnson, Geo. A., bacteria in Columbus sewage, 297 Johnson, Geo. A., contact bed operation, 450 sludge measurements, 429 Joint outlet sewer, New Jersey, leakage, 39 Joints, asphalt, 250 cement, 248, 249 clay pipe, 245 effect of size of socket, 245 filtite, 251 G-K compound, 250, 251 Jointlte, 251 linseed oil preparations, 250 Marbleoid, 251 Stanford, 249 sulphur and sand, 249, 250 tar, 250 underdraln, 207 Junctions, bellmouth, 146 careful design needed, 8, 145 flat-top, 147 Philadelphia, opp. 148 K Kansas City, Mo., density, 29 Kellerman, Karl F., copper sulphate as algi- cide and disinfectant, 528 Key West, Fla., tidal range, 317 Kinnicutt, L. P., sludge measurements at Worcester, 429 Knauff, rainfall formula, 66 Kops, J. de Bruyn, rainfall formula, 66 Kulchllng, Emll, rainfall formula, 66 run-off formula, opp. 85 Kutter, W. R., formula for velocity in chan- nels, 101, 103, 104^107, 121, 122 Lacustrine current, effect of winds on, 320, 322 Lake Erie, 321, 328 Lake Michigan, 320, 322 Milwaukee investigations, 322, 323, 324, 325 not so marked as river currents, 20 Lagging, see Forms. Lakes, currents of, 20, 320, 321-326 dissolved oxygen in, 313 Lamphole, defined, 5 Lancaster, Pa., sewer sections, 112, 118 Land drain, defined, 3 Lanpbear, Plainfield contact beds, 457, 458 sludge drying at Worcester, 422 Lap blocks, 221 Lateral sewer, adequacy, 72 defined, 2 minimum size, 7 Laying, brick, 266 Inverts, 271 pipe, 246 backfilUng. 253, 254, 255 584 INDEX Laying' pipe, jointing, 248, 249, 250, 251 supporting, 251, 252 use of buggy and lifting hook, 247 wedge blocks, 251 in wet trench, 252 Lawrence Experiment Station, aeration tanks, 411 bacteria in sewage, 296, 297 colloidal matter in sewage, 423 contact beds, 450, 454, 463, 464, 465, 466 intermittent filters, 507, 508 originated activated sludge process, 395 oxygen requirement to stabilize sewage, 407 sand standard, 504 sludge density in sedimentation tanks, 383 trickling filters, 471, 475, 476, 477 Leakage, see Infiltration. Lederer, Dr. Arthur, stability of effluents in cold weather, 403 Ledge, see Rock. Leicester, Mass., efficiency of sand filters, 520 operation of intermittent filters, 508 Liabihty for inadequate sewers, 21 for lack of repairs, 22 Lindsay, Ont., fertilizer value of air-dried Imhoff tank sludge, 445 Linseed oil jointing compounds, 250 Filtite, 251 G-K compound, 250 Jointite, 251 Marbleoid, 251 Lipman and Burgess, sludge as fertilizer, 444, 445 Loads, concentrated, 240 dead and live, 119 on pipes, 237, 238 uniformly distributed, 240 London, chemical precipitation, 372, 429 outfalls, 8 sludge disposal at sea, 438 Louisville, Ky., density, 28, 29 form for manhole base, 278 forms for semi-elliptical sewer, 277 full round wooden forms, 274, 275 industrial flow allowed for, 47 rainfall curve, 66 sewer sections, 114, 118 street detritus, 99 Lumber for sheeting, 231, 234 Lynn, Mass., density of population, 43 discharge of residential sewage, 44 M McMath, Robt. E., run-off formula, 80, 81, 83, 84, 85, opp. 85 Madison, Wis., infiltration into sewers, 36, Main sewer, adequacy, 72 defined, 2 designed for future conditions, 23 selection of size, 7 Main trap, see Trap, main. Maiden, Mass., leakage into sewers, 39 Malvern, Eng,, revolving distributor, 480 Manchester, Eng., activated sludge studies, 395, 399 contact beds, 450, 453, 454, 455, 456, 462, 463 sludge disposal at sea, 438 Manchester, Mass., maximum daily water consumption, 32 Manhattan, Borough of, abnormal growth, 26 catch basin, 130 combined sewerage, 12 discharges into Hudson and East rivers, 19 manhole, 142, 144 Manhole castings, see Castings, manhole Manholes, at junctions, 145 brick, 278, 279 Cincinnati, 136 concrete base for brick, 278, 279 covers, 142 defined, 5 double, 138, 139 drop, 141 Newark, 139 flushing, 156 for Gloversville trickling filters, 493 form for base, 278 inverts, 137 location, 134 materials of construction, 135, 136, 138 Newark, 135, 139 on inverted siphons, 149 permit compensations for curvature, 120 spacing at Worcester, 134 St. Louis, 136 steps, 143, 144 tops, 137 Mansfield, Mass., water consumption, 32 Mansfield, Ohio, analysis of raw sewage, 307 of water, 307 contact beds, 455, 457 septic tank outlet, 366, 369 Maps, St. Louis, opp. 73 Springfield, opp. 48, opp. 74 Marion, Mass., efficiency of sand filters, 520 intermittent filter operation, 508 Markmann, P. J., formula for curvature allowances, 120 Marlborough, Mass., efficiency of sand filters, 520 intermittent filter operation, 508 leakage into sewers, 39 sewage distributor, 513 sewage flow, 40 INDEX 585 Marston, Anson, bending moment for pipe, 240 Fairmont septic tank, 367 modulus of rupture of pipe, 241 studies on pipe loads, 237, 238, 241 Massachusetts, bacterial standard for drinking water, 529 infiltration into sewers, 36 law reliability for inadequate sewers, 22 sewage flow, 40 sewer sections, 111, 116 Medfield, Mass., leakage into sewers, 39 Memphis, Tenn., manhole invert, 137 Merrimack River, dilution requirements, 314, 315, 316 Merrimac catch basin cover, 132 Metcalf, Leonard, water consumption studies, 30 Metcalf and Eddy, application of run-off formula, opp. 85 classification of rock excavation, 191 double manhole for Hopedale, 138, 139 rainfall formula, 64, 66 semi-eUiptical sewer, 258 value of n in Kutter formula for curved channels, 121 Metering to prevent water waste, 30, 31 Metropolitan Sewerage System (Boston), catenary sewer section, 259 Deer Island Outlet, 328 disposal by dilution, 328 float measurements in harbor, 319 Gothic catenary basket-handle sections, 110, 260 infiltration into sewers, 37, 39 Moon Island outlet, 319, 328 Peddocks Island outlet, 319, 328 ratio of wateif to sewage, 30 removal of screenings, 338, 350 screenings, 350 Ward St. screen chamber, 341, 342, 350 Metropohtan Sewerage Commission of New York, experience with catch basins, 129 float measurements, 318 salinity observations, 320 velocities necessary to move solids, 96 Miami Conservancy District, conclusion reeffect of curvature, 121 Miles acid process 447 Milford, Mass., efficiency of sand filters, 520 intermittent filter operation, 508 sewage flow, 40 Miller flush tanks, 157 dimensions and capacities, 161 Milwaukee, Wis., activated sludge process, aeration period, 413 air required, 409 dewatering sludge, 424, 441 difTusers, 414 Dorr thickener, 437 Milwaukee, Wis., drying sludge, 443 horizontal sedimentation tanks, 419 preliminary screening and grit removal, 405 production of active sludge, 399 progressive changes, 404 rate of sedimentation, 410 reaeration not advantageous, 402 return sludge, 404 screen, 350 screenings removed, 350 solids in sludge, 401, 421 standard of purification, 404 volume of sludge, 422 winter operation, 403 Bay, effect of winds on chlorine distri- bution, 324 temperatm-es of bottom waters, 325 vertical distribution of bacteria, 323 cage racks, 343 density of population, 28 industrial flow allowance, 47 sewage flow from residential districts? 42, 43 water consumption, average, 32 classified, 29 proportion reaching sewers, 29 Minneapolis, Minn., depth of wellholes, 141 dry weather outlet, 171 pumping sewage, 171 Monongahela River, acidity of, 527 Morse, H. S., maximum flow gage, 89 Mount Vernon, N. Y., dash plates, 483 Dorrco screen, 345, 346 sludge flow through pipes, 370 Murphy, E. C, run-off formula, opp. 85 N Narragansett Bay, algse in, 303 Natick, Mass., efficiency of sand filters, 520 intermittent filter operation, 508 leakage into sewers, 39 sewage flow, 40 Neponset Valley Interceptor, flow of industrial wastes allowed, 47 New Bedford, population growth, 26, 27 New Jersey State Board of Health, recom- mended minimum sewer grades, 98 New Orleans, complete sewage system designed, 10 infiltration, 37 leakage into sewers, 39 rainfall curve, 66 New York, density of population, 28 phenomenal rainfall, 67 screen, 350 screenings, 350 tidal range, 317 New York Harbor, dilution requirements, 315, 316 float measurements, 318 586 INDEX New York Harbor, salinity of water, 320 sewage discharge into, 312 specific gravity of water, 320 tidal range, 317 Newark, N. J., density of population, 28 drop manhole, 139 manhole, 135 Newton, N. J., dosing apparatus (Ansonia) , 518 Nitrates, bacterial formers of, 300 determination of, 289 nitrogen cycle, 286, 398 see also Analyses, chemical. Nitrites, bacterial formers of, 300 determination of, 289 nitrogen cycle, 286, 398 see also Analyses, chemical Nitrogen, content of activated sludge, 423, 425, 432 content of air-dried sludges, 445 sludges at Worcester, 432 cycle, 286 denitrification, 286 fertilizer ingredients, 443, 444 fixation, 286 Kjeldahl method, 287 hberation, 286 organic, 287 in sewages of American, English and German cities, 306 significance of, 287 Nordell, Carl, diffuser plates at Milwaukee, 414 North Attleboro, Mass., arrangement of sand filters, 512 dosing apparatus, 516, 517 eflficiency of sand filters, 512 gravity removal of sludge, 433 intermittent filter operation, 508 settling tanks, 433 sewage flow, 40 North Berwick, catch basin cover, 132 North Brookfield, Mass., leakage into sewers, 39 Northbridge, Mass., efl&ciency of sand filters, 520 intermittent filter operation, 508 sewage flow, 40 Norwood, Mass., efficiency of sand filters, 620 intermittent filter operation, 508 sewage flow, 40 Nozzles, fixed, 483 head variation, 487 spacing, 487 Taylor, 482, 483, 485 Worcester, 483, 485, 486, 487, 488 Oakland, Cal., density of population, 29 Ocean Grove, N. J., infiltration into sewers, 37 Odors, accompany some tank treatments, 21 determination of, 295 from bodies of water, 316 from Imhoff tanks, 377, 379 from intermittent filters, 521 from septic tank efiluents, 494 from sludge beds, 370 from trickling filters, 494 of Lake Erie water, 329 production in sulphur cycle, 298, 299 removal by activated sludge process, 403 scale for Worcester sewage, 295 Ohio, degree of river pollution, 315 Orange, fertilizer value of air-dried sludge, 445 Organic matter, see also Solids. action of bacteria on, 18, 287, 297, 299, 403, 503 constituents of, 282 danger from in bodies of water, 312 decomposition of, 313 dissolved by lime, 372 effect of nitrogen cycle on, 286 in grit chambers, 331 hydrolysis, 287 more important than mineral, 282 nitrogen in, 286 oxidation of, 287, 299, 403, 503 in septic tanks, 362, 363, 364 in sewage, 18 Outfall sewer, Baltimore, 12 Borough of Manhattan, 12 defined, 2 grade affected by method of disposal, 19 reduction of cost, 8 Outlet chamber, design-problem, 154 Outlets, 169 defined, 6 Deer Island, 328 duplex, 171 Imhoff tank, 390 Minneapolis, 171 Moon Island, 319, 328 Peddocks Island, 319, 328 precipitation tanks, 375 reUef, 95 Koohester, 169 septic tank, 366, 368, 369 Overflows, on combined sewers, 20 storm, 6, 162, 167 Oxidation, effected by bacteria, 299 of organic matter in activated sludge process, 403 Oxygen, absorption by quiet and running water, 326 absorption by septic tank effluents, 362 biochemical demand, 292 nitrate method, 293 relative stability method, 292 consumed, determination, 289, 290 INDEX 587 Oxygen, consumed in sewage of American, English and German cities, 306 dissolved determination, 281, 289, 294, 329 in water, sewage and efl9uent, 495 presence necessary in activated sludge process, 39S for action of intermittent filters, 503 requirements of activated sludge, 407 in streams and lakes, 313, 326 327 Parker, P. A. M., defines coagulation, 397 Parmley, W. C, describes Cleveland storm overflow, 167 run-off formula, SO, 81, 82 Pasadena, fertilizer value of air-dried Imhoff tank sludge, 445 Passaic Valley Sewer Comm. allowance for industrial wastes, 47 restricted as to quality of efl9uent, 312 Pasteur, Louis, classification of bacteria, 296 Patapsco River, receives Baltimore sewage, 11 Paterson, N. J., industrial wastes allowed for, 47 Payment, classification of excavated mate- rial, 191 for excavation per linear foot, 209 method governs measurements in quan- tity surveys, 185 see also Cost estimating. Pearse, Langdon, proportion of activated sludge in Chicago Stockyard tanks, 406 vertical sedimentation tanks for stock- yards, 419 Peoria, 111., leakage into sewers, 39 Percolating filter, term abandoned, 17 Phelps, Earle B., New York Harbor dilu- tion requirement, 315 Philadelphia, Pa., bacteria in sewage, 297 baffled sedimentation tanks, 35S density of population, 28, 44 flow of sewage, 44, 45 growths on filters, 500 intensity of rainfall, 63 junction, 146, opp. 148 rainfall curve, 66 sewer sections, 111, 114, 115 Webster's sludge study, 383, 384 Phillips, Asa E., reports operation of Wash- ington regulator, 166 Phosphoric acid, fertilizer base, 443 Pierson, Geo. S., Sturgis septic tanks, 367 Pipe, bedding in concrete, 241 buggy, 247 cast iron, inspection, 190 cement, manufacture, 246 modulus of rupture, 241 Pipe, cracking under pressure, 237 drain tile, bearing strength, 241 jointing, 248, 249, 250, 251 laying, 246 modulus of rupture, 241 multiple for inverted siphons, 151 sewers, 240 bedding in concrete, 241 cement modulus of rupture, 241 clay, 2^2, 243, 244 concrete, 246 cracking under pressure, 237 deep and wide socket versus standard, 245 defined, 5 discussed, 240 drain tile, bearing strength, 241 in wet trench, 252 jointing, 248, 249, 250, 251 laying, 246, 247 niodulus of rupture, 241 recording locations, 186 strength, 240 supporting, 251, 252 thicknesses listed, 242, 243 vitrified, 241, 242, 243, 244, 245 sizes for inverted siphons, 150, 151 vitrified, 241, 242. 243, 244, 245 bearing strength, 240, 241 bedding in concrete, 241 deep and wide versus standard socket, 245 dimensions specified, 243, 244 inspection, 190 high velocities undesirable, 100 jointing, 248, 249. 250, 251 laying, 246, 247 lengths, 243, 244 manufacture, 241 rejection, 243 resists abrasion, 99 salt glaze, 244 seasoning, 242 specifications, 245 supporting, 251, 252 thicknesses, 242, 243 Pittsburgh, Pa., density, 28 junction, 147, 148 orifice feed tank, 535 Pittsfleld, Mass., efficiency of sand filters, 520 intermittent filter operation, 508 sewage flow, 40 Plainfield, N. J., contact beds, 455, 457, 462 flies on Biters, 500 Plankton, algce, 302 copper sulphate treatment, 528 diatoms, 302 effect on Cincinnati stream, 302 euglenas, 303 general description, 280 588 INDEX Plankton, habitata, 302, 303 in Narragansett Bay, 303 mission of, 2S1 Platforms, to support pipe, 251, 252 Plumbing, defined, 3 Poling boards, 227, 228 Pollution, by storm water, 18 defined, 14 Ponds, absorption of oxygen, 326, 327 dissolved oxygen in 313 temperature changes in deep, 326 Population, Amherst, 508 Andover, 508 Attleboro, 508 Brockton, 508 Clinton, 508 Concord, 508 density, 27 exceptional growth, 25, 26 four assumptions as to growth, 24, 25, 26, 27 Framingham, 508 Gardner, 508 growth of American cities, 25, 29 of New Bedford, 26 Hopedale, 508 Hudson, 508 Leicester, 508 Marion, 508 Marlborough, 508 Milford, 508 Natick, 508 North Attleborough, 508 Northbridge, 508 Norwood, 508 per acre, 28 Baltimore, 28 Boston, 28 Buffalo, 28 Chicago, 28 Cleveland, 28 Detroit, 28 Harrisburg, 28 Hoboken, 28 Milwaukee, 28 New York, 28 Newark, 28 Philadelphia, 28 Providence, 28 St. Louis, 28 San Francisco, 28 Savannah, 28 percentage rate of growth, 24 Pittsfield, 508 Southbridge, 508 Spencer, 508 Westboro, 508 Worcester, 508 Portland, Ore, density of population, 29 Potash in fertilizers, 443 Potter, Alexander, travelling distributor, 482 Pratt, R. Winthrop, aeration tanks, 411 concentration of activated sludge, 401 septic tanks at Washington, Pa,, 366 sludge produced in activated sludge experiments, 421 winter operation of activated sludge process at Cleveland, 403 Precipitation, Chestnut Hill records, 51, 60 curves, Baltimore, 66 Boston, 58, 66 Brackenbury, 66 Bureau of Surveys, 66 Chicago, 66 de Bruyn Kops, 66 Denver, 66 Dorr, 66 Eastern U. S., 66 Gregory, 66 Grunsky, 66 Hendrick, 66 Hill, 66 Horner, 66 Knauff, 66 Kuichling, 66 Louisville, 66 Metcalf and Eddy, 66 New Orleans, 66 Philadelphia, 66 St. Louis, 66 San Francisco, 66 Savannah, 66 Sherman, 66 Spokane, 66 Talbot, 66 form of curve, 65 Galveston record, 62 intensity of, 59 phenomenal, 65 New York, 67 periods of maximum, 70 rates and amounts in specified periods, 70 relation between time and intensity, 63, 64, 65 Preliminary investigations for sewerage design, 178, 179 field work, 178 maps, 178, 180, 181 office work, 180 Presses, for dewatering sludge, Besco-ter- meer, 424, 441 Bushnell, 439, 440 plate, 424 Shaefer-ter-Meer, 424, 441, 442 squeeze, 424 Prices, fertilizer, 446 fluctuation in commodity, 1910-1921, vii in fertilizer, 425, 426 grease, 447 of sewage as fertilizer, 546 INDEX 589 Prices, see also Cost data. ProblemB, application of rational method, 74, 75, 76, 77, 78. 79, 80, 81 map for use with, opp. 74 design of inverted siphon, 152, 153, 154 hydraulics of sewers, 124, 125 Springfield separate system, 49, 50, 51, 52, 53, 54 map for use with, opp. 48 tabulation of data for, 51, 53, 78, 79 Profiles, see Forms. Providence, R. I., analyses of raw sewage, 307 of water, 307 calcium hypochlorite, 531, 532 catch basin, 131 chemical precipitation, 372, 373, 375 sludge, 428 density of population, 28, 29 disinfection, 537 filter pressing, 439 hypochlorite plant, 536 industrial wastes allowance, 47 infiltration into sewers 36 method of determining, efficiency of sedimentation, 284 precipitation tanks, 395 ratio of water supply to sewage, 30 sludge disposal at sea, 438 suspended matter in settled sewage, 307 Puddling, backfilling, 253 Pumping, in wet trench work, 206, 208 sewage, 171 at Baltimore, 11, 12, 171 economic considerations, 171, 172 at MinneapoHs, 171 screening auxiliary to, 172 station at Boston, 174, 175, 176 at Waltham, 173, 176 storage auxiliary to, 172 at Washington, 171 Pumps, to handle unscreened sewage, 176 rating at Baltimore, 12 for wet trench work, 206 centrifugal, 206 diaphragm, 206 reciprocating, 206 steam vacuum, 206 Putrescibility^ American Public Health Association test, 291 defined, 14 tests, 290, 291, 292 Q Quantity, see Volume. Queens, Borough of, typical ciroillar sewer section, 112 Quicksand, defined, 213, 214 laying pipe in, 252 methods of dealing with, 214, 215 Quicksand, pockets in, 233 rarely rises in sheeted trench, 218 Quincy, Mass., ratio of water to sewage, 30 R Racks, cage, 341 design, 337, 338, 339 fixed. 337, 338 cleaning, 339 Frankfort, 343 Geiger, 343 inclined easily cleaned, 338, 339 Link-bar, 344 mechanical cleaning, 339, 340 movable, 341, 342 object of, 330 placing, 336 wing, 14, 343 cleaning, 343 see also Screens. Radial system of sewerage, 10 Rain gages, automatic, 55 FitzGerald, 57 Friez, 56 setting, 58 Rainfall, see Precipitation. Ramming, backfilling, 254 Rangers, 220, 221, 224, 229 bracing, 221, 234 cross-bracing, 220 defined, 213 on curves, 230 length, 221 placing, 220 placing lap blocks, 221 rectangular, 234 size, 233 spacing, 220, 224 stock recommended, 224 Rankine, W. J. M., formula for earth pres- sures, 216 Rapid sand filtration, requires high head, 21 Rational method, application of — problem, 74, 75, 76, 77, 78, 79, 80, 81 of estimating run-off, 67 examples of use, 73 Rats, as scavengers, 280 Reading, Pa., butterfly valve on trickling filters. 498 infiltration into sewers, 36, 39 removal of screenings, 350 Weand screen, 345, 350 Recorder, Builders Iron Foundry, water level, 87 Sanborn Co. water level, 88 Regulating devices, Boston, 163 Coffin regulator, 162 defined, 6 described, 162, 163 Washington, D. C, 164, 165 590 INDEX Relative stability, defined, 14 Relief outlets, increase hydraulic gradients in combined sewers, 95 sew^s, defined, 2 Residential sewage disposal, reason for omitting, viii Residue on evaporation, see Solids. Ribs, see Forms. Richmond, Va., sewer section, 118 Rochester, N. Y., density of population, 28, 29 grit chamber, 334 outlet, 169 screen, 350 screenings, 350 Rock, blasting, 202 character influences amount of explo- sive, 199 excavation, 198 classified, 191 sheeting, 214 underdrairdng, 207 hand drilling, 198, 199 ledge, 198 machine drilling, 200 treacherous if seamy, 214 Rodding, in soil survey, 180 through manholes, 150 Rolling backfilling, 254 Roof water, defined, 4 discharge into separate sewers, 9 Roofs, for sedimentation tanks, 358 for septic tanks, 368 Rosewater, Andrew, flush tank design, 160 Roughing basins at Worcester, 354 Royal Commission on Sewage Disposal, advertizes Glasgow sludge as fertilizer, 446 bacteria in English sewage, 297 contact beds, 450, 451, 452, 453, 454 dilution standard, 48 sewage farming, 542, 543 trickling filters, 474, 475 Run-off, at St. Louis, 73, 74 classified according to surface, 72 coefficient, 69-72 defined, 4 dry weather at Chicago, 41 effect of rate of rainfall on, 71 factor, 69-72 formulas, comparison, 80, 81, 82, 83, 84, opp. 85 quality determines disposal, 17 rational method of estimating, 67 retardation due to storage, 71 St. Louis, cableway for trench work, 197 density of population, 28 intensity of rainfall, 64 St- Louis, manhole, 136 periods of maximum precipitation, 70 rainfall curve, 66 run-off from residential districts, 74 sewer section, 117 sewerage map, opp. 73 street detritus, 99 use of rational method, 73 water supply contaminated by Chicago sewage, 329 St. Paul, Minn., density of population, 29 Salford, Eng., sludge disposal at sea, 438 Salinity, effect on sedimentation, 316, 317 of sea water, 320 use of salinoiheter, 320 Sampling, sewage, 281 water, 281 San Diego, Cal., tidal range, 317 San Francisco, Cal., density of population, 28 intensity of rainfall, 65 rainfall curve, 66 tidal range, 317 Sanborn Co., water level recorder, 88 Sand, effective size, 504 for intermittent filters, 503 inspection, 189 mechanical analysis, 505, 506, 507 pockets in, 233 quick, defined, 213 renewal on intermittent filters, 523, 524 requires sheeting, 213 sieving, 505 uniformity coefficient, 504 water-bearing, underdraining, 207 Sanitary sewer, term discontinued, 2 Savannah, Ga., density of population, 28 periods of maximum precipitation, 70 rainfall curve, 66 tidal range, 317 Screening sewage, coarse, 336 fine, 336 preliminary to pumping, 172 theory of, 335 Screenings, defined, 16 disposal of, 343 quantities at Boston, 350 Milwaukee, 350 New York, 350 Reading, Pa., 350 Rochester, 350 Worcester, 349 removal, 339, 340, 341, 346, 347, 348 Screens, defined, 14 band. 14 bar, 14 belt, 348 cage, 14, 341 Carshalton, 349 coarse, 14, 336 disc, 14, 347, 348 INDEX 591 Soreens, Dorrco, 345, 346 drum, 14. 344, 345 fine. 14. 336 fixed, 337, 338 for sand sieving, 506 Frankfort, 343 Geiger, 343 Link-bar. 344 medium, 14 mesh, 14 object of, 330 parallel-bar. 336 placing, 337 removal by, 336 Eiensch-Wurl, 347 Weand, 344, 345 wing. 14, 343 Scum, boards for Imhoff tanks, 391 for sedimentation tanks. 358 submerged. 391 defined, 16 ImhofE tank, 386, 392, 393 prevention of excessive, 393 septic tank, 363 Seattle, Wash., density of population, 29 manhole invert, 137 tidal range, 317 Sedimentation, in aeration tank of acti- vated sludge plant, 419 at Atlanta, 308 chamber of ImhofF tank, 391 in fresh and in salt water, 316, 317 at Chicago, 308 at Columbus, 308 at Gloversville, 355, 360 at Providence, 308 at Worcester, 352, 354 capabilities of hypothetical sewage, 351 detention period, 355, 356 determination of efficiency at Provi- dence, 2S4 factors affecting, 352, 353 part of activated sludge process, 418 percentage in short periods, 308 rate at Milwaukee, 410 removal of suspended solids by, 284, 351, 352 secondary, 495 sewAge maintained in natural condition, 395 of solids in bodies of water, 313, 316, 317 fiU-and-draw method, 356 horizontal fiow, 356 in standing versus moving water, 327 tanks, act as sedimentation basins, 351 baffles, 357, 358 character of sludge, 354, 395, 429, 430 cleaning' 359, 433 costs, 362 cylindrical, 354 Sedimentation, defined, 15, 351 density of sludge, 383 design, 356, 360 detention period, 355, 356, 381 function, 351 at Gloversville, 355 at North Attleboro, 434 at Toronto, 360 at Worcester, 353 horizontal Sow, 356 in series, 354 number needed, 356 outlet pipe, 358 rectangular, 354 roofing, 358 scum boards, 358 secondary tank sludge as fertilizer 445 types, 354 velocity through, 353, 419 vertical flow, 355, 359, 360, 361, 36i weirs, 357 with activated sludge process, 418 419 vertical flow, 355, 359 with chemicals, 370-374 Separate sewers defined, 2 see Separate system, in Baltimore, 10-12 in Brooklyn, 13 sludge digestion, defined, 16 system, advantages of, 9 caution reallowance for storm water 48 compared with combined, 8 deflned, 3 deposits in, 100 design problem, 49 disadvantages of, 9 minimum grades allowable, 98 supplemented by storm drains, 9 Septic tanks, arrangements of, 365 bacterial action in, 362 bottom slopes, 369 cleaning, 433 at Columbus, 365, 366, 367 defined, 15 draw-off pipes, 369 effluents, 362, 364, 494 at Fairmont, 365, 366, 367 inlets, 366, 368 at Mansfield, 366, 369 number needed, 365, 366 outlets, 366, 368, 369 roofing, 368 scum, 363 shear gates on sludge outlets, 370 size considerations, 366 sludge, character, 362, 363, 364, 395 430, 445 volume, 363 592 INDEX Septic tanks, storage capacity, 364 at Sturgis, 365, 366, 367 value of sludge as fertilizer, 445 at Washington, Pa., 365, 366, 368 Settling solids, defined, 14 Sewage, aeration, 395 analyses, see Analyses of sewage; An- alyses, bacteriological; Analyses, chemical, changes caused by bacteria, 296 chemical and biological character, 280 composition, 282, 303, 304, 305, 306, 307 Brockton, 307 Chicago, 307 general discussion, 282 influences period of contact in acti- vated sludge process, 407 Mansfield, 307 of American, English and German cities, 306 Providence, 307 sources of total solids, 303 suspended matter, 385 Worcester, 307 concentration, general discussion, 282 condition favoring disposal by dilution, 312 general discussion, 282 decomposition with loss of oxygen, 299 disinfection, 527 defined, 13 method influences sewage plan, 19 residential not included, viii see also Sewage treatment, digposal by dilution, 310 at Boston, 328 at Chicago, 329 at Cleveland, 328 conditions favoring, 311, 312 defined, 310 discharge at single point, 19 into Great Lakes, 20 into streams, 19 into tidal waters, 20 effect of temperature, 316 most natural method, 310 nature of, 311 percentage in United States, 311 preliminary studies necessary, 327 proportional flow requirements, 313, 314, 315 domestic defined, 2 effect of agitation, 396 fertilizer value, 542 flow, see Velocities, volume of sewage, fresh, 299 general characteristics discussed, 282 method of estimating volume, 23 num.ber of bacteria in, 296, 297 oxidation, defined, 16 pxirification, term abandoned, 13 Sewage, ratio of, to water supply, 30 septic, 299 solids in raw, 280, 330 stale, 299 treatment, activated sludge process, 395-426 before disposal, 19 chemical precipitation, 351, 370-376 contact beds, 448-466 defined, 13 dilution, 21, 310-329 disinfection, 527-540 Emscher tank, 377-394 grit chambers, 330-335 Imhoff tanks, 377-394 intermittent filters, 502—526 irrigation, 541-550 method determines location, 21 methods mentioned, 18 necessary at times, 17 previous to disposal by dilution, 310, 311 racks, 330, 336-344 screens, 330, 344-350 sedimentation, 351-362 septic tanks, 351-370 at several as compared with single point, 21 suitable sites, 19, 20 temperatiure affects, 316, 402, 403 trickling filters, 467-501 water content, 541 Sewer air, see Sewer gas. appurtenances, 126, see Individual items, capacity, by application of Kutter and Hazen and Williams formulas, 102 effect of undue size, 22 governed by future conditions, 23 surcharging, 93 construction, backfilling, brick, 255, 266 concrete, 255 costs, general, 262, 560 excavation, 190 imperviousness, 264 inspection, 188 jointing pipe, 248, 249, 250, 251 laying pipe, 246 lining,' 265 locating specials, 186 materials, 119, 262 payment, 185 pipe, 240 preliminary investigations, 178 quantity determination, 185 in rock trench, 264 sheeting and bracing, 212 stability, 263, 264 surveying, 178 in tunnel, 262 under drainage, 207, 251 INDEX 593 Sewer construction, use of cradles, 245 defined, 2 gas, provision against, 4 grades, 7, 179, 184 affected by method of disposal, 19 hydraulic, 93, 94 at invert, 94 marked on battens, 183 notes, 184 on steep hills, 12 surveying for, 182, 183 sections, analysis, 119, 261 basket handle, 110, 260 catenary, 105, 116, 259, 262 circular, 103, 110, 111, 112, 255, 256, 262 cunette, 118 delta, 106, 260 egg-shaped, opp. 102, 104, 115, 116, 256, 257, 262 excavation required for various types, 262, 263 Gothic, 110, 260 horseshoe, 105, 113, 260, 262, 264 hydrauhc elements, 103-109 influenced by construction methods, 261 inverted egg-shaped, 116 in tunnel, 259, 262 parabolic, 106, 260, 262 recovery of velocity head in transi- tion sections, 124 rectangular, 104, 260, 262 selection of, 109 semi-circular, 107, 117, 260, 262, 263, 264 semi-elliptical, 106. 114, 257, 258, 259 stability of inverts, 258 U-shaped, 118, 260 with brick arch, 111 systems, combined versus separate, 8 defined, 3 general description, 7 infiltration into, 35, 36, .37, 38, 39 influence of method of disposal and treatment, 19 of topography, 10 mapping, 180, 181 population served by, in United States, 311 see also Combined system; Separate system. Sewerage, definitions, 1-3 works, defined, 3 Sewers, appurtenances of, 126 brick, 255, 266 circular with brick arch, 111 compensation for curvature in, 120 concrete, 255 condition of interior, mirror method, 179 38 Sewers, depth, 7 at Baltimore, 12 design under minimum flow conditions 97 discharge of circular, opp. 102 egg-shaped, opp. 102 factor of safety in design, 92 flight, 100, 141 Baltimore, 141 flushing, 155 future conditions govern design, 23 hydraulics of, 92 imperviousness, 264 inadequate, 21 infiltration into, 35, 36, 37, 38, 39 liabihty for inadequate, 21 municipal liability for inadequate, 21 pipe, 240 population served in U. S., 311 rats in, 280 special structures for, 126 stability, 263, 264 subject to external forces, 119, 237 surcharging defined, 93 survey of, 179 Shaw, A. L., observation of inverted siphon, 151 Shedd, J. Herbert, dilution requirement, 314 Sheeting, beveled, 224 box, 226, 227 bracing, 212, 220, 226, 229, 230 care of banks, 233 defined, 213 depth driven, 222 driving, 222, 225, 235 earth pressures, 216 economies in, 230, 232, 233, 234 horizontal, 226 hydrostatic pressure, 218 inspection, 189 jetting, 230 length of, 225 lumber required, 231, 234 materials requiring, 213 necessity of in trenches, 212, 216 notched planks used, 225 overcomes lack of adhesion, 215, 216 plank caps, 235 clamp, 236 poling boards, 227, 228 puUing, 236 skeleton, 229 stay bracing, 229 steel sheet piUng, 235 stock for, 234 time element important, 216 trenches, 212, 213 typical, 223, 224, 232 use of hay, 233 vertical, 220, 221, 222, 223, 224. 227 water pressure on, 218 594 INDEX Sheeting, window, 226 Sheffield, Eng., agitation of activated sludge, 409 Sherman, C. W., rainfall formula, 66 Shoring, see Sheeting. Siphons, for dosing intermittent filters, 514, S15, 516, 517, 518 inverted, see Inverted siphons. Miller, 157, 161 timed, 458, 460, 461 for trickling filters, 497, 498 Van Vranken, 160 Slant, defined, 6 Slate bed, defined, 17 Sleek, defined, 16 Slopes, dangers of steep, 100 minimum in separate sewers, 98 of sewers, 97 Sludge, activated, 395-426, 430, 432 analysis of Worcester, 432 bacterial digestion, 402 beds 16, 422, 423, 437-439 burning, 439 characteristics, 429 chemical precipitation, 395, 428, 429, 430, 432 comparison by moisture content, 383, 384 composition of air-dried, 445 concentration, 401, 421, 428 defined, 15, 427 degreasing, 447 density, 382, 383, 384, 428 deposits on banks and shores, 312, 313, 316, 317 dewatering, 16, 424, 425, 439, 440, 441, 443, 446 disposal, 370, 437, 438 on beds, 422, 423, 437 for fertilizer, 438, 439 in lagoons, 438 at sea, 437 in trenches, 438 drying, mechanical, 425, 443 on beds, 422, 423, 437 fertiUzer value, 424, 443, 444, 445, 446 Fidler apparatus, 435 fiuidity, 370, 422, 431 Imhoff tank, 378, 379, 380, 391, 395, 431, 432 measurements, 409, 411, 429 moisture content, 382, 383, 384, 385, 401, 421, 422, 438, 446 nitrogen content, 423, 432 odor, 403, 430, 431 plain sedimentation, 354, 395, 429, 430, 432 pressing, 424, 439, 440, 441, 442, 446 pumping, 349, 394, 422, 437 removal by gravity, 434, 435 hand cleaning, 433 Sludge removal from Imhoff tanks, 393 394, 432-437 by mechanical means, 433 from sedimentation tanks, 354, 432 from septic tanks, 364, 365, 369, 432 by squeegee, 432 secondary sedimentation, 431, 432 septic tank, 362, 363, 395, 430 slopes, 432-435 specific gravity, 382, 383, 428 sumps, 433 volume, 427 volumetric determinations, 410, 427, 428, 429 weight, 428 Sludge cake, defined, 16 concentration, defined, 16 dewatering, see Sludge, defined, 16 digestion, defined, 16 drying, bed defined, 16 cost, 446 defined, 16 mechanical, 425 moisture in sludges, 382 odors during, 370 on beds, 422, 423 see also Sludge, dewatjring. pressing, cost, 446 defined, 16 described, 424, 441, 442 Socket, deep and wide versus standard, 245 dimensions of clay pipe, 243, 244 omitted in Washington, 245 Soil pipe, defined, 4 Solids, care in determination of, 283 determinations of, 283 dissolved, 283, 306 hydrolysis, 287 in colloidal state, 285 in sewage of American, English and German cities, 305, 306 methods of determining settling, 284 oxidation, 287, 299, 313 percentage in raw sewage, 280 removal by filtration as compared with sedimentation, 284 by grit chambers, 330 by racks, 330, 336 by screens, 330, 336 by sedimentation, 351, 352, 356, 360, 381 residue on evaporation, 283 settleable in hypothetical sewage, 351 sources of sewage solids, 303, 304, 305 suspended, 283, 306, 312, 360, 385 Southampton, Eng., sludge disposal at sea, 438 Southbridge, Mass., efficiency of sand filters, 520 intermittent filter operation, 508 INDEX 596 Special structures, see Individual items. Specials, establishing location, 186 Specific gravity, effects susceptibility to velocity, 96 Spencer, Mass., efficiency of sand filters, 520 intermittent filter operation, 508 Spiders on trickling filters, 500 Spigot, dimensions of clay pipe, 243 Spokane, Wash., rainfall curve, 66 Springfield, Mass., cost diagram for sewers, 561, 564, 565 hypothetical problem of sewer design, 49. 50, 51, 52, 53, 54 application of rational method, 75-81 maps for use with problems, opp. 48, opp. 74 Sprinkler nozzle, defined, 17 Sprinkhng filter, term abandoned, 17 Squeegees, for cleaning sedimentation chambers, 392 Stability, defined, 14 of effluents in cold weather, 403 of inverts, 258, 259, 260 numbers, Am. Pub. Health Association, 292 tests, 290, 291, 292 of trickling filter effluents at Lawrence, 495 see also analyses chemical. Stacks, defined, 4 Stearns, Frederic P., dilution requirement, 313, 314 used butterfly valve on trickling filter, 498 Steel, reinforcing, inspection, 189, see Concrete reinforcement. Sterilization, defined, 17, see Disinfection. Stockbridge, analysis of sewage, 308 Stone, crushed, inspection, 189, see Contact beds. Stoppages, in laterals, 7 in plumbing fixtures, 3 see also Deposits. Stoppers for closing branch sewers, 252, 253 Storage basins, auxiliary to pumping sewage, 172 Storm drains, allowance for maximum veloe- ity in, 99 defined, 2 in Baltimore, 10-12 method of tabulating data for, 78, 79 overflows, at Cleveland, 167, 168 defined, 6, 162 described, 167 supplement separate sewers, 9 water, defined, 2 discharge into separate sewers, 9 flows from large areas, 86 in combined system, 9, 19 methods of determining quantity, 55 Storm drains, water, provision in intercep- ters, 47 run-off formulas, opp. 85 separation of from house sewage, 19 treatment before discharge, IS Storm-water overflow sewer, defined, 2 Streams, absorption of oxygen by, 326, 327 Chicago Drainage Canal, 329 Cuyahoga River pollution, 328 danger of deposits in, 312, 313 dilution requirements, 313, 314, 315 Merrimack River pollution, 314, 315, 316 Monongahela, acidity, 527 Ohio streams pollution, 315 sewage discharge into, 19, 310, 31 1 temperature of infiuences disposal con- ditions, 316 tidal, 317 Street plan, influences sewerage plan, 12 see also Maps. Street wash, danger of deposits from, 47 deflned, 4 discharge into separate sewers, 9 discharge into streams, IS quaUty, 18 Sturgis, Mich., septic tank, 365, 366, 367 Sub-mains, defined, 2 depth, 7 prevention of backing up in, 163 selection of size, 7 Sub-surface irrigation, defined, 16 Sulphur and sand joints, 249, 250 cycle, 298 Surcharging, defined, 93 Surface irrigation, defined, 16 water, defined, 2, see Storm water. Surveys, base lines, 187 bench marks, 179, 182, 185, 187 elevations, 179, 184 grade marks, 183 location of specials, 185, 186 notes in field, 179, 183, 184 offset fine, 181 plotting notes, 180 preliminary, 178 quantity, 185 reference marks, 181 running profiles, 179 running sewer line, 181, 182 soil, 180 structural, 186 transferring line into trench, 182 Suspended solids, defined, 14 Syracuse, N. Y., manhole invert, 137 sewer sections, 114, 118 Talbot, A. N., bending moment for pipe, 240 formula for rainfall intensity, 63, 66 strength of pipe, 241 596 INDEX Tamping backfilling, 254, 255 Tank treatments, defined, 14 some create offensive odors, 21 see Sewage treatment. Tankage, fertilizer value, 445 Tanks, dosing, at Fitchburg, 485, 486, 489, 497 Tar joints, 250 Taylor, Wm. Gavin, motion of sewage in trickling filter, 479 Temperature, changes in deep ponds, 326 during slaking, 373 effect on bacterial action, 402 effect on disposal by dilution, 316 effect of winds on in Milwaukee Bay, 323 Templates for brick laying, 266 for concrete inverts, 271 Thermophone, use of 321 Tide gates, 168, 170 Dodd type, 170 function of, 162 Tidal range at various places, 317 Toledo, Oiiio, density of population, 29 Topography influences sewerage plan, 7, 12, 13 Toronto, Ont., cleaning fixed racks, 339, 340 sewer section, 112 value of activated sludge as fertilizer, 445 vertical flow sedimentation tanks, 360, 361 Trade wastes, term discontinued, 2 Traps, main, defined, 3 omitted, 4 on catch basins, 130, 131 Travis, Dr. Wm. Owen, tank design, 377 Travis tank, defined, 15 described, 377 Trenching machinery, advantage over handwork, 193 Austin trencher, 194 Buckeye traction ditcher, 194, 195 cableway, 195, 197 Carson, 195, 196 ditcher, 193 trencher, 193 Trickhng filters, at Andover, 500 Baltimore, 489, 491, 492, 498 Calvert, 481, 493 Columbus, 489. 490, 494 Dorking, 499 England, 474, 475, 480 at Fitchburg, 467, 473, 484, 489, 490, 491, 492, 497, 500 Gloveraville, 491, 492, 496, 500 Lawrence Experiment Station, 471, 475, 476, 477, 482 Malvern, 480 Mount Vernon, 483 Philadelphia, 500 Trickling filters, Plainfield, 500 Reading, 498 Springfield, 482 Waterbury, Conn., 491 accomplishment, 469 aerobic conditions, 467, 468 bacterial action, 468 character of influent, 472 clogging, 473, 479, 493 compared with contact beds, 500 continuous operation, 487, 488 cost, 473 defined, 17 depth, 472, 473, 474, 477, 478 described, 467 distribution of sewage, 469, 487 distributors, dosing tanks, 485 486 gravity, 482 nozzles, 482-487, 498, 499 revolving, 479, 480 siphons, 496 splash plates, 481 travelling, 480, 481, 482, 483 dosing, 473, 474, 488, 496 effluent, 469, 470, 476, 495 floor, 492 floor types, 491 growths, 473, 500 head on nozzles, 487 intermittent operation, 487, 488 lateral distribution in beds, 479 material of construction, 470, 471 472, 476 odors, 494 performance, 469, 470, 474, 475, 476 477, 478 preliminary treatment of influent 494 removal of bacteria, 469, 470 rest period, 488, 489 risers, 490 sedimentation of effluent, 495, 49( size and shape,. 494 stabiUty test, 495 theory, 467, 468 time of passage through, 471, 472 474, 478 underdrainage, 490, 491, 492 unloading, 468 valves, 498, 499 varying head, 487 ventilation, 493 winter operation, 468, 498 worms, 499 Trunk sewer defined, 2 Turneaure and Maurer, application o. elastic theory, 119 Typhoid fever, epidemics, 301 modes of transmittal, 301 patients discharge B, typhosus, 301 INDEX 597 u Umatilla Condiiit, gagings, 120 Underdrains, defined, 5 for wet excavation, 207 inspection, 189 United States Bureau of Surveys, rainfall formula, 66 United States Hydrographic Service, study of currents, 321 United States Public Health Service, definition of biochemical oxygen demand, 292 United States Reclamation Service studies, losses due to curvature, 120 recovery of velocity head, 123 University of California, experiments of Lipman and Burgess on fertilizer value of sludge, 444, 445 University of Illinois, studies, Bartow and Mohlmann, 398, 399 Crawford and Bartow, 407 diffuser tests of Bartow and Mickle, 415 sludge produced by activated sludge process, 421 Urea, gives off nitrogen when heated, 288 nitrogen cycle, 286 Valves, butterfly, 499 varying head, 498 Weand, 498 Velocities of flow, absorption of head, 123 at different depths; 108 dangers of high, 99, 100 depend on head and friction, 93 formulas for computing, 100 comparison of Kutter and Hazen and Williams, 102 Hazen and Williams, 101 Kutter, 101, 104 head, 123, 124 in cast iron pipes, 93 in circular sewers, opp. 102, opp. 107 in egg-shaped sewers, opp. 102, opp. 107 in inverted siphons, 150 in junctions, 145 in partly filled sewers, 108 maximum, 99 minimum, 96 necessary to move solids, 96 should be maintained, 98 transporting power, 95 through contact beds, 449, 450 through grit chambers, 331, 332 through sedimentation tanks, 350, 355, 356, 357, 360 through sludge pipes, 422 Velocities of flow, through trickling filters, 471, 472, 474, 478 Ventilation of house sewers, 4 through perforated manhole covers, 6, 7 Venturi meter, for measurements in acti- vated sludge process, 411 recovery of velocity head in, 124 Vitrified pipe sewers, see Pipe, vitrified. Volume of sewage, at Amherst, 508 at Andover, 508 at Attleboro, 508 at Brockton, 508 at Cincinnati, 46 at Clinton, 508 at Concord, 508 at Framingham, 508 at Gardner, 508 at Hopedale, 508 at Hudson, 508 at Leicester, 508 at Lynn, 43, 44 at Marion, 508 at Marlboro, 508 at Milford, 508 at Natick, 508 at North Attleborough, 508 at Northbridge, 508 at Norwood, 508 at Pittsfield, 508 Southbridge, 508 Spencer, 508 Westboro, 508 Worcester, 508 discharge at different depths, 108 factor of safety allowance, 93 from commercial districts, 45 from districts in Chicago, 41 from industrial districts, 47 from partly filled districts, 108 from residential districts, 43 gaging, 86, 150 hourly flow, 42 in American, iE^nglish and German cities, 306 method of estimating, 23 per day for Massachusetts cities, 40 ratio of to water supply, 30 uncertainty in estimating, 92 under dry weather conditions, 98 Vouasoir arch method, 261 W Wachusett Aqueduct section, 113 Waltham, Mass., sewage pumping station, 173, 176 Washington, D. C, cage racks, 343 cradles, 245 exceptional growth, 24 periods of maximum precipitation, 70 598 INDEX ■Washington, D. C, pipe joints, 245 pumping sewage, 171 regulating chamber on Piney Creek sewer, 164, 165 sewer section, 118 Washington, Pa., septic tank inlet, 366 outlet, 366 tanks, 365, 366, 367 Waste of water, prevention by metering, 31 pipe, defined, 4 Water, bacterial standards, 301 composition, see Analyses of water. fresh versus salt to receive eflBuents, 316, 317 limitations of chemical analyses, 281 oxygen absorption by quiet and running, 326 salinity of sea, 320 supply, amount for flushing sewers, 156 consumption- — classification, 3 1 committee on, 30, 32 fluctuations, 32 in cities, 30 in different parts of city, 31 prevention of waste by metering, 31 proportion reaching sewers, 28 ratio of to sewage, 30 transmits typhoid, 301 transporting power, 95, 96, 97 Waterbury, Conn., sewer section, 113 trickling filter floor, 491, 492 Watson, Jn. D., vertical flow tank design, 360 Weand, O. M., butterfly valve, 498 Webster, Geo. S., bacteria in Philadelphia sewage, 297 sludge at Philadelphia, 383, 384 Wedge blocks, objectionable with sewer pipe, 251 on brick sewer construction, 268 Weirs, for overflow at Cleveland, 168 on Baltimore Imhoff tank, 386, 387 on sedimentation tanks, 357 Wellesley, Mass., water consumption, 32 Wellholes, 141 Cleveland, 140 defined, 5 Westboro, Mass., efficiency of sand filters, 520 Westboro, Mass., intermittent filter opera- tion, 508 leakage into sewers, 39 sewage flow, 40 Williams, Gardner S., Hazen and Williams formula, 101 Williams, Gardner S., Hubbell and Fenkell, velocity of flow studies, 93 Wilmington, Del., experience with inverted siphon, 151 sewer sections, 112, 117 Wind, effect on chlorine distribution, 324 on lake currents, 320 on temperatures of bottom waters, 325 on vertical distribution of bacteria in waters. 323 Window in sheeting, defined, 226 Wisner, Geo. M., costs of sedimentation tanks, 362 reports on Chicago Drainage Canal, 329 Worcester, Mass., activated sludge process found impracticable, 405, 422 analysis of raw sewage, 307, 308 of sludges, 432 of water, 307 chemical precipitation at, 372, 374, 375, 428 efficiency of sand filters, 520 fertilizer value of air-dried Imhoff tank sludge, 445 filter pressing, 439, 441 grit chambers, 332, 333 Imhoff tank sludge, 385 infiltration into sewers, 37, 38, 39 intermittent filter operation, 508, 524 lime added in filter pressing, 439 measuring sludge, 429 nature of street detritus, 99 nozzles, 483, 485 precipitation tanks, 375 ratio of water to sewage, 3D scale of odors, 295 secondary sedimentation tank sludge, 431 sedimentation tank, 353 sewage flow, 40 sewer section, 115, 116 spacing of manholes, 134 Worms in trickling filters, 499 CONSERVATION Ilii