CORNELL UNIVERSITY LIBRARY FINE ARTS LIBRARY CORNELL UNIVERSITY LIBRARY 924 099 175 246 Cornell University Library The original of tiiis 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/cletails/cu31924099175246 In compliance with current copyright law, Cornell University Library produced this replacement volume on paper that meets the ANSI Standard Z39.48-1992 to replace the irreparably deteriorated original. 2004 DESIGN OF THE TURKISH BATH S By J. J. COSGROVE Author of 'PRINCIPLES AND PRACTICE OF PLUMBING' "SEWAGE PURIFICATION AND DISPOSAL" "HISTORY OF SANITATION" "WROUGHT PIPE DRAINAGE SYSTEMS" "PLUMBING ESTIMATES AND CONTRACTS" "PLUMBING PLANS AND SPECIFICATIONS" Published by Standard ^amtsLVS tPfe. Co. Pittsburgh, U. S. A . PREFACE I N PRESENTING this volume the author adds another to a set of books devoted exclusively to plumbing and sanitation. The several books included in the set are ' 'Principles and Practice of Plumbing," "Wrought Pipe Drainage Systems," "Sewage Purification and Disposal," "Plumbing Plans and Specifica- tions, " " Plumbing Estimates and Contracts, ' ' "History of Sanitation, " and "Design of the Turkish Bath." These books are not disjointed fragments put out from time to time, without reference to one another, but all are parts of one comprehensive whole, the result of a sys- tematic division of subjects planned before any of the books were written. The object was to present in one volume the principles that underlie the practice of plumbing, and how those principles are applied to practice. In another, a thorough explanation of the wrought pipe systems of drain- age together with an exhaustive description of how to handle and manipulate iron, brass and nickel-plated pipe. A third volume was to explain the art of preparing plumbing plans and writing specifications, and a fourth the principles of estimating, contracting and conducting a business. Closely related to plumbing is sewage disposal and general sanitation, so a fifth volume was devoted to the purification of sewage, then, that the plumber might have a greater historical knowledge of his calling, a history of sanitation was written and pubhshed. In "Design of the Turkish Bath" it was the author's aim to present in as simple and comprehensive a manner as possible the method of constructing various types of Turkish Baths, their details and principles, together with descriptions of the materials and appliances best adapted to the purpose This volume contains rules, tables and data for de- signing and proportioning Turkish Baths. The principles and data worked out by experiment and experience are so widely scattered through separate reports, public and private documents that they are not in available form. For this reason and owing to the present active interest in the sub- ject it is hoped that the "Design of the Turkish Bath" will fill the want in the field of Sanitary Engineering literature. These books, the author believes, cover very fully the various subjects, and contain such technical data as has long been needed by those interested in Plumbing and Sanitation. J. J. COSGROVE Philadelphia, Pa. February 15th, 1913. PUBLISHER'S NOTE iN THE publisher's note, printed in the vari- ous books which preceded this volume, we en- deavored to make clear our connection with the publication of technical works on plumb- ing and sanitation, and for the benefit of the readers of this volume, we herevsrith repeat the story of our relation as Publisher's and Manufacturer's. The primary object of our organization is, as uni- versally known, to manufacture and market "Standard" Plumbing Fixtures, Brass Goods, and other products made in our factories. In the development of an organization to accomplish this result, there has been established an Adver- tising and Publishing Department of no small proportions, and "Design of the Turkish Bath" is a part of the work of this department. One of the most widely known productions of the Publishing Department is the monthly magazine, "Modern Sanitation," which was established in June, 1904. From this came the publication, first in serial form and later as books, of "Principles and Practice of Plumbing," "Sewage Purification and Disposal," "History of Sanitation,'' "Wrought Pipe Drainage Systems," "Plumbing Plans and Specifications," "Plumbing Estimates and Contracts" and "Design of the Turkish Bath." These books are all the work of the same author, Mr. J. J. Cosgrove, who is recog- nized as one of the leading authorities on the questions upon which he writes. Mr. Cosgrove's first work was "Principles and Prac- tice of Plumbing." This book has been phenomenally suc- cessful and has been adopted as a text book in more than 40 of the largest universities and colleges in the United States. Several of his other works are similarly used either in the form of text books or as books which students in plumbing and architecture are advised to read. In "Principles and Practice of Plumbing," "Sewage Purification and Disposal," "History of Sanitation,'' "Wrought Pipe Drainage Systems," "Plumbing Plans and Specifications," "Plumbing Estimates and Contracts," and "Design of the Turkish Bath," we feel that the literature of the craft has been enriched in an enduring manner and that we have fully justified our appearance in the field of publishers as amply as we have our standing as manufac- turers of a world wide known and used product. Staitdard ^anitats'iPfe.Co. Pittsburgh, U. S. A. Publishing Department. TABLE OF CONTENTS Page Universality of the Sweat Bath 1 Hygiene of the Bath 13 Layout of a Turkish Bath 33 Insulating Turkish Bath Rooms 43 Heating Turkish Bath Rooms 57 Proportioning the Heating System 71 Loss of Heat from Buildings 87 Ventilation of Turkish Baths 103 Fans or Blowers 121 Air-Distributing Systems 145 Turkish Bath Details 157 Example of a Turkish Bath 183 LIST OF TABLES Table Page I. Heat Emission from Horizontal-Tube Radiation.. 79 II. Properties of Saturated Steam 82-83 III. Heat Loss from Windows 90 IV. Heat Loss from Skylights 92 V. Heat Loss from Walls 94 VI. Heat Loss from Floors and Ceilings 96 VII. Flue Area for Different Volumes and Velocities of Air 112 VIII. Size of Round and Square Pipes for given Areas.. 114 IX. Volume of Air Discharged at Different Pressures 127 X. Capacities of Straight-Blade Centrifugal Fans.. 129 XI. Capacities of Multivan Fans 130-131 XII. Size and Capacity of Cone Fans 133 XIII. Heat emitted by Coils to Air from Fans 138 XIV. Size of Steam and Return Pipes for Coils 139 XV. Properties of Air 142 XVI. Relative Humidity of Air 154 XVIL Size of Drain Pipes 164 LIST OF ILLUSTRATIONS FIG. PAGE The Fleischman Baths, Frontispiece 1 Ruins of Baths of Caracalla 2 2 Ruins, Thermae of Caracalla 3 3 Mosaic from the Floor of the Baths of Caracalla 4 4 Frame for Toholua or Sweat House 5 5 Sweat House used by Navajo Indians 6 6 Old Sweat House in Ireland 9 7 Microscopic View of Healthy Skin 14 8 Oil Gland, Oil Duct, and Hair 15 9 Sweat Pores in Palm of Hand 16 10 Sweat Ducts and Glands.. 16 11 Conventional Floor Plan of Bath 34 12 Fireproof Insulation for Masonry Walls 45 13 Wooden Insulation for Masonry Walls 46 14 Insulation for Frame Walls 47 15 Insulation for Fireproof Ceilings 49 16 Insulation for Wooden Ceilingfs 50 17 Insulation for Ground Floor 52 18 Insulation for Windows. .. 53 19 Insulation for Doors 64 20 Continuous Flat Steam Coil 64 21 Return Manifold Steam Coil 66 22 Angle Wall Steam Coil ... 67 FIG. PAGE 23 Wall Radiators 68 24 Steam Pipe through Wall 77 25 Example of Ventilating System 108 26 Air Filter 116 27 Peerless Air Washer 119 28 Centrifugal Fan 122 29 Cone Fan 123 30 Pipe Coil Heater 135 31 Reheater in Single Duct. . . 146 32 Double Duct 147 33 Inlet and Outlet Registers 149 34 Automatic Regulator 152 35 Humidostat 155 36 Plunge Tank 158 37 Hanging Plunge Bath 159 38 Waste Connections to Plunge Tank 161 39 Hand Rail Support 165 40 Heater Connections to Plunge Bath 168 41 Feed- water Heater Con- nections to Plunge Bath 171 42 Entrance to the Dressing Room 184 43 The Calidarium, Hot Room 185 44 Steam or Vapor Room 186 45 The Laving Rooms 187 46 Douche Bath 188 47 The Plunge Bath 189 48 Massage Room 190 49 Pedicure Department 191 50 The Solarium 192 DESIGN OF THE TURKISH BATH CHAPTER 1 UNIVERSALITY OF THE SWEAT BATH ROMAN BATHS H A T is known to people of the pres- ent day as the "Turkish Bath" is merely the highly evolved form of sw^eat bath ' ' practised for ages by both civihzed and savage races.' It would seem that human experience the world over has been the same, and widely scattered races of people, deducing certain truths from their personal experiences, early learned that if the pores of the skin were kept open, so that a copious flow of perspiration could be easily induced, beneficial results and improved health were sure to reward the effort. This led to communities establishing bath houses, crude or elaborate according to the wealth of the citizens or the advancement they had made in the building arts. In Rome, no doubt, sweat baths reached the highest form of development. There the benefits of the sweat bath were known to the public at large, who availed themselves of the advantages 1 D e s i g f the Turkish Bath of the public baths, which were held in such high and well deserved esteem. PHny, an early writer, credits the general good health of the Romans of that period to the frequent, almost daily, use of the hot room, and is authority for the statement that during the six hundred years of the public baths the Romans needed no medicine but the Therm £6. Fit:. 1 Ruins of Baths of Caracalla Ruins of the Roman baths, see Figs. 1, 2 and 3, still exist, which show the former luxury and splendor of these establishments; and volumes have been written describing not only the general plans and details of the structures, but recounting, likewise, the habits and routine of the bathers, the use of the various compartments and the prac- tices followed by different physicians. From all that can be learned of the Roman practice of bathing it would seem that the bather 2 Design of the Turkish Bath first exposed himself to the temperature of the hot rooms, and, after going through the regular course of perspiration there, washed himself thoroughly with warm water, then closed the pores of the skin by the application of cold water, after which he was thoroughly massaged and the skin was anointed with oils, emollients, or skin foods. A striking coincidence in this regard is that wherever the sweat house or hot bath has been established practically the same course of treat- Fin. 2 Ruins, ThermiE of Caracalla ment was followed, showing that human exper- ience in all quarters of the globe taught the neces- sity of closing the pores after sweating and re- placing the natural oils of the body removed by bathing. SWEAT BATHS OF THE INDIANS From the early Romans to the American Indians is a far cry, yet no more striking example 3 Design of the Turkish Bath of the widespread use of sweat baths could be cited than to tell the story of the American In- dians' bathing establishments. Every Indian village had its sweat house when the white man first came among them, so that the idea must have originated in America at the same time that it was in vogue in Rome, and could not have been borrowed from the white people. The sweat house of the Indians, as a rule, was built Fig. 3 Mosaic from the Floor of the Baths of Caracalla after the stj-le of an ordinary Indian lodge, as may be seen in Fig. 4, with the exception that no open- ing was left near the top to serve as a smoke flue and ventilator. On the contrary, the framework of wood was covered with animal skins, sewed Design of the Turkish Bath toi^^ether with tight seams to prevent the passage of air from within or without. Inside of the sweat house was built a rectangular heater made of stone walls about six feet long, three feet wide, and three feet deep. Every lodge in the village had a woven willow basket or crib, which fitted the heater on which it rested. The basket was made large enough for a grown person to recline on ft *■■»• /'// ! ' '.. • ■ " , - J — ^^ - - Fig. 4 Frame for Tuh<.>lua, or Sweat House of Indians almost at full length and was supported in place by cross poles which prevented the crib from drop- ping down into the heater. When a brave wanted to take a bath a squaw built a fire outside of the sweat house and in this fire heated a number of stones to a red heat. These stones she then piled inside of the heater in the sweat house, arranged the crib in place, put a 5 Design of the Turkish Bath mat woven of wild sage and other aromatic herbs into the crib and the bath was ready for the brave. On the mat of wild herbs in the basket the Indian placed himself in a sitting posture and the squav/ then poured cold water over the stones. The vapor, charged with the scent of the herbs, arising from the heater soon enveloped the bather, causing him to break out into a profuse perspiration, which continued as long as hot stones were placed in the Fig. 5 Sweat House Used by Navajo Indiana heater. When the Indian had sweated to his heart's content he arose from the crib, ran to the river and plunged into the water. Emerging from the river he wrapped himself in a buffalo robe, ran to his lodge, where he was rubbed down and oiled with bear grease, then laid down, with his Design of the Turkish Bath feet to the fire, to sleep, wrapped in the buffalo robe with the fur turned inside. The sweat house of the Indians was not always made of poles covered with animal skins. The style of lodge used for this purpose differed with the tribes and with their habits of living. Among the nomadic tribes, which were constantly moving from place to place, the tepee was made of animal skins as shown in the preceeding illustration, but among the more settled tribes sweat houses built of sods or sun-dried bricks, or dug out of the earth, were used instead. An illustration of a sweat house of this description belonging to the Navajo Indians may be seen in Fig. 5. This sweat house differs but little from caves used for a simi- lar purpose by the natives on the far-away British Isles. When the sweat bath of the Indians is com- pared with that of the Romans it is found to differ chiefly in its crudeness. So far as the routine ob- served is concerned there is practically no dif- ference between the two. First, the brave sub- jected himself to the heat of the sweat room, which took the place of the tepid and thermae chambers in the Roman baths. The next process, that of washing in hot water, was omitted, prob- ably for want of suitable facilities. The third operation, consisting of closing the pores, was accomplished by plunging in the river, which the Indian used as the Roman did his plunge bath. The rubbing down or massage, of the Indian and subsequent anointing with oil in no wise differed 7 Design of the Turkish Bath from that of the Roman, outside of the oils or unguents used. There is one process, however, in which the bath of the Indian differs materially from that of the Roman, and it will be well here to note the difference, for upon this one process rests the chief distinction between what are now known respect- fully as Turkish baths and Russian baths. In the Roman or Turkish baths, both as practised now and by the early Romans, the bather was subjected to a dry heat in the sweat rooms. Fire and hot gases cir- culated through flues in the walls of the old Roman baths and played beneath the floor of the hot room, so that the interiors of these rooms were like the inside of a baker's oven — dry, but extremely hot. In the Indian sweat house, on the other hand, the room was filled with vapor, so that instead of be- ing a dry heat to which the bather was exposed it was distinctly a wet one. That is the chief dif- ference between the Turkish bath and the Russian bath, to be described later on. The Turkish bath induces perspiration by means of dry heat, the Russian bath by means of vapor. IRISH SWEAT BATHS In Scotland and Ireland a sweat bath, very similar to that practised by the Indians, was likewise in vogue, and the fame of these baths became so great that it spread to Europe, where even at the present time in some localities sweat baths are known as Roman-Irish baths. The Irish bath houses, however, were very crude affairs, not at all to be compared with the baths of Rome, as may Design of the Turkish Bath l>e seen by a reference to Fig. 6, which shows a typical Irish dug-out sweat house. The baths were built close to the ocean or lake oi' running water, so that, like the Indians, the bather would have plenty of water for the operation and a pool in which to plunge when through with the heating Fig. 6. Old Sweat House in Ireland. process. In the Irish sweat house a fire was built of peat on the floor and the fire was maintained until the walls were heated to a high temperature. The fire was then withdrawn, the bather entered and the hole by which he entered was closed to 9 Design of the Turkish Bath keep the interior from cooling. Water was passed in to him as often as he wished it, to supply the moisture wrung from his body by the high tem- perature' of the sweat house, or to wash himself, as the case may have required. NATIVE RUSSIAN BATHS The Russian bath, as practised in that country, differed but little from the bath of the Indians, as may be seen from the following description of the native Russian baths, taken from Kohl's "Russia": "The passage from the door is divided into two, behind the check takers' port, one for the male and one for the female guests. We first enter an open space in which a set of men are sit- ting in a state of nudity on benches, those who have already bathed are dressing, while those who are going to undergo the process are taking off their clothes. Round this space or apartment are the doors leading to vapor rooms. The bather is ushered into them and finds himself in a room full of vapor, which is surrounded by a wooden platform rising in steps to near the roof of the room. The bather is made to lie down on one of the lower benches, and gradually to ascend to the higher and hotter ones. The first sensation on entering the room amounts almost to a feeling of suffocation. After you have been subjected for some time to a tem- perature which may rise to 145° the transpiration reaches its full activity and the sensation is very pleasant. The bath attendants come and flog you 10 Design of the Turkish Bath with birch twigs, cover you with the lather of soap, afterward rub it off, and then hold you over a jet of ice-cold water. The shock is great but is fol- lowed by a pleasant feeling of great comfort and of alleviation of any rheumatic pains you may have had. In regular estabhshments you go after this and lie down on a bed for a time before issuing forth. But the Russians often dress in the open air, and instead of using the jet of cold water go and roll themselves in the snow." In the numerous villages throughout the rural districts the same means employed by the early Indians are still used by the Russians for heating their bath houses or rooms. The bath house con- sists usually of two rooms, one for dressing and undressing, and the other the hot or vapor room. In the vapor room a heating stove is located, and on this stove or in its oven ordinary paving stones are heated very hot. Water is then poured over these stones to generate the vapor which fills the room. In Japan the hot-water bath and sometimes a vapor bath are used, the object of either being to open the pores and cleanse the blood through the sweat glands in the skin. It will thus bei seen that people widely remote from one another — so remote, in fact, that they could not have adopted the practice from one cer- tain place — each as the result of their own exper- ience, proved to their own satisfaction the value of sweat baths. If these people were right in their deductions, as they certainly were, there must be good reasons why heat baths are beneficial to the 11 Design of the Turkish Bath health. Such, indeed, is the case, but in order to understand fully the benefits derived from the Turkish or Russian baths it will be necessary to point out first some of the chief characteristics of the skin and describe the functions of the numer- ous glands embedded therein. 12 CHAPTER HYGIENE OF THE BATH FUNCTION OF THE BATH THE SKIN T R U C T U R E of the Skin.— If a section of a piece of healthy skin be viewed through a microscope it will be found, as shown in Pig. 7, to be a many-functioned organ instead of a mere covering to protect the external surface of the body and support the internal organs. The outer, or scarf skin, will be seen to be made up of numerous scalelike cells of a horny consistency, which are being shed constantly and replaced by younger cells from below. Immediately beneath the outer, or scarf skin, is a layer of pigment cells. It is the difference in quantity and color of pig- ment deposited in these cells which gives to differ- ent races their characteristic colors. Indenting this pigment layer at frequent in- tervals, so frequent, in fact, that fifty-seven thous- and may be grouped together in one square inch of 13 Design of the Turkish Bath space, are little elevations, or centers, where blood vessels and nerve fibers terminate. Underneath the terminal of nerves and blood vessels is the true skin, which is sensitive. Beneath the true skin are fat cells embedded in fleshy tissue. Bedded deep in the true skin e are a myriad of sweat glands with their little / ducts leading through the various layers to # the surf ace of the body. The roots of hairs are/ likewise embedded in L it e e e ac ons tvhere Pi^meniCe Oil elands — True 5kin -^ ^i ^5weot DuiLi Sweat C/and3, Fat Cells. Fip. 7 Microscopic View of Healthy Skin the true skin, and the stems extend out through oil ducts, which serve to conduct forth the oil secreted by the glands. A sectional view through an oil gland, oil duct and hair is shown, greatly enlarged, in Fig. 8. The root of the hair has its origin below the oil gland, where it communicates with a little nerve cell to give it the sensation of 14 Design of the Turkish Bath touch, and with a blood vessel to supply it with nourishment. Up through the center of the oil gland grows the hair and passes out to the surface of the skin, through the little oil duct, through which the gland pours forth its lubricating and softening fluid to the surface of the body. Oil glands, which are quite numer- ous, are scattered over the entire surface of the body, a gland and duct being found wherever there is a hair, as well as in many places where there are no hairs. The total number of hairs on the human body may be judged from the fact that there are estim.ated to be one hundred and twenty thousand hairs on a normal scalp. More important, perhaps, than the oil glands and ducts are Fig. 8 Oil Gland, Oil Duct and Hair the numerous sweat glands embedded in the skin and the minute ducts which conduct their fluids out of the body. On the cheeks there are only about five hundred and fifty sweat glands per square inch; on the forehead there are twelve hundred glands per inch, and on the soles of the feet and palms of the hands, where they are the most numerous of any part of the body, they number as high as twenty-seven hundred per square inch. The little sweat tubes where they pierce the outer skin of the palms of the hands can be seen by a microscope dotting the curves, circles and whorls which make up the in- tricate patterns of the skin. These tubes can be 15 Design of the Turkish Bath Fig. 9. Sweat Porea in Palm of Hand. seen, greatly enlarged, in Fig. 9. According to careful computation there are about one million five hundred thousand sweat glands in the entire body, the total length of which has been variously esti- mated at from 2i to 28 lineal miles. When the number is conservatively esti- mated, however, the total length of perspiratory tubes will be found in the individual of average size to be less than 3 lineal miles. If the coils of the little sweat glands shown in Fig. 10 were un- raveled and stretched out in a straight line it would be found that the combined length of sweat duct and coils of the gland measured about one-fifteenth inch. If the total number, one million five hundred thousand, were then placed end to end, they would be found to measure about 166,666 lineal inches, or 2.63 miles, in length. It will be noticed that near the mouth of the tube the little sweat ducts are not straight, but take a tortuous or spiral course, like the coils of a spring or the spirals of a corkscrew. Functions of the Skin. — When speaking of the skin, not only the scarf-skin and cutus, or true skin are meant, but likewise are included the various organs, glands, ducts, nerves and blood 16 Fig. 10. Sweat Glands and Ducts. Design of the Turkish Bath vessels embedded in the skin. The outer, or scarf- skin, being intended principally as a protective covering for the body, is admirably suited to that purpose. Made up as it is of numerous layers' of scalelike cells, composed of a horny substance, like the hoofs and horns of cattle, it is tough and will withstand an incredible amount of wear, while at the same time it is soft and pliable. This layer of the skin contains no blood vessels, so that it will not bleed when cut or scratched, and as it pos- sesses no nerve centers is utterly devoid of feeling. The outer scales of this layer are constantly peeling off, exposing younger and more tender ones below, which are better suited to normal conditions. When, however, certain parts of the body are ex- posed to great wear — as, for example, the palms of the hands— a thick layer of horny scales or callus forms to protect those parts. The dead scales, which have become detached from the skin, or which hang like ragged particles to the surface, require periodical removing or they will putrefy on the body and perhaps clog some of the oil ducts or sweat pores. To keep the outer skin, or scarf-skin, in a soft, smooth and pliable condition the various oil glands pour out their secretions of fatty matter, or oil, to lubricate the surface of the body. When for any reason the action of the oil glands is interfered with or the ducts become ob- structed no oil can be poured out onto the skin, which then becomes harsh and rough to the touch and, probably, chapped. The chapped hands most boys have in winter weather will readily present 17 Design of the Turkish Bath themselves to most people as an example of such lack of activity of the oil glands, vsrhile the remedy, application of oil or grease to the rough parts, is simply doing artificially what nature does automati- cally, and without our being conscious of the fact. A turkish bath will likewise remedy the chapped condition by cleansing the surface of the skin and allowing the oil and sweat glands to perform their normal functions. The pigment cells under the scarf-skin play a more important part in maintaining perfect health in the individual than would seem possible to the casual observer. The function of the pigment de- posited in these cells is to prevent the penetration of harmful rays of light into the deeper tissues of the body. The action of the nervous organism in filling these pigment cells when necessary for the protection of an individual will be recalled in the " tanning" of dark-complexioned people when ex- posed to strong sunlight, and the sunburn, with subsequent tanning, of light-colored people when exposed to the direct rays of the sun. So far as the bath is concerned the pigment cells are of no further interest, for the bath has no effect on the cells and the cells in no way interfere with the bath. In institutions where light baths are given, on the contrary, pigment, or the lack of pigment, in the cells would have to be considered, for not only has the pigment, or lack of pigment, a direct effect on the person subjected to the light, but, conversely, the intensity and color of light ought to be tempered to the color and amount of pigment in the cells. 18 Design of the Turkish Bath The nerve centers which terminate in little elevations just beneath the layer of pigment cells impart to the skin the sense of touch, so that it will be sensitive and respond to influences of heat, cold, electricity or friction. The blood vessels, working in harmony with the nerve centers, drive the blood from the surface of the body at times, then, responding to some reaction, like heat or massage, draw it again to the surface and the skin takes on a healthy glow. Of the numerous functions performed by a healthy skin there is, perhaps, no one gland, duct or vessel which plays a more important part than the little sweat glands. Through their pores these little glands, when in a healthy condition, throw off from the blood about two pints, or two pounds, of waste matter daily. In doing so the action of the glands is continuous, not intermittent, and the waste matter thrown off is generally in the form of invisable vapor, though when the body becomes warm from exercise or through excessive tempera- ture of the atmosphere the vapor becomes visible in the form of perspiration. In this respect the sweat glands perform the glandular functions of the kidneys, and when the sweat glands refuse to perform their functions an extra burden is imposed upon the kidneys. This in itself would not be so serious a matter if the sweat glands and kidneys performed the same functions and were interchange- able in that respect, as they seem to be. As a matter of fact, however, they are not, for the func- tion of the kidneys is to void fluids of an acid reaction, that being the normal reaction of urine, 19 Design of the Turkish Bath while the sweat glands excrete an alkaline solution, that being the normal reaction of perspiration. By exciting either of the organs they can be made ap- parently to perform the function of the other, but it is extremely doubtful if they actually do so. By exciting the sweat glands the fluid which would normally be voided by the kidneys can be carried off from the systems in the form of perspiration ; but what becomes of the poisonous or deleterious acid constituents of that urine — are they carried off likewise by the sweat glands, which were not designed for the purpose ? It would seem not ; and, if not, such impurities will remain in the blood, accumulating with time, until the health of the individual becomes undermined. On the other hand, by exciting the kidneys much of the fluid re- quired for perspiration may be carried off as urine. But, as the kidneys are intended only to filter out the acid wastes, what becomes of the alkaline mat- ter which should be carried off through the sweat ducts ? It would seem that each of the organs — the kidneys and the sweat glands — have separate and distinct functions to perform, and that one cannot successfully perform the functions of the other; even if it could it would seem an unwise policy by neglect of the person to require the kid- neys to carry off daily the two pounds of matter ordinarily excreted by the sweat glands, or the sweat glands to carry off the waste ordinarily voided by the kidneys. That the kidneys or other organs cannot carry off the poisonous matters ex- creted by the sweat glands is proved by the fact that if the skin be varnished death will quickly 20 Design of the Turkish Bath result, due, no doubt, to the retention of some poisonous substance the nature and production of which are not understood; while the kidneys can be put out of service for a much longer time with- out fatal results. To say that the sweat glands throw off two pounds of matter daily in the form of invisible perspiration is to state a fact without interpreting it. The bald statement lacks perspective to show its true size and proportion. A better understand- ing of this important function of the sweat glands can be had when it is known that the average amount of matter voided by the kidneys in twenty- four hours is approximately two and one-quarter pounds, or but slightly more than that thrown off by the sweat glands; while the solid fecal matter discharged by the average person in twenty-four hours is less than one-quarter pound and the vapor carried off by the lungs during the same period of time is approximately one pound. There is another function of the skin in which the sweat glands play the most important part, and that is in maintaining a uniform and normal temperature of the body under all the variations of temperature to which individuals are exposed, so that, whether in the tropics, near the equator, or in the frozen north searching for the pole, the temperature of any individual would remain con- stant. The reason why temperature remains con- stant is because when exposed to a temperature greater than 98° Fahrenheit, or when the body temperature is raised by exercise, the little sweat ducts open wide their mouths and pour forth a 21 Design of the Turkish Bath stream of moisture proportioned to the tempera- ture or the exertion. The cooHng effect of this moisture evaporating from the surface of the body keeps down the temperature, which otherwise would rise to a dangerous or fatal degree. Owing to this wise provision of nature man can live with- out the least discomfort when exposed to incredible extremes of dry heat. For instance, a man can sit with complete immunity in a hot-air bath raised to a temperature sufficiently high to bake bread or cook meat, and men have habitually sat without the slightest inconvenience in the hot room of a Turkish bath where the temperature was 240° Fahr- enheit, or 28° above the boiling point of water. Many instances could be cited of persons whose occupations exposed them to temperatures of from 250° to 280° Fahrenheit for long periods of time, and others who have endured temperatures of 360° Fahrenheit for shorter periods of time, but these may be dispensed with and the case of Cha- bert alone mentioned. Chabert, who was known to the public as the "Fire King," was in the habit of entering an oven, the temperature of which was from 400° to 600° Fahrenheit, and remaining there a considerable length of time. When exposed to a temperature of less than normal body heat, on the other hand, the sweat glands cease pouring forth their fluid, there is con- sequently no evaporation from the skin to reduce the temperature of the body, and the internal com- bustion is required to furnish only enough heat to replace that lost by radiation and convection. 22 Design of the Turkish Bath In addition to the functions of the skin already enumerated the skin possesses also a respiratory function, giving off a small amount of carbon diox- ide and taking in a small quantity of oxygen. In this respect the skin performs to a slight extent the function of the lungs. The skin can also ab- sorb slight amounts of water and other fluids, but these last-named functions are of no great im- portance from the bathing standpoint. Benefits of Bathing. — Bathing is beneficial to a person in a number of ways and affects, more or less, all of the glands, vessels, fibers and cells of which the skin is composed. These, in turn, react upon the larger and more important organs of the body, thereby toning up the entire system. As has already been mentioned, dead scales, or cells, are constantly being shed from the outer surface of the scarf-skin. Some of these scales, however, are not cast off at once, but cling to the younger cells beneath or hang in ragged fragments to the body until removed by friction or other mechanical means. These scales are dead organic matter and unless removed will putrefy on the person, besides, in many cases, interfering with or obstructing the oil ducts and sweat pores. The putrefactive bac- teria engaged in breaking down the skin scales might contain, among their number, some harm- ful kinds which would cause illness or death if in- troduced to the body through a puncture of the skin. It will be readily seen, therefore, that a bath of any kind, taken regularly, will remove this skin as fast as it is shed, thereby preventing the mouths of oil ducts and sweat glands becoming clogged by Design of the Turkish Bath cast-off materials, will soften and remove the dead or dying skin which clings to the younger cells be- neath, and will keep the skin physically, if not sur- gically, clean, thereby minimizing the danger from infection through a cut or bruise. The beneficial influence of water not only upon the skin but likewise upon the nervous and circu- lating systems, as well as the internal organs of the bather, is due principally, to the temperatures of the water in which the bath is taken. Baths must, therefore, be considered according to their temperatures and modes of application, and the effect of hot and cold baths as well as vapor and hot-air baths will next be explained. Cold Water Baths. — Generally speaking, it may be said that the effect of a cold bath is to close the pores of the skin, contract the capillaries, driving the blood from the surface of the body to the interior, leaving the skin white and bloodless. The functions of the skin being thereby tempor- arily arrested, the temperature of the blood rises from 2° to 4° Fahrenheit, which is equivalent to fever heat. Immediately after the bath, provided it has not been prolonged to an unreasonable extent, a reaction sets in, the sweat pores of the skin open, the capillaries expand and the overheated blood rushes back to the surface, bringing with it a healthy glow and a grateful feeling of warmth. So long as the healthy reaction can be induced, cold baths are beneficial to an individual, but if the skin remains white or turns blue the shock is too great and a warmer bath should be resorted to. 24 Design of the Turkish Bath Baths in water from 65° to 55° Fahrenheit are considered cold, and anything below 55° Fahrenheit is considered very cold. Very cold baths cannot be borne long without ill effect, and baths which lower the temperature of the skin to 9° Fahrenheit may be endured for a very short time, but any further reduction of the temperature is liable to prove fatal. Baths in water from 80° to 65° Fahr- enheit are considered cool baths. When immersed in a bath of cold water the temperature of which is above 50° Fahrenheit there is a diminution of the temperature of the skin and tissues near the surface of the body and the tem- perature of the blood rises about 4° Fahrenheit. At the same time there is a slight shock exper- ienced from the water, and if the cold is intense it induces a feeling of numbness in the skin, which becomes pale, due to contraction of the capillaries, which sends the blood to the internal organs. As would be expected the cold bath likewise affects the central nervous system and the heart and lungs, as may be seen by the tremor of the limbs, the gasping for air, and the general depres- sion which follows, due to the pulse beat becoming weaker. After the bath reaction sets in, bringing blood and warmth to the surface of the body. The colder the water and more powerful the depressing effect the quicker and more active will be the re- action, provided the individual is strong enough to withstand the shock. Tepid Baths. — Any bath taken at a tempera- ture between 80° and 92° Fahrenheit is a tepid bath. The effects of the tepid bath are not so 25 Design of the Turkish Bath numerous nor so far reaching as those of other temperatures, so that tepid baths can be borne for hours without ill results. The effects of the tepid bath are confined to the surface and do not reach the internal organs or nervous and circulating systems. There is no reaction whatever following the tepid bath, and the body and blood tempera- tures remain unchanged. On account of the absence of shock or stimulus of the internal organs tepid baths are best for people of weak constitu- tions or weak hearts. They are the least beneficial of all the baths, however, for those of strong, robust constitutions, as their beneficial influence is confined simply to cleansing the skin. Warm Baths. — Temperatures from 92° to 98° Fahrenheit include those of the warm bath. There is no difference in effect between a warm bath and a hot bath; the only difference is in the degree of action and reaction excited. In the warm bath the effect is not confined to the surface, but is propagated to the internal organs, which causes an increased flow of blood to the surface and an increased frequency of the pulse beat. It seems likewise to stimulate slightly the building up or renewal of new tissue. Hot Baths. — ^A bath at any temperature be- tween 98° and 104° Fahrenheit is a hot bath. When a person is immersed in a hot bath a transfer of heat takes place from the warmer to the cooler medium— that is, from the hot water to the bather —while at the same time the evaporation becomes checked, and the combined effort of the transfer 26 Design of the Turkish Bath of heat and checking of evaporation increases the body temperature of the bather. In the effort to keep down the body tempera- ture the capillaries expand and blood rushes to the surface of the body, the skin becomes congested and the accumulated body heat finally bursts forth, causing a profuse perspiration, while at the same time the pulse beat increases and the respiration becomes quickened. It will thus be seen that in a hot bath the nervous and circulating systems be- come affected, which in turn react upon the inter- nal organs. After the bath reaction sets in, the capillaries contract, all excess blood leaves the sur- face and the air at ordinary temperature feels cold to the skin. Comparison of Hot and Cold Baths. — In com- paring the effects of hot and cold baths upon normal individuals it might be stated as a general rule that the effects are opposite to each other; cold baths, on the one hand, tend to check perspi- ration, while hot baths favor it. It is believed, but not conclusively proven, that cold baths, by stimulating the internal organs, increase the re- action of the gastric and other fluids of the stomach and alimentary canal, while hot baths, on the other hand, tend rather to retard such activity. All baths, whether hot or cold, but particularly the latter, favor the secretion of urine. Hot baths cause dilation of the capillaries and a rush of blood to the skin. When the stimulus of heat is withdrawn the capillaries contract and all excess blood flows away from the skin. 27 Design of the Turkish Bath The coid bath, on the contrary, first contracts the capillaries and forces the blood to the interior, then when the reaction sets in the overheated blood flows back to the skin through the dilated capil- laries. A warm bath raises the temperature of the body by transferring heat to it and at the same time preventing evaporation and radiation of heat from the body. " The cold bath reduces the tem- perature of the surface of the body by withdrawing heat from it but raises the temperature of the blood. The hot bath draws the blood to the surface while the cold bath drives it to the interior. In either case there is increased oxidation, or waste of tissue, but with the warm bath there is less demand made upon the system, because oxidation depends chiefly upon increased heat, which in this case is supplied by the water. The reason that a hot bath seems refreshing to an exhausted person when he could not take a cold one may be that the heat supplied by the water helps the process of oxidation without any tax on the system. The hot bath can be borne longer than a cold bath and a tepid bath can be borne longer than either. Vapor Baths. — So far the effects of water baths only have been considered, no mention hav- ing been made of vapor or hot-air baths. It should be borne in mind, however, that the body bears changes of temperature of air, or even of vapor, much better than it does of water, because water being a better conductor of heat than either air or 28 Design of the Turkish Bath vapor brings more heat to the body or carries off a greater amount, as the case may be, than would air or vapor of equal temperature. The vapor bath, on account of the less specific heat, does not act as quickly as water on the body, but once the action does begin it causes a profuse perspiration and acts powerfully in cleansing the skin. Vapor baths can be borne hotter than water baths, but not for so long, for the vapor being of higher temperature than the bather, and being charged with moisture, not only prevents evapora- tion from the skin but likewise radiation from the body. In consequence of this the temperature of the vapor bath, while far less than that of the hot- air bath, heats the blood considerably more, besides impeding the respiration by depositing moisture in the bronchial tubes. A vapor bath can be borne for a much longer time when the vapor is not in- haled. Ordinarily, however, when the vapor is inhaled a temperature of more than 125° Fahrenheit cannot be borne with comfort. The effect of the vapor bath is much the same as that of the hot bath, with the one exception, perhaps, that it causes a more profuse perspiration, and in that one difference lies the greater value of the vapor bath. The vapor bath is the chief feature of the Russian bath, while the hot-air bath is the distinguishing feature of the Turkish bath. Hot-air Baths. — Hot-air baths possess all the advantages of the hot-water bath, and some other advantages that hot- water baths, as well as other forms of bathing, do not possess. It is those latter 29 Design of the Turkish Bath qualities which give the hot air, or Turkish, bath its hygienic value. One feature of the hot-air bath lies in the air virhich is inhaled. Unlike the air of a vapor bath it is not charged with vapor, so that moisture can- not be deposited in the bronchial tubes. Indeed, on account of the dryness of the air it increases instead of retards the evaporation from the lungs, while, being of a higher temperature than the body, there is no tax on the system heating the inspired air. The greater benefit of the hot air bath, no doubt, is due to the profuse perspiration it induces without raising the temperature of the body over several degrees Fahrenheit. This profuse perspir- ation is beneficial in several ways. By exciting the glands it keeps them in a healthy and active con- dition; the fluids they pour forth flush the millions of little sweat ducts so that they will be perfectly clean, and last but not least, the fluid poured forth in the form of perspiration carries in solution waste products of the cell activity so that the blood is puriiied by the process. It is in its function as a blood purifier, then, that the hot-air or Turkish bath differs most from ordinary baths, and it is in this function that the chief value lies, for any other form of bath will clean the person, stimulate the organs and excite the circulating systems. Routine of the Turkish Bath.— The Turkish bath, as now practised, consists first of exposure of the naked body to dry, hot air until a profuse per- spiration has been induced; next massage followed 30 Design of the Turkish Bath by a thorough scrubbing, with brush, soap and hot water; then a cooling shower to close the pores, which may or may not be followed by a cold plunge. The last stage consists of drying the body and resting. Of the various operations undergone by a bather the exposure in the hot room is the most important and the massage is of the least benefit. When properly performed massage is of real bene- fit to a person, but slapping of the body with open palms to make a loud noise as commonly practised in Turkish bath houses, is of little or no value. In its simplicity, then, the Turkish bath con- sists of exposure in the hot room, followed by knead- ing, rubbing, massage and a thorough washing, and this simple operation can be performed in the house as fully and with every bit as much benefit as in a public Turkish bath house, provided in addition to the ordinary bath room, there is a hot room or cabinet in which to induce perspiration. In the public bath room, however, numerous rooms and accessories are necessary in order to accommodate a number of patrons at one time and make the op- eration of bathing a luxury as well as a benefit. Routine of the Russian Bath. — The Russian bath, as now practised, differs but slightly from that of the Turkish bath, the chief difference being in the use of a vapor room instead of the hot-air room used in the Turkish bath. After exposure in the vapor room until a perspiration has been in- duced the bather is put through the operations of kneading, rubbing, massage, scrubbing, shower, plunge, drying and rest as in the former case. 31 Design of the Turkish Bath In most Turkish bath establishments a vapor room is provided in addition to the hot-air rooms, so that patrons may take either a Turkish or a Russian bath, or combine the two. Vapor and hot-air baths are so beneficial to a person that it is to be regretted that provision is not more frequently made for them in the home. The matter would be simple to effect, for a small room or cabinet could easily be provided and fitted with means for heating and ventilating, so that the inmates of the house could have both the bene- fit and the luxury of a Turkish bath. 32 LAYOUT OF A TURKISH BATH m m ARRANGEMENT AND RELATION OF THE VARIOUS PARTS ONVENTIONAL Floor Plan — The conventional layout of a simple little Turkish-bath establishment is shown in Fig. 11. In a large bathing establishment many features would be included which are not shown here. For in- stance, a barber shop, lounging room, pedicure booth, hydrotherapeutic department, vapor room, sun parlor, light baths and massage room might be included, or any one or more of them. Further- more, instead of combining the shower, plunge and shampoo, or laving rooms, all in one, as was done in the present instance, separate rooms or com- partments would or could be provided for each. The example here illustrated is given because of its simplicity and because it shows at the same time all the necessary elements of the bath. Dress- ing rooms containing a cot, bed or couch are nec- 33 Design of the Turkish Bath ■4 CL Thihd Hot Room CccoiiD Hot P-qom TemptrtCure ISCT " Ji -'Marble ^^ntfWe ->1arbie Sbempoo First Hot Roon ">. Temperature KOT / Fig. 11 Ojnventional Floor Plan of Bath Design of the Turkish Bath essary in which to dress and undress, as well as to rest in and possibly sleep after the bath. Besides the bed or couch each room should contain a chair and dressing table, with mirror and hooks on which to hang clothes. Each door should be provided with a Yale or other lock which cannot be easily picked, and the top of the rooms should be covered with a wire netting unless the walls extend to near the ceiling. At some convenient place in the estab- lishment toilet facilities for the bathers should be provided. The number of fixtures need not be large, one water closet and one urinal being suffi- cient for each fifteen persons that can be accom- modated at one time. From the dressing room the bather passes into that part of the establishment where the actul pro- cess of bathing is undergone. There are first the the hot rooms, next the shampoo compartments, then the needle shower and spray bath and, finally, the plunge, each of which will be considered separately THE HOT ROOMS Number of Hot Rooms. — In the bathing es- tablishment of the ancient Romans the bather passed in successive stages through three hot rooms, known respectively as the tepidarium, or warm room, calidarium, or hot room, and the loconium, or hottest room. By thus accustoming himself to the temperature of one room before passing to the next hotter one he was able to en- dure a much higher temperature and, consequently, derived much greater benefit from the bath than 35 Design of the Turkish Bath if he were exposed only to the heat of one room and the temperature in that room was as hot as he could endure. This ancient practice of the Romans has not been improved upon to this day, and the best re- sults will be obtained in Turkish bath design by arranging three hot rooms en suite, as shown in the example illustrated. The human system can accustom itself to ex- tremes of temperature and pressure if the changes leading up to the extremes are easy and gradual, not sudden and severe. Everybody is familiar with the danger of passing suddenly from ordinary at- mospheric pressure into compressed air of two or more atmospheres, or of emerging suddenly from the higher to the lower pressure, while at the same time the transition can be made without danger if the change is gradual, so that the body can be- come accustomed and adjust itself to one pressure before being exposed to another higher or lower one. In the same way, no doubt, the body is put under a severe strain by exposing it suddenly in a room heated to an extremely high temperature, while, on the other hand, if the high temperature be reached gradually, by passing from one hot room to a hotter one, until the maximum temperature is finally reached, the effect will be beneficial. The Indians seem to have understood this fact as well as the Romans; they accomplished the same result by entering their sweat house when it was but slightly above the normal temperature and remain- ing until it had reached the maximum. In spite of 36 Design of the Turkish Bath the benefit derived from a gradual exposure to the highest temperature the tendency of the times in Turkish bath design is to provide but one, or some- times two, hot rooms for the bathers. Two hot rooms are not so bad, but the cutting down of the number of hot rooms to one cannot be commended unless the gradual heating is achieved in some other way. For instance, if the room has a high ceiling the difference in temperature between the floor level and the ceiling would be great, and different temperatures could be attained by the bathers if balconies were placed at different elevations in the room. But there is a different way of looking at the matter. People differ greatly in the amount and degree of heat they can stand. Some have reached their limit at 140° Fahrenheit while others delight in the higher temperature of 250° Fahrenheit, and if the bath is to be considered as a pleasure and luxury the different temperatures desired by the various patrons should be maintained, and this, of course, is impossible when there is only one hot room. Again, Turkish baths are often used for reducing flesh and as remedial measures in case of sickness, and for such cases the higher-temperature rooms will be required even though most patrons desire only the low-temperature room. Temperature of Hot Rooms. — The tempera- tures to be maintained in the various hot rooms of a Turkish bath have been well established by usage. In the first hot room the temperature may vary from 130° to 150° Fahrenheit with an average of 140° Fahrenheit. In the second hot room the tem- 37 Design of the Turkish Bath perature may vary between 170° and 190° Fahren- heit with a fair average of 180° Fahrenheit; while in the hottest room the temperature will average 250° Fahrenheit and may vary from 240° to 260° Fahrenheit. In bath establishments where there is but one hot room it would seem that a temperature equal to the second hot room would be about right, while in establishments having two hot rooms temperatures similar to the first and second hot rooms would prove satisfactory. Of course, in fitting up such a room sufficient radiation would be supplied so that hotter temperatures could be maintained if necessary, and the radiation would be so controlled that the various units could be cut off or turned on at will to regulate the amount of heat by adding to or taking from the heating surface. In speaking of the temperature of the hot rooms it is assumed that the thermometer readings are taken at an elevation of 6 feet 6 inches above the floor line. Not only the temperature of the hot rooms, but, likewise, the temperature of the other parts of the bathing establishment must be considered when designing a Turkish bath. In the shampoo com- partments, likewise the shower room and plunge bath, the temperature of the air should be kept at about 110° Fahrenheit. Owing to the lack of cloth- ing and the rapid evaporation from the body there would be a feeling of chilliness in these rooms if they were maintained at lower temperatures. In the dressing rooms, on the other hand, the bathers are provided with sufficient clothing to keep them 38 Design of the Turkish Bath warm, and in order that they get the greatest, amount of good from the bath, the temperature there should be kept at about 65° Fahrenheit and there should be a constant change of pure, fresh air. Size and Proportion of Hot Rooms. — The size of the hot rooms for a Turkish bath will generally depend upon the size of the building and the mag- nitude of the establishment. Seldom will they be made so small that the amount of space to be allotted to each patron will have to be considered, for Turkish baths are built more as a luxury than as a convenience or necessity, consequently ample room will be provided for all compartments. Nevertheless, there will be cases where that prob- lem arises, and it will be well to consider just what amount of space will be necessary for each person. In the first place, a certain amount of space will have to be reserved for passages, and, after allowing for such aisles, a fair allowance for each person would be a floor space occupying a rectangle 6 feet long by 4 feet wide. Such an al- lowance would give approximately 2 feet between chairs and would allow about 18 inches over the average length of reclining chair, such as are gen- erally used in the hot rooms of bathing establish- ments. Next in importance to the amount of space to be allotted to each patron is the proportion of the various hot rooms to one another. Experience shows that of the numerous patrons of a bathing establishment only a certain percentage of them will pass into the hotter rooms. All of the patrons, however, enter the first hot room, so that this compartment should occupy from 39 Design of the Turkish Bath one-half to three-fifths of the total space allotted to the hot rooms. Of the remaining space a little over one-half, or about three-fifths, should be al- lotted to the second hot room, and the balance of the space to the hottest compartment. The hottest compartment, it would seem unnecessary to point out, is made the smallest because fewer people will reach this room than will expose themselves to the heat of the other rooms. Drainage for Hot Rooms.— As a rule it is bet- ter not to place a floor drain in any of the three hot rooms of a Turkish bath. Owing to the dry heat in these compartments and the absence of water, except when cleaning and scrubbing the rooms, the traps would be liable to become emptied by evaporation, thus allowing drain air to enter and vitiate the air within the rooms, which should be kept as pure and fresh as possible. Instead of placing a floor drain within the sweat rooms the better plan is to locate one just outside, as shown in the illustration, then give the entire floor of the three hot rooms a slight slope towards this drain. Water Supply for Hot Rooms. — The only sup- ply of water ever piped to the hot rooms of a Turkish bath is a line of filtered water to a drink- ing fountain, located in the first hot room. On account of the amount of moisture lost during the routine of the bath, particularly when in the hot rooms, it is advisable to drink as much water as possible, so that transpired perspiration will not take place at the cost of the living cells, which might be robbed of their much-needed fluids. On account of the faucet and some of the piping being 40 Design of the Turkish Bath located within the hot room the water from such a drinking fountain will be found slightly warm, particularly if not allowed to run for a short time before drinking. Warm water, however, promotes perspiration, while cold water tends to check it, so that those who like warm water or, at worst, can drink it, will be benefited by so doing. Ventilation of the Hot Rooms.— The ventilation of the hot rooms is of the utmost importance to the bathers. The methods of ventilation are taken up in a subsequent chapter and all that is necessary here is to point out, generally, what is good and what is bad in this respect. It will be remembered that the object of the bath is to gradually raise the temperature of the body until all the sweat ducts open wide and pour forth the stored up impurities from the blood. The greater comfort with which this can be accomplished, the better it will be for the bather in every way. Lack of air is a discom- fort which would defeat much of the object of the bath, and a supply of unsuitable air cannot be considered in the nature of ventilation. It is a well-known fact that heating air to a high tempera- ture deprives it of some of its most refreshing qualities. Note, for instance, the invigorating cool air of early Spring, Autumn, or the sea shore, as compared with the enervating hot air of July and August. There is the same quantity of air avail- able at all times, but the quality differs greatly under different conditions. The benefits of the hot bath would, perhaps, be greatly enhanced if the bather could draw in his lungs copious drafts of fresh cool air from the outside while sitting exposed 41 Design of the Turkish Bath to the radiant heat of the hot room. Such a con- dition would _ simulate the condition of an early Spring morning when seated in a protected nook of a garden exposed to the sun enjoying the invigor- ating glow from a sun bath— radiant heat— while breathing in pure cool air of Spring. Perhaps sometime we will be able to provide similar conditions in the hot rooms of a Turkish bath. A hood or vent duct in which the heads are immersed in fresh cool air, while the bodies are exposed to the radiant heat of the steam coils, or a supply of fresh cool air supplied in some other way to the nostrils, would seem to add greatly to the comfort of the bathers in the hot rooms, and if to the comfort, then to the benefit. Some experi- ments would have to be made before such a method would be practicable, but the benefits that would be derived warrant keeping the possibilities in mind. is'llif 42 INSULATING TURKISH BATH ROOMS NSULATION of Hot Rooms— In order that the hot rooms of a Turkish bath may be economically maintained at their proper temperatures it is absolutely necessary to thoroughly insulate the vails and ceilings so that heat will not be constantly radiated to the surrounding atmos- phere or penetrate to other parts of the building. Usually a Turkish bath establishment is but one of many industries grouped together under one roof, and if heat from the bath rooms were permitted to escape to other parts of the building it would not only entail an extra and unnecessary expense on the proprietor of the bathing establishment but would, furthermore, prove extremely annoying and, perhaps, expensive to the rest of the tenants. When selecting insulating materials it should be borne in mind that dead-air space is among the very best of heat insulators or retardants. If the air has room for circulation, however, the insulat- ing effect will be destroyed. That is why materi- als like mineral wool, which of themselves are 4;j Design of the Turkis h Bath heat conductors, will, when packed in an air space, prove effective for insulating walls or ceilings. The mineral wool when packed into the space forms a multitude of cells which contain air, but, owing to the cellular formation of the mass, the air is dead and cannot circulate. Hollow tiles like- wise prove good insulators. In the case of hollow tiles, unlike spaces filled with mineral wool, there is local circulation, but of such a sluggish and re- stricted nature that the air is practically dead. Besides mineral wool any fibrous or granular material which will not pack too tight will prove satisfactory for filling purposes. Among the ma- terials successfully used may be mentioned coal cinders, hair felt, granulated cork, charcoal and sawdust. Materials like granulated cork and char- coal rank highest for the purpose for the reason that in themselves they are not good conductors of heat, while materials like mineral wool and cinders are, to a limited extent. On the other hand^- cin- ders and mineral wool are fireproof, while some of the better insulating materials are not, and in de- signing an establishment all these facts must be taken into consideration. Concrete is a much better insulator of heat than is generally supposed. This fact was recently demonstrated very forcibly at Boston, Mass., where, for five hours, a floor 5 inches thick was subjected to an intensely hot fire, varying in tem- perature from 1, 700° to 2, 000° Fahrenheit. During the entire period of the experiment there was not a moment when the hand could not have been held 44 Design of the Turkish Bath Brick Wall. Z Coats ofPitcJr. fHoiiow tile. 4" Mineral V/ool. S'Hollow Tile. Cemei Plaster or Tile. on the top surface of the concrete without fear of being burned. Insulating Fireproof Walls. — Certain methods for insulating walls, floors and ceilings against the loss of heat have been adopted in practice for the reasons that they are simple, effective and applic- able to all conditions. The method commonly adopted for insulating walls of stone, brick or con- crete is shown in Fig. 12. The inside of the wall to be insulated is first given two good coats of pitch, tar, asphaltum or some like elas- tic and moisture- proof ma- terial to prevent damp- .:'->.-■ ness striking into the wall and to keep down as low as possible the loss of heat due to winds. Inside of the coat of pitch is laid a wall of hollow tile to provide a partial dead-air space. Four inches inside of the first wall of hollow tile is laid another, 3 inches thick, and between the two courses of tile the space is filled with min- eral wool. The inner surface of the 3-inch hollow tiles can then be plastered with cement mortar for a finish, or, if a better finish is desired, the sur- face can be covered with marble or tiles. Instead 45 Fig. 12. Fireproof Insulation for Masonry Walls. Design of the Turkish Bath of the inner course of hollow tiles, glazed bricks may be laid, and the space between the bricks and the first course of hollow tile filled in with mineral wool or some other filling material, as previously explained. Further, wherever necessary or desir- able to do so, one course of hollow brick can be omitted and the wall otherwise insulated just as described. Wooden Insulation for Brickv\rork. — ^Wood is not so desirable a material as hollow tile for insu- lating hot rooms, but, whenever for any reason wood is the chosen material, the rooms can be in- sulated as shown in £>rickwall y^a Coofs ofPikh. /- .^r.P '3/rws . /6 "c-C '^^ Double /"sheathing I With S hyers afpape> / . 2'W5tudi /&'c-cnVed bpJween v/ith tvmera/Woof. --Double I" sheathing with [J 2 layers of paper bcfw. ^j'kZ^Sfrit^s l6"C''c Ceffienl p/asf-er orqiajea fi/e on eK p. metal lath. "'befiY. .Fig. 13. The brickwork, X; stone wall or cement con- crete wall, as the case may be, is first treated to two coats of pitch, asphalt or tar. After the pitch has been applied regular 1 inch by 2 inch furring strips are nailed to the wall at distances of 16 inches from center to center. A horizontal course of i-inch sheathing is nailed to the furring strips, a double layer of good building paper tacked to the sheathing, and another course of sheathing, laid diagonally, is placed on top of the building paper and nailed to the furring strips through the first course of sheath- 46 Fig. 13 "Wooden Insulation for Masonry Walls Design of the Turkish Bath ing. Regular stock, 2x4, studding are then set up on 16-inch centers and nailed to the sheathing and the face of the studding covered with a double course of sheathing with building paper between, as in the first instance. The space between the studding is then filled with mineral wool, or some other heat insulator. Furring strips are next nailed to the surface of the last course of sheathing, and expand- ed metal lathing fastened to the strips, so that the surface is ready for a three-coat layer of cement plaster or for a finish of glazed tile. If the masonry walls are pretty thick, or are other- wise well insulated, the inner layer of wooden sheathing may be omit- ted, and the expanded metal lath nailed to the face of the 2x4 studding. The sheathing used should be well matched stock free from knot holes or shakes, and should be of some wood which will not twist and warp under the action of heat. Insulation of Walls in Frame Buildings. — The method of insulating walls in frame buildings may be seen in Fig. 14. 47 ^-/"Ndvelfv 8idin^4"3leep&rs. /2"C//7cfer3. S — ■6'3l-or7e. Fig. 17 Insulation for Ground Floor would not be desirable on account of the odor from hot asphalt and the fact that the flooring would become soft from the heat. It might be well to add that in case the floor is to be of tile or cement that the woodwork would be left out entirely and the concrete and cinders depended upon for the necessary insulation. Insulation of Windows.--When windows or skylights open into any of the rooms of a Turkish 52 Design of the Turkish Bath bath which are heated above ordinary room tem- peratures great care must be exercised to prevent the loss of heat through the glass. The ordinary w^indow sash with single-thick or double-thick glass would never do for this purpose. Sash and frames of special construction must be provided. The method of insulating windows is shown in Fig. 18. Ventilation to the rooms is provided for, indepen- dent of the windows, so the sash need not be made to raise and lower like those of ordinary windows. That being true, the sash are made to fit the entire window opening and are set in place so that the Fig. 18 Insulation fop Windows leakage of air around the stops is reduced to the minimum. It will be noticed that instead of one sash there are three sash in this frame, and each sash is glazed with two panes of glass, so that all told there are six thicknesses of glass and five air spaces for the heat to pass through. If the window exposure of a Turkish bath is at the south or in a sheltered part of the east, or if the country is a comparatively mild one, two sash, with four 53 Design of the Turkish Bath thicknesses of glass and three air spaces, would probably be sufficient. In glazing windows for Turkish bath rooms it is customary not to use putty and window tacks, as is commonly done in glazing, but to use adjustable stops with felt or chamois linings, so the glass will have room and freedom to expand and contract without opening crevices for the leakage of air. To avoid breakage the panes of glass should not be very large, and rolled glass should be used in preference to ordinary window glass, as rolled glass is tougher and less liable to break. =L ■ Double sheafhincf wifh 2 /ayers paper°b efvyeer7 Mineraf ivaoL Fie. 19 Insulation for Doors Insulation of Doors. — Wherever a passageway leads from one of the hot rooms of a Turkish bath to a cool room — that is, one maintained at the ordi- nary temperature of living rooms — double doors with a lobby between will be found advisable to prevent the transmission of heat from the hot to the cooler rooms. In other parts of the bathing establishment the opening from one room to another of higher temperature may be closed with 54 Design of the Turkish B a t h a single door which is well insulated against the passage of heat. An insulated door is shown in Fig. 19. This door has three double layers of boards forming a double air space, and three layers of good building paper are placed between each double thickness of board. In the illustration an ordinary door is shown without attempt at ornamentation. When, how- ever, the bathing establishment is to be elaborately finished the doors may be surfaced with the wood used for the interior trim and may be paneled to correspond with any detail. If for any reason it is found necessary or desirable to place glass in the doors it can be done by observing the precautions necessary for window insulation. It might be well to have the upper part of the door leading to the first hot room made of glass, so that bath attendants from the outside can see if everything is all right with the bathers without having to open the door, as would be necessary if the glass were omitted. Patented Heat Insulating Materials.— In the foregoing pages the methods of insulating against the loss or leakage of heat, and the doors and windows used for this purpose, have of necessity been stock goods obtainable in any locality, and easily assembled from goods at hand. The princi- ples outlined are the principles which must be observed in any kind of insulating methods or materials, whether ordinary methods or patented ones. However, there are certain patented heat insulating materials now in common use which 55 Design of the Turkish Bath have been made with the one object of preventing heat loss. Likewise there are patented doors and windows, and in many cases insulating materials, doors and windows which are manufactured for the purpose but not patented, and often those will prove not only cheaper but better than the ordinary methods outlined, the difference depending on the method of construction and the location of the building. The practice of heat insulation has been brought to its highest state of perfection in the ice-making and refrigeration industry, where heat loss must be reduced to the minimum if an eco- nomical operating plant is expected. When taking up the design of the Turkish bath, therefor, the manufactured materials for insulating buildings, walls, floors, ceilings, windows and doors can be found advertised and described in publications devoted to ice-making and cold storage, to which it will be advisable to refer. 56 CHAPTER HEATING TURKISH BATH ROOMS ETHODS of Heating the Sweat Rooms.— Up to the present time the Turkish bath has been referred to in this work as a hot-air bath. As a matter of fact, however, the Turkish bath is a heat bath pure and simple, the tempera- ture alone being depended upon to bring about the beneficial results sought in the sweat rooms, re- gardless of the manner in which the heat is applied. For convenience the term hot-air bath will con- tinue to be used, although it must be understood that this term is meant simply to distinguish hot baths in air from hot baths in water. So far as heating of the sweat rooms of a Turkish bath is concerned, in the best method used, the direct radiation method, the air is heated only incidentally, not with a view of transmitting the heat from the air to the bather. In the hot-air method, on the other hand, air is used as a vehicle of heat, and those baths are not only heat baths, but are rightly named hot-air baths. It should be born in mind, however, that the object of a Turkish 57 Design of the Turkish Bath bath is to apply heat to the body, not necessarily air or hot air. There are two systems now in use for heating Turkish sweat rooms. The first, the hot-air meth- od, is similar to heating an ordinary dwelling by means of a furnace, only the rooms must be heated to a much higher temperature. In the second method direct radiation is employed, which may be likened to the heat from a muffle kiln where earth- enware is being fired. The second method, heating by direct radiation, was the means employed by the ancient Romans for heating their bathing es- tablishments, and ia the method, greatly modified but not improved, which is followed at the present time in the United States. In Great Britain, on the other hand, the hot-air method is the one com- monly followed. In Turkish bath rooms, heated by radiation, the heat is transmitted direct to the bather from the walls, floor and ceiling of the rooms which are heated to a high temperature from outside, or from a system of steam pipes or radiators placed within the rooms. In this respect the Turkish hot rooms are very similiar to the bake oven of an ordinary kitchen range, which is heated from outside and in which only an occasional change of air takes place. In the sweat rooms, however, provision is made for ventilation, which will be explained later. The old Roman baths were constructed with the furnace underneath the sweat rooms and so de- signed that the flames and hot gases, after playing upon the ceiling, which formed the floor of the hot rooms, passed up through hollow tiles or flues 58 Design of the Tu r k i s h Bath formed in the walls of the sweat rooms, so that radiant heat flowed to the bather from the walls, the floor and the ceiling of the rooms. In the British hot-air method of heating, a steady stream of air of the temperature desired is passed through the hot rooms, the volume and velocity being sufficient to maintain the tempera- tures of the several rooms without causing draught. Each method of heating has its advantages and disadvantages, but it seems to be pretty gen- erally conceded that radiant heat is the most beneficial form of heat and for bathing purposes better than hot air. It would seem at first thought that heat is a condition or vibration which is the same no matter from what source it is derived or in what form the energy is available. Such is not the case, however, as may be seen by a comparison of the two methods of heating under consideration. Take, for example, first, the interior of a hot room heated by radiation from the walls, floor and ceiling, and bear in mind that heat always flows from the hotter to the cooler substance and that radiant heat passes through air without heating it. In that case, the walls, floor and ceiling being ap- proximately of the same temperature, there would be but little transfer of heat from one point to another within the room; on the contrary, the heat would be, in a nature, "bound," or at rest. Let a bather then step into the room which, assume, is heated to a temperature of 200° Fahrenheit. The moment he does so the radiant heat from all parts of the enclosure bombards him with a multitude of heat waves which strike him from every quarter 59 Design of the Turkish Bath and play upon him with an intensity equal to the difference between his own body temperature and the temperature of the walls. In addition to the heat waves, it is believed, but not fully demon- strated, that electro-thermal currents also pass from the waves to the bather through the dry air of the room. In short, all of the excess heat above the body temperature contained in the walls, floor and ceiling of the rooms focus their rays upon the bather and concentrate their combined force upon his person, giving him its undivided attention. In the second case, when the rooms are heated with hot air, all heat contained in the walls, floor, ceiling and air of the rooms must be supplied from the current of air pouring in at the register. That being the case, as soon as air enters the room the heat imparted to it by the furnace is divided, part flowing to the bather, part supplying the heat lost from walls, floor and ceiling, and part passing out through the ventilator register. In- stead of the heat being concentrated on the bather it is scattered and much of its beneficial effect is lost. Besides the concentration of heat waves in the one case and dissipation of them in the other, there is a real benefit derived from radiant heat which cannot be secured from heat by convection. What the quality of radiant heat is, which makes it so superior to other forms of heating for bathing pur- poses, it is impossible to say, further than that experience demonstrated the fact without being able to point out the cause. Further, there is a vast difference in the quality of heat radiated from 60 Design of the Turkish Bath different substances. There is nothing really bet- ter as a radiating surface than ordinary firebricks. From this material a soft heat is given off, differ- ing in quality from that obtained from iron or any other metallic surface. It would naturally follow from the foregoing statement that the radiant heat given off by steam pipes is vastly inferior to that from radiating surfaces of firebricks, yet, in spite of this fact, there is little likelihood of firebricks coming into extensive use for this purpose, al- though they may be used in the more expensive establishments. The simplicity, economy and con- venience with which steam can be used as a heating medium, and the variety of forms of heat- ing surfaces which can be adapted to any and all possible conditions, makes this the accepted method of heating the sweat rooms of Turkish bath estab- lishments. There is another reason why radiant heat is better than hot air for heating the sweat rooms. In the bath the object is to keep the head cool and the feet warm. To accomplish this bathers some- times place a cold, wet towel on the head and sit with their feet in a vessel of hot water. When the bath is heated by direct radiation the heating surface can be so disposed that most of the heat will be at the bottom of the rooms. With such a disposition the feet will be equally as warm, if not warmer, than the head. Where the pipes are concealed back of firebrick walls and under firebrick floors most of the heat can be radi- ated from the floor and the balance from the side walls. ~ When the bath is heated by hot air, on the 61 Design of the Turkish Bath other hand, whether from a furnace or by means of indirect steam, the coldest air in the rooms will be found near the floor, so that the head might be hot while the feet are cold. Usually there is a dif- ference of several degrees between the tempera- ture at the floor line and at the head level, and this difference in temperature is in the reverse order from what it should be. Steam will be found so necessary around a bathing establishment for operating the machinery to run ventilating fans, laundry machinery and for drying clothes that there is but little likelihood of the hot-air method, with all its disadvantages, be- ing used in this country, consequently the hot-air method of heating sweat rooms will not be con- sidered. Heating Sweat Rooms with Steam Coils. — The simplest method of heating the sweat rooms of a Turkish bath with steam, and the one commonly followed in practice, is to cover the walls and ceil- ings of the rooms with a grid of steam pipes made up in the form of coils. In arranging the heating surface in the several rooms it is well to bear in mind the fact that the greatest amount of heat is required near the floor line, so that if enough radi- ation will not be required to cover both the walls and the ceilings the walls should be covered first and whatever excess radiation then remains can be placed under the ceiling. For convenience in regulating the temperature of the rooms the pipe coils should be put up in sections, so that any or all of them can be cut out of service at any time, 62 Design of the Turkish Bath either with a view of regulating the temperature of the room or for repairs. In proportioning the heating surface sufficient radiation will have to be supplied to heat the various compartments to their regular temperatures in the severest winter weather. Such an amount of radi- ation, of course, would be too much in the summer time, or at intermediate periods between the two extremes, so the best way to arrange the coils is to provide a certain amount of radiation for the sum- mer months, which can be kept in service at all times, winter and summer, then have additional sections in reserve which can be put in service as the weather grows cold until, in midwinter, all the sections are in use. By such an arrangement the heating system is made sufficiently elastic to take care of the sweat rooms in any and all kinds of weather without constantly fussing with the valves and without the annoyance experienced when steam has been shut off all or a great part of the system because the rooms are too hot, only to have them go to the other extreme and fall below their regular temperature. Types of Coils for the Sweat Rooms. — What- ever type of coil is used for heating the sweat rooms, provision for expansion and drainage must be made when building them and putting them in place. If the pipes are screwed together rigid, without room for expansion, they will bend and buckle when the steam is turned on and the tre- mendous stress under which the coil is placed might burst some of the fittings. Such an accident would not be a small matter by any means, for high- 63 Design of the Turkish Bath pressure steam is required for heating the rooms, and if high-pressure steam were to escape from a burst steam coil while bathers were in the room some of them would doubtless be badly scalded be- fore they could make their escape. In order to prevent rattling and snapping sounds within the coils when the steam is turned on, all the pipes should be given a pitch of from 5 inch to 1 inch in 10 feet, sloping continuously from where the steam enters toward the outlet of the coil. If the coils are not properly drained there will not only be a disagreeable rattling and snap- M 9l Fig. 20 Continuous Flat Steam Coil ping sound within, but the coils will not heat as they should and the rooms cannot be maintained at their proper temperature. A continuous flat coil extensively used as a radiating surface is shown in Fig. 20. In this type of coil, which is also known as a return bend coil, the circulation is direct and positive, steam enter- ing the top pipe and traveling continuously, alternately right and left with the water of con- densation, toward the outlet at the bottom. When properly proportioned this is a very efficient form 64 Design of the Turkish Bath of radiating surface, but it must be properly pro- portioned or it will prove more or less a failure. Once the steam enters this type of coil it must travel continuously throughout its entire length, and if the coil be extremely long, contains a great number of bends and be exposed in a fairly cold place, the steam will be condensed in the upper loops of the coil, thereby parting with most of its available heat, after which the water of conden- sation will have to flow through the balance of the coil, where it will part with but little of the heat it contains. It will thus be seen that the lower part of this type of coil under such conditions would be of but little service. Further, owing to the dis- tances the water of condensation would have to travel and the friction of the pipe and bends, the lower part of the coil, particularly if made of 1-inch pipe, would be likely to become overtaxed, or air may be accumulated in the return bends to impede or wholly stop circulation. This form of coil can safely be used for small spaces in the sweat room, but if fairly large coils are to be used it is well to see that they are made of large size pipes, IJ or 2 inches in diameter, and that the pipes all have a good pitch from where steam enters to the return con- nection. Return manifold coils are of more direct cir- culation than are continuous coils, consequently they are the more efficient type. A return mani- fold coil is shown in Fig. 21. Steam enters the top end of the upper manifold header and flows almost simultaneously into the several steam loops of the coil. At the end of the 65 Design of the Turkish B a t h steam loops the condensation and vapor flow to the return portion of the coil and pass through the several pipes to the return header, thence through the return pipe to the boiler. It will be noticed that with this type of coil the steam does not have to pass consecutively through all the pipes, but, on the contrary, each steam pipe has a cor- responding return loop so that the steam has to pass through only one loop, or twice the length of the coil, before reaching the return header. 'm/AWM'M//,;m Fig. 21 Return Manifold Steam Coil Such an arrangement of the pipes insures per- fect and immediate drainage of the coil, so that good live steam can fill the pipes at all times, thereby insuring the maximum efficiency. Pro- vision is made for expansion and contraction by having separate headers for the steam and return pipes, which allows considerable spring to the horizontal steam and return pipes of the coil. 66 Design of the Turkish Bath Angle wall coils are very convenient in the sweat room for the reason that they can be ex- tended in one piece around two, or even three, sides of the compartment. An angle wall coil is shown in Fig. 22. This illustration shows a coil intended for only two sides of a room, but by replacing one of the manifolds Fig. 22 Angle Wall Steam Coil with elbows and providing pieces of pipe from those elbows to the mani- fold tiie coil could be extended around a break in the wall or around a third side of the room. In angle wall coils the flow of steam is short and direct, simply from the inlet manifold through one full length of the parallel pipes to the return manifold. Expansion and contraction of the coil are taken care of by the elbows and pipes, which will give or 67 Design of the Turkish Bath spring a sufficient amount to compensate for any contraction or expansion of the metal. Instead of heating the sweat rooms of Turkish baths with pipe coils they may be heated with cast- iron radiators or any other form of steam-heating surface. If cast-iron radiators are to be used wall radiators similar to those shown in Fig. 23 will be found both convenient and satisfactory, particu- larly for all the upper rows, which do not rest on the floor but must be secured to the walls. Wall radiators possess no advantages whatso- ever over pipe coils, so far as heating efficiency is concerned, and in point of cost will prove more ex- pensive. On the other hand, a wall surface //////////Ayz/liiyMi/Mi:,:. FJS. 23 Wall Radiators covered with well-designed wall radiators presents a more pleasing appearance to the eyes than a grid of steam pipes covering a like surface. Heating with Concealed Radiators. — ^Whileit is customary to place the radiating surface for the sweat rooms of a Turkish bath in the compartment it is to heat, such a practice is not necessary and is followed more for economic than for other reasons. It cannot be said that wall coils or radiators add anything to the ornamental features of a room, so whenever it is desired, visible radiation may be omitted and concealed radiation installed in its stead. To do so, curtain walls, preferably of fire- 68 Design of the Turkish Bath bricks, which may be enameled on the inside surface, are built in front of the steam coils — or, in other words, the coils are set up in a space provided for them back of the firebrick curtain walls and under the floor. Sufficient heating surface is sup- plied to raise the temperature of the firebricks, which then radiate heat to the bathers. The surface back of the steam coils must be well insulated and reflectors may be used to throw the heat against the firebrick walls. Heating a Sweat Room in the Home.— If it were realized how easy it is to fit up a sweat room in the home, no home would be without the benefit of a Turkish bath. As has been pointed out before, no agency but heat is needed for the purpose, water being obtainable in almost every home of to-day. The shower bath is fast becoming an indispensable part of every bath room, and a combined shower bath and sweat room can easily be constructed. Partition off from the main bath room a small square or round enclosure, about three or four feet square, or of the same diameter. Line the walls with tile, marble or porcelain, and set a porcelain or porcelain lined receptor in the floor. Then fit up this small enclosure with sufficient heating surface so that when the steam or hot water is turned on the compartment can be raised in temperature say to 140" Fahrenheit. With a chair placed inside of this enclosure it makes a perfectly satisfactory and efficient sweat room, and the cold shower can be used instead of the plunge for closing the pores after heat enough has been applied. Wall coils around the sides of the enclosure, either concealed 69 Design of the Turkish Bath or exposed will furnish the heat necessary for the operation. In case the building is heated with hot air, instead of wall coils an extra large hot-air register with an extra large hot air duct direct from the furnace to the hot room will supply the necessary heat, which can be shut off when the bath is not in use. In old buildings where no provision has been made for a shower bath or a sweat room, a bath cabinet can be had which will serve the purpose very well. These bath cabinets are more in the nature of Russian baths than Turkish baths, as vapor is used as the heating medium. A valuable feature of these vapor cabinets is the fact that the head is outside the cabinet so that pure cool air is being inhaled, while heat is being applied to the body inside of the cabinet enclosure. 70 PROPORTIONING THE HEATING SYSTEM m m PRINCIPLES OF HEATING PROPERTIES OF HEAT RANSFER of Heat.— When two bodies of different temperatures are brought near to each other a transfer of heat takes place from the hotter to the colder body. This tendency toward maintaining an equilibrium of temperature is universal, and the transfer of heat may take place in any of three different ways — by conduc- tion, by convection or by radiation. Conduction of Heat. — Conduction of heat is the progressive movement of heat through a sub- stance without perceptible movement of the mole- cules. If one end of a poker be held in a fire the other end will become heated by conduction. The hot rooms of a Turkish bath lose their heat by conduction, by convection and by radiation. ' If the walls are not properly insulated a great quantity 71 Design of the Turkish Bath of heat will flow by conduction from the inner to the outer surface of the wall, where it may radiate to surrounding objects of lower temperatures or be cooled by convection, currents of air carrying: off the excess heat. Convection of Heat. — Convection of heat is the transfer of heat by movement or circulation of the substance to be heated or cooled, which by constantly changing its position brings all parts consecutively into contact with the heating or cool- ing medium. Water in a vessel placed on a stove is heated by convection, or circulation, the constant movement of the water bringing all particles to the bottom, where the heat is applied. Water in a range boiler is likewise heated by convection, the constant circulation of water to and from the waterback bringing all particles there to be heated, and this heated water gives off part of its heat to that in the boiler when they are brought into con- tact. In like manner the sweat rooms of Turkish baths warmed by hot air are heated by convection, whether the air from the rooms is circulated through the heater— which, of course, should not be done— or whether fresh air from the outside is forced into the rooms. The air becomes heated by contact with the heating surface in the furnace, and when poured into the sweat rooms parts with some of its heat to the walls, floor and ceiling of the enclosure. In like manner heat is lost from the sweat rooms of a Turkish bath by currents of cold air blowing against the walls, thereby carrying off the 72 Design of the Turkish Bath heat that has been transferred by conduction from the interior of the rooms. Radiation of Heat. — Radiation of heat is the transfer of heat through space from a warm body to one of lower temperature, without the use of intermediate agencies. For example, the earth is warmed by radiation from the sun. Radiant heat does not heat the air through which it passes; it travels direct and in straight lines until intercepted and reflected or absorbed by some other body. The cooler body will reflect or absorb all the heat rays it intercepts, and the sum of the absorption and reflection equals the total of the intercepted rays. When a bather stands naked in the sweat room of a Turkish bath radiant heat from the steam coils passes direct to his body without losing any of its intensity to the air through which it passes. Heat and Temperature. — In all natural pheno- mena there are two distinct conditions or properties which can generally be measured or gauged. For instance, there are both pressure and quantity of water, the pressure being measured in pounds per square inch and the quantity in gallons or cubic feet. Electricity is measured in volts and amperes,, Sound, likewise, has its intensity and volume, and the condition known as heat has two properties — heat and temperature. The temperature of a sub- stance may be considered the intensity of the energy, and the heat itself as the quantity. The intensity of heat may be measured by means of a thermometer, ordinary temperatures being indi- cated by the common household thermometer. The quantity of heat, on the contrary, cannot be meas- 73 Design of the Turkish Bath ured by instruments or by any of the known standards for weight or quantity, such as pounds or gallons, so a new standard had to be devised which would measure the quantity of heat by the effect it would produce. The standard for the measurement of heat is known as a British thermal unit. British Thermal Unit. — A British thermal unit, usually abbreviated B. T. U., is the quantity of heat required to raise the temperature of one pound of water from 62° to 63° Fahrenheit. In practice it is taken as the quantity of heat required to raise one pound of water 1° Fahrenheit. British thermal units are likewise commonly called heat units. Heat Emitted by Radiators and Coils. — The quantity of heat transmitted from coils or radiators to air is practically independent of the materials used, so long as they are under similar conditions of internal and external temperature. That is, one square foot of cast-iron surface will give off practi- cally the same quantity of heat as one square foot of wrought-pipe surface, if both the cast-iron radiator and the pipe coil are supplied with heat from steam of the same pressure and are exposed in rooms having equal temperatures. Further, under the same conditions the quantity of heat given off from a hot water radiator will be equal to that given off from an equal size steam radiator. Again, the rate at which heat is emitted from radiating surfaces is not affected by the size of the tubes, or sections, of the radiators, or the diame- 74 Design of the Turkish Bath ters of pipe used in making the coils, so long as the pipes, loops or sections are large enough to afford good circulation. Painting or bronzing radiating surfaces has a marked effect on the quantity of heat emitted, the effect depending on the character of the finished surface. As a rule, it may be stated that painting radiating surfaces with a dull finish or a rough bronzing increases the total quantity of heat given off but decreases the amount of radiant heat emitted. This is important in Turkish bath work, for the reason that in the sweat rooms it is radiant heat, not connected heat that is wanted. However, there seems no practical way of overcoming the difficulty, for it is necessary to coat the pipes and radiating surfaces with something to keep them from rusting, even though it were not desirable for ornamental purposes. Paints or coatings of any kind having a glossy surface, particularly a glossy white finish, should be avoided, as it reduces not only the radiant heat but the total quantity of heat emitted. Radiators or coils having but one row of tubes are more effective in still air than those having two or more rows. This fact should be borne in mind when designing the heating sytem for the Turkish bath, and only one-tube coils used unless enough heating surface cannot be installed without doub- ling up the rows, in which case the double-row coils should be near the floor. Horizontal coils are more effective than vertical coils in still air, so that all the piping in the sweat room should be run in a horizontal position. 75 Design of the T urkish Bath In still air, such as obtains in all places where direct radiation is used, low radiators or coils are more effective than tall ones of equal surface. This fact has an important bearing on the heating of Turkish bath rooms, for the reason that in the sweat rooms the coils generally extend from floor to ceiling. The loss in effectiveness is general, radiant heat and convective heat both falling below the normal. The reason for this is that only part of each steam pipe in the coil is so placed as to make available all of the heat it emits. Part of the heat given off as radiation strikes the next lower and higher pipes in the coil and is reflected to the emitting pipe or to the v/all back of the coil, and another part of the heat radiates to the wall to which the coil is attached, and from there is re- flected back to the pipe which emitted it, or to the one next higher or lower in the coil. The convected heat suffers in a hke degree. The coolest air in the room is near the bottom, and the cooler the air the greater the heat emission of the pipe; consequently, when air at the bottom of the coil becomes heated it flows up toward the ceiling alongside of the steam coil, growing hotter as it progresses, thereby forming a film of hot air in front of the coils, which absorbs but little heat and at the same time prevents cooler air from com- ing in contact with the pipes. Emmissive Capacity of Radiators and Coils.— The actual quantity of heat given off by a radiator or coil depends upon the diflference in tem- perature between the steam within the pipe and the air surrounding the coil or radiator. 76 Design of the Turkish Bath If, for instance, a portion of a steam pipe con- taining one square foot of radiating surface were extended through the insulated side wall of a build- ing into still air, so long as the temperature of the air and steam remained constant the steam pipe would give off, hour after hour, an equal amount of heat. If a well-insulated box were next placed over the steam pipe, as shown in Fig. 24, and the steam were maintained at the same temperature, heat would begin to radiate to the walls of the box, and air passing over the surface of the steam pipe would abstract more heat, which would be given off, assume, at the rate of 300 B. T. U. per hour. At that rate of emission the box and the air con- tained therein would soon begin to rise in tempera- ture, and, in proportion as the temperature of the box and air increased, the steam Pipe Through Wall emissive Capacity of the steam pipe would decrease correspondingly until a point was reached where just enough heat would be given off to supply that lost by radiation and convection from the outside of the box. When this point was reached, it would be found that instead of 300 B. T. U. , the heat emission of the pipe had dropped to perhaps 30 B. T. U. per hour. The actual amount of heat given off per square foot of surface has been found by experiment to be constant for each degree of difference between the 77 Design of the Turkish Bath temperature of the air surrounding the coils and the temperature of the steam within the pipes, and from this experimental data rules have been de- vised for finding the amount of radiating surface required to heat given spaces. The rate of emission and the number of heat units (B. T. U. ) emitted per hour per square foot of radiating surface for various degrees of differ- ence between steam and air, in horizontal tube radiation, in still air, may be found in Table I. To better understand the value of the table apply it to the heating of the sweat room of a Turkish bath. For example, to heat the hottest room in a Turkish bath to 240° Fahrenheit would require steam at a temperature of about 260° Fah- renheit. The reason for this is that with a less difference in temperature than 20° the heat emission drops so low that a much greater amount of radia- tion would be required to maintain the usual tem- perature of the rooms than if a greater difference in temperature be maintained. Note, for instance, in the table, that for a dif- ference of 10° temperature only about one-half heat unit is given off per hour for each degree differ- ence in temperature between the steam and the air of the room, while with a difference of tem- perature of 20°, 1.11 heat units are given off from the same amount of radiating surface in an equal period of time for each degree difference in tem- perature. It may be assumed, then, that the steam used for heating Turkish bath rooms will have a tem- perature at least 20° higher than that to which the 78 Design of the Turkish Bath Z o < < ca H < H 2 O N o o u< Z o a j=i . t •=§ II Is s i. e 3 One-Inch Pipe Coil 12 Inches High Single Row Pipes lOi-N 8 34 55 77 100 124 148 173 199 228 257 287 318 346 378 410 443 476 612 652 660 762 872 1.020 1.150 1,295 oo 3 E 3 6 Two-Inch Pipe Coil 24 Inches High Massed Pipes co-*t- ■(= E 'o O Four-Inch Pipe Coil 40 Inches High Single How Pipes 6.2 24.9 39.7 56.2 73.0 90. 108. 127. 147. 167. 188. 208. 230. 252. 276. 300. 324. 348. 375. 393. 477. 667. 637. 742. 810. 945. C E 'o U Six-Inches Pipe Coil 40 Inches High Massed Pipes 6.50 20.2 35. 49.6 64.6 79.8 95.2 112. 128. 147. 166. 184. 203. 223. 244. 26.5. 286. 307. 330. 366. 420. 493. ' 563. 654. 7411. 835. c 1 cq 0) y c Is £ M CJ — S 2 = C a 13 "5 w ■p I in c £ 'o U -^ c B 'o O One-Inch Pipe Coil 12 Inches High Single Row Pipes 0.85 1.72 1.84 1.92 2,01 2.06 2.12 2.18 2.24 2.28 2.34 2.39 2.44 2.48 2.53 2.67 2.62 2.65 2.71 2.76 2.91 3.06 3.22 3.37 3.50 3.66 -Inch Coil 24 sHigh ssed pes sggsssssgs^sssssssgg^ssssgsa Two Pipe Inche Ma P CO c £ 3 Four-Inch Pipe Coil 40 Inches High Single Row Pipes 0.62 1,26 1.34 1.40 1.46 1.50 1.54 1.58 1.63 1.66 1.71 1.74 1.78 1.81 184 1.87 1.91 1.94 1.98 2.01 2.12 2.23 2.34 2.45 2.55 2.67 CSl 3 B 'o Q Six-Inch Pipe Coil 40 Inches High Massed Pipes lgS2SSS3SS^i;SS5SSsSEag?9£5S;S^Sg; 1-1 c g •q O Difference in Temperature Between Air and Steam in Degrees, Fahr. OOOOOOOOOCOOOOOOOOOOLOOL.'SCIOO ~z a c a U'3 79 Design of the Turkish Bath hottest room will have to be heated, and a greater temperature of steam than 20° difference will be found even better. For the sake of explaining the table, then, assume a temperature of the air of 140° Fahrenheit and a temperature of steam of 260° for one of the hot rooms in the bath, which is to be maintained to a temperature of 140° Fahrenheit. The temperature of the steam is then 260°, that of the air in contact is 140°, and the difference between the two temperatures would be 260° minus 140°, or 120° Fahrenheit. Glancing down column 1 of the table to the difference of temperature rated as 120° and following the horizontal line to column 3, it will be found that 1.74 heat units are given off per hour from each square foot of radiating surface for each degree difference in temperature between the steam and the air. To find the total amount of heat given off per hour from one square foot of radiating surface with that many degrees difference in temperature all that is necessary is to multiply 120, the difference in temperature, by 1.74, the rate of emission, which gives 208 B. T. U. per hour. Instead of multiplying, the same infor- mation will be obtained by following the horizontal line to column 7, which gives the total amount of heat emitted per hour for one square foot of this size of coil at a difference of temperature of 120°. It will be noticed in the foregoing explanation that the coil is only 40 inches in height and gives off 1.74 heat units per hour, while the 12-inch single row radiation in column 5 of this table gives off 2.39 heat units per hour, or .65 heat unit more 80 Design of the Turkish Bath than an equal radiating surface 40 inches in height. The same law will hold good throughout and the higher the coils extend the less heat will be given off per hour from one square foot of surface. If, therefore, the coils to be installed will extend from the floor to the ceiling, or a good portion of the way there, allowance would have to be made for that fact, and a less rating given the surface, or the coils will be overrated. Perhaps in the ex- ample used in explanation an allowance of 1. 5 heat units per hour per square foot for each degree dif- ference would have been more nearly right. Un- fortunately, there are no experimental data avail- able showing the actual difference for very high coils. The size of radiator or coil in square feet re- quired to supply a certain number of heat units per hour, when the temperature between the steam and air is known, can be found by means of the following rule : Rule. — To find the number of square feet of radiation required to supply a certain number of heat units per hour, divide the total number of heat units to be supplied by the total number of heat units given off per hour from one square foot of surface at the stated difference of temperature. Expressed as a Formula: h — = r c In which h = heat units to be supplied; c = total heat units per hour given off by one square foot of 81 Design of the Turkish Bath cioicic»o5OTa:coooi:-t-l:-t-t-t-t-c-(D(oeo«too«De0«(O»"^"3 h'h.\ 1-1 (^ . jco 1--CO use- O c — •" " ■a (M W D3 CO CO CO--I' "I r-^usi-5LOUi«)^to— 1-^.— lOOOOOOOOOOOCOO _ cot-'— (-^t-i-icO'ai3^i~fMU5I?-asOi~'COCO-^U50^l>-t-t-t-t-WtC'S oi CO U3 tD i^^ ffi o •-! oa -^ i^ w i~ CO o t-H oi CO* •^' iQ to tr^ CO ci o i-f ci co' -w" . '3'^-:r"S'Tt<-^mLOiDmu3Lnmm'XieD^«DeD<£>i;DixicD«>t-r-i>t-t-t- CaO-JWWNe4'MOaNCJNe«lCMWCOMCMNCSlMWC;t-;C»oqa3 00COCOOOOOC-t-;CS 0'fl«u5i~cooioc^Jco-fl'int-^oDCi'oT-'Mcdrj!ij^«>[>-'cooioi-icgco - -c-t-i—c-i— cococoooaacooocooaosoic^oioioioocsoooooo >aMactr-»-(ioCT»cot-.-i-*coojma>cvnnaic^iflcoc^3moo^T)-i>^«cooi-H:omoooc^Ti't-oi.-i-^ mco**-=j'-^-^ioi/iioiO'x>tDtDa;)Oiffl)aioo 00C3OOO00OOOO0O00000OO00O0OOOTH.-1 -* s ? CD0>0i0>00'.DlOC0OC0mN00mi--*0iUDOU3Om0S-*01C0t- Oh5 '^W c-coooot^D-coowiACJeoc-coooTfot-Tj^t-tejt-us-^c^ir-icioaiOi iniocooco'^^'coi-ioit-^co- ^ CJtDOlfH^OJC-^COoqcOtDOii-lMCOCOeQoaOOOeDCOOtOCOOOiJ'Oi'xJ'OO 0'-'Ti*cidwcp«DO>r-<-^(dcoi-icoirtc-oii-Icd'^eDodot--ieo^eDi>^0)0 OOOOOOOOOlCiaiOlOOOO'-H.-'— •-—(.— lOaNMCJCaCOCOCOCOCOCOCOrj" .-Hi-lTHt-l.-l.-ti-Hi-IOJ(r4CMWOJCMCM01NWNCJWCc-t-«5eDi;DcousintnTfTj;Tjj 1.--I U--I •— ' I— 1 >— 1 T— I '-' ■ ' >— ' — J O O CTi 05 t- QD '^ CO N COCOCOCOOTCOCOojcOCOMCOCOeOCOmNiNC^MMMN t-O^ONCOlftC-0004rHC^-01C>)TpiX)00OCaWI>aii-HC0inl>0^M-!I'tD00OtN-^C-0:ti-HMlOt-00— i-fCOM towto«>ir-i^-t~-r-i--oqcC2'3t»ovoiO:iOiOiooooOi-H'--i'-H'Hc:iC)M'a' WNtMNCJCMOKMNC -lIMNfNIWNCOCOCOCOCOCOCOCOCOeOCOCOCOC' i'tDooo(Ncoiflt-o^.-HiM' od 00 od t> l> ix> 5C007COCO:OrOrO!WCOCO OS lO M 00 ■<* r c-^ od oj oi o --i r-i I CO CO CO CO -^ " -l>ododc?;ooI--!l-^MC^'edco-v^Oln^D«D^^^-"o6coccc^it>^~< 3COC003COCOCOCOCOCOCOCOCOCO(rOCC!:OCOCOCOCOCOCOCOCOTOr?COCO u^cDt-ooaiOi-^wcQ-^ujtoc-coasO'-iMco-^ustot-cocnO'-icvico-^iococ-oocio ^ z;z^ oo TH.-Hr-l■MI--^c<^NCMC^l?JW(^^]5^l?«cJCOmC-air-ICO'*tDCOO'— "COIO'X>COO>— icoiotsoc— . — ro^^to C0inC-OM*TtX'00.-HC0l/3C— OiW'^00C0C00000(?J0JCS01(71OOOO'-l'-ii-li-"'-H{NCSCMNC0C0C0C0C0-*-3!"*Tj< ^u." l^UilOCD ■^t-oeoinoOi--ieo^cO'-HeO'X5oorHo:irDooocob^c-oc»)-^cDa5T-Hc^iot>a>rHTi<^oc — m-j-^sco c-t^t-c-t-c— t-t— t-c-t-t-c-c-t-c— i>c— 5ooocoooooo'^ooc«cooooooococ»oooocoa55iocx>coro ajcooocot>iHiftOJcoc-T-^l^^«5lrtlOT)"cocv3WoiI--5oQoioiodo6^^l>^OlOlc^ CTaCiff>05CTlCiOiC5aid010iOaOiOOOOCOOOOOOOOOOOCO( _ - - - - - COO0COCOCOCO00aJCOCO000OCO0000(»00CO0000COO0O0c»000OO000CO00JjC»000000O0Q000XCOCO lO-*^COMi--^OJC-. 'X7C~X;cO CJ l> od OS O f-^' M M CO V lO ^ t~ t-^ QO' si O r-! I-! C\l CO -rjl lO 1/3 ^ C-^ 00 00 5i O t^t-t-C^OOOOOOOOOOOOOQOCOQCCOOOOgjCiOlOiOsOiCr. OiCTSCHiCnOlCriOOOO^^O ^ — o ^oo OlWWc^^wNc^NWMc3Mc^Sc>JWMNMNl^Jl?4c>^c■ac^t^Jc-^c^lMC^lcocococococo TZ :z rocococo la-^CONi-^OaiC-tC-^COr- ICiL— lOCOrHOlt--- tDt^odciOr-i'-HMCO'TlA<:D«Jl>-cdoioOi OOOO' ^-^^ . : J'C>301t--*C-'COU5C<10HCiMO:OC';aitDO .__. ICMCOCOmO(DCDt-OOCJC10>— IrHWCOC^Tirtlrt^Dt- ^-^__ iMOJOJCMOJNCVlC.lC' 8 = 32, substituting these values in the formula: 188 X .41 X 32 = 2361.6 B. T. U. (Answer). Loss of Heat from Skylights. — The loss of heat from skylights varies but little in quantity from the loss of heat from windows, the difference being a trifle greater for skylights, due, probably, to the fact that heated air can get away quicker from a sloping surface than from a perpendicular one, thereby bringing a greater amount of air into contact with the skylight, with consequently greater cooling effect. The loss of heat from sky • lights of various layers with air spaces between can be found in Table IV. This table, like the pre- ceding one, gives the rate of loss in heat units per 91 Design of the Turkish Bath hour per square foot for each degree difference in temperature between the air inside and that out- side. TABLE IV.— HEAT LOSS FROM SKYLIGHTS Chabacter of Skylight Number of Heat Units Given off per Hour from each Sq. Foot of Surface for Each Degree Differ- ence Between the Temperature of Air Outside and Inside. 1.118 Two layers with air space between .. . 62 Three layers, with two air spaces between Four layers, with three air spaces between Five layers, with four air spaces between .53 .44 .36 To find the total amount of heat loss use the rule, or formula, given, simply substituting values given in the table of heat loss from skylights for those of windows. Exposure of skylights to the direct rays of the sun would have a beneficial effect in keeping the heat loss down during the daytime of bright weather, but, as the interior compartments of the building must be maintained at the regular tem- temperatures night and day during all kinds of weather, the beneficial effect of the sun in this re- spect cannot be considered. In the two foregoing tables the heat losses have been given for windows and skylights having five thicknesses of glass. It is doubtful, however, if, under ordinary conditions, more than three thicknesses will be required except in the most northern parts of the United States, including 92 Design of the Turkish Bath Alaska, and Canada. In such climates, where the mercury often falls to 20° below zero and stays there for weeks at a time, economy might dictate more than three layers of glass. Further, if in other parts of the United States a Turkish bath house is to be built in an exposed place, with a pre- vailing strong north or western wind blowing against the exposed walls of the hot rooms in which windows are located, more than three thicknesses of window glass might be advisable. Loss of Heat from Walls. — The loss of heat from walls of various kinds and thicknesses can be found in Table V. In this table the values given are per hour per square foot of surface for each degree difference between the temperature of the air inside and that outside. From the various thicknesses of walls and the number of air spaces listed in the table it ought to be possible to find some one construction suf- ficiently similar to the one being planned to give an approximate heat loss to use as a coefficient. If more air spaces are provided than shown in the table a deduction of from .07 to .08 heat unit will be about right for each additional air space. Air spaces were not listed in the table for walls of greater thickness than 24 inches, for the reasons that in present-day construction it is not likely that anything but foundation walls will exceed that thickness, and if the Turkish bath apart- ments are to be located below the street level, where the walls are over 28 inches thick, the heat loss will be so small that an air space would hardly be necessary. 93 Design of the Turkish Bath TABLE v.— HEAT LOSSES FROM WALLS Character of Walls Thick- ness of Walls in Inches Heat Units Lost per Hour from Each Sq. Foot of Surface for Each Degree Differ- ence Between the Temperature of Air Inside and Outside. Ordinary wall of frame building.. 4 .24 Solid stone, brick or concrete 4 .68 Solid stone, brick or concrete 8 .46 Stone, brick or concrete, with air space 8 38 Solid brick, stone or concrete 12 .32 Stone, brick or concrete, with air space 12 25 Solid stone, brick or concrete 16 .26 Stone, brick or concrete, with air space 16 20 Solid stone, brick or concrete 20 .23 Stone, brick or concrete, etc., with airspace 20 16 Stone, brick or concrete, with two air spaces 20 .09 Solid stone, brick or concrete 24 .20 Stone, brick or concrete, with air space 24 13 Stone, brick or concrete, with two air spaces 24 .06 Solid stone, brick or concrete 28 .174 Solid stone, brick or concrete 32 .15 Solid stone, brick or concrete 36 .129 Solid stone, brick or concrete 40 .115 An important consideration to bear in mind is that all of the values for heat losses given in the foregoing tables are for still air in sheltered places. If the building is to have a northern or western exposure and be subjected to strong winds an al- lowance of 10 per cent, should be added to the 94 Design of the Turkish Bath values of heat losses for the walls, windows or skylights so exposed. Loss of Heat from Floors and Ceilings.— The loss of heat from ceilings of certain construc- tion will be found greater than the loss of heat from floors of similar construction. This difference in heat loss is due to two causes. In the first place, as heat tends to escape upward, any which is conducted downward through the floors has less chance of escaping by convection than has heat which has passed upward to the floor above through the ceiling. In consequence of this fact the rate of transmission or heat losses upward through the ceiling is greater than that downward through the floor. Further, the temperature of the air at the top of a Turkish hot room will be found to be from 10° to 15" Fahrenheit higher than the air near the floor, the exact difference depending greatly upon the height of ceiling, and as there is a less difference of temperature between the air at the floor line and that of the ceiling below than be- tween the air at the ceiling line and that of the floor above there would naturally be a less loss of heat per square foot, even though the rate of loss were the same for both floors and ceilings. The loss of heat through floors and ceilings of wooden and fireproof construction can be found in Table VI. The values given in this table are rather vague and indefinite, as not pointing out the character of fireproof construction referred to. However, the coefficients will serve as a guide to the designer, who must use them with judgment. The values 95 Design of the Turkish Bath TABLE VI. HEAT LOSS FROM FLOORS AND CEILINGS Character of Surface Heat Units Lost per Hour from Each Sq, Foot of Surface for Each Degree Differ- ence Between the Temperature of Air on Opposite Sides. Wooden floor, beams planked over; for .083 Wooden floor, beams celled for ceiling surface .104 Fireproof construction, floored over as flooring .124 Fireproof construction, floored over as .145 given are for still air, and as the floors and ceilings cannot possibly be exposed to wind currents under ordinary conditions no allowance need be made as in the case of the preceding tables. CALCULATING HEATING SURFACE Amount of Radiation Required.— Ha ving the rate of heat emission from radiators and coils and the loss of heat from the walls of buildings, the next step is to determine the amount of heat- ing surface which will be required to supply that heat loss. This, perhaps, can be explained best by a concrete example, showing how to attack the problem. Assume, then, that a Turkish-bath es- tablishment is to have but one hot room, which will be heated to a temperature of 180° Fahrenheit. This room will be 30 feet long, 20 feet wide and 15 feet high. One side and one end, each having le- ge Design of the Turkish Bath inch solid brick walls, are exposed in northerly and westerly directions and will be swept by strong winds. A window 6 by 8 feet, with three layers of glass, is located in the side and one in the end wall. The other two walls are of 8-inch brick and adjoin compartments which will be heated to 110° Fahrenheit. The floor and ceiling are of fireproof construction and are exposed in rooms having temperatures of 70° Fahrenheit. The problem, then, is to find the amount of radiating surface re- quired to keep the rooms at a temperature of 180° in zero weather, with steam supplied under a pres- sure of 60 pounds per square inch. Solving the problem, the following square feet of surface will be found: 654 square feet of 16- inch outside solid wall; 96 square feet of 3-layer window surface; 750 square feet of 8-inch solid in- terior wall surface; 600 square feet of fireproof ceiling surface; 600 square feet of fireproof floor surface, and the following temperature differences will be ascertained: Difference between interior air and outside air, 180° Fahrenheit; difference between air of sweat room and other compartments, 180° — 110° = 70°; difference between air of sweat room and that on other sides of floor and ceilings, 180° — 70° = 110°. The rate of transmission through 16-inch solid walls is .26 heat unit per hour per square foot for each degree difference in temperature between the air inside and outside. This difference of tempera- ture is 180°, therefore each square foot of exposed wall surface will lose .26 X 180, or 46.80 heat units 97 Design of th e Turkish Bath per hour in still air. But owing to the exposures of the walls 10 per cent, must be added to the heat loss in still air, so that the rate of heat loss from the outside walls will equal 46.80 X 1.10, or 51.48 heat units per hour from each square foot of sur- face. As there are 654 square feet of exposed wall surface in the building the total heat loss from the outside walls will be 654 X 51.48, or 33,667 heat units per hour. The rate of heat loss through three layers of glass is .41 heat unit per hour from each square foot of surface for each degree difference between the temperature of the air inside and outside. At a temperature difference of 180° the heat loss per square foot per hour would be 180 X 41, or 73.80 heat units. But the windows are exposed as are the walls, so that 10 per cent, must be added, bringing the heat loss up to 73.80 X 1.10, or 81.18 heat units per hour from each square foot of surface. There are 96 square feet of glass surface exposed, so the total heat loss from windows would equal 96 X 81. 18, or 7, 793 heat units per hour. Heat loss through 8-inch solid brick walls is at the rate of .46 heat unit per hour per square foot of surface for each degree difference between the air on the opposite sides of the wall. In this case, the difference in temperature is 180° — 110°, or 70 Fahrenheit. At a difference of 70° of tempera- ture and a heat loss of .46 heat unit the total heat loss per hour from one square foot of surface would be 70 X .46, or 32.20 heat units. This wall is protected so that no allowance need be made for 98 Design of the Turkish Bath exposures or winds, consequently the" total heat loss from the entire wall would be 750 X 32.20, or 24,250 heat units per hour. There are 600 square feet of fireproof floor surface, losing heat at the rate of .124 B. T. U. per hour per square foot of surface for each de- gree difference in temperature, and there is a dif- ference in temperature of 180° — 70", or 110°. The heat loss per hour, then, would be 110° X . 124, or 13.64 heat units per square foot of surface, and 13.64 X 600, the total number of square feet of surface, gives 8,184 heat units as the total loss through the flooring. The ceiling has the same number of square feet of surface as the floor, and the difference in temperature is the same, the rate of loss only being different. If, then, the rate of transmission for ceilings, .145, be substituted for .124 in the pre- ceding explanation or solution for floor surface, it will give 110 X ,145 X 600, or 9,570 heat units as the total loss from the entire ceiling surface. Combining the heat losses from the several sources then gives: Heat Units Heat loss from 650 square feet of 16-inch wall sur- face 33,667 Heat loss from 96 square feet of three-layer window surface 7,793 Heat loss from 750 square feet of 8-inch brick wall surface 24,250 Heat loss from 600 square feet fireproof floor sur- face 8,184 Heat loss from 600 square feet fireproof ceiling sur- face 9,570 Total heat loss from the compartment 83,464 99 Design of the Turkish Bath It might be well to point out here that as 24,250 heat units per hour pass from the hot room to an adjoining compartment an equal number of heat units may be deducted from the number required to heat that compartment. The total number of heat units lost, which equal the number of heat units required to heat the sweat room (that required for ventilation ex- cepted), have now been determined, and it only- remains to find out how many square feet of heat- ing surface will be required to supply that amount of heat. Referring back to the rule for finding the number of square feet of heating surface re- quired to supply a certain number of heat units per hour, it will be found that all that is necessary is to divide the total heat loss per hour by the total number of heat units given off per hour from one square foot of heating surface at the difference of temperature between the air of the room and the steam in the pipes. In the present explanation steam is to be sup- plied at a pressure of 60 pounds per square inch, which corresponds to a temperature of 307° Fahren- heit, and the difference between that temperature and the temperature of the room, which will be 180°, is equal to 307° - 180°, or 127° Fahrenheit. At this difference of temperature one square foot of coils 40 inches high will emit 230 heat units per hour; then, 83,464 ^ 230 = 362 square feet of radiation. If the coils are higher than 40 inches an al- lowance would have to be made for that fact and 100 Design of the Turkish Bath a lower heat emission assumed, which would in- crease the total heating surface required. In the foregoing explanation no allowance has been made for the heat required for the air in the rooms, as that is a separate item and will be treated separately under the title of "Ventilation". Effect of Exposure on Heat Loss. — Too much emphasis cannot be laid on the effect of exposure on the heat loss from buildings. Experience has demonstrated that in a number of cases. There are numerous instances where a building in a protected position has been heated with steam or hot water and the results were most satisfactory, a tempera- ture of 70° Fahrenheit being easily maintained inside of the building during zero weather. Those same buildings were then duplicated in other localities which were more exposed. The same owners, the same architects, the same builders and the same heating contractors in each case insured the buildings and the installations being identical. Yet, when subjected to the test of extremely windy or zero weather, the plants failed to heat properly. It might be well to further point out that one heating plant cannot well be duplicated in another building having the same exposure, unless the dup- lications are carried to the most minute extreme, so far as radiators, boilers and pipes are concerned. For instance, if the original building had a boiler with a capacity of 1500 square feet of direct radi- ation of the five column 45 inch type, it would not do to use the same size boiler and the same size steam mains, then change from 1500 square feet of five column 45 inch radiation to the same 101 Design of the Turkish Bath number of square feet of 20 inch single column radiation. Owir g to the greater heat emissive capacity of the low type single column radiation, 1500 feet of it would be equal to over 2000 square feet of the larger size radiators, and the boilers and steam mains would not be of sufficient size to supply that amount. It is generally supposed that windy weather cannot affect the heating of a building warmed with steam or hot water. That is a mistake, how- ever, while a steam or hot water heated building is not affected, to so great an extent as one heated with hot air, still it is affected, and this is proven by the facts that the windward rooms are never as hot as the protected rooms, and that a building which is heated perfectly by a plant when located in a sheltered position, if exactly duphcated but exposed in a strong wind-swept location will not be heated properly with the same amount of radiation. 102 VENTILATION OF TURKISH BATHS EGESSITY for Ventilation. - Thorough ventilation of a Turkish- bath establishment is of the utmost importance if the beneficial effect ex- pected of a course of baths is to be realized. The dry heat so necessary for perfect evaporation from the person of the bathers cannot be maintained if a stream of dry air is not con- stantly passed through the sweat rooms to carry off the moisture exhaled and respired by sweat glands and lungs. Without ventilation the air of the hot rooms would soon become so charged with moisture that much of the benefit of a Turkish bath would be lost. In the other rooms of the es- tablishment pure air is as desirable as in the sweat rooms. In the dressing room or separate resting room, where the bathers rest after a course of treatment in the baths, ventilation is absolutely necessary or the patrons will awake with a heavy, dull feeling due to lack of pure air, which will erroneously be blamed to the weakening or de- bilitating effects of the bath. 103 Design of the Turkish Bath No less important than supplying dry air to carry off the moisture from the body and provide a plentiful supply of oxygen for the lungs is the removal of the organic matter, watery vapor and carbon dioxide thrown off by the lungs and skin of the bathers. Each person who undergoes the process of the bath rids his person of from several ounces to several pounds of organic matter in the hot rooms. This matter is as objectionable as other emanations from the body, and if allowed to remain in the rooms to accumulate and putrefy would make the compartments objectionable, to say the least. Quantity of Fresh Air Required. — The quantity of fresh air required for ventilation can easily be foretold when the number of persons the bath is designed to accommodate simultaneously is known. The object of ventilation is to keep the air within the room as fresh and pure as the air outside. This is impossible in practice, however, or, at least, impracticable, owing to the size of ducts and apparatus required and the cost, so the reasonable and logical alternative is to keep the air within the rooms up to a certain standard of purity. To maintain this degree of purity in the sweat rooms of a Turkish bath at least 50 cubic feet of air per minute should be supplied for each occupant the rooms are designed to accommo- date. The number of persons the sweat rooms will accommodate can be found, as previously ex- plained, by allowing for each a floor space 4 feet by 6 feet. Another method of determining the amount of air required for ventilation is to allow 104 Design of the Turkish Bath for a certain number of changes per hour of the air within the rooms. This can easily be done by finding the cubical contents of the room and mul- tiplying it by the number of air changes per hour. One change every six minutes, or ten changes per hour, will be about right for ordinary conditions. Checking up those two methods for computing the amount of air required for ventilation it will be seen that they agree perfectly for the conditions given, and would be approximately the same for any other reasonable height of ceiling or size of floor space. Assuming a floor space of 4 feet by 6 feet and a height of ceiling of 12J feet would equal a cubical content of 300 cubic feet for each person. At one change every six minutes, or ten changes of air per hour, 3, 000 cubic feet of air would be re- quired for each person per hour, according to the air-change-per-cubical-contents method of compu- tation. According to the air-per-person method 50 cubic feet of air per minute would be required, or 50 X 60 = 3,000 cubic feet per hour per person, and as each person would have allotted to him a floor space 4 feet by 6 feet it will be seen that the same amount of air will be required per person by both methods. An allowance of 3, 000 cubic feet of air per hour per person will generally be found sufficient for all rooms, even the hot rooms. No less supply than that, however, should be tolerated. In fact, if any change whatever is made in the quantity of air supplied it should be in favor of a larger, not a smaller amount, an allowance of 4,000 cubic feet per person per hour being a more desirable quantity. 105 Design of the Turkish Bath Method of Ventilating Turkish Baths — In order to move with regularity and certainty the large amount of air required for ventilation some mechanical means must be employed. Natural ventilation through flues and fireplaces, or ventila- tion by aspiration — that is, by means of coils or heaters placed in ventilator shafts — are not suf- ficiently reliable for the purpose. Further, the cost of ventilation by means of aspirating shafts would be far greater than the cost of moving the required amount of air by means of fans or blow- ers. There are two methods of ventilating rooms by means of fans or blowers, which are known, re- spectively, as the exhaust method and the plenum method. In the exhaust method the fan, or blower, is installed in connection with the venti- lating shafts, and the air is drawn from the rooms. This method is unsatisfactory for the reason that the operation of the fan creates a partial vacuum or lowers the pressure within the rooms which are to be ventilated and cold air leaks in around win- dows, doors and other openings, creating draughts within the apartments. Such a leakage inward is objectionable in ordinary buildings and doubly ob- jectionable in the hot rooms of a Turkish bath, for the least draught of cold air in this highly heated atmosphere will be felt and seem excessively cold to the bathers. Further, with the exhaust method of ventilation there is no way to govern the quality or place of introduction of the air, which might be drawn from a contaminated source, thereby being as bad, or worse, than the air which is exhausted. 106 Design of the Turkish Bath The plenum system is the best method of ven- tilation for the Turkish bath. In the plenum sys- tem air is drawn from the outside atmosphere at a selected source, heated by steam coils and forced, by means of a blower, into the several rooms to be ventilated, the excess air being forced out above the roof through specially constructed ventilator shafts. With this system the air within the rooms is under a slight pressure, consequently the leak- age around windows, doors and other openings is outward, so that draughts will not be felt by the bathers; the temperature of the air and points of admission are completely under control, the action is positive, and the air can be heated to any desired temperature before passing into the rooms. Example of Ventilating System. — The con- ventional arrangement of the ventilating apparatus and air-distributing system for a Turkish bath is shown in Fig. 25. Cold air is drawn through the screened inlet duct to the inside of a fresh-air chamber, where it passes successively through an air filter, a tempering coil, a fan or blower and a heating coil to the distributing ducts and flues. The arrangement shown in the illustration may be modified in many ways, and would be, according to the system of ventilation adopted. For instance, instead of forcing the air through the heating coils the fan or blower might draw the air through by suction and force it into the distributing ducts. In that case, of course, the tempering coil would be omitted. Again, if an air washer were used in- stead of a dry-air filter the washer would be placed between the tempering coils and the fan so the air 107 Design of the Turkish Bath 108 Design of the Turkish Bath would be slightly heated before passing through the washer, to prevent the water from being frozen in cold weather. Instead of a vertical steam en- gine a horizontal engine, either direct connected or belted, may be used for driving the fan, or in- stead of a steam engine an electric motor, either belted or direct connected, could be employed. Further, instead of a single duct system of air dis- tribution the double duct system may, and no doubt would be used, while a disc or cone fan could be employed under some conditions instead of the centrifugul fan or blower indicated. The illustration shows only the general ar- rangement and relative positions of the apparatus. The sizes, capacities and construction of the sev- eral parts will be described in detail later on. Fresh-Air Inlet. — As the object of ventila- tion is to introduce pure air into the various com- partments of the bathing establishment, the very first consideration should be to obtain the air in as pure a state as possible and free from dirt, dust, soot or other substances. This can best be done by locating the source of supply in an open space where there is a free circulation of air and placing the mouth of the intake several feet above the ground. In many cases the fresh-air supply is taken in through a screened window. Such a practice is not objectionable when the window is above the basement or ground floor, or even when taken in at the basement or ground-floor window if the yard outside is in every sense clean and the ground sodded or otherwise covered with a clean sui-face and free from dirt and traffic. Ordinarily, 109 Design of the Turkish Bath however, in business districts, where Turkish-bath establishments are generally located, the yard, court, alley, shaft or area from which the fresh- air inlet is to be taken is not wholly free from dust and dirt, not to mention contamination and vitia- tion of the air from other causes. In such cases the better practice is to extend the fresh-air inlet 18 or 20 feet above the ground level, or in some cases it might be desirable to locate it above the roof, in either case turning the inlet face down- ward, so that dust and dirt, snow or rain, will not settle therein, and cover the opening with a screen or grating to keep paper and other light sub- stances from being drawn within by the current of air. It might seem as though the point, or place, of intake would not matter so much when air filters or air washers are used. It should be borne in mind, however, that filters and washers remove only the mechanical impurities from the air; they do not purify it chemically by supplying oxygen which has been depleted or remove the excess of carbon dioxide or other deleterious gases. Even if it did there would be no logic in supplying air from a notoriously polluted source, depending upon me- chanical apparatus to purify it, thereby placing upon the apparatus the extra burden which it would have in removing the mechanical impurities, such as dust, soot and dirt, which would be in ex- cess over that in air taken at a more favorable point. Sometimes, in order to place the fresh-air in- take at a favorable point, it will necessitate the 110 Design of the Turkish Bath construction of brick or concrete ducts underground clear across the building from the point of intake to the fresh-air chamber. When such is the case manholes should be provided in the underground ducts to facilitate cleaning. Size of Fresh-Air-Inlet Duct or Shaft. — The size of inlet duct for a building is determined by the total number of cubic feet of air required per minute for ventilation and the velocity at which the air will be delivered. In ordinary build- ings — that is, those used for educational, business, medical and factory purposes — the velocity of air throughout the heating and ventilating system varies from 900 feet per minute, in some cases, to as high as 2,500 feet per minute in other cases. High velocities are sometimes necessary for eco- nomic reasons in order to keep down both the size and the cost of the distributing ducts, while in other cases high velocities are necessary to insure a thorough diffusion throughout the entire space to be heated and ventilated. In Turkish-bath design, however, high velocities must be avoided, and the highest rate at which air should pass through any of the ducts or flues is, perhaps, 900 feet per minute. If fresh-air-intake ducts, flues or shafts are proportioned to supply the total quantity of air required at this velocity they will not be found too small and will be large enough for any condi- tions which might arise. When the total number of cubic feet of air required to heat a Turkish bath has been calculated the area of duct in square inches required to conduct that amount of air at 111 Design of the Turkish Bath Pi < b O tn u p u o J w > D Z -a; D J S O > > CO ca <: ei fe b oi O fe Q U bs 5 o OS <: OS <: J b eoiAoo coi-'soo COU500 coi^ yn rtursoo ccloco cowoo _ccioco ^wirtco o^ t-i CO LQ of c "^i -^ C-- c^ ■— ' CO ■-^' cc c; oi L'? I^-' ci T-H -^ %c v:^ ~ CO Ln [— C"' M ■^ is c/3 — " CO lo ,_l,_i,_(,_10JC'JM>)C^COCOCOCO'T-^-T'-t'-Tl/5lJ3i-Oi_-5'-i — '-^'^-5C:-C-C-L--C0CC:03:>Ci to -^ :c CJ 'JS ■<*' C^ C-] CD -^ CO (M --D -^ 00 OJ IS -^ CO N '-D "^ 00 OJ IC ^ -T CO OJ ^D ciM-^'^Ci-^— ■■j::>i-^'co'jfcoo'cOi-'r'o6c:Ni/^[>^o'oi'^l>ciw-^tsC5.--H'^'^ -— i'^.-(^i— it-oqt-Mt—cat-cococo(»eocococ?i-Tf--<iMa)-foocMoO'^o*^C'ia>-^o«3wco-^c;ccKMOo-i'OtctMco-^oocJoo-*o (McocO'i'-^m'S'^c-t-oociaiocr^.cjcMcoco'q'i-ourstD'^c^oooocTici^— --ic^M"-^ r-l.-<,-(,HfHr--i.-ir-lr-i-— i.-lMCMMCMCMCgOJ!M(MNOgcOCOCOCOCOCCCO : ffii oo c- -ii lo - -lCgW(MC-JN0JiMC4MCOCOCOC0C0COCO;O'^-^ NCgC'X'-flMOgOCCM'MOCOO'^MOOOeO-^CMOCO^-rMOCOO-fC'SOOOtD-^'MO t-ci =:c•lrr-^oocn.-Hcolntr-cooc]'*^at~cl^oo^n^(»C(^JTl.ln^-c^rHeo^SOT = w omoiraoioomoiooiooioou3Cimou3oiflcnooiaoiaou5ousoiri^-m(;!!i OWU3I— cscoinc-c^-c^iij^t-ocviirtt-ocgioc-owinc-cwinc-ocviSt^oeoir. ii-^ .-lr-l.-ii-JWCQCOMC>3'>*-<*-V!j"miOlOlO^«)tCi(OC-t-C-t-OOOOOOOOClSoiCiO 112 Design of the Turkish Bath any velocity from 200 feet per minute to 1,600 feet per minute can be found in Table VII. Table VII does not give areas for volumes below 100 cubic feet per minute or above 1,000 cubic feet per minute. If areas for quantities larger or smaller than these amounts are required they can be determined readily by multiplying the volume of air to be moved by the area required for the given velocity. For instance, if 6,525 cubic feet of air are to be delivered at a velocity of 900 feet per minute, by multiplying the area (160 square inches) required to move 1,000 cubic feet at a rate of 900 feet per minute by 6. 5, vi^hich is the number of 1,000 volumes to be moved, vv^ill give 160 X 6. 5, or 1, 040 square inches area required in the pipe. If on the other hand, the volume of air to be moved be less than 100 feet, for instance, .65 cubic feet per minute at a velocity of 900 feet per minute, the area (16 square inches) required to deliver 100 cubic feet at that velocity would be multiplied by the quantity of air to be delivered. Thus, 16 X .65 = 10.4 square inches area in the pipe. In this case, of course, the quantity .65 would be treated as a decimal, as it is only 65-100 of the unit amount multiplied. In like manner, when quantities containing odd parts of thousands are multiplied the odd parts must be pointed off as decimal places. Having the area of pipe required in square inches the actual dimensions of the pipe, either for round pipe or square, can be determined by means of Table VIII. 113 Design of the Turkish B a t h Ot'OlCnOOit^CICOOOOdi-Jt-'CDi-'-O^li-'TOOCTCLTiOoai^ME^j-J..^^ c3 rf». CO en it- cj cc c£) «s t^ -a u; i.#^C43Co^^^3^-'l— 'C ^<:DtD00Co-J-JOlO*c^C7l.^^ti.co&^^^to^- WCnOlCnOlCTlCJlCT C71 cji en ui u» > H-o 9. CO £ "tJ < p g S 3 01 "o o hW [^^^^^h-'Ml— 'I— 'OOoeD«D'-DiX)Ooc>ocooo--n-3-q-ji3^oiOic?ac7icncncJioitC>.>i^i(^*i.i^ci3Cococo 'XO:iCn4iJi.Ji-*-C7iCl--3tDtO*--Jh-'UiCDjt^JOtf^pC3WO^plWpOt-JOOOh--'M rom"iCkfnM»--'j-kji:oboboi--'boioIci.wbiN)b3^c-jboJi^bi(D~^ allU^^^DO^^-Jl--'a^oooH-'!^^l^^^--'Oco«^^-*-q^50l^3^^^^i.cJ^OiClcnM^^co^P-cDlf*.ool-'&^wC3 3 C0» CD "O o cnoioitnaimcjioi < 3 g S Jig CF3 3 C-J^ B (t s 3 "S o _.2.m 3 3 3 *i> o a " w ■aa ro^^^s^^^^^^^^^o^^^^^oco^o^^^^^^^3^o^^^^^^H -~30CO-ai>3 01l- ao3eDaii>5000Gii *.cow^o^^a3lf^o^-qe©^^*iOQClc;loa^ocn^ooot^^^o «3 3 3 mi» •jt' bi' in" en' bi' oi' en' en' en' en' 'en' en' bi' cn' w' en* en' en* en' en' Oh-t-pppjo oooooo-j-affjci c.-:3 0i'i■^^'=^c^^--3W^D4i.ocll--'-awco■evoenl--'-JWOOOsoen^D--J^^colP.coen>--'C5MWl£ <32 E >5-"3 3 j^ »5 3 p,{B » COB- has 3 " " O !?3 S =^ 3 3 S fn n It " m 114 Design of the Turkish Bath Suppose, for instance, that a pipe containing an area of 1,040 square inches is required. Glancing down the column of areas to 1,046, which is the nearest value, it will be found that a round pipe having an inside diameter of 36i inches would con- tain that amount of area, and a square pipe having sides 32J inches an equal area. Air Filters— From whatever source the sup- ply of air for ventilation is obtained there will be more or less soot, dust, smoke and other impurities contained therein. If the velocity at intake is very high many particles of light floating matter will be drawn toward the mouth of the duct by the current of air rushing in, and, unless removed, will be discharged into the rooms where the bath- ers are assembled, to soil their persons after being cleansed. It might seem as though the small amount of dust or dirt in a unit quantity of air would be too insignificant for consideration, but it must be borne in mind that the unit quantity is multiplied time and time again, so that the insig- nificant amount of suspended matter becomes an important and noticeable quantity, sufficient to make dusty the rooms of a bathing establishment if the air is not filtered or washed before being discharged into the bathing compartments. A "Whitley" apparatus, designed for filtering air before it is delivered to the fan, is shown in Fig. 26. In the illustration the filter is shown housed in a sheet-metal casing, although it may be enclosed by masonry walls, wood, or any other suitable material. 115 Design of the Turkish Bath The drawing shows the folds or convolutions of the filtering material in a vertical position. In- stead of vertical folds, however, the folds of the fabric may be horizontal, as would be necessary Air Filter if a filter of this type were housed in a horizontal flue. The principle of the filter consists in inter- posing a coarse woven fabric between the inlet to the fresh-air chamber and the fan, so that the 116 Design of the Turkish Bath mechanical impurities will be strained out, as it were, before the air enters the distributing sys- tems. As the fabric will interpose considerable resistence to the flow of air through its meshes the surface of the filtering material must be much larger than the area of the inlet duct to allow for the flow of a sufficient volume of air without ex- cessive friction. Ordinarily, in this type of filter an allowance of 100 to 1 is made. That is, the area of filter sur- face is one hundred times greater than the area of the fresh-air-inlet shaft or duct. In order to get this proportionate area of filter surface in a small space the fabric is folded back and forth as indi- dicated in the drawing. In operation the filtering material must be cleaned occasionally or the meshes of the fabric will become so clogged with dirt that air will not pass through in sufficient volume for efficient ven- tilation. The fabric used for an air filter must be of coarse mesh, like cheesecloth or burlap, but of tough fiber, to stand the strain of air passing through and the cutting and tearing effect of the particles it stops. Air Washers.— Wet filters, or air washers, are more extensively used in mechanical ventila- tion than are dry filters of types similar to the one just described, for the reason that air washers not only purify the air of all dirt and dust, but impart much needed moisture or humidity to it at the same time. This is important in all classes of buildings except Turkish-bath establishments, for 117 Design of the Turkish Bath a humid air is required for efficient ventilation. In the hot rooms of a Turkish bath, on the con- trary, a dry atmosphere with high evaporative capacity is desired, and the addition of moisture to the air is not beneficial. For all other rooms throughout the building, however, an atmosphere with the right degree of humidity should be main- tained. These different requirements of the vari- ous compartments of a Turkish bath should be kept in mind when designing the ventilating sys- tem, so that air for each room will be supplied not only in sufficient quantity but likewise of suitable quality. If an air washer is used for removing the dirt and dust from the air used for ventilation some means should be employed to remove all the moisture possible from the air before it reaches the hot rooms. Four moisture extractors instead of the usual three would probably reduce the excess moisture to such a degree that while it would be suitable for all other compartments in the estab- lishment the relative humidity in the hot rooms would be sufficiently low when the temperature of the air had been raised to the required degree. On the other hand, if a dry filter be used means must be provided to furnish humidity to those compartments throughout the building where a normal or relatively humid atmosphere is desired. A Peerless air washer is shown in Fig. 27. The entire apparatus consists of a tempering coil, spray chamber with spray nozzles, moisture ex- tractors and centrifugal pump. The pump is belt- connected to the engine which drives the fan, so that when the fan is in operation the pump and 118 Design of the Turkish Bath 119 Design of the Turkish Bath air washer will be in operation also. Water is maintained at a certain level in the sump, or tank, at the bottom of the spray chamber, by means of a ball cock connected with the water supply, and the water in this sump, or tank, is circulated through the spray nozzles by means of the centrifugal pump. The bottom of the sump, or tank, is sloped to a point where it is connected to the sewer, a valve or gate controlling the outlet, so that all dirt re- moved from the air can periodically be washed into the sewer. The spray nozzles are so designed and located that when in operation they fill the spray chamber with a fine spray of water, through which the in-rushing air must pass to reach the fan. Between the spray chamber and the fan is interposed a set of three or more extractor units to eliminate all excess water from the air. Tem- pering coils are provided to raise the temperature of the fresh air to 60° or 70° Fahrenheit to prevent freezing of the water in the spray chamber. As the amount of moisture that will be absorbed by air depends upon its temperature, increasing with an increase of temperature and decreasing with a decrease of temperature, it might be well in Turk- ish-bath design to provide less heating surface in the tempering coils so that they would heat the air to a temperature not above 45° Fahrenheit in zero weather. The relatively small amount of moisture then contained in the air of the hot rooms heated to from 140° to 250° Fahrenheit would not be noticeable. 120 FANS OR BLOWERS ELECTION of a Fan— There are only two types of fans which are suitable for ventilation purposes in Turkish bath buildings. One is the centrifugal fan, the other is the cone fan. A centrifugal fan with bottom discharge and full housing of metal is shown in Fig. 28. This is the design commonly adopted for the small and medium fans, such as would be required in a mod- erate-sized establishment. The fan in this illus- tration is direct connected to an upright vertical engine, making what is called a steam fan. Steam fans may be had with any conceivable style of housing, with the discharge outlet at any con- venient point in the casing, and driven by any kind of prime mover. When designing the venti- lation system the best way is to consult the cata- logues of fan manufacturers to find the design of fan suitable for the purpose. A cone fan is shown in Fig. 29. This fan is illustrated with a pulley for belt drive, although, like the centrifugal fan, it may be driven by a belt, 121 Design of the Turkish Bath steam engine or electric motor, care being taken to keep the apparatus out of the way of air supply and delivery. Cone fans and centrifugal fans each have their own field of usefulness. In small establishments, where the air delivery ducts are direct, of large area and short runs, so there will be but little frictional re- sistance of- fered to the flow of air, the cone fan will prove the more satisfactory, as its oper- ation is much less noisy than that of the centrif u- , gal fan. On the other hand, in large estab- lishments, or in any kind of an establishment where the air must be forced a considerable dis- tance through ducts and flues, the centrifugal fan will prove the more efficient, owing to its better delivery against pressure. 122 Fig. 28 Centrifugal Fan Design of the Turkish Bath When centrifugal fans must be located close to the public rooms of a Turkish bath a canvas accordion-fold section can be installed between the fan and the main duct, to overcome the effect of vibrations which otherwise cannot be silenced. Under no consideration should the engine be lo- cated inside of the fresh- air room or in close prox- imity to the air ducts or the throb of the engine can be heard all through the building, a condition which would be very an- noyiag to many of the patrons. When the centrifugal fan is to be used the Fig. 29 Cone Fan steam-heating coils may be so located with refer- ence to the rest of the apparatus that the air either will be drawn through the coils to the fan or so it will be forced through after leaving the fan. With the cone fan, however, the air must be drawn through the coils; it cannot successfully be forced 123 Design of the Turkish Bath through. Cone fans are suitable only for deliver- ing large volumes of air at low velocities, a condi- tion vi^hich obtains in Turkish bath ventilation, where a speed of 1,000 feet per minute is about the highest that will be required in any part of the system. They can be set without casings and are best set in the circular opening of a wall which serves as the inlet from the fresh-air shaft and coil room. Whether a cone fan or a centrifugal fan will prove the better for a certain installation can only be told after a full study of all the related condi- tions bearing on the case. Method of Operating Fan — For general work, on account of its noiseless operation, the electric motor is coming into extensive use as a motor apparatus for driving fans, but in Turkish bath design it is questionable whether the motor will ever supplant the steam engine. The princi- pal reason for this lies in the fact that it is more economical to use the steam engine. Every Turk- ish bath establishment must have a high-pressure steam plant to supply heat to the sweat rooms, operate laundry and other machinery, and, having the available high pressure, true economy lies in using it, as the exhaust steam from the engine can be used for heating the tempering coils. The exhaust steam may likewise be used for heating the main-heater coils, or as many sections of them as there is exhaust steam to supply, if the air in the ventilation system is simply raised to the temperature required for the dressing and other rooms of the building having normal tem- 124 Design of the Turkish Bath peratures, and the air for the sweat rooms raised to their respective temperatures by means of re- heaters. This is, no doubt, the method which would be adopted, so that steam engines would naturally be the best prime movers for ventilation fans. The type of steam engine, whether upright or horizontal, direct connected or pulleyed for belt drive, must be determined in each case from all data available. Ordinarily, where quietness of operation is not essential — as, for instance, when the engine and fan are remote from the bath rooms— a direct connected upright engine will prove the more satisfactory, as it will cost less, occupy less space and be less likely to get out of order than a belt-connected engine. On the other hand, when the fan and engine are so situated or located that noiselessness of operation is essential, a horizontal-engine belt connected to the fan should be provided. When a horizontal engine with belt pulley is installed an auxiliary pulley for operating the cen- trifugal pump that circulates water through the air washer can be provided. Speed of Fans. — The speed at which fans can be operated in Turkish bath ventilators depends upon the greatest speed at which a fan can be operated without making an objectionable noise. This limit has been found to be a peripheral velocity of about 3, 500 feet per minute for straight- blade fans, and about 4,000 feet per minute for multivane fans, and in order to be well within the noise-producing velocity it is well not to operate 125 Design of the Turkish Bath the fans at a greater speed than from 2,800 feet per minute for straight-blade fans, and 3,500 feet per minute for niultivane fans. Indeed, greater economy will result from using larger fans, run at slow speed, than small fans run at high speed, to deliver a given volume of air, while at the same time a smoother-running, less-noisy apparatus will result. It must be borne in mind, however, that true economy does not lie in going to the other ex- treme and installing a huge fan to run at an extremely low speed. No matter what size fan is used it must be run at sufficient speed to overcome the resistance due to friction in the ducts and flues and at the same time maintain a sufficient pres- sure to move the air along to the rooms. This ex- cess pressure must be very slight, however, or too high velocity will be established throughout the distributing system. The fans should never be calculated to run at a velocity which would pro- duce a greater pressure than that required to bal- ance a column of water one-half inch high, which is equal to a pressure of 0.29 ounce per square inch. The velocities produced in a duct 100 feet long and 1 foot in diameter by various pressures in ex- cess of the atmosphere can be found in Table IX. This table was calculated for a barometric pressure of 30 inches and a temperature of 60" Fahrenheit. For other temperature the results would have to be multiplied by suitable factors, but as 60° Fahrenheit is about the temperature at which the air would ordinarily reach the fan for ventilation purposes the values in the table may 126 Design of the Turkish Bath TABLE IX.* VOLUME OF AIR DISCHARGED AT VARIOUS PRESSURES Difference of Pressure Velocity in Feet per Spcond Through Eound Pipe 100 Feet Long and One Foot in Diameter Inches of Water Ounces per Square Inch By Accurate Formula By Appromimate Formula. li = V 2 — gh. Using a Ccefficient of 0.7 .01 .006 4.3 4.6 .05 .030 9.6 9.5 .1 .058 14.5 14.5 .2 .116 19.4 20.5 .3 .174 23.6 25.1 .4 .2.12 27.4 29.1 .5 .289 30.5 32.5 .6 .347 34.0 35.2 .7 .405 36.0 38.3 .8 .463 39.2 40.7 .9 .512 41 43.7 1.0 .579 43.0 45.7 2.0 1.158 61.1 65.2 3.0 1.303 78.0 78.2 4.0 2.316 85.3 91.1 5.0 2.895 8S.2 103.3 6.0 3.474 104.0 113.3 7.0 4.053 114.0 122.1 8.0 4.622 121.0 130.6 9.0 5.221 128.0 138.8 10.0 5.790 136.0 145.7 11.0 6.369 142.0 153.0 12.0 6.948 143.0 159.6 'Carpenter. be taken as given and will be found sufficiently correct. Size and Capacity of Centrifugal Fans.— The size and capacity of fan required for ventila- tion purposes will depend always upon the volume of air to be delivered per minute. When that quantity is determined the selection of a fan which 127 Design of the Turkish Bath will deliver that amount of air is more a matter of judicious selection from manufacturers' catalogues than a matter of calculation. The theory of de- sign of fans is of no importance to the architect, for there are various stock types and sizes which can be had and the variety of size and design is so large that a stock apparatus can be had for almost any conceivable use. All that is necessary is to select one having a sufficiently high rating to de- liver the amount of air required at a low velocity. In selecting a fan from manufacturers' catalogues, however, the designer must not lose sight of the fact that manufacturers' ratings are generally based on the free discharge of air into the atmos- phere. In actual practice, forcing air through a system of ducts and flues, a fan will deliver only from 50 to 65 per cent, of this theoretical quantity, depending on the length and size of the ducts. Ordinarly, in selecting a fan for ventilation pur- poses, it is assumed that the fan will deliver only 50 per cent, of its rated capacity, and one is speci- fied which has a capacity just double that required. The capacities at different revolutions and air pressures, also the dimensions and horse-power of Sturtevant centrifugal straight-blade fans, can be found in Table X. A careful examination of the table will show the economy of using large fans operated at mod- erate speed over smaller fans operated at greater speed. Take, for example, the discharge of 50,000 cubic feet of air per hour into the atmosphere. By glancing at the table it will be seen that a 5 J by 3 foot fan, operated at a velocity of 300 revolu- 128 Design of the Turkish Bath CO z ■< h O b Bi H Z u a Q <: n H K O H en b O CO H U -s! cu Tl< •^ J3M0 J^ 8SJ0II Ol (N O 00 s S ?? g a^jnuij^ J3d i % ^ 1 s g ^ § s S o 133^ aiqno ^ ^ ^ 3 c^' r' !^ O CO a^nuipi jad s '^ " Buot^jnioAay^ CO oo s (N S CO rS ^ ^ u C ttaaj 0[qn3 "* ^' ffi C-i" ^ s fS N ^' o S (M CM a},nuii(j jad > '3 aoB^ OS S a a s S 3 S CO jai^aureiQ 00 CO CD CO eg 00 s saqouj ui g ^ S S 3 s N 00 g eaqouj ui t- SJ O S r- -*« -*> t^aim JO ja^auiBici a Baqoui ui (D (M '^H CO 00 „ 3 00 o UB^ JO ja^auiBid s s a -H ;aa^ ui azig M K >i X X n X »< M 1 J? to t- 00 m s a 3 s 129 Design of the Turkish Bath <; a a w o Q l-H When Burning 40 Lbs. Coal I'er Sq. Ft. of Grate per Hour ■d'-R O -* LT. ?.! 2.911 6.70 3.64 6.97 3.93 7.73 4.81 9.46 5.14 10.10 6.25 12.30 8.09 16.90 9.97 19.1,0 11.50 12.60 14.10 27.80 T-l CO CiL-^ rHCO Vol. of! Gases 600" F. 2,940 4,630 3,140 4,860 4,250 6,560 4,540 7,000 5,770 8.900 6,170 9,520 7,610 1-1,691 18,040 12,400 11,900 18,360 12,730 19,620 16,900 26, lot 18,101 27,901 co'm" CM CO £ 2 CO o -*' CO CJ CO PS feg 1,464 1,746 1,657 1,868 O .-< Cl CO rH lO C- CO 1,036 1,243 1,109 1,330 909 1,090 971 1,165 c-oo t-ds OC- CDC- mo si When Burning 25 Lbs. Coal per Sq. Ft. of Grate per Hour •JH 05[Baa 1.40 2.76 1.69 3.12 2.01 3.95 2.29 4.61 PS Si3 cdc^ -^00 5.55 10.90 6.36 12.60 8.04 16.80 9.10 17.90 tea as Vol. of Gases BOO" F. g§s§ 3.750 5,790 3,930 6,060 5,110 7,890 5,360 8.260 6,650 10,260 6,960 10.710 10,550 16,250 11.000 17,000 16,000 23,100 15.600 24,100 -w -* dCO CM CO « c^ S 1,290 1,548 1,346 1,616 1,072 1,286 1,120 1,344 920 1,102 961 1,152 806 966 842 1.010 irtCO CO C HJ" t- t- O CD t- OOO CO -* CD t- lO CD ID CD ^s ^m When Burning 18 Lbs. Coal per Sq. Ft. of Grate per Hour ■d H a>lEJa .86 1.70 1.04 2.04 1.26 2.46 1.49 2.93 1.70 3.34 2.03 4.00 M CO COM (M-* oiui 3.51 6.90 4.19 8.24 4.97 9.78 5.96 11.70 CO CO •^ 00 Sffi 00 in Vol. of Gases 500° F. 2,220 3.430 2,360 3.630 3,210 4,960 3,400 5,250 4,360 6,720 4,620 7,130 5,670 8,750 6,040 9,310 9,000 13,870 9,530 14.700 12,800 19,700 13,500 20,900 « »^S 1,100 1,320 1,167 1,400 917 1,100 971 1,165 com OCO *^^ ESS -tj" 1/3 CO O U3C0 lO t- 11 IS CO-* 5S ■a-r B c ? (u dH a5(BJa S3 gS O O CC CO tr-ta 0«3 'rH ' rA 1.04 2.04 1.27 2.49 1.36 2.66 1.66 3.25 WCO C«t- rHrH USO CO ID l»t>- '^CO in cj III! THof f-TcM 2,240 3,450 2,390 3,690 3,050 4,700 3,250 5,020 3,940 6,110 4,230 6,530 6,200 9,650 6,620 10,200 8,940 13,770 9,530 14,700 II CMO^j r-t tH 11 eo'c" — ICM «p^s 639 766 684 820 COC- incM ■V lO lOCO oo to rH 03 CO-* -*HJ< coco CO-* Si^ cf. Co c-ico -dH a^lEjg 'tH ' rA .62 1.21 1.04 2.04 .81 1.60 136 2,66 1.28 2.51 2.12 4.16 moi "^cd Vol. at High Mech. Eff. 1,310 2,020 1.550 2,390 1,890 2,920 2,240 3,450 2,530 3,900 8,050 4.700 3,350 5,160 3,940 6,110 5,230 8,070 6,200 9,550 7,640 11,620 8,940 13,770 OO in o CM CO o m" iHi-f cm" CO* K ».:§ ^ I- CDCM CD t- t-O o 00 oa CO -^ rf CO CD m CD CD t- mm oot- !53 3S ■v-y -a- m cgys crj o coco CO -1" CM CO coco CM 00 CM CM TCI S?3 Public Building with Heaters dH ojiEja CO in c^i XI Cl ^^ CO CO com corH COCD -^ Ci ' rA 'rA .91 1.79 1.28 2.51 1.31 2.68 1.84 3.61 r-icO cj-* Vol. at High Mech. Eff. 1,170 1,805 1,310 2,020 1,690 2,610 1,890 2.920 2,300 3,560 2,530 3,900 8,000 4,620 3,350 5,160 OO o o 6,760 10.400 7,540 11.620 oi-i? o'lD « &; s O CD CD OS CO c:^ ■^t- in CO CO [^ coo ooo 0000 -V rr 3S S!§ CO Hi< rrrr Ol CO coco 00= 0-* -3" Cl C— CI CJCvJ C3CO ?i5j S5 lis •d -H 35[BJa ssss ssss rHCO COM rHO rHO 0] 00 00 rH CO '^ ca u5 CDrH -*rH ■*0> 00 CD CO -* CO CM gg rA rHCJ ,-HCO '-^ cg-^ Vol. at High Mech. Eff. 825 1,275 1,310 2,020 1,190 1,840 1,890 2,920 1,630 2,610 2.530 3,900 2,120 3,260 3,350 5,160 3,310 6,100 5,230 8,070 4,770 7,350 7,540 11,620 cocj CJ CO o"io k'^S ON OiOS tHOi ooo CO ■* lO !S cMco -^m nil Ol CM coco rH CM CM CO r-cD C-lCl Exhaust or Ordinary Ventilating Systems Without Heater or Air Screens dH ssss S§SS3 CDi-f CD--I OrH i-ICO gsss SSS3S CDCM CO— 1 1-ICO -^ Ci !33 cot CD W -;l Vol. at High Mech. Eff. 587 905 825 1.275 m m o o ^o o^ rH i-TrH 1.150 1.770 1.630 2,510 1,600 2,310 2,120 3,260 oo oo TP ^ ■•-* Z> CO J3 COi-J_ CM CO co'm 3,370 6,200 4,770 7,350 O C o- -5" O cd'o W Cl| g oico -r-^ Tfca -#0 M CM CO -J' C-OO CMtH O -" cam CIOl c-i -0 r-( t^ CD C- 00 l-H mo ^■N eico m ^ m CO 35 SS 121 145 171 206 OC< t-CD aDuB .2 S .s S s s •I 1 S S s s CO •* lO o r- 00 3-, ] Design of the Turkish Bath oo c" d CM -^J" oo CMua S5 oo IHCO to .-5 ss 00 to cot- 5S 00 S8 §8 00 T-tO tod toco S8 tON gs c- in eg 00 itj 00 t- (M"of CO-fl« S8 COlO §1 "*S Ii S^" S"£" si" is "lO 0" is cncg ~o to sss m"--r 00 CO cgec Si C4CO §8 S"S IS" 8S t> l/S CO to is cQ'm" Cvl cgS ^g 13 ■*U3 ss ss 1-H t- S5S CC.H tHCO woo coin iHCO ss gs ss gs gs 88 8S 3g ss s^ gs ss t-'co 26,700 41,200 27,900 43,000 88 co'oa" CQU3 lO"* coin 00 COI> ■* tD Q si gf:" 11 1000 88 S"g" gg t- t- ii gg" 00 ?23" ""i ii i-s Ie TJ1TJ< ^s CO ■* coco 00m S5S CJ CO ii ^53 SSI OS 00 ss ScO ii is to t35 as mo eg eg 00 l> (COS dS "* -* OOCM CD to to CM* ooo toed §S d d ^8 cot-^ si wm 88 c-co ss" gg 88 g| gg ss CO to sg Kf2 oo gs NCO odrf tN"* is" ii ii 11 88 ""S ""°" gg eg gg mo s"s" 88 CO_C^ tOO) 00 gs CTiCO c-cg 00 gg 00 CO CO"* CO"* ii is SE3 OiliO CJCO §8 OJCO 5g2 eg CO cgc^ 5g eg eg com eg eg "*ai t- to SS COOi SS -* 00 eg to m ss too C-;CO toco «JO 00 in gs 100 oaoa "* ■£> ^00 mo to eg ID d ESS mo 00 t^ c-o ooi (MCg 00 to" l-( gg ss * eg la ino 58 T-ico ss rHCO' s"i t-LO ,-(tN sg .cgni. 00 si ss" ggss .-!_cgm_ sa ii _!Hoi_ 88 it «?o» ii 00 co" oo oo 00 t- 88 0-3 eg Ti< ss" i-H t"^ 88 s"i" 88 too eo-^" s§ c 1 to 00 en 2K 3S t-00 SS3 too . d oj d S >?. opcj T-l OJ , ii r. P rt 3 C *- o •uW 3« 2 c . £ !U M d o 01 Design of th e Turkish Bath tions per minute, will discharge 49,524 cubic feet; a 6 by 3i foot fan will discharge 54,966 cubic feet, when operating at a speed of 238 revolutions per minute; a 7 by 4 foot fan will discharge almost 60,000 cubic feet per minute when operating at a speed of 166 revolutions per minute, and an 8 by 4 foot fan will discharge 47,486 cubic feet, at a velocity of 103 revolutions per minute. In the case of the 5J-foot fan, however, it requires 9.7 horse power to operate it at the speed of 300 revo- lutions per minute, while the 6-foot fan will de- liver approximately the same quantity of air with an expenditure of only 7.1 horse power; the 7 foot with a consumption of 5.8 horse power; the 8 foot with 2.3 horse power. So far as horse power is concerned, therefore, economy would lie in using a large-size fan run at a comparatively low rate of speed. There are other reasons, however, which necessitate the use of large fans at slow speed in Turkish-bath ventilation. As was previously pointed out, in order to prevent noisy operation of the apparatus the fan must be run at a less speed than 3, 500 feet per minute, and preferably not at greater speed than 2,800 feet per minute. This requirement would eliminate the two smaller fans mentioned, which run at peripheral velocities re- spectively of 5,484 and 4,485 feet per minute to de- liver the required volume of air, while the large fans run at speeds of 3,633 and 2,588 feet re- spectively. There is a further consideration, also, which eliminates those two fans. As formerly stated, a fan should be operated so as not to pro- duce a greater pressure than i ounce, with 132 Design of the Turkish Bath preferably J -ounce pressure, and both of the smaller fans above mentioned, which were run at speeds of 238 and 300 feet per minute, produced pressures respectively of f ounce and 1 ounce per square inch. Multivane Fans.— Multivane fans of the cen- trifugal type are becoming extensively used for ventilation purposes, as they are more efficient and less noisy than straight-blade fans when deliver- ing equal volumes of air. The sizes, capacities, uses and various data regarding the Sturtevant multivane fans may be seen in Table XI. The size of fan listed in the first column is simply a size number and does not refer to the diameter of fan. Size and Capacities of Cone Fans. — The capacities of cone fans run at various speeds are given in Table XII. In this table the quantities of air delivered are for actual working conditions through ventilation ducts and flues and at pres- sures ranging from about 3-16 ounce to less than I ounce per square inch. By maintaining a speed midway between the two extremes stated would give a suitable velocity for the purpose required. TABLE XII.— SIZE AND CAPACITY OF CONE FANS Diameter of Fan in Feet Revolutions Per Minute Delivery in Cubic Feet Per Minute Pressure in Inches of Water Horse Power i 250 to 600 6,400 to 12,800 .31 to 1.23 3 5 200 to 400 10,000 to 20,000 .31 to 1.23 6 6 167 to 333 14,400 to 28,800 .31 to 1.23 7i 7 140 to 280 20,000 to 40,000 .31 to 1.23 10 8 125 to 250 26,500 to 61,000 .31 to 1.23 15 10 100 to 200 40,000 to 80,000 .31 to 1.23 20 12 83 to 167 67,500 to 115,000 .31 to 1.23 25 133 Design of the Turkish Bath Cone fans must always be so installed that they can draw air through the heater coils and dis- charge it into a plenum chamber, from which the various ducts are taken. Owing to the low pres- sure against which this type of fan is designed to operate it cannot very well blow the air for venti- lation through the heater coils. Further, it is de- sirable, in case an electric motor is used, to have the fan and motor on the house side of the heater, so the parts will he warm and the oil for the bear- ings in a free condition to run. Cone fans with direct-connected motors are quite noiseless in opera- tion. When direct-connected motors are used they should be both dust proof and moisture proof. Electric motors for fans are generally of the direct- current type, wound for 110, 220 or 500 volts. They may be had direct connected to the fans, connected by means of belts or by means of quiet- running chains. When a motor is to be connected to a fan by belting or chain the motor should be mounted on a sliding bed, so that slack in the belt can be taken up by slightly shifting the motor. COILS AND HEATERS Main-Heater Coils . — The most satisfactory and efficient form of main heater for heating and ventilating purposes is a pipe-coil heater made up of 1-inch wrought pipe with cast-iron steam fit- tings, and screwed into cast-iron base sections, as shown in Fig. 30. In making up a pipe-coil heater the vertical runs of pipe are staggered so that air will have to take a zigzag course through the coil, thus bringing all the air into direct contact with 134 Design of the Turkish Bath ihe heating surface. The coil bases rest on an angle-iron foundation, which is provided at the end opposite the steam inlet with a series of expansion balls or rollers, which permit of easy movement of the base sections under the variations due to ex- pansion and contraction. The steam coils should be divided into sections, which can be shut off or turned on, according to the require- ments of the weather, and when steam engines are used in the building to operate the fan or for other pur- poses some of the sec- tions should be so connected that exhaust steam can be used. Sometimes the heaters are located at lower elevations than the steam, boiler, or, at all events, at such low levels that water of condensation can- not be returned to the boilers. When such is the case, or when exhaust steam is used, either atmos- Fig. 30 Pipe Coil Heater 135 Design of the Turkish Bath pheric steam traps must be used in connection with the heaters, so that the water of condensa- tion can discharge to waste, or return pumps or other apparatus must be provided to return the water of condensation to the boilers. Automatic air valves are generally used for keeping the coils free from air. Cast-iron radiators may be used instead of pipe coils, but they are not so satisfactory, and are not extensively used. Heat Emitted by Coils. — There is a vast dif- ference between the amount of heat emitted by radiators and coils in still air, such as the air of rooms heated by direct radiation, and the amount of heat that will be emitted by the same surface of radiator or coils under forced blast from a fan. The reason for this is that the heat contained in the steam within the coils can be supplied much faster than it can be absorbed by the air surround- ing the coils, no matter how fast, within practical hmits, it is moved; it stands to reason, therefore, that the more air brought into contact with the coils the more heat will be carried off; conse- quently the more will be emitted by the steam. The actual heat emitted by coils, then, depends not only on the difference in temperature between the air and the steam, as in direct radiation, but, fur- thermore, on the velocity of air forced through the coils. In practice the quantity of heat given off per square foot of surface in a coil depends further on the. depth or number of rows in the coil. The rate of emission is greatest when there is only one row of pipes in a coil, but the total quantity of heat 136 Design of the Turkish Bath emitted with one row of pipe is not sufficient for heating purposes, so that a number of rows must be provided, even though the average rate of emission is greatly reduced thereby. It has been found that the first section of the heater, four pipes deep in a forced-blast heater coil, emits about 40 per cent, of the total heat; the second section of four rows about 25 per cent.; the third about 15 per cent. ; the fourth about 10 per cent. ; the fifth about 6 per cent. , and the sixth about 4 per cent. It will be observed, therefore, that after six sec- tions of four rows each, or twenty-four rows of pipes, have been passed that little or no further heat is absorbed by the air. At all events, the amount is so small that twenty-four rows of pipes may be considered as the maximum depth of heater coil. The amount of heat transmitted per hour at different velocities per square foot of surface for each degree difference between the air in contact with the coils and the steam within can be found in Table XIII. It will be noticed that as the velocity of the air increases the amount of heat given off by the coils Hkewise increases, so that a one-row pipe coil which emits only 8.2 heat units per hour at an air velocity of 360 feet per minute will give off 19.5 heat units at an air velocity of 3,600 feet per minute. In Turkish bath ventilation advantage cannot be taken of that fact, however, for a velocity of 600 feet per minute through the coil is about all that should be calculated.. At this velocity, with a coil sixteen rows deep and steam 137 Design of the Turkish Bath at a temperature of 300° Fahrenheit in zero weather, each square foot of heating surface would give off an average of 300 X 6 = 1,800 heat units per hour. This would be equivalent to a condensa- tion of about If pounds of steam per square foot per hour. Heaters made up of cast-iron indirect pin-type radiators will not give off more than one-half the amount of heat that will be emitted by an equal amount of surface of 1-inch pipe coil. TABLE XIII. HEAT EMITTED BY COILS TO AIR FROM FANS Heat Units Emitted per Hour per Square Foot per Degree Difference Between the Temperature of Air in Contact with the Coils and the Steam Within Rows of Pipe Velocity of Air in Feet per Minute 360 600 722 960 1,200 1,800 2,400 3,iiiin 1 8.2 9.9 10.7 12.0 13.1 15.5 17.6 ]9..'> 4 7.1 8.5 9.2 10.3 11.3 13.4 15.2 ' 17.(1 8 6.1 7.3 7.8 8.7 9.5 11.3 12.9 j 14.5 12 5.5 6.6 7.0 7.8 8.4 9-9 11.1 12.S 16 5.1 6.0 6.4 7.0 7.6 8.8 9.9 11.6 20 4.8 6.5 5.9 6.5 7.1 8.2 9.2 10.9 24 4.3 5.3 5.6 6.2 6.7 7.7 8.6 10.2 Steam Connections to Coils.— The size of steam and return pipes for heater connections not over 100 feet long, for both high and low-pressure steam can be found in Table XIV. In this table the amount of radiating surface is given both in lineal and in square feet. In computing the num- ber of square feet in a coil of 1-inch pipe, 3 lineal feet of pipe were allowed for each square foot of surface. As a matter of fact there are 148.68 138 Design of the Turkish Bath square inches in 3 lineal feet of wrought pipe, but that is close enough to 144 square inches for all practical purposes. In proportioning coils to use exhaust steam from 75 to 100 feet of 1-inch pipe per horse power should be allowed, and the supply connection should be not less than 3 inches for engines up to 30 horse power; 4 inches for engines from 30 to 50 horse power; 5 inches for engines of 50 to 80 horse power; 6 inches for 80 to 120 horse power, and 8 inches for engines of 160 to 220 horse power. TABLE XIV. SIZE OF STEAM AND RETURN PIPE FOR COILS Square Feet of Pipe in Coil High Pressure Low Pressure Lineal Feet of 1-Inch Pipe in Heater or Coil Diameter of Steam Pipe in Inches Diameter of Return Pipe in Inches Diameter of Steam Pipe in Inches Diameter of Return Pipe in Inches inn to l.fiOO 33 to 333 U 1 3J 21 l,nnoto2,noo 333 to 666 2 li 4 3 2,n00 to 3,0(10 666 to 1,000 24 li 41 3 3,000 to 4,000 1,000 to 1,333 2i li 5 3i 4,000 to 6,000 1,333 to 1,666 3 2 6 3* 5,00(:i to 6,000 1,666 to 2,on0 3 2 7 * Area of Heaters.— In proportioning the heater coil it is well to allow a free area between the pipes of about double the area of the fan out- let. There are two reasons for this. In the first place an allowance of about 50 per cent, greater area should be made on account of friction, if the same velocity were to be maintained through the heater as through the fan. But in Turkish bath work a slower velocity is desired, owing to the fact that air must be discharged through the sys- 139 Design of the Turkish Bath tem at an exceedingly slow speed, in order to avoid draughts. In the fresh air inlet and through the fan a velocity of about 900 feet per minute is satis- factory, but through the heater coil the air must be passed at a much slower rate, a velocity of about 600 feet per minute being about right. By increasing the free area of space between the coils, then, by about double the area of the fan dis- charge, or perhaps a little larger, would bring the velocity down to about the right limit. Even should the velocity through the coil drop slightly below 600 feet per minute there would be no per- ceptible loss of heat, for the rate of emission for a drop of speed of 25 feet would not average over .1 of a heat unit per square foot of surface. Heat Required for Ventilation.— In ordi- nary heating and ventilation work sufficient heated air must be supplied to replace the heat lost by convection and radiation from the walls, the heat lost by leakage around doors, windows and other openings and that carried off through the ventila- tion flues. In Turkish bath work, however, the principal rooms are heated by direct radiation, so that all that will be required of the heater coils connected with the ventilation system is to heat the air forced into the rooms. Once the air reaches the rooms, instead of being cooled by loss through the walls it is maintained at its incoming tempera- ture, if not actually raised in temperature by the steam coils within. If the dressing rooms, sleeping rooms and all other compartments outside of the sweat rooms, shampoo room and plunge room are to be heated 140 Design of the Turkish Bath by direct radiation, then the only provision that need be made in the ventilation system is to heat enough air for ventilation purposes. If, on the other hand, the rooms outside of the active bath- ing rooms are to be heated as well as ventilated by the hot blast system, then allowance will have to be made to supply a sufficient volume of air at a sufficiently high temperature to heat the walls, floor, ceilings, cubical contents, and replace that heat lost by leakage, convection and radiation. Once the volume of air to be heated has been de- termined the size of coil required to heat that amount of air can be easily calculated. In doing so Table XV. , giving the number of cubic feet of air warmed 1° by 1 heat unit, will be found con- venient. PROPERTIES OF AIR Temperature of Air for Ventilation.— The air supply for ventilation must enter the several compartments at or above the temperature of the rooms, according to the system of ventilation adopted. If ordinary rooms are to be both heated and ventilated by the hot-blast system the air de- livered to the rooms must have a temperature of about 120° in order to maintain the temperature of the compartments at 70° Fahrenheit. When the air is used for ventilation purposes only, on the other hand, it need not be heated any higher than the temperature of the compartment into which it is discharged. There are so many different tem- peratures to be maintained in the various compart- ments of a Turkish bath establishment that some 141 Design of the Turkish Bath < b O a H Si u Oh o a: a. I > w pa < Oi Elastic Face of Vapor in Inches of Mercury (Regnault) .044 .074 .118 .181 .267 .388 .522 .656 .764 .785 1.092 1.501 1.929 2.036 2.731 3.621 4.762 6.165 7.930 10.099 12.768 15.960 19.828 24.468 29.921 CO Cubic Feet of Saturated Air Warmed lo per Heat Unit 48.7 50.0 51.0 61.8 52.8 63.8 64.9 65.7 56.6 56.7 66.8 57.1 57.8 57.8 58.5 59.1 59.9 611.6 60.5 60.4 60.3 59.5 - Cubic Feet of Dry Air Warmed 1° Fahrenheit by One Heat Unit 48.5 60.1 61.1 52.1 53.2 54.0 55.0 56.2 57.3 68.6 67.2 68.5 59.1 69.5 60.6 61.7 62,5 63.7 65.0 62.2 67.1 68.0 68.9 68.5 71.4 iX> Heat Units Absorbed by 1 Cubic Foot of Saturated Air per Degree Fahrenheit 02054 02006 01963 01924 01884 01848 01822 01812 01794 01790 01770 01751 0)735 01731 01711 01691 01670 01662 01664 01656 01658 01687 vo Heat Units Absorbed by 1 Cubic Foot of Dry Air per Degree Fahrenheit ooqSoooocoocoooooSooocooo -^ Cubic Feet of Vapor from 1 Pound of Water at Pressure Stated in Column 9 3,289 2,252 1,695 1,227 1,135 882 819 600 444 356 334 253 194 161 118 93.3 74.5 69.2 48.6 39.8 32.7 27.1 CO C3 ■g -M § rt o m a <> •g!l 1,092.4 646.1 406.4 263.81 178.18 122.17 92.27 S4.79 64.59 69.54 42.35 30.40 23.66 21.98 15.99 11.65 8.49 6.18 4.45 3.15 2.16 1.402 .357 .OIM ■Bs>, « rt !- r-;c.i^ Temperature of Air in Degrees Fahrenheit CC^tCJC^CvlNO(NOCM(NINC;'M'>J(M(McqOJ'>7MrM'M'M'M ^ CM CO -Tj- lo y? -^ L— t- 00 Oi o C' •-< oi 1:0 "^ ir:i X L- UK :r: c — ( 142 Design of the Turkish Bath method must be adopted either to temper the heated air going- to some rooms so it will not be too warm or to reheat the air for other rooms so it will not be too cold. The size of coil to use in connec- tion with the fan cannot well be determined until this matter is decided upon. If the intention is to heat all air to the temperature of the hottest sweat room and temper with cold air that going to the cooler compartments, then a quantity of heat equal to that required to warm the air used for ventila- tion in the several rooms will have to be supplied by the coil. If, on the other hand, reheaters will be used, the air for ventilation will have to be raised by the main coil only to about lb" or 80 Fahrenheit, depending on how far it will have to travel and the probable loss of heat before reach- ing the several room. In that case the extra heat required for the sweat and the other bath rooms will have to be supplied by auxiliary heaters, as will be explained later. It might be inferred from what has already been said in another chapter that it is better to use reheaters to raise the temperature of the air to the right degree than to raise the temperature of all the air to a very high degree, then temper it with cool air before discharging it into the various rooms. Very hot air seems to be burnt. It is possible that as water is an almost universal solvent for minerals and liquids, air is likewise an almost universal solvent for gases; and that just as potable waters often carry beneficial foreign matters in solution, so air carries other gases which in their natural condition are beneficial to the system. But these 143 De.sign of the Turkish Bath voletile properties of the air are very elusive, and a high temperature is liable to burn them out or drive them off, thus depriving the air for ventil- ation of some of its most valuable qualities or properties. There is no denying the fact that no matter how much air w^e heat and force into a building in form of ventilation, the air of the interior of the building never approximates in w^holesomeness and invigorating qualities of the air outside, particularly the crisp air of a Spring or Autumn morning. 144 AIR-DISTRIBUTING SYSTEMS INGLE-DUCT System— There are two systems of air distribution com- monly used in heating and ventilating work, which are known, respectively, as the single-duct system and the double-duct system. The single-duct system is shown in Fig. 31 and would be used in Turkish- bath buildings where the sleeping, dressing and other rooms, outside of those devoted to the bath proper, would be heated and ventilated by the hot- blast system. In such cases the air for heating and ventilating these rooms would be raised to a temperature not exceeding 140° Fahrenheit, and more probably not to a greater temperature than 120° Fahrenheit, the exact temperature depending on how far the air had to be forced through ducts and flues, the amount of heat it would lose in passage and the excess heat required to replace that lost by radiation and convection from the walls of the rooms. Such temperatures would not be sufficiently high for the sweat rooms of the bath, however, so that in order to raise the temperature of the air to the right degree for those compart- 145 Design of t he Turkish Bath ments indirect stacks, or re-heaters, would have to be provided at the base of the flues, as indicated in the illustration. Sufficient surface is provided in these stacks to raise the temperature of the air for the several rooms to the required temperatures. Cast-iron indirect radiators are shown in this illus- tration, although pipe coils of suitable design, or any other kind of heating surface, may be used. Fig. 31 Reheater in Single Duct Douhle-Duct System. — The application of a double-duct systerh of ventilation is shovi^n in Fig. 32. Two ducts are run to each flue, one duct above the other, the under duct usually being the cold one. At each vertical flue a mixing damper is provided so that cold air, hot air, or air of any temperature between the two extremes can be had 146 Design of the Turkish Bath in the rooms by simply manipulating the dampers. Such a system would be necessary where the air for the sweat rooms was heated to the required temperature for the hottest of these compart- ments, so that air for the other rooms could be tempered with cold air to the several temperatures required in the various rooms The ducts for ventila- tion pur- poses, whether of the single or double-duct system, may be run as a single main or in a di- vided main with bran- ches to the various flues, or separate ducts may be run from the plenum chamber to the several flues. Square or rectangular ducts are generally used in preference to round ones, notwithstanding the fact that round ducts are cheaper, stronger and offer less frictional resistance to the flow of air. The ducts should be covered with several thick- nesses of asbestos paper, corrugated asbestos air- cell covering, or some equally good heat-insulating material, and in exposed places the covering can be finished by drawing canvas tight over the cov- 147 Fia. 32 Double Duct Design of the Turkish Bath ering and painting the canvas to harmonize with the surroundings. The loss of heat from uncov- ered ducts, with velocities ordinarily used in prac- tice, will average about 7 per cent, per 100 lineal feet, while with the slow velocities required for Turkish-bath work will average perhaps 10 per cent, per 100 lineal feet. This heat loss can be re- duced to about 3 per cent, by properly covering the ducts. Velocity of Air in Ducts and Flues.— The velocity of air in the ventilating system of a Turk- ish bath must be considerably lower than velocities in other types of buildings, for in the highly heated temperature of the sweat room the least waft of air is felt to be cold. Velocities which will be found safe in the various parts of the system may be had by proportioning the ducts and flues so that with a velocity of 900 feet per minute through the fan and 600 feet per minute through the heater coils the air will pass with velocities of 500 feet per minute through the main duct, 400 feet per minute through the branches, 300 feet per minute through the vertical stacks and enter the hot rooms at velocities not above 200 feet per minute. For the other parts of the building much higher velocities would be permissible, even as high as 900 to 1,200 feet per minute in the main ducts, 700 to 1,000 feet per minute in the branches, 500 to 800 feet in the vertical flues, and 300 to 400 feet through the registers. As the highest velocity available, how- ever, is 600 feet per minute through the heater, this part of the system can be proportioned for a velocity of only 600 feet per minute through the 148 Design of the Turkish Bath 1 l-l /i\\\ I / /iVW I I/A main duct, 500 feet per minute through the branches, 400 in the vertical stacks and 300 through the registers. The air to other parts of the building could be moved with the same veloci- ties as for the hot rooms, but the faster the air is moved the less is the percentage of loss of heat from the ducts and the smaller the size of flues and ducts required. Location of Inlet and Outlet Registers. — The location of inlet and outlet registers is of the utmost im- portance in ventila- tion, for if they are improper- ly placed the system vsri 1 1 be a failure, no matter how well propor- tioned and installed the other parts of the work may be. The best location for inlet and outlet registers is shown in Fig. 33. The inlet register is located the higher of the two, and should be at least 8 feet above the floor level; on the other hand, as the object of the outlet regis- ter is to remove continuously the bottom layer of air it should be placed at the floor level, and in the 149 ■\\///x\\\// /i\\\///y %' Fig. 33 Inlet and Outlet Registers Design of the Turkish Bath hot rooms by the floor level is meant actually so, not several inches above the floor level. In the other rooms of the compartment a few inches more or less do not matter so much. In the hot rooms long, narrow registers, 6 or 8 inches deep, are preferable to narrower but higher designs. The coldest air in the rooms is right at the floor line, and the coldest air is what should be removed. In addition to the respective elevations of the two registers it will be noticed they are placed in the same wall. This is found to give a better distribu- tion of air than any other arrangement which can be devised. It is well, though, instead of having the inlet register on line directly above the outlet, to have one register at one end of the wall and the other register near the other end when the design of the building will readily lend itself to such an arrangement. Inside walls are preferable to outside walls to run the flues in, as there will be less heat lost by radiation and convection from inside Myalls. Size of Registers. — In order to give the air a fair inlet to the rooms and flues, register faces must be provided having a free area equal in capacity to that of the flues. It is found by experi- ment and in practice that the resistance offered by ordinary commercial registers decreases their actual capacities to such an extent that the area of open- ings must exceed the area of the flues by 33 per cent. In the hot rooms of a Turkish bath, where the velocities are low, in order to interpose as little resistance as possible large openings in the register face will be found preferable to small openings, 150 Design of the Turkish Bath and the area of the openings may be with benefit 50 per cent, greater than the area of the flues. REGULATION OF TEMPERATURE Hand Regulation of Temperature. — All the heating coils used in connection with the Turkish bath should be installed in sections, so that one, several, or all can be cut out or turned on as occa- sion requires to maintain the right temperature in the several rooms. Each section must be suitably valved for this purpose, and, as there are numerous attendants keeping close watch on temperature and other conditions conducive to the welfare of the bathers, the regulation of temperature may well be left to them, and ordinary steam valves only installed to control the several sections of coils. Automatic Regulation of Temperature.— If the amount of money available for building and equipping the bath will warrant the extra outlay the temperature of the several rooms may be main- tained automatically by means of a Johnson or other type of temperature regulator. The Johnson regulator consists of a thermometer or thermo- stat placed in the room where the temperature is to be regulated, an air compressor and air tank in the basement or other convenient part of the building and a diaphragm valve operated by compressed air on the several steam coils. The temperature is controlled by the difference in ex- pansion between two strips of different metals. Expansion of the metals opens a small air valve which admits compressed air to the top of a dia- phragm valve, thus closing the valve and shutting 151 Design of the Turkish Bath off steam from that coil. Contraction of the strips of metal closes the air valve, relieves the air from the diaphragm valve and permits that valve to open, thus turning steam into the coil. Ther- mostatic temperature controllers will operate within a few degrees of temperature, so that the air of a room can be maintained practically at a constant temperature. An apparatus for automatically controlling the supply of hot air to rooms heated by the h o t - blast sys- t e m of h eating and venti- lation is shown in Fig. 34. This is op- erated in much the same manner as the diaphragm steam valves are reg-ulated, the only difference being that a motor is used to operate the dampers, whereas a diaphragm valve is used to turn on and shut off steam from the steam coils. A motor can be had which will shut off com- pletely the supply of hot air and turn on the cold 152 Fie. 34 Automatic Regulator Design of the Turkish Bath or tempered air whenever the temperature of the air within the room warrants such a change, or a motor can be provided which will gradually open and close the dampers or hold them at such angles that the right proportions of hot and cold air will be delivered at all times. AIR MOISTURE FOR TURKISH BATHS Outside of the hot rooms, where the air is preferably dry, the other compartments of a Turk- ish bath should be provided with sufficient moisture to maintain the normal humidity of the atmos- phere. The normal humidity is about 60 to 70 per cent, of saturation of the relative humidity. By relative humidity is meant the percentage of moisture contained in air at different temperatures. For example, at 10° Fahrenheit and 60 per cent, of saturation 1 cubic foot of air would contain .466 grains of water, while at a temperature of 70° Fahrenheit 1 cubic foot of air would contain 4.788 grains of moisture when containing 60 per cent, of saturation. The amount of water, it will be seen, varies thus with the temperature for certain per- centages of saturation, and the amounts so con- tained at different temperatures, but the same per- centage of saturation represents the relative hu- midity at different temperatures. The actual amount of moisture that air will hold depends upon its temperature. When the air is so charged that it will hold no more it is saturated, or contains 100 per cent, at that temperature. Adding more moisture to the air or lowering its temperature will cause a precipitation or condensation— as, for in- 153 Design of the Turkish Bath stance, in rain or dew, which are merely a precipi- tation of moisture due to over saturation or fall in temperature. On the other hand, if air is saturated and its temperature is then raised the percentage of saturation falls and the air can absorb more moisture. The relative humidity of air at different temperatures can be found in Table XVI. TABLE XVI.—RELATIVE HUMIDITY OF AIR Weight of Water in Grams per Cubic Foot of Air S it a 0) Percentage of Saturation 40 SO 60 70 80 911 KHI —10 .114 .142 .171 .200 .228 .256 .28.-, — 5 .148 .186 .222 .259 .296 .333 .i.lu , .192 .240 .289 .337 .385 .433 .481 + 5 .244 .305 .366 .427 .488 .549 .610 10 .310 .388 .466 .543 .621 .698 .776 IB .394 .493 .692 .6H0 .789 .887 .986 20 .494 .618 .741 .864 .988 1.112 1.235 25 .620 .776 .931 1.086 1.241 1.396 1.551 30 .774 .968 1.161 1.354 1.548 1.742 1.935 35 .946 1.183 1.420 1.656 1.893 2.129 2.36<; 40 1.140 1.424 1.709 1.994 2 279 2.564 2 849 45 1.366 1.707 2.048 2.390 2.731 3.073 3.414 60 1.630 2.038 2.446 2.853 3,2i;l 3.6G8 4.076 55 1.940 2.424 2.909 3.,394 3 879 4.364 4.84!) 60 2.298 2.872 3.447 4.022 4.596 5.170 5.745 65 2.713 3,391 4.069 4.747 5.426 6.104 6.782 70 3.192 3.990 4.788 5.586 6.384 7.1S2 7.98(1 75 3.742 4.678 5.614 6,549 7.485 8.420 9.356 80 4.374 6.467 6.560 7.654 8.747 9.841 10.934 85 5.094 6.368 7.642 8.915 10.189 11.462 12.736 90 5.916 7.395 8.874 10.353 11.832 13 311 14.790 95 6.850 8.662 10.274 11987 13.699 16.412 17.124 ino 7.906 9.883 11.860 13 836 15.813 17.789 19.766 105 9.100 11.375 13.650 15.925 18.200 20.475 22.760 110 10.445 13.056 15.967 18.278 20.890 23.501 26.112 115 n.955 14.943 17.932 20.921 23 910 26.898 29.887 120 13.646 17.0.57 20.468 23.880 27.880 30.703 34.115 From the table it will be seen that if air is introduced to a building during zero weather when the humidity of the outside air is normal or contains about 60 per cent, of moisture, 1 cubic foot of air at that temperature would contain only 154 Design of the Turkish Bath .289, or slightly over one-quarter grain of water. When this cubic foot of air is raised in temperature to 70° Fahrenheit, however, it will require 4.788, or over four and three-quarters grains of moisture, to raise it to the relative and normal humidity of 60-per-cent. saturation. This moisture must be supplied to the air in some manner or a feeling of discomfort will ensue and moisture will be absorbed from the woodwork of interior trim, furniture, or from whatever other source it is available. In large bathing establish- ments, where a fountain is provided in the resting room, sufficient moisture will be supplied from this source, which at the same time can be made an 155 Design of the Turkish Bath ornament of the room. When there is no fountain, on the other hand, a humidostat should be pro- vided. Humidostats.— A humidifier for moistening air is shown in Fig. 35. It consists of a shallow pan placed in a hot air duct and kept filled with water through a ball cock supplied from the city water. A steam coil is immersed in the water and the supply of steam to the coil is controlled by a diaphragm valve, which in turn is operated by a humidostat located in one of the rooms of the build- ing. When the humidity in the room where the humidostat is located falls below a certain percent- age of saturation an air valve is opened which operates the diaphragm valve attached to the steam coil, opening the valve so steam can enter. The heat of the steam vaporizes some of the water, which is absorbed by the air for ventilation passing over it, and the air within the rooms is raised to its right percentage of humidity. As soon as this point is reached the humidostat again shuts off the supply of steam to the coil and the evaporation of water from the pan drops to the minimum again. 156 TURKISH BATH DETAILS PLUNGE BATHS lONSTRUCTION of Plunge Baths.- The general construction of a plunge bath depends somewhat upon the place where it will be located. When the pool will be located in the base- ment, first floor, or in any other place where it will rest upon the ground, it may be made as shown in Fig. 36. A concrete tank with walls 10 inches or more thick, according to size and depth, reinforced with expanded metal, is waterproofed by applying a coating of asphalt, hydroline, hydrex, biolite, or other good waterproofing material about f-inch thick, laid hot in burlap of 8 to 12-ounce weight, then while still hot covered with another f-inch coating of the waterproofing material and burlap course, and pressed and broomed to a perfect bond. After the tank has been waterproofed it may be tested by filling with water and allowing to stand for several days to see if leaks are disclosed. Where pipes pass through the walls the pipes 157 Design of the Turkish Bath should be covered with the waterproofing material, then set in place, and the material carried clear through the walls alongside of the pipes. After the walls have been properly water- proofed the tank may then be lined with an 8-inch wall of white glazed bricks as a finish. Sometimes Fig. 36 Plunge Tank only 4-inch walls of brick are used, but in such cases, if there is danger in back pressure from ground water, the site will have to be under- drained. Instead of using bricks, the waterproof course may be laid between two walls of concrete, and the inner surface of the tank covered with 158 Design of the Turkish Bath 159 Design of the Turkish Bath glazed tile. No lime mortar should be used around the plunge, but cement mortar used instead. The plunge bath should be made sloping toward the center or to one end to facilitate drain- age, and so a sufficient depth of water will be avail- able for those to dive in who can swim, while at other points the water will be sufficiently shallow for those who cannot swim. In small size bathing establishments a depth of 3i to 4 feet at the shal- low end and 5 to 6 feet at the deep part will be found sufficient, while in larger plunge baths a depth of 7 feet will not be too much. A gutter ex- tending along one entire side of the plunge tank will be found convenient for use as an overflow and so the top layer of water can be skimmed or floated off by turning on the water, thus carrying off floating impurities from the surface. The gut- ter will also serve as a cuspidor for those who find such a fixture necessary. Around the tank a rail may well be run at points where there is danger of accidental falling into the water, and inside of the tank a hand rail is usually run for the bathers to grasp. Steps, of course, must be provided to facilitate entering and leaving the water. When the Turkish-bath establishment is to be located in the upper floors of a building some means must be provided for supporting the weight of the plunge bath and lining it so as to prevent leaking. One method of building a hanging plunge bath is shown in Fig. 37. The bearing walls must be made sufficiently strong to sustain the tremendous weight of the tank and water, then double I beams of suit- able strength will be required at frequent intervals 160 Design of the Turkish Bath to support the framework for the tank. The frame- work will be made of I beams well riveted together so as to form base and side supports, and inside of this framework a steel tank must be built as a casing for the plunge. This steel tank must be Fig. 38 Waste Connection to Plunge Tank made of sufficiently heavy plates to sustain the weight of lining and water between supports with- out bending or giving, and should be well coated on the outside with some good metallic paint to prevent corrosion. Inside the tank should be 161 Design of the Turkish Bath waterproofed with a couple of good coats of water- proofing and burlap as previously explained, then if perfectly tight may be concreted and lined with tile, glass or marble, or may be finished with an 8- inch lining of glazed bricks, laid up in cement mor- tar. On the outside of the upright I beams form- ing the frame for the plunge tank angle iron may be riveted to support a channel beam for the floor construction; outside of the method of supporting the tank this construction of plunge bath does not differ essentially from one built in the ground, other than slight modification of piping details made necessary by the suspended construction. DETAILS OF THE PLUNGE BATH Waste and Overflow Connection.— The method of connecting the waste and overflow pipes to a plunge bath is shown in Fig. 38. The drain pipe is trapped with a running trap placed inside of, a manhole outside of the tank, and on the tank side of the trap a Y branch is provided for the overflow connection which drains the gutter. The waste pipe proper is valved to hold the water in the tank, and an extension rod extends upward to the wheel located in a valve box at the floor level so the valve can be easily opened or closed. All pipe, where bedded in the masonry of the tank, should be coated with the waterproofing material in order to form a water-tight joint or bond at these points. The overflow pipe should be made large enough to carry off as much water as can be discharged 162 Design of the Turkish Bath into the tank by the supply pipe so there will be no danger of the tank overflowing, and the waste pipe should be made sufficiently large so that the tank can be emptied in an hour or less. The size of drain pipe capable of emptying the tank in this length of time can be found in Table XVII. The quantities given in the table are in cubic feet per minute, so to find the discharging capaci- ties of the pipes per hour the quantities must be multipHed by 60, as there are 60 minutes in an hour. Instead of doing this the cubical contents of the plunge bath may be divided by 60, which will give the rate at which it must be emptied per minute, then a pipe of that capacity can be selected from the table. The quantities in the table are in cubic feet. These, however, can be converted into gallons by multiplying them by 7.5, which is ap- proximately the number of gallons in a cubic foot. Sometimes plunge tanks are located below the level of the street sewer, in which case the con- tents of the tank must be elevated to the sewers by means of pumps or compressed air ejectors. Ordinarily a centrifugal pump direct connected to an electric motor and located in a sump adjoining the plunge bath or at some other convenient point nearby will be found the best method for raising the water. Hand-rail Supports. — One method of pro- viding anchors for the support of hand rails is shown in Fig. 39. A piece of wrought pipe flat- tened on one end to keep from turning, and threaded and bent to the right angle, is built into 163 Design of the Turkish Bath a. Z < a b O a N > X a J < « ., fl^ «) p 3 3"S.S . S^I- 5, J, s.a If X OJ ^1 "S o .. S 01 n 5 3I5.S 01 " P..S ro C3 " 3 P 2 3 ol.S -l2 5! 3 oj'f a.S 4) ,, m qj ' a.S St: b 5 3 <;.£ «T^ 0..S >MJ3 E 4J Mt-i-f^OCOOjmOOi-lM tPCOCOCOCOMOJCvKNCMWi-H COCQCOOJOaOJNMIMt-l'-i'-i M(NOOOOOOOt-C-tO«5(D COCOCMlM(N(MC>)>-6 a i "■- S 3 pi'§-s.s 'P..S ■ a p 2 3^.5 >.$ a oooooooooomo m O K> !£> t- 10 t-^ ■ — -.31 COCO 05000 -#■< _ 00 t- CO N T- OOOOOOOOOlOOCO Mt-Jr-- NOJCvli-d-HiHiHi-li-liHT-li- coo>-Hi— ii-(ocooto-*eoco t— rHt-^^nCitDcooeic—u:) miia-Tf-^-^cQcococooJoaoj ooocooousaiooooio ■^comooi(£>-^ccicoc-mio lO'd'Tr-itcoeococowNNpj Ift0t-U50>COU5031AOOOSO >-(MOCOi-ceDy3COU3t-tf>iO id t-^ ',D CO (B O <^ to -H (D C, ,„ OOCOOMNC-OOOWO'-H i-icococa'.D-'riMcot-m-^roj lOTj'^COCOCOCOCMNCvlOJCM dccccccccccc 164 Design of the Turkish Bath the wall. Then when the glazed bricks or tile, as the case may be, are in place the hand rail can be finished by putting the cast-brass skew plates on the anchor pipes close to the wall, slipping short pieces of brass tubing over the pipe, as shown in the illustration, then screwing into place the cast- brass ring heads. Instead of using iron pipe to bed in the walls, pieces of iron bar may be used, and either threaded for the ring head as in the fCorrcrefe. ii~»^";J' J3'a-^4'.<>"i"kl;; \ yi ross head. Fig. 39 Hand-Rail Support case of iron pipe, or drilled and tapped for bolting the ring head to the iron bar. Copper or brass pipe may likewise be used instead of the wrought pipe and will be found to last longer. Water Supply for Plunge Bath. — The' water-supply pipe for the plunge bath should be made sufficiently large so that the tank can be filled in at least one hour's time. This necessitates the tank being out of use about two and one-half 165 Design of. the Turkish Bath hours, allowing one hour for emptying, one hour for filling and thirty minutes for cleaning. To facilitate cleaning, a hose connection about two and one-half inches in diameter, located close to the tank, will be found convenient. Small tanks in medium-sized institutions can be emptied and filled in less than two and one-half hours, but in the larger institutions it is not advisable to have the plunge bath out of use for a greater length of time, for patrons are likely to put in an appearance for a bath at any hour of the day or night. Unless the city water supply is filtered all water used for the plunge should be filtered before being discharged into the tank. In this connec- tion, however, it is well to point out that a per- fectly sterile water is not necessary so long as it is clear, assuming, of course, that the city water sup- ply is wholesome and comes from an unpolluted source. Domestic filters are made for use with and without coagulating materials, according to the use to which the water will be put. For drinking and household use the filters are designed to not only clear the water, but render it sterile as well, by removing all bacteria. In industrial establish- ments and other places where the water is not for drinking purposes, or will be boiled before use — as, for instance, in canning works — all that is necessary is to clarify the water of all organic im- purities. Where the water is to be made sterile, coagulant and coagulating apparatus must be used in connection with the filter, but for Turkish-bath work, outside of the drinking fountains, a clear water is all that will be necessary. Two filters, 166 Design of the Turkish Bath therefore, may well be supplied where the water is not already filtered— one large filter, or a battery of two large filters, to supply the bath house proper, and a small filter with coagulating appa- ratus to sterilize the water for drinking fountains. Heating Water for Plunge Baths. — Ordi- narily in moderate climates there will be no neces- sity for heating the water in the plunge baths. The object of the plunge is to close the pores and ad- minister the tonic effect of a cold bath, and both these functions will be better performed the colder the water in the tank, within reasonable limits. A temperature of 65° Fahrenheit is not too cold for the water in the plunge of a Turkish bath, although for an ordinary swimming pool, such as is some- times built as an adjunct to a Turkish bath, a tem- perature of 75° Fahrenheit is more suitable. There are a number of ways in which water may be heated, but very few of them are applica- ble to Turkish-bath work. Heating the water by injecting steam into the tank is objectionable for the reason that the process is more or less noisy. Heating by means of submerged coils in the plunge tank is objectionable for the reasons that the pipes are not only unsightly and in the way but, further, unless made of brass or copper, will rust and at all times prove to be dirt catchers. Within a tank where so many people bathe no nooks or crannies should be permitted which will catch, hold or con- ceal filth. Submerged coils, therefore, may be considered too objectionable for use in the plunge tank. 167 Design of the Turkish Bath Steam or fire may be used for heating the water in a tank, but whichever kind of heat is used it should be so apphed that water from the tank will have to circulate through the heater to be warmed. Heating with Hot-water Heater. — When the installation is such that the heater can be Heating Plunge with -Return Pipe. '^^'^'^ heater located at a lower elevation than the tank, so it will stand at least partly full of water when the tank is empty, a hot-water heater of j^ usual type may be used and will, probably, prove the simplest and best. The conventional method of connecting up a water heater to warm the water in a plunge bath is shown in Fig. 40. To use this kind of heater it is necessary that it be located as low down as the bottom of the tank, and preferably lower. A branch of cold-water 168 Design of the Turkish Bath supply is then connected to the heater and the main supply continued on to the tank so that water can be supplied direct, or through the heater. Under no consideration should a valve be placed in either the flovsr or return pipe between the tank and heater. In small tanks only one inlet and outlet connection will be necessary, while in larger tanks multiple connections are advisable. The flow and return connections to the tank may be made at one side, as shown in the illustration, or in extremely large tanks the flow pipes may be connected to one side and the return pipes to the opposite side of the tank. With this type of heater to fill the tank the cold water can be turned on to the heater, the valve a opened and a fire lighted, thereby heating the water as it flows through. After the tank is full the cold-water valve can be shut off, the valve a between the flow and return pipe closed and the water in the tank circulated through the heater by gravity until the right temperature has been reached. If the bi-pass controlled by the valve a were omitted hot water would not flow out of the heater when the tank was filling, the cold water would short circuit right through the return pipe to the tank, while steam would blow through the flow pipe. Placing a valve or check in the return pipe would overcome this, but, as previously stated, checks or valves of any description are objectionable in the flow and return pipes. Heating Water with Feed-water Heaters. —When the plunge bath or swimming pool is situ- ated at a lower level than the heater the water 169 Design of the Turkish Bath cannot be circulated by gravity, but must be pumped through the heater. Instead of using a heater in such cases the water could be heated by discharging steam into it, but, as before stated, that practice is noisy, and there is more or less odor from the steam which would be objectionable in the plunge room of a Turkish bath. One method of heating water at a lower level is shown in Fig. 41. In this method an ordinary feed-water heater, heated by a steam coil, is used, and the water circulated through it by means of a pump. When filling the tank the water can be passed through the feed-water heater, using live steam; then, when the tank is filled, the water can be circulated through the feed-water heater by means of the pump, and the exhaust steam from the pump used for heating the water. It will be observed that the feed water is connected with both the live steam pipe from the boiler and the exhaust steam from the pump, so that either may be used, and the pipes are all properly valved with this object in view. In the exhaust pipe to the roof there is a back pressure valve, so that in case the steam consumption in the pump exceeds that of the feed-water heater the excess can escape to the atmosphere without imposing an unnecessary back pressure on the pump. The illustration shows the pump connected with a bi-pass to sewer, so the water can be dis- charged to waste instead of circulating it through the heater in case it is desired to empty the tank. This bi-pass will not be necessary, however, if the 170 Design of the Turkish Bath 171 Design of the Turkish Bath bottom of the plunge is above the level of the sew^er. The water of condensation from the steam coil in the feed-water heater can be returned to the boiler through the return steam trap, or dis- charged to the sewer through the atmospheric steam trap. Heat to Be Supplied to Plunge Baths.— There are two conditions to be considered in the heating of water for swimming pools and plunge bath. They are, first, the heating of the water in the pool to the required temperature, and, second, maintaining the water at that temperature. Ordi- narily, in Turkish-bath work, all that will be neces- sary is to raise the temperature of the water to the right degree. The pool or plunge is generally located in a room heated to about 110" Fahrenheit, so that instead of heat being lost by radiation, con- duction and convection, as it would be in a colder place, it is actually gained from the higher tem- perature of the surroundings. All that will be necessary in such case, then, is to heat the water to a temperature of about 65" Fahrenheit if the city water delivered is of a colder temperature. Swimming pools, on the other hand, which are heated to a temperature of 75° or 80" and are ex- posed in places where there is danger of loss of heat to surrounding objects or a cooling by evapo- ration and convection, must have provision made for maintaining the temperature. The loss of heat will depend upon the difference of temperature between the water in the tank and the earth or 172 Design of the Turkish Bath air outside and the conductivity of the walls, as in the case of heat losses from rooms, and the same values may be used. It is not likely, however, that such a refinement of calculation will be neces- sary, for the apparatus required for heating the water in the tank will prove of such large capacity that when operated at its slowest rate it will be sufficiently large to maintain the temperature of the water at the right degree. To find the amount of heat required to heat the water in the tank the number of heat units necessary must be determined. This will depend upon the cubical quantity of water and the num- ber of degrees it must be raised in temperature. It must be borne in mind that a heat unit is the quantity of heat required to raise the temperature of one pound of water one degree Fahrenheit — so to find the number of heat units the cubical con- tents of water must be reduced to pounds of water. This can be done by multiplying the number of cubic feet of water by 62.355; or if the contents are in gallons, by multiplying by 8.33, which is the weight of one gallon of water. Having the num- ber of pounds of water, multiplying it by the raise in temperature will give the heat units required. For instance, if the water is to be raised in tem- perature from 55" to 75" Fahrenheit, a range of 20°, then multiplying the total number of pounds of water by 20 will give the heat units required. The heat required to warm the water in a tank can be found by this rule: 173 Design of the Turkish Bath Rule. — Multiply the weight of water in pounds by the rise in temperature of the water. The product will be the total heat units required. Expressed as a Formula: wt = h. In which w = weight of water in pounds; t = de- grees temperature water is to be raised; h = heat units required. Example.— A plunge bath is 30 feet long, 20 feet wide and has an average depth of water of 5 feet. How many heat units will be required to raise the temperature of the water from 50"" to 75" Fahrenheit ? Solution.— A volume of water 30 x 20 x 5 feet contains 3,000 cubic feet, or 62.355 X 3,000 = 187,- 065 pounds. The difference between 75° and 50 = 25". Substituting these values in the formula: 187, 065 X 25 = 4, 676, 625 heat units. (Answer. ) Knowing the quantity of water to be heated, the next step is to find the size of apparatus to heat that quantity of water, or the capacities of water heaters and steam coils. Capacity of Water Heaters. — The capacity of a water heater depends upon the amount of coal it can efficiently burn during a given period of time and the conductivity and thickness of the walls of the firebox. Boiler iron is a better con- ductor of heat than cast iron, therefore a boiler- iron heater of given surface will heat more water in an hour than will a cast-iron heater of equal surface, the amount of coal burned and the in- tensity of the fire in both cases being equal. The amount of coal economically burned in a heater de- 174 Design of the Turkish Bath pends upon the area of grate and size of the smoke flue. Heaters burn from three to six pounds and will probably average four pounds of coal per hour per square foot of grate surface. The total heat of combustion of a pound of coal of average com- position is 14.133 B. T. U. Of this amount, hovp- ever, a large percentage passes up the chimney as hot gases, so that under ordinary conditions only about 8,000 B. T. U. are actually transmitted to the water. Therefore, in calculating the capacity of a heater, the area of grate surface, amount of coal efficiently burned and the available B. T. U. in a pound of coal are the limiting factors. Archi- tects and plumbers should determine for them- selves, by calculation, the heating capacity of a heater, and not rely upon manufacturers' ratings. This is made necessary by the lack of uniformity among manufacturers in the rating of their heaters, which differ from one another in some cases over 100 per cent, for equal area of grates. Some part of that percentage might be accounted for by the difference of construction, which gives some heaters greater heating surface than others, but, making due allowance for the improved design of some heaters, they will invariably be found over- rated, while the run of heaters are overrated from 20 to 50 per cent. The capacity of heaters can be calculated by means of the rule or formula fol- lowing: When the quantity of water to be heated per hour is known the size of grate required can be found by the following rule: 175 Design of the Turkish Bath Rule. — Multiply the weight of water in pounds by the number of degrees' rise in temperature, and divide the product by the number of pounds of coal burned per hour per square foot of grate sur- face times the number of heat units transmitted to the water from one pound of coal. The result will be the area in square feet of grate required. Expressed as a Formula: g = wt ^ cu. In which w = weight in pounds of water to be heated; t = degrees Fahrenheit water is to be raised; c = pounds of coal burned per hour per square foot of grate; u = units of heat absorbed by water from each pound of coal; g = area of grate in feet. Example — What size of grate will be required to heat 300 gallons of water per hour from 62" to 212° Fahrenheit, 1 gallon weighing 8.3 pounds? SoLUTION.-^50ii^#,X(|l|^62) ^ 7,7 b X 6,000 square feet grate surface. (Answer. ) In the above solution six pounds of coal was assumed as the consumption per square foot of grate surface because the maximum rating of the heater is desired. The capacity of a water heater of known dimensions can be ascertained by the following rule: Rule.— Multiply the consumption of coal per square foot of grate surface by the number of B. T. U. transmitted to the water from each pound of coal, then by the number of square feet of surface 176 Design of the Turkish Bath in the grate, and divide the product by the weight of one gallon of water times the degrees of tem- perature the water is raised. Expressed as a Formula: q = gcw -^ pt. In which g = size of grate in square feet; c = pounds of coal burned per hour per square foot of grate surface; u = units of heat absorbed by the water from each pound of coal ; p = 8. 3 weight of 1 pound of water; t = degrees Fahrenheit water is raised; q = quantity of water in gallons heated per hour. Example — How many gallons of water can be heated from 62° to 212" Fahrenheit in a heater with 7.7 square feet grate surface? „ 7.7 X 6 X 8,000 „„„ o . ■ , Solution. - g.g. x (212- 62) = 296.8. (Ans. ) Smoke Flues. — It is important that a good chimney flue, straight and smooth inside and pro- portioned to the area of the grate, be provided for each water heater. No other smoke pipe should be permitted to connect to this flue, nor should other openings to it be permitted, as they would spoil the chimney draught. Smoke flues should be cased with flue linings to give them a smooth in- terior surface. The best form for flue linings is round or oval, as smoke and hot gases pass up with less frictional resistance in a round flue than in a square one. Square flues are much more efficient than rectangular ones, on account of the less surface exposed for a given area of flue; for instance, a flue 12 x 12 inches has an area of 144 177 Design of the Turkish Bath square inches and a perimeter of only 48 inches, while a flue 8 x 18 inches having an equal area has a perimeter of 52 inches, thus presenting four ad- ditional inches to offer resistance. No satisfactory formula was ever devised to calculate the area of smoke flues under varying conditions. A simple empirical rule that will be found satisfactory for determining the area of flues for water heaters follows: Rule. — Allow for smoke flue one-eighth the sectional area of heater grate. Example. — What size of smoke flue will be re- quired for a water heater containing 4 square feet of grate ? Solution. — 4 square feet = 576 square inches. i of 576 = 72 square inches = area of smoke flue. The nearest sizes of commercial flue linings are: square, 8i X 8 J inches = 72.25 square inches; round, 10 X 10 X .7584 = 78.54 square inches. Capacity of Steam Coils. — Steam coils for plunge baths and swimming pools are not advisable, as previously pointed out, although when necessary they may be used and may be made of copper, brass, or iron pipe, although brass or copper are the more suitable. Copper and brass pipes last longer than iron and transmit more heat to the water per square foot of heating surface, thereby requiring less room in the tank, and, besides, are easier to keep clean, as they do not rust. For these reasons either copper or brass coils are pre- ferable to iron pipe coils. The size of steam coil in square feet required to heat a certain quantity 178 Design of the Turkish Bath of water in a given time can be found by the following rule: Rule. —Multiply the weight of water in pounds by the number of degrees temperature Fahrenheit the water is to be raised and divide the product by the coefficient of transmission times the difference between the temperature of the steam and the average temperature of the water. Expressed as a Formula: s = wr ^ c (T - t). In which s = surface of copper or iron pipe in square feet; w = weight in pounds of water to be heated; r = rise in temperature of water; t = aver- age temperature of the water in contact with coils; T = temperature of steam; c = coefficient of trans- mission. The value of c for copper is 300 B. T. U. and for iron 200 B. T. U. transmitted per hour per square foot of surface for each degree difference between the temperature of the steam and the average temperature of water. In computing the heating surface of copper or iron pipe in steam coils the inner circumference of the pipe must be taken, as that is the real heating surface to which heat is applied. The average temperature of the water in contact with the coil is taken as the temperature of the water. Example. — How many square feet of heating surface will be required in a copper coil to heat 300 gallons of water per hour from 50° to 200° Fahren- heit with steam 15 pounds pressure? Solution.— 300X8.3=2490 pounds of water to be heated; 200° - 50° = 150° = rise in temper- 179 Design of the Turkish Bath ature of water; 150" -^ 2 + 50" = 125" = average temperature of water; 250" = temperature of steam at 15 pounds gauge pressure (Table II) ; 250°— 125° = difference between temperature of steam and average temperature of water. Substituting these values in the formula: ^ = 30'^f25'o^5) = ^ ^ '^"^'^ ^^^* °^ '=°"- (Answer.) Some convenient constants for steam coils that produce approximations sufficiently accurate for most purposes follow. The values will be found safe : W = gallons water to heat per hour. Wh-10 = square feet iron pipe coil required for exhaust steam. W -=- 15 = square feet copper coil required for exhaust steam. Wx .07 = square feet iron pipe coil for 5 pounds steam pressure. W X . 045 = square feet copper pipe coil for 5 pounds steam pressure. Wx .05 = square feet iron pipe coil for 25 pounds steam pressure. W X .035 = square feet copper pipe coil for 25 pounds steam pressure. Wx .04= square feet iron pipe coil for 50 pounds steam pressure. W X .025 = square feet copper pipe coil for 50 pounds steam pressure. Wx. 03 ^square feet iron pipe coil for 75 pounds steam pressure. 180 Design of the Turkish Bath WX. 02= square feet copper pipe coil for 75 pounds steam pressure. Taking the foregoing example for comparison the nearest value to 15 pounds steam is 25 pounds, and the cot fficient for copper pipe at this temper- ature is .035. Hence 300 X .035 = 10.5 square feet. (Answer. ) Heating Water by Steam in Contact.— The quickest and most economical way to heat water with steam is to bring the steam into direct contact with the water. This method is used ex- tensively to heat water in swimming pools where noise is not objectionable, vats for industrial pur- poses, dish washing, etc., and is usually accom- plished by forcing steam through a perforated pipe or steam nozzle located near the bottom of the tank and submerged by the water. When per- forated pipes are used for this purpose they should be of brass or copper to prevent corrosion, and the combined area of the perforations should be at least eight times the area of pipe to equal it in capacity. Exhaust steam from pumps, engines or other apparatus that is liable to contain oil or grease is not suitable for this purpose. When steam is brought in contact with water in an open vessel steam bubbles are formed, rise toward the surface and collapse with a report. For this reason water is heated by steam in direct contact through perforated pipes only when noise is not objectionable. Steam Required to Heat Water. — The weight of steam required to heat a given quantity 181 Design of the Turkish Bath of water when brought into contact with it can be found by the following rule: Rule. Multiply the number of pounds of water to be heated by the number of degrees tem- perature the water is to be raised, and divide the product by the total heat of steam at the pressure it is to be used, less the sensible heat at atmospheric pressure. Expressed as a Formula: s = wh^(L-l) In which s = weight of steam in pounds; w = pounds of water to be heated; h = degrees Fahren- heit water is to be heated; L = total heat of steam at pressure used; 1 = sensible heat at atmospheric pressure. Example.— How many pounds of steam at 70 pounds pressure will be required to heat 7,500 pounds of water from 48° Fahrenheit to boiling point? SOLUTlON.-'^-^^5g^2-48)^ 1^237 pounds of steam. (Answer. ) An empirical rule that is sufficiently approxi- mate for most purposes is to allow one pound of steam for six pounds of water to be heated. Tak- ing the above example then 7,500-^6 = 1,250 pounds of steam. (Answer. ) The temperature of steam at different pres- sures can be found in Table II. To use this table add 14.7 to the reading of the gauge to get abso- lute pressure; the quantities desired will be opposite this figure. 182 EXAMPLE OF A TURKISH BATH }0 far, the design of the Turkish bath has been treated mainly from an en- gineering standpoint, no consideration having been given to the architectural features of the building interior. In- deed, little can be said on that subject, for the architectural treatment must of necessity differ with the amount at the disposal of the designer. The best way, perhaps, to treat the matter of architectural design, is to show views of the in- terior of one of the finest Turkish bath buildings in the country, the Fleishman Baths, of New York, designed by Buchman and Fox of that city. The entrance to the dressing rooms is shown in Fig. 42. There are 200 well ventilated dressing rooms in the group, each provided with a large comfortable divan, for sleeping or resting after the bath. From the dressing room the bathers go to the first hot room, or tepidarium, shown in Fig. 43, which is kept at a temperature of 150 degrees Fahrenheit. From this room, those who require a greater heat can pass to the second hot room, or calidarium, 183 Design of the Turkish Bath where the temperature is maintained at 180 degrees Fahrenheit. A stay of from ten to fifteen minutes in these rooms is generally sufficient for persons who are not taking the baths for remedial purposes. From the dry hot rooms the bather can pass to the vapor room, Fig. 44, if he wishes, or those who W' "^'fM^€ Entrance to the Dressing Rooms wish to take a Russian bath without entering the hot rooms start here in their course of treatment. This room is kept filled with steam vapor at a temperature of 112 degrees Fahrenheit, which is about as high a moist temperature as the average person can stand with comfort. 184 Design of the Turkish Bath In the shampooing rooms, Fig. 45, the bathers are washed and scrubbed and kneaded, until all the loose and dead skin or epidermas hanging to the body is removed and washed away. Cold and hot water, and sea salt if it is wanted, combine to clean the hair and body, and cleanse the little pores of the skin of all foreign matter. There are four- Fig. 43 The Calidarium, Hot Rooms teen of these little rooms, and in this department are also the various needle, rain, douche, sitz and other scientific bath equipments. A douche bath, and the manner of using it also, are shown in Fig. 46. The operator standing at his control table, about twenty feet from the bather, can throw a stream of water on various parts of the bather's body. By manipulation of 185 Design of the Turkish Bath Cr