■ ft. - V - sc RESEARCH LIBRARY THE GETTY RESEARCH INSTITUTE JOHN MOORE ANDREAS COLOR CHEMISTRY LIBRARY FOUNDATION ERRATA TO BURNING CLAY WARES Page 8, twelfth line from the bottom, insert “in” after “build up” to read “build up in water etc.” Page 8, fifth line from bottom, “some” should be “come.” Page 14, fourteenth line from top “potach” should be “potash.” Page 14, eleventh line from bottom, insert a comma “(,)” after “mass” and omit the following “and” to read “portion of the clay mass, the fused material etc.” Page 15, second line under the heading of “The Chief Minerals in Clay,” change the comma “(,)” after feldspar” to a semi-colon “(;).” Page 18, twenty-second line into twenty-third line, “prod- uct” should be “pro-duct.” Page 22, fourth line under “Lime” heading, omit “as’’ to read, “serious menace in clay burning operations.” Page 24, twelfth line from bottom, “motely” should be “motley.” Page 30, fifteenth line from bottom. “Fe 2 0 2 ” should be “Fe 2 0 3 .” Page 56, seventeenth line from bottom, “Besemer” should be “Bessemer.” Page 61, seventh line from top should read, “carbon mon- oxide and nitrogen; from a deeper bed we get carbon.” The semi-colon and the words following it replace the period and the words “The first result is wasteful in”. Page 61, ninth line from top, put space between “mon- oxide and”. Page 63, twenty-second line from top, put “s” after com- bination”, making it plural. Page 67, fifth line under head “Specific Heat”, change “of” to “or” to read “in calories or B. t. u. etc.” Page 69, Heading “Thermal Capacities of Cases” should read “Thermal Capacities of Gases.” Page 69, in the table of “Thermal Capacities of Gases,” in the eighth line of figures beginning “1432” the number in the fourth column should be “784” instead of “748”. Page 73, fourth line from top, the first number should be “.8173” instead of “.9173”. Page 76, twenty-fourth line under heading “Kiln Tem- peratures,” add “(See pages 69, 70 and 71).” to a foot note as follows: “Does not include excess air moisture = 126.7 B. t. u.” Page 76, twenty-fifth line, after this put a referring Page 77, first line under the table, “readily” should be “really get.” Page 77, second line from bottom, “giving” should be “given.” Page 79, put a “*” at the end of the 21st line referring to a footnote as follows: “Note — Recent practice uses pressure in the oil tank.” Page 81, eleventh line from bottom, “molecule” should read “molecules.” Page 82, first line, “bases” should be “gases.” Page 96, second line under heading “Sulphur in the Ash,” “19 per cent.” should read “10 per cent.” Page 98, at the end of first paragraph add, “(See page 69) Page 98, ninth line from bottom, “45 H to CH 4 ” should read “45 H to 35 CH 4 ” Page 99, second and third line from the bottom put a x period “(.)” after “ ” then spaces between this and “x = 2.2843 nitrogen.” 76.5 Page 101, ninth line below the first table, “dtermined” should be “determined.” Page 101, tenth line below the first table following “coal” add “(See page 69)” Page 102, second line from top “71” should be “17.” Page 103, sixth line below Heading “Kiln Temperatures, etc.” add “(See Page 100).” Page 103, second line from bottom put a period “(.)” after x “- ” and put “x = .6071” in the last line since it otherwise 16 cannot be spaced. Page 104, fourth line put a period (decimal point) before “4480” which should read “.4480.” Page 104, in the table toward the bottom of the page, the Figure heading the first column should be “2232” instead of “2230.” The fourth figure in this column should be “1593” instead of “1577.” The last figure (total) should be “6928” instead of “6912.” In the second column of figures change “1786” to “1804” and the last total from “7806” to “7824.” Page 105, fourth line from top, change “2.3975” to “2.4194.” Page 105, in the table toward the center of the page make the following changes: First column of figures, fourth line change “1923” to “1940”; sixth line change “7421” to “7438”; ninth line change “8376” to “8393.’ Second column of figures, fourth line change “2158” to “2177”; sixth line change “8193” to “8212”; ninth line change “9237” to “9256.” Third column of figures, fourth line change “2402” to “2424”; sixth line change “8906” to “9008”; ninth line change “10120” to “10142.’ Fourth column of figures, fourth line change “2661” to “2686”; sixth line change “9813” to “9838.” In the second line below this table change “8376” to “8393.” Page 106, twelfth line “klin” should be “kiln.” Page 113, fourth line above heading “Velocity Head” change “W” to “w” reading “w = weight of a cu. ft. of air at 32°” Page 118, second line in table near the center of the page, “Great Resistance” should be “Grate Resistance.” Page 119, in center of page after “— 25.8 feet” add “(See page 113)” Similarly in the eighth line from bottom after “18.62 feet per second” add “(See page 111)” Similarly in the fifth line from bottom after “= 1.6” add “(See page 118)” Page 120, first line add “to” to the end of this line to read “whereas relative to air at 62 degrees, etc.” In the ninth line from top after “= .06H” add “(See page 114)” Page 123, second line from top add “(See page 70).” Page 123, eighteenth line change “is” to “are” and in the same line change “compartment” to “compartments.” Page 123, fifth line from bottom, change “inch” to “inches” and to the end of this line add “(See page 113)” Page 124, third line after “= .4H” add “(See page 118)” Page 124, sixth line put a decimal point “(.)” before “6.” Page 125, fourteenth line from bottom “drift” should be “draft.” Page 129, twentieth line from bottom “repuired” should be “required.” Page 168, figure 52 is upside down. Page 169, eleventh line from bottom, “oftner” should be “oftener.” Page 174, twenty-sixtji line from bottom, “care” should be “cars”; fourteenth line from bottom, “converer” should be “converor”; eighth line from bottom, “conveyer” should be “conveyor.” Page 175, tenth line from top change “conveyer” to “con- veyor”; fourteenth line from top, change “conveyer” to “con- veror”; twenty-second line from top, change “unites” to “units.” Page 235, nineteenth line from top should follow the six- teenth line to read as follows, beginning with sixteenth line: “drain tile are set in alternate benches, but the tile benches are enclosed with the usual setting of bricks on the heads” then follows the seventeenth line, “to preserve the continuity, etc.” Page 249, fifteenth line from top, “Schnatolla” should be “Schmatolla.” Page 254, fifth line from bottom, change “compartmenet” to “compartment.” Page 284, third line from bottom change “gast” to “gas.” Page 285, twentieth line should read, “ — Assume six com- ' partments — one and two” instead of “one had two.” Page 286, fourth line from top, the first word should be “heim’s” instead of „heims.” Page 329, seventh line from top change “lecing” to “lecting.” Digitized by the Internet Archive in 2016 https://archive.org/details/burningclaywaresOOIove BURNING CLAY WARES By ELLIS LOVEJOY, E. M. Member and Ex-President American Ceramic Society. Member American Institute of Mining Engineers, National Brick Manufacturers’ Association. THIRD EDITION T. A. RANDALL & CO. Publishers Indianapolis, Ind. PREFACE, In telling a story one must have an audience and if the story is worth while it must be adapted to the audience. A learned professor after solving a difficult problem ex- pressed pleasure that it had no practical application. A story from him would be interesting to a few brother scientists, but it would be over the heads of us common mor- tals. The author of the following series has no such high attainment, however much he might wish it, but in his forty years’ preparation as student, clayworker, and engineer in the clay and refractory industries, he has come across many things one would like to know. In writing the articles his first consideration was the audience. To whom shall he write? The thought occurred to him — Why not write to himself? These are the things he would liked to have known when as a beginner in the manufacture of clay wares he was blun- dering through difficulty after difficulty. How big they looked to him before they were overcome, but how they dwindled and were soon forgotten afterward. As the years went by he gained confidence with experience, but new problems were coming up to be solved, other difficulties to be overcome, and he is now telling himself how he could have solved these prob- lems and overcome the difficulties, or how futile were his efforts. As an engineer in the clayworking industries, the field of his experiences greatly widened, and many interesting prob- lems have been presented to him for solution. In his story he has become reminiscent, with himself as the listener. In writing the story he came upon many problems which he found in his experience he had only partially solved and he must work them out to prepare his monthly lecture to himself. The work has been interesting. When one is relating his ex- periences, it is always interesting to him, and if he is talking to himself, it is equally interesting to the one who listens. And so the story has been written by a clayworker to a clayworker, and if other clayworkers find something of inter- est, something of value, in the series, the author will be greatly pleased and rewarded for his effort. A great deal more could have been said and some may wish that it had been included, although it would have made the series unduly long. Others may find the series all too long and will feel that much could have been omitted without any loss to the industry. Both are right, but the author is not concerned because he has just been talking to himself. ELLIS LOVEJOY, E. M. TABLE OF CONTENTS. Page CHAPTER I Clays and Their Mineral Contents Kaolinite Fluxes and Eutectics Bond The Chief Minerals in Clay Kaolin Quartz Mica Lime Feldsar Carbon Iron Materials Magnesia CHAPTER II The Burning Process Watersmoking Oxidation Shrinkage Vitrification Flashing Cooling CHAPTER III Burning Behavior of Clays Color Changes Behavior of Various Types of Clays. . . . Salt Glazing Causes of Blisters Pimples or Rough Pipe Crazing and Cracking Bloating and Black Coring CHAPTER IV Fuel and Combustion The Need of Better Understanding Fuels and Their Burning Properties . . . Measure of Heat Unit Heat Losses in Kiln Burning Evaporation of Water Unburned Carbon in Ash Heat of Ash Incomplete Combustion and Excess Air Specific Ileat Calorific Determination Carbon-Dioxide in Combustion Cases . . Radiation Losses Kiln Temperatures Advantages of Fuel Oil Producer Gas Producer Operation of Producer Carbon in Ash 7 8 9 14 15 15 19 21 22 20 26 28 32 33 33 34 35 36 37 40 42 42 43 50 53 54 54 55 56 56 57 58 60 64 65 65 65 67 69 74 75 76 77 81 81 82 95 CHAPTER V Page Sulphur in Ash 96 Steam for Blast 96 Evaporation of Moisture 97 Tar and Soot 97 Calorific Value 101 Kiln Temperatures from Producer Gas 106 CHAPTER VI Stacks 106 Kiln Stack Problems 106 Intensity of Draft 108 Area of Stack 110 Total Draft Intensity 112 Velocity Head 114 Friction Head 114 Maximum Efficiency Temperature 115 Stack Heads and Weights of Gas Moved 117 Stack Furnaces 117 Height of Stack for Periodic Kiln 118 Continuous Kiln Stacks 123 Construction of Stacks 126 Induced Draft 128 CHAPTER VII Furnaces 136 Secondary Air 137 Coking Table Furnaces 139 Pit Furnaces 141 Unclassified Furnaces 143 Position and Size of Furnace 145 Comparison of Furnace Areas 148 Furnace Doors . 150 Construction of Furnace 151 CHAPTER VIII Kilns 153 Classification of Kilns 153 Periodic Open Top Kilns 154 Kiln Sheds 157 Setting 157 Furnaces and Burning 164 Coaling 171 Machine Handling and Setting 173 Open Top Continuous Kilns. 176 Periodic Crowned Updraft Kilns 183 Down-Draft Periodic Kilns 187 Rectangular Down-Draft Kilns 187 Multiple Stack Kilns 188-204 Rectangular Kilns with Outside Stacks 190 Features in the Construction of Rectangular Kilns 194 Sand Pockets 196 Setting 197 Round Down-Draft Kilns 199 Types of Round Kilns 201 General Construction of Kiln Bottoms 201 Center Stack Kilns 201 Single Outside Stack Kilns Banding Round Kilns Up and Down Draft Kilns Horizontal Draft Kilns Muffle Kilns Muffle Kiln Construction CHAPTER IX Some Notes on Setting Setting for Flame Effect Setting Common or Face Brick Setting for Flashed Brick Setting in Bungs for Salt Glazing Setting Roofing Tile Setting for Vitrified Tile Setting for Spanish Tile and Finials Setting Terra Cotta Setting Sewer Pipe Setting Elbows, Branches, Tees, Traps, Etc Setting Enameled Brick Setting Silica Brick Setting Magnesite Brick Setting Pottery, Porcelain, Abrasives, Etc CHAPTER X The Continuous Kiln Economizer Kilns in General Kiln Dampers Open Top Economizer Kilns Ring or Tunnel Kilns Zig Zag Kiln Compartment Kilns Kiln Arrangement Producer Gas Economizer Kiln CHAPTER XI Car Tunnel Kiln Advantages and Disadvantages Products Successfully Burned Early Car Tunnel Kilns Drayton Kiln Faugeron Kiln Heat Balances of Car Tunnel Kilns Dressier Kiln Hoffman Kiln Comparative Fuel Consumption Zwerman Kiln Harrop Kiln Other Car Tunnel Kilns CHAPTER XII Burning a Down Draft Kiln Coloration, Discoloration and Other Burning Effects Scum Fireflashing APPENDIX Equalization Tables Tables of Rectangles in Equivalent Circles Page . 206 . 214 . 216 . 223 . 225 . 231 234 236 237 239 241 242 243 244 244 245 245 245 246 247 248 250 251 256 257 257 263 264 275 278 294 294 295 295 298 302 303 307 311 311 313 318 322 323 328 328 332 337 339 Copyrighted 1920-1922 by T. A. EANDALL & CO. BURNING CLAY WARES. 7 BURNING CLAY WARES By ELLIS LOVEJOY CHAPTER I. CLAYS AND THEIR MINERAL CONTENTS. A CLAY IS NOT a mineral in a strict sense of the term mineral but instead is a complex mixture of minerals. In order to carry out the burning process intelligently we should have some understanding of the minerals with which we are dealing and their behavior under fire. Each clay is a separate problem because it differs from other clays in its mineral content. The same mineral in dif- ferent clay does not give the same effect in each. Iron, for instance, may produce a deep red color even through a long vitrification range; it may develop a deep red changing to a dark brown; the color may be a pale red or even a buff at certain temperatures while at higher temperatures the ware may become peppered with black iron spots; associated with lime it may produce a pale red at low temepratures, a buff at medium temperatures, and a green at high temperatures. The amount of iron may be the same in each case and the different results are due to the chemical and physical char- acter of the iron minerals and their reaction with associated minerals. The common base of all clays is kaolinite, a hydrated sili- cate of alumina. There are other hydrated silicates which closely resemble kaolinite in their physical properties, par- ticularly plasticity, but so far as our burning problems are concerned we need only consider kaolinite. 8 BURNING CLAY WARES. Kaolinite comes from the disintegration of crystalline rocks, especially feldspar. The latter is a silicate of alumina with some base, such as potash, soda, lime, magnesia, iron, etc. In the disintegration process, silica in part is set free, kaolinite is formed, and the base is set free to combine with some acid radical to form sulphates, carbonates, etc. Kaolinite is considered a final product from the disintegro- tion of feldspar, but there is evidence to show that under conditions not now understood, kaolinite is disintegrated to form bauxite (hydrated alumina) and silica. From the pure feldspar, then, we have several derived minerals any or all of which may be found in the clay mass, but feldspar is only a part, often a very small part, of crystalline rocks. It is associated with quartz, mica, hornblendes, augite, and a long series of other minerals, usually complex, which disintegrate to form other derivative minerals, and all of these minerals, both original and derived, may be found in the clay body. Finally, as the resulting minerals are washed from their source to their final resting place, they are not only under- going mineralogical changes, but they are being sifted and sorted by moving water and deposited as sands and clays, or on the contrary, perhaps, they are being mixed with the waste from other sources and the complexity of the mass increased to develop into trouble for the clayworker. On the one hand we may have pure kaolin and on the other common clay, and between there will be all gradations from the one to the other. Nor is the story fully told — it is too long to tell here — since the development of the mineral mass we call clay only begins with the disintegration of the rocks. The deposits may gradually build up water teeming with life the re- mains of which go into the clay bed to be reckoned with later in our kiln problems; they may subsequently become land surfaces covered with vegetable growth; again be buried to great depths and hardened by heat and pressure, the vege- table matter converted into lignite, coal, graphite, and the pores of the clay mass filled with the distillates of the ani- mal and vegetable remains ; again some to the surface by some wrinkle in the earth’s crust, be eroded and removed to other locations. A flood from the north may bring one bed, superimposed by a bed left by a flood from the east, and this covered by a BURNING CLAY WARES. 9 flood deposit from the west, each bed perhaps widely differ- ent. Not only will the beds be one above the other but each succeeding flood may cut gullies in the beds already depos- ited and fill them with its material. The clays when deeply buried and subjected to pressure and heat are converted into shales but without losing any of their properties; under greater pressure and heat the shales become schists and slate and lose their plasticity which is of first importance to the clayworker. This property is largely irrevocably lost and the schists where exposed crop out as massive rocks. Many of them, however, still contain the rem- nants of the original clay-forming minerals and these are being decomposed to break down the schists and give to them some measure of plasticity which they must have to be useful to the clayworker. Ground waters are circulating through the clay beds puri- fying some and defiling others. Iron minerals are being dis- integrated and the iron converted into oxides, sulphites, sul- phates, and carbonates; lime, magnesia and the alkalies are being changed into sulphates and carbonates. Geologists have never given us a satisfactory theory to explain the buff burning and refractory clays found in the coal measures. The latest theory is that advanced by Mr. Wilber Stout, of the Ohio Geological Survey, in Yol. XVII, Transactions of the American Ceramic Society. The theory in brief is that coal beds have completely oxidized and the remaining ash hydrated which would give a clay having the composition of coal ash. It would seem as if everything in nature leads up to clay or plays a part in the development of clay. The clayworker’s problem is to take these clays, un- limited in their variations, and burn them into a limited number of wares. He is dealing with materials the fusion temperature of which vary from cone 010 to cone 42. Fluxes and Eutectics. The common conception of a flux is that of an easily fusi- ble chemical or salt which melts at a low temperature and dissolves the refractory materials to be fused, either with or without chemical reactions although chemical combination frequently takes place resulting in the formation of a differ- ent compound or mineral. 10 BURNING CLAY WARES. We have considered the potash, etc., in minerals as the flux, hut the study of solutions has developed results which compel us to modify our original conceptions. It is the min- eral that is the flux and not the mineral content considered separately, moreover, the fluxing effect is due to an intimate mixture of the minerals to be fused. Feldspar, mica, etc., will act as fluxes at the fusing temperature of these minerals, but they will do more than this if finely ground and intimately mixed with the minerals to be fluxed, namely, they will act as fluxes at temperatures below their fusion point. As has been shown by Seger and others, if we mix kaolin and silica we develop without any flux a mixture more fusi- ble than either alone. With successive additions of silica to kaolin the fusion temperature drops from cone 35 to about cone 26, but further additions increase the temperature re- quirement. This minimum fusion point is called the eutectic and the mixture which fuses at the minimum temperature is the eutectic mixture. In this mixture there is no flux as the term commonly im- plies, but the mixture itself acts as a flux. In any mixture of kaolin and silica some part of it will be in the eutectic pro- portion and this will start the fusion. The most fusible mixture in a clay body, regardless of the content of a fluxing salt, such as potash, is the flux, and while this mixture may contain the alkalies it does not nec- essarily contain all the alkalies in the clay mass; indeed, it may contain a very small part of them. The fluxes in a clay body are not potash, soda, lime, iron, etc., but instead they are more or less complex minerals and mixtures of minerals. The bases are a part of these minerals and enter into the mixtures, and, no doubt, aid in developing more fusible eutec- tic mixtures, but as constituents of minerals which do not enter into the fusible mixture they play no part in the fusion. Paving brick which we class as vitrified often come through the kiln heavily scummed and the lime in the scum though an active flux under certain conditions had no part in the fusion which produced the vitrification. In the form of an oxide the lime undoubtedly would have been a part of the fused matrix but as a sulphate it remains free to appear as scum. Mica is often seen in well-burned bricks and its fusible BURNING CLAY WARES. 11 base has not entered into the fused matrix which makes the bond. Potash, soda, and magnesia are found in wall efflorescence and there is ample evidence to show that they come from the clay ware. They are fusible at very low temperatures and we often wonder how they could have escaped the fire influence. It is likely that the heat has affected some of the minerals which did not enter into the fusion and that in consequence they disintegrate under weather influence, thus setting free the base to develop efflorescence, and it is also likely that some of the fused matrix is not stable and sim- ilarly disintegrates. The leaching out of these bases has no appreciable effect on the bond of the ware and if it has none whatever, which remains to be proven, then we must con- clude that the bases in the efflorescence come from minerals which were not a part of the fused matrix. A mineralogicai examination of vitrified ware will undoubtedly reveal a num- ber of minerals, containing alkalies, which have not become a part of the fused matrix. An important factor in fusible mixtures is a fine state of division and intimate association of the minerals. Two clays may be chemically and mineralogically identi- cal yet behave widely different in the burning. In the one all the minerals may be in a fine state of divi- sion and thoroughly mixed and the action of one mineral upon the other in a maximum degree is possible, while in the other, all or any one of the minerals may be in coarse frag- ments which would prevent the close association essential to the development of a mixture fusible at a lower temperature than any of the minerals. Some minerals develop more than one eutectic mixture. The addition of one to the other in increasing amounts lowers the fusion point to a certain minimum. Further additions raises the fusion point for a period, after which continued additions again lowers the fusion point to a second eutectic followed by a second rise. In clay wares we do not have mixtures of simple minerals with which it is possible to determine the eutectic points. If we could plot the eutectic points in terms of temperature and viscosity we would find sudden drops in the viscosity curve as illustrated in the following assumed curve, Figure 1. The temperature would increase perhaps in a straight 12 BURNING CLAY WARES. line. The viscosity with the beginning of fusion would be very high and would decrease with increase of temperature but not in a straight line nor in a uniform curve. As eutectic points “A,” “B” and “C” were approached there would be a rapid drop in the curve, but beyond these points the down- ward tendency of the curve would be slower. Suppose we are burning paving bricks and in one material the soaking heat required to burn the bricks to the bottom of the kiln is approaching a eutectic point, the effect will be a rapid softening of the ware and distortion, in other words a short vitrification range. In another material the final result may be obtained between two eutectic points where the vitri- fication range would be longer, the rate of softening would be slower. Clay wares, particularly the common wares, are not burned to complete fusion. Complete vitrification under any defini- tion applicable to clay wares is simply filling the pores of the clay mass with fused material. Three per cent, has been suggested as the maximum limit of absorption for vitrified wares, but this is too low. Many vitrified wares run five to six per cent, absorption and some as high as ten per cent, and find acceptance in the markets. The fusion in common wares is far from complete and is stopped short of the point where deformation under weight becomes serious, otherwise the ware would be ruined, but in vitrified wares we go so near the limit that the development or non-development of a eutectic mixture at the finishing temeprature may be the BURNING CLAY WARES. 13 factor which determines whether the material is practical or not for such ware. By the way of an illustration, let us consider a single min- eral — lime, for instance — in a clay body. Lime is very basic and an active flux, but potash and soda are even more active. Clayworkers are familiar with the difficulty in burning a limey clay. We have been accustomed to say that lime is not active at a low temperature but at a critical higher tempera- ture it becomes exceedingly active which results in the sud- den failure of the clay mass under fire. Any flux increases in activity with increase in temperature but seemingly at a 10 o .'O Temperature Figure 2. slower rate than lime. We know from the color of the ware that lime has some fluxing action at low kiln temperatures. At some critical higher temperature the fluxing action is so rapid that the temperature range between soft ware and dis- torted ware is too short for commercial operation. It may be simple fluxing action causing the sudden change, but it is likely that a eutectic mixture, equivalent to a sudden advance in temperature, is the cause of the marked decrease in vis- cosity. We illustrate the effect in sketch, Fig. 2, in which the upper curve is the theoretical viscosity of a simple fluxing action under advancing temepratures while the lower curve repre- 14 BURNING CLAY WARES. sents the viscosity of the same flux including the develop- ment of a eutectic mixture. Lime with pure kaolin is known to develop two or three eutectic mixtures while feldspar and magnesia do not develop any eutectic. These tests are, of course, on pure materials and do not necessarily apply to such complex mixtures as we have in clay, but undoubtedly, the eutectic tendency of a mineral has material effect on the burning behavior of a clay mass. From 1800 to 2000 degrees Fahrenheit potassium, mag- nesium and calcium minerals may be acting as fluxes at the same rate. Between 2000 and 2100 degrees the calcium min- eral may develop a eutectic and down goes the ware, while with a magnesium or potach mineral no eutectic is devel- oped and the rate of fusion is a continuation of that at the lower temperature. With one mineral we get a degree of fusion within one hundred degrees, which may require several hundred degrees with either of the other minerals. The bond in a common clay ware is the cementing to- gether of the grains of the clay mass. The grains may be coated with finely divided material either as dust in dry pressed ware, or sludge in suspension coating the grains as the water in mud ware evaporates. This finely divided material is the flux which starts the fusion and which hardens in cooling to a permanent cementing material. The fused material will naturally collect where the angular points of the clay grains are in closest contact. It is possible, even likely, that where the clay grains are in close contact that there will be surface fusion because of the contact of the two surfaces, but it is of no moment whether the fusion is due to fine material coating the grains or to close contact of the grains. In a clay where the initial bond is due to fusion of some portion of the clay mass and the fused material serves as a cement to hold the mass together and to resist the action of weather agencies. As the temperature advances the volume of fused material increases by solution of the more refractory minerals and by the development of other eutectic mixtures, and the pores of the clay mass are filled with the fused material. In the initial bond we have an aggregate of grains cemented together ; next a fused matrix in which are embedded the grains of more refractory material; finally a mass approaching glassiness in its structure. Some clays are made up of quite similar ma- BURNING CLAY WARES. 15 terials, all of which enter into the fusible mixture and de- velop a glassy body quickly. Other clays are made up of widely differing refractory materials and the most refractory grains are slow in dissolving, resulting in a granitoid body. In the fusion process there is not only solution, but also chemical reaction, and in the cooling process other definite minerals are formed in the matrix which may play an im- portant part in the structure and toughness of the bonding mass. The Chief Minerals in Clay. The chief minerals in clay other than kaolinite and often far in excess of the kaolinite, are : quartz, mica, feldspar, iron as protoxide, sesquioxide, sulphate, sulphide and carbonate; lime as carbonate and sulphate; carbon as coal, bitumen, and vegetable matter; magnesia as carbonate and silicate. There is a long list of minerals which might be enum- erated, but the above list includes the minerals which are the most common and which have marked effect upon the burning behavior of clays. Kaolin. Kaolin, when pure, fuses at cone 36 (3362 degrees F.) Note — Cones are ceramic mixtures in convenient form used to denote the fusion point of clay bodies, but they are not ac- curate in the determination of temperatures, and should not be used for this purpose. Cones will fuse at higher or lower temperatures according as the heat is applied quickly or slowly and the low temperature cones are materially influenced by the character of the kiln atmosphere. Since cones are ceramic mixtures they are especially valuable in measuring the tem- perature required by other ceramic mixtures because both are similarly affected by kiln conditions. Kaolinte is the refractory base of clays, although lime and magnesia are more refractory but they occur as impurities and, being strongly alkaline, act as fluxes in conjunction with more acid minerals. Where lime is excessive in amount, exceeding the kaolin base, its action is that of a refractory, but in the ordinary clays it is simply a flux. Bauxite fuses at cone 42 and is much more refractory than kaolin, but it should be considered as a separate mineral, since it does not commonly appear in clays. Where conditions have been favorable for its development it is found under clay 16 BURNING CLAY WARES. beds, in clay beds and above clay beds, and apparently has been derived from the clay, but its occurrence is relatively rare. The property attributed to kaolin which makes it valuable to the clayworker is plasticity, but as a matter of fact pure kaolins are not highly plastic. All that can be said is that we do not know “why is plasticity.” Impure kaolins, such as ball Figure 3. clay, are highly plastic, and still more impure clays, such as gumbo, are even more plastic. Flint clays, which in composi- tion are often pure kaolins, have no plasticity. Some very plastic clays when made into ware and dried develop a very hard body, almost rock-like, while others equally plastic develop relatively tender bodies in drying. The illustration No. 3 shows the most plastic clay we have BURNING CLAY WARES. 17 ever tested. A good plastic clay flowing from a one-inch die will ordinarily break under its own weight when unsupported at about eight inches. The material in question, which was a shale, did not break, but bent down until it touched the ground, thus giving the free end a support. It was then run out, as shown in illustration No. 4, support- Figure 4. ing the free end, and the bar was over five feet long before it broke.* The full length bar is shown on the floor in illustra- tion No. 3 with a three-foot rule back of it. The dried ware from this shale had good strength, but nothing unusual. * Since the above was written we have tested a common limey clay which, under the conditions of Figure 4, exceeded eight feet in length before breaking. 18 BURNING CLAY WARES. Very plastic days, such as ball clays and gumbo, are very difficult to dry without cracking, but the shale illustrated was first class in its drying behavior. Our opinion of plasticity is that it is due to some physical property of the kaolin, probably extreme fineness of grain ; that the grains of kaolin possess high adhesive power and at the same time selective, as we know to be true in Fuller’s earth; that the impurities coating the grains of kaolin serve as a lubricant when moistened, permitting slippage of the grains; that these impurities when the clay is dried serve as the cementing material to bond the clay mass together and the strength of the dried ware is dependent upon the character of the impurities; that the drying behavior of the clay mass is dependent upon the degree to which the pores of the clay mass are filled by the impurities, or by colloidal material. The study and discussion of plasticity is greatly interesting scientists, and it is to be hoped that the result of their work will give us a theory the application of which will enable us to develop or reduce plasticity to any desired degree, and to overcome the troubles which are associated with it. Plasticity is not a burning problem except in so far as it develops faults in the ware which appear in the burned prod- uct and because of the burning. It is the important property which makes clays valuable in the ceramic industries, and we know that plasticity in the clay is due to the kaolin, although pure kaolin is less plastic than many clays. We are not con- cerned whether the plasticity is a property of the kaolin or induced in the kaolin by salts and other materials. Kaolin is a hydrous alumina silicate containing 13.9 per cent, of chemically combined water which comes off in the burning at temperatures between 800 and 900 degrees F. Common clays have chemically combined water in propor- tion to the content of kaolin or other hydrous minerals, and the quantity varies from to 3 per cent, up to 14 or more per cent. Clay takes up a relatively large amount of water, commonly called free water or moisture to distinguish it from the chem- ical or combined water, and this free water is so closely held by the clay that seldom can it be squeezed out in the manu- facturing processes, although frequently the pressure to which the clay is subjected in forming the ware is quite heavy. This free water is removed largely in the drying, but the removal of the last traces of it is an important stage in the burning process. BURNING CLAY WARES. 19 Kaolin is subject to a large reduction in volume as the burning progresses, and likewise clay in proportion to the con- tent of kaolin. Shrinkage is not a property peculiar to kaolin, but it is an important factor in developing a dense ware with- out carrying the burning to an extreme degree of fusion, and it is fortunate that kaolin possesses this property in high degree. Shrinkage is often the cause of considerable loss in the burning in that it decreases the difficulty of keeping the ware in place and also in that provision must be made to relieve the strains developed by it to prevent rupture of the ware, but at the same time it aids in the operation of burning in that it is an excellent measure of the progress of the burn. Quartz. The most common mineral in clay, other than kaolinite, is quartz, or silica sand. Quartz is highly refractory, though less so than kaolinite, and it has an important place in re- fractory products. It fuses to a glass at about cone 25 (2966° F.) but it re- tains its shape to temperatures above cone 30. Its final fusion temperature has not been accurately determined except to the extent that it is completely fused at temperatures below the fusion temperature of kaolin te. The impression seems to prevail among clayw T orkers that quartz, or its equivalent in sand, is more refractory than clay and that any addition of sand will increase the refractoriness of the clay. This is true of ordinary clays which have a low refractory value, but it is not true of pure clays which approximate kaolinite in their composition. On the contrary, silica when added to such pure clays results in a mixture which fuses at a lower temperature than either the clay or the silica alone, as Seger has shown. The addition of silica to a very limey clay — one containing 20 to 30 per cent, of lime, which is difficult to burn because of the high lime content — would serve to dilute the lime and at the same time assist in the development of a fusible mixture. The size of the grain of the quartz will have material in- fluence on the rate of fusion and in a finely divided state will more rapidly develop a fusible mixture with the othei clay minerals. Quartz, as sand, is used to reduce plasticity by dilution, to counteract shrinkage both in drying and in burning, and aids in overcoming lamination, particularly if it is coarse and angular. 20 BURNING CLAY WARES. In some instances fine sand may lower the burning tem- perature required, but as a rule, it does not improve the burn- ing behavior, frequently quite the contrary, while coarse sand often shows a decided improvement. The coarse sand reduces the lamination of the clay ware, increases the pore spaces which are important in the drying, and serves as a mechanical bonding material, thus improving the quality of the ware to be burned, which results in better burned ware. In the burning the large grains may be only partially ab- sorbed into the fused matrix and the undissolved remnants serve as binders and give the ware a granitoid texture, which has the greatest resistance to shocks and to temperature changes, especially in vitrified products. Fine sand has less effect on lamination and does not improve the drying quali- ties, or at least whatever gain there may be in porosity is largely offset by the weaker bond. In burning to vitrification, as the term is understood in common clay wares, the fine sand is largely absorbed into the fused matrix and the texture of the burned product is glassy in character and frequently subject to heavy loss in cooling cracks. Silica under heat undergoes physical changes which min- eralogists recognize as distinct mineral forms. To the clay- worker the feature of importance is expansion. Up to 1600 degrees F. silica decreases in specific gravity and increases in volume about 14 per cent. At higher temperatures it in- creases in specific gravity and probably accompanied by a corresponding decrease in volume. Kiln temperatures generally exceed 1600 degrees F., and in this behavior of silica may be a possible explanation why so many fine-grained silicious clays do not have a clear ringing sound when burned. The silica in the burning first expands materially then per- haps shrinks slightly, and in the cooling expands slightly then shrinks. It is the cooling expansion which we would suspect as the cause of the trouble. If the grains are round, as they likely are in the fine-grained alluvial clays, the expansion cannot be taken up by the pore spaces and there will be a slight rup- ture of the bond in consequence. Naturally, such a product will not have the clear ringing sound of a perfectly bonded product. BURNING CLAY WARES. 21 Mica. Mica is found in nearly all clays and some clays are largely mica, as for instance, some of the micaceous sands of New Jersey, which formerly were put on the market as “Kaolin.” A sample of kaolin recently tested had 80 per cent, of mica and silica, the mica largely predominating. The percentage of kaolin was very small. Mica is a silicate of alumina with some base. There are potash, soda, lithia, iron, lime, and magnesia micas besides micas containing two or more of these bases. The clay will contain the mica which has resisted the weathering influences in greatest degree, and the potash mica — muscovite — is the most common. Muscovite is said to fuse at cone 13, which is above com- mon kiln temperatures. Addition of kaolin even up to twenty per cent, seems to have little effect on the fusibility, in other words, does not readily develop a fusible mixture. This explains why mica is so often seen in burned clay wares. The state of division undoubtedly has considerable effect in producing a fusible mixture, and Stull, Yol. IV, Transactions American Ceramic Society, has shown that finely ground mica exerts a fluxing action below cone 4 and that alone it vitrifies to a non-absorbent body below cone 4. Reike, a German inves- tigator, on the other hand, found that mixtures of mica and kaolin in percentages of twenty per cent, mica to eighty per cent, kaolin only reduced the fusion point of the kaolin from cone 35 to cone 34. Mixtures containing as high as forty per cent, of mica only lowered the fusion point to cone 32. It seems inconsistent that a mineral that fuses at cone 13 should, when mixed in large amounts with a clay fusing at cone 35, have less effect than the mixture of two highly refractory bodies, such as kaolin and silica, but the explanation is that the mineral developed bears no relation in its fusing point to the fusing points of the original minerals. This explains why the micaceous sands of New Jersey, mixed with kaolin, pro- duce excellent fire bricks. Stull’s work shows vitrification at ordinary kiln temperatures, but there may be and evidently is a long range between vitrification and fusion. Mica, then, in a fine state of division aids in vitrification at ordinary kiln temperatures but requires a higher tempera- ture when present in plates visible to the naked eye. The point is that a fine state of division permits an inti- mate mixture and this is essential in developing a fusible 22 BURNING CLAY WARES. body. Lumps of kaolin and quartz in a crucible in the pro- portions of the most fusible mixture will not fuse at the tem- perature required for that mixture but if finely ground and intimately mixed a eutectic mixture is developed. Mica cannot be considered as having any serious effect in the burning behavior but under some conditions it will im- prove the burning qualities. The behavior of mica indicates how variable may be the problems in burning such complex material as clay. Lime Lime occurs in clays in several mineral forms — as a car- bonate (limestone), as a carbonate with magnesia (dolomite), as a sulphate (gypsum), as a constituent of silicate minerals — and it is a serious menace as in clay burning operations. Lime carbonate as pebbles burns to caustic lime, which, when exposed to the weather, slakes, swells and ruptures the product. In large pieces deeply embedded in the ware, the rupture of the ware through hydration of the lime is com- plete, but small pieces cannot exert sufficient pressure to cause rupture except when near the surface, and in such instances circular discs are flaked off. This surface effect of lime is commonly termed ‘‘popping.” The dissociation temperature of lime carbonate is about cone 014 (1526° F.), which is lower than commercial kiln temperatures, and in consequence any lime carbonate in the clay is certain to be converted into lime, and “popping” follows. Lime sulphate is present in clays as gypsum, and it is also developed by the oxidation of sulphide minerals result- ing in sulphuric acid which reacts with the lime carbonate to form the sulphate. The dirty white coating which comes to the surface of clay wares in the drying is largely lime sulphate. Lime sulphate dissociates at higher temperatures than the carbonate, especially under oxidizing kiln conditions, and in consequence in many kiln operations passes through the kiln unchanged. The proof of this is the “scum” on many of our burned wares. Clays which “pop” when burned in periodic kilns may not do so when burned in continuous kilns, especially those which have no advanced heating or water-smoking flue, in which the water smoking is done with gases ladened with sulphur. Under such conditions the lime carbonate is converted into BURNING CLAY WARES. 23 sulphate, which, as above mentioned, does not dissociate at low temperatures, and at higher temperatures any lime set free enters into and becomes a part of a fusible mixture and is permanently locked up in some silicate formation. Lime in a finely divided state or as a constituent of a silicate mineral or any mineral which enters into a fusible mixture is a serious element in clay bodies, because of the fusibility of the mixture. It has been shown by investigators of fusible mixtures that lime with kaolin develops two or three eutectic mixtures defining the term in the sense that it is the most fusible mixture of any mixture of the same min- erals in proportions approximating those of the eutectic mix- ture, which is our interpretation of eutectic. The effect of these eutectic mixtures, or if you please, con- sider it simply from the standpoint of a simple flux becoming active at a critical temperature, is a rapid development of vitrification — a decrease in viscosity. In any commercial kiln there is a difference of tempera- ture, and a clay to be of commercial value must produce marketable ware within such differences of temperature. In up-draft kilns we must hold the temperature in the bottom at a maximum until the heat works to the top, but we never attain the same temperature in the top as that in the bottom. Similarly in down-draft kilns, we must hold the heat in the top until we can get the ware burned to the bottom. We consider three cones the minimum limit of variation of tem- peratures in down-draft kilns, and no other type of kiln has such a small limit. This statement requires some explanation. It is possible to attain the same temperature in the top and bottom of a down-draft kiln, but at considerable expense in fuel and time, which the majority of wares will not stand. Up-and-down draft kilns develop more uniform temperatures, but their application is limited. If the clay, to produce marketable ware, will only stand a range of two cones, or one cone, we must sacrifice the ware in one part of the kiln to get good ware in another part. This is the weakness of limey clays. The lime develops mixtures which fuse so rapidly that in many instances it is impossible to get a properly bonded ware throughout the kiln. Under a subsequent discussion of the burning behavior of clays will be found curves, in which are shown the burning ranges of a number of clays, including limey clays. 24 BURNING CLAY WARES. One cannot determine how serious the lime trouble may be from a casual examination of clays. We have seen shales interstratified with limestone in a prohibitive degree, provided the limestone developed the usual effect, but which in use did not “pop,” and the burning range was satisfactory for com- mercial operation. Our conclusion was, that the limestone was impure, probably earthy, and that the granules of this material burned to a lime-alumina-silicate, in itself quite in- fusible and which would not slake. The segregation of the lime prevented the formation of a fusible mixture of it with the other clay ingredients. Lime in small percentages in clay shows little effect, at least in practical operations. As the lime content increases we experience the serious fluxing difficulty, but when it be- comes excessive it acts as a refractory and carries the burn- ing over the fluxing range ; or, to be consistent, a low tem- perature fusible mixture is impossible with such excess of lime. A marked feature of lime is the color produced. A clay containing sufficient iron to burn red will, when impregnated with lime, burn to a buff, changing at higher temperatures to a yellow-green and finally to a decided green. At very low temperatures, before the lime enters into the fusible mixture in any effective degree, the color of the ware is red. The red color is due to the iron, and disappears when the iron begins to combine with the lime and silica as a lime-iron-silicate, while the final green color is due to a full development of the lime iron body. The color and its mutations are characteristic of a limey clay. A limey clay seldom produces a pleasing face building color. While red, buff and green are satisfactory building colors, limey clays are not satisfactory, in that we cannot produce any uniformity of color, and the result is motely effect in the wall. Particularly is this true in continuous kilns where the product will be red streaked with buff, and vice versa. There is a large excess of air in a continuous kiln, and the kiln atmosphere is as nearly oxidizing all the time as is possible in any kiln. Ferric oxide, which is red and gives the color to red wares, does not enter into chemical combination in the ferric state, at least not with lime, and the effect of the kiln atmosphere is to maintain this ferric condition. It is a contest between the oxidizing kiln atmosphere and the dissociating influence of a fusible mixture. BURNING CLAY WARES. 25 As the temperature advances the oxidizing conditions are weakened, and the effect of the fusible mixture strengthened; but in many instances the fusion does not go far enough to fully include the iron, and in consequence the streakiness of the product is more marked in a continuous kiln. It is pos- sible to overcome this effect by dampering the kiln and thus produce a reducing condition in the kiln atmosphere, which reduces the iron and puts it in condition to combine readily with the lime and silica. The usual effect of a limey clay product in a wall is that of a ware that has been coated with mortar and then cleaned with a scraper or skutch. There are exceptions, however, and we have seen excellent buff and green wares sorted from limey clay products ; nor must the pale red products be overlooked. In the latter the lime mixture absorbs only part of the iron, or possibly some mixture other than lime develops the bond at a lower temperature than that of the lime eutectic, and the iron only partially enters into this other mixture. Clay is too complex to draw any conclusions, except in regard to very pronounced phenomena. Gypsum (lime sulphate), as has been noted, is of frequent occurrence in clays and also that it dissociates at a higher temperature than lime carbonate. The dissociation begins below 1800° F., but becomes very rapid about this temperature and up to 2000° F. Limey clay products in the burning frequently begin to fail at about cone 02 (2030° F.), and in consequence the burn- ing is done at a lower temperature than this. We have seen in limey clay products crackled surfaces, especially where the ware has been exposed to the flame around the bag. This may be due to increased shrinkage under ex- posure to flame temperature, but, the ware not being de- formed, the exposure to such flame temperature must have been only for short periods, however many times it may have been repeated. It is not unlikely that the scum on the surface of the ware, being reduced as it usually is around the bags of a kiln, introduces -additional flux into the surface layers, caus- ing largely increased shrinkage, and that the crackling is due to this rather than the general effect of the fusible mixture in the body of the ware. 26 BURNING CLAY WARES. Feldspar. Feldspar has been noted as one of the common minerals in clay. The potash feldspar — orthoclase — is probably most commonly found in clays first, because it is more abundant in granite rocks ; and, second, perhaps because it may resist weathering influences in greater degree than the other feld- spars. Feldspar enters largely in ceramic bodies as a fluxing mineral, and it is a desirable mineral in any clay body. With pure clay it does not develop any eutectic mixture and its effect as a flux is more or less proportional to the advance in temperature. The fusion temperature of commercial potash feldspar is listed by different authorties from cone 4 to cone 9 but in ceramic bodies it undoubtedly develops mixtures fusible at lower temperatures. Cone 4 to cone 9, however, are not un- usual temperatures in commercial kilns, even on common wares, and felspar must be ranked as an important mineral in clays in the development of a permanent bond and in vitrification. Fortunately, it is a safe material for this pur- pose. Carbon. Carbon occurs in clays as vegetable matter, such at root- lets, mould, etc. ; as bituminous matter high in volatile gases, such as lignite, bituminous coal, oil and oil residues ; as non- volatile carbon, such as graphite. It also occurs as a mineral constituent as in carbonates. Carbon plays an important part in the burning process, and it is often the cause of considerable trouble, though in limited amounts it may be beneficial. (1) To whatever extent it is present in the clay it may assist in the burning, and indeed it is frequently added to the clay for this very purpose. (2) It retards the oxidation of iron, and because of this factor we often get into serious trouble. (3) It produces a porous product. The Hudson River district and Chicago are noted examples of the use of carbon for no other purpose than to assist in the burning. In one locality coal dust is added to the extent of a little over one per cent, of the clay, and the double coaled product which is for casing, etc., has about thirteen per cent, of coal. Sawdust is frequently used, but more especially to increase the porosity of the product. Coke, anthracite and semi- BURNING CLAY WARES. 27 anthracite are the usual types of coal added to clays, and they the generally safe, while bituminous coal, because of the volatile gas, the rapid combustion, and the relative high tem- perature developed are likely to cause bloating and black cor- ing. Lignite is used in a number of localities and has proven to be excellent material in spite of the fact that it is highly gaseous. These materials, under proper control, not only aid in the burning, but they serve as “grog” to reduce lamination. The serious difficulty of carbon in clay is its retardation of the oxidation of other minerals. If is a reducing agent, and until it has been burned out there can be no oxidation of the other minerals. If the combustion of the carbon is rapid and it develops temperatures sufficient to fuse the clay in contact with the carbon, bloating and black coring result. The carbon can only burn when supplied with air, and the fusion of the clay mass produces an impervious body, thus shutting off the air supply. The entrapped intensely heated carbon will take oxy- gen from any minerals in the clay mass containing oxygen and the gas thus developed being unable to escape forms bubbles of blebs, increasing in size because of the expansive force of the gas under advancing temperatures. In burning carbonaceous clays it is necessary to remove the carbon at low temperatures, which means at a slow rate, since the carbon itself may develop a fusing temperature in the clay mass. There are many carbonaceous clays which cannot be burned in the ordinary manner. The black shales in Ohio and the black cretaceous clays in New Jersey are examples of clays difficult to burn because of their carbon content. As soon as the carbon in the clay becomes ignited it burns at such a rate as to produce fusion and the result is the same as if the clay had been badly overburned in the ordinary way. In burning such clays the first step is to heat up the mass by furnace fires until the carbon in the clay is ignited, then the fires are drawn or allowed to die out, the furnaces are closed up and daubed, thus shutting out the air supply, except that which may leak in through the walls, and because of the reduced air supply the combustion of the carbon is very slow and high tem- peratures do not develop in consequence. After the carbon is thus burned out the furnaces are again put into use and the kiln burned off in the usual manner. The bloating is due entirely to the gas in a fused mass, and the black core is partly due to unconsumed carbon and 28 BURNING CLAY WARES. partly to reduced iron in combination with silica, forming a black iron silicate. The black color of a clay is not proof that the clay will be unsafe in its burning behavior. The clay may be refractory and resist the fusing tendency of the carbon, or the carbon may be of such character or in such condition that it- will not develop a fusing heat. All that can be said is that a black clay must be re- garded with suspicion until it has been proven to be safe burning. Iron Minerals. Iron is a very common mineral in clay and also in many products very important. It occurs in a number of common mineral forms, such as oxides, carbonate, sulphide, sulphate, and as a constituent of silicate minerals. The oxides are of first importance. Clays and shales may be red or yellow near the surface, changing to blue in depth, and these colors are largely due to iron, particularly the red and yellow. The red ferric oxide — hematite — is the most per- manent form and the most common, and to it we owe the red color of our ceramic products. The yellow is a hydrous ferric oxide — limonite — which burns to red. The blue may be ferrous oxide, but the presence of some carbon would give the blue color in greater degree than iron. There are some very red clays in which the red color is said to be due to an algae and the raw clays are a deeper red than the burned product. Such clays are not common, how- ever, and the ordinary red clays burn to a much deeper red than the raw clay, except they also contain lime. The depth of the red color in the burned product in any case is due, first to the amount of ferric oxide present or produced in the burn- ing, second to the degree of fineness and dissemination through the clay mass. If we soak a mass of white burning clay with a strong solu- tion of ferrous sulphate, dry and burn it, we will get a brilliant deep red color; if we mix the clay with powdered hematite ore, introducing the same amount of iron ore as before, we will get a brownish red color, but far less deep and less brilliant than in the first instance. If we use iron scale, iron sulphide, or metallic iron in granular form, the amount of iron still the same, we will have a buff product speckled with black spots. The sulphate of iron is readily soluble, and in solution it penetrates to all parts of the clay mass, and when the water is driven off each grain of clay is coated with a film of iron BURNING CLAY WARES. 29 sulphate which dissociates at a low temperature to ferric oxide and thus a maximum color effect is obtained from a minimum quantity of iron. The powdered ore and clay are merely a mixture and may be likened to a “pepper and salt” effect — at least to red pepper and salt, and the coloring effect of the iron is greatly reduced. Larger grains of iron ore or iron will produce red or black spots, but the natural color of the clay predominates. The thin film of sulphate readily oxidizes to the ferric state and the surface of a grain of iron also will oxidize, but the oxidized outer face protects the inner core from oxidation and the result is a more or less brown color, due to the close association of the black ferrous and red ferric oxides. Sulphides of iron burn black because the sulphur prevents the oxidation of the residual ferrous oxide, and the latter easily combines with silica to form a black silicate. Iron sulphide — pyrite, marcasite and pyrrhotite — introduces a troublesome problem in burning clay wares, and hardly a clay is entirely free from this mineral. It occurs most frequently as a concretion varying in size from granules, invisible to the naked eye, to lumps an inch or more in length. In this granular form it has little effect in producing color, even though in the burning process it becomes ferric oxide. Ware will often come from the kiln with a flaked or “popped” surface, and in the center at the bottom of each disc will be a red, brown, or black mineral grain. This mineral grain is the residual of pyrite or iron carbonate granule. There is a reduction in volume of these minerals in the dissociation stages, but the shrinkage of the clay mass closes up the space. In the subsequent oxidation of the iron mineral the increase in volume in consequence of oxidation often ex- erts enough pressure to flake off the surface of the ware in circular discs within the radius of influence of the pressure, similar to the behavior of lime pebbles. The sulphur in pyrite begins to pass off at very low tem- peratures, but the evolution does not become rapid until a temperature of about 700 deg. F. is reached, and above this temperature the rate of expulsion increases with the tempera- ture up to about 1300 deg. F., but we cannot completely drive off the sulphur at this temperature. The “blue smoke” from a kiln stack is evidence of the sulphur coming off, and we have seen operations where the temperatures were advanced by stages holding the temperature constant at each stage until the “blue smoke” became very light, but which again became heavy with an advance in the temperature. 30 BURNING CLAY WARES. The sulphur comes off at all temperatures and there are some dissociation temperatures where the expulsion is par- ticularly heavy, but we know that in many instances some sulphur remains in the ware up to the highest kiln tempera- tures. In this dissociation behavior of iron sulphide lies one of our serious burning problems. After the ware reaches the fusion point, the sulphur gas still coming off but unable to escape, bloats the ware. If the grains of pyrite are near the surface the bloating effect will develop large blisters on the surface of the ware. The effect of pyrite in clay has been admirably shown by H. B. Hender- son in a paper on “Clay Testing,” read before the National Brick Manufacturers’ Association at the annual meeting, Feb- ruary, 1916, and published in The Clay-Worker in the follow- in March. We have mentioned the bloating on account of carbon gas, and sulphur gas, or any gas, will have the same effect under similar conditions. The sulphur blebs are likely to be larger than the carbon blebs because the pyrite granules are often larger than the carbon grains and a larger volume of gas is developed in the immediate vicinity of the granule. Bloating, or vesicular structure, occurs in many clays when the temperature in burning is carried too high, and while it is due to a gas formation it is not always sulphur or carbon gases. In fact, very few clays fuse quietly to a dense glassy body. In the fusion process as the different minerals enter the fusion mixture there is a chemical change taking place and frequently a gas given off in consequence. For example, ferric oxide — Fe 2 0 2 — enters into combination with other min- erals only after it is reduced to ferrous oxide — FeO — and in the reduction, or in the reaction or fusion which combines the ferrous oxide and silica, one molecule of oxygen is given off. Two things are happening at the same time, namely ; a mineral which does not readily enter into a fusible mixture is being converted to one which does act as a flux, thus pro- ducing or aiding fusion, and a gas is being given off which entrapped in the fused mass causes the vesicular structure. The fusion of minerals which in fusing do not give off any gas would be quiet and there would be no bleb structure. Such minerals would tend to have a very long, safe vitrifica- tion range within the clayworker’s definition of the term, while on the other hand clays that develop a large volume of gas in fusing and at the same time greatly increase the BURNING CLAY WARES. 31 fusibility of the fusible mixture by the addition of other min- erals would have a very short vitrification range. There are all degrees in between these two extremes. As previously stated, the red color in our clay products is due to the ferric iron present and the intensity of the color is determined by the state of division of the mineral. If we start with a clay containing ferrous and ferric oxides, or only ferrous oxide, the first step is the oxidation of the ferrous oxide to the ferric. This begins at a very low temperature and the rate of oxidation increases rapidly up to a temperature of about 1300 deg. F., provided there is no carbon present. The carbon, as has been stated, prevents the oxidation of the iron and must first be burned out. As oxidation proceeds the amount of ferric oxide increases as the ferrous oxide decreases, and in consequence the depth of the red color increases correspondingly. We will have the red color so long as the ferric oxide can be maintained in a free state. At high temperatures, even though there is an excess of air in the combustion gases, oxidation progresses very slowly, practically ceases, indeed, some reduction takes place. The incomplete combustion gases satisfy themselevs with the oxy- gen from the minerals in the clay mass given off in the process of fusion, in preference to the oxygen in the air accompanying the gases. At high temperatures then the ferric oxide is being reduced to ferrous, which combines with silica to form a black iron silicate. The color of the ware changes from red, to red-brown, to brown, to dark brown, to black. In the later stages of the fusion undoubtedly the iron in the silicate min- erals is having effect in darkening the color. There are some results which are not readily explainable. In some clays the red color continues even to complete vitrification ; in others the red color changes to a brown almost with the beginning of vitrification and deepens to a black as vitrification advances. If we color a clay with a solution of iron — sulphate for instance — a red color will develop and continue to a high temperature, but if we use powdered hematite the brown color will appear at a relatively low temperature and darken to black at higher temperatures. In other words, if the iron is in a chemical state of divi- sion the color is red, but if in a mechanical division the result becomes black. We do not know whether iron can go into combination 52 BURNING CLAY WARES. and retain its ferric form ; if so, this would explain the red color under conditions which ordinarily develop, brown or black. In one instance the ferric oxide may simply enter into solution in the fused mass without losing its identity, and in the other instance it loses its identity by combination. It may be that in the chemical state of division the iron coat- ing the grains of the clay mass is in too small a quantity relative to the mass to develop a fusible mixture, or one that is permanent, and in cooling the iron oxidizes to the red color, while in the mechanical state of division such a volume of the black ferrous silicate is formed that the red color is fully mantled. Whatever the explanation, some clays will remain red to complete vitrification while others will not. Magnesia. Magnesia occurs in clays as a carbonate (magnesite), and associated with lime (dolomite). It is a common constituent of many silicate minerals, and is nearly always present in clays in small amounts. It has been claimed that it is an excellent flux in that it acts slowly and lengthens the vitrification range, reduces the tendency to warpage, and produces a tougher product. Its addition to paving brick materials has been recommended. Our experience with it has not been encouraging. We have found that, when present in quantity such as would with lime give a very short vitrification range, its behavior is very simi- lar to that of lime. It is better than lime in that at just the right temperature an excellent paving block is obtained, but the vitrification range was too short to be practical and the distortion was excessive. Our work is, however, not conclusive. Parmelee and Bleininger (Yol. XVI. Trans. Am. Cer. Society) state that in porcelain bodies and slags magnesia in mixtures has a longer fusion range than lime and develops a tougher body subject to less distortion. All tests show the beneficial effect of magnesia compared with lime in the production of slag bodies, but in vitrified common wares we do not carry the fusion to such a degree, and it remains to be proven whether the addition of magnesia to a vitrifying material containing neither magnesia nor lime in appreciable quantities will lengthen the vitrifying range and toughen the resulting product. BURNING CLAY WARES. 33 CHAPTER II. THE BURNING PROCESS. T HE BURNING PROCESS may be divided into a number of stages as follows: (1) Drying, commonly called watersmoking. C2) Oxidation and dehydration. (3) Bonding or shrinkage. (4) Vitrification. (5) Annealing and cooling. Watersmoking. Few clay wares, particularly the common wares, are fully dry when set in the kiln. Hygroscopic water is not driven off at atmospheric temperature or even at the boiling point of water ; in fact, wares from dryers having temperatures as high as 300 degrees F. seem to contain some moisture, as shown by the operation of the kiln. Generally the wares are far from dry through imperfect dryer operation, and some wares, such as dry pressed bricks, are placed direct from the machine into the kiln wherein the drying is accomplished. The watersmoking is slow or rapid as the ware will permit. Dry-pressed wares require from five to thirty days slow dry- ing in the kiln, from five to eight days being the customary period. Mud products, which have been previously dried, can be watersmoked quickly in from twelve to seventy-two hours. The watersmoking is accomplished by low fires in the kiln furnaces. Wood is frequently used for this purpose, to avoid scumming and sooting. Scumming is caused by the combination of sulphur gases, moisture and lime or other minerals in the clay. The clay may not contain lime in sufficient quantity, in which event there will be no scumming, and watersmoking may be done with a sulphurous fuel, or it may be that scumming is less serious than the cost of a special fuel to prevent it. Scumming 34 BURNING CLAY WARES. is largely a dryer trouble, and the development of a little more in the kiln may be of no consequence. The combination of cold ware, moisture and a smoky gas deposits soot which often fills the draft spaces among the ware and stops the draft. This is a common occurrence in watersmoking with bituminous coal. Some factories watersmoke with anthracite, coke, or smoke- less coal and finish the burn with bituminous coal, but many factories find it possible to use bituminous coal from start to finish. In the watersmoking period we desire to heat up the ware, evaporate and remove the moisture, and, to accomplish this, particularly the removal of the moisture, it is essential that the draft be strong. Unfortunately, it is a period of weak draft, because the kiln, stack and ware are cold and the fire is low. There is an advantage in having two or four periodic kilns connected with a single stack having an individual flue for each kiln. Under such an arrangement the chances of having a hot stack to start the watersmoking are more favorable. In some instances where the stack is a single one, it is provided with a small furnace in the base to heat up the stack and the gases contained therein. A steam jet also may be used for the same purpose. Oxidization. As soon as the watersmoking is completed, and, in fact, in many products during the watersmoking, the temperature is advanced to that required in oxidation. The several periods in the burning process are not dis- tinctly separate but always overlap more or less. Oxidation begins in the later stages of the watersmoking and continues into the shrinkage stage, but the greater part of the oxidation occurs at a low red heat — 800 degrees F. to 1300 degrees F. The oxidation includes the oxidation of the carbon, the decarbonization of the carbonates, the desulphuri- zation of the sulphides, the oxidation of the ferrous oxide and the dehydration of the hydrated minerals, particularly the kaolinite. Sulphur begins to come off from pyrite at low tempera- tures, but the expulsion does not become rapid until a tem- perature of 700 degrees F. is reached, which is below a visible red heat. The greater part of the sulphur comes off between 700 degrees and 1300 degrees. BURNING CLAY WARES. 35 Carbon ignites at about 850 degrees F. and the oxidation continues up to about 2000 degrees F. It is, of course, im- portant that the carbon be oxidized at low temperatures ; first, because the oxidation is more rapid at low temperatures, and second, because many clays fuse at lower temperatures than 2000 degrees, and bloating would result if the carbon were not expelled before fusion begins. It is also important that the ferrous oxides, both those originally in the clay and those de- veloped from the sulphides and carbonates, be oxidized before a fusing temeprature is reached, and this oxidation cannot take place until the carbon is removed. The carbonates dissociate between 700 degrees F. and 1600 degrees F. (iron carbonate, 732 degrees F. ; magnesium car- bonate, 1380 degrees F., and calcium carbonate, 1526 degrees F.). Even the maximum dissociation temperature is 300 to 400 degrees below the finishing temperature of an average low burning temperature ware. It is our experience that dehydration of the clay base — the expulsion of the chemically combined water — takes place be- tween 800 degrees F. and 900 degrees F., but several authori- ties claim that dehydration begins at a much lower tempera- ture. There are other hydrous minerals besides kaolin in a clay mass, and undoubtedly some of these dehydrate at much lower temperatures than kaolin. We know that a temperature of several hundred degrees is necessary to remove completely the hygroscopic water, but the water may contain some soluble salts in solution or may easily be acidulated with sulphuric acid, which would require higher temperatures to remove the water content and this hygroscopic water may be mistaken for combined water. There is a marked decrease in the weight of the mass at about 850 degrees F., and this we attribute to the dissociation of the kaolin base. Combined water and sulphur are both coming off rapidly at about the same temperature, and this is the “blue smoke” period of the burning, although the “blue smoke,” which is due to the sulphur, continues long after the combined water has been driven off. Nine hundred degrees is a very low red heat, barely visible through the kiln peep- hole. Shrinkage Period The shrinkage or bonding period covers the fusion from the start, up to that required for a thoroughly bonded ware, and it marks the burning range of the clay. We distinguish between burning range and vitrification range in the same 36 BURNING CLAY WARES. sense that we distinguish between ordinary ware and vitri- fied ware. The average common clay begins to shrink in a measurable degree at about cone 010 (1742 degrees F.), but seldom is the fusion sufliciently advanced to develop a permanent bond until cone 07 (1850 degrees F.) is down, and more frequently not until cone 04 is turned. The maximum temperatures are rather a question of fuel cost than the limit of the clay. We have seen common clay products which required a tempera- ture up to cone 10. Shale products range from cone 04 to cone 5; No. 2 fire clay products from cone 1 to cone 9 ; No. 1 fire clay from cone 5 to cone 14 ; silica and magnesite bricks, cone 20. The tem- peratures are totals and include the vitrification range in vitrified products. The vitrification period is simply an extension of the bond- ing period to a more complete fusion. Vitrification. The term vitrification is used very loosely in the ceramic industries. Sewer pipe and paving brick bodies are vitrified, yet an examination of the body in the majority of instances shows a large percentage of mineral grains which have not entered into the fusion. These vitrified bodies will absorb water up to 10 per cent, in some instances, and 3 per cent, to 5 per cent, absorption is of common occurrence. Porcelains are also vitrified, but there is a big gap between porcelain and vitrified bricks. The only reason for assigning a burning period to vitrifica- tion is that there is a definite product classed as vitrified, such as paving bricks, sewer pipe and electrical conduits, and there is a limitation in the clays which will produce such products. There is no change in the burning process — we merely ad- vance the heat to the vitrifying point and hold it until all the ware in the kiln is properly burned. It would be more proper to say heat soaking period, since the production of all hard- burned ware or vitrified ware requires a period of heat soak- ing in order to get some degree of uniformity throughout the kiln. We may burn common bricks and impervious face bricks to the same temperature — let us say cone 1 — but when this cone is reached the common brick kiln is held for a very BURNING CLAY WARES. 37 short period thereafter, while the face brick kiln is held for a longer period. In the common brick kiln we may have a variation of seven to eight cones between the bottom and top of the kiln, and yet have a satisfactory ware throughout the kiln. In the face brick kiln we must hold the heat until the difference in tem- perature is reduced to three or at most four cones to get a satisfactory product. The only difference in vitrified products is that the finishing heat is nearer the failure point of the material, and in consequence requires greater care. In this connection, we wish to make a note in regard to fuel consumption. We often get information that certain products are burned to cone 1, for example, with 300 pounds of coal per ton, while someone else is doing the same work with 200 pounds. Why? Assume that the coal is the same, the kiln the same, and the burners equally efficient ; the difference may be en- tirely due to a soaking heat, given by one and not by the other. There are differences in the clay, in the coal, in the kilns, both type and size, in the draft, in the efficiency of the burners, and in the results to be obtained from the kiln, al- though cone 1 may be the temperature in both instances. To make any comparison, all this data must be given and taken into consideration. Flashing. Flashing might properly be included as a burning period, since in many instances it is the final operation in the burning. The purpose of flashing is to produce a color effect different from that which naturally results from ordinary burning. A flashed color on red burning wares is brown to gun metal black ; on buff burning clays a golden yellow to brown. Flashing is accomplished by closing the fires toward the finish of the burn, thus shutting out the secondary air, re- sulting in a strongly reducing kiln atmosphere. The common method is to begin flashing twelve to twenty- four hours before the finish of the burning. The fires are closed for a period of two to six hours, followed by a longer or equal period of clear fires, and the alternations are repeated until the end of the burn. In some operations the reducing conditions are started earlier in the burning and it is found that a sufficient depth of flash can be obtained in this way 38 BURNING CLAY WARES. without carrying the temperatures to such a high degree, but the results in this method of firing are less brilliant and in consequence do not find the same favor on the market. Flashing in all its variations is not fully understood. The reducing action of the kiln gases will keep the iron minerals in the ferrous oxide state or convert them into ferrous oxide and, as has been shown, the iron in this state readily enters into silicate combination, producing a brown to black ferrous silicate. This will explain the brown to gun metal black colors from red burning clays, but does not explain the golden yellow colors which develop on the face of buff burning clays. The iron content in many buff burning clays is largely seg- regated in grains and the ordinary reducing effect is evident in the development of the black spots throughout the body of the ware, but the golden yellow color is a surface effect and does not penetrate the ware. Moreover, only the faces that have been exposed to the flame or to the moving combustion gases have the “fire flashed” color. Even in the red burning wares the color on the faces exposed to the flame is more brilliant than those not so exposed. This greater brilliancy may be due to a greater degree of fusion but we believe that the contact of the combustion gases with the surface has had effect just as in the buff burning ware. It is said that flashed colors are due to some chemical change in the iron content in the clay, likely ferrous silicate, which would occur under reducing conditions, and that in the cooling there is a superficial oxidation of the iron which would give the golden color. It has been our opinion that the surface flash is in some degree due to a surface accumulation from the combustion gases. We have seen masses of iron or iron slag a half inch thick built up on the surface of bricks which have been exposed to a flame jet through a crack in the bag wall, and crusts on the crowns of high temperature kilns are familiar to many clay workers. The faces of heavily flashed bricks have a roughness sensi- ble to touch which is not apparent on the backs. Where the flame passes through a small aperture, as the checker in closely set brick ware, the flash on the under brick is fan- shaped, just as the gas would naturally flare out after leaving such an aperture and the flash is the mark of the moving gas. Flashing is evidently a flame phenomenon, and is pre* BURNING CLAY WARES. 39 sumably due to iron. At least we may say that a clay which does not flash readily will flash if magnetic iron ore is added to it. Clays containing manganese also flash in greater de- gree than the buff burning clay to which the manganese is added, but as the manganese is impure and contains iron, the flashing may be due to the iron content rather than to the manganese. Whether the flash is due to a chemical change in the iron content in the clay, to a volatization of the grains of iron in the clay or to a surface accumulation from the gases is not important except the determination of the question might lead to more intelligent development of flashed effects. The flashing is more intense the nearer the ware is to the fire, but this gives no light on the problem because temperature and reduction are factors in producing the flash, and these prevail in higher degree near the fire. If the flashed ware is cooled very quickly, the flashed color does not appear, and the flash is more pronounced ag the cool- ing is slower, within reasonable limits. This is merely a question of oxidation. All kiln burners are familiar with the difference in color of the draw tests and the properly cooled ware. The browns cool to reds, the blue, gray or green to buff. This difference in color is more marked in flashed ware, perhaps, because of a finer state of division of the mineral which produces the flash and a greater susceptibility to oxidation of the fine ma- terial. It is also known that the flash on a fire clay product can be burned off, in part at least, by reburning the ware under oxidizing conditions. The process of manufacture has considerable effect on the flashed color and only rarely can the same color be produced on the same clay ware made up by different processes. Many dry-pressed bricks have a beautiful golden flash, while the stiff-mud bricks from the same clay lack the golden color and often are dull brown. There are, however, good flashed stiff- mud bricks, and the difference is only one of degree in the effect on the same clay made by different processes. The dry-pressed flashed bricks have retained a strong hold in the market partly because of the fine color and partly because they are more impervious than other dry-pressed bricks, but the use of the plain red, buff and gray dry-pressed bricks has been waning. 40 BURNING CLAY WARES. Henderson in Yol. 1, No. 3, Jour. American Ceramic So- ciety, has shown that flashing on fire clay bodies and in salt glazing is due to an amber-colored hexagonal crystal in the de- velopment of which carbon has played an important role, although the crystal is not graphite. This explains several phenomena. The essential smoky flame supplies the carbon which the surface absorbs. The crystals develop in cooling, and hot ware or ware chilled suddenly will not show the flashed color. If we remove the carbon by subsequent oxida- tion the crystals do not recur in the following cooling stage. Cooling. Cooling is not a stage in the burning, but in many wares proper cooling in order to anneal the ware is an important factor in the production of sound tough ware. When the firing ceases, particularly in vitrified wares, the mass is ra a state of semi-fusion, and the rate of cooling has material effect on the physical structure. Minerals are forming and crystals developing which serve to lace the mass together and give it greater resistance to shocks. In some measure we may compare the fused mass with glass which, as is well known, must be annealed to relieve the strains and give it the toughness without which it would be of little practical value. We have held that the annealing is accomplished at a low red heat and that we may safely cool rapidly from the finishing temperature down to a red heat when the rate of cooling should be slower in order to properly anneal the ware. Kiln burners are familiar with the snapping and cracking that can be heard in the kiln as the cooling takes place and these evi- dences of the relief of cooling strains are only heard in the later stages of the cooling process. Wares that are subject to cooling cracks are particularly benefited by thorough annealing, and such wares should be entirely cooled by conduction and radiation and not by con- vection. The kilns should be closed and daubed and the dampers closed during the annealing stage. Cooling cracks are easily recognized. They never open up and a casual ob- servation does not reveal them. Ringing two bricks together will show whether the ware is cracked or not, and if cracked, an examination will reveal a fine hair-like crack, sometimes nearly through the ware. In fact, the ware may be entirely cracked through and when picked up falls into two pieces, yet the crack is not seen until some movement has caused a BURNING CLAY WARES. 41 separation of the two pieces. The top courses in many paving brick products are rejected because they are too brittle, although they are the hardest bricks in the kiln, and except for the brittleness should stand the best test. We frequently find cooling cracks in such bricks, but they are difficult to see, as the bricks come from the kiln. After exposure to the weather, when the cracks have been developed by the infiltra- tion of dirty water, they are very evident. Some wares, including vitrified products, are apparently not improved by slow cooling, and the cooling fans can be connected, or the wickets opened and the dampers raised as soon as the burning is finished and the cooling from begin- ning to end carried out as rapidly as possible. Theoretically any ware will be improved by slow cooling, but in view of the practical evidence it would be foolish to urge slow cooling on a theoretical consideration. Every manufacturer, however, should determine whether his ware is improved by slow cooling and to what extent. It may be cheaper to stand the cooling losses than the cost of the slower operation, and, of course, that which shows the most profitable operation should be adopted. 42 BURNING CLAY WARES, CHAPTER III. BURNING BEHAVIOR OF CLAYS. T HE BURNING behavior of clays is shown by the color changes with advancing temperature, such as : Light red, red, dark red, brown, black ; light red, red, dark red, dark red, etc., the red color continuing even to complete vitrification ; cream buff, buff, greenish buff, green or gray ; light red, buff, greenish buff, green, dark green ; cream white to grayish white. Under reducing kiln atmospherers besides the browns and gun-metal blacks which we get from red burning clays, there are a number of color effects, such as saffron and dull green, besides the partially reduced products in which we have brown edges and red centers. The flashed colors tend to the russets in buff burning clays, usually speckled with black iron spots, while the red burning clays have the browns and gun-metal blacks. The color is primarily due to the character of the clay, but the variation in color marks the progress in the burning. Burners are instructed to burn this kiln hard for gun metals, that kiln for reds, or red centers, another for a light flash, a heavy flash, bluish buffs, light or dark grays, etc., etc. The difficulty of burning is dependent upon the character of the clay. A clay with a very short burning range requires great care in getting the temperature high enough to pro- duce satisfactory ware, yet without exceeding the safe burn- ing limit of the clay. Face building wares are troublesome because of the wide variety of color effects which must be produced and which require different burning treatment. Vitrified wares require temperatures approaching the max- imum limit and the only favorable factor is that each kiln receives the same treatment. Clays which are used for com- BURNING CLAY WARES. 43 mon wares, such as common brick, drain tile, fire-proofing, etc., are often impractical for face brick products or vitrified wares, but on the other hand, clays which are suitable for the latter products are easily burned when made into the former products. The burning behavior of clays is nicely illustrated by their change in size under increasing temperatures, as the following illustrations show. The shrinkage is indicated in per cent, on the vertical line on the left. The zero line marks the original size of the test piece. The advancing temperatures are shown from left to right on the base line and are indicated by cones. Illustration Fig. 5 shows an ideal shrinkage curve indica- tive of the burning behavior. At cone 07 the shrinkage has advanced nearly 1 per cent, and at a rate of about 1 per cent, per cone advance in temperature. The rate of shrink- 44 BURNING CLAY WARES. age decreases gradually until at cone 5 the shrinkage is com- plete and there is no change in the size from cone 5 to cone 7. We do not know how much higher the temperature could be carried without damage to the ware. The ware is steel hard at cone 04 and the burning range for all hard ware is from cone 04 to cone 7 — in all ten cones and perhaps more. A burner would have to be very careless or incompetent not to get good results from such a clay. Compare this with Fig. 6. The latter begins to bond and shrink at a lower temperature than No. 5, as is evident from the fact that the shrinkage is over 2 y 2 per cent, at cone 07. At cone 02 the curve is upward, which means that shrinkage has ceased and swelling begun. At cone 3 the sample has lost 5 per cent, of the total shrinkage through bloating. The maxi- mum safe burning temperature is cone 02. The burning range would b*> U 0 g ft c 3 0) M 03 a> 0 O KA fl ■3 § m u o3 a> H 03 EH < § X ^ x 232 .564 1.354 1.918 13.7 11.8 13.56 9.10 432 .726 1.742 2.467 24.0 21.5 14.40 12.90 632 .888 2.131 3.019 30.5 27.5 14.98 13.50 832 1.048 2.515 3.553 35.1 31.5 14.57 13.07 1032 1.210 2.904 4.114 38.4 34.3 13.82 12.34 1232 1.368 3.283 4.651 40.9 36.3 12.97 11.51 1432 1.531 3.674 5.205 42.9 37.7 12.18 10.71 Stack Furnaces. Some kiln builders use a small furnace directly connected with the stack to increase the stack intensity. This it does, but at the same time it introduces an additional volume of gas and thus uses some of the power which it creates. To whatever extent air enters through the stack furnace, to the same extent is the draft through the kiln lessened, tempera- BURNING CLAY WARES. in ture not considered. The size of the furnace is not impor- tant, but the opening into the stack should be of such small size that the resistance equals or exceeds the kiln resistance, depending, of course, on the volume of hot gas from the furnace required to heat the volume of gas from the kiln to the desired temperature. There is a legitimate use for such a furnace in the early stages of the burning when the stack power is very low, and yet at the same time we need a strong draft to sweep the moisture out of the kiln during the early stages of water- smoking, and during oxidation when we need a large excess of air for the oxidation. Two or more kilns on one stack with a flue for each kiln have the advantage that the stack is kept hot and in starting a kiln we have the value of this temperature, which the kiln in question could not give during the early stages of firing. Two kilns should never be connected with a single stack flue. When we have one kiln connected it has the full power of the stack controlled, of course, by a damper. When we connect a second kiln we rob the first kiln of some velocity of movement of gases through the kiln, since we must divide the volume of gases moved proportionately between the two kilns, and besides the second kiln being cold, lowers the available head of the stack. Such conditions are undesir- able if for no other reason than the irregularity of the opera- tion. Height of Stack for Periodic Kiln. Data is lacking to develop any definite rule to be used in the determination of the heights of kiln stacks. The stack temperature and the kiln resistance are the two items necessary in order to establish such a rule. These can be determined in any operating plant, but the data may apply only measurably to another plant in which the setting is different, offering greater or less resistance; in which a different type of furnace is to be used; in which the quantity of fuel burned per hour is greater or less; in which there is a more or less complicated kiln bottom. If we had data from a number of kiln operations under different conditions, one approximating the desired condition could be selected and with some corrections, perhaps, could be used, but such data is entirely lacking. 118 BURNING CLAY WARES. We can get some idea of the approximate height from the following method of calculation. The formula for total stack intensity is Wt. of Air — Wt. of gas I = H ( ). Wt. of Air. V 2 d The formula for velocity heads is h = 2g. hHK and for friction head, F = . S We wish to determine the height, H. The problem should be solved for some average atmos- pheric temperature, let us assume 62 degrees, at which a cubic foot of air will weigh .076 pound. The gas at 32 de- grees weighs .087 pound. Prof. Gale, in a test of a boiler, found the total static pressure to be 0.48 inch water pressure distributed as fol- lows: Entrance Velocity 0.5% Great Resistance 36.6% Boiler Tubes, etc 49.5% Discharge Velocity 3.4% Flue to Stack 2.4% Stack Friction 7.6% 100 . 0 % Stacks based on a kiln resistance equal to boiler resistance would make it possible to burn the coal at a rate of 15 pounds of coal per hour per square foot of grate surface, which is good boiler practice. The rate of combustion in kiln work is scarcely one-half the rate in boiler work, and if we consider the stack solely from the standpoint of the rate of combus- tion, the stack heights could be materially reduced. It must be remembered, however, that the average tem- perature of the gases in a kiln is much higher than in a boiler equipment and the frictional resistance is greater in consequence, leaving less available stack intensity for the furnace resistance. A test of a gas burning kiln stack showed .2-inch stack intensity. There was no furnace resistance, but, to the con- trary, the gas was introduced under pressure, and this pres- sure was available to assist in overcoming the kiln resistance. The sum of gas pressure and the stack intensity exceeded BURNING CLAY WARES. 119 • .35-inch of water pressure, and the test indicates that a coal burning kiln set with bricks will offer a resistance greater than .4-inch of water pressure. It also must be borne in mind that during the water smok- ing and oxidation stages, which are prior to the maximum intensity stage of the stack, there is a large volume of water vapor, and often sulphur-dioxide and carbon-dioxide from the ware, which must be considered. Water vapor being lighter than air increases the stack intensity, but the other two are approximately double the weight of air. Finally, in the best practice it is customary to partially close the damper during the finishing stages of the burning, thus introducing a resistance which must be offset by less kiln resistance, or, in other words, checking the kiln draft The stack must be designed for the worst condition of the burning operation, and controlled for the more favorable con- ditions. We will not be far astray if we adopt the boiler data for our periodic kilns. The two items of grate resistance and tube resistance total 86 per cent., which in round numbers require a stack intensity of .4-inch of water, plus the stack requirement. This kiln requirement in feet of air will be .4 X 774 = 25.8 feet 12 The total intensity required will be 25.8 -f velocity head + friction head, which will equal Wt. of Air — Wt. of Gas. H ( ). W t. of Air. If we assume a stack temperature of 1,432 degrees, the second member of the above equation becomes : .076 — .087 X .284 X H, which reduces to .675H. .076 The velocity in the stack, as previously determined, is 18.62 feet per second. The velocity head becomes 18.64 2 X .284 X 1.06 = 1.6 64.3 Note — We wish to determine the velocity head in feet of air at 62 degrees, and therefore the densities as previously given must be multiplied by 1 + .002 t = 1.06, because .284 120 BURNING CLAY WARES. is the density relative to air at 32 degrees, whereas relative air at 62 degrees it will be .284 X 1.06 = .301. We have already determined the velocity heads relative to 32 degrees and need only to correct these values for 62 degrees, namely, 1.531 X 1.06 = 1.6 feet velocity head for 1,432 degrees. For a stack 2 feet 6 inches in diameter, or square, the friction head is H 1.6 X X .1 = .06 H. 2.5 We now have : Total stack intensity = .675 H Kiln and furnace resistance =25.8 Velocity head = 1.6 Friction head = .06 H We wish to find the value of H. The last three items equal the first, which gives us the equation, .675 H = 25.8 + 1.6 + .06 H, from which we find H = 45 feet. The distance of the stack from the kiln introduces flue resistance equivalent to the stack friction per foot, which we have found to be .06. We must introduce this into the equation, and let us as- sume flue lengths of 15 feet and 300 feet. For the first equation we have: .675 H = 25.8 + 1.6 + .06 H + .06 X 15. H = 46 feet. For the second equation: .675 H = 25.8 + 1.6 + .06 H + .06 X 300. H = 74 feet. In kiln wall stacks we have even higher temperatures than 1,432 degrees. Since the stack is separated from the interior of the kiln by a 4%-inch wall, or at most 9 inches, and since combustion gases are burned in the stack at least a part of each firing period, the stack temperature will approximate the kiln temperature. Wall stacks then may be correspond- ingly low. On the other hand, the greater the distance of the stack from the kiln, the cooler the gases and the higher the stack requirement. The following table gives the heights for different stack temperatures, in the first column without flue resistance, in the second including 15 feet of flue, and in the third including 300 feet of flue. BURNING CLAY WARES. 121 Temp. Height Height Height in stack without flue 15 ft. flue 300 ft. flue 432 94 95 125 632 66 67 93 832 56 57 80 1032 50 51 77 1232 47 48 76 1432 45 46 74 Stacks within 15 feet of the kiln will usually have tem- peratures between 832 and 1032 degrees, and stacks 300 or more feet away will have the temperature reduced to 632 degrees or less. In working out the proper height of stack the engineer must take into consideration and allow for a number of fac- tors which cannot be included in a typical calculation. The above calculations are based on the consumption of .18 pound of coal per second. An increased coal consumption will have no effect on the grate resistance, since it would involve increased grate area, but it does involve iu creased kiln resistance because we would thereby increase the volume of gases passing through the kiln, thus increasing the velocity which is a factor of the friction. We provide a larger stack area for increased volumes of gas, but we cannot, or do not, provide a larger free area in the kiln, and therefore the stack should be higher to overcome the increased kiln resistance. On the other hand, we cannot increase the velocity of the gases through the kiln without in some degree increasing the stack temperature, and this will lower the height of stack required. We assumed a kiln resistance based on a limited boiler test, and this may not be a reasonable assumption. Our justification is that the results approximate good practice in clay ware burning. Our assumption, further, is for a down draft kiln,, set with brick. Hollow ware would have less resistance and would not require as high a stack. We have determined the weight of the kiln gas at 32 de- grees to be .087 pound per cubic foot, and noted that the weight varies between .085 and .087, and if we use the lignter weight we increase the stack intensity and reduce the height of the stack. Our calculations are based on coal burning furnaces, which offer resistance and require additional stack height in con* 122 BURNING CLAY WARES. sequence. Where the fuel is natural gas, producer gas or oil, it is introduced into the furnace box under pressure, thus not only eliminating furnace resistance, but actually developing a pressure available to overcome kiln resistance, lessening the work to be done by the stack and reducing the required stack height correspondingly. The stack construction is important. The intensity of the stack depends upon the average temperature of the gases in the stack, and the temperature is lowered by increased radiation and leaks. The situation also must be considered, the height and position of adjacent buildings or hills and the direction of prevailing winds. The height of kiln wall stacks varies from 6 to 8 feet above the kiln crown, or a total of 18 to 24 feet. Stacks just out- side the kiln walls range in height from 25 to 50 feet depend- ing upon the type of kiln. We have built rectangular kilns 100 feet long with stacks at the end, and the long draft flue had to be reckoned with in deciding the height of the stack. Stacks 15 to 18 feet away from the kiln are built from 40 to 60 feet high, not considering long draft flues inside kiln, as mentioned above. When the stacks are several hundred feet away from the kiln — a single stack for a battery of kilns — the height should be determined for the most distant kiln and will be from 80 to 100 feet high. Continuous Kiln Stack. A modern chambered continuous kiln is a series of down draft compartments comparable with a periodic kiln. Each compartment has bags, checkered floor, under floor flues, and the setting is the same as in down draft kilns. The furnace resistance of a coal fired periodic kiln is eliminated in the continuous kiln. From the limited data available, it would not be safe to figure on less than 0.3 inch water pressure for each compart- ment of a continuous kiln during its maximum temperature period. We will assume capacity of 30,000 brick per day, and a fuel (coal) consumption of 400 pounds per thousand brick, making 12,000 pounds per day, or .139 pounds per second. We will also assume 100 per cent, excess air. BURNING CLAY WARES. 123 The Hocking Valley coal, per pound, under such condi- tions, will give us at 32 degrees : Dry Air 114.3 cubic feet C0 2 20.6 cubic feet N 90.3 cubic feet S0 2 3 cubic feet Water Vapor 13.5 cubic feet 239.0 cubic feet The total weight of the gas is 19.5 pounds, and the weight per cubic foot is .082 pound, which gives a density of 1.02. The cubic feet of gas are as follows : Per Pound Per Second Entering Air 236 cubic feet 32.8 cubic feet Combustion Gas 239 cubic feet 33.2 cubic feet We will assume the following temperature in seven con- nected chambers, viz., 232, 932, 1632, 2132, 1232, 832, 632, and an average stack temperature of 432. The first three com- partments is the combustion compartment ; the last three are heating up. Correcting the air and gas for temperature, assuming that the air enters at 62 degrees, we determine the following rela- tive velocities per second : 1.32, 2.64, 3.96, 4.97, 3.25, 2.48, 2.1. The densities will be : 0.75, 0.32, 0.25, 0.2, 0.3, 0.4, 0.48. The relative resistance is as the velocities squared times the densities, and in this way we get the following results : 1.31, 2.58, 3.92, 4.94, 3.17, 2.46, 2.12. The total relative resistance is the sum of the above, and is 20.5. The combustion compartment has a relative resistance of 4.94 and the actual resistance assumed is 0.3. The total kiln resistance will be in the proportion of 4 94 20.5 .3 x x = 1.245 inch water pressure = 80.3 feet of air. The volume of gas per second at 32 degrees is 33.2 cubic feet, which at the stack temperature becomes 59.76 cubic feet. If we assume a stack diameter of 4 feet, the velocity in the stack will be 4.8 feet per second. 124 BURNING CLAY WARES. The total stack intensity will be : .076 — 0.455 H = .4H .076 The velocity head will be: 4.8 2 X 6 .215 64.3 The friction head is : .0215 H = .0054 H We get the equation : .4 H = 80.3 + .215 + -0054 H, From which: H = 204 feet. In order to build a stable stack for such heights the diam- eter must be large which reduces the velocity and friction heads to a negligible factor, and we may determine the height from the total intensity and the required pressure which in the above example would give 201 feet for the height of the stack. We have figured on seven compartments, which is a prac- tical minimum often exceeded, and for a greater number of compartments we must increase the height of the stack. The modern compartment continuous kiln is complicated compared with earlier types of kilns, and the resistance is correspondingly greater, which explains why we have resorted to induced draft. The earlier kilns had solid floors, no checker work, no under floor flues, and the compartments were con- nected by relatively large direct openings. The ware was set with flues, longitudinal in the lower part of the kiln, and verti- cal feed holes at short intervals throughout the mass of ware. Tunnel kilns of the Hoffman type have solid floors, and no division walls, and flues are set in the ware. The resistance in such a kiln would be less although the gases are pulled through a longer distance. The latter type of kiln may be likened to a long, tortuous, exceedingly rough flue. If we figured on the basis of equivalent 0.2 resistance per compart- ment in a tunnel kiln, the stack would be 136 feet high. A resistance of 0.25 per compartment would require a stack 158 feet high. Stacks for the earlier types of compartment kilns and the tunnel kilns vary from 125 feet to 190 feet in height. In the BURNING CLAY WARES. 125 earlier types of compartment kilns the stack was frequently included within the kiln walls and its base was kept hot by conduction from the adjacent compartments. In both types the main draft flue was above the ground in the longitudinal center of the kiln, with the compartments or tunnels on either side, and thus the stack gases averaged a higher temperature than that entering the main draft flue, which increased the draft intensity. In spite of these aids, we frequently found it necessary with stacks of minimum height to open a damper into the first compartment ahead of the combustion compart- ment in order to strengthen the draft by increasing the tem- perature of the stack gases. The discussion brings up a point worthy of mention. In modern kilns we by-pass the air for water smoking, the pur- pose of which is to insure the water smoking keeping pace with the burning, to reduce the scumming difficulty and to overcome the swelling peculiar to a continuous kiln product. The hot air is taken from a cooling compartment and con- ducted through a flue direct to the compartments ahead. The air leaving the cooling compartment has an excessive force and the flue offers little resistance. The total resistance in this circuit may be the resistance of a single compartment, or two compartments, or three, depending upon whether they are connected to the main draft flue singly or in series. The temperature in these compartments is relatively low, and the resistance less in consequence. These low resistance compartments are connected with a stack or fan adapted to overcome the resistance through a longer series of higher temperature compartments. The effect is to weaken the drift intensity, in consequence of the reduced stack temperature, and because of the increased volume, although this is a small matter if the stack area is amply large. The proper adjustment is obtained by introducing damper resistance, but this is uncertain at best. The control of two distinct and widely differing operations with a single equipment emphasizes the importance of greater flexibility, which a fan gives, and it is but a step to the use of two fans, one for each operation We have not attempted to go into the details of the prob- lem. For instance, the clay will contain 5 to 10 per cent, of combined water and 3 per cent, or more of hygroscopic water, which as vapor in the gas will lower the density, and lessen 126 BURNING CLAY WARES. the height of stack required. On the other hand, carbon or sulphur in the clay would increase the density of the gas, unless they were included in the combustibles and provided for, otherwise they would convert excess air into heavier car- bon and sulphur dioxides. Construction of Stacks. The bearing weight of soil, clay, gravel, etc., are usually taken as follows: One ton per square foot for soft wet soil. Two tons per square foot for firm, wet soil and sand. Three tons per square foot for firm, dry soil, clay or fine sand. Four tons per square foot for dry, hard, coarse sand, clay or gravel. This leaves a wide margin for the exercise of one’s judg- ment, and the rule should be to err on the safe side. Hotop’s formulas for the depth and size of stack founda- tions and foundation base plates are as follows: Depth of foundation, one-eighth of the height of the stack above ground. Breadth of foundation, one-eighth of the total height of stack. Thickness of foundation plate (below the flue entrance), 1.6 + .01 H. A 60-foot stack, by these rules, would have a depth below ground of 7.5 feet. The breadth would be 8.4 feet, and the thickness of the base plate 2.24 feet. A better method is to estimate the weight of the stack and determine the foundation dimensions from this data. A 60-foot stack with a 4%-inch lining and outside walls 13 inches, 9 inches and 4 y 2 inches thick in sections each 20 feet high and a foundation depth of 7 feet will weight 70 tons. The maximum breadth of the base for soft, wet soil will be 8.2 feet. On the basis of 2 tons per square foot, the re- quired breadth of the foundation is 5.8 feet, but if the stack has a 30-inch flue the side at the base will be 6 feet 5 inches, and with proper set-off at the bottom, together with a proper slope or steps in the base, we will get the maximum breadth. The angle of the slope should not be less than 30 degrees from the vertical. BURNING CLAY WARES. 127 Concrete should not be used in the stack foundation above the bottom of the kiln draft flue, nor, indeed, nearer than 12 inches to the bottom of the flue, on account of the high temperatures which develop in the stack. If the ground is soft and wet, it is desirable to carry the foundation deeper in order to use a monolithic concrete base plate,' but as a rule, if the ground is suitable for kilns which have underground flues to a depth of from 4 to 8 feet, the stack foundation will be on hard, dry soil or gravel and there is little need of a massive base plate. The foundation above the base plate should be built of hard burned brick laid in cement mortar. Every kiln stack should have an independent fire brick inwall. We have stack temperatures exceeding 1,000 degrees, and if the wall is single the expansion inside lifts the outer brick from their bed, and once loosened, they are easily sep- arated laterally, developing zig-zag cracks from top to bottom. The fire brick inwall need not be over 4 V 2 inches thick unless the stack is very high, and the brick should be laid in a thin bed of fire clay mortar in the lower part of the stack and adding some cement to the clay mortar for the upper part. The end joints may be heavier, in fact, they should be, to provide locally for the expansion of each brick. The top of the inwall should be several inches below the top of the outer wall, so that the expansion will not lift any protective cover placed on the outer wall. The outer wall at the base should be set back about 6 inches from the inwall and be carried up with a batter inside and out until the inside is approximately 2 inches from the inwall, then drop off one brick in the thickness of the wall without breaking the continuity of the batter of the outside face and carry this wall to the proper height to drop off an- other brick in the thickness. Tne height of each section should be about 25 feet and not exceeding 30 feet. A 50-foot stack may start with a 9-inch wall, dropping off to a 4% inch, and similarly a 60-foot wall, but the latter preferably should have three sections, 13 inches, 9 inches, and 4 y 2 inches thickness of wall respectively. Stacks exceeding 75 feet in height should have no section less than a 9-inch wall, not considering the fire brick inwall. The outside wall should be laid in cement or lime-cement mortar. An excellent lime mortar for kiln work is made of lime and ground furnace clinkers instead of sand. The clink- ers are much sharper than sand and have some hydraulic property. 128 BURNING CLAY WARES. The top of the outer wall should have a suitable cap, either iron or reinforced concrete, and this should overhang inside to cover the space between the outer and inner wall and par- tially cover the inner wall. It is customary to put holes in the base of the outer wall for the admission of air to keep the inside face of the wall cool. At intervals of four to six feet there should be a brick projection from the center of the inner face of the outer wall to the stack lining wall, but not bonded into the lining wall. It is better to make this projection about three courses high and to chamfer the top and bottom edges. The purpose of these projections is to stay the fire brick lining, yet at the same time not to interfere with the rise and fall of the lining as it expands and contracts under changes of temperature. Fig. No. 18 illustrates a properly constructed stack. Induced Draft. The use of fans in place of stacks is now very common. They are used almost exclusively in continuous kiln opera- BURNING CLAY WARES. 129 tions, and are beginning to replace periodic kiln stacks. The advantages of fan drafts are: 1. Less first cost. 2. Flexibility, in that by speeding up the fan, we increase the intensity of the draft and the volume of gases handled. 3. Independence of atmospheric conditions. In continuous kiln operations it is imperative that the stacks be replaced by fans in order to increase the rate of burning. There is a limit to the height of a stack, if for no other reason than the cooling of the gases to a temperature at which the weight of cubic foot of gas is equal to or greater than the weight of a cubic foot of outside air. To make a higher stack effective it would be necessary to take the gases from the kiln at a higher temperature, which would involve a loss of heat. The need is not so imperative in periodic kilns because of the high stack temperatures, but in such kilns, in view of the fact that we need greater intensity than a stack will give during the earlier stages of the burning, the use of a fan has decided advantages over a stack. The loss in heat in a pe- riodic kiln stack is a large percentage of the total fuel, and this loss is unavoidable in a stack, since the heat is repuired to create the draft. With fan draft we may make use of this heat, and extended use of fans for periodic kilns will lead to the development of uses for the waste heat in the gases. Dryers have been built in which the combustion gases from the kilns are taken through flues in the dryer and the heat applied to drying the ware. Here is one direct use of the waste heat, which with fan draft is readily adaptable, but which is not applicable to stack draft because of the increased resistance introduced by the dryer flues and the decreased stack intensity in consequence of the lower temperature stack gases. Such an application of waste heat to radiated heat dryers is but a step in advance of the present direct firing. The collection of the waste heat in the combustion gases in one or two kiln installations has been accomplished by an economizer, and this idea is worthy of further development. It would be practical to use the heat thus collected for primary or secondary air in the kiln furnaces. Some designs have been worked out to generate steam for power with the waste gases under the kiln floor, which would 130 BURNING CLAY WARES. not be satisfactory because of the cooling effect on the floor, but there can be no objection to using the heat for such pur- pose. Such conservation of waste heat is in practical opera- tion. The selection of the fan for induced draft requires study and thought. We must first determine the volume of gas to be moved, provide for maximum conditions, and make some allowance for leakage. The theory of fan performance is a little puzzling to a layman, and each problem should be put up to a manufac- turer of fans, but unfortunately we cannot give them satisfac- tory data. All tables of fan performances presented in fan catalogues are for ventilation and are only measurably appli- cable to clayworking conditions. A brief discussion of fan operation will not be out of place. Mergue, Rateau and others discuss the problem from the standpoint of effective area or epuivalent orifice. In Fig. 19, the large square represents the outlet of a fan. The shaded area, which is 40 to 50 per cent, of the total area, is the effective area, or, in other words, it is the area of opening within which we can maintain a constant static pressure without changing the speed of the fan. If we assume a desired static pressure of one inch of water for any operation, we cannot get this with a fully opened outlet. If we close the outlet entirely, the fan will develop a pressure of one inch within the closed space. When this pressure is attained the back- ward pressure through the fan just balances the forward pres- sure by the fan. If we open the outlet, increasingly up to an outlet equal to the shaded area, the pressure within the space will remain* practically constant, and the effect is an increase in volume of air. If we are designing an equipment with several distributed outlets as a dryer, for instance, the outlets should have an equivalent area equal to about 40 per cent, of the fan outlet, and thus we will get approximately the volumes listed and the pressure throughout the main dis- tributing duct will approximate the estimated pressure. If we have a fan connected with itself by an encircling duct, as shown in Fig. 20, there will be no static pressure in the duct if it is equal to the fan outlet throughout the circuit, friction not considered. If we introduce a diaphragm as at Fig. 19. BURNING CLAY WARES. 131 “A,” this when completely closed enables the fan to develop its maximum pressure, say, one inch, between the diaphragm and the fan and this static pressure relatively will be main- tained up to a 40 per cent, diaphragm opening, above which the pressure falls off rapidly to zero at the full opening. If we move the diaphragm to “B,” there is no change in prin- ciple, except we say that the fan is overcoming a resistance Fig. 20. of one inch instead of developing a like pressure. Suppose now we remove the air from the duct, there being nothing to move, no pressure can be developed. If we introduce a little gas we can develop a slight pressure, increasing with the density of the gas until at atmospheric conditions we again develop the maximum pressure. The point of this is that the pressure given in catalogue data is based on atmospheric conditions, and if we wish to handle a hot, rarified gas, we must select an equipment with 132 BURNING CLAY WARES. a higher listed pressure in order to overcome the equivalent resistance. It is simply a question of force equals mass times velocity. If we increase or decrease the mass we increase or decrease the force which we measure as pressure. Effective areas assume that the pressure is constant up to 40 per cent, opening, but this is not strictly true. The per cent, pressure curve for a steel plate fan may be represented by the solid line curve shown in Fig. 21. With no outlet we get the initial pressure, which increases slightly with an opening approaching 20 per cent. Then follows a pres- sure with slight change up to 50 per cent, opening, beyond which there is a rapid decline in pressure. The effective area is this range of slight change, and we preferably take the up- per limit in order to get a maximum volume. The dotted curve illustrates the behavior of a multi-vane fan. First a slight drop, then a rise, followed by the effective area between 30 and 60 per cent, opening. In working out a problem for a pressure system, we select a fan with the required volume and pressure capacity and proportion the delivery outlets to suit the fan. BURNING CLAY WARES. 133 We cannot do this for induced draft, which would mean constructing the kiln and setting the ware to correspond with the fan. We must determine the kiln resistance, the volume, density and temperature of the gas. In pressure estimates we figure on 40 per cent, effective area, but since we do not know what the area is in induced work, we should err on the safe side and base our estimates on an area not exceeding 30 per cent. The determined resistance must be increased to cor- respond with the decrease in gas density compared with air, then select from the fan tables the fan which will deliver the volume of gas, temperature considered, and which will de- velop a static pressure equal to the resistance. This pres- sure is a matter of speed, and the selection should be an aver- age speed which gives the desired flexibility for a fan equip- ment. In our periodic kiln stack problem we assumed a kiln resistance of .4 inch of water. Let us say that the kiln re- sistance, including flue to fans, turns in flue, etc., totals .5 inch resistance and that the gases reach the fan with a tempera- ture of 832 degrees. We estimated on the basis of .181 pounds of coal per second, which in round numbers would be 11 pounds per minute developing about 1970 cu. ft. of gas. This increases to 5122 cu. ft., or allowing for leakage, assume 6000 cu. ft., at 832 degrees, and the density becomes .38. The re- sistance then should be estimated 1.32 inches water pressure. The fan (steel plate) should have a 4y 2 -foot wheel speed 350 r.p.m., or a 5-foot wheel speed 300 r.p.m. If we should figure on four kilns, the fan should have a 9-foot wheel with a speed of 190 r.p.m. This gives an excellent starting point in the selection of the fan, and the results from available data and carefully considered assumptions will apply to our clayworking opera- tions. In the selection of a single fan for the double operation involved in continuous kiln operation, we must make wide allowance for imperfect operation. We can determine within a reasonable approximation the volume of gases from the burning operation, and, given the temperature, we can deter- mine the air volume of the water-smoking, but we cannot deter- mine the uncertain control of the latter. No harm results if we use ten times the required volume in the water-smoking except the decrease in the profits through excessive waste of heat. 134 BURNING CLAY WARES. It can be shown that it requires nearly, perhaps twice, as much air for the water-smoking as that required for the burn- ing. Assume 30,000 bricks containing 3 per cent, moisture aDd 2 per cent, hygroscopic water heated up to an average tem- perature of 432 degrees by air entering at 632 degrees and leaving at 332 degrees. 180.000 pounds brickwork X .2 X 185 6,660,000 B.t.u. 171.000 pounds dry clayware X .2 X 370 12,654,000 B.t.u. 5,400 pounds water X 150 810,000 B.t.u. 5,400 pounds water to vapor X 970 5,238,000 B.t.u. 5,400 pounds water vapor X .48 X 220 570,240 B.t.u. 3,600 pounds water (hygroscopic) X 370 1,332,000 B.t.u. 3,600 pounds water X (966 — .7 X 200).... 2,973,600 B.t.u. 30,237,840 B.t.u. Less 3,600 X .49 X 100 • 176,400 B.t.u. Total heat requirement 30,061,440 B.t.u. If the air enters the kiln 80 per cent, saturated, the thermal value from each pound will be 72.5 B. t. u., and to give the above heat requirement we must have 405,537 pounds of air per day, or 4.7 pounds per second, or 58 cubic feet per second. The moisture from the ware adds 2 cubic feet, making 60 cubic feet per second. We found the air for combustion to be 33 cubic feet per second, and to this 2 cubic feet of com- bined water vapor must be added, making 35 cubic feet In one instance where the theoretical combined volume of gases did not exceed 18,000 cubic feet per minute, the actual volume coming through the fan exceeded 50,000 cubic feet, showing a big loss in heat and power. In the problem of stack height we found a required pres- sure of 1.245 inches of water. If the gases entering the fan have a temperature of 432 degrees, the density will be ap- proximately 0.56-inch. We have 35 cubic feet of gas per second from the combustion compartments and 60 cubic feet from the water-smoking, making a total of 95 cubic feet of gas per second, or 5,700 cubic feet per minute. At 432 de- grees this becomes 9,920 cubic feet. We must select a fan 1.245 speed which will give a static pressure of = 2.4 inches .56 and the volume .will be 9,920 cubic feet plus allowance for leakage and bad control. BURNING CLAY WARES. 135 Theoretically, this would require a steel plate fan with a 6-foot wheel, running at something over 400 r. p. m. If we allow three times the estimated volume for leakage and bad control, which was the fact in the test cited, the opera- tion will require a fan with a 9-foot wheel, running about 250 r. p. m. 136 BURNING CLAY WARES. CHAPTER VII. FURNACES. T HE FURNACE is the most important part of a kiln and upon it largely depends the economy and success of the burning. There are two general types — the flat grate and the in- clined grate, or more properly, pit furnaces — which include nearly all the furnaces in common use in clayware kuns. The common flat-grate furnace, such as we find in boiler furnaces, is too well known to need any description. The selection of the grate bar best adapted to the purpose is an important item in economical operations and many plants may net a comfortable profit by attention to this detail. The ordinary double parallel bar has not good distribution of the metal and in consequence its life is short and the waste in fuel excessive. The herring-bone bars with wide side ribs are better, although objectionable in that the proximity of the ribs in adjacent bars does not give full opportunity for cool- ing, which is necessary to keep the bars from sagging in the center. A better type has a single central deep rib with eheckerwork top, Fig. 22. The rocking grates, of which there are several on the market, will prove a profitable investment over the usual fixed type of bar. The average clayworker looks askance at the cost of an installation, but views with complacency a seri- ous fuel loss simply because the loss is charged to burning and does not appear on the books debited to the proper opera- BURNING CLAY WARES. 137 tion. Clayworkers would find it profitable to analyze the ash for carbon, a simple and inexpensive analysis — and then by tests of bars and furnaces select that which shows the least loss. A double furnace (Figs. 23 and 24) has proven very satis- factory in burning ware where constant oxidizing conditions are required and, in fact, such a furnace is preferable to a single furnace in any operation. The twin furnaces have a single bag, and they are fired alternately. When one is freshly fired, the distillation gases are a maximum and the secondary air is insufficient for complete combustion. The adjacent fur- nace has in tlie meantime reached the burning coke stage, likely with some excess air. The products from both furnaces mingle more or less in the single bag and that which one lacks in air and temperature is supplied by the other. If the fur- Fig. 23. Fig. 24. naces are clinkered at proper intervals, one will be at its maxi- mum heating efficiency when the other is approaching the period of lowest efficiency, followed by chilling effect of clinker- ing and filling with fresh fuel. Secondary Air. Secondary air is essential for complete combustion, and its admission to the furnace has been the subject of considerable study and numerous patents. A few holes in the furnace door are the simplest arrangements and likewise the least satis- factory. Leaving the doors partly open is less satisfactory, because of the carelessness of the operators. In such oper- ation we have found the openings varying from half an inch to half the width of the firing hole. 138 BURNING CLAY WARES. Setting the door to leave a space between the bottom of the door and the firing plate is better, since it introduces air at the bottom, but being a fixed opening makes it objection- able, especially where alternating oxidizing and reducing con- ditions were required. Leaving a space between the door plate and the ends of the grate bars is still better (see Fig. 24) in that it brings the air into immediate contact with the burn- ing coal, and it can be readily closed by a loose plate or even by coal tailing out on the door plate. A number of patented constructions have been introduced which look good on paper, some of which have merit in practi- cal use. Fig. 25 and Fig. 26 illustrate one method of introducing hot secondary air, which has been extensively used. The furnace lining is separated from the back wall by a Fig. 25. Fig. 26. four-inch space and this space connects with the outside air through a hole in the front wall, and similar holes in the furnace lining introduce the air into the furnace. The air is heated by the furnace walls before entering the furnace and it can be shut off in any degree by closing, partially or entirely, the opening in the outside wall. We had extended experience with this design, but we do not know what benefit resulted. Perhaps, when first installed it was effective, but after a time when the furnace doors be- came warped and no longer close fitting, and the secondary air ports became clogged with ash and clinker, it is likely that the resistance to the entering air was less around the doors than through the air ports, and that the secondary air entered by this route rather than by the one provided. BURNING CLAY WARES. 139 Another method which is more promising introduces the air through the ash pit walls from the inside, which rises in the space between the furnace lining and wall and enters a similar space between the inner and outer crown arches, thence through ports into the bag, as shown in Fig. 27. This method has the advantage over the preceding one in that the upcast would act as a stack, and possibly there would be some aspi- rating effect of the gases rising in the bag. Another method, shown in Fig. 28, introduces the air from the ash pit through the base of bag wall, but if it is effective, it is bad in that the air would tend to cool the bottom of the bag wall, whereas we depend upon the heat conducted through the bottom of the bag wall to finish the ware on the kiln floor near the bag wall. Several designs have been worked out to use the heat i . Fig. 27. Fig. 28. from the flues under the downdraft kiln floor by means of parallel adjacent air ducts leading to the furnaces — in one instance for the generation of power steam — but these efforts are usually unsatisfactory in that they tend to cool the part of the kiln we have the greatest difficulty in heating up. Coking Table Furnaces. The coking table furnace is used in many plants for bitu- minous fuel. Fig. 29 illustrates a single furnace with a double coking table, and Fig. 30 shows a double furnace with a single coking table. The coal is first placed on the coking table, then before 140 BURNING CLAY WARES. each firing the coke is pushed off the table and falls on the grates where the combustion is completed. A type with the coking table in front is shown in Fig. 31, which has been used to some extent in kiln furnaces and also in boiler furnaces. This furnace can be clinkered without Fig. 29. Fig. 30. opening the firing door and secondary air enters through the space between the coking table arch and the grate bars. The objection to these furnaces is that they involve extra labor on the part of the fireman and increase the period during which the firing door is open. In firing a flat grate the fire- man spreads the coal over the furnace by a dexterous move- ment of the shovel, but to level the coke after it leaves the Fig. 31. Fig. 32. table requires a separate operation before again charging the coking table. The McManigal furnace, shown in Fig. 32, combines a coking table with the pit type of furnace, which simplifies the operation in that the pit furnace requires no leveling. Sec- BURNING CLAY WARES. 141 ondary air in the natural draft furnaces is admitted through small horizontal ducts in the coking table as shown by the dotted line. Pit Furnaces. A pit furnace is in effect a gas producer. The most com- mon form is the inclined grate bar, illustrated in Fig. 33, which resembles the original Siemens producer. The grates are usually flat bars of iron hooked over a bearing bar at the mouth of the furnace and extending to within about one foot of the pit floor. The bars are set at an angle varying from 30 to 60 degrees Figure 33. from the vertical, and their purpose is less that of grate bars than that of a supporting plate. The coal is fired at the top, and the secondary air supply is regulated by partially or completely closing the fire mouth with coal. As the coke in the pit burns and its volume be- comes reduced, the coal on the plate slips downward or is forced down with the shovel prior to introducing a fresh sup- ply of coal. Thus the coal passes through the several pro- ducer zones — distillation, dissociation, combustion and ash. The primary air largely enters through the ash tailing out below the grate bars. The bed of coal is necessarily deep, and the gas rising from it is lean producer gas from coke. This mixes with the distillate gas and secondary air in the combustion space over the bed of coal where the final com- bustion takes place. The usual construction of the furnace is with a straight, horizontal arch, but Hull in Trans. A. C S. shows the advan- 142 BURNING CLAY WARES. tage of stepping down the arch toward the bag, as shown by the dotted lines in Fig. 33. This, or a simple drop arch on the inner ring, forces the air downward keeping it in close touch with the coal and insures better mixing of the air and gases. It is not essential that we get the distinctive results of a producer since the furnace is a part of the kiln and we get all the heat of the combustion except that lost by radiation from the front, regardless of the particular zone in which the heat is developed. There are many modifications of this simple furnace, some good and some not so good. Frequently a solid plate of brick work or fire clay blocks is used instead of the grate bars and our preference is for such a solid plate as shown in Fig. 34 and Fig. 35. It enables us Fig. 34. Fig. 35. to reduce the size of the furnace mouth and gives us better control of the secondary air. The Paul Beers furnace is illustrated in Fig. 36 and needs no explanation, except to call attention to the secondary air port in the drop arch, which, since it can be readily closed, gives the burner better control. The McManigal furnace, Fig. 32, as noted, has a coking table, and this table is the distillation zone. McManigal has also introduced a forced draft feature, and the air for primary combustion is heated by the radiation from the kiln crown. A second crown is sprung over the first with a space between the two. Air, by means of a fan, is forced through the top crown vent into the space between the two crowns, and thence at the spring level is conducted through pipes to the furnaces and introduced into the ash pit, which is provided with a tight door. See Fig. 37. In this way he BURNING CLAY WARES. 143 collects the heat loss by radiation through the crown and returns it to the furnaces. Unclassified Furnaces. The Boss system of burning is a forced draft application. The furnace is simply an arched rectangular opening in the kiln wall. On the ground level in the furnace is placed a flat, rectangular cast-iron box about 12 inches by 24 inches with perforated top, or more correctly, with numerous air ports of special design in the top. One end of the casting has a con- nection for the blast pipe. The advantage is that slack coal or slack and nut coal can be burned at a rapid rate, whereas it would not be possible to use such fuel in a natural draft furnace. The following data relative to the Boss system is an ab- Fig. 36. stract from the Transactions of the National Brick Manufac- turers’ Association. Subject — “Forced Draft in Up-draft Kilns.” “The product is common bricks burned in updraft kilns. The kilns are sixteen arches long and 21 feet wide, and the bricks are set 42 courses high in the usual manner and platted. The burners (grates) are six inches wide and two feet long, and each burner has 14 spaces 5^ inches long and %-inch wide through which the air passes. The burners are placed in the fire box level with the kiln floor. The fan is 4 1 / £ feet in diameter and is driven by a six or eight-horse power engine. A 24-inch duct runs at right angles to the kilns and there 144 BURNING CLAY WARES. is a 12-inch duct down the side of each kiln, and from these side ducts 3-inch pipes lead to the furnaces. The draft is regulated by the speed of the fan and is measured by a “TJ” tube water gage. The fan suffices for four kilns. In starting a kiln the pressure is put on, then a shovelful of coals is thrown into each firehole, followed by a shovelful of slack coal. The initial pressure is one inch to 1 % inches. During the water smoking the fire doors are left open and the platting on top is up. When the dampness is off the pressure is increased to 1% inches to 2 inches and the firing consists of one-half a small shovelful of slack coal in each hole at 8-minute intervals. In about two days when the fire is through the bricks, the platting is tightened. The firing then is about 10-minute intervals using the same amount of coal. The total time required is six or seven days, during which six to seven bushels of coal per thousand bricks are used, with a result of 80 to 85 per cent, hard bricks.” The advantages claimed for any pressure system are: (1) BURNING CLAY WARES. 145 More rapid combustion, quicker burns in consequence, thus reducing the gross kiln radiation loss. (2) Pressure distri- bution of the hot gases throughout the kiln, thus getting more uniform results. A step-grate furnace, Fig. 38, combines a grate-bar furnace in maximum degree with a pit furnace in limited degree. The purpose is to get an excessive bar surface within a limited space and at the same time get some degree of a producer condition to develop unconsumed gases to take up the excess air coming through the bar spaces. This type of furnace is particularly useful in burning low-grade fuels, such as lignite. A number of furnaces, which carry out the producer gas Fig. 38. principle, have been designed, but they are seldom used in this country in clayware burning. Position and Size of Furnaces. The possibility of burning clay wares economically depends largely upon the furnace power of the kiln, and in this im- portant item engineers differ widely. We have seen twelve furnaces in a 30-foot diameter round kiln to burn ware to cone 08 ; eighteen furnaces in an equivalent kiln for cone 1 tem- peratures ; twelve furnaces for cone 7 to 10 ; ten furnaces for cone 1, also for cone 9 ; ten to twelve furnaces for cone 20 ; eight furnaces in a 37-foot kiln to cone 18. One authority, whose opinion we value, states that if it were possible he would have a ring of fire entirely around the kiln and at the level of the top of the bag. Another authority would have the grate level four or more feet below the kiln floor level if practicable. 146 BURNING CLAY WARES. The grate level should not be above the kiln floor level if for no other reason than that leakage from the ash pit through the base of the bag wall will keep the temperature in this part of the kiln below the required temperature resulting in soft ware around the bottom of the bags. We believe there is an advantage in the burning by placing the grate level be- low the kiln floor level, but the advantage is less than the dis- advantage of the depressed firing pit. A grateless furnace gives a level yard throughout and we would not go below this level with this type of furnace. A flat-bar furnace, or any furnace with separate ash pit, requires a depressed firing pit to get the grate bars down to the kiln floor level, and we may readily go 6 inches to 12 inches deeper to get the bar level below the floor level. Furnaces should not be wider than 36 inches nor longer than 48 inches in order to clean them readily. It is better to increase the number of furnaces rather than the size. Grateless furnaces vary in width from 20 inches to 30 inches, depending upon the coal. If the coal is high in ash and clinkers badly, the wider furnace should be used in order to insure sufficient free area for the required combustion air. A number of years ago, after investigation of many kiln operations, we adopted the rule that for cone 1 the flat grate area should be 1.5 per cent, of the cubical setting space in the kilns, with an increase of .05 per cent, for each additional cone. Pit furnaces which get air through the pit area and which are not restricted by grate bars should be about two- thirds the size of the flat-bar furnaces. In designing a fur- nace we should always take into consideration the character of the fuel, the kiln tonnage and the time required to burn. A practical rate of combustion is eight pounds of coal per foot of flat grate area per hour. This may be increased to twelve pounds in burning hollow ware, or other ware not closely set as bricks, in consequence of lessened kiln resist- ance applicable to overcome the furnace resistance, but in burning closely set ware the rate is often less than eight pounds. As a basis for such determination the following general data for bituminous coal per ton of burned ware may be used : Fuel. Time. Common bricks 150 to 400 lbs. of coal 5 to 8 days Face and sewer bricks 300 to 700 lbs. of coal 6 to 12 days Paving blocks 500 to 800 lbs. of coal 6 to 12 days Drain tile 350 to 800 lbs. of coal 2 to 5 days Hollow blocks 300 to 600 lbs. of coal 3 to 5 days Sewer pipe 800 to 1,500 lbs. of coal 5 to 8 days Terra cotta 1,200 to 1,800 lbs. of coal 3 to 5 days BURNING CLAY WARES. 147 These requirements are on the assumption of an average quality of coal, and the differences are due to variations in temperature required, to furnace and kiln construction and to the care taken in the firing operation. A dirty, high ash, badly clinkering coal, will require a larger grate area in order to maintain a proper free area and similarly, a low value lignite must have provision for burning two or three times the estimated quality. The tonnage and character of the setting must also receive some consideration as well as the time required to burn. Some kilns are stacked to the crown with ware, while other similar ware requires lower setting. A kiln set with 90 tons of drain tile will require larger furnaces than one set with 60 tons. Paving blocks in some instances are burned in four and one-half to five days, but in others require ten to thirteen days with little difference in temperature requirement. Obviously, the former should have greater grate area. The question might arise, why not increase the stack height and burn at a faster rate, to which it may be replied that such high rate of combustion, reducing as it would the per cent, radiation loss, would give a higher flame temperature in the top of the kiln and in consequence a rapid absorption of heat by the top ware to its ruin. When we have attained the finishing temperature on top we simply wish to hold it by gases of approximately the same temperature, and work this temperature to the bottom of the kiln. We could get the desired bottom temperature quicker with a higher rate of combustion, but to do so would over- burn the top, perhaps ruin it, and in any event the result from top to bottom would be less uniform. To get uniform results in a down draft kiln we must work with a so-called balanced draft, namely, a draft which will merely keep the furnaces cleared of their gases and which if ever so slightly reduced will cause the furnaces to smoke around the feed holes. This applies particularly to the later stages of the burning when the ware on top has attained the finishing temperature. During the early stages of burning — water smoking and heating up — when the stack intensity is low, we should use all the draft we can get, and in the majority of instances more would oe desirable. Greaves-Walker bases the grate area on the kiln area. 148 BURNING CLAY WARES. His maximum furnace has one foot of grate area to four feet of floor area and the minimum has one foot of grate to eight feet of floor. He adopts one foot of grate to 7.5 feet of floor as the best average ratio. For sewer pipe and similar salt-glazed ware he gives a ratio between 1 to 6 and 1 to 8, the latter having the preference. If the furnace area is too small, there will be an excessive wnste of fuel: (1) Because of a required longer firing period. (2) Because’ of frequent stirring of the fires and repeated clinkering to get a necessary higher rate of combustion re- sulting in a higher carbon content in the ash. An excess grate area within reasonable limits can be con- trolled to a minimum rate of combustion by longer intervals between firing periods and by allowing the clinker to accumu- late and thus reduce the free area. Comparison of Furnace Areas. A 30-foot round down-draft kiln has about 650 feet of floor area, exclusive of the bag walls. An average setting height of 9 feet will give 5,850 cubic feet. On the basis of the first rule, the area of grate surface of such a kiln for cone 1 ware will be 8.8 square feet, or nine furnaces of 10 square feet area each, reducing to eight furnaces for temperatures below cone 1. Cone 5 temperature ware will require ten fur- naces ; cone 10 temperature, twelve furnaces ; cone 26 tempera- ture, sixteen furnaces. A ratio of one foot of grate area to four feet of floor area would require sixteen furnaces, and a ratio of 1 to 8 would require eight furnaces. Either of these rules in the hands of a competent engineer, who will take into consideration the kiln tonnage, time re- quired to burn, and the quantity of fuel required per ton, will give a kiln and furnace design well within reasonable limits of a practical and economical rate of combustion. Let us consider furnace areas for a 30-foot round kiln on the basis of fuel and time given on a preceding page. We will assume 20 per cent, of the time is required for the water- smoking, during which time the rate of combustion is slow, and we will make no allowance for the fuel used during this period. The results are as follows: BURNING CLAY WARES. 149 Common Brick- — 240 Tons. No. of Coal Time Grate Area Furnaces 150 lbs. coal, 4 days 5 400 lbs. coal, 4 days 12 400 lbs. coal, 6.4 days 8 Average 8 Face Brick — 210 Tons. 300 lbs. coal, 4.8 days 7 700 lbs. coal, 4.8 days 16 700 lbs. coal, 9.6 days 8 Average . .103 sq. ft. 10 Paving Blocks- — 200 Tons. 500 lbs. coal, 4.8 days . .109 sq. ft. 11 800 lbs. coal, 4.8 days . . 173 sq. ft. 17 800 lbs. coal, 9.6 days , . . 87 sq. ft. 9 Average 12 Drain Tile — 75 Tons. 350 lbs. coal, 1.6 days . . 53 sq. ft. 5 800 lbs. coal, 1.6 days 19 800 lbs. coal, 4 days 8 Average 11 Sewer Pipe — 60 Tons. 800 lbs. coal, 4 days 6 1500 lbs. coal, 4 days , . . 117 sq. ft. 12 1500 lbs. coal, 6.4 days , .. 73 sq. ft. 7 Average 84 sq. ft. 8 The drain tile estimate is open to criticism because only small kilns can be burned off in two days, and in no case would the maximum quantity of fuel be burned in this short period of time, and the other intermediate estimates are open to the same criticism, but in less degree. These estimates bring out two facts: (1) The combustion rate of 8 pounds of coal per square foot of grate area per hour is a maximum for average prac- tice, because if we were to figure a higher rate of combustion, the minimum-fuel-minimum-time operation would require an absurdly small grate area not found in practice anywhere. The maximum-fuel-maximum-time operation gives results ap- 150 BURNING CLAY WARES. proximating general practice, being under, however, rather than over; and if we were to increase the rate of combus- tion 50 per cent, the grate area would be less than that we know to be practical. The intermediate figures would indi- cate a higher rate of combustion, but it seldom happens that the maximum quantity of fuel is burned in the minimum period of time upon which assumption these intermediate figures are based. (2) The rules given for grate areas will give the proper furnace power if one will take into consideration any unusual conditions and make allowance for them. Furnace Doors. The furnace door is the most important feature of the fur- nace, because it is the only movable feature and easily gets out of shape, with the result that large volumes of air are being drawn into the furnace around the door, and of this we have no control. An ill-fitting door wastes in fuel each burn the cost of two good doors. A furnace of the inclined grate bar type without a door is preferable to a door which does not fulfill its purpose. The former we control by banking up the coal, and if the com- bustion gases are investigated to determine the most eco- nomical firing conditions, the fire mouths can be banked to maintain this condition. Secondary air is necessary, but it is important that we have it under control, which we cannot have through an irregu- lar opening around a badly warped door or door frame. Terra cotta, pottery and abrasive products kilns all under roof and enclosed generally use a substantial cast iron frame and swinging door similar to a boiler furnace door, the door being lined with a cast iron perforated shield set away from the door, or a fire clay block may be used. Such doors are not satisfactory for outside kilns exposed to all atmospheric conditions, but it must be mentioned that the doors used in outside kilns are frequently much less sub- stantial than the doors used on the above mentioned kilns. Whether the trouble is due to a flimsy door equipment or to severe conditions, the fact remains that cast iron swing doors have been largely abandoned in the construction of outside kilns, except in scoved and up-draft kilns. We frequently see thin sheet iron plates used for furnace doors, which are in a condition to discredit a scrap pile and which attached to a kiln are wasting the profits of the busi- ness. BURNING CLAY WARES. 151 A cast iron plate, ribbed to reduce warpage, and grooved on the bottom or with lugs to slide on a projecting angle plate makes a fairly good door. The front of the furnace wall is battered so the plate retains its place by its own weight. A good door is made of a shallow cast iron box lined with thin fire clay brick. This is hung by a trolley to the top wall of the furnace front, or it may be hung by a wire attached to a hook in the kiln wall higher up, and projecting from the wall just enough to clear the front wall of the furnace. Simi- larly brick or a clay block may be bound with an iron band and hung on a trolley or by a wire as above mentioned. A large fireclay slab bound with an iron band to be raised and lowered by a chain or flexible steel rope, using a grooved pulley and counter-balanced, is an excellent door. This is the type of door used in heating furnaces. Construction of Furnace. The furnace lining should be built of the best quality of fire brick laid in fire clay mortar of the same quality as the brick, and the bed joint should be as thin as it is possible to make it. The wall should be not less than 9 inches thick and built independent of the kiln wall. It is customary in building the main kiln wall to complete and bond in the furnace wall as part of the main wall, par- ticularly the main wall lining and the furnace lining. This is essential in order to turn the inner furnace crown satisfac- torily. The furnace walls may be carried up with the main walls, and crowned without bonding the former into the latter. With such a construction, when repairs become necessary the furnace walls may be removed and rebuilt without tearing an irregular, jagged hole in the main wall, which is difficult to fill up in a substantial manner. The furnace inner crown should be a single course of arch brick and separated from the outer supporting crown by an expansion space. A 9-inch crown of wedge brick or a bonded arch has the objection that the inner ends of the wedge brick or the inner course of the bonded arch will shrink under the intense heat, and the inner points will break off, or the inner courses slip down and out, to the destruction of the crown. The outer arch is the kiln wall support, and there should be no occasion to remove it in making furnace repairs. More substantial arches can be built of special shapes, using two or at most three blocks for the skew and arch. A single brick rowlock arch over the special block arch will sup- 152 BURNING CLAY WARES. port the main wall and provide vertical expansion space for the furnace walls and crown. When furnaces project from the main wall, they should be held in place by a band or channel plate. Round kilns are usually built with a hub which includes the furnace. The furnaces thus are prevented from spread- ing by the kiln walls, but in addition there should be a band at the grate bar level. Projecting furnaces are frequently seen in rectangular kilns, and they may readily be held in place by a channel plate at the grate bar level, extending the full length of the kiln. Between the furnaces, stay rods connecting the channel to the main vertical kiln buck stays hold the channel in place and thus brace the furnace wall. It is a good plan to batter the furnace front, which reduces the tendency to draw off, and also the doors will hug the wall instead of hanging away from it as happens when the wall overhangs. BURNING CLAY WARES. 153 CHAPTER VIII. KILNS. K ILNS MAY BE classified as follows, and while the classification given below is somewhat incongrous and open to criticsm, still it has the merit of arrang- ing the kilns in some relation to the wares produced. I. Open-top kilns. (1) Periodic in operation. (2) Continuous in operation. (A) Ring kilns. ( B ) Chambered kilns. II. Crowned kilns. (1) Open fire. (A) Periodic in operation. (a) Up draft (b) Down draft. (c) Up and down draft. (d) Horizontal draft. (B) Continuous in operation. (a) Ring kilns — horizontal draft. (b) Chambered kilns — down draft. (c) Car tunnel kilns. Horizontal draft. Down draft. (2) Muffled kilns. (A) Periodic in operation. (B) Continuous (car tunnel) in operation. The first general division applies to common bricks, hollow bricks or blocks, and drain tile in small degree. The second general division includes all other types of kilns and the first subdivision covers a broad field and in- cludes every line of ware from common bricks to poreclain. The white wares, however, are inclosed in saggers which are multiple muffles, and we might properly duplicate several of 154 BURNING CLAY WARES. the types in the first subdivision under muffled kilns, which would leave the first division applicable to common wares — bricks of all kinds, fire-proofing and other unglazed hollow ware. It would also include salt-glazed wares and slip-glazed ware such as stoneware — in fact, any ware that can be burned in contact with the fire gases. The second subdivision, then, would cover terra cotta and other sensitive glazed wares, white ware, art pottery, etc. Periodic Open-top Kilns. The rectangular open-top periodic kilns, known as “scove,” “clamp” and “up-draft” kilns, are suitable for common bricks only. It was in such kilns that the ancient clayworkers burned their brick hard, as they were commanded to do, and practically the same kiln is used today in the great centers of the common brick industry. The term scoved kiln is applied to the mass of bricks in- stead of to the structure inclosing the bricks, in that we indi- cate the location as the “kiln yard” or “kiln shed” or “kiln of bricks.” Indeed, the scoved kiln hardly justifies a name being given to it as a kiln structure. The bricks are set in rectangular form, with fire arches, etc., and then cased with bricks, sometimes unburned bricks, and the casing daubed with mud. When the kiln is burned, cooled, and shipping begins, the casing is thrown down, or sometimes loaded out with the burned product, and the ground is cleared for a second kiln. In some instances, especially where coal is the fuel and grate bars are required in consequence, the casing wall is made permanent to a height w T hich includes the fire mouths and arches, and the usual casing starts on top of this wall. The ordinary casing is eight inches thick up to two-thirds of the height of the setting, then drops off to four inches. Frequently the bottom of the casing wall is twelve inches thick to the top of the fire mouths, then drops off to eight inches and finally to four inches. The casings are laid up dry, then daubed to prevent or reduce the air leakage. When the kiln is finished and ready to fire, the top is completely platted with bricks laid flat. The bricks in the first course of platting are placed tightly end to end, but the rows are spaced so that each course of the top platting, with its bricks at right angles to those in the first course, will center on the rows in the under course. BURNING CLAY WARES. 155 After the platting is in place, or as it is placed, each alternate third or fourth brick is raised to provide draft. For conven- ience in lifting the platting, and particularly in replacing (“tightening”) it after the “heat is through” to the top, it is an advantage to set on edge each alternate brick in the top course which gives a finger hold on the projecting edge. Usu- ally, however, the alternate loosened bricks are not lifted out, but instead one end is raised and the brick slipped endwise, and it can be hooked back into place when the time comes to tighten the platting. The cost of casing and platting exclusive of the bricks used varies from $0.15 to $0.30 per thousand bricks set in the kiln. This is a serious item in the cost of the product, and the ques- tion may be asked : How can such kilns compete with other types of kilns? The cost of installation is very small in comparison with other kilns, thus eliminating interest, depreciation and up- keep, and the capacity per kiln is unlimited, varying from 100,000 or less to a million or more bricks, usually ranging between 250,000 and 700,000 bricks. They are more economical in fuel than other types of periodic kilns. The kilns are placed alongside the loading tracks or docks, giving short runs for loading, especially in view of the fact that the bricks may be taken from any point along the kiln to the car or boat placed opposite such point, whereas in per- manent kiln structures the route must be through the kiln entrance. The kilns being open and rectangular, the setting and draw- ing can be done at a minimum cost. Finally, because of the small cost of the kiln — in fact, noth- ing more than the cost of any storage space — the bricks may remain in the kiln until they are sold, thus saving a double handling. The bricks are set in benches, the legs of which have a width of 2 y 2 bricks (21 inches) or 3 bricks (26 inches) or 3 y 2 bricks or 4 bricks, with a space of 12 to 16 inches between the legs for the fire arch. The height of the setting is com- monly 42 courses of bricks on edge, but in some instances 54 courses are the setting height. The former height is estab- lished by the ability of a man (tosser) to toss the bricks in units of 4 or 5 bricks to the setter. Higher setting requires an extra man standing on a lower bench to receive and toss the bricks to the setter working on an upper bench. The av- 156 BURNING CLAY WARES. erage 3-brick bench kilns, set 42 courses high, hold about 20,000 bricks in each bench, varying from 15,000 to 30,000, depending upon the width of the kiln, and higher setting will increase the quantity correspondingly. The length of the kiln is simply a matter of convenience or yard arrangement and is designated as so many benches, arches or “eyes.” In the large common-brick centers the kiln shed is a thou- sand or more feet long. The kilns are set longitudinally in the shed and the coal and shipping track are parallel to the kiln on each side either under the shed or just outside. Where the shipping is by water, the kiln shed and kiln are close to the dock and parallel to it. The setting begins at one end of the shed, although not essential, and depends upon whether the old stock has been moved. As soon as sufficient benches to constitute an econom- ical firing unit (in one district four arches are the unit for one burner, day and night, in another five arches) are set, the end of the kiln is closed, cased and daubed, and the fires started. The sides of the kiln are cased as the setting pro- gresses. A second, third, etc., kiln continues the setting to the other end of the shed. Meanwhile the first kiln will be burned, cooled and loaded out, and the setting returns to the original starting point. If the market is good the shed will be kept clear for the setting until late in the fall, in some instances throughout the winter where winter work is possible. If the bricks are not moved, the season closes when the setting overtakes the drawing and the shed is full. This is the common-brick con- tinuous kiln operation, but the individual kiln is periodic and not a continuous kiln, as we ordinarily and properly des- ignate continuous kilns. The arrangement of the kilns depends upon the layout of the yard. Sometimes the kilns are endwise to the loading track on one side and to the factory on the the other side, and some plants — partly rail shipping and partly direct delivery by wagons, or all wagon delivery — have the kilns distributed around the factory on two or three sides. If scoved kilns or the permanent wall up-draft kilns pro- duced as good a product as other types of kilns, they would find a wider use and would be difficult to displace. Unfortu- nately, they do not give the results. The arch bricks, which BURNING CLAY WARES. 157 in effect are a continuation of the furnace walls, are checked, spalled and blackened by the fire on the end exposed to the fire, and where the shrinkage is considerable they are wedge- shaped. The top bricks, often several courses deep, and the sides and ends are almost invariably soft, and thus the per- centage of high-grade, hard, uniform-colored bricks is greatly reduced, in the face of a market continually calling for the better quality product. Kiln Sheds. We frequently find factories where the kilns are not pro- tected by sheds. The bricks are set in the open, and after the first bench is set and cased, it is covered with boards or often with canvas, which is a very unsatisfactory arrangement, re- sulting in considerable loss during inclement weather and frequent delays in the setting operations. The usual arrangement is a large shed with board roof completely covering the kiln, with lower shed roofs on either side to cover the fire pits. The main shed roof has a monitor in the center to take away the gases and watersmoke escap- ing from the top of the kiln. Unless the shed roof is high above the kiln, it is necessary to remove the roof boards dur- ing the high fire stages, particularly toward the latter period of the firing, when the top courses reach the finishing tem- perature. Such a shed is illustrated in Fig. 46. Setting. It would too greatly extend this chapter to describe and illustrate in detail the various settings of scove kilns. Each 158 BURNING CLAY WARES. district has some special feature in some part of the setting which may be an advantage with the fuel used and the prod- uct burned, but which would be of little or no value in other districts. Figs. 39, 40, 41 and 42 illustrate common-brick setting in one district. The end bench is usually one and one-half brick. All the bricks forming the arch are headers. The first course is set single or in pairs, as shown, spaced and lined up to a straight edge on the arch side. This is backed up by two stretchers to carry the tie when the step-off course is reached. These stretchers should not be placed tight end to end, but instead should be separated about 2 inches to form vertical flues, and care should be taken in setting these bricks that the flues con- nect with the spacing between some of the header courses, in order to get fire to the end wall. The second, third, fifth and sixth courses are set headers, in pairs, and spaced as shown in Fig. 40. The fourth and sev- enth courses are set tight. On top of the seventh course each course is projected into the arch space, which is closed on the twelfth or fourteenth course, depending upon the width of the arch. The eighth, ninth, eleventh and twelfth courses are in pairs, and spaced. The tenth and thirteenth courses are tight. This setting, it will be noted, closes the top of the arch tightly, as shown in plan, Fig. 41, and the flames cannot take a direct course upward through the top of the arch, but instead are deflected to either side and rise through open checker work. This method of setting just above the arch closure course is an excellent one, but not common practice. Where it is desired to have some draft upward through BURNING CLAY WARES. 159 the top of the arch, the closure is on the fourteenth course, which is spaced, and the tight courses under it not being in touch end to end, provide draft spaces. The end bench, from the seventh course, is built up ver- tically on the outside by means of skin tied bricks (set at an angle), stretchers and headers, or broken bricks, as may work out best. When the closure course is set on the end bench, the second bench is started. It is considered best not to have the benches in whole brick lengths — two, three or four, as the case may be — but instead to make them two and one-half bricks, three and one- half, etc. This permits the use of a single or double stretcher in each course, which, should there be any tendency for the flame to draw through from one arch to the next, serves as a baffle to deflect the gases upward. Experienced brickmakers consider this important, but it is a question. (Very satisfactory re- sults have been obtained by a setting which we will describe, the purpose of which is to provide flues connecting the sev- eral arches.) The arch facings are a duplicate of the end bench, and the space between them is filled with stretchers and headers as may bond to the best advantage, and all spaced either singly or in pairs. When the second bench is brought to the level of the end bench, the setters stand on the second bench and the setting is carried to higher levels, usually in three-brick benches, which make a good working width, and, when bonded in, form a substantial column. From about midway of the height to the top, the end bench and sides are drawn in to form a battered wall. The end and sides out to the face of the setting are built up first in simple checker work, carefully bonded, leaving a rectangular space in the heart of the kiln. This central space is sometimes filled with all skintled set- ting, as shown in Fig. 42, the advantage of which is rapid work. A skillful setter can toss five or more bricks into place and separate them at the same time. Checker work, Fig. 39, the bricks being set singly in alternate courses of headers and stretchers, is common practice. In such method the setting is built up in blocks three brick square, and the setter can work from the front in setting header courses and 160 BURNING CLAY WARES. from the side in setting stretchers, and thus does not have to bend his wrists, and can handle four or five brick at a time, as in the skintling method. Where the bricks have a ten- dency to kiln mark, or become discolored by the flame, and it is desired to get a higher percentage of first quality facing bricks, the headers are set in pairs two courses high, and faced, alternating with double stretcher courses similarly set, or a single stretcher course. Figs. 43, 44 and 45 show a special setting mentioned above. i =j j 13 iO Figure 43. The arches are faced with header courses, as in the first method of setting, as shown in Fig. 44, set singly or in pairs. In this we find considerable difference in practice, some local- ities setting the spaced arch bricks in pairs, single courses, others in pairs, double courses, faced, others single bricks in single courses, others single bricks in double courses, faced, and others a combination of single and double courses. The feature of the setting in Figs. 43 to 45 is that the interior of each leg is set solid with stretchers from arch to arch, but across BURNING CLAY WARES. 161 the kiln each block of stretchers is spaced about iy 2 inches from the next block and the setting is worked out to connect these flues with the spaces between the header courses in the arch facing, thus connecting each arch with the adjacent arches. The setting above the arches is such as to give a maximum quantity of face bricks. 162 BURNING CLAY WARES. In some districts the two top courses under the platting are set closer than the usual spacing, the purpose being to check the flow of the gases at the point of escape and give greater opportunity to recover the heat from the gases. This is the purpose of the platting, and the closer setting of the top two courses increases the effect of the platting. In some yards it is customary in the later stages of the burning to cover the top of the kiln with dirt or ashes to hold back the heat, and such covering results in a harder burned product on top, but the work of putting on and removing the dirt, and the dirt and dust sifting through the bricks when they are being drawn, offsets the value of the harder product and the use of such dirt covering is not common. However, when a hot spot appears on top, indicated by ex- cessive settling, such spot is covered with dirt to check the draft at that point and to drive the heat to the surrounding cooler areas. An up-draft kiln, illustrated in Fig. 46, is an open-top kiln, virtually a scove or clamp kiln, inclosed in heavy, permanent walls. The advantages over a clamp kiln are : (1) Saving in labor required to scove and daub a clamp kiln. BURNING CLAY WARES. 163 (2) Less radiation loss. (3) Permanent furnaces and in consequence a better type. (4) The platting bricks may be piled on top of the per- manent walls, whereas in scored kilns they must be tossed up and down each burn. (5) The heavy walls hold the set brick in place, elimi- nating the racking back to batter the heads, and there is no danger of the mass of brick spreading to a dangerous degree, drawing outward, and sections falling, thus exposing the burn- ing bricks, as occasionally happens to a scoved kiln. There is, of course, the cost of installation, but this is quickly offset by the saving in labor. The necessity of removing the brick through the doorways in the ends of the kilns is a disadvantage in some situations, in that it increases the average distance the burned bricks must be moved, and likewise increases the average distance the unburned bricks must be moved in getting them from the drying hack to the setting face. The most general arrangement is to place the kiln longi- tudinally between the dryer and the loading tracks. The av- erage distance the ware must be moved within the kiln walls is half the length of the kiln, whereas the average distance in the scove kiln is half the width. Where the burned product is loaded into wagons which can be backed into the kiln to the working face, the increased distance is of no consequence. Where the loading is into cars or boats, portable gravity carriers can be used in moving the burned product, which practically eliminates the item of dis- tance cost, and the advantage of the scoved kiln is then almost negligible. The setting in an up-draft kiln is the same as a scoved kiln, except that the outside is plumb and in close touch with the kiln wall. Frequently the outside courses are set 2 on 1, to give greater flue space near the walls. The setting varies from 3 on 1 (three bricks on edge in the length of one brick) to 11 on 5, depending on the length and thickness of the bricks, the usual setting being either 3 on 1 or 8 on 3. Three on one is close setting, and more open setting ranges through 8 on 3, 5 on 2, 7 on 3, 11 on 5 and 2 on 1. The general setting is seldom more open than 8 on 3, and the wider settings given are only used in parts of the kiln where the burning has shown that draft is needed to get uniform results. The ad- 164 BURNING CLAY WARES. vantage of 3 on 1 or 8 on 3 setting is that the work can be carried up in three brick benches over which the setters can reach nicely. Sizes of common bricks vary widely, and the setting varies accordingly. The standard for common bricks adopted by the N. B. M. A. is 8 : *4x4x2 : ‘4.* The New England and Hudson river bricks range between 7%x3%x2 and 8x3%x2 y 8 . The Western and Southern product runs as large as 8%x4 1 / 4x2 y 2 . Furnaces and Burning. Scoved kilns were originally burned with wood and the furnaces were merely openings through the wall, connecting with the arch. As wood became scarce and the use of coal was necessary, a short, flat grate was introduced, and to get the heat to the center an occasional stick of wood was pushed into the arch as near the center as possible, or lumps of coal were thrown in over the grate fires to the center. Thus part of the combustion took place on the grate bars and part on the solid kiln floor. The up-draft kilns with heavy walls give opportunity for permanent furnaces of a better type. The most common is a simple flat bar box furnace with a door, ash pit and throat connecting with the kiln arch, the grate bar being level with the kiln floor, and the coal storage and firing pit depressed 18 inches to 24 inches to the level of the ash-pit floor. The furnaces are 12 inches to 16 inches wide and 36 inches to 48 inches long, and are spaced to correspond with the arches of the kiln — 36 inches to 42 inches. There is a differ- ence of opinion in regard to the. furnace throats. Some hold that the throat should be greatly reduced, as small as 6x6 inches, to introduce the combustion gases as a jet, which serves to project them to the center of the kilns. We have, however, seen excellent results with furnace throats the full size of the furnace and practically the size of the arch. We are of the opinion that the experience and intelligence of the burner has more to do with the results than an open or re- stricted throat. We frequently find a single furnace on each side of the kiln, arranged for two arches, as shown in Fig. 47, and the ♦Footnote. — Since the above was written the N. B. M. A. has adopted for common brick the size, 8x2 *4x3% inches. For face brick, the size, 8x2 1 / 4x3% inches. BURNING CLAY WARES. 165 flat bar coking table furnace, previously illustrated, has been extensively used in up-draft kilns. Sometimes three arches are connected with a single furnace in a similar manner. Considerable skill in burning is required to get satisfac- tory results from up-draft kilns. It must be borne in mind that the mass of set bricks is the stack, and if any portion becomes unduly heated the draft will be to this point. Chim- neys of over-burned, frequently fused bricks, surrounded by masses of under-burned bricks, are of frequent occurrence, and any burner who can get uniform results throughout the kiln should be kept on the job. If the kiln is wide and the draft weak, the heads will become unduly heated and the center cannot be brought up to a finishing heat. The restricted Figure 47. throat, above mentioned, is used to overcome this difficulty. Where wood is used, crossfiring is common practice. When the kiln is ready to fire, the fires are started on one side only, for a short period. This serves to heat up the brick along this side, and after this is accomplished the furnaces are closed by sheet iron from plates and daubed. The firing then is from the other side for a period of four to six hours. The strongest draft is on the side first fired, and the tendency is to draw the gases through the arch to this side, thus heat- ing the arch the full width of the kiln, but in spite of this the side next to the fire rapidly gains and becomes the hottest. After about six hours the fires are reversed. Two to four days of such cross-firing gets off the watersmoke and heats up 166 BURNING CLAY WARES. the kiln from side to side, with the heads slightly in advance of the center. Then both sides are fired, some wood being pushed to the center to maintain the desired condition until the heat works to the top and the kiln is finished. In some instances the cross-firing is continued to the end of the burn. Cross-firing is not used in coal burning, and in some op- erations a dead wall is set in the center of the arch to insure independent control of each furnace. To get independent control of the center and heads, we have seen furnaces con- BURNING CLAY WARES. 167 structed as shown in Fig. 48. There are two throats, one di- rect into the arch and one into a flue under the arch floor with an inlet six or more feet from the kiln wall. These throats may be plugged with loose bricks as the condition of the kiln requires. Another method of getting the heat distributed through- Figure 49. Figure 50. out the arch, in common use in some sections, is a long grate bar extending ten or more feet into the arch on each side as shown in plan in Fig. 49, each grate of the same length, or alternate long and short grates, as shown in Fig. 50. A longitudinal section of such grates is shown in Fig. 51. The grates are straight, flat bars, full length, and set on edge, or short sections of cast gridiron grates. The 168 BURNING CLAY WARES. long bars are supported by notched bearing bars, as shown in Fig. 52, which span the ash pit and are spaced about three feet apart. The ash pits must be kept clean, otherwise the bars and supports will quickly burn off. This is done with a long-handled swivel scoop, as shown in Fig. 53. Kilns have been constructed with the furnaces projecting into the kiln as shown in Fig. 54, and one design has an open- ing in the top of the furnace arch, as shown, besides the regular throat into the kiln arch. This type of furnace, with- — t r + Figure 51. I ■■■■■■■■! fill oar Figure 52. out the top opening, found some use in Pennsylvania with anthracite coal for fuel. The heat conducted through the furnace crown aided in bringing up the heads and keeping them in advance of the center, which is important, but the objection is the bad setting in consequence of the permanent non-setting bench supporting part of the mass of bricks. A crude pit furnace shown in Fig. 55 is used in one or two installations and is said to give excellent results, but that it should do so is almost inconceivable. Natural gas, oil and producer gas are readily applied to BURNING CLAY WARES. 169 ■up-draft kiln-firing, as one can see. The pressure behind such fuel serves to drive the gases to the center and thus get heat throughout the arch. Mr. J. D. Pratt, in a paper before the Wisconsin Clay Man- ufacturers’ Association, gave brief instructions for firing an up-draft kiln, from which we take the following excerpts : “We use double platting, put on in checker-board fashion, and raise every fourth brick, start wood fires in the ash pit with the ash pit wide open and the furnace doors closed, and continue this firing until the kiln has a good draft, then raise fires onto the grates. Fire with smokeless coal, with ash pits and furnace doors wide open. After the kiln is dry, com- mence to heat up with the coal regularly used in burning. Increase the heat gradually, keeping the ash pits open and the furnace doors closed, except about an inch for air and over- draft. If the center is hard to get, set it more open. After the kiln is hot all over the top, and the platting down, com- mence the hot firing. It is important to admit the proper amount of air and no more, and this is what the burner has to look after. Keep the grates and ash pits clean so the fire shows bright in the ash pits. Keep the furnace door open three-quarters of an inch to one inch. Keep the grates cov- ered with not more than three to four inches of coal, and fire light and often. Clinker with the furnace hot every six hours, and oftner if necessary. Never open the door to drive the heat to the center. Keep the furnaces as hot as the arches will stand and keep even fires. Summarized, Mr. Pratt’s instructions are : Use judgment, give the operation careful attention, keep the fires in good condition and at their maximum efficiency, and go slow. The summary will apply to any firing, but the specific directions do not apply. In the Hudson river district it is common practice to “fol- low up the watersmoking” — in other words, do not wait until the steam is off before raising the heat, but to drive the heat 170 BURNING CLAY WARES. to the maximum temperature under the steam. Such practice would ruin the product in many districts. One operation uses very heavy firing at compartively long intervals. The effect is distillation of a large volume of gas following the firing, which is carried into the kiln and burned in contact with the ware to whatever extent it is possible to get the necessary oxygen. Some oxygen will come through the furnaces with the gas, some from leakage through the Figure 54. walls, and some from reduction of the minerals in the clay. As the fires burn down the temperature advances, then be- comes stationary as the air excess increases, resulting in oxi- dation. The result of this reduction, smoking, heating, oxi- dizing process, is a chocolate colored product, and the burns are quite uniform and hard throughout the kiln, without an excessive fuel consumption. Such firing, however, would ruin BURNING CLAY WARES. 171 the arch bricks made from other clays, and with the arches down the product above them could not be satisfactorily burned. Firing is a problem of the clay as well as the fuel, and unless one has specific knowledge of both, he cannot give de- tailed directions in regard to the firing. Having determined the proper method for a particular operation, the economy in fuel and the character of the results depend upon the care and intelligence given to the work. It may be mentioned that attention should be given to the entire kiln and not to the furnaces alone. Cold spots must be looked for and worked out, by regulation of the fires, by the use of wood or coal in the arches under the cold spot, or by freeing the draft through the platting, or checking the draft around the spot. Hot chimneys, if taken in time, may Figure 55. be checked by covering the platting above them with dirt, thus diverting the draft to the surrounding areas. The speed of the burning depends largely upon the safe burning range of the clay. If the range is short, we must take a longer time to work the heat to the top, whereas with a wide range we may carry a higher temperature in the arches, resulting in rapid heat absorption throughout the kiln, since the rate of such absorption varies with the differ- ence in temperature of ware and gases. Coaling. Mixing coal screenings with the clay is a common practice in several districts, and anthracite and coke are best for this purpose. Bituminous coal may not be used, at least not gen- 172 BURNING CLAY WARES. eraliy, but lignite is practical. The anthracite and coke con- tain very little volatile matter, and in consequence do not cause bloating. Bituminous coal is not only highly gaseous, but it burns rapidly and develops a high temperature, result- ing in fusion of the particles of clay in contact with each granule of coal before the coal is fully consumed. The gas generated by distillation, or combustion, or reduction of the clay minerals, becomes entrapped in the fused mass, and its expansion under advancing temperature forms blebs, the com- bined effect of which causes bloating. Lignite is used for coaling, but we do not know how gen- erally it may be used. It is more gaseous than bituminous coal, and is rapid burning, but usually, being high in ash and moisture, it does not develop the heat that we get from bitu- minous coal, and unless there is fusion of the clay mass, bloating does not result. Likely, lignite could not be used in a clay which fuses at a low temperature. Sawdust has been extensively used, but more to develop porosity than to aid in the burning. It, too, is highly gaseous, but does not cause bloating. In England, we understand, bricks are sometimes burned by the coal mixed with the clay and setting the bricks with coal dust. The mass is cased up and ignited, and the problem would be to regulate the air admission to insure a combustion rate sufficient to develop a needed temperature, and not to exceed such temperature. It applies, to burning bricks, the principle of charcoal burning. In this country coaling is used only as an aid, and it is supplemented by furnace firing. The amount of coal varies in different districts between 50 and 100 pounds per thousand bricks. In one district the amount used is 66 pounds per thousand bricks. Double-coaled bricks are set around the heads, over the top and in the casing, where green bricks are used to scove clamp kilns. The term “double coaling” does not mean twice the amount used in single coaling, but simply means a larger amount, which varies from several times to ten times the amount used in single coaling. The coal is usually added to the clay in the pug mill or soak pit, but we have seen one operation where the clay is piled in the clay shed to a certain depth, then covered with the required depth of coal, followed by a second lot of clay, and the mass is then cut down and fed into the preparing machinery., BURNING CLAY WARES. 173 Machine Handling and Setting. A thousand bricks will require 6,000 to 7,000 pounds of clay. This must be dug, prepared, and manufactured into bricks. These bricks contain approximately 1,500 pounds of water which must be driven off, requiring an average of 300 pounds of coal. The dried bricks must be set in the kiln, followed by burning, which requires 500 or more pounds of coal, and finally the product is removed from the kiln and loaded for shipment. There are six handlings — taking off the green bricks and placing them on cars, tossing, setting, drawing (which involves tossing), loading on barrows, loading from barrows to cars. Two bricks at a time are handled in the first operation, two to five in the second and third opera- tion, four to six in the fourth, fifth and sixth operations. Consider a stiff mud factory with a kiln shed 1,000 feet long at the end of which is a dryer 120 feet long, and beyond this an average distance of 80 feet to the machine. The bricks are handled on cars in units of 400 to 700 and the average dis- tance traveled to and into the kiln and return will be around 1,500 feet, including four transfers. The burned bricks are handled in units of 100 to 120 an average distance of 150 feet, including return, or 750 feet for 500 bricks, making a total distance of 2,250 feet for each unit of 500 bricks. For each 1,000 bricks, we handle three tons of clay in small units six times and travel a distance of one-half to three-quarters of a mile. We also move a quarter of a ton of coal and shovel it twice or three times. It is not surprising that the profits are small at the pre- vailing prices. It is surprising that we have been so slow in adapting mechanical devices to do the greater part of this work. A majority of yards today operate with dryer cars, carts and wheelbarrows, by hand and horse power. The work is heavy and the wages paid are low, compared with other in- dustries and in consequence brick yards have difficulty in hold- ing sufficient labor to do the work in times of prosperity. A number of larger yards use electric power to move the product from the dryer to the kilns, and the electric truck is being tried out for the movement of the burned product. Mr. Lemon Parker, in the Transactions of the N. B. M. A., gives the following data relative to handling a variety of burned clay ware : 174 BURNING CLAY WARES. Distance, ft. 501 285 420 Barrows. ft. Cost, per ton. Distance, ft. Electric Trucks. Cost, per ton. $0,256 0.129 0.197 557 483 478 537 504 $0.16 0.121 0.145 0.179 0.176 av. 402 $0,194 average 512 $0,156 The barrow work for 512 feet on the same basis as for 402 feet will cost $0,247 per ton. The use of electric haulage will lessen the cost of moving bricks, but does not touch the several handling operations. A portable elevator has been used to elevate the cars of bricks in the kiln to the setting level, thus reducing in some degree the setting labor, but its particular advantage would be in higher setting. It would eliminate the tossing in so far as a setter could reach the bricks on the car without changing his position. When the bricks on the far end of the care are be- yond reach, it is better to have a tosser than to require the setter to walk back and forth in order to reach and set the bricks. Furthermore, a setter can place more bricks when they are tossed to him than he could if he had to pick them from the car. There would be some gain in labor, particularly if the setting was of such height as to require an intermediate tosser, but considering the time required to move the eleva- tors, etc., the net gain would not be large. The Scott System, installed in a number of yards, extends the take-off belt (from a stiff mud machine) the full length of the kilns, in one instance over 800 feet. Opposite each kiln a cross conveyer carries the bricks into the kiln. One man can transfer 50,000 bricks per day from the long take-off belt to the cross conveyor. Take off and dryer transfer men are thus eliminated, except the one man above mentioned. The bricks are set in the kiln 6 to 8 courses high all over the kiln floor. The conveyor is centered longitudinally in the kiln and within reach of the conveyer, the setters can pick the bricks from the belt and set them, but when the setting is be- yond reach a tosser is needed. Whether there is any gain in the tossing and setting labor by this method over the ordinary method may be questioned, but there is a decided gain in the pace set by the machine. The bricks have to be set as fast as they are delivered to the kiln, or if not they are carried through the kiln and dumped off from the end of the conveyor, BURNING CLAY WARES. 175 at the end of the kiln outside, where they are mute evidence of slack work or insufficient setting force. Men, as a rule, do not like to be “snowed under” or “buried,” as tl saying is, and will work harder, within a reasonable limit, to have a clean record at the end of the day, than they will where the onus of a bad record may be shifted to any slacker in the crew r . The bricks so set are dried at night in the kiln, or two nights and a day may be used, if the setting is in two kilns, alternately. The second day, the conveyer is raised to permit a second setting of 6 or 8 courses on top of the first setting, and third, fourth and fifth settings follow, in the same man- ner, until the kiln is filled to the top. It was intended to use the same conveyer system in re- moving the burned bricks from the kiln, but this does not work out well, as one can readily see. The first shipments from a kiln would be all soft top bricks, followed by practically all hards, and finishing with all arch bricks. Such shipments do not properly fill the orders. The Fiske System, introduced a number of years ago in New England was a radical departure from established meth- ods. The bricks (stiff mud) were racked in unites of 1,500 on forms on the factory floor at the take off belt. An electric traveling crane picked up the mass of bricks, by means of a lifting rig equipped with a series of fingers spaced to corre- spond v 7 ith the setting on the forms. Each unit was lowered by the crane into a dryer, then conveyed to the kiln, and set in place by the crane, and after being burned, it was the in- tention to similarly remove the bricks from the kiln and con- vey them to the shipping station. This system required an open-top kiln, or one with a removable crown, as in the initial installation. One crane man and one helper sufficed to put the bricks into the dryer and thence into the kiln. The crane was also used to handle the dryer and kiln covers and bring in the kiln coal. Following the Fiske initial installation, a somewhat similar system was developed, in the West, for dry pressed bricks. The bricks were handled in smaller units and taken from the press to and into the kiln on a single rail overhead trolley (telegraph track) and a kiln crane was used to place the bricks. The outcome of these systems is the machine setting, used chiefly in Chicago. The bricks are placed on cars from the 176 BURNING CLAY WARES. take off belts in units of 600 to 1,000 bricks, and the car setting corresponds to the required kiln setting — arch and bench units, head units, casing and regular setting units. The cars are taken from the dryer to the kiln on transfer and kiln tracks, in the usual manner, except that motor trams are used for the haulage, handling several cars at each trip. A traveling lift- ing crane with a fingered lift picks up the units from the cars and places them in the kiln in the proper place. A simple crane handling in use has the usual car system, delivering the cars of bricks to the kiln, where they are picked up by a crane and placed at the setting face convenient to the tossers and setters, the setting being done by hand. Platforms with turn tables are placed in the kiln at intervals convenient for the setters. The cars of dried bricks are placed on these platforms, and in setting the bricks, as soon as one side or end of each car is emptied the car may be turned on the table, bringing the other side or end within reach of the setters. After the bricks are burned, the kiln is emptied by piling the bricks on platforms which, when loaded, are picked up by the crane, carried to the car and dumped, the platforms being returned to the kiln. The tossers, both in setting and unloading the kiln, are largely eliminated, and the wheelers, from the kiln to the car, are replaced by one crane man and a car trimmer. The up-draft kiln, or any open-top kiln, is especially adapted to machine operations. Open-top Continuous Kilns. The advantages of the periodic clamp and up-draft kilns have been pointed out — low first cost, sanitary conditions, large capacity, economy in fuel, availability for storage, and adaptability to mechanical devices for handling the product. The objections are the commonly large percentage of soft bricks, the frequently inferior arch bricks, and the excessive loss in broken bricks. There is an insistent demand for an open-top continuous kiln, particularly in the South. Such a kiln is comparatively low in cost ; it is as sanitary as the up-draft kiln, which is an important feature in hot climates ; the capacity, single fired, is limited to from 30,000 to 50,000 bricks per day, but this is sufficient for the average plant; being regenerative, the kiln is more economical in fuel than a periodic kiln ; it is adapted BURNING CLAY WARES. 177 to crane setting and drawing; it gives a larger percentage of hard bricks and largely eliminates arch bricks. Several years ago an attempt was made in New England to operate a scoved kiln as a continuous regenerative kiln. There were no arches across the kiln, but instead the bricks were set to form a series of longitudinal flues in the bottom as in a crowned ring kiln setting and vertical flues extended from the trace flues to the top of the brick setting. These vertical flues were spaced about 42 inches longitudinally and transversely, and were used as feed holes. The kiln was cased and platted in the usual maimer, except the tops of the feed holes were provided with suitable caps. At intervals of approximately 16 feet corresponding to sec- tions of a continuous kiln, were underground transverse draft flues connected with a lateral main (fan) draft flue outside m Figure 56. the kiln. The draft flues within the kiln area had graduated openings in the kiln floor. The operation was identically the same as that of an open- top continuous kiln or a crowned ring kiln. It could hardly be expected that such a radical departure from the usual scove kiln operation would result in a perfect burn in the first attempt, but as far as could be determined from an examination of the product in both types of kilns on the same yard, the advantage was with the continuous opera- tion. The burn was lacking in hardness but otherwise was better than that in the regular scove kiln, and undoubtedly had the test been repeated, there would have been improve- ment in the results. 178 BURNING CLAY WARES. It is but a step from such a crude continuous operation to the open -top continuous kilns which are successfully operated in Europe, and there are a few successful installations in this country, 'besides several failures. A kiln principle is not to be condemned because of initial failures to adapt it to American conditions, otherwise we would not now have in operation the several hundred continuous kilns of various types. There have been a number of disappointments and several flat fail- ures in the development of continuous kilns, but in spite of these, the regenerative principle has won its way and now is a large factor in the ceramic industries. The Chmelewski (Finland) kiln, patented in this country, and offered to the clayworkers several years ago, is an open top continuous kiln. It is shown in Fig. 56 (plan), Fig. 57 (cross section) and Fig. 58 (longitudinal sections). It is sim- ply a ring kiln, without a crown. In the original kiln, the tunnel is about 12 feet wide and 10 feet high, in sections 16 to 20 feet long, each section con- trolled by a single inlet, at the floor level, into the main draft flue between the parallel tunnels. The sections are closed by sheet iron plates lowered from the top into a slot provided in the brick setting. The bricks are set to form longitudinal flues in the bottom spaced about three feet apart, and each flue is 8 inches to 9 inches wide and the height of four (Ameri- can) bricks. Cross flues, 10 inches wide and 16 inches high, and spaced five feet apart are also set in the bottom. These cross flues correspond with the arched openings in the outside kiln wall. Vertical flues spaced about three feet apart, extend from the top of the longitudinal flues to the top of the setting. These flues are about six inches square and serve as stacks BURNING CLAY WARES. 179 in the early stages of the drying and as feed holes and fire ducts during the burning. The set bricks are covered with two courses of platting and these with several inches of dirt. Openings are left in the platting corresponding with the vertical flues and these are fitted with collars and caps over the compartments under fire. After a section is set and platted, fires are started in the small arches in the outside walls connecting with the cross flues. The vertical flues open through the platting, giving the necessary draft, and the drying can be controlled by these openings. By closing those on the outer wall side and open- ing those on the inner wall side, the heat will be drawn com- pletely across the kiln, or to any degree desired, by closing the openings on the inner side and opening those toward the outer wall. When the bricks are dry enough and hot enough not to be injured by the combustion gases, not to soot, etc., the damper between the drying section and the sections subject to com- bustion gases is drawn, and at the same time, the independent Figure 59. fires are extinguished and the small arches closed. Three or more sections will be drying at the same time, the number depending upon the time required for this work, in order to keep pace with the burning. The coal firing is through the holes in the platting and is the same as any ring kiln firing — a small quantity of fuel in each hole at intervals of 12 to 20 minutes. The setting is also the same as in a ring kiln (Hoffman) except the independent drying flues. 180 BURNING CLAY WARES. The Otto Bock open top continuous kiln shown in Fig. 59 (cross section), differs from the Chmelewski kiln in that the draft flues are on top immediately under the platting, and in- stead of a single draft outlet for each section, the outlets are a series of openings about five inches square, spaced about ten inches on centers. The tunnel may be below the ground level, as shown, or on the ground level, as drainage conditions may require. The tunnels are about 12 feet wide and 5 feet high, and the bricks (German size) are set 10 courses high (about 4 feet). The kiln is fully continuous, having parallel tunnels with cross tunnels connecting the ends. The usual layout has nomi- nally, 16 compartments each about 16 feet long, but since Figure 60. there are no doorways nor flues at intervals, which would des- ignate sections of the kiln, the division into sections is simply a matter of working units. There are the usual sections — heating up, burning, cooling, drawing and setting, in connected continuous operation. After a section is set it is closed by a paper damper, and two such sections are watersmoking. The operation is as follows : sections 1 and 2 are watersmoking ; section 3, setting ; section 1 1 : OTMMHMe b b B n n e b fei "b "£ "ePsT rTi eV id” e" b b B'b 1-1-14- 14 l-i - w Figure 61. 4, drawing, and the air for combustion enters in this section ; sections 5, 6, 7, 8, 9, 10 and 11 are cooling, and the air as it passes from section 4 to section 11 advances in temperature to a red heat ; sections 12, 13 and 14 are burning, and the products of combustion pass through sections 15 and 16, which are heating up. The dirt platting from section 4 is being re- moved and placed on section 3 as the work of drawing and setting progresses. The K. W. Klose kiln is a modification of and improve- ment on the Bock kiln, and is illustrated in Fig. 60 (cross- section), Fig. 61 (longitudinal section) and Fig. 62 (view BURNING CLAY WARES. 181 plan). The draft outlets are a series as in the Bock kiln, except they are larger in size, not so closely spaced, and are placed midway in the vertical kiln wall. (Note.- — In the Bock kiln we have many small draft openings immediately under the platting ; the Klose kiln has fewer and larger draft openings approximately midway between the platting and the kiln floor ; the Chmelewski kiln has a single draft outlet for each section located near the end of the section on the kiln floor. With such extremes in existing types of the open-top kiln, one would conclude that the success of the kiln depends more upon skillful operation than on the design.) The paral- lel tunnels are connected at the ends by flues instead of con- tinuing the tunnels across the ends. The central flue is the main draft flue, with a stack or fan at one end of the kiln, and connections to this flue and the kiln outlets are by means of Figure 62. portable goosenecks. In the outer walls are openings corre- sponding to the draft outlets in the inner wall, and these con- nect with an underground advanced heat flue completely sur- rounding the kiln, the continuity of which is broken at each end by dampers. The bricks are set in the usual manner with longitudinal (“trace”) flues in the bottom corresponding to the openings in the end walls, and vertical flues connect these bottom flues with the platting feed holes. The operation is the same as in the Bock kiln, or any direct coal-fired ring kiln, the processes of drawing, setting, platting, cooling, burning and heating up being continuous and advanc- ing section by section. The initial water-smoking is independent and is done with hot air from the cooling sections. The water-smoking sections 182 BURNING CLAY WARES. are separated from the heating op and setting sections by pa- per shields pasted to the bricks. Cooling sections of the kiln are connected with the advanced heating flue by goosenecks, and at the same time the draft outlets in the inner kiln wall are connected by goosenecks with the main draft flue. The hot air from the cooling sections is drawn into the advanced heat flue and is by-passed therein direct to the water-smoking sec- tions ahead of the heating up chambers or backward around the drawing and setting sections, as may be the shortest cut, and the dampers in the heating-up flue give this control. The goosenecks connecting the water-smoking sections with the heating flue introduce the hot air into these sections, and simi- lar connections with the draft flue on the opposite side of the water-smoking sections complete the circuit to the fan. Fig. 62 illustrates the advanced heating flue for water- smoking applicable to the kiln in question, and to any contin- uous kiln, with such modifications as the kiln construction may require. A chambered type of open top continuous kiln, in principle, is shown in Fig. 63 (longitudinal section), Fig. 64 (cross-sec- tion) and Fig. 65 (plan). The usual method of building this kiln is to have each row of compartments in a separate battery with working space be- tween. The main draft flue will then be underground between the two batteries of kilns. The transfer tracks from the dryer will be centered in this working space, with stub tracks into each compartment. BURNING CLAY WARES. 133 The burned ware is removed through doorways in the op- posite ends of the compartments. The ware is set in the usuai checkers with feed holes corresponding with holes in the plat- ting, and is direct coal fired. The bricks may be set 38 to 42 Figure 65. courses high and burned hard from top to bottom. One in- stallation of a somewhat similar type of kiln uses producer gas, but the kiln is best adapted to direct coal firing. This kiln has shown very satisfactory results and extended use of the kiln is promising. Periodic Crowned Updraft Kilns. A crowned updraft kiln is seldom used in the ceramic in- dustries except in burning pottery. We see no reason for the extensive and continued use of the updraft kiln in the pottery industry, except that the operation is largely controlled by tradition. In the near past to ancient times it has been an in- dustry of secrecy, of formulas handed down from father to son for generations, of processes carefully guarded, and the up- draft kiln, adopted in the early periods of the industry, has come down to the present time along with the other features of the work. There may be good reasons for the continued use 184 BURNING CLAY WARES. of this type of kiln for this industry, but we fail to appreciate the full force of them. The fires are below the floor level and a portion of the gas passes through flues under the kiln floor and rises through a center well hole in the floor, then upward among the ware, and escapes through a vent in the crown. Low bags inside the kiln, connecting with the furnace throats, direct a part of the gas from the furnaces upward into the kiln space, and from the top of the bag the gas takes a vertical and diagonal course to the draft outlet in the center of the crown. Thus we have a volume of gas rising from the center well hole in the floor and volumes from each bag. In some instances there are a series of holes in the floor introducing the gases into the kiln. Montgomery and Gray, in Yol. XIV. Trans. American Cera- mic Society, present data relative to the fuel consumption in up-draft and down-draft pottery kilns. Gray shows a fuel consumption of about 9 tons of coal per burn in four up-draft kilns, for temperatures of cone 8, and Montgomery gives the same tonnage in a center-stack down-draft kiln, burning to cone 11. The kilns are practically the same in cubic capacity. A single instance is not sufficient evidence of the advantage of one type over the other, nor do we consider the small dif- ference in fuel as an important item. The uniformity of burn is the chief concern, and we are of the opinion that the down- draft principle will make the best showing in this respect. Watts, in Yol. Y, Trans. American Ceramic Society, makes the statement: “Any one who has studied the workings of an up- draft white ware kiln knows that the outside ring is solid-flat ware. Between this and the Center is a great area that is too soft for flat ware and too hard for hollow ware. The ware taken from this area is never right. Why can we not use a center-stack down-draft kiln, which will give uniform results, and burn only such ware in a kiln as can be all burned at the same temperature?” From an engineering standpoint, without reference to pot- tery ware, the down-draft principle has a decided advantage over the up-draft. The first objection to the up-draft kiln, it seems to us. is the cost of maintaining the under floor flues, subjected, as they are, to high temperatures and carrying the load of the ware. This objection has had effect on the development of under floor fired kilns for the common ware industry, and many up-and-down-draft kilns have been converted into down- BURNING CLAY WARES. 185 drafts on this account, although a number of them gave very uniform results. The floor of an up-draft kiln or any under floor fired type must be above the factory floor level, and all the ware has to be carried up into the kiln and brought down after being burned, thus increasing the labor cost. The radiation loss from any kiln is greatest from the crown. It is greater from the crown of a down-draft kiln, be- cause the temperature of the gases under the crown of a down- draft is higher than that of an up-draft, and, considered from this standpoint alone, the up-draft kiln would show economy in fuel, but the purpose of a kiln is, first, to burn the ware uniformly, and fuel economy is of second consideration. It must be remembered that while the radiation from the top of a down-draft kiln is greater than in an up-draft, the conduction losses on the bottom are less. The top of an up-draft kiln is most distant from the source of heat; the gases there have a minimum temperature and a maximum radiation loss. The down-draft kiln has a maximum temperature in the top to off- 186 BURNING CLAY WARES. set the radiation loss. The bottom of the kiln, where the tem- perature of the gases is the least, has a minimum radiation (conduction) loss, and besides has the benefit of the ring of furnaces around the base of the kiln. The burning of open-set ware is materially advanced by pressure. We can get hard ware under weight in the bottom of the kiln at a lower tem- perature than we can in the top of a kiln where the product has no superincumbent load. This, of course, is of no conse- Figure 67. quence in pottery ware enclosed in saggers, but it is impor- tant in burning common wares which are stacked up in the kiln, one piece on another, the load increasing from top to bottom. Finally, another difficulty of the up-draft kiln in burning common ware is that the proximity of the ware in the bottom of the kiln to the fires subjects it to an intense heat and to reducing conditions. The former intensified by the overburden carried by the bottom ware results in over- burned, distorted ware, and the latter causes discoloration. BURNING CLAY WARES. 187 We can get more uniform common ware burns in down- draft kilns than in up-draft, and the latter type of kiln is sel- dom found except in the pottery industries including abrasive products. We have seen one or two fire brick up-draft kiln installa- tions which were scarcely more than crowned “up-draft” kilns. The crown replaced the platting, and the arches were perma- nent trenches stepped over with the fire bricks to be burned. Fig. 66 (sectional elevation) and Fig. 67 (plan) illustrate an up-draft pottery kiln, after Riddle, Vol. XIII, Trans. Ameri- can Ceramic Society. Down-Draft Periodic Kiln. The down-draft kiln is the type of kiln most widely used and in it are burned the greatest variety of products — com- mon bricks, face bricks, fire bricks, paving blocks, drain tile and fire-proofing, salt-glazed conduits, sewer pipe and building blocks, stoneware, terra cotta in some degree, and the smaller lines of special ware. A few years ago it was a simple matter for an engineer to select the proper kiln principle for any industry, and his chief problem was to adapt the type selected to the particular industry. Today his biggest problem is to determine the type of kiln. Machine setting demands an open top kiln, and if the open top continuous kiln, with its economy in fuel and sani- tariness, can be developed to give the better results required by critical markets, the down-draft kiln must largely abandon the common brick field. The higher types of continuous kiln are steadily invading the pre-eminent field of the periodic kiln. The latter, how- ever, will never be abandoned, but instead will be further de- veloped and specialized to fill a very important place in the ceramic industries. Rectangular Down-Draft Kilns. The rectangular kiln has a possible larger capacity than the round kiln and permits a better yard arrangement. Kilns have been built 150 feet long, holding approximately 300,000 standard size bricks. The further advantages of a rectangular kiln are ; — the uniform setting from end to end, and wide door- ways permitting double tracks in consequence of which the setting crew is not delayed by switching cars in and out, or wagons can be backed in to the working face. 188 BURNING CLAY WARES. Multiple Stack Kilns. The difficulty in working out a plan for a long kiln is in getting an equal draft distribution, and this, in the early de- velopment of large capacity kilns, led to the adoption of mul- tiple kiln wall stacks, as illustrated in Fig. 68 (cross section) and Fig. 69 (plan). The objection to the wall stack is that Figures 68 and 69. it weakens the kiln wall, and the expansion distorts the stacks, frequently partially closing them and thus in a measure defeating the purpose of the multiple stack in that if the stacks are not alike the intensity of the draft will vary. A feature of the wall stack is that the stacks will become heated up early in the burning process, thus increasing the draft BURNING CLAY WARES. 189 for the water-smoking and oxidation when strong draft is needed and frequently not available in outside stacks. It is also claimed that the wall stacks keep the side walls hot throughout the burn and that we derive some benefit from the heat in the waste gases. This claim may be questioned. After the gases have been drawn from the kiln bottom, the chances are that the temperature increases in their passage through the wall stack and instead of giving up waste heat to the kiln they draw heat away from it. The plan shows a double row of perforated floor bricks in the longitudinal center of the kiln, but the floor can easily be made full perforated or perforated at intervals across the floor as may be desired. There is a difference of opinion in regard to the floor arrangement. For hollow ware the floor should be fully perforated with very small slots, but for bricks the semi-solid floor is considered the best. We found that with a fully perforated floor that we got better results after the under flues were filled with sand except over the center flue, and this led to the use of the plan as shown in Fig. 69. Some engineers hold that it is best to have a complete circu- lation of the waste gases under the floor, and this is our opinion, and the plan shown can be arranged for such com- plete circulation. The several floor arrangements above suggested have been tried out in actual practice, and each has its advocates. The Laubscher rectangular kiln has a solid floor with out- lets at intervals, and there is complete circulation under the floor. The main draft flue leading to the individual stacks is under the main kiln wall. The furnaces are the inclined grate bar type and the hot coals and ashes rest upon the arch of the draft flue, the crown of which is only four inches thick. The effect of this is to heat up the draft flue almost at the beginning of the firing, thus giving stronger draft in the early stages of the burning, as well as throughout the burns. The Dennis kiln is a multiple stack kiln, unique in that it is, practically, two kilns in one. Two distinct compartments are built side by side and con- nected. The stacks are in the wall between the two compart- ments, and the furnaces are in the outside walls on either side. The Yates kiln carries the wall stacks up over the crown from each side to the center, terminating in a series of stacks. Since the crown is the part of the kiln to become heated up 190 BURNING CLAY WARES. first, it would conduct this heat to the stack flues resting on the crown and thus develop a strong draft early in the burn- ing process, particularly in the last stages of the water-smoking and in the oxidation period. Rectangular Kilns with Outside Stacks. The biggest problem in designing rectangular kilns with outside stacks has been to get an equal draft in all parts of the kiln. Kilns of limited length have been built with a central longitudinal flue in the kiln bottom with a cross flue leading Figure 70. to the stack as shown in Fig. 70, and one double stack would serve two kilns. The draft in such kilns is weak at the ends, where even with uniform draft we have difficulty in main- taining the temperature. Double cross flues as shown by the dotted lines in Fig. 70 also are used and these divide the kiln into three sections. The several under floor outlets into the main longitudinal flue are, or should be, adjusted in size to give a theoretically uniform draft from the stack cross flue to the most distant points in the main kiln flue. It is, how- ever, impossible to adjust the sizes of these outlets to get an accurate distribution of the draft, but any proper adjustment helps. Kilns of these types are doing good work in the hands of competent burners who appreciate the importance of keep- ing the end fires in better condition to offset the weaker draft. BURNING CLAY WARES, 191 Too frequently, however, the burner gives each furnace the same treatment in firing and clinkering and expects the kiln to overcome his shortcomings. If it does not the fault lies with the kiln. Many plants are laid out with the stacks at the ends of the kilns, or a large single stack, or fan, for kiln control through the ends of the kilns. Sometimes the kilns have a single flue through the center of the kiln longitudinally and the draft is all from one end. That such a plan is decidedly bad needs no argument. Sim- plicity in kiln design is desirable, but such a plan sacrifices results for simplicity. A diaphragm in the kiln flue as shown in Fig. 71, serves to divert the draft from the end toward, or to, the center and gives a center draft as in the preceding kiln with stack at the side. A better end draft worked out by Richardson is shown in Fig. 72. The kiln has two central flues with perforated floor as shown in Fig. 69, and these are divided into three sections. These sections on each side are controlled by blind draft flues extending beyond the kiln wall to the stack or fan flue, and each draft flue has a damper control just outside the kiln wall. Short blind flues connect the kiln perforated floor 192 BURNING CLAY WARES. Figures 73 and 74. BURNING CLAY WARES. 193 flues with the draft flues. The kiln floor is solid except over the central kiln flues, similar to Fig. 69. This arrangement gives individual control over six sections of the kiln area, and there is full circulation of the kiln gases under the solid floor. A balanced draft kiln with a stack at each end is illus- trated in Fig. 73 and Fig. 74. The feature of the kiln is the two main under floor draft flues in front of the bag walls. These connect at their opposite ends with their respective stacks. Alternating blind cross collecting flues extend across the kiln from the draft flue, and these connect with parallel perforated floor flues. The collecting flue outlets into the draft flues are graduated in size from the stacks to the op- posite ends of the kilns to give a theoretically even draft from end to end of the kiln, in so far as is possible. Uniform draft, however, is not dependent upon this gradation of the collecting flue outlets. Assume that, in spite of the gradations, the draft is strongest from a collecting flue at the end of the kiln nearest its stack. The weakest draft then will be from the alternate collecting flue into the opposite draft flue, since it is the greatest distance from its stack. We have then the extreme end of the kiln controlled by the weakest and strongest draft in adjacent collecting flues, and the op- posite end has a like control. As we advance from one end to the other, the strong drafts on one side are becoming weaker, and the weak drafts on the opposite side becoming stronger. The aim of the design is to have the drafts equal throughout but whether this is accomplished or not, the sums of the draft from adjacent pairs of collecting flues are equal from end to end. As an illustration, represent the strong draft in the collecting cross flue at one end of the kiln by 10 and the weak draft in the adjacent collecting cross flue to the opposite side by 2, then the draft from the next collecting flue on the strong side may be 9 and the weak draft on the opposite side 3. The draft from the next pair will be 8 and 4, then 7 and 5, etc., the sum in every instance being 12. One might think that the strong draft would pull from the weak, in other words, an up-draft through the weak, and a down-draft through the strong. This, of course, is impossible, because, no matter how weak the draft from any collecting flue may be it has a big stack developing it and the strong draft flue will satisfy itself from the non-resisting gases com- 194 BURNING CLAY WARES. ing from the furnaces. The strong draft flue will get the largest volume of the furnace gases, but as it approaches its limit capacity the weak draft flue will begin to get its share. The collecting flues are spaced 36 inches on centers, so that the maximum distance from one strong draft flue, if there be such, to the next flue on the same side is 6 feet and midway between is a weak draft flue getting some of the gases. The plan shown in Fig. 74 has been modified in several ways. The collecting flues may have a perforated floor, thus eliminating the adjacent connected perforated floor flue, and getting thereby a simpler bottom. The perforated floor collect- ing flues may be spaced 18 inches on centers, thus getting a perforated floor for hollow-ware — fully perforated by lowering the draft flues and extending the floor flues to the bag wall — or they may be spaced to correspond with a three-brick bench, which is common practice in floors for brick setting. Instead of the perforated floor flues across the kiln, we may have longitudinal perforated floor flues as shown in Fig. 69 by simply introducing the main draft flues with their corresponding stacks, in place of the multiple wall stacks. Features in the Construction of Rectangular Kilns. Prof. C. B. Harrop’s article on “Kiln Expansion and Brac- ing, read before the National Brick Manufacturers’ Associa- tion and published in the Transactions and also in pamphlet form, should be in the files of every clayworker for ready reference and use in constructing kilns, and if his instruc- tions are followed, the bulged kilns which he illustrates will be less in evidence in clayworking factories. In designing a kiln one should have in mind that every- thing entering into the construction will expand except the fire clay mortar and under burned fire bricks, and further that no restraint short of crushing the materials will pre- vent the expansion. Provision for expansion should be a kiln draftsman’s slogan. The main walls of a continuous kiln are usually built solid, and if we do not provide for expansion the end walls will be crowded out and cracked resulting in excessive leakage. The best method of reducing this difficulty is by using a heavy vertical joint in the fire brick work. As the bricks expand the mortar joint shrinks, and thus we control the expansion locally and avoid the ac- . BURNING CLAY WARES. 195 cumulated effects, at least for a long period of operation. Division walls should not be bonded into main walls, but instead should be recessed into the main wall with a space for expansion at the end of the division walls. In one instance the end walls were thick and battered and the side walls were recessed into the end walls. Long flues and feather walls may have an occasional lapped or toothed joint with expansion space to take up any expan- sion not controlled by the mortar joint. It is very important that the expansion of long flues or walls connected with cross feather walls be controlled locally, otherwise the creeping of the flue will carry the feather walls, which must in conse- quence be frequently replaced. Rectangular kiln crowns are built in sections 12 feet to 16 feet long, and usually the sections are separated by a 2 inch to 4 inch expansion space. This is absurd. The kiln wall has no such joint, and the crown will move with the wall, no more, no less. It never creeps on the wall. It will crowd end- wise in the rise and may even close up the expansion joint however large this joint may be, but in doing so it will crack in one or more places between the expansion joints. The large expansion joint is a provision to facilitate cracking and serves no other purpose. The crown should be in sections to avoid rebuilding the entire crown when one section requires renewal, but there should be no space between the sections, at least not to ex- ceed y 2 inch. We have seen arches 60 feet long in one section and without a crack in them after several years’ use. Where flues extend through the main wall, such as main draft flues, an opening should be provided in the wall and arched over, but the opening should not hug the flue wall too closely. A heavy clay mortar joint between the walls of the opening and the flue walls will permit the flue walls to move should they creep any, and such a joint will crush and crumble away under the expansion and movement of the flue and thus protect the main walls. Where a stack is close to a kiln wall and connected with a long flue through the kiln we usually break the flue under the kiln wall and use a stub extension of the flue from the break to the stack to safeguard the stack. The kiln flue may enter the kiln wall opening and be broken off, then the stub flue ex- tends from inside the kiln wall opening to the stack opening. 108 BURNING CLAY WARES. Half circle arches on rectangular kilns are not satisfactory. They have a marked tendency to bulge on the quarters and flatten in the center. The crown should be the segment of a circle arch with a radius that will give a rise between % and y 3 of the width of the kiln. The best crown construction is built of wedge bricks to properly turn the circle, but such bricks have to be made especially for the work and are seldom available. Wedge bricks alternating with standards are satisfactory, but crowns should not be built of all standards where a heavy mortar joint is required to keep the bed joint radial. Prof. Harrop, in his kiln expansion paper previously cited, calls attention to crown skew backs constructed of cut skews and built up with wedge brick, and shows that the arch thrust is greater with the latter skew than with the former, and this is the only point in his article wherein we do not heartily agree. The rise of the arch is a factor in the determination of the thrust, and using the same rise for the crown circle, he has a flatter arch for the wedge skew than for the cut skew, and consequently gets a larger thrust. If we make the total kiln height the same in both cases, and lower the spring line to compensate for the higher wedge skew instead of flattening the crown circle, we will get practically the same thrust on the wedge skew distributed over a larger area. The tendency in a wedge skew is a turning movement, while that in a cut skew is to slip horizontally. In our experience, we have seen cut skews forced out on the wall in many instances, but never a wedge skew. Sand Pockets. The under floor flue system should be arranged to provide for an accumulation of sand, thus keeping the draft flues and outlets to their maximum efficiency, or provision should be made for easy cleaning. We occasionally lose a kiln of ware and the excuse is that the bottom is choked up with sand. In some yards the loss of a kiln is the signal that the bottom needs cleaning and it is baxl practice. The perforated floor flues in every case should be deeper than their outlets into the draft flues, and the depth of the sand accumulation should be frequently measured to insure timely cleaning and prevent the accumulation from getting over into the covered draft flues, where cleaning is impossible except to tear up the entire floor. BURNING CLAY WARES. 197 Kilns with central longitudinal perforated floor flues as shown in Fig. 69 and Fig. 62 are easily cleaned. In one fac- tory, the slotted floor bricks were loose and were taken up after each burn and reset in setting the bricks. If these flues are built deeper than the openings into the draft flues as shown in Fig. 68, no cleaning will be required for a period of several months to a year. Fully perforated floors are not so easily cleaned, and some types are particularly difficult. The latter should have ample pockets for a long period of operation. In some types, as Fig. 70, with arches continuing the feather walls over the draft flue to support the slotted floor bricks, the flue may be made large enough to permit a workman to enter it from the end, and the cross flues sloped so the sand will flow down into the central flue as shown in Fig. 75. We consider such a method a last resort because the large flues, particularly the arches, are liable to get out of shape and the high feather walls to draw over, and in consequence repairs to the floor are as frequent as cleaning the flues. Setting. The setting in a rectangular kiln is very simple. If the bags are straight flash walls, or segments of large circles, the setting is uniform from end to end of the kiln. Square or half circle bags require that the space between the bags be filled, but this independent of the main setting. Kilns with solid floors and limited perforated floor space must be set in such a way as not to restrict the floor outlets, and also that the spaces in the bottom courses will serve as flues leading direct to the floor slots. For example, the setting on the floor shown in Fig. 69 should start with stretcher 198 BURNING CLAY WARES, courses on the floor and they should be so spaced as to rest squarely on the slotted floor bricks in the center of the kiln. The setting for Fig. 74 would start with headers spaced to suit the slotted floor bricks and preferably a single course. Figure 76. Fig. 76 shows face brick setting and incidentally a good burn. The lower part of the kiln is set with roman bricks (114x4x12) and the upper part with standards. The latter are set 8 on 3 and care has been taken that each column, three brick square, is independent. The bricks are perfectly BURNING CLAY WARES. 199 faced as the illustration shows. It also shows the racking back from the top of the bags and the flash brick set on the step-backs to protect the bricks from the flame and to force the gases into the crown space. Brick setting in down-draft kilns varies from 24 courses high to 32 courses, and the height of the kiln for such ware should not exceed 13 feet in the center. For setting 28 courses high, as shown in the illustration, the crown space in the center is about 3 feet high before the bricks are burned and on either side the bricks are about 12 inches from the crown. If the setting is lower than this the kiln should be correspond- ingly lower and vice versa. Hollow ware, including sewer pipe, is set higher than bricks and the kilns for such are built 15 to 18 feet high in the center. Round Down-Draft Kilns. The round down-draft kiln is more widely used than any other type, and not without reason. The furnaces are uniformly spaced around the outer peri- phery of the kiln, and each controls a sector of the kiln area. The idea is that the greatest mass of the ware is nearest the source of heat, and, as the distance from the latter increases, the mass to be burned decreases to the vanishing point in the center of the kiln. In practice, however, the operation does not carry out the sector idea. The hot gases rise from the bags around the inner wall of the kiln into the space between the ware and the crown, then pass downward through the ware. Other reasons for the wide use of the circular kiln are : (1) Lower initial cost per ton capacity. (2) Simple and efficient banding. (3) Less mass in the kiln construction relative to the capacity tonnage and also proportionately less radiating sur- face, thus reducing the heat losses. (4) Greater substantiality and in consequence less upkeep cost. (5) Advantages in setting several lines of wares. A circle incloses a greater area than that of any other form having an outline equal to that of a circle, and it follows that the mass of brickwork in the kiln and the wall surface rela- tive to the area will be a minimum in circular kilns. A single band encircling the kiln wall suffices to hold the kiln together, although additional bands are desirable, and it is evident 200 BURNING CLAY WARES. to any one that a spherical or spheroidal crown is less liable to distortion than a circular segment. The advantages in setting undoubtedly are frequently im- portant factors in the selection of the round kiln. Sewer pipes, for instance, are set in a circle because the large sizes must be handled with a crane, and preferably should be equi-distant from the furnaces. The spaces between the bags (“Pockets”) are filled with smaller sizes — 4-inch to 10-inch, preferably 6-inch to 8-inch sizes, which have less tendency to topple over than the 4-inch and which will stand more severe burning treatment than the 10-inch. Inside this circle will come two rings of 6-inch to 10-inch sizes as a shield for the larger sizes. Then follow the large pipes in one or two rings, finishing the center with such sizes as can be set by hand. Drain tile are similarly set, especially where large sizes are manufactured, but smaller sizes are frequently set in parallel benches, as in rectangular kilns. Bricks are often set in circles, although this is less con- venient than in the straight benches. The advantages of the circular setting is that, if the benches of bricks are liable to roll, as frequently happens with high-shrinkage clays and those having short burning range, they cannot carry other benches with them, because circular setting distributes the tendency of the movement radially in all directions and checks it. Pottery kilns are built in circular form — in the up-draft type to get better distribution of heat around the ware, as well as up through the center, and also for uniform burning of two or more kinds of ware, since each ring is subjected to the same heat conditions ; in the down-draft because of the center stack, which is the most popular type of down-draft pottery kiln. Muffle kilns are preferably circular because of the greater substantiality of the crowns, and the center-stack type, which necessitates a circular form, is widely used. The problem of uniform draft conditions is more easily solved for the circular kiln bottom than the rectangular. We have called attention to the advantage of a rectangular kiln for car setting, namely, that two tracks may be laid in the kiln, thus keeping a loaded car at the working face, where- as, in the round kiln the empty car must be switched out before a loaded one may be brought in, and the setters are idle during this switching period. If the switching takes one BURNING CLAY WARES. 201 minute for each car the setters will be idle one hour per day in setting 30,000 bricks. This can be partially overcome if the yard arrangement is such that the cars may enter from one side of the kiln and leave from the other. There will be no interruption in the work except in filling the center space. A small advantage of this arrangement is that the cars are paral- lel with the setting face instead of end on. Another disadvantage of the round kiln for car operation may be mentioned. In a rectangular kiln the car of ware is always at the setting face and alternately on each side of the center of the kiln, thus being in. very close touch with the work, while in a round kiln the maximum distance of the car from the setting face is half the kiln diameter, and whether the ware is tossed or carried the labor cost is increased. Types of Round Kilns. We frequently distinguish kiln types by the stacks, as cen- ter stack, multiple wall stacks, multiple stacks outside the walls, and quite as often designate the type by the kiln bot- tom, as center well, radial flues, ring flue, cross-head flue, etc. The “and-so-forth” includes many kilns with distinct flue ar- rangement — a number of them good, but which have not found wide enough use to give them a name. We will not attempt to follow the development of the round kiln, but instead will simply present a few of the better known types, without taking up the numerous modifications which have been designed and patented as improvements upon the original, except such modification has features, in our opinion, worthy of mention. General Construction of Kiln Bottoms. The kiln bottom in general has three sets of flues : . (1) Perforated floor flues. (2) Collecting flues. (3) Main draft flues. The collecting flues may be on the same level as the per- forated floor flues, with connecting openings in the division walls, or they may be under the floor flues, with connecting openings in the division diaphragm. Some kiln bottoms have the flues all on the same level, being merely a system of open work over the entire bottom. Center Stack Kilns. The center stack type of kiln, illustrated in Fig. 77 and Fig. 78, is widely used in small brick and tile plants, and .with 202 BURNING CLAY WARES. modifications in the floor, is recommended for pottery. It also is extensively used in muffle kiln construction. The floor shown in the illustrations is semi-perforated and especially adapted for bricks. The perforated floor flues are adjacent to and parallel with the collecting flues and the latter have outlets through their floors into the under radial flues Figure 78. leading to the stack. The floor and collecting flues should be deep enough so .that the connecting openings in the division walls (not shown in the illustrations) need not extend to the bottom of the flues. This forms a pocket for sand in the bot- tom of the perforated floor flues, and if the slotted floor blocks are taken up occasionally and the pockets cleaned out, no sand BURNING CLAW WARES. 203 can get over into the parallel collecting flues nor into under- lying radial draft flues. If the slotted floor blocks are set loosely they can be easily removed and replaced, and it is a small matter to keep the flues clean, or if the radial flues are above the outside ground level they may be extended through the kiln wall and cleaned from the outside at any time with- out disturbing the kiln floor. For tile or other hollow ware the floor should be fully per- forated, which requires that the collecting flues should have slotted floor blocks as well as the perforated floor flues. For this type of floor the floor flues may be a series of concentric circles. The fully perforated floor eliminates the sand pockets, but by making the flues deep and protecting the draft outlets by bridge walls on each side of each outlet we get sand pockets between the outlets. Then if we use solid instead of slotted blocks in the floor immediately over the draft outlets, the sand cannot get into the under radial draft flues. The advantages of a center stack are : (1) Low stack and low construction cost. (2) Strong draft early in the burning. (3) Uniform draft conditions. The objections are : (1) Kiln space taken up by the stack. (2) Interference of the stack with the setting. The first objection is of small moment. In some wares we have difficulty in getting the center, and the stack occupies this space, thus shortening the time required to bum, or in the same time we get a more uniform burn. The second objection is serious in many operations, espe- cially where cars are used to bring in and remove the ware. It is of no consequence, however, for ware carried by hand or on trucks or barrows. A simple center stack kiln for common bricks is shown in 204 BURNING CLAY WARES. Fig. 79. The kiln has a solid floor, upon which is built the center stack, with openings in its base. The bricks are set with flues corresponding with the stack openings. In this way what ordinarily constitutes the kiln bottom becomes a salable product, and the fuel expended in heating up the floor is ap- plicable to the ware, thus reducing the cost per ton ; in other words, it increases the capacity of the kiln without increasing the size or the amount of fuel and labor required in the burning. Figure 80. Multiple Stack Kilns. The Eudaly kiln has been more widely used than any other individual kiln, perhaps, because it was introduced at a time when there were fewer types from which to choose, but it had merit and was a distinct advance over the average kiln in use at that time. It was built round and rectangular, and its chief competitor in round kilns was the center-well kiln, largely used in sewer pipe manufacture. The Eudaly kiln is illus- trated in the lower half of Fig. 80. The bottom is divided into sectors corresponding to the fur- BURNING CLAY WARES. 205 naces, and each sector is controlled by an individual wall stack. The central ring of the kiln is also controlled by a sep- arate wall stack, connected with the ring by a blind flue. The feather walls are circular and are stepped over the radial cen- ters of the sectors to form flues to the several stacks, and all being on the same level, the bottom is shallow compared with that of a center well type or any type with a double set of Figure 82. flues. The floor is fully perforated, or was in the earlier in- stallations, which was considered essential for proper heat distribution. There have been many modifications of the Eudaly kiln, to get a more substantial bottom, or a semi-perforated or solid floor, but the credit for the wide use of this type of kiln be- longs to Mr. Eudaly. 206 BURNING CLAY WARES. The upper half of Fig. 80 illustrates the change from a Eudaly floor to a semi-solid floor. If the kiln was small in diameter, only the radial flues were needed for the perforated floor, but where the diameter was large, lateral flues were ex- tended from the radial flues, forming a broken ring flue. A multiple stack kiln of a distinct type is shown in Fig. 81 and Fig. 82. There are four stacks, which may be in the wall or outside, and from each of these a branching draft flue con- trols one-quarter of the kiln. Above the draft flues are the perforated floor and collecting flues for brick setting, but, as previously noted, this type of floor may be readily changed to a fully perforated floor. The radial branches of the draft flues extend nearly to the center of the kiln and control the center and intermediate area, while the lateral branches cover the circumferential ring. A ring flue could be used instead of the angled branches, but it would require a right turn, whereas since the angled branches have an obtuse turn the draft in the three branches will be more nearly equal in consequence. Perfect adjust- ment of the kiln is possible by changing the sizes of the several openings from the collecting flues into the draft flues. The Hook down-draft kiln bottom, illustrated in a view plan, with the perforated floor removed, in Fig. 83, is unique in its arrangement and in the practical uniformity of the draft in such a simple plan. The illustration shows the entire flue system, and the flues shown are covered with perforated floor blocks. The feather walls are checker work except the top course. The circular feather walls in front of the stack serve as baffles where the draft would be the strongest, and by pro- portioning the checker work and varying the openings in the top floor, a uniform draft is claimed for the entire floor area. Single Outside Stack Kilns. The center well type of outside stack kilns, Fig. 84 and Fig. 85, should be mentioned first because of its continued ex- tensive use, particularly in the sewer pipe industry. The arrangement of the kiln may be the same as that of BURNING CLAY WARES. 207 the center stack kiln, the difference being that the waste gases are drawn off from below instead of from above the well inlets. Since there is no inside stack to be considered, the well can be made much larger and the inlets correspondingly larger or in greater number. The principle of the kiln is that the draft should be strong- est in the center because of the greater distance from the furnaces. It is remarkable what a difference of opinion there is in regard to the position of the inlets into the draft flues. The center stack and center well types have proven by their ex- Figure 86. 208 BURNING CLAY WARES. tensive use the value of strong central draft. On the other hand a kiln largely used in the west has the draft openings around the kiln wall. Here we have two extremes in success- ful operation, and from the results obtained it would be diffi- cult to determine which type is the best. The cross head flue type of bottom comes second, perhaps first, in view of its rapid development and the firm place it has in the clay industry. It has not displaced the center well type, but it has encroached largely upon the particular field of the latter, and in other lines it by far has the preference. Its simplicity and efficiency appeal to the clayworker. This type is illustrated in Fig. 86 and Fig. 87. As the name indicates, there is a main, arched draft or col- lecting flue across the kiln and from the center of this and usually at right angles to it, is the main draft flue to the stack or fan. On either side of and over the cross flue are collecting and perforated floor flues. The former enter the cross flue by openings in the side wall of the cross flue, and the latter connect with the former by openings in division walls. Vari- ous plans are worked out to overcome any lack of uniformity in the draft. In the plan shown the distribution over the kiln area is regulated by the position of the division wall openings connecting the collecting and perforated floor flues. The longest collecting flues, in the central axis of the kiln and con- trolling the largest floor area, enter the crosshead flue nearest the stack flue connection and in consequence have the strongest draft, which seems essential since they must move the largest BURNING CLAY WARES. 209 volume of gases. As tlie distance towards the ends of the cross flue increases, and the draft intensity presumably weak- ens, the area drained by the collecting flues lessens, and a practical balance is maintained. Some builders claim that since the intensity of the draft is greatest in the center of the cross flue, the inlets from the central collecting flues should be of minimum size, increasing toward the ends of the cross flue. Others hold that the inlets into the cross flue should correspond to the floor area drained, which would make central connections large and the end con- nections small. Both plans are followed with good results. The kind of ware and setting has a great influence on the Figure 88. draft, and it is a good plan to make all the openings — both those into the cross flue and those in the division walls larger than the plan shows then close them with loose bricks to the proper size. Should the draft not be uniform, the sizes of the various openings can be easily changed by the placement or removal of the throat bricks. Thus the kiln can be ad- justed to correspond to the conditions, and the adjustment is permanent. Fully perforated floors preferably have the collecting flues below the floor flues, which gives opportunity for graduated openings from the floor flues into the collecting flues. The western kiln, previously mentioned, is of this latter type ex- cept it does not have graduated openings distributed over the kiln area through the diaphragm between the floor and the collecting flues, but instead there is a single peripheral open- 210 BURNING CLAY WARES. in g from each perforated floor flue into each underlying col- lecting flue. All the gases escaping through the floor must travel to the kiln wall, then down into the collecting flues and back to the cross flue. In other words, the draft is entirely in the outer ring of the kiln. There are numerous modifications of the cross head flue type of kiln. Fig. 88 shows a combination of a cross head flue with a ring collecting flue. The cross flue is divided into three parts, each controlled by a separate draft flue extended to the out- side of the kiln, thence a single flue to the stack. The claim for this plan is individual control of the center and sides. The design is an effort to get the distribution of multiple stacks with a single outside stack. Fig. 89 and Fig. 90 show a kiln presented by Greaves- Figure 89. Walker, which may be classed as a cross head flue type with the difference that both the main flue and cross flue are con- nected with the collecting flue system. The features of the kiln are, the sloping collecting flues as in the kiln illustrated in Fig. 75, and the large main flue, which may be entered from the outside to remove the accumulated sand without disturb- ing the kiln floor. Comparison of this kiln with the two preceding kilns again brings out a marked difference in the draft. The kiln being considered has its strongest draft on a quarter point in the kiln circumference, decreasing in intensity toward the oppo- site quarter, and still further decreasing in intensity toward BURNING CLAY WARES. 211 the ends of the cross flue, in consequence of the right turn. If such a kiln will give satisfactory results, and it is said to do so, why complicate a kiln bottom with multiple draft flues and minute adjustments as in the two preceding kilns? We believe that there are extremes in complications of bottoms to get theoretically uniform heat distribution, but barring such extremes, the kiln with the best distribution will give the bet- ter results, leaving, as it does, less to the skill and intelligence of the burner. A third modification of the cross flue developed cn the Pa- cific coast, illustrated by Fig. 91 and Fig. 92, eliminates the blind draft flue, which is also eliminated in the Greaves- Figure 91. Walker kiln, and instead of the direct and transverse flues, uses two direct flues. This plan has individual control of the halves of the kiln, combining the idea of individual control of the first mentioned modification, with the self-cleaning idea of the last mentioned kiln. In this kiln we have the draft all from one side, but it 212 BURNING CLAY WARES. would be better to have the draft connections on opposite sides of the kiln, thus giving, in a measure, a balanced draft. As it is, the least kiln area on one side has a maximum draft and on the other side a minimum. Control on opposite sides would make both sides alike and while not theoretically per- fect, yet considering the results obtained from the second modi- fication, it, with the change suggested, would give practical uniformity. We illustrated an up-draft potter’s kiln in Figs. 66 and 67. The central stack kiln as shown in Figs. 77 and 78, or with other arrangement of under floor flue system, is considered by many the best type of kiln for pottery work. The bottom of a simple down-draft potter’s kiln with wall stacks is shown in Fig. 93. The under floor flues are circular with openings in the walls to form radial flues on the same level, leading to the wall stack flues, as in the Eudaly kiln. This kiln is practically a Eudaly kiln adapted to pottery work and the kiln properly belongs with the multiple wall stack kilns except that individual stacks are not built on the kiln BURNING CLAY WARES. 213 wall, but instead, a large stack, the full diameter of the kiln at its base tapering to the requisite opening in the top, is used. Such a stack is shown in Fig. 66 and Fig. 94. The lat- ter illustration together with Fig. 95, is from a kiln designed Figure 94. by C. B. Harrop. The plan, Fig. 95, shows three levels, one through the lower draft flues, one through the floor flues, and Figure 95. one above the floor level. The vertical view shows a section through a furnace and one through a draft flue. The waste gases circulating through the upper flue level are drawn to the center, then down and out to the wall flues through the 214 BURNING CLAY WARES. lower level. The floor openings are graduated in size, being smallest near the center, to give uniform draft over the floor area. Prof. Harrop suggests that the floor system could be simplified without serious loss by taking the gases to the wall flues direct from the upper level, thus eliminating the lower set of flues. The large stack typical of a “bottle” (pottery) kiln is not always necessary. For instance, in the up-draft kiln, Fig. G6, a small central stack above the crown of the kiln can be carried on I beams supported by the kiln walls. Banding Round Kilns. We haye differences of opinion in regard to the proper banding of a round kiln, varying from a single band at the crown skew level to a complete steel casing, including the furnaces. Regarding the advantages of the latter, Prof. Har- rop in “Kiln Expansion and Bracing,” says, in substance: (1) The kiln w T alls are kept plumb. (2) There is no leakage. (3) The walls are protected from the elements. (4) Steel radiates 12 per cent, less heat than fire bricks. No one, we think, will question the advantage of a steel casing and only the cost of installation prevents its universal adoption. The usual thought in relation to kiln bands is, that the chief purpose is to hold the walls in place and prevent the collapse of the crown, for this we do not need steel casing. Harrop figures that a band y 8 inch by 6 inches is strong enough to withstand the thrust of a 30-foot kiln crown, but unless the kiln walls are properly laid to take care of the wall expansion, the strain on the band may be greater than that due to the crown thrust. We have had 5-16 inch by 6 inch, and % inch by 10 inch bands burst on smaller kilns than 30 feet diameter. We found it necessary, where the banding consisted of sev- eral narrow bands, to adjust the tension during the first heat- ing up of the kiln. At intervals of about 30 minutes each band was loosened to a normal tension and thus we provided for the wall expansion. After the kiln cooled, we found all the bands loose, showing that the kiln wall practically carried the crown thrust. A very extensive practice is to use a single band 24 inches to 36 inches wide as the height of the wall above the kiln door will permit, sometimes cutting out the lower side of the band BURNING CLAY WARES. 215 to conform to the door arch. Such a band is put in place when the wall has reached the proper height, fully riveted, thus eliminating bolts and lugs, and the wall is carried up to the final height inside the band. Such a band is amply strong for both crown thrust and wall expansion. It is questionable however, whether such a band is the best practice. It is good in that it gives a wide margin of safety, and in so far as it approximates a steel casing, but unless it is supplemented by lower bands the walls will bulge and crack, thus opening passage ways for cold air in the lower part of the kiln, where the inward leakage is greatest and where (in down- draft kilns) we have the greatest difficulty in attaining the de- sired temperature. It would be better practice to case the lower part of the kiln wall with a wide band and support the crown with a nar- row band, but since the ash pit, fire doors, and wicket opening complicate the use of a wide band around the bottom, it is more practical to use a number of narrow bands which can be placed to miss the openings. One band can be placed with its top edge flush with the kiln floor or ash pit floors. A second band can be placed at the grate bar level. A third band at the top of the fire mouth, and above this other bands can be spaced at close intervals up to and including the crown skew level. The bands below the wicket arches can be carried across the wicket opening by long bolts, or better, a steel frame can be placed at the wicket jambs, and the bands within the wicket height attached to this frame. Such a frame can be made of wide channel bars set flush with the wicket jambs and riveted to the bands below the floor level and above the wicket arch, and to these channels the intermediate kiln bands can be riveted, thus leaving the wickets clear. The bands should be spaced, so far as possible, to include every course of bricks as shown in Fig. 96, and where this is not possible, as in the lower kiln wall, vertical bands spaced 24 inches to 36 inches as shown in Fig. 97, should be used. In one instance a kiln was cased with 12 inch bricks on end and the joints covered with a narrow band as shown in Fig. 98. A kiln properly banded will require 9 to 13 bands and, con- sidering the cost of a kiln, and the losses resulting from a distorted wall, one can readily afford proper bands, if indeed he cannot go further and completely case the kiln wall with steel, and back up the casing with a porous insulating brick. 216 BURNING CLAY WARES. and finally cover the crown deeply with ashes or some better material. Such a kiln construction will result in : (1) Direct fuel economy. (2) Quicker bums with further fuel saving and less labor. (3) Better results. Many clayworkers save a little in the cost of installation Fig. 96. Fig. 97. Fig. 98. and then for all time complain that the cost of burning and upkeep eats up all the profits of the business. Up-and-Down Draft Kilns. There have been a number of up-and-down-draft kilns de- veloped, but as a rule they have, not shown sufficient merit over the simple up-draft or down-draft types to lead to any wide use, except in some one particular line of ware for which the kiln was developed. The chief difficulty with them has been that the ware in the bottom of the kiln was subjected to the maximum flame temperature, resulting in overburning the bottom courses of ware or if not overburning, at least discoloring in consequence of flame contact and reduction. Another difficulty in some of the types has been to maintain the combustion flues, subjected as they were to such intense temperature and constructed of the usual kiln fire brick, often inferior in quality. The early up-and-down-draft attempts were very simple, as shown in Fig. 99, adapted to a rectangular kiln. An under floor flue connected the furnace with the interior of the kiln, BURNING CLAY WARES. 217 and a damper controlled the bag outlet from the furnace. With this damper closed, and the throat to the under floor flue open, the gases passed under the floor, then up through the ware until caught and drawn down by the force of the draft. The draft flues might be parallel to the combustion flues extending from the bag wall to the center flue, instead of a single central draft flue with perforated top as illustrated. Round kilns work out equally well with a series of radial flues — one set from the furnaces to the center for combustion flues, and the alternate flues from midway between the bags to the center for the draft flues, connecting with a center well, center stack, or wall stacks. The throat to the under combustion flue could be easily fully closed with a fire clay block, or partially with fire bricks, as the up-draft feature might require. Up-draft water smoking is considered the most desirable, and a number of down-draft kilns have been designed for such up-draft work to be followed by down-draft burning. The original Eudaly kiln, Fig. 80, had this feature. The radial flues were extended through the kiln walls and boxed in outside the walls. By closing the stack dampers, furnace doors and ashpits, opening the crown vent, and using the open ends of the radial flues as furnaces or air inlets the operation became up-draft, either for watersmoking or cooling the kiln, Fig. 99, Fig. 100. 218 BURNING CLAY WARES. but so far as our observation goes, little use was made of this feature of the kiln. Any radial or cross head flue kiln could be adapted to this purpose, and several kilns offered to clayworkers have had the up-draft water smoking feature. So far as the up-draft feature in the burning is concerned, practically the same results are obtained by openings through the lower part of the bag wall, and we frequently find such bag wall construction, generally, however, by building the bag wall in open checker work. The Stewart kiln was a popular one in drain tile burning and to some extent in other lines of ware. Had the kiln gen- erally been built of better material its use would have con- tinued and been extended, but many installations were built Figure 101. of very inferior fire bricks instead of the best obtainable, and in consequence there were many complaints in regard to the failure of the combustion flues after a short period of use, sometimes in the first bum. Fig. 100 and Fig. 101 illustrate the principle of this kiln. The furnaces are on opposite sides of the kiln and stag- gered to bring them in line with the combustion flues. Each furnace has two flues leading from it, under the floor, to a bag wall on the opposite side. The adjacent pair of flues leads from a furnace on the opposite side to a bag across the kiln, and thus, the floor system consists of alternate pairs of flues carrying gases in opposite directions, each pair having its fur- nace and corresponding bag. The floor is solid throughout and BURNING CLAY WARES. 219 the outlets from the kiln to the wall stacks are in the kiln wall at or near the floor level. The ware in the lower part of the kiln is burned by heat conducted through the floor, and the ware in the upper part of the kiln by direct contact with the gases from the bags. It could hardly be called an up-and-down-draft kiln, but the effect was the same in that the top and bottom were burned at the same time, and equally hard burned. The ware which receives the least heat, and which is the last to finish burning, is in or near the vertical center of the kiln. The kiln was first designed and most generally built as a Figure 102. round kiln to which the principle is less adapted than to a rectangular shape. In the round kiln, it was necessary to set the ware with flues in the bottom to insure any degree of uni- form draft over the kiln area. This feature was decidedly ob- jectionable especially for drain tile for which the kiln was largely used. The principle works out nicely in the rectangular kiln by alternating the furnaces in singles or in pairs with a cross draft flue to wall stacks as shown in Fig. 102. This enables us to build the kiln any desired length for large capacities, which was not possible in the original kiln, with its fully solid floor and stack inlets above the floor level. We found it possible in such a rectangular kiln, to burn entire kilns of special ware 220 BURNING CLAY WARES. which in the regular down-draft kiln could only be burned in the upper part of the kiln, and no special setting was required in the bottom as in the original kiln. Above Fig. 103. Below Fig. 104. A recent up-and-down-draft kiln has a unique feature which may overcome the difficulty of the earlier types. In the latter the up-and-down-drafts were obtained by a distribution of the BURNING CLAY WARES. 221 heat from one set of furnaces, and the intense heat essential for down-draft had a serious effect on the ware when diverted to the up-draft, especially in view of the proximity of the ware to the furnace. The kiln in question, shown in Fig. 103 and Fig. 104, uses two sets of furnaces, one for up-draft and one for down-draft, and each may be fired as the conditions require. The gases from the furnaces for up-draft are led through main flues and distributed in cross flues with perforated covers (kiln floor) throughout the under floor system, then rise into the kiln and come in contact with the ware. The stacks are approximately on the quarters of the kiln, and the inlets from the kiln are in the kiln wall three or more feet above the floor. The down-draft gases follow the usual course, up through bags and down through the ware, except that they do not pass to and through the floor as in an ordinary down-draft kiln, but instead are drawn off to the stack inlet above mentioned. The up-draft furnaces may be used for up- draft water smoking by closing the down-draft furnaces, open- ing the crown vent and closing the stack dampers. The illus- tration shows the up-draft furnaces in the base of the stack and by discontinuing the use of the stack furnaces except for draft intensity, with proper damper adjustment the burning operation could be entirely down-draft. The course of gases from the down-draft furnaces would be, up through the bags, down through the ware and perforated floor, into the collect- ing flues, thence to the main cross flues, through the up-draft furnaces, and up through the stacks. The furnace in the stack base with the connecting damper gives opportunity to heat the stack for increased draft to any degree and at any time during the burning operation. The Gamble and Bryan up-and-down-draft pottery kiln is illustrated in Fig. 105 and Fig. 106. In general the firing is that of a simple up-draft pottery kiln as shown in Fig. 66 and Fig. 67, and it could be used exclusively for up-draft wmrk with the crown vent open and the wall flue dampers closed. Be- tween the radial flues from the furnaces is a duplicate set of radial flues, with floor inlets, connected with the wall flues. It is only necessary to close the crown vent and open the wall flue dampers to convert the operation to down-draft. Another similar up-and-down-pottery kiln has the stacks for down-draft operation outside the kiln wall, but leading into the 222 BURNING CLAY WARES. Above Fig. 105. Below Fig. 106. kiln main stack above the kiln crown and connected at the base with the bottom of the kiln. We have mentioned the difficulty of maintaining under floor flues for burning bricks by up-draft, but it must be noted that BURNING CLAY WARES. 223 pottery kilns are largely of this type, without serious failure in this respect, although the temperatures required are much higher. The only explanation is that pottery kilns are built of better material. The comparison in favor of the pottery kiln might be carried still further, for one has only to compare pottery and muffle kilns with the average brick and tile kiln to note the difference in the structures in every particular. If pot- tery kilns were built of the same materials used in common ware kilns, the pottery industry would be in a sad plight. Horizontal Draft Kilns. A description of a horizontal draft periodic kiln would not be necessary except for our classification of kilns, because such kilns find little use in this country. We do not know of any distinctively horizontal-draft kiln at the present time, but we have seen one or two such kilns in the past which have since been abandoned. It is an early type of crowned kiln and perhaps a natural adaptation of a furnace, a hearth and a stack. Fig. 107 illustrates an early horizontal-draft kiln and little explanation is needed. There are three or more furnaces in the front, one of which is in the doorway and after each burn is torn out to provide an entrance to the kiln. The kilns were built tapering on the sides and crown from front to stack presumably on the principle that as the gas gave up its heat, became lower in temperature and less in volume, the mass of ware to be burned decreased correspondingly. One would think that the reverse would be the proper principle wherein the high temperature gas and large volume would quickly pass the restricted mass of ware, leaving a greater volume of heat and a slower movement for the larger mass of ware. Whatever the correct principle the horizontal-draft kiln has little merit. The nearest approach to a horizontal draft is a kiln used in burning fire bricks, shown in Fig. 108. It is properly a down- draft kiln but an effort is made to get a horizontal draft. It is equivalent to two horizontal kilns placed back to back. In each end are three furnaces, one being in the doorway. The bags are low and built of wide open checker work for the horizontal draft. Midway in the kiln is a cross main draft flue leading to a stack outside the kiln wall. The kilns may be built singly, but usually they are in bat- teries of several kilns in touch with each other and it is only 224 BURNING CLAY WARES. necessary to brace the outside walls of the end kilns. A fur- ther advantage of the battery construction is the elimination of some radiation loss from the sides of the kilns. The kilns are necessarily limited in capacity, and size considered it is not possible to get as uniform temperature throughout the kiln as in the more modern down-draft kilns. The best evi- dence of the inefficiency of the kiln is that the newer fire brick plants controlled by the same corporations have built the round down-draft kiln. It has been our opinion for a long time that some type of a continuous kiln had its most promising field in the manu- facture of fire bricks because of the high temperatures re- quired and excessive fuel loss in consequence, but there has been very little development along this line. One factory used for many years a gas-fired semi-continuous ring (tunnel) kiln, now dismantled, and another factory built two gas-fired chambered kilns of a Scotch design, but abandoned them after a short period. Also a small ring kiln, direct coal fired, on an Illinois fire brick yard has been wrecked after many years use. In spite of the several attempts to use regenerative kilns in this industry and the unsatisfactory results, we do not be- lieve this type of kiln has had a fair trial in this line, and we still hold the same opinion in regard to the usefulness of the continuous kiln in the fire brick industry. BURNING CLAY WARES. 225 At the present time three car tunnel kilns are being tried out in fire brick manufacture, two of which are proven to be successful, and it is to be hoped that the problem of fuel economy in fire brick burning has been solved in these kilns. Muffle Kilns. Muffle kilns are used to burn terra cotta, enameled bricks, and other wares, which must be protected from contact with the furnace gases, and in consequence burned by radiant heat from a muffle wall separating the ware chamber and furnace flues. It is apparent that, to be economical, the muffle walls must be as thin as possible and the arrangement of the kiln ducts must provide for a complete encircling of the muffle by the furnace gases. These are the problems of the muffle kiln. The up-draft muffle kiln needs no illustration. If we take an up-draft pottery kiln such as that shown in Figs. 66 and 67, and build a muffle inside of it, we will have an up-draft muffle kiln. If the kiln is small, no center flue is required, but we retain the under floor flues connected with the furnaces although there is no positive draft movement under the floor. These flues extend across the kiln from furnace to furnace and the floor is heated by radiant heat from the furnaces and by con- vection through pulsations of the gases as the pressure condi- tions in the several furnaces vary from time to time. When 226 BURNING CLAY WARES. there is a large volume of gas in one furnace the pressure forces some of the gas under the floor, perhaps fully across to the opposite side, until the pressure all around is equalized. The central opening or any openings between the under floor flues and the kiln chamber of the pottery kiln are closed when converted into a muffle kiln, and the bag walls are con- verted to an annular ring carried nearly to the height of the kiln wall and completed with a crown a few inches below the kiln crown. The movement of the furnace’s gases is up through the annular space between the muffle wall and kiln wall, then through the space between the muffle crown and kiln crown to a central vent in the kiln crown. The stack may be the usual pottery stack resting on the kiln wall and tapered to the required opening at the proper height, or it may be a smaller structure supported by I beams, or by the kiln crown, or by a third crown sprung from the kiln walls for this special purpose. The latter mentioned support is illustrated in Fig. 115. For large up-draft kilns, the central opening in the floor of the pottery kiln is retained and extended to the top of the muffle crown by a circular wall. The movement of the gases under the floor and up through the central flue is positive, and the mass of ware within the muffle is heated by conduction through this flue wall and radiation from it to the ware as BURNING CLAY WARES. 227 well as by conduction through and radiation from the circum- ferential muffle wall. Such up-draft designs are preferable for small kilns, particularly for short flame fuels. A rectangular muffle kiln developed in England and used in Canada in the manufacture of enameled bricks is illustrated in Fig. 109, Fig. 110, and Fig. 111. Fig. 109 is a vertical section through the furnace and Fig. 110 is a corresponding section through the flue adjacent to the furnace. Fig. Ill is a com- bined view plan below and above the floor. The gases from the furnaces pass under the muffle floor to the opposite side, then up through a flue space between the Figure 111. muffle wall and kiln wall, and this flue continues over the muffle crown and down to the floor on the furnace side. When the gases reach this point they pass through an opening into the adjacent parallel flue, then rise, pass over the muffle crown, and down to the floor on the opposite side where they escape into the wall draft flue, up and out into the stack. Thus the gases make a complete circuit of the muffle and reverse except they do not pass under the floor a second time, but in- stead are drawn off into the stack at the floor level. The muffle floor support consists of a number of single brick piers which permit complete circulation under the floor, 228 BURNING CLAY WARES. and this circulation may be controlled by dampering the indi- vidual wall draft flues. The kiln is necessarily narrow to insure a substantial Figure 112. muffle construction. The construction is simple and it has been proven practical to reduce the thickness of the muffle walls to 1 V 2 inches. A down-draft, or one might call it an up-and-down-draft, Figure 113. muffle kiln is illustrated in Fig. 112, which shows the plan through the lower return flues and that above the muffle floor. Fig. 113 is a vertical section through a furnace on one side and a doorway on the opposite side. The latter shows BURNING CLAY WARES. the stack within the central muffle flue, but this construction is not essential. After the gases have reached the inlet at the base of the stack, they may be drawn off through the stack as shown, or drawn down into a central well, and to an outside stack or fan through an underground draft flue. The situation is the same as that of a central stack and central well open fire periodic down-draft kiln, where either plan of removing the gases may be used, and both methods are used in the muffle construction. The circular feather floor walls shown in the plan provide direct flues to the center, but above the diaphragm these walls are staggered as shown in the vertical section, and the gases in their passage from the center to the circumference must take a sinuous course before dropping into the lower direct flues. This insures full contact with the muffle floor and a maximum absorption of heat. The stack is shown as a con- tinuous structure from the kiln bottom, but the usual method is to independently support the portion of the stack above the kiln crown as previously noted. The movement of gases from the furnaces is up the annular space between the muffle and the kiln wall, then over to the central annular space between the muffle and stack, down which they pass to the flues below the floor, spread out to the circumference of the floor flues and are drawn down and back to the stack or well through the lower flue system. A long flame coal, by means of which we get more or less gas combustion throughout the circuit, is essential in securing uniform burns. It is a question whether the central stack is as good as the central well with underground draft flue. It insures strong draft almost from the beginning of the burning, and, as shown in the discussion of stacks, does not require the height of an outside stack. On the other side of the question, it decreases the size of the muffle or increases the diameter of the kiln for any given size or muffle. The chief point, it seems to us, is that the gases have given up their heat value when they enter the lower flues, and as they rise through the stack, having a lower temperature than the descending gases in the annular space between the stack and the muffle, there will be a flow of heat from the hotter gases through the stack wall and this heat wflll be taken up and carried away by the stack gases. In other words it seems to us that after the gases have ceased 230 BURNING CLAY WARES. to be valuable for heating the muffle, that it is a mistake to bring them again in touch with it where the only effect would be to lower the temperature of the muffle gases. Another up-and-down-draft muffle kiln is shown in Fig. 114. Figure 114. Figure 115. In this kiln the annular space between the muffle and kiln wall is divided into sections, up in front of the furnaces, as shown in the section through the furnace, and down between BURNING CLAY WARES. 231 the furnaces. The gases from the furnaces will rise and he projected into the crown space, then caught and be drawn down through the alternate flues between the furnaces, pass under the floor, up the central flue and into the stack. Fig. 115 shows two half sections of a more complicated kiln with practically a double annular space. The inner space into which the furnaces deliver their gases is a complete circle but the outer space is from furnace to furnace — practically broad flues between the furnaces. The movement of the gases is indicated by the arrows. Muffle Kiln Construction. It is evident that to get a satisfactory life from a muffle kiln, it must be built of the best materials laid up with great care. When we consider, that the muffle walls are only 214 — T 1 Sr "7 £7ee Figure 116. inches thick and the muffle crown not more than four inches thick and often as thin as 214 inches, that the latter is pierced by a number of steam ports extending through both of the kiln crowns, and that it must make and retain a close con- nection with the muffle flue, we appreciate the need of the best possible construction. If the kiln is not properly banded and its walls bulge, as frequently happens in common ware kilns, the muffle wall will follow the kiln wall and become cracked and distorted to its ruin. If the fire bricks are not properly burned the shrinkage in use will settle the muffle and tear the crown away from the steam ports. Should the central flue not settle 232 BURNING CLAY WARES. to the same degree, or should settle at all without being fol- lowed by the crown, there will be a separation of flue and crown. The average life of a well-built muffle is from forty to fifty burns, and it is remarkable that the structure should remain intact this long. A common method of building the muffle wall is shown in Fig. 116, all the bricks being preferably special shapes unless the diameter of the kilns is relatively large. The outer wall is backed up by the kiln wall, and cannot crowd out unless Plan Figure 117. El ev. Figure 119. the kiln wall gives way, nor can it come in because of the circle. The inner wall cannot crowd in because of the circle, but it can work outward. An improvement on such a wall is shown in Fig. 117, where instead of ordinary wedge bricks for the ties, a special shape with shoulders is used. An advantage of this plan is that inner and outer wall may use the same shape, or a standard brick if the kiln diameter is not too small. BURNING CLAY WARES. 233 The central flue may be built of circle bricks, but a better construction is illustrated in Fig. 119, showing a tongued and grooved circle block. This makes a strong structure, the walls of which may be very thin. Crowns are often built of half bricks in wedge and key shapes, but a thinner crown is possible by using tongued and grooved blocks somewhat similar to those in Fig. 119. These blocks, of course, are special shapes, and each ring has a different radial pitch which requires a special shape. If the kilns are of a standard size the repair supplies become a simple matter, but with a number of sizes of kilns in no way standardized the problem of repairs is increasingly diffi- cult. A later method of muffle wall construction is shown in Fig. 118. The walls are built of hollow tile made of the best grade of No. 1 fire clay. The base of the muffle wall in any construction is built thicker to withstand the cutting action of the flame and slag- ging action of the ash. This thicker wall also protects the ware from the intense furnace heat. The thin muffle wall of whatever construction rests upon this base wall, but for the hollow tile wall a ring distributing flue around the base is necessary to get the gases fully around the kiln circle. 234 BURNING CLAY WARES. CHAPTER IX. SOME NOTES ON SETTING. W E HAVE mentioned the setting and discussed it briefly in connection with the descriptions of the kilns, and we will not here take up the ordinary setting, but instead will take up some special features. A large factor in common brick setting is the possibility of rapid work even at some expense in quality, and for com- mon bricks the skintle method, or alternate headers and stretchers, has the preference. Face bricks, on the other hand, require a setting which will give the maximum quantity of first quality, and the nat- ural color product is set faced in alternate double courses of headers and stretchers, while the fire flashed product is set flat and the setting is such as gives the most uniform ana maximum exposure of the faces to the kiln gases. The inexperienced clayworker does not see any reason why flat set bricks may not be in alternate headers and stretchers in single courses or a number of courses, and he cannot under- stand the importance of the complicated setting used in many instances. We have seen such simple alternate flat setting and the results were as unsatisfactory as the setting was primitive. The chief loss in the kiln output either in broken bricks or culls comes from the binder courses and the base and cap courses, and the setting should be worked out to keep this loss to a minimum. Some brick products break easily when subjected to kiln strains, or distort (kiln mark) under weight and particular attention must be given to their set- ting. We have seen kilns of brick irregularly bonded through- out and the strength of the product was such that the whole mass was drawn together in shrinking without material loss in broken bricks. The manufacturer of such bricks is par- ticularly fortunate in his material. Other clays develop such BURNING CLAY WARES. 235 weak products that any bonding in larger masses than three brick benches or any lipping of one brick on another, results in a broken product. In down-draft kiln setting, where the ware extends above the bags, the front exposed to the flames must be racked back to prevent it from drawing over and falling into the bags, and the open setting in such fronts must be partially or completely closed to force the gases over the top. Three brick benches, racking back above the bags, and protection ot the exposed fronts, are shown in Fig. 76. Setting in up-draft kilns has been fully discussed and illustrated in connection with the kilns. In addition it may be mentioned that some small yards frequently burn drain tile in such kilns along with bricks. The arches are set with bricks in the usual manner. Above the arches, brick and drain tile are set in alternate benches, but the tile benches to preserve the continuity of the kiln walls. Such a com- bination of bricks and tiles is not to be recommended. The are enclosed with the usual setting of bricks on the heads tiles, because of their thin walls, burn more quickly than the bricks, and since the columns of tiles form chimneys with relatively low resistance, the tendency of the draft is through the tile benches, where it is least needed and most likely to do damage. We have seen warped overburned tile enclosed by benches 01 underburned bricks. In the setting of such a combination care should be taken in setting the arches to reduce the draft spaces in the brick setting under the tile benches and thus force the gases into the brick benches* Variations in down-draft kiln brick setting should not be necessary if the kiln is properly designed and in good con- dition. This has been mentioned in discussing down-draft kiln bottoms, and a study of the bottoms will show that in some of them it may be very necessary to correct the defi- ciency of the bottom by variation in the setting. It is pos- sible to materially increase the degree of uniformity of the burning by the setting, but if the kiln bottom is of a good type and the clay has a fair burning range, there should be little need of variation in the setting. We have experi- mented quite a little along this line. Starting with uniform setting of 8 on 3 in a down-draft kiln we change the upper half of the kiln to 7 on 3, on the principle of less resistance and a rapid movement of the gases in the upper part of the kiln, thus conserving the heat for the lower part. 236 BURNING CLAY WARES. We next went to the opposite extreme and set the lower half 7 on 3, and the upper 8 on 3, and topping out with two courses set practically tight. We provided a square vertical flue in the center, and midway of the height of the setting were distributing flues from the vertical flue to the outer circle of the setting. The principle involved was to burn the upper muffled mass of bricks by conduction and the lower mass by convection. The heat distribution was satisfactory, but considerable loss resulted from the distributing flues, and also some damage accrued in consequence of the excess weight above and reduced support below. We finally came back to the original setting of 8 on 3 throughout the kiln, but by this time we had learned that there were greater possi- bilities in the handling of the fires and control of the kiln draft than in the different methods of setting. Figure 120 nnn nnn ITT - - y- - * -- -- Figure 121 Setting for Flame Effects. Bricks to be fire flashed are set on the flat and as above noted, it is important that the setting be such as to permit an equal flame contact with the faces of the bricks. The color effects are very sensitive to the action of the kiln gases. Where the bricks are set in checker work the gases, after passing the checkers, flare out and produce a fan-shaped flame effect on the faces of the underlying bricks. Walter A. Hull’s paper, “On the Burning of Rough Tex- ture Shale Bricks,” in Vol, XVI, Trans. American Ceramic Society, should be read by every clayworker interested in rough texture face bricks. The paper is a detailed descrip- tion of methods of burning to get any desired color effects, but it incidentally discusses the setting. Fig. 120, plan, and Fig. 121, front elevation, show a simple setting for flashed bricks. As shown in the plan the bench BURNING CLAY WARES. 237 starts with a stretcher and three headers, or two headers for a two and one-half brick bench. This plan is carried up five to six courses or more as the uniformity of the brick will permit and then reversed for a duplicate number of courses, and thus alternating to the top of the setting. It is desirable to have the setting work out to form independent columns 3 *4 bricks by 3, or 4 or 5 bricks. The plan shows a 3*4x5 brick column, but will also work out 3*4 x 4 with a little wider spacing than that shown. Frequently the headers are set part single and part double, as in Fig. 122 — the double courses set back to back give greater stability to the column and such setting permits us to reduce the size of the columns from that shown. For instance, if the third and fourth courses from the left in Fig. 120 are brought together, the column would work out four stretchers and the setting would be four single headers alternating with one double header. The benches are spaced about 2*4 inches, and to stay them 4 1 =ZB — — Figure 122 ■FI 1 II 4 — — Figure 123 it is necessary to project headers at intervals across this space to abut against the adjacent bench, as seen in Fig. 124, and the double courses are desirable for this purpose, in that no broken bricks are required to fill up the space left vacant by the projection. In the sketches the bricks are shown set tight, which would give one flashed end in each 3*4 brick. This is not a sufficient proportion of quoins for many jobs, and the number may be increased to 100 per cent, by separating the bricks. If the header courses are set away from the stretchers the usual space we will get two flashed ends from each 3*4 bricks. If the loose end header is set away from the adja- cent header we increase the flashed ends to 3 out of a pos- sible 3*4, and one of them would have both ends flashed, which is necessary for pier and pilaster work, and the stretch- ers may be spaced as shown in Fig. 123, thus giving a flashed 238 BURNING CLAY WARES. end on each brick. In some instances where it is desirable to have the benches even bricks — three or four, as may be required — the first setting has a stretcher course front and rear and second tier will be all headers, thus alternating to the top. The setting shown in Figs. 122 to 125, inclusive, is more complicated, but it has greater stability. The bottom course is set, as shown in Fig. 122, merely to give a full bearing for the superimposed stretchers. This bottom course may be omitted, and the setting started with the regular layout, as shown in Fig. 123. This is carried up four to six or more courses, then comes a single tie course, as shown in Fig. 125 or Fig. 126, and if Fig. 125 is the plan followed, this is topped by a single course reversing Fig. 122. Then follow four to six courses, as in Fig. 123 reversed. The bench tie course, Fig. 124, or Fig. 125, may be repeated at this level, or may be omitted until some higher level is reached, depending upon the need for ties. The complete setting will be as follows : First course as in Fig. 122 ; 2nd, 3rd, 4th and 5th courses as Fig. 123 ; 6th course, Fig. 125; 7th course, Fig. 122 reversed; 8th, 9th, 10th and 11th courses, Fig. 123 reversed; 12th course, Fig. 125 re- versed. An occasional header is projected from the plan Fig. 122, or Fig. 123, as shown in Fig. 124. The 12th course com- pletes the cycle, and the order is repeated to the full height of the setting. If we use the tie shown in Fig. 126, the setting will be as follows: First course, Fig. 122; 2nd, 3rd, 4th and 5th courses, Fig. 123; 6th course, Fig. 126; 7th, 8th, 9th, 10th and 11th courses, Fig. 123 reversed; 12th course, Fig. 126, if close tying is necessary, or if not, then the 12th course is as Fig. 122, followed by Fig. 123 in the 13th, 14th, 15th and 16th courses, then perhaps the tie, Fig. 126, followed by Fig. 123 reversed. BURNING CLAY WARES. 239 A simple setting for flashed brick is shown in Fig. 127, front view, and Fig. 128, side view. The setting is started in two courses set on edge. Flat setting is carried up in two brick benches and the brick are spaced so the flat setting -M 1 3 f~1 — r~ i 3 -FT 1 i 3 Figure 126 will break joints with the tie stretchers. This is also very open setting and expensive in kiln room, but the loss in this way may be partially offset by quicker burning. The ties run Fig. 127, Front View. Fig. 128, Side View- through the full width of the setting, and as they always result in culls, it is important that the flat courses in each tier be set as high as practical to reduce the number of tie courses to a minimum. 240 BURNING CLAY WARES. About one-third and three-fourths the total height of the setting, or as frequently and at such heights as may be found necessary, a double tie course is introduced — one the regular cross tie and the other a through bench tie, as shown in Fig. 128. Usually these bench ties are placed at levels convenient for the setters— the first tie coming at the limit height of setting from the floor level, and upon this tie the setters stand to carry the setting to the higher levels. The ten- dency to rolling is greatest at the top of the setting and the upper bench tie should be near the top, even though three bench ties may be necessary, but a third bench tie is objec- tionable from a working standpoint, because it introduces 1 l £ *4up i hrgFTgF m Q Fig. 129, Front View. Fig. 130, Side View another set-back over which the setters must reach to set the top tiers. A setting similar to the preceding is shown in Fig. 129, front view ,and Fig. 130, side view. As in the preceding the flat courses are all headers, but by using a double tie course for each tier the spacing is much closer. The double tie course gives greater stability, and is preferable where the rolling and twisting tendency is excessive, although the num- ber of culls is increased. However, considering the total BURNING CLAY WARES. 241 mass, the proportion of culls in this setting will be only slightly greater than that in the preceding method. A method of setting in bungs for salt glazing is shown in Fig. 131, front view and Fig. 132 plan. The bungs are tied together by headers spanning the intervening space. Such setting greatly reduces the capacity of the kiln, but for salt glazing the results are very satisfactory. We frequently have to burn bricks set flat, such as orna- ]- i L_ — J — 1 H e3Q — i3Cb tr in _b LI Above, Fig. 131, Front View. Below, Fig. 132, Plan. ! ! 1 1 • » JTTT 1 f a !!!__ ■fr in n Above, Fig. 133, Front View. Below, Fig. 134, Plan. mental shapes, enameled bricks, etc., the faces of which must be protected from the flames, and the usual “boxing in” method for such work is illustrated in Fig. 133, front view and Fig. 134 view plan. Each bench is started with a course of bricks on edge, and on this is set a flat face and tie course of standard bricks, as shown in the plan to carry and protect the special shapes. Each tie course is the same as that shown in the plan. The special shapes, or enameled bricks, are set with their faces toward each other, and only the backs are 242 BURNING CLAY WARES. exposed in the larger draft spaces between each double tier. The benches are set in close touch with each other, and may be tied together to maintain ciose contact, ana tnus prevent the gases from passing down between the benches and baf- fling back into the narrow spaces between the faces of the bricks. The top tier of bricks is covered with the usual tie and protecting course, and this may be doubled on top and carried across from bench to bench. One can readily see that as the bricks shrink there is some danger of the gases getting in between the ends of the special bricks, but it is unusual for them to penetrate as far as the faces, and for stretcher face bricks the protection is ample. Quoins, how- ever, are frequently damaged on the heads, and to protect them the bricks are set in triplets instead of doubles. The quoins are placed in the middle tier and the stretchers in the outer tiers on each side. Roofing tiles of the porous type are set in open kilns with- out supports or stands. Such tiles are made of clays which shrink very little and which are not liable to warpage, dis- tortion and kiln marking. The floor of the kiln is covered with two or more courses of bricks on edge to provide draft space and circulation under the mass of tiles. The tiles so set are of the interlocking type, and no matter how closely they may be nested, the projecting lugs and locks provide draft space. In setting, a bunch of tiles are nested as closely as possible and the number of tiles in a bunch is such that the thickness of the mass corresponds to the length. The setting across the kiln in each bench consists of alter- nate bunches of headers and stretchers, the tiles resting on their edges, and are packed as closely as possible and wedged against the kiln walls. The second course reverses the first. If the first course starts with a bunch of headers the second course will start with a stretcher bunch, and the third course will duplicate the first, thus alternating to file top. A face of set tiles will have a checker board appearance. Squares, or rectangles, the size of a tile, showing ends of the tiles, will alternate with squares showing the side of the tile. The alternation is also carried out in the benches. If the first bench starts with a bunch of headers, the next bench will have stretchers. Any tendency to roll sideways is limited to a single header bench of ten tiles, more or less, since the stretcher benches prevent the extension of the rolling tendency. Similarly, any BURNING CLAY WARES. 243 forward or backward rolling in the stretcher blocks is checked by header blocks in the adjacent benches. It is essential that each course be tightly wedged against the kiln and bag walls, and that the tiles be in touch with each other through- out the mass of the setting. The rolling then is limited to the shrinkage, which in porous tiles is very slight— frequently none whatever. In some instances where there is no shrinkage, or the kiln is narrow, it is only necessary to put in an occasional block of stretchers to prevent rolling. Vitrified tiles, and such tiles as nest perfectly — shingle tiles and Spanish or “S” tiles, without lugs or locks, must be set in “stands.” The stands are fire clay plates, 2 y 2 inches thick, with the other dimensions corresponding to the width and length of the tiles usually around 12 inches by 16 inches. These plates are set up to form a series of pigeon-holes, or “boxes,” as shown in Fig. 135, in which the tiles are placed. The boxes are braced against the kiln walls, or where this is not practical, as in some parts of round kilns, and above the bags, an occasional blind box is set in the stand by simply placing the flat side of the vertical plate parallel with the 244 BURNING CLAY WARES. face and back of the stand, which we also illustrate in Fig. 135. The stands rest on bricks on the kiln floor, which insure draft and circulation under the stands and spacing the stands several inches insures draft space from top to bottom of the kiln. The tiles are placed in the boxes on the flat, on end, or on edge, as may be best for any particular shape. Shingle tile, for instance, may be set flat, or on edge in a solid mass in stands of a suitable height, as shown in the bottom tier in the sketch, but any clay worker will appreciate the difficulty of burning to vitrification a solid block of clay with- out cracking, or checking, bloating, or block coring. Such solid setting is not practical except the entire kiln is so set, and the needed time and care given to the burning. It would be a serious waste of fuel and time to burn a kiln largely filled with open set ware, at the rate required for a few stands packed with tile in a solid mass. Shingle tile may be set open by using strips of clay to separate alternate pairs of faced tile. Several methods of setting shingle tile are shown in the sketch. Spanish tiles are set on end in pairs, closely nested, and the pairs are separated by embedding them in strips of clay, and supporting them in a vertical position by similar strips of clay on top. Interlocking tile are also embedded in strips of clay, and are set on edge or on end as they may best fit the stands or suffer the least deformation in burning. Finials and other roof trimmings are either set in the boxes, or on top of the stands. Terra Cotta is also set in stands called floors, but the boxes are much larger and the floors cannot be built up in uniform sized boxes on account of the variations in the sizes of the ware. The floor blocks must have larger dimensions in order to support the weight which larger boxes involve. There is no standard size for the floor plates and posts. The floor plates are commonly three inches thick and two feet square. The posts are thick walled hollow tiles, about six inches square and vary in length up to thirty inches. Low floors may use short posts, or lacking these one or more posts may be placed on edge and extra high floors may have the posts capped by one on edge. The aim, however, is to have the structure as open as possible to get the full benefit of the radiation from the walls. Manufacturers differ in regard to the best kiln height for terra cotta. The older and more com- mon kiln is 15 to 18 feet high, and there may be five floors. This involves considerable work in building the floors, set- BURNING CLAY WARES. 245 ting and drawing the ware and taking down the floors. Others advise low kilns, even as low as merely head room for the workmen, and limiting the floors to one, or two at most. Economy in heat is claimed for the high kilns and economy in labor for the low kilns. Sewer pipes are set with the spigot end down and the large sizes are placed on rings of green ware to take up the shrinkage strain. The setting is in a series of circles on account of the use of the crane for handling large sizes. As has been previously described, the spaces between the bags are filled with small sizes, three lengths high or sometimes four. Then come rings of intermediate sizes, and inside these are the large pipes, which usually may not be set next to the bags where they would be within the variable influence of the high temperature gases leaving the bags, resulting in serious loss. The setting height is usually four standard lengths, or equivalent in longer lengths. The setting is such that the sockets are in touch, which, because of the circular setting, is a sufficient brace, especially in view of the stability of the columns of large pipe. Smaller sizes are set inside the larger sizes — “stuffing” — but there must be sufficient dif- ference in size to insure a clear annular space between the two, otherwise the inside of one and the outside of the other will not have a good glaze. Elbows, branches, tees, traps, etc., are largely set on top. Tees and branches may be set in the regular columns, but elbows, traps, etc., must be set on top. Some of these shapes may be set in the top course without any difficulty, the spigot resting in the socket of the standard pipe below. Elbows, if set singly with the spigot end down, would be top heavy on account of the socket and would be liable to topple. We have seen them set in this way and the two braced against each other with a clot of clay to keep them apart for draft space. The safest setting for elbows is to use a ring of clay in the socket of the lower pipe, the socket of the elbow resting on the projecting ring, and the elbow may be tipped at an angle to equalize the weight on the supporting column, at the same time insuring a more direct course for the salt fumes. In various simple ways the top of the kiln may be set with spe- cials, often without any bracing, while unstable pieces may be braced, one against another, by clots and strips of clay. Enameled bricks are occasionally set in stands or in sag- gers, especially where the enamel has a tendency to run and 246 BURNING CLAY WARES. form a thick edge very noticeable in the wall in certain angles of reflected light. In stands or saggers they may be set with the enamel edges up and the enamel spreads evenly. Silica bricks were formerly set in the usual checker fashion and this method may still be used in a number of yards. In several yards a bench type of setting shown in Fig. 136 is the modern practice. A course of burned bricks is set on the floor in four brick benches the full width of the kiln, and these bricks are spaced to permit the escape of the kiln gases to Figure 136. Figure 137. Silica Brick Setting. Silica and Magnesite Brick Setting. the kiln floor outlets. This course of bricks is covered with a flat course of burned bricks laid tight, and upon this floor begins the setting of the green ware. The bricks are set headers throughout except at the ends of the benches where stretchers are necessary in order not to rack back except as desired. The setting is carried up to a height of about eigh- teen courses and covered with a flat, tight course of burned bricks. On this upper floor are usually set the shapes, blocks, etc., interspersed with standard bricks. The top is BURNING CLAY WARES. 247 finished with a single, or several courses of standards on edge. The setting is much higher than the usual face and paving brick setting, being around thirty-six courses on edge, where the latter vary between twenty-four and thirty courses. The four brick benches are spaced not more than three inches and these spaces are the draft flues from top to bot- tom of the kiln. It is seen from the setting that there can be no draft down through the mass of ware, but each bench is completely surrounded by the kiln gases, and the ware within the bench is burned chiefly by conduction and radiation, and by convection only to whatever extent the gases are baffled back and forth by the varying gas pressure in the draft spaces. The setting is perhaps a consequence of the burning which differs materially from the ordinary burning process. The method followed is to use deep furnaces with free egress into the kiln, and to heavily charge the furnaces with fuel each firing period. In consequence the kiln and flue system even to the top of the stack are filled with unburned gas. The purpose is to burn the gas in contact with the ware from top to bottom of the kiln, thus getting flame temperature. At least during some stage between the firing periods there is perfect combustion conditions and maximum temperatures and the setting is practically such as to quickly develop the temperature in the bottom of the kiln. Magnesite bricks are set with silica bricks. Where both products are made in the same factory we hare seen the magnesite bricks set with silica bricks header and stretcher in the usual checker fashion, except that the header courses were alternate silica and magnesite, and the stretchers were all silica. The setting was such that the stretchers broke joints on the silica headers. The silica bricks being slightly larger than the magnesite and expanding in the heating up, the magnesite bricks carried no weight. In other words, the silica bricks were set to form a series of pigeon holes or boxes in alternate courses, and in these boxes were placed the magnesite bricks — one in each space spanned by the silica stretchers. Some factories make only magnesite bricks, and these are set in burned silica brick stands. One method of such set- ting is shown in Fig. 137. The boxes are built up of silica bricks and covered tightly with flat courses of silica bricks, in four brick benches. Each box is filled with magnesite bricks 248 BURNING CLAY WARES. placed on end. In the upper part of the kiln there are set all magnesite bricks to whatever height they will stand, usually six to eight courses, in alternate finger spaced and tight courses. By such setting the kiln content is about 40 per cent, magnesite. The burning is the same as that of silica bricks and the temperatures required are around cone 20 — 2780 degrees. Pottery, porcelain, abrasives, etc., are burned in saggers and the problems involved are those of filling the saggers, supporting the ware, etc., and are beyond the province of these articles. The filled saggers are set in the kilns ill “bungs,” and the chief difficulty is to place the “bungs” in the proper position in the kiln to get the desired temperature for the ware enclosed. The setting in a ring kiln and in a direct coal-fired cham- bered kiln require some knowledge of the operation of such kilns, particularly the ring kiln. The draft in a ring kiln is largely horizontal, and unless the fires advance properly there is danger of losing them, and greatly delaying the burning of the ware, or of getting unsatisfactory results. The heat should be kept in advance in the bottom of the kiln, which is difficult to do if the top of the kiln is set very open, or if there is too much free area over the top of the ware. The tendency in such instances is that the gases from the trace flues will rise vertically from the coal and then turn forward in the upper part of the kiln, completing the combustion there, and thus finishing the ware in the upper part of the kiln before the bottom ware is finished. We then get the top burned first, and have a maxi- mum temperature on top in advance of the fire below. If the setting is relatively close on top, the air and gases are forced to follow the trace flues, thus keeping the heat in advance on the floor of the kiln, and the upper ware may be finished by radiation, conduction and convection, and keep in pace with the bottom ware. A ring kiln constructed with drop arches at intervals of about twelve feet has a decided advantage, in that the arch acts as a damper to retard the flow of the gases in the open space between the ware and the crown of the kiln, and forces them down among the ware. Sometimes a dead wall of green ware is set in conjunction with the drop arch, thus actually dividing the tunnel into separate compartments, connected. BURNING CLAY WARES. 249 one with the other by the trace flues. This gives a fresh start in every section, and it is impossible for the top to get much in advance of the fires in the trace flues. To the con- trary, it may be necessary to set the ware to collect a por- tion of the fuel at different levels in order to get a uniform burn. The bricks forming the vertical firing flues are stag- gered in such a way as to collect the fuel on the edge of the projecting bricks, and as it spills it is caught by lower courses of bricks step by step to the lower part of the kiln, thus spreading it out in a fan or inverted “V” shape, prac- tically covering the width of the kiln. Sometimes loose bricks are set in the vertical flues, which at any stage in the firing may be turned flat to provide a shelf for the collection of fuel at any level or series of levels. Otto Bock and Ernst Schnatolla, and other German engi neers, describe and illustrate methods of setting and kiln con- struction which give, in a ring kiln, firing compartment sep- arate from the ware and in some instances convert the kiln into separate compartments. American clayworkers are not much interested in these changes. Some of the changes in- volve a lot of dead work which would be prohibitive in Amer- ican factories, while others are adapted to small kilns only, which are much more common in Europe than in America. American practice requires large kilns and a maximum capacity of some kind of ware, and we cannot afford to muss up our kilns with a lot of equipment or a variety of wares not essential in other types of kilns. Instead we scrap the kiln and build the better type. 250 BURNING CLAY WARES. CHAPTER X. THE CONTINUOUS KILN. I T IS a question whether we should not adopt a more dis- tinctive name. The common up-draft kilns as operated in the Hudson River and Chicago districts are continuous in their operation and are often called continuous kilns. The setting, followed by the burning, cooling and drawing, pro- gresses in stages from end to end of the kiln shed, a prac- tically continuous operation, though not in the sense as we apply it. The designation “continuous” is used, however, and leads to some confusion. The connection and continued operation of a number of periodic kilns in order to make use of the waste heat is just as much continuous as the operation of a battery of compart- ments. Such a method of burning is usually termed a “Sys- tem” and bears the name of its promoter. We designate a single battery of compartments, as semi- continuous, in consequence of the independence of successive burns. Each burn starts at one end of the series of com- partments and is completed at the other end. Subsequent burns are merely repetitions and are entirely independent of th preceding operation. The operation of the kiln as a whole is periodic, but the several compartments are in series, and the progress of the fires during each burn is continuous, as we use the term. If we introduce a return heating flue the semi- continuous kiln becomes fully continuous in its operation, but its designation is the same. If we build a duplicate series of compartments for the re- turn, the double series becomes a continuous kiln, or in other words, two semi-continuous kilns are a continuous kiln. The term continuous applies to the method of operation, hut instead we should have a name descriptive of the prin- ciple involved. Continuous kilns are sometimes called regenerative, and this is a better term, in that it refers to the principle. Strictly BURNING CLAY WARES. 251 speaking, there is no regeneration of the gases, no renewal of the heat value, but the term regenerative has become firmly fixed in metallurgical operations, and means heating up the incoming gases by the waste heat from the hearth, which is identically the same principle as our continuous kilns. Economizer, which in steam and combustion engineering means recovering for use in the boiler or hearth the other- wise waste “heat from it, is a better term than either continu- ous or regenerative. Regenerator in metallurgy is identical with economizer in steam engineering, but the latter is the most consistent. The continuous up-draft kiln operation mentioned above could not legitimately be called a regenerator or economizer operation. The semi-continuous and continuous kilns are truly econo- mizers. The periodic down-draft kilns in series are regenerative, but we would not designate them as a kiln, instead and X»roperly, when adapted to the utilization of their own waste heat, they become an economizer system. There are many economizer kilns in use and in prospect in this country. It is not our purpose merely to describe them. Rather we wish to show the development of the prin- ciples and the chief modifications. There are a number of excellent kilns which we will not mention simply because they differ slightly from others herein presented. We have in mind that the presentation of principles may be of some assistance to economizer kiln operators in improv- ing their kilns. The study of the development of kilns is interesting and one can frequently trace the idea back to other simple types of kilns and improvements are often com- bination of existing types. This leads to progress in the devel- opment to which purpose this presentation is dedicated. Economizer Kilns in General. The great factor in favor of the economizer kilns is the saving of fuel. Fifty per cent, saving is a conservative esti- mate, and many operations will show 60 to 70 per cent. The development of these kilns in this country was slow for the following reasons : 1) American coal is high grade, and in the past cheap. Any requisite temperature can be obtained from our coals by direct cold air combustion and there is no need to adopt econ- omizer kilns to get temperatures from low grade fuel. 252 BURNING CLAY WARES. (2) The early kilns were small — in general having a weekly capacity scarcely equal to the daily output of an Amer- ican factory. (3) High priced American labor. It is frequently claimed that the labor cost of operating an economizer kiln is less than that of a battery of periodic kiln of the same capacity. This claim can be substantiated only in isolated instances, and certainly not in the operation of the small early kilns. As the capacities of the conomizer kilns increase from 5,000 to 20.000 bricks per day up to present capacities in excess of 100.000 bricks per day, there will undoubtedly be some econ- omy in labor, but taking the average over the country, past and present, the records will show no economy in labor, and in many instances an increased labor cost sometimes offsetting the saving in fuel, though on an average the net balance is in favor of the economizer kiln. However, labor cost had its effect in retarding the growth of the kiln. (4) Unsatisfactory operation of early kilns. Our first kilns were of foreign design and we simply enlarged them to meet our needs and thereby got into difficulties. Moreover, the small foreign kilns burned two and even three kinds of ware to accommodate the variable temperatures, whereas the Amer- ican output in the enlarged kilns was of one kind only. (5) America demands a better quality of ware than that in foreign countries. Our facades are a surprise to visiting engineers. Many of these effects were impossible in the earlier kilns, and some of them are impractical in the modern kiln, or at least the modern kiln has not yet proven its adaptability. Besides this there were excessive, scumming difficulties, and a swelling of the product due to bloating in consequence of lack of oxidization, also streaked edges in certain line of face bricks. These troubles are attributed to sulphur gases, the sulphur not only coming from the coal, but often in greater quantity from the ware itself, thus developing a much larger volume of sulphur gas than in a down-draft kiln. Limey clays, which ordinarily burn a pale red to greenish buff, and in mass altogether unsightly. The lime-iron-silicate red, sometimes buff, with all intermediate shades irregularly intermingled, in fact individual bricks are streaked red and buff, and in mass altogether unsightly. The lime-iron-silicate buff color is materially aided in its development by reducing kiln conditions, which were easily obtainable and during some stages of the firing normally prevailed in down-draft kilns, BURNING CLAY WARES. 253 while in the economizer kilns the large excess of air main- tained unfavorable oxidizing conditions. Seger explained the cause of this behavior and how to correct it. (6) The cost of the kiln. Many maunfacturers balked at the high cost of an initial installation, although piece-meal they invested a larger sum in periodic kilns. While the development in this country has been slow, yet it has gone steadily forward and today the kiln occupies a prominent place in our industries. The capacities have been enlarged to meet our requirements and the results have been improved until in several lines of common wares they equal those of the down-draft kiln. In consequence of this advance the economizer kiln has largely entered the field of the down- draft kiln. In common bricks, which were largely burned in up-draft kilns, the desire for a longer campaign, the demand for a better product, the need for reduction in cost, however small it might be, to meet competition, and the necessity of elimi- nating the smoke and gas nuisance, especially in city districts, have led to the replacement of up-draft kilns by economizer kilns. Face brick have taken up the kiln in some degree, but not extensively because of the difficulty in producing the varied color effcts which the trade demands. Salt glazing has not been successfully done in the econo- mizer kilns, although one or two kilns have been built for this product, the results have been unsatisfactory, and other lines of ware had to be developed to keep the kilns in opera- tion. Several Mendheim kilns in Germany are said to be pro- ducing salt glazed ware, but this kiln in the type used for salt glazing is in a measure a periodic down-draft kiln and it does not fully use the economizer principle. We are informed that when the salt glazing period is reached in any compartment it is disconnected, salted and shut off and so remains until cool and emptied. Years ago the writer experimented with an economizer kiln in salt glazing and had some excellent results, but the operation as a whole was not satisfactory and the work discontinued. It seemed, however, to be possible, and it is strange that in the intervening twenty-five years the problem has not been solved. It requires a kiln so designed that the compartment being salted can be cut out of the series during the salting period and afterwards until the salt fumes have been driven off. We have seen plans of a kiln in which 254 BURNING CLAY WARES. such operation is possible, but we do not know of any kiln having been built. The fire brick industry was the first to take up the econo- mizer kiln in this country and the last to make any extensive use of it. A gas fired tunnel kiln was built in Maryland more than thirty years ago, and also a small kiln was built in Illinois. These were the first so far as we know, and were in use until within the last year or two. The third was built in Ohio in 1884-5 for face bricks, but the product was changed to fire bricks and later came back to face bricks, although they were made from fire clay and differed from fire bricks only in size. This kiln is no longer in use. Two gas-fired kilns were built in Pennsylvania in 1891 and one in Ohio the same year for fire bricks, but all were abandoned after a short trial. About this time, or a year or two later, two economizer kilns were built in Missouri for fire bricks, but neither are in operation today. There are a number of econo- mizer kilns burning other products made from fire clays and which are used in burning fire bricks when the demand justifies, but the fire brick companies making fire bricks ex- clusively have not taken kindly to the economizer kiln. It is inexplicable that the industry which first took up the economizer kiln in this country and for which such a kiln is particularly fitted because of the possible high temperature from economizer operation, should not have successfully developed it, especially in view of the fact that refractory ware is less exacting in its temperature range than other lines of ware for which the economizer kiln has proven successful. Recently the fire brick industry has taken up the car tunnel kiln and there are four or five installations which are running successfully. Pottery manufacturers have taken little or no interest in the economizer kilns until recently they have taken up the car tunnel kiln and are leading representatives in the use of this kiln. The terra-cotta field was not invaded by the economizer kiln except in a single instance in 1891, and the kiln in ques- tion was shortly abandoned. Both the ring kiln and the compartmenet kilns are exten- sively used in the manufacture of common bricks, face bricks, paving bricks and hollow ware. The compartment finds a wider field in face bricks, but since the higher product includes the lower, it is equally efficient in the lower grade wares. BURNING CLAY WARES. 255 The determination of the type of kiln, tunnel or compart- ment for any product is a question which requires considera- tion. The advocates of the car tunnel kiln voice the opinion that it will displace the simple tunnel and compartment types, but this is inconceivable. The compartment kiln is a more advanced type than the tunnel kiln and it will have the preference in higher grade wares. In its simplest form the chief advantage is a greater degree of down-draft, which is better than horizontal draft. In the more advanced types the draft is fully down, and the firing is not in contact with the ware. We thus get a better heat distribution and a cleaner product. In fact, the double gas compartment kilns are identically rectangular down-draft kilns with the economy of heat recuperation. The setting is uniform throughout and there are no trace or vertical flues in the set ware. Moreover, the compartments are large, per- mitting the use of a double track within the kiln for setting w r ork, and the wicket work is reduced to the minimum. In a tunnel kiln, turntables or transfer cars must be used inside the kiln and only one car at the setting face is prac- tical. This must be unloaded and removed before a second car can be placed, thus retarding the rate of setting. The labor per ton of ware is slightly greater in the operation of the tunnel kiln than is in the compartment kiln. The compartment kiln, on the other hand, costs more than the tunnel kiln, and its maintenance is greater, which offsets more or less the greater labor cost in the tunnel kiln. The tunnel kiln probably leads in the number of kilns in use in this country, but so close do they run together that one is not justified in making a positive statement. Each kiln has its field of work and, though the fields overlap, yet they are not identical. The tunnel kiln is used in the manufacture of common bricks, fire-proofing extensively, paving blocks and drain tile. The latter product because of the setting diffi- culty is preferably burned in a compartment kiln. The com- partment kiln is used for the above mentioned products, and in addition is predominant in face bricks and roofing tile. The several types are now firmly established in this coun- try and the high price of coal during the war has induced manufacturers of clay ware to study the kilns and give them much greater attention than in the past. It is likely that there will be some recession in fuel costs, but the price will never return to that before the war, and the result will be much more rapid advance in the use of economizer kilns. 256 BURNING CLAY WARES. Kiln Dampers. The damper in an economizer kiln has been a perplexing problem. The relative vacuity greatly increases the leakage which materially interferes with the operation of the kiln. In the older types of kilns and in some modern kilns, the dampers are seated, and the flues being within the kiln walls and frequently subjected to a red heat, the deterioration of the dampers is rapid and their effectiveness materially nullified. We have used flanged circular covers with an annular flanged seat, sealing with sand, but the covers would bulge, the frames warp, and the sand be carried away by the strong draft. Fire clay blocks on a flat seat were better, but were subject to frequent breakage and were difficult to replace. Cast iron dampers in a flat seat were still better chiefly, however, in that the construction was such that the dampers could be readily replaced, and this we consider an important factor in any kind of a permanent damper. A sliding damper is practically worthless in an economizer kiln, with the possible exception of a heavy fire clay block, or blocks in a cast iron frame, set at such an angle that the weight of the damper will hold it firmly on the seat. A widely used damper is a conical valve in a seat adapted to it. The seat may be a fire clay block or cast iron, and the valve is usually cast iron. This type of damper holds its shape fairly well, but at best it is far from satisfactory. The hood or goose-neck is the only satisfactory connec- tion between kiln and flues. When it is removed and the openings covered and sealed, the disconnection is absolute. There may be some leakage around the covers, but they are fully exposed on the kiln top or on the ground and can be readily inspected and tested. The use of hoods has solved another problem. The flue system in a modern kiln is com- plicated and there are frequently cross-overs. Ordinarily, in making such cross-overs, one set of flues must be under or over the other, but with hoods for the connections the flues may usually be on the same level. It frequently happens that one flue serves two or more purposes, for instance, to supply gas during the burning, for hot air during the cooling, and perhaps for draft in the ad- vanced compartments. The hood or goose-neck will make each of these connections without any possibility of inter- ference from the other two. BURNING CLAY WARES. 257 Open-top Economizer Kilns. The open-top kiln has been illustrated and described under the head of up-draft kilns. Ring or Tunnel Kilns. Efforts to develop an economizer operation during the lat- ter part of the eighteenth century and the early years of the nineteenth century are recorded, but the first kiln having the basic principle of the ring kiln was patented by Arnold in 1839, and although a failure, yet to it may be traced the cause of the nullification of Hoffman’s later patent. Arnold used a horizontal transverse flue for each section in the bottom of, or under, the ware, in which the firing was done, and the position of this flue was fixed. The gases rising from this combustion flue passed forward through heating and watersmoking sections until the heat value was practically exhausted, when they were drawn off into the stack. The combustion air entered through the open sections and advanced through the cooling sections to the burning section. The Hoffman kiln had no fixed combustion flue or shaft, but instead, vertical firing shafts were provided in setting the ware, as in the tunnel kilns now in use. The Hullman kiln, patened in 1854, was distinctly annular since it has the annular tunnel enclosing an annular smoke flue with the smoke stack in the center. The Hoffman kiln was patented in 1858 by Friedrich Hoff- man and A. Licht, and though the patent was twice extended, it was finally abrogated in 1870. The credit for the ring or tunnel kiln, known the world over as the Hoffman type of economizer, belongs to Hoffman be- cause he made the kiln a success, whether he originated the basic idea or not. The annular plan has been replaced by the rectangular arrangement of the tunnels either as a single construction with a division wall separating the tunnels or as separate parallel tunnels connected at the ends by circular tunnels (a flattened ellipse) or simple cross-over flues. Fig. 138 is a cross section and Fig. 139 a plan of a Hoffman type of an economizer kiln. The kiln illustrated is a marked advance over the original Hoffman kiln, or any kiln designed by Hoffman. There have been many modifications of and im- provements on the original kiln. As previously mentioned, the original kiln had a circular tunnel, with included smoke flue and central stack. This gave way to the oblong and rec- 258 BURNING CLAY WARES. tangular arrangement of the tunnel, and finally to separate tunnels — independent single kilns — connected at the ends, thus combining them into a fully continuous kiln. The earlier rectangular kilns, and some of the modern ones, have the draft flue in the longitudinal division wall, with inlets from the tunnel at the floor level, connecting with the main draft flue by uptakes, and the draft is controlled by dampers op- erated from the top of the kiln, as shown in dotted lines in Fig. 138. The stack, as stated, in the circular kilns was in the center, and in the development of the oblong kiln it was natural to place the stack somewhere in the center wall between the parallel tunnels. If the stack draft is to be used, the stack should be included within the kiln walls, because we carry the gases through the ware until they have given up their heat and become fully saturated with moisture. The operation of Figure 138. the kiln depends upon the buoyancy of the gases leaving the kiln, which constitutes the draft,- and where the stack is within the kiln walls, its base is kept hot by conduction and the draft is stronger in consequence. The importance of this is evident from the previous discussion of stacks. In some kilns the draft is from the top by using a hood (goose-neck) over the feed holes on top of the kiln or over special draft holes through the crown and extending the hood to cover a corresponding opening through the top pavement of the kiln into the central draft flue. Where the stack (or fan) is outside the kiln walls, a down take flue draws the gases into a transverse flue under the kiln floor and thence to the stack or fan. In the above described draft arrangement the draft con- nections are through the inner wall, which has the effect of drawing the heat away from the outer wall, where, in conse- BURNING CLAY WARES. 259 Figure 139. 260 BURNING CLAY WARES. quence of radiation losses and leakage, it is essential the draft should be the strongest. The draft connection is always a number of compartments ahead of the burning compartment and one sided draft has no effect on the burning, but it does affect the water-smoking and heating up and thus retards the rate of progress. To correct this one kiln had the draft opening in the outer wall, then down and back under the kiln to the central division wall and up into the main draft flue. Then followed the placing of the draft flue under the tun- nel floors with connections through the outer or inner wall; later we find the draft flue under the outer wall; and finally it is placed outside the kiln walls with goose-neck connections to the kiln. These changes are partially illustrated by Fig. 158, Fig. 159, and Fig. 160. In Fig. 158 we have the annular kiln, an- nular smoke flue, and central stack. In Fig. 159 we have the early rectangular form with a longitudinal central draft flue and the stack within the kiln wall. When the stack was placed outside the kiln wall, the arrangement in its final devel- opment for a single kiln became that shown in Fig. 139, and the plan of the modern kiln is shown in Fig. 160 with the detail of Fig. 139. Dampers which were often necessary when the flues were within the kiln wall gave a lot of trouble and always leaked more or less, and the final location of the flues outside the kiln walls with goose-neck connections was the perfect solu- tion of the damper problem. The earlier kiln had no advanced heating flue by means of which hot air is by-passed from the cooling sections to the water smoking sections, thus overcoming or reducing the scum- ming and swelling difficulties, and increasing the drying effi- ciency, thereby correspondingly increasing the rate of burning. The advanced heating flues are in the upper kiln walls and connections from the cooling sections to the water smoking sections are by means of hoods covering feed holes in the top of the tunnel and corresponding holes in the top of the ad- vanced heating flue. This is illustrated in the description of open top continuous kilns. In the kiln illustrated the advanced heating flue is in the outer wall, but it may be placed in the wall between the tunnels, especially if the draft flue is else- where. Where the main draft flue is in the mid-wall, we may BURNING CLAY WARES. 261 still have the advanced heating flue in the same wall, either alongside or above the draft flue. We speak of sections and compartments in the tunnel kiln, but in the tunnel kiln itself there are no divisions. For con- venience in setting and unloading the kiln wickets are spaced from 12 feet to 16 feet apart, and the draft connections are correspondingly spaced. In setting the kilns we fill the tunnel from one wicket to the next, and blanket the draft with paper pasted to the face of the set bricks. This constitutes a section and each section has a draft connection, and so long as the paper dampers remain in place, each section may be water smoked by the hot air brought from the cooling sections through the advanced heating flue. When this stage of the process is finished the connection with the advanced heating flue is broken, and holes are torn in the paper damper nearest the fire sections and as the fire advances, the damper is ignited and thus completely removed. In one kiln, perhaps in several designs, a drop or apron arch, shown in the illustration, is introduced in the kiln crown at intervals of 12 feet to 16 feet. This arch drops 12 inches to 15 inches below the kiln crown, and it is simply a diaphragm to deflect the gases and air downward among the burning and cooling bricks. Without this arch, especially where the ware is not set close to the kiln crown and where the settle is con- siderable, we have a continuous free passage between the crown and the ware along which the air will move and the purpose of the drop arch is to break up and deflect this movement. Many engineers hold that all the heat in the kiln is useful in the kiln operation, and that no heat should be taken for outside work. One prominent tunnel kiln designer provides openings in the kiln crown connecting with wall and under- ground flues leading to an independent dryer and uses the heat of the cooling sections for drying purposes outside the kiln walls. Any heat thus taken from the kiln, must be re- placed by fuel burned in the kiln, but in view of the fact that combustion in the kiln is absolutely complete and smokeless, while that in an independent furnace is very imperfect, it is reasonable that it is economical to generate heat for drying purposes in the kiln. Such a kiln will show higher fuel con- sumption than a kiln from which no heat is taken, but if we add to the latter the fuel required to dry the ware, the balance will be in favor of the former. 262 BURNING CLAY WARES. The same kiln designer introduces furnaces in the wickets and these use cold air, as do the furnaces in any periodic kiln. The top firing, as in any direct coal fired kiln, is through feed holes in the crown, vertical shafts in the set ware, into trace flues in the set ware at the floor level. These fires are kept in advance and serve to burn the ware in the bottom of the kiln and may be advanced as rapidly as the bottom ware can be burned. Following these, the side fires are started and the flame from them surrounds the mass of ware, sides and top and completes the operation. There is much to be said, pro and con, relative to such a kiln which departs so radically from the principles of an econo- mizer kiln, but in view of the fact that the kiln has been adoted by American clayworkers in greater degree than any other economizer kiln, and that it is successfully burning the difficult paving brick product, the merit of the kiln must be conceded. The self-contained rectangular kilns require that the ware shall go in and come out the same doorway in the face wall of the kiln. If depressed railroad tracks are put on each side of the kiln to handle the burned ware, as it comes from the kiln, and since transfer cars paralleling the depressed tracks are needed to handle the ware from the dryer to the kiln, there is considerable interference in the work. The modern kiln, having independent tunnels, with two wickets on opposite sides for each section, permits the green ware to be handled in the space between the batteries, while the burned ware leaves the kiln through the wickets in the outside walls. The tunnel kiln is not as convenient for setting as the chambered kiln. The ware is delivered to the kiln over a trans- fer track, and thence into the kiln on a spur track, and from this to a transfer car inside the kiln to the working face. When a car is unloaded it must be transferred back to the wicket, and out before another load can be brought in. In some instances a portable turntable is used in the kiln in the place of the kiln transfer track and car, but this is simply a matter of preference and is without effect on the periodic setting operations. The compartment kiln with wide wickets permits the use of a double track as in rectangular periodic kilns, and no time need be lost by the setting gang in consequence of switching cars in and out. BURNING CLAY WARES. 263 The setting in tunnel kilns requires trace flues in the bottom and vertical flues corresponding with the top feed holes. A similar flue system in the set ware is required in the direct coal fired chambered type of kilns, but the gas fired chambered kiln has the same setting as in a down-draft kiln. A tunnel kiln should not have fewer than 16 sections and preferably 20 or 22 sections. From 6 to 9 sections are cooling, 2 to 3 sections burning, 3 to 4 sections heating up, 2 to 3 sec- tions water-smoking and two or more sections for working space. Occasionally a kiln is built of such a number of sec- tions that it may be double fired, or three, four or five sets of fires, but such kilns are merely combinations of several kilns in a single construction, and each fire has its related cooling, heating up, water-smoking, and working sections, and it is oper- ated as an independent kiln. A description of the many ring kilns tried out, in use, or on the market in this country would be voluminous. They all start with the Hoffman kiln and the improvements consist of a rearrangement of the air and gas flues, the addition of advanced heating flues, drop arches, perforated floors, perma- nent trace flues, outside furnaces, adaptation to producer gas, oil, etc. We do not wish to intimate that the changes and additions have not improved the kiln — quite the contrary. The kiln in its highest development occupies a prominent position in the clayworking industries at the present time, probably ranking first among economizer kilns in the quantity of output. Zig-Zag Kiln. The zig-zag kiln is merely a rearrangement of the ring kiln, giving greater compactness. Such a kiln is illustrated in Fig. 140. There are several modifications of this type of kiln both in the arrangement of the several pseudo chambers, and in the floor and draft exits. In some the plan is simply a single battery, with an underground return flue. This has the advantage that the ware enters on one side and is removed from the opposite side. In other plans the battery of compartments is single, but the return flue is above ground and has the same cross section as the other compartments, in other words, a longitudinal com- partment connects the ends of the battery of compartments, making the operation of the kiln fully continuous. Others are the double battery type, shown in the illustra- tion. 264 BURNING CLAY WARES. The kiln never found favor in this country, so far as we know, and a detailed description of any of the type would not interest American clayworkers. It seems strange that the zig-zag kiln, with its greater compactness and lower radiation loss, has not found some acceptance in this country. Compartment Kilns. The compartment kiln is in a measure a development of the tunnel kiln. The earlier types of tunnel kilns were direct coal fired and the burning fuel came in contact with the ware. This would not be objectionable for a number of common wares, L9_ _i . _9 1 1 _ ~ i~_ p_ i"_v. Fig. 140. but for other wares it was essential that the fuel be burned in a separate compartment. The simplest and earliest effort to separate the high-grade ware from the fuel was to build the trace and vertical firing flues of common bricks and to fill the space between with the higher grade ware, but this involved the manufacture of two or more kinds of ware simultaneously, complicated the setting and increased the labor cost. The first step toward a compartment kiln was to introduce, in the ring kiln, division walls at intervals, the walls being double, with space between. In this space were placed so- called step grates upon which the fuel collected and openings in the division walls provided for the admission of air on one side and the exit of the gases on the other. BURNING CLAY WARES. 265 Fig. 141, view plan below the crown, and Fig. 148, elevation, illustrate such a division wall-firing pocket. In a modern American ring kiln drop or apron arches are introduced in the tunnel crown at suitable compartment inter- vals. The green bricks are set in front of the aprons to form Fig. 141. Fig. 142. a solid wall except openings through the base corresponding to trace flues. This green brick wall is virtually a division wall. The bricks may shrink and the wall settle to the depth of the deep arch without opening a direct passage from compartment 266 BURNING CLAY WARES. to compartment. Back of each apron wall are slots in the kiln side walls, and in line with these and inserted into them is a second solid green brick wall across the tunnel, but drop ping off below the kiln crown. This second wall is built of bricks set herring bone fashion so that under fire shrinkage the bricks will settle together and maintain a relatively tight wall. The extension of this wall into the side wall slots pre- vents any openings around the ends of the walls. This second wall diverts the gases from the trace flues to the kiln crown, and the initial wall brings them back to the trace flues in the next compartment. The firing is with producer gas introduced through a series of ports a short distance back of the deflecting division wall. Fig. 143, longitudinal section, illustrates the above described division of a ring kiln into compartments. An English kiln introduced into this country divided the Fig. 143. tunnel into a series of compartments by permanent division walls. Each division wall was perforated at the base by a number of ports, thus connecting the compartments. In front of each division wall ; that is, on the firing side, was placed a low box to receive the fuel. A section through one port in the base of the division wall is shown in Fig. 144. The hot air from the rear cooling compartment is split as it passes through the division wall ports. Part of it enters the fuel box at the floor level and is drawn up through the fuel, thus giving the primary combustion. The second portion enters above the fuel box and combines with the products of primary combustion. The operation is that of a gas producer, and it may be said that, crude and simple as it was, little fault could BURNING CLAY WARES. 267 Fig. 145. 268 BURNING CLAY WARES. be found with this feature of the kiln. In a subsequent con- struction, the division walls became the support of the com- partment arches, following the usual construction of a com- partment kiln. Another (German) method of combustion apart from the ware is shown in Fig. 145, a view above the floor on the left, Fig. 146. and similarly below the floor on the right, and Fig. 146, a vertical section. Here we have side firing spaces with firing plates to receive the fuel, similar to the step grates, with the hot air entering from below. The use of step grates is com- mon practice in Germany, if we may judge from the literature, while in this country they have not proven successful. A German counterpart of the English kiln (Fig. 144) is illustrated in Fig. 147 and Fig. 148. It was necessary to use lump coal in the English kiln, whereas in the German kiln with BURNING CLAY WARES. 269 the step grates, powdered, or at least fine coal is essential, and a further essential is that the coal should be easily in- flammable and highly gaseous. Brown coal, lignite and similar volatile and gaseous fuels are especially adapted to the step grate combustion furnaces. This is probably one reason why the step grates in any application have not found favor in this country. In the German kiln above illustrated the coal is dropped through feed holes in the kiln crown and collects on the fire clay plates. Hot air from cooling compartments in the rear enters the burning compartment through the ports in the divi- sion wall above and below the plates holding the fuel. The hot air from below heats the plates, gasifies the coal and finally unites with the residual carbon in its passage over the top of the plates. The secondary hot air from the upper ports completes the Fig. 149. combustion. The operation is crude, but the final results are perfect in so far as complete combustion is concerned. The step grate principle in other kilns is further carried out by increasing the height of the fire wall and putting in a series of step grates, as shown in Fig. 142 and Fig. 147. The next advance is the double fired step grate kiln, shown in Fig. 149. The arrows show the movement of the air under the floor, to the fire bags, thence up around the step grates, over into the kiln, and down through the ware and kiln floor, repeating the operation in the succeeding compartments until the draft outlet is reached. Any clayworker will understand that the distributing air flues in any compartment, one of which is shown in the illustration, alternate with perforated floor 270 BURNING CLAY WARES. flues leading into the under main draft flue, thence forward into the distributing flues in the next compartment. The col- lecting flue in one compartment becomes the distributing flue in the next compartment. The same principle of collection and distribution of the air and combustion gases is used in later producer gas double fired kilns with marked success. An early kiln developed in this country is shown in Fig. 150, section, and Fig. 151, a partial plan of two compartments of the kiln including the central longitudinal division wall. The floor is solid. The division walls have ports in the bottom spaced about 27 inches. The draft connection is in the corner of each compartment, with an up-take flue into the bottom of a main draft flue. The kiln is simplicity itself. The ware is set away from the division walls on each side, thus forming combustion spaces in which the firing is partly done through the outer feed holes. Under the intermediate holes the ware BURNING CLAY WARES. 271 is set to form step grates, thus spreading out the fuel fan- wise in the mass of ware from top to bottom. The combustion gases from the lateral spaces are partly drawn through the lower courses of the ware by means of trace flues set in the ware, and partly pass over the top of the ware, thus burning the top and bottom. The fuel within the mass of ware performs its w T ork locally, and a fairly uni- form result is obtained throughout. These results, however, depend largely upon intelligent setting and firing. A weakness of compartment kilns is the tendency of the division walls to lean in the direction of the draft, and in consequence throw the arch, as shown in Fig. 152, and after a few years’ use the kiln has to be rebuilt, which, considering the cost, is a serious matter. The life of the kiln depends upon the thickness and solidity of the division walls, their height, the quality of the structural material and workmanship, the rise of the crown and the temperatures attained. It has been suggested that the kiln should be designed so that the firing can be reversed annually or semi-annually. If a kiln is well built the slight displacement which would occur in six months would be corrected by a reversal of the direction of the firing for a like period or longer, as may be required. A simple construction, such as shown in Fig. 150, could easily be designed for such a reversal of the draft. It would only be necessary to carry the compartment draft con- nections through the middle wall, as shown by the dotted lines in Fig. 151, and block them up on one side or the other with temporary walls. A reversal of the firing then would only require that the kiln be burned off and emptied, immediately refilling and firing in the opposite direction, changing, of course, the temporary w r alls blocking the draft ports. In a kiln with the draft connections in the corners of the 272 BURNING CLAY WARES. compartments, one would say that there would be a diagonal tendency in the movement of the kiln gases. This is true, but it does not materially affect the burning compartment, because the draft control is always several compartments ahead of the firing compartment. The cooling effect of the wicket and outside wall forming the end of the compartment would result in underburned ware were it not for heavier firing in the end feed holes, and even then we often do not get the ends as fully burned as the remainder of the compartment. The diagonal draft does effect the water-smoking and heat- ing up and lagging in this work often continues through the burning. Fig. 153, section, and Fig. 154, view plan of the division wall between compartments, illustrate a modification of the preceding kiln to eliminate the diagonal draft. Between the ports in the division wall connecting the compartments are up-take flues into the bottom of a cross collecting flue which in turn enters a longitudinal main draft flue. If the sizes of the division wall flues are properly proportioned we are assured of uniform draft from end to end of the burning com- partment. It would, of course, be a simple matter to provide dampers for each of the up-take draft flues, but this is not con- sidered necessary. Even imperfect operation so far breaks up BURNING CLAY WARES. 273 the diagonal tendency that any material or noticeable retarda- tion is impossible. Another method of overcoming the diagonal draft and at the same time in some degree counteract the cooling effect of the outside wall, is to put a draft outlet in the outer as well as in the inner wall with a down-take from the former to an under-ground cross flue leading to and connecting with the main draft flue, or an up-take to an upper cross flue. Such a flue arrangement with proper draft control enables us to have the draft all to the outer wall, all to the inner wall, or balanced as may be desired. When we consider that any damper within a kiln structure leaks more or less, and in con- sequence there is always some draft through the damper con- Figure 155. nection of the burning compartment, it becomes more important that the tendency of the draft should be toward the outer wall, where radiation losses are the greatest. In kilns which have been designed and put in use, one can generally trace a development from the simpler form to over- come some of the difficulties, but occasionally there is a marked departure from earlier types. Fig. 155, section of the kiln, Fig. 156, plan, and Fig. 157, 274 BURNING CLAY WARES. section of the division wall, show an American kiln which differs materially from the general plan. The movement of the air and gases in the compartments is longitudinally of the compartment, reversing in the succeed- Figure 156. ing compartment, in other words, it is that of a zig-zag kiln, and it could very properly be termed a zig-zag compartment kiln. Presumably the firing in each compartment is more or less progressive. As the fires slacken, or cease in the entering end of the firing compartment, air becomes available to start or increase the fires toward the other end. The connection from compartment to compartment is a single opening at one end of each division wall, alternating so BURNING CLAY WARES. 275 that the movement of the air and gases is forward through one compartment and backward through the next, continuing thus from the entering cooling compartment to the exit water- smoking compartment. The collecting draft flues are between kiln crown arches, above the crowns, immediately over the division walls, but each extends only half the length of the compartments, and they alternate in the successive compart- ments. These flues are connected on the underside of the kiln crown to the adjacent firing holes, and at the outer ends con- nect with a down-take flue in the outside wall, thence into an under-ground main draft flue. The compartment connection is in one end of the division wall and the draft connection is in the other end, the position alternating in the consecutive walls. Trace and vertical firing flues are set in the ware and the firing is the usual top coal operation, with the addition that, through holes in the side and wicket walls corresponding with the trace flues in the ware, the accumulations of ashes in the trace flues under the feed hole flues can be raked out, or lev- eled down and thus keep the active combustion near the kiln floor and provide free passage for hot air through the lower courses of the ware at all times. The operation of the kiln is in outline as follows : The air enters as usual through the compartment from which the ware is being drawn, and traverses the several cooling compartments longitudinally, forward and backward, finally entering one end of the burning compartment where it comes in contact with the fuel. The combustion gases continue the forward movement through the heating-up compartment into the water-smoking compartment, where they are drawn off through the connected feed hole shafts into the collecting draft flue, thence to the down-take and main draft flue. The top draft is claimed as a feature of the kiln, in that as the vapor rises from the water-smoking ware it is drawn off to the stack instead of being drawn down through the ware and condensing thereon as frequently happens in down-draft opera- tion. Kiln Arrangement. A factory should be designed so that there will be a con- tinuous forward movement from the clay supply to the car loaded with the finished ware. This does not mean that the operation must be strung out in a line, quite the contrary, but 276 BURNING CLAY WARES. it does mean that the product shall not cross itself at any point and thus interfere with the general forward movement. The original ring kiln was annular (Fig. 158) and decidedly awkward for the movement of ware which can be most eco- nomically moved in straight lines and right turns. These early kilns were probably filled and emptied by means of wheelbarrows, or perhaps the ware was carried in and out by hand. A circular track could be put around the kiln with a turntable opposite each doorway, or better, a circular transfer track with a turntable car, but with any arrangement the kiln is not adapted to modern methods. The plan was changed to an oblong shape (Fig. 159), which is an improvement, but still there is more or less inter- ference in the setting and unloading since the ware must come Fig. 158. Fig. 159. out the same doorway through which it is taken in and the setting and unloading operations are not far apart. This double operation on the same side of the kiln is complicated by the fact that the transfer tracks for green ware are slightly de- pressed and the loading tracks for the finished ware deeply depressed. One or the other must be crossed by the moving ware, and there are frequent delays and numerous minor acci- dents. The final and satisfactory plan (Fig. 160) has the tunnels separated and the ware enters through the inner doorways and is taken out through the outer doorways. The early compartment kilns (Fig. 161) have the same diffi- culty as the ring kilns shown in Fig. 159, and to overcome this, particularly where it was impractical to have a spur track on each side of the kiln, so-called semi-continuous kilns were built (Fig. 162) and the operation became continuous by means BURNING CLAY WARES. 277 of a return flue connecting the ends of the kiln. It is prac- tical to build a kiln with ten or even twelve compartments and get satisfactory operation through such a return flue. It is but an extension of the semi-continuous kiln to build Fig. 161. 278 BURNING CLAY WARES. a parallel duplicate battery of compartments, connecting the two batteries of compartments by cross-over flues and thus get the fully continuous kiln as shown in the dotted lines in Fig. 162, which gives the same yard arrangement as Fig. 160. The zig-zag kiln illustrated in Fig. 140 has a large part of the ware coming out the same doorway through which it entered, but a single battery kiln with a return flue may be built as shown in Fig. 163. Such a kiln develops a very long T V Fig. 162. tunnel in a comparatively short space — in other words, a com- pact kiln and factory arrangement. Producer Gas Economizer Kilns. We do not know when producer gas was first used in this country in the production of clay wares. At Mt. Savage, Md., there was in operation in the early eighties, and up to within a few years ago, a tunnel kiln fired with producer gas. The producer was built on a car and traveled forward as the fires progressed. It was scarcely more than a large port- able box furnace, without, it is our recollection either steam or air pressure, in which the combustion was very imperfect BURNING CLAY WARES. 279 from either the standpoint of complete combustion or producer gas development, but being movable and at all times in close touch with the burning compartment, thus getting full benefit of the sensible heat in the combustion gases, it was not impor- tant that a high-grade gas be developed. Producer gas operation in foreign countries was much earlier than in this country. Mendheim developed a producer gas kiln and put it in operation in 1867. The first kiln was very simple. If we were to take the kiln illustrated in Fig. 150 and introduce a gas flue under or in front of the division wall below the kiln floor level, with ports in the crown of this flue to mate with the hot air ports in the division wall, and build a bag wall in front of the division wall, we will have the first type of producer gas compartment kiln. The early introduction of producer gas in tunnel kilns was 1 Fig. 163. equally simple, and in this operation it is doubtful whether we have progressed any. In the early installation a series of cross fines carry the gas from the main gas flue outside the kiln wall to and under the tunnel floor and ports in the top of these flues correspond- ing to the ordinary feed holes in the kiln crown deliver the gas into the kiln tunnel. As the setting of the ware progresses each gas port is converted into a vertical flue to the top of the setting by the use of perforated clay pipes closed at the top. The combustion of the gas jets from the pipes is com- pleted by the hot air coming forward through the cooling ware in the tunnel. This is a German method described by Schmatolla. 280 BURNING CLAY WARES. Bock shows the same method except that the gas is intro- duced from above through the feed holes. An American method shown in Fig. 143 and described in connection therewith introduces the gas through a series of ports in the kiln crown from side wall to side wall. A drop brick or baffle in front of each port serves to spread the gas and air. The heat is forced to the bottom of the kiln, as shown and described. The original Mendheim compartment kiln, which has been repeatedly illustrated and described, carries the air forward under the division wall and across the compartment in a series of parallel flues. Between these are gas flues supplied with gas from under cross flues connecting with outside main gas flue. The gas and air are brought together through small openings in the kiln floor. The firing is over all parts of the kiln floor and in direct contact with the ware, resulting in overbumed ware in the bottom and underburned ware on top. This type of kiln has been superseded by the later and more satisfactory down-draft types of kilns. Such a later kiln is illustrated in Fig. 164, section, and Fig. 165, view plan. The gas is delivered around the kiln in underground flues, and cross flues under the kiln floor deliver the gas to ports in the bottom of the bags. The hot air is brought forward from the cooling compartments through underfloor flues corresponding with the gas ports. The draft connection is at the floor level in the corner of each com- partment. The Dunnachie, a Scotch kiln, Fig. 166, had three installa- tions in this country in 1891 ; one, of two kilns, in Pittsburgh, Pa., and one in Portsmouth, O., for fire bricks, and one in Perth Amboy, N. J., for terra cotta, but all of them have long since been dismantled. The gas was introduced through a single BURNING CLAY WARES. 281 flue across each compartment, instead of two flues as in the previously described kiln. There was no combustion bag, and instead of four gas ports there were a series of small ports in the kiln floor immediately in front of the division wall. The air for combustion was collected behind the division wall, as in Fig. 164, with the difference that the transverse flue was under the division wall and had two damper controlled ports through which the air was delivered into an upper division wall transverse flue from which at the floor level were a series of ports corresponding with the gas ports. The air entered horizontally, and projecting into each stream of air there was a vertical jet of gas from below. The ware was set away from the division wall to form a combustion space equivalent to a bag space. The illustration and description applies to the kiln as introduced into this country in 1891, and we do not know what changes and improvements may have been made in the kiln up to date, but so far as we know the three kilns men- tioned were the only ones ever tried out in this country. The Youngren (English) kilns, illustrated in Fig. 167, longi- 282 BURNING CLAY WARES. tudinal section, and Fig. 168, cross section, introduced into this country a number of years ago, marks the beginning of suc- cessful operation of producer gas fired compartment econo- mizer kilns. Whatever the merit of a foreign kiln it requires time, patience, ability and money to adapt it to Ainerican con- ditions. We took English and German plans worked out for capacities of 30,000 to 60,000 bricks per week and simply en- Fig. 167. larged them to get similar outputs per day, and the result in most instances were disastrous. Mr. Youngren’s unbounded faith in his kiln, his recognition and admission of its faults, his ability to overcome such faults, and his untiring energy in presenting the kiln to American clayworkers have won for the kiln the prominent position it holds today. The number of kilns of the Youngren type which have been BURNING CLAY WARES. 283 built in this country has given opportunity to develop the faults and correct them in so far as the type will permit and to bring the kiln to its highest efficiency. The kiln has been improved not only by those promoting it, but by those operating it, and at the present time there are more installations of Youngren kilns under several names than any other compartment continuous kiln. A number of the designers of compartment kilns on the market have been oper- ators of Youngren kilns and their kilns are extensions of and improvements on the Youngren kiln. The promoters of the Youngren kiln have developed new plans which are patented, yet the ground work of these new kilns must be credited to Youngren, although they must be counted as improvements and not as infringements. The most marked advance is in the double fired type. The Youngren kiln is illustrated in Fig. 167, section, and Fig. 168, section of the division wall. The gas is carried through a pipe on the kiln wall or a flue within the kiln wall, and delivered to cross flues between the kiln arches below the kiln pavement through a goose-neck conection or by means of a valve. The position of the gas flue will depend upon whether the kiln is a single battery of compartments (semi-continuous), a double battery in a single, unit, or a double battery in two units. In the modern gas fired kiln we must have (1) a main draft flue the full length of the kiln and extending to the draft fan, (2) cross draft flues for each compartment, (3) a main gas flue, (4) cross gas flues to each compartment, (5) an ad- vanced heating flue, (6) cross-over flues at each end connect- ing the batteries, or a return flue connecting the ends of a single battery. The double unit type is particularly compli- cated, in that there must be two cross-over flues at each end 284 BURNING CLAY WARES. — one connecting the compartments and one continuing the advanced heating fine. Such a multiplicity of flues requires varied arrangements to cover the several modifications of the modern kiln. In the Youngren kiln the flues delivering the gas from the main gas flue to the several compartments are centered in the spandrels of the crown arches. From the bottom of these flues are down-take flues in the division walls, with ports into the bags at the base. Each of these individual gas flues is con- trolled by a valve. The original Youngren kiln, in order to reduce the number of control valves, branched the individual gas flues so that each supplied four ports, or, in other words, four fires, as shown on the left of the sectional drawing, Fig. 168. The modern kiln uses individual straight flues for each fire, as shown in the illustration, and the reason for this is evident to any one who has had experience in the use of pro- ducer gas. It is doubtful whether we should consider the mod- ern kiln as a Youngren kiln. The kiln has been modified and improved and such improvements covered by patents, some of which are recent (1918). The air from the cooling compartments, in the original plan, is drawn down through the ware, though a perforated floor, into parallel collecting flues leading forward through the divi- sion wall into the base of the bag, thus delivering the hot air below the gas port. A question discussed by engineers is whether the gas should be below the air, or vice versa. It is generally conceded that the lighter fluid should be underneath. The maximum flow of a gas in a vertical flue is in the center. If the lighter gas enters the flue above the heavier it will pre-empt the center of the flue and the heavier gas will be drawn up as an envelope of the lighter gas. On the other hand, if the lighter gas is below the heavier it will, because of its greater force, overtake the heavier gas and crowd it aside to get into the path of least resistance, which is the center of the flue. In this way there is a more intimate mixture of gas and air. It is generally assumed that gas is lighter than air and con- sequently should be under the air in a combustion operation. At 62 deg. F. air weighs .075 pound per cubic foot and producer gas about .065 pound. If the air enters at 1,500 deg. F. and the gast at 1,000 deg., the relative weights are .0196 and .0216 — the air being the lighter, and theoretically should be under the gas. It is the author’s opinion that it makes little or no BURNING CLAY WARES. 285 difference, and one might present arguments favorable to put- ting the heavier gas under the lighter. For example, complete combustion at the entry ports is not desirable, because it generates an intense heat at those points, and it is difficult to get the heat over into the kiln. In many instances it is desirable to carry the heat in part latently from the furnace to the ware and complete the development in con- tact with the ware. The modern kiln developed from the Youngren, but no longer the Youngren, is a marked advance over the kiln, as shown in Fig. 167. In that illustration it is seen that the hot air or gas passes from one compartment into the next, and if it were desired it would be practical to have fully perforated floors. In one recent plan the under-floor collecting flues are so arranged that air from cooling compartments may be deliv- ered to at least two burning compartments, and the products of combustion are carried to compartments ahead of the burn- ing compartments, so as not to interfere with the combustion in the second burning compartment. The arrangement is quite simple. Assume six compartments — one had two cooling, three and four burning and five and six heating up. The hot air from compartment one is carried in an under-floor blind flue and delivered at the gas ports in three, and similarly the hot air from two is delivered to four, while the products of com- bustion from three and four are delivered respectively into five and six. A plan is also worked out so that air from one cool- ing compartment may be divided and delivered into two burning compartments without any interference of the combustion gases. A double fired kiln is developed without changing the gas delivery of the Youngren kiln, by simply placing the gas ports alternately on opposite sides of the division wall, with bag walls on each side. The under-floor collecting flues are ar- ranged to deliver air to the several bags. The latest arrangement departs materially from the Young- ren kiln. In fact, one might say that nothing remains except the position of the gas port and the method of delivering the air in the base of the bag below the gas port. This, however, is an incontestible feature of the Youngren kiln. The gas main is now outside the kiln wall and under ground. A series of cross flues, corresponding to the number of gas ports, adjacent to and parallel with the sides of the division walls, underground, in fact, below the level of the under-floor 286 BURNING CLAY WARES. flues, with short right turns to the centers of the division walls and up-takes to the gas ports, deliver the gas to the ports. This method of delivering the gas to the kiln is used in Mend- heims kiln, Fig. 165, in the Dunnachie kiln, Fig. 166, and as will be seen in the Richardson kiln. The chief difference and the vital factor in each kiln is the method of getting the air and gas together at the point of combustion. One may consider the air and gas port as a burner, and it is in the burner that the several kilns men- tioned materially differ. In fact, the validity of the patents largely depends upon the burner. The advantages of a double fired kiln are: (1) Progressive equal heating of the division walls which will prevent, or meas- urably reduce, the drawing over tendency illustrated in Fig. 352. (2) It insures more uniform burns. The Underwood compartment kiln has no openings through the division walls. In all compartment kilns the draft flue is centered in each compartment and connects with a main draft flue outside the kiln walls. The hot air, or combustion gas, is collected in perforated floor flues and carried forward through the division walls, thus connecting the several com- partments. The Underwood kiln collects the air or gas in the usual manner, but when the division wall is reached a flue parallel to the wall carries the air outside the kiln wall into an air main, and it is brought back into the kiln through a similar flue on the opposite side of the wall and delivered through ports into the bases of the bags. The main gas flu’s is underground outside the kiln walls, and the gas is deliv- ered under the kiln through a cross flue for each compartment in front of the bag walls. The gas and air delivery flues are Fig. 169. BURNING CLAY WARES. 287 superposed — the former being underneath and from it the gas is conducted to the gas ports in the division walls, as shown in Fig. 169. The operation of the air flue is similar to that of an ad- vanced heating flue. A compartment, or any number of com- partments, may be by-passed should it be necessary to do so for repairs, or should it be desirable to partially isolate a burning compartment for reducing conditions or salt glazing. When the burning has advanced to the point where reducing Fig. 171. conditions are desired, or for salting, the air supply to the burning compartment could be reduced to any required degree and by connecting this compartment with the main draft flue and disconnecting the air escape, the operation of the com- partment in question would, except in the air supply, be inde- pendent of the other compartments without in any way inter- fering with the operation of the latter. The Richardson kiln is a double fired compartment kiln, 288 BURNING CLAY WARES. Fig. 172. Fig. 173. BURNING CLAY WARES. 289 shown in Fig. 170, section, Fig. 171, plan, Fig. 172, section through the gas connections, and Fig. 173, sectional perspec- tive. As will be seen from the plan the under-floor system con- sists of alternate open and covered collecting flues with the usual central draft flue in each compartment. The collecting flues are arranged in a staggered manner — the continuation of an open flue in one compartment becom- ing the closed flue in the following compartment. Each flue includes two compartments within its length. The open end collects the air, and the covered end in the next compartment delivers the air to the combustion bags. The combustion gases are similarly collected and carried to the heating-up and water- smoking compartments, and finally are drawn off through the draft flue in the most advanced compartment. Back draft is prevented by pasting paper to form a cover on top of the bags in the compartment ahead of the draft connection, and this method of dampering continuous kilns is common practice in both the compartment and the tunnel type of kilns. The main gas flue is outside the kiln wall and a series of small flues, underground, on each side of each compartment, parallel with and adjacent to the division walls, deliver the gas into the bags under the air. Each small gas flue serves two bags (except the flues supplying the end bags) and each air collecting flue likewise serves two bags. It has long been held that we cannot get reducing conditions in an economizer kiln. In such kilns we get complete combus- tion and use a large excess of air. This excess occasions no loss, because it is heated by the cooling ware and the heat is given up in the advanced compartments before the gases are drawn off into the stack. Mr. Richardson has succeeded in flashing face bricks in a very simple manner. The draft in the burning compartment is checked, which cuts down the volume of air entering this compartment. The gas connections into the succeeding heat- ing-up compartment are opened to the outside air. The air thus entering this compartment not only further checks the draft and incoming hot air in the burning compartment, but completes the combustion of the unburned gases from the compartment under reducing conditions. The gas comes in under its own pressure and with the air supply cut off or greatly reduced, the burning compartment is filled with un- burned gas, thus getting the reducing conditions necessary for flashing. 290 BURNING CLAY WARES. The unhurried gas passes into the next compartment in advance, where it comes in touch with the air entering through the producer gas ports, its combustion completed, and the tem- perature in the heating up compartment is correspondingly increased. The gas connection, admission and distribution is shown in Fig. 172. BURNING CLAY WARES. 291 The Legg design is illustrated in Fig. 174, plan, and Fig. 175, sections. In several of the compartment kilns one can readily trace a development from an earlier kiln, and there is a marked similarity in the general arrangement of the several flues which constitute the kiln, and in the operation of the kiln. The Legg kiln is a distinct departure from this general plan. It is double fired, as will be seen from the plan, and the gas is introduced from the outside through the kiln wall into the bag space on each side of each compartment. In each com- partment there is a longitudinal draft flue as usual. The hot air for combustion is drawn down through a perforated floor into the draft flue and from this is carried forward througn a flue at each end and delivered from below into the bag space. The producer gas enters the bag spaces horizontally and the air rises vertically into the gas and both are carried forward Fig. 175. into the bag space of the next compartment. The top of the bag space is crowned, forming an enclosed combustion space, and in this crown are a series of ports for the escape of the burned gases into the compartments, carrying with them the heat developed by the combustion. The division walls are solid above the floor level, and also below the floor level except the two hot air flues above men- tioned. The plan illustrated is semi-continuous and the return hot air flue is centered under the compartments, returning the hot air from the last compartment (on the right) to the first com- partment (on the left), and thence it is distributed through special flues to the four comers of the first compartment. The sketches show the several main flues underground out- side the kiln wall — draft (“D”), gas (“G’”), and advanced 292 BURNING CLAY WARES. heating (“AH”) — but the actual location may be determined by circumstances and be varied in consequence. Seemingly it would be better to place the advanced heating flue on top of the kiln or in the upper part of the kiln wall, collecting and delivering the hot air for drying through hoods from man- holes in the kiln crown, in the usual manner — at least this method would offer less resistance. The plan shown is for producer gas, natural gas, or oil, but it seems to the author that gas producer furnaces could be attached to the gas openings on each side of the kiln and Fig. 176. thus get direct coal firing without any producer loss or any requirement in steam to operate the producer. Such a plan was worked out a number of years ago, and theoretically, so far as the plan was developed, no difficulty was encountered. A section through one of the division walls of the Goldner kiln is shown in Fig. 176. The bag wall is arched over as in the Legg design, with vents in the top corresponding to the feed holes in the kiln crown. The hot air from one compart- ment is collected and carried forward into the bag wall space BURNING CLAY WARES. 203 of the following compartment by means of perforated floor and under floor flues. For natural gas or oil firing and one might add, powdered coal, the burner is inserted in the feed hole and the downward current of gas comes in contact with the upward current of air and the final combustion takes place under the kiln crown. The spandrell flue — one for each compartment — primarily is for producer gas supplied from a main in or on the kiln wall as in the Youngren kiln. Damper controlled ports in the base of the gas flue lead the gas into the low r er part of the feed hole, thence into the compartment and into contact with the secondary air. This spandrell flue also serves as a col- lecting flue for advanced heating. The connections to the gas main and the main advanced heating flues are by hoods or goose-necks. After a compartment is burned the spandrell flue is cut off from the gas flue, and after the cooling has progressed to the point where heat may safely be taken for water-smok- ing, connection is made to the advanced heating flue and hot air is drawm from the compartment through the feed hole and gas port into the spandrell flue, thence into the advanced heat- ing flue and by-passed to the compartments ahead of the burn- ing compartments entering the latter in the same manner as the gas is introduced. The sketch shows the air port in the top of the bag as being directly under the feed hole, but it should be between the feed holes and alternate throughout the length of the compartment. Thus a table is formed under the feed holes for coal firing, or for salt in salt glazing. 294 BURNING CLAY WARES. CHAPTER XI. CAR TUNNEL KILN. HE CAR TUNNEL KILN idea is more than 150 year 3 old, but the impetus to the development of the kiln came from the Bock kiln, presented to foreign clayworkers about forty-five years ago. There are several advantages in a car tunnel kiln which have kept the idea alive in spite of the numerous early failures. These may be enumerated as follows : (1) The economizer principle, with its marked fuel econ- omy. (2) Centralization of the fuel. (3) A fixed hot zone reducing the kiln absorption loss. (4) Very narrow tunnels which are easily maintained, and in consequence the kiln upkeep is relatively slight. (5) Only the limited combustion zone requires high refrac- tory construction, thus lessening the initial cost. (6) Eliminating two handlings of the ware; namely, the setting and the drawing. There are some offsets to these advantages : (1) The fuel economy is less than in a large tunnel or compartment economizer kiln. It is a question of the relation of the mass of ware to the kiln mass, in which the large kilns have the advantage. (2) The initial cost of the car equipment will more or less even up the difference in kiln cost, and its upkeep will in some measure counter-balance the greater maintenance of the other types of kilns. (3) The car tunnel kiln has not yet been adapted to a wide application. The failure of the Bock kiln, and two or more early attempts in this country, was due to the water- smoking and oxidation, and it is not yet proven that this diffi- culty is sufficiently overcome to cover a wide range of clay materials. One kiln has introduced the possibilities of an advanced BURNING CLAY WARES. 295 heating flue, but it has not yet been tried out. We are of the opinion that this feature of the kiln must come if we are to adapt the kilns to a number of common wares. One can readily see that if a tender clay requires one to several days to water- smoke and only a few hours to burn, or relatively long periods for oxidation and correspondingly short periods for burning we must either hold the fire on the burned ware an unduly period or build the kiln a prohibitive length. Dressier uses two tunnels — one the kiln and one for water- smoking and heating up — in fact, the complete plan has in each unit a third tunnel for drying, and the earlier Bock kiln also had two tunnels. American practice will demand that the several operations, perhaps exclusive of drying, be accom- plished in a single tunnel. There are upwards of fifty car tunnel kilns now in opera- tion in this country, and more projected, and the majority of them are burning pottery wares. In several instances biscuit ware is burned in the top saggers and glost ware in the bottom saggers in order to adapt the ware to the differences in kiln temperature. Here is a point that becomes very important in the manufacture of certain common wares, such as paving blocks, face bricks, where color is important, and products from clays having very short burning ranges. Products Successfully Burned. There are several car tunnel kilns operating successfully on fire brick products, but this product does not involve the difficulties in drying and oxidizing that we find in common clays, nor does it require the uniformity of temperature that other products require. There are two kilns, at this writing, operating on bricks made from shale which may or may not be difficult material to water-smoke and oxidize, but in this product we have the common clay problems and the success of the operation will be a matter of deep interest to clay- workers. When the car tunnel kiln is adapted and its operation ad- justed to a ware, there is undoubtedly a balance in its favor over the other types of kilns. Early Car Tunnel Kilns. The Bock car tunnel kiln is shown in Fig. 177, longitudinal section, Fig. 178, plan, and Fig. 179, cross section. (Note: These drawings and all the car tunnel kiln draw- ings are not to scale, being shortened without proportionately 296 BURNING CLAY WARES. Figure 178 BURNING CLAY WARES. 297 reducing the width and height in order to fully illustrate the principles. Car tunnel kilns vary in length from 200 feet to 350 feet. The widths, exclusive of any side wall flues or spaces, range from four feet to eight feet and twelve feet is claimed to be practical. The heights above the car decks to the under side of the crowns are around five feet. It may also be noted that in the sketches no attempt is made to show the details of construction, since they are merely intended to show the principles of the several kilns.) A detailed description of the Bock kiln is hardly necessary. Briefly, the cars are covered with a refractory floor upon which the ware is placed. An apron attached to the car floor or to the top of the car frame and a sand-filled trough in the kiln wall cut off the metal car and the under tunnel from the firing tunnel. This is the ordinary sand seal found in several kilns. The air supply enters under the cars at the stack end, flows 298 BURNING CLAY WARES. the length of the kiln, thus keeping the cars cool, and enters the firing tunnel at the opposite end, thence traveling succes- sively through the cooling, combustion, and heating-up zones. A feature of the kiln which has been applied to dryers and in modified form to other kilns, is the metal diaphragm, forming the inner side walls at the receiving (stack) end. The purpose of this is to prevent condensation on the cold wet ware, and at the same time recover the heat from the waste gases, as well as the latent heat from the condensing water vapor. The diaphragm flues open into the firing tunnel at the point where the waste gases are assumed to become fully saturated, and here the gases are removed from contact with the ware, but the heat from them passes through the metal plate to the ware by conduction and radiation. The kiln was direct coal fired through the feed holes in the crown of the firing zone, similar to a top-coal fired tunnel (ring) kiln. The Siemens-Hesse car tunnel kiln, which ap- peared three or four years after the Bock kiln, was virtually a Bock kiln adapted to producer gas. The gas was introduced into the combustion zone through the inner side walls virtually in the same manner as in the present day kilns. It is interesting to trace the development of the car tunnel kiln. The Pechine kiln had the fire at one end and one can see in it an effort to develop a tunnel kiln from an end fired periodic kiln such as were used in the early days. Borrie moved the fires to the center of the tunnel. Mean- while, or later, the Hoffman ring kiln is developed, and Bock merely converts it into a car tunnel kiln, in a measure com- bining the Borrie and Hoffman kilns. An outline of the Drayton kiln is shown in Fig. 180, longi- tudinal section, Fig. 181, plan and Fig. 182, cross section through the furnace. The kiln is in reality a compartment car tunnel kiln with advanced heating flue. (Note: The illustrations do not show the full number of compartments. A normal kiln will have four compartments cooling, one burning, three heating up, and two water-smoking, corresponding to a ten-section compartment economizer kiln.) The kiln is divided into sections, holding three cars each, more or less as may be decided in designing the kiln. Three cars are put into the kiln at one time together with a short car carrying a division wall which in effect is equivalent to (jr tar.cpmtiiiition BURNING CLAY WARES. 299 300 BURNING CLAY WARES. a division wall with its accompanying bag or flash wall in a compartment economizer kiln, as shown in Fig. 180. The car furniture is constructed to form under floor flues which connect the compartments through vertical flues in the division walls. In the main walls on each side of the division wall cars throughout the burning and heating-up zones are built up sliding vertical dampers, and through a slot in the crown of each division wall car is a horizontal fire clay damper. When three cars of ware, more or less as the plan may be, followed by a division wall car, are put into the kiln, the division wall cars come opposite the damper slots. Exact placing is necessary only to the extent that the damper slots Figure 182. shall come within the limits of the division walls, and since the division walls may be made any thickness, ample leeway may be provided for the variation in the position of the cars. Following each lot of cars, the side dampers are shoved in and the top dampers lowered, to contact with the division walls. In the cooling zone up to the combustion zone there are no dampers, but beyond, each zone is fully dampered by vertical swinging metal dampers in the side walls where the heat does not require fire clay construction. The operation is as follows : A fan forces air into the cooling zone, through the ware, through the car floor flues, and under the cars, up to the com- bustion zone division wall. The air under the cars passes for- ward and is delivered under the furnace grates. The com- bustion gases pass down through the ware, up through the BURNING CLAY WARES 301 302 BURNING CLAY WARES. division wall flues, into the next compartment, and similarly through several compartments, and finally are drawn off by the draft fan. Hot air is diverted from the cooling zone through a by- pass flue to the water-smoking compartments. The advantages which the kiln offer are : Independent com- partment control, down-draft operation and separate water- smoking. The Faugeron kiln, illustrated in Fig. 183, longitudinal section, Fig. 184, plan, and Fig. 185, cross section through the combustion zone, showing air inlets, is being introduced in this country in the manufacture of fire bricks and pottery. The kiln is divided into sections by drop arches, or more Fig. 185. correctly by step downs in the tunnel crown. The walls are hollow on each side from the center of drop arches to points midway between them, and the inner walls are perforated at the level of the top of the car floor. The ware on the car is set to fit the tunnel under the drop arches closely so that in passing these points the free area is so restricted that the air or gases will be at least in a measure forced out into the hol- low space in the walls to return to the tunnel after by-passing the restricted space. One will readily understand the theo- retical movement of the air and gases. The tendency always is to rise into the crown space over the cars and to move forward under the influence of the draft. When the air reaches a drop arch it is forced downward, and since it cannot readily BURNING CLAY WARES. 303 pass forward under the drop arch, nor alongside the cars, it must pass down among the ware out through the wall perfora- tions into the hollow space, and then after passing the drop arch section, it is forced to re-enter the tunnel at the car floor level and naturally rises into the crown space. This sinuous course, even though the operation is imperfect, suffices to bring the air, or gases, fully in contact with the ware. In the combustion zone, direct coal fired, though not necessarily so, the hot air is distributed under the grate bars for primary combustion and through the bridge and diaphragm walls for secondary combustion. The combustion gases enter the tunnel through the diaphragm wall at the car floor level and thence take a sinuous course to the draft outlet, thus giving up to the green ware the waste heat just as the entering air collects the heat from the cooling ware. It may be noted that the ware must be of such a character and so set as to offer greater resistance to the passage of the air and gas under the drop arches than through the wall slots and by-pass flues. A heat balance of such a kiln by Prof. C. B. Harrop in Vol. XIX, Trans. American Ceramic Society, shows the fol- lowing results : 3.27% Unconsumed combustible matter in the ash pits. 35.83% stack losses, including dehydration. 17.27% in cars and ware leaving the kiln. 45.00% radiation loss. 101.37% total. The results show greater consumption than fuel supply, but considering that all the determination, including radiation, were direct, the results check up very closely and are more satisfactory than data in which some important item is deter- mined by difference. The usual method is to determine radia- tion and kiln losses by difference and thus hide the errors which may have been made in the direct determination of the other factors. The heat balance of a producer gas fired economizer kiln by Prof. R. K. Hursh, University of Illinois, in Yol. I, No. S, Journal American Ceramic Society, is as follows: .6% unconsumed combustible matter in ash. 14.7% producer loss. 32.4% stack loss. 43.6% consumed by the ware. 8.7% radiation and kiln loss. 304 BURNING CLAY WARES. It is to be regretted that scientists have not adopted some definite interpretations of heat balances, and it is also to be regretted that the results are not given in pounds of fuel. A. V. Bleininger, in Heat Balances of Industrial Kilns, gave the percentage fuel consumptions required for the ware as follows : Sewer pipe, 5.7 per cent. ; paving bricks, 11.3 per cent. ; terra cotta, 12.6 per cent, and 8 per cent.; common bricks, 19.6 per cent. Prof. Harrop, it will be noted, does not give any heat re- quired by the ware except that removed from the kiln in the cooling bricks and the dehydration losses. It is beyond our present knowledge to estimate the heat actually required to burn clay wares. We must heat the ware to a certain temperature, and during this heating certain pyro- chemical changes take place which may or may not require heat, but probably very little heat is required to effect these changes or given up by them. The estimated heat required then is based on the temperature and specific heat of the clay and ware, taking into consideration, of course, the heat re- quired for dehydration. The heat required for any given ware will be the same in any type of kiln and the economy of the kiln will depend upon the other factors, particularly the possibility of recovering the heat in the ware. In Bleininger’s kilns there was no recovery of the heat from the ware, at least not for use in the kilns, and when the burns were finished the ware held the heat value given in the percentages. The percentages are low simply because the total fuel re- quired for the operations was high. In two of the five periodic kilns the radiation loss in per cent, was less than Harrop found for the car tunnel kiln, but in pounds of fuel the periodic kilns will show a higher loss than the car tunnel kiln. Prof. Hursh, in the compartment economizer kiln test, gives the ware requirement at 43.6 per cent., which considering the total amount of fuel used and the kiln temperatures attained, will compare approximately with Bleininger’s results, but all of this heat is given back to the kiln operation and the heat actually retained by the ware is practically none. In conse- quence of setting aside 43.6 per cent, of the total fuel for the ware, Prof. Hursh gets by difference the remarkably low radia- tion loss of 8.7 per cent. He has not taken into account the BURNING CLAY WARES. 305 heat from the cooling ware which is carried forward into the burning compartments and the actual heat lost by radiation during the period of the test will be that given, plus the heat derived from the cooling ware. Clayworkers are deeply interested in the study of kilns, and it is unfortunate that any technical data should be mislead- ing. On the face of the returns from the several tests the clay worker would naturally give the preference to an economizer kiln in which the radiation loss is only 8.7 per cent, of the remarkably low total fuel consumption. In pounds of coal this loss would be very small. Compared with this, the 45 per cent, radiation loss from Prof. Harrop’s car tunnel kiln with its higher total fuel consumption, makes a poor showing. The car tunnel kiln may be less economical in fuel than the compartment or other type of economizer kiln, but there is no such itemized difference as the two heat balances show, and there would.be still less difference were the same ware being burned. Both kilns can reduce the temperature of the gases to a minimum before they are drawn off into the stack and in stack losses on the same ware there should be little difference. Both types can be thoroughly insulated and the radiation loss should not widely vary, while in the ground and kiln absorption loss the car tunnel has the preference because the hot zone is fixed and there is no cold wall mass to be heated up as in the com- partment kiln. The ware from each, if the kilns are properly designed, will approximate atmospheric temperature before it leaves the kiln, and the heat actually consumed by the ware will be the same in either type of kiln. The car tunnel kiln tested by Prof. Harrop is 197 feet long, 4 feet 4 inches wide inside the tunnel, and 8 feet high from the rail to the under side of the crown in the center. The output in fire bricks is about 30 tons per day burned to cone 8. A subsequent report by another authority shows temperatures of cone 13 to cone 14, which are the tempera- tures claimed for the kiln in question. The coal consumption during the test by Prof. Harrop was 226 pounds of coal per ton of ware, the heat value of the coal being 13815 B.t.u. The other authority mentioned reported 250 pounds of coal per ton of ware burned to cone 14. These results are not comparable with the results given for the compartment kiln by Prof. Hursh, nor would they be if 306 BURNING CLAY WARES, BURNING CLAY WARES. 307 the latter had distributed his results as did Prof. Harrop and given them in pounds of coal, simply because the wares are different and the temperatures attained are widely apart. It is really immaterial what method of distributing the data is adopted, but there should be a standard method. We would prefer to have the results in pounds of fuel and placed where they belong, but since the fuel required to heat the ware is approximately the same for each cone temperature the per- centage attributed to it becomes a comparative measure of the efficiency of the kiln. For instance, in Prof. Hursh’s heat balance the kiln efficiency ranks high because 43.6 per cent, of the total fuel is attributed to the ware, while the kilns tested by Mr. Bleininger would rank low since only from 5 per cent, to 20 per cent, of the total fuel is attributed to the ware. In other words, the latter kilns, if the ware had been the same, would require from two to nine times as much fuel per ton of ware as the former. The Dressier kiln is illustrated in Fig. 186, a diagrammatic plan, Fig. 187, a cross section through the combustion zone, and Fig. 188, cross sections through the heating up and cool- ing zones. The heating up zones varies in length from 50 feet to 120 feet and may be longer if desired. It is in this zone that dehy- dration and oxidation take place, and the successful accom- plishment of these operations is an important factor in the wide application of the kiln. Pottery and sanitary wares are being burned successfully and economically in this kiln, and also fire clay products, but in these wares the oxidation diffi- culties are a minimum. Some of the common clay wares are exceedingly difficult to oxidize. The heat in the gases from the combustion zone is recovered by conduction, radiation, and convection in the heating up zone. These gases do not come in contact with the ware, but instead are carried through the piping shown in the cross section of the heating up zone. The volume of air in the heating up end of the tunnel is very small, being that introduced by leakage, but any volume can readily be introduced and this, it is said, may be carried forward to any desired point and there drawn off and by- passed to the furnace bench. This would provide the air essential to thorough oxidation. The rate of oxidation is determined by the drop in tem- perature of the combustion gases. If the oxidation is too rapid, we might move the cars forward at a slower rate, but 308 BURNING CLAY WARES. this would involve holding the ware longer in the combustion zone, which would not be desirable if for no other reason than the reduction in capacity without corresponding reduction in fuel consumption. The volume of gases and consequent heat supply cannot be changed because this depends upon the com- bustion requirement, but the drop in temperature requisite for any oxidation can be adjusted (in a longer tunnel) by the character, number, size, and shape of the waste gas flues and piping. Longer stretches of fire clay ducts with thicker walls with a single large iron pipe extension to the draft outlet in- stead of the several pipes shown, will lengthen the heating up zone and result in a more gradual drop in temperature as may be required. In the ordinary economizer kiln the combustion gases pass forward into the compartments of green ware, or at least into the compartments in which the ware is being heated up and oxidized, and the ware is subjected to the ill effects of the combustion gases, particularly sulphur, including the sulphur from the ware which often exceeds that from the fuel. The bad effects of sulphur are especially severe where the kiln is not equipped with an advanced heating flue. The operation of the Dressier car tunnel kiln differs ma- terially. The air for combustion enters at the delivery end and passes through the tunnel in contact with the cooling ware, up to the combustion zone. In this zone, on each side, is a combustion duct (furnace bench) built of fire bricks and lined with carborundum. The gas, if from a producer, is delivered to the kiln through a cross duct under the kiln base, and rises through vertical ports into the combustion ducts. The air is drawn into the combustion ducts through ports in the furnace bench and enters the ducts behind the gas, although there is, or may be, air admission for secondary combustion in front of the gas port, or for a second gas port. In fact, the relative positions of the air and gas entries may be anything the de- signer desires. The combustion ducts connect directly with the carborundum double walled combustion chambers shown in Fig. 187. The central ducts in these chambers carry the combustion gases, while the surrounding transverse flues are for air circulation and heat convection in the heating up zone. Iron pipes extend the combustion ducts to the draft outlet. It is seen that the air enters in the cooling end, is heated by the cooling ware, enters the combustion ducts near the BURNING CLAY WARES. 309 Fig. 187. Fig. 188. 310 BURNING CLAY WARES. longitudinal center of the kiln, and the combustion gases are drawn off through the heating up end by means of ducts and pipes. It is evident that no air except that from leakage enters the heating up end of the tunnel, although air in greater quan- tity, if desired for oxidation, may be introduced through this end. Any gas or vapor from the ware in the heating up zone cannot pass back over the cold, possibly wet, entering ware, but instead must move forward to the combustion benches, where it commingles with the hot air from the cooling end of the tunnel unless it be drawn off at some intermediate point. In the heating up zone are the hot waste gas ducts and pipes, and any air in the tunnel will tend to rise through the circulating channels and among the pipes into the tunnel crown space and then drop through the cooler ware. Since the air is replaced by leakage to some extent and in larger volume intermittently when the end doors are opened to introduce cars of green ware, there will be in conjunction with the transverse circulation, a forward movement, or in other words, theoreti- cally, there will be a slow spiral forward movement of the air through the heating up tunnel, which may be materially in- creased if desired. The cooling zone has iron pipes on each side, open to the outside in the end wall, and connected with a fan at the end near the combustion zone. Through these, air is drawn and heated for use in independent dryers or for other purposes in the factory. A baffle wall is placed between the air pipes and the hot ware to develop the same circulating conditions as in the other end of the tunnel ; namely, a transverse circulation of the air in conjunction with its forward movement in the tunnel, and the volume of air is that required for combustion, together with more or less excess. It is of first importance, of course, to conserve the heat in the cooling ware for the kiln operation, but any excess can be utilized for other work with the added advantage of getting the ware cooled, which is not always the case where the cool- ing end of the tunnel is limited in length. When the kiln temperature to be attained is not excessive a larger volume of heat may be drawn off for other work, even to the extent, if desired, that the heat so removed must be replaced in the kiln by the consumption of increased quan- tity of gas. It is simply a question of whether it is more economical to generate heat for outside purposes in the kiln BURNING CLAY WARES. 311 or independently. In this instance the kiln has the prefer- ence most decidedly. Where high temperatures are required in the kiln the first consideration must be conservation of heat from the cooling ware to augment the heat from combustion, and heat for out- side work becomes secondary. Dressler’s idea in its completeness has three tunnels. One is the kiln proper. The second is a pre-heater and the heat is supplied by the hot air and by the products of combustion from the firing tunnel, the former introduced into the second tunnel direct and the latter drawn through wall pipes. The temperature in the pre-heater is said to be 100 deg. C. to 250 deg. C. The air from the pre-heating tunnel is drawn through side pipes in a third, or drying tunnel, and not only is the available sensible heat recovered, but the vapor from the ware in the pre-heater, which represents a large part of the heat, is condensed in the dryer tunnel pipes and its latent heat given up. The data relative to the fuel consumption is varied, but does not give us a fair comparison with that of other types of kilns. We get the following data from a paper by A. Bigot in Vol. XV, Transactions English Ceramic Society. Percentage of fuel value attributed to the ware in three types of kilns: The Hoffman kiln, it will be understood, is the ordinary tunnel, or ring kiln, and the periodic is presumably a down- draft. If the above data is correct, then a unit of fuel will burn one ton of ware in the periodic kiln, a ton and one-half in the Hoffman, and more than three tons in the Dressier. The data is interesting, but not acceptable. We should know whether the ware was the same, and also we should know the size and character of the periodic kiln. On the same yard we have found down-draft kilns using two and more times as much fuel per ton of ware than was being used in other kilns, simply due to the difference in size of the kiln. Mr. C. J. Kirk, in Yol. XVIII, Transactions American Cera- mic Society, states that 15,000 pounds of sanitary ware are being burned daily with 25,000 cu. ft. of natural gas, whereas Dressier Hoffman Periodic 49.6 per cent. 22.9 per cent. 15.0 per cent. 312 BURNING CLAY WARES. BURNING CLAY WARES. 313 the same weight of ware formerly burned in periodic kilns required 250,000 to 300,000 cu. ft. of natural gas. These figures are startling and cannot be accepted without some explanation of the gas firing in the periodic kilns. Nat- tural gas is a very difficult fuel, economy considered, in a periodic kiln. The flame is short and very intense. It is neces- sary to protect the furnace brick work and to get the heat over into the kiln, to introduce an excessive quantity of air through the furnaces to sweep the heat out of the furnace. The stack losses under such circumstances are excessive. It is possible to fire a kiln indefintely without getting sufficient heat from the furnaces into the kiln through lack of excess that the gas combustion temperature is never high enough to burn the ware. A pottery manufacturer puts the difference in gas consump- tion in the two types of kilns in a single sentence ; namely, “Gas bills for periodic kiln operation were $3,000.00 per month, and now for Dressier kiln operation they are $1,000.00 per month.” * This is a saving of 66 per cent., and it may be accepted as practical. Ordinary tunnel and compartment kilns show a saving of 50 per cent, to 75 per cent, over the periodic kiln, and the car tunnel kilns have the same regenerative principles as the ordinary continuous kilns and should show similar sav- ing over the periodic kiln. A Dressier kiln is to be installed in an Ohio factory to burn common bricks, hollow bricks and drain tile. The kiln will be about 300 feet long, and it is expected to turn out approxi- mately 100 tons of hollow ware per day. The cars are to be 6 feet by 8 feet, and the setting will be 4 feet high, perhaps 5 feet. It is expected to pull the cars at the rate of one every forty-five minutes. The operation will be continuous and extra cars will be provided to maintain the operation of the kiln throughout the night and over Sunday. The manufacturers of common clay wares will watch the result of this operation with a good deal of interest. Up to date there have been several attempts to burn such common wares in car tunnel kilns, but without satisfactory results. The Zwermann single tunnel kiln is illustrated in Fig. 189, longitudinal section, Fig. 190, a diagrammatic plan, and Fig. 191, a cross section through the burning zone. The sketches are broken to reduce them to a suitable length for illustration, but the furnace section is shown full length. 314 BURNING CLAY WARES. In the cooling end are shown flues in the side wall, or rather forming a part of the side wall, and also it will be noted that the tunnel crown is double, thus forming a crown flue. The lower flues in each side and the crown flue have no part in the kiln operation except cooling. These flues, which are one hundred feet long, more or less, are open to the outside air at the delivery end of the kiln, and near the furnace section they terminate in a fan which draws the air through these flues and delivers it into a dryer or to other purposes in the factory independent of the kiln. The other flues in the side walls, one on each side of the tunnel, have a pressure fan connection at the delivery end of Fig. 191. the kiln, which forces air through the flues and delivers it to the furnace for combustion. The combustion gases are drawn off through a fan near the receiving end of the kiln. The dimensions of the kiln are, pre-heating zone approxi- mately 150 feet long, furnace zone nearly 60 feet and cooling zone about 140 feet. The tunnel is 6 feet 4 inches wide and 6 feet 6 inches high above the trucks. The lengths of the sev- eral zones and their total length will vary in different kilns, depending upon the character of the work to be done. The furnaces are staggered except the final two burners, and it would seem that the purpose of this staggered arrange^ ment is to get the equivalent of cross firing in the arches of up-draft kilns. The furnace box or arch is built on the truck BURNING CLAY WARES. 315 transversely — one centered on each truck — and the ware is piled on each side and over the arch. In this we plainly have the idea of the arches in an up-draft field kiln. If the ware being burned in the tunnel kiln were bricks the setting would be such as to form the combustion arches from the bricks as in a scove kiln. As the cars advance the arches come opposite, first one furnace on one side, then the next furnace on the opposite side, repeating to the end of the furnace section where the finishing touches to the ware are given by a pair of oppo- site furnaces. A later design by Zwermann, shown in Fig. 192, cross sec- tions, and Fig. 193, plan, has a double tunnel, operating in opposite directions. At either end there will be green ware going into one tunnel and burned ware coming out of the other tunnel. Near the end where water-smoking takes place in one tunnel and cooling in the other the division wall be- tween the two tunnels is thin, in fact, toward the end where the temperature of the cooling ware is a low red heat or less, the division wall is steel plate stiffened by pilaster walls, which also support the crowns. On one side of this wall we have cooling ware coming out and on the other side green ware going in, with the reverse at the opposite end of the kiln. By convection the heat from the cooling ware is brought into contact with the thin division wall and by conduction it is carried through this wall into the adjacent tunnel where 316 BURNING CLAY WARES. baffles direct the currents of air and take the heat from the wall to the green ware. The advancing green ware next passes into a muffled sec- tion of the tunnel. The combustion gases from the furnace zone are drawn back through tubes and the forward moving ware is heated up by the air in the tunnel circulating among these waste gas ducts and taking the heat therefrom to the ware. This muffled section is the heating-up zone. From the muffled zone the ware passes the furnaces, where it comes in direct contact with the flame from the furnaces or burners and the combustion gases pass down among the ware to the Fig. 194. car floor level, thence back to the draft flue in the side wall and forward into the heating-up ducts above mentioned and out through the stack. The ware goes forward into the cooling section, where the heat is given up to the ware coming in through the other tunnel. The two tunnels are identically alike except the sev- eral zones are reversed. Along the central portion of the kiln the crown is double with a space between, and the air for secondary combustion is drawn through this annular space and becomes heated be- fore bringing it into touch with the gas. Another car tunnel kiln which is finding use in this country is illustrated in Fig. 194, a cross section through the cooling zone looking toward the furnace zone. The kiln is very sim- ple. It has a straight tunnel except an enlarged section for the furnaces. There is no sand seal, but instead the car fur- BURNING CLAY WARES. 317 niture recesses loosely into a groove in the side walls. The furnace section is simply a wider span overlapping the tunnel width, which is boxed in on sides and ends to form a combus- tion box on each side of the kiln. The fires are in the ends of this box and the flame is parallel with the moving cars until deflected by the furnace crown. In several installations there are no flash or perforated walls between the firebox and the tunnel. In fact, it is said that the curved crown of the furnace section reflects the heat to the ware to better purpose than is possible through direction by means of flash walls. Under the hot sections of the kiln and underneath the cars is a longitudinal open duct which is connected with the out- side air by under cross ducts with vertical risers to the ground level outside the kiln wall. The connection between the cross duct and the longitudinal duct is dampered, and the only air normally entering the longitudinal duct is by leakage, but this can be increased to any degree by opening the damper connec- tion. The purpose of these ducts is to supply sufficient air to protect the cars, and this air, since there is no sand seal, rises into the tunnel proper. It is, therefore, desirable that the volume of air entering in this way should be kept to a mini- mum, and this will depend upon the amount required to keep the cars cool. The walls of the cooling end of the kiln are of a hollow construction and the vertical flues formed within the walls by this construction are open to the air on top, and likewise at the bottom through ports in the outside wall. The outside air thus can freely enter these wall flues at the bottom, and naturally rising, it escapes through the top ports. The inwall is made thin and the heat from the cooling ware is conducted through the wall, picked up by the air currents in the wall flues and removed from the kiln. No use is made of this waste heat except as it serves to heat the fac- tory in cold weather, but of course it could be readily collected and applied to any purpose in the factory operation, or be carried forward and used in the furnaces. In one plan pro- vision is made for the use of the air in the combustion by sim- ply collecting it by down-draft into a flue and carrying it for- ward to the furnace. This provision, however, has not been installed in the kiln in question. In the pre-heating end there are several openings at inter- vals into the stack ducts, and by dampering the gases may be drawn off before reaching the end of the kiln, or they may 318 BURNING CLAY WARES. be carried to the end, thus shortening or extending the tem- perature conditions in the pre-heating end of the kiln. Another kiln has combustion chambers or tubes on each side of the ware, in principle equivalent to the Dressier; namely, that the combustion takes place in a tube, and the ware is burned by convection of the heat conducted through the tube walls. Instead of using the double-walled combustion and circulating construction, extending it to the end of tile kiln by a simple duct and tubes, as in the Dressier, the com- bustion gases are taken through economizers on top of the kiln, and the air for combustion is drawn from these econo- mizers and delivered to the burners and into the combustion tubes. Hot air is also drawn from the economizers and deliv- ered into the tunnel under the combustion tubes, circulating around them, then over and down among the ware, and finally it is drawn off through the car floor and out of the kiln, and may be returned to the economizer, thus making a complete circuit. The claim is made that the positive circulation around the combustion tubes and among the ware is essential to uni- form results, but the equipment is complicated, and if results can be obtained with a simpler arrangement, as they seem to be, the advantages will be decidedly in favor of the simple construction. The Harrop kiln illustrated in Fig. 195 (plan), Fig. 196 (cross section), Fig. 197 (side view showing furnaces), and Fig. 198 (temperature curve), has several new features in car tunnel kiln construction and operation. The kiln is a simple direct-fired tunnel type with staggered furnaces near the center and the fuel may be coal, oil or gas. There are no flash walls in front of the furnaces but instead the aim of the construction is to direct the com- bustion gases from the furnaces among the ware by deflect- ing walls which also serve as radiating surfaces. The circulation of the gases among the ware in the heating up zone is simply and effectively attained by alternate off-sets in the side walls, by battering these walls vertically and by drop arches in the kiln crown. Gases will flow into free areas and hot gases rise to the crown space. The alternating free areas in consequence of the recessed side walls cause the gases to follow a sinuous horizontal course in their passage to the draft ports, passing, in part at least, from one enlarged space in the kiln wall across the tunnel to BURNING CLAY WARES 319 -Jo -3afU.V2-MIJ.31 Q X 0 U ui X Li X D H < X Li QL 5 UI H 320 BURNING CLAY WARES. the next following open area in the opposite wall and repeat- ing to the end of the staggered wall construction. As the gases rise into the crown space the battered wall construction increases the resistance and thus retards the upward movement; in other words, the gases are squeezed downward and inward among the ware, and any gases col- lecing in the crown space are deflected downward by the drop arches. The cars are sand-sealed in the usual manner and the air under the cars is introduced in jets by fan pressure. A balanced air pressure is maintained in the tunnel proper and the under car space. This is accomplished by means of dams in the under car space which increase the resistance to the flow of the air and this resistance is measurably equivalent to the resistance of the passage of the air in the upper tunnel and thus a balance is maintained throughout the length of the kiln. Air for cooling the ware and for secondary combustion is introduced at the discharge end of the kiln by the same fan which supplies the under tunnel air. It has been found that the hot air in the cooling end of the tunnel tends to drift back to the discharge end in the upper part of the kiln, while the in-going air is an under current in the lower level of the tunnel. This unsatisfactory air movement is BURNING CLAY WARES. 321 counteracted by introducing the cold air horizontally under the crown by means of a metal inner crown sheet annular with the crown and several inches below it. The air fan has several functions: To introduce air into the tunnel for cooling and secondary combustion; to put the cooling air under the cars; to force air through a duct on top of the kiln for primary combustion use in the furnaces or to be admitted to any part of the kiln for any purpose. The draft fan is at the charging end of the kiln and connects with the kiln through an under cross duct with vertical ducts to horizontal ducts in the side walls at the car deck level. There are a series of ports in these horizontal ducts which enable the operator to remove the gases practically at the discharge end or at any point in the length of the ducts. Figure 197 This is a provision for adjustment in the kiln operation to adapt it to the need of the ware in the early stages of the water-smoking period, and once adjusted no further changes are likely to be required. The draft conditions in the kiln are controlled at the fan. The heat curve, Fig. 198, shows practically a straight line increase in temperature through the heating up, or oxidation zone to the furnace zone, then it is held with slight variation through the furnace zone, dropping rapidly through the higher temperature of the cooling, and then slowly through the annealing section. This curve is taken from a kiln burning porcelain in which the problems of water- smoking and oxidation are far different from those in many common wares. Kilns to burn such wares must make better provision for oxidation in order that this part of the burning may keep pace with the final burning. 322 BURNING CLAY WARES. Professor Harrop meets this requirement by introducing a preliminary set of furnaces near the point where oxidation begins, and provides for the introduction of pre-heated air between the main furnace zone and the preliminary furnaces. The length of the kiln is increased to the extent of this oxidation section. The heat curve will have the gradual rise through the water-smoking zone as shown in Fig. 198, up to the oxidation zone, then a slow rise through the oxida- tion zone into the main furnace zone, followed by the quick cooling to the annealing temperature, then slow cooling to the end. Clayworkers recognize the fact that in the majority of wares, the cooling from high temperatures to a low red heat may be done rapidly without damage to the ware, and this rapid cooling beyond the furnace zone as shown in the heat curve is accomplished by the introduction of air in this section of the kiln through flues provided for this purpose, but if the ware will not stand such treatment the air may be cut out. There is an interesting car tunnel kiln in successful opera- tion on small pieces of electrical porcelain. The ware is placed on fire clay trays on the cars. The kiln is an open fire type, but instead of a direct flame contact the heat is deflected to the ware, which is exposed to the combustion gases. It may be noted that two of the kilns previously mentioned make use of deflected heat. The car tunnel kiln is developing along three lines : (1) The open fire kiln, with the combustion gases in direct contact with the ware, and the latter may be in open setting or enclosed in saggers. (2) The open fire kiln in which deflecting walls are used to direct the combustion gases to the ware set either exposed or in saggers. (3) The muffled kiln, in which the ware may be set in sag- gers, but only as a matter of convenience in handling it. The first two types are very closely akin. The same kiln may have a perforated flash wall between the furnace and the ware and thus operate on the principle of the first type, or the perforated wall may be removed and deflecting walls built in such a way as to deflect the gases from their natural course down on or among the ware. BURNING CLAY WARES. 323 CHAPTER XII. BURNING A DOWN DRAFT KILN. DOWN draft kiln operation has four distinct stages, — water-smoking, oxidation, heating up, and driving the heat to the bottom, or heat soaking. Fire flashing is a fifth stage in the burning. The ware from the dryer is never fully dry. There remains a fraction of a per cent, up to three or more per cent, of moisture and two to three per cent, of free water in such form that it cannot be removed by dryer tempera- tures. Wares are often set before they are apparently dry, but even though they are seemingly dry, they contain water ap- proximating five per cent, which must be removed in the kiln operation. The removal of this water content is the water-smoking stage in the burning, and the rate of its re- moval depends upon the behavior of the ware under such drying conditions and upon the efficiency of the kiln con- struction. Slabbing of sewer pipe and large drain tile starts in the water-smoking; steaming with its resultant more or less rotten ware and subsequent discoloration, is a water-smoking trouble. With some wares the fires may be advanced as rapidly as it is possible to build them up, and in other wares, espe- cially dry pressed products, the kiln temperatures must be held very low for a period of five or more days, and in some instances up to thirty days, to safely remove the moisture. The rate of water-smoking must be determined for each ware, and the fires manipulated to attain the desired results. The usual custom is to build a small fire in the ash pit of each furnace, preferably of wood, coke, smokeless coal, or gas, but soft coals are often used. The objection to soft coal is the large volume of smoke which soots the damp ware and often the draft throughout the burn is choked by the sooted draft spaces. The furnace doors or mouths should be left open and as much draft maintained as possible. Un- fortunately where the draft is by a stack we have very weak draft during the water-smoking stage in consequence of the stack being cold. 324 BURNING CLAY WARES. A kiln with a single stack is objectional, and in building a battery of kilns it is better to have one stack for two, three, or four kilns, but with an independent flue for each kiln. In this way we get the benefit of conducted heat from the combustion gases of the burning kilns, but in any event such a stack does not cool to the extent of a single stack, and we get the advantage of this heat in starting the water- smoking. If the situation is such that single stacks are unavoidable, they should be built with small furnaces at the base to heat up the stack and thus give us a good draft through the kiln in the early stages of the water-smoking. A better plan would be to have a fan connection for the start of the kiln burning. Fig. 199 shows the heat curves for two kilns, both burn- ing bricks, and as it happens both clays stand a lot of abuse in water-smoking and oxidation. These curves are shown in the solid line and the dotted line. Practically complete removal of the moisture is attained at a temperature around 400 deg. Fah. to 450 deg. In the kiln shown by the solid line the water-smoking was finished in 16 hours and the dotted line kiln was ready for oxidation in 24 hours. If one has a pyrometer — and every kiln should be equipped with such an instrument — and the burner has determined the safe rate of advance of the heat in the top of the kiln, he can safely control the fires to get satisfactory results, but if he has no such guide, he must feel his way by observation of the volume of steam coming from the top of the stack and by means of an iron rod thrust into the kiln at different levels, noting the condensation on the rod. If the moisture driven from the ware is not removed from the kiln, but instead condenses and soaks into the ware in the bottom of the ki n, the result will be a lot of damaged ware in the bottom, and we have seen many burns in which it was evident the water-smoking had been advanced too rapidly in that the kiln floor, kiln flues, and stack, did not amply provide for the removal of the steam. Where such con- ditions exist the water-smoking must be correspondingly slower to prevent steaming the bottom ware. The dash and dot curve in Fig. 199 illustrates slower water-smoking in which we have indicated 68 hours for this stage in- the process and for some wares this would be rapid work. In the other extreme, one operation — solid bricks set high and close — the burning was completed in 72 hours, but few clays will stand such rapid work. The kiln in this instance had the advantage of fan draft in the water-smoking and throughout the burn, and the clay stood a lot of abuse in water-smoking, besides requiring no oxidation, and also it had such a long burning range that it was not necessary to follow the usual procedure in working the heat to the bottom but instead the fires were kept under full head until the bricks in the bottom of the kiln were hard. BURNING CLAY WARES. 325 The usual burning periods are three to five days for hollow ware, five to nine days for stiff mud bricks, and eight to twelve days for dry pressed bricks. In general practice the fires are built up to full fire by the time the water-smoke is off, and with full fire begins the oxidation. The oxidation consists in burning out the carbon in the clay and roasting out the sulphur, otherwise the product will be black cored and in the finishing stages the product will be bloated. The kiln is under full fire and with all the draft it can get, but the fires are kept open to take in as much excess air as possible. If the furnaces are of Figure 199. the grateless type we do not close the firing mouths with coal as in the later stages of the burning, or if the furnaces are provided with doors, the latter are left partly open. The firing should be light and frequent, keeping in mind that we wish to advance the heat slowly and to admit as much excess air as may be taken in. It is the hot air that burns out the carbon and sulphur in the clay. We could carry the heat up to 800 degrees, more or less, and hold it until 326 BURNING CLAY WARES. oxidation was complete, but it is better to keep the heat advanced slowly to the end of the oxidation and then jump as quickly as possible to the final stage of the burning. The “blue smoke” from the stack, following, the water- smoke, is the visual evidence of the oxidation, and the end is determined by trial pieces taken from the kiln and noting whether the black core has fully disappeared. In one opera- tion we noticed that after the black core had disappeared, there remained a saffron colored core which had not pre- viously been considered. Extending the oxidation period 24 hours until the yellow core was dissipated, resulted in an improvement of five or more points in the rattler test in this paving brick product. It is therefore not always safe to assume that oxidation Is complete with the disappearance of the black core. Oxidation takes place rapidly at a low red heat and in general the heat should not be allowed to get much above 1200 degrees until after the black core is fully out. Carbon particularly should be driven out at low temperatures be- cause at higher temperatures there is a greater tendency to cracking of the gases and the deposition of carbon instead of its removal, and at high temperatures regardless of the c uantity of excess air, carbon in the clay cannot be burned out. Sulphur may be removed at higher temperatures but it must be out before vitrification begins. In- one factory using a shale high in carbon and sulphur it was the practice to hold the temperature around 1200 degrees until the carbon was out and with it a lot of the sulphur, and this was in- dicated by the color of the smoke from the stack. With the disappearance of the blue smoke, the tempera- ture was advanced to 1,400 degrees, which resulted in a large volume of blue smoke, and this temperature was held until the smoke was off, when the temperatures were further ad- vanced to 1600 degrees and again held till the blue smoke was off. The combined water in the clay begins to come off rapidly around 800 degrees but it does not require any change in the burning operation. Following the removal of the combined water, the ware begins to shrink, if it has any shrinkage, and the progress of the burn in many factories, especially those not equipped with pyrometers, is determined by the rate of the settle of the mass of ware. The guides in burn- ing, besides the pyrometer, are the water-smoke, the blue smoke, the black and yellow cores, the settle, and trial pieces. When oxidation is complete the fires should be closed at the furnace mouths and the heat may be advanced to the finishing stage as rapidly as possible. This is the final heating up and it is the heavy firing period during the burning. The fires should be kept clean and in the best possible condition to get a maximum value out of the fuel BURNING CLAY WARES. 327 When the finishing temperature is attained, or nearly so, on top we begin the process of working the heat to the bottom. The dampers are lowered to check the draft and thus we increase the pressure in the kiln, or more properly equalize the pressure and under this equal pressure the heat drifts to all parts of the kiln. The heat should be held at the finishing temperature, or practically so, throughout this final period, and this can be done by firing lighter, and oftener, if necessary. The furnace mouths, if the furnaces are the grateless or inclined grate type, should be lightly closed with coal, permitting the bed to burn down quickly to a small opening over the fuel bed. A single shovelful, or two at most, suffices for each fire and it is seldom necessary to lessen the intervals of firing, in fact we may often increase the interval. If it is a new operation one must work slowly through all the stages of the burning. It is better to take ten or more days in the early burns than to discourage the investors in the factory with bad burns. The curve to be followed is along the line of the dash and dot curve shown in Fig. 199, and one may go even slower than this depending upon the exhibits of the burn previously mentioned. If the first burn is successful throughout, one may begin to shorten the periods, one at a time. A day more or less may be clipped from the water-smoking and repeated until the bottom ware shows some effect of steaming. This is beyond the limit and the period must be increased until no steaming is apparent. The next step is to shorten the oxidation period in the same manner until some bloated ware shows that the oxidation has been too fast. The dash and dot curve shows relatively slow oxidation. The heating up following the oxidation is usually simply a matter of getting the most out of the fuel, and the final tempera- ture requires careful handling until satisfactory trials are obtained from the bottom of the kiln. Fire flashing usually starts toward the end of the finish- ing heat and merely consists in working with dampers lowered in a greater degree and keeping the fires fully closed all the time. It is often designated as the “Smoking” stage because during it the volume of smoke is greatly increased, but it is more properly a reduction stage and smoke is not essential but unavoidable. The great danger is that we may advance the temperature beyond that which the ware will safely stand. In consequence of this danger, we must, with many clays, work with an excess reducing condition, first to get the reducing condition which produces the flash, and second to get it in such degree that the flame temperature and thereby the temperature in the kiln is not increased. In the water-smoking and oxidation stages we are work- ing with low temperatures and large air excess; in the heat- ing we endeaver to get as nearly perfect combustion as pos- sible to get a maximum heat value out of the fuel; in the 328 BURNING CLAY WARES. soaking period we work with high temperatures, slow draft, and open fires and the excess air keeps the temperature from running too high; in the flashing we used closed fires, slow draft and insufficient air which prevents excessive temperature. If, in the flashing, we changed the soaking pe- riod firing back to the heating up firing except retaining the slow draft, we would rapidly advance the temperature which few wares would stand. Instead we go beyond the heating up firing and so damper the furnaces with coal that the com- bustion is far below perfect conditions and the flame tem- peratures are lower and thus we can control the kiln tem- peratures, but in doing so we get a lot of smoke we do not need. The flashing period varies from 12 to 24 hours, depending upon the clay and the degree of flash desired. COLORATION, DISCOLORATION, AND OTHER BURNING There are many burning effects, intentional and otherwise, concerning which frequent inquiries are received, and a dis- cussion of a few of the more common ones may be pertinent. Scum. The almost universal bug-bear of the clay-worker is scum. It comes up again and again as a kiln trouble because it be- comes apparent in the kiln, and we have made many in- vestigations of it. It is chiefly a dryer trouble which has been discussed in “Scumming and Efflorescence” by the author and there are several available published articles relative to it. The ordinary scum is a dirty white coating which comes to the surface of the ware in the drying and becomes per- manent in the burning. Where such trouble occurs the first step should be to determine whether it originates in the dryer or in the kiln. If the product is brick, a brief study of the scummed cross bars on the faces and backs of the bricks, corresponding to the setting, will determine the origin of the trouble. If these marks correspond with the setting on the dryer cars and not with the kiln setting, the trouble goes back to the dryer with which we are not concerned in the burning. The examination of the burned product in the kiln fre- quently reveals that the coating occurs in both dryer and kiln, but occasionally it is clearly a kiln trouble. In the majority of instances the scumming minerals are in the clay or shale and the water used in pugging. The common scum is sulphate of lime (gypsum). Opera- tors have often observed glassy plates in the clay or shale which to them “looked like isinglass” but instead these crystal plates are gypsum, and the proof is that they can be scratched by one’s finger nail. Ground waters almost invariably carry some sulphate of lime which produces a tough adhesive scale in steam boilers. BURNING CLAY WARES. 329 If a careful investigation were made it is doubtful if there would be found one shale out of every hundred that is free from gypsum. If the scum minerals are so universal it may be asked why it is that many clays and shales in use do not scum, or at most only occasionally. The explanation is that the clay mass absorbs more or less of the salts of whatever kind and in drying, these absorbed salts do not come to the surface. The surface scum is the excess salts in the clay mass. The elements of scum are therefore usually found in the clay or water, or both. It may be present in neither in sufficient quantity to cause surface scum, but the base element may be in the clay and the acid element in the water, or vice versa, and when the two are brought together an excess of scum mineral may be developed. Scum often develops in the dryer when combustion gases either direct or by leakage, get into the dryer. The shale or clay may contain lime insoluble in water — lime carbon- ate, for instance — and this is converted into sulphate by sulphur gases in a humid atmosphere in contact with the drying ware. If the ware is not scummed in the dryer, but does scum in the kiln, the investigator should determine whether the effect is not due to setting wet ware. If the kiln fuel is coal, we get sulphur gases from it and if the ware is set wet, we get the dryer conditions above mentioned, and the correction is evident. It may be asked how sulphur gas and water vapor, es- sentially a gas, in the kiln among the ware but not within the ware, can penetrate the ware, disintegrate the lime minerals and dissolve the lime and subsequently come to the surface in a liquid form, which it must do to bring the dissolved salts to the surface? Sulphuric acid forms when sulphur dioxide and water vapor are mixed or when the dioxide is brought in contact with a damp surface. It volatilizes at a much higher temperature than water, and although water vapor may be leaving the ware, any sulphuric acid developed is being absorbed by the ware and it with any dissolved salts are brought to the surface later in the water-smoking period. In this connection, but aside from the subject under dis- cussion, we may mention a sulphur gas trouble which was experienced by a number of the early users of continuous kilns, before advanced heating flues were adopted. The ware when burned to vitrification, or approaching vitrification, was swelled, — not bloated as we understand bloating, — yet perfect in shape but a larger size than the same ware burned in periodic kilns. The soft burned ware was normal in size but the hard burned product often was larger than the dry ware although the normal burning shrinkage should have been six or more per cent. In such kilns we had not only the sulphur from the fuel but often a far greater volume of sulphur gas from the shale or clay and these gases were going forward into low tempera- 330 BURNING CLAY WARES. ture water-smoking compartments. The conditions for the development of sulphuric acid were ideal, and the effect on the product was as noted above. No satisfactory explanation of this has ever been made, so far as we know. Undoubtedly mineral sulphates were developed in the ware which had little or no effect upon the soft burned ware, and it likely that at vitrifying tempera- tures, the sulphates were dissociated, and likely we got a result similar to black coring and bloating with which all clayworkers are familiar, yet apparently there was no visible vesicular structure such as we get in bloated ware. In- cidentally it may be mentioned that clays containing lime pebbles are often burned in continuous kilns and the burned product does not “pop,” as it does when burned in periodic kilns. Returning to the subject of scumming, it is evident that the ordinary scum develops in the dryer, and if the ware is properly dried and not scummed, it will not scum in the kiln. Two rare discolorations have come to our attention, — one of them in several operations and the other in a single in- stance. The former is an evanescent white coating which appears on the surface of the bricks, — even vitrified bricks, — after they are burned. When the kiln is first opened and still hot, the ware is clean but as cooling proceeds the white coating appears. In one instance the bricks were clean when taken from the kiln but became heavily coated shortly after, without moisture except such dampness as they might gather from the atmosphere during a summer day. An analysis of this coating showed it to be an alkaline sulphate, — an alum if you please, but it is not fully clear to us how this coating develops. In three instances we found that the trouble appeared when the bricks has been burned with wet dirty coal from an old mine. In two of these , instances we suggested that clean, freshly mined coal be tried and the result was that the trouble was overcome. One operator after getting clean ware from good coal, went back to the cheaper coal which he got from a near by old mine, and the efflorescence reap- peared. In one operation the product was a vitrified flashed brick and it is inconceivable that the efflorescence could have come from within the product, especially in view of the fact that there is no moisture except atmospheric vapor. Evidently the alkaline sulphate is a surface coating and it seems to us that it must come from the coal ash. Its visibility and evanescence are due to the action of am- monia in the atmosphere, the effect of which is to produce a white effloresence on the alkaline sulphate. There likely is present in the surface of the ware an invis- ible crystalline alkaline sulphate which when acted upon by ammonia produces a white flocculent ammonia alum, and this is the white coating which appears under the conditions given BURNING CLAY WARES. 331 above. It is very soluble and quickly disappears in the presence of excess moisture. In this connection we may mention a vitrified flashed brick product which does not effloresce in the wall, but in two jobs effloresced badly, and these two jobs were ice manu- facturing plants. This efflorescence, it seems to us, is analo- gous to the above described phenomenon, because of the ammonia vapor in the ice factories. The white coating which we have described and which is due to a condition developed in the kiln, must not be con fused with the white, yellow and green coatings which appear on wares on the yard and often after the product is laid in the wall. These are quite another story which goes back to the clays when the coatings are developed from the pro- duct and not from extraneous sources, nor are these efflores- cences related to scumming, at least not closely related. The other unusual coating mentioned previously had all the appearance of common ordinary scum. An examination of the bricks in the kiln showed masses of the ware badly scummed and the scum flared out on the sides of the bricks where the gases came up through the checker work and the white cross bars on the backs of the bricks corresponded to the kiln setting. It was clearly a case of kiln scumming. The product from the dryer was clean in spite of the fact that the drying was done with combustion gases from the boilers and gas from the producers from which gas was also drawn to burn the kilns. This was the puzzling problem. If the combustion gases would not scum the product in drying why should they develop scumming in the kiln work? A mineralogical exami- nation by an authority in this work revealed that the coating was almost entirely silica, — just white sand. If this is correct the explanation is simple. The bricks were closely set in an updraft kiln and platted with a tight flat course instead of the usual spaced course covered by a tight course, and the openings were small and far apart instead of the customary opening of each alternate brick in the tight course. The result was that the bricks, especially in the patches where the draft was sluggish, were steamed in conjunction with sulphur gases. Any chemist is familiar with the rational analysis of clays by which practically all the minerals in the clay except free silica are dissolved by hot sulphuric acid. This, it seems to us, is what happened in the above described kiln operation. The surface minerals except sand were dissolved and carried into the bricks and the final evaporation took place below the surface, or at least under the sand, leaving the latter as a dirty white coating. Manufacturers of clay wares are familiar with a number of discolorations in consequence of steaming in a sulphur acid atmosphere. In the manufacture of gray bricks using manganese for the gray color, light edges bordered by a streak darker than 332 BURNING CLAY WARES. the normal color of the brick, were frequently, and on some yards invariably produced, when water-smoking with coal or coke. We have had similar experience with light red bricks especially when burning the product in continuous kilns with- out advanced heating flues and occasionally in down draft kilns when the draft was sluggish, or the kiln bottom became choked by soot or crushing of the bottom courses. We have often seen buff bricks badly streaked and coated with red, — the sides and cross bars of the setting and any part of the product from which the final evaporation took place. Fire-proofing from buff burning clays when set on solid kiln bottoms often develop a reddish color in the bottom tiles. All of these effects are produced by sulphur gas and moisture. The reddish colors in the buff burning product is due to the disintegration and solution of the iron minerals by the sulphuric acid and the iron stain is left as a red oxide coating on the surface of the ware or within the ware as the final evaporation of the acid moisture takes place from the surface or below the surface. The streaked edges in the gray bricks were due to the solution of the manganese which was taken into the bricks and finally left as a streak at a depth below the surface where the moisture was converted into vapor. Many discolorations are due to sulphur gases and conden- sation on the ware in the colder parts of the kiln, and im- provement, if not complete correction will come from slower water-smoking, better circulation, more open kiln bottoms, and stronger draft. Fire Flashing. Fire flashing is a color effect produced by reducing kiln atmospheres, but more than this, the flashed face must be in contact with the moving gases, or in other words, in the path of the flame. A buff burning clay in a reducing kiln atmosphere but not in flame contact will develop a grayish color usually speckled with iron yet when exposed to direct flame contact the result is a russet colored flash. The grayish color in the reducing atmosphere is due to the reduction of the iron to the ferrous state and its combination with silica in which the larger aggregates produce visible black spots of the iron sili- cate and similarly the disseminated and finely divided iron, but the specks are so minute that they are invisible and yet to them is due the grayish color, and the effect will fully per- meate the mass if sufficient time is given for the reduction. When the surface of the ware is exposed to flame com tact the result is a russet colored flash and exclusively a surface effect. Mr. H. B. Henderson in a paper before the American Ceramic Society has shown that the russet color is due to BURNING CLAY WARES. 333 amber colored hexagonal crystals which owe their develop- ment to carbon, — likely graphitic carbon, — from the kiln gases, and probably in a nascent state from the cracking of the gases, especially methane. The color of salt glazed ware and unglazed fire flashed ware is due to such crystals and analyses of several such glazes show a small content of carbon. It is not believed that carbon enters largely into the con- struction of the crystals nor that the color is due to carbon, but instead it is the agent which starts the crystal and perhaps continues to influence its growth throughout the period of development. The color is thought to be due to iron either dissolved in the crystal or as a constitutent of the crystal. The intensity of the color may be due to the solution of the iron. A small content of iron in glass sand has no color effect on the sand but when the sand is fused into glass the color effect of the iron is very marked, and similarly the iron which gives a buff color to fire clay wares will be intensely deeper in a crystal glaze, or it may be that the color is largely due to light refraction. With this problem we are not concerned. The analyses of the glazes do not justify the assumption that carbon is a major constituent of the crystals, and the only explanation is that it is the agent or principle directing the development of the crystals. There are several phenomena in the production of flashed effects. If a fire flashed buff burning brick is taken from the hot kiln in the final stages of the flashing process, it does not show the flashed color but instead has the grayish color of a simple reduction. Mr. J. Parker B. Fiske in an early re- port of the American Ceramic Society shows the advancing depth in the flashed color of the bricks taken from the kiln at intervals during the cooling process. Time is essential to permit the growth of the crystals. We have taken deeply flashed bricks and reburned them in the top of a continuous kiln where oxidizing conditions were a maximum and when the bricks were taken from the kiln, the flashed color had largely disappeared and instead we had the grayish color of a reduced surface. The explana- tion is that we have burned out the carbon and without its activity, the crystals cannot develop, but complete oxidation is difficult and there remains a trace of the flashed color. One company in flashing fire clay products gets an unsatis- factory dark color, but not the desired russet color. The operation in the factory is to finish the burning with clear fires and as strongly oxidizing condition as possible, and in this way the dark unsatisfactory color is overcome, leaving a light russet flash. Fire clay products can be flashed at relatively low tem- peratures by starting the flashing treatment early in the 334 BURNING CLAY WARES. burning operation and alternating with oxidizing conditions, but it is our observation that such flashes are lustreless and lack the life and brilliancy of the properly developed flash. Why we do not know. Our mental process runs to the theory that the true crystal flash is in some way due to graphitic carbon from cracked gases; that cracking, of methane par- ticularly, occurs at higher temperatures; that soot developed in the furnace and at low temperatures and held mechanically in the gases may be absorbed by the ware but does not enter into crystal development. It may have the effect of darkening the surface of the ware but the true flash development is limited, or hindered, and its brilliancy is dimmed. Smoke as we understand the term is not essential to the production of the flashed color. We have gotten an intense flash on a light buff burning ware with natural gas and throughout the burn there was no smoke in evidence. The true flash comes only when the surface of the ware is exposed to the flame. If the ware is set in saggers, or muffles, however porous they may be, thus permitting the passage of kiln gases, we do not get the flash because the walls of the saggar or muffle collect the graphite. A hole or crack in the muffle wall which permits tongues of flame to impinge on the surface of the ware will produce a flash wherever the flames touches. Bricks closely set end to end may have deeply flashed faces but the ends will not be flashed except shrinkage provides an opening, and if this opening is small, the flash will enter only a short distance and the remainder of the end of the brick will not be flashed. As a rule, — invariably so far as we know, — we cannot get the deep golden flash on a stiff mud product such as we get on a dry pressed product which explains why the dry pressed flashed product has held its market which in other color effects it has been largely displaced by the stiff mud product. The deep flash is easily obtained on the salt glazed stiff mud product. The quantity of absorbed carbon seems to influence the crystal development although it is not necessarily a consti- tuent of the crystal. We have shown that we can reduce the depth of the flash by oxidation, and it follows that in- creased carbon content will deepen the flash. If this be true we have an explanation of the difficulty in flashing a stiff mud product. Compared with the surface of the stiff mud product, that of the dry pressed product is very porous, and in consequence the surface exposure within a given area is much greater. Since the flash is a surface phenomenon any increase in the depth of the surface will give corresponding increase in the development of the flash. The viscous surface of the salt glazed ware will take up the carbon and continually present a fresh surface for further accumulation. The intention of the above discussion is not so much an attempt to explain the cause of flashing concerning which BURNING CLAY WARES. 335 there is little definite knowledge and much conjecture, but instead we wished to present the phenomena in so far as we know them in order that a better procedure may be de- veloped in the production of these color effects. The best flashed effects are produced at high temperatures, — cone 3 to cone 9, preferably the higher temperature. Reduc- ing kiln atmospheres are commonly employed but in this re- spect we have a field worthy of investigation. Excessive reduction, and by that we mean an air supply far below the theoretical requirement for perfect combustion, may produce a darker color, but it is less brilliant than a lighter reduction. In fact one may produce the golden flash at high temperatures with an excess of air in the combustion gases. Flame contact with the surface of the ware is necessary. The slower the cooling, the deeper the color of the flash, due to the greater crystal development. Unsatisfactory color effects, due, per- haps, to excessive reduction, sometimes may be corrected by finishing the burning under so called oxidizing conditions. Fire flashing, or more properly, reduction of red burning clays is a very different effect from the flash on fire clay products. In the red burning clays the brown to gun metal color is throughout the mass and only more brilliant on the surface in consequence of more fusion or glazing effect. Whether there is any graphitic absorption and crystal development or not is not known but considering the behavior of fire clays at low temperatures, it is likely there is very little true flashing in red burning clays and any slight true flash is mantled by the deep brown to black color of the reduced iron in the clay mass and its combination with silica. In limey clays we get a green color which is as truly a flash as the brown and blacks in red burning clays. In the one the green color is due to the development of lime-iron-silicate and in the other the brown is due to iron-silicate. In a red burning clay under reducing kiln conditions the red oxide of iron is reduced to the ferrous oxide which acts as a flux and combines with silica, likely also with alumina, to form a black iron silicate, or iron alumina silicate to which the dark color is due. If the reduction is slight the amount of black iron silicate developed suffices merely to darken the natural red color to a brown, becoming darker with increased reduction and finally attains the dark gun metal black in which the iron silicate predominates. The fused matrix will take into combination or solution other minerals which give color effects other than the brown to gun metal black, and thus we may get the olive greens, saffrons, and purples. If the flashings is done quickly the re- duction may not penetrate to the center of the mass of the ware and thus in bricks faced in the kiln, we get the dark edges and red centers. The widest variation in color effects comes from clays that scum badly and irregularly. The olive greens are likely 836 BURNING CLAY WARES. due to lime content in consequence of the reduction of the lime sulphate (scum) and combination of the lime with the iron silicate, just as a limey clay produces a green color when hard burned. Such a green color will be on the surface only, except the color comes from sufficient lime distribu- tion throughout the clay mass to develop the lime-iron green. The saffron colors we have attributed to sulphur effects in the water-smoking and in incomplete oxidation. We have noted the influence of sulphur in producing unusual colors and we believe that some of the peculiar color effects in reduced products may come from sulphur. One yard formerly getting a percentage of unusual colors including saffrons no longer produces them since the kiln bottoms, setting, and kiln drafts have been corrected to get better and quicker burns. Copyrighted 1920 by T. A. Randall & Co. BURNING CLAY WARES. 337 EQUALIZATION TABLES E QUALIZATION tables are used by engineers in design- ing structures involving the movement of fluids, but their use is not limited to engineers. The farmer and drain tile maker should use them in determining the proper sizes of mains and laterals. The factory superintendent who plans the kilns, dryers, stacks, hot air and gas systems, should have at hand the tables and not follow rules of thumb, assump- tions, guesses, which are only too common in clay-working factory construction, and often the cause of inefficient opera- tion and considerable loss. The following tables are used in our engineering work, but have been re-calculated and extended for this publication. The equalization table, appearing on opposite page, is for circular pipes of given diameters. The numbers to the left are the diameters of the mains and the numbers at the top are the diameters of the laterals. The numbers in the body of the table are the equivalent num- ber of laterals in each main. For example, if we wish to use six-inch laterals and desire to know how many of them are equivalent to a 20-inch main, we find on the first page of the tables in the column headed by 6 opposite the number 20, the number 20.3, which is the number of smaller circles equivalent in carrying capacity to the larger circle. Assume that we have a 30-inch main and wish to know how many 8-inch pipes may be taken off from it. In the tables opposite 30 in the column to the left and under the column headed 8 we find 28, which is the number of pipes required. Again we have 32 pipes 9 inches in diameter and wish to know the size of the main. We drop in column headed by 9 338 BURNING CLAY WARES. to the number 32, and opposite this in the column on the left we find that the main should be 36 inches in diameter. The table is extended' to cover 48-inch mains and 36-inch laterals, but it is possible to get the data for any sizes. Sup- pose we have a 60-inch main and wish to know how many 36-inch laterals it will require to equal it. Sixty is beyond the limit of the table, but if we halve it and the size of the lateral, the number of pipes is the same. Halving the two numbers gives us a 30-inch main and an 18-incli lateral. Opposite 30 and under 18 we find the number of laterals to be 3 6. Or we may quarter the sizes of the pipes, or divide them by any number to get sizes within the limits of the table and the results will be practically the same. For example, if we quarter the above given main and lat- eral w T e have 15 inches for the main and 9 inches for the lateral and from the table we find that 3.6 of the latter are equal to the former. Conversely, if we have given the number of lateral pipes and the number exceeds the limit of the table, we may deter- mine the size of the main pipe for half the size of the laterals and double the result, as for instance sixty 10-inch laterals to find the main. We look in column headed 5 and find that 62 pipes (the nearest number to 60) require a 26-inch main, and therefore a 52-inch main will be the proper size for approxi- mately 60 pipes 10 inches in diameter. The problem is the same for large mains, as for example, assume we have a 60-inch main and wish to take off 100 lat- erals, what will be the size of the laterals? Since 60 is not in the table we will take 30 and we find that 100 is between 154 in the column headed by 4, and 88 in the column is headed by 5, but it is nearer the latter than the former. Evidently the size of the lateral is between 9 and 10 and the size used will depend upon whether one wishes the carrying capacity of the main or the laterals to be in excess. If one desires the exact size, he must resort to the formula which is the ratio of the square roots of the fifth powers of the diameters of the circles. Where the numbers are small one may interpolate in the tables and get a close approxima- tion of the exact size, but interpolation is liable to wide error where the numbers are high. BURNING CLAY WARES. 339 Tables of Rectangles in Equivalents Circles HE ducts with which the engineer has to deal are quite as often rectangular as circular, and the following tables in conjunction with the Equalization Tables enables one to get the necessary data for rectangles. The numbers across the top and in the left-hand column of each table are the respective sides of the rectangles. The numbers in the body of the tables are the diameters of equiva- lent circles. The use of the tables is best shown by illustrations. Suppose we have a square stack for a down-draft kiln and wish to properly proportion the kiln floor ducts, and also to make the main draft flue a rectangle with unequal sides. We will assume the stack to be 36 inches square. The main draft flue in the kiln should be equivalent in size. In table 8, oppo- site 36 on the left, in the column headed by 36, we find the diameter of the equivalent circle to be 39.7. We may be lim- ited in the depth of the main draft flue, or the arrangement may limit us in the width. Assume that we may make the draft flue 48 inches deep, what should be the other dimension? In table 6, opposite 48, in the left-hand column, we find 39.3 under the column headed by 27, and 40.1 in the column headed by 28. Since the circle equivalent to the stack is 39.7 inches in diameter, the width of the draft flue having a depth of 48 inches will be between 27 inches and 28 inches. We may assume a width of 24 inches for the draft fine and in table 6 in the column headed by 24 we find 39.7 to be the equivalent circle for 56 inches, which is the proper depth for the flue if the width is 24 inches. Let us assume that there are twenty kiln floor ducts to be drained by the main draft flue and the stack. The stack and 340 BURNING CLAY WARES. draft flue equivalent circle is 39,7 inches, as shown above. We wish to find the the proper size for twenty rectangular floor ducts. We turn to the equalization tables and find that a 40-inch circle is equivalent to twenty 12-inch circles. Twelve inches then is the diameter of the equivalent circle for the ducts. In building these ducts one or the other dimension will be lim- ited. For instance, we may wish to use a 13-inch floor block and the width of the duct should be 9 inches. What depth must the duct be to equal a 12-inch circle? In the column headed by 9 in the rectangular table 1 we find 11.9 to be the nearest circle to 12 inches, and the other di- mension of the duct corresponding to this circle is 13 inches. The proper size for the kiln floor ducts is therefore 13 inches by 9 inches. It may be that the engineer is not limited in either width or depth, or he may be limiited in both dimen- sions, and wishes to get a compromise rectangle which will best meet the limits. The equivalent circle is 12 inches. He may start with a square which he will find to be 11 inches in table 3, equivalent to a 12 1-inch circle. Starting with the lat- ter circle in the body of the table and running up diagonally to the right he will find a 12-inch circle equivalent to a 10-incli by 12-inch rectangle, or a 11.9 circle is equivalent to a 9-inch by 13-inch rectangle, and it is also equivalent to an 8-inch by 15-inch rectangle, or a 12.1-incli circle is equivalent to a 7-inch by 18-inch rectangle. He may also work diagonally down to the left and pick out the dimensions which best suit his con- ditions. For instance, in table 1, in the last column, we find that a 12-inch circle is equivalent to a 10-inch by 12-inch rec- tangle, or, in the next column; a 11.9-inch circle is equivalent to a 9-inch by 13-inch rectangle, and, next, to an 8-inch by 15- inch rectangle. Further scanning shows that the approximate size of the rectangle may be 7 inches by 18 inches, 6 inches by 21 inches, 5 inches by 27 inches, or, turning to table 2, we find the rectangle may be 4 inches by 36 or 37 inches, 3 inches by 55 inches. He may wish narrow slots 2 inches wide, but finds that the depth exceeds the limits of the table. He may double the dimension, and find the corresponding other di- mension and doubling this get approximately the depth re- quired. In table 2 for a width of 4 inches the depth will be 36 inches or 37 inches and the proper depth for a 2-inch slot would be 72 inches or 74 inches. In such extreme dimensions there is always likely to be an error since the table is only worked out to the nearest tenth. BURNING CLAY WARES. 341 It usually happens that the tables are exceeded only in larger tlues and the error in such instances is slight. The tables may be checked up by taking dimensions within the limits and doubling and halving them. For example, a duct 40 by 60, as may be seen in table 12, has an equivalent circle of 53.7 inches, or the same is found in table 8. A 20 by 30 duct, which would be halving the dimensions of the above larger duct, has an equivalent circle of 26.9, as shown in table 5 or in table 3. Doubling this equivalent circle, we get 53.8, which practically checks the 40 by 60 dimensions. If there- fore we had an 80-inch by 90-inch duct we would look up the equivalent circle for a 40 by 45 duct which we find in table 8 or table 10 to be 46.7, and the equivalent circle, for the larger duct will be double this, or 93.4. The several problems are as follows : (1) Given a square duct or a rectangular one of fixed di- mensions, we wish to find equivalent ducts having different dimensions. This is done, as has been shown, by finding in the tables of rectangles the equivalent circle for the duct given, and picking out equivalent circles diagonally up to the right or down to the left and taking the dimensions corre- sponding. (2) Given a main duct, we wish to find the proper size of a fixed number of smaller ducts. This problem has been fully worked out above. (3) As a corollary to No. 2, we have small ducts or fixed dimensions and wish to know how many of them can be turned into a larger duct of fixed dimensions. We first find the equivalent circle of the small duct and also that of the large duct from the tables of rectangles and from the equalization table determine how many of the smaller circles equal the larger circle. (4) We have given the dimensions and number of small ducts and wish to determine the size of the main duct. We first find the equivalent circle for the small ducts from the table of rectangles and then in the equalization table find the proper circle for the number of smaller circles, and fin- ally, assuming one dimension of the main duct, we find in the table of rectangles the other dimension corresponding to the equivalent circle as found. Should the dimension thus found be too large or too small for our purpose, we can change it by assuming a different di- 342 BURNING CLAY WARES. mension in the first instance, and thus we may pick out from the table of rectangles, dimensions, corresponding to the equivalent circle, which suit our purpose. Should the sizes of the ducts exceed the limits of the tables, we can bring them within the limits by halving, and get the correct sizes by doubling the dimensions thus found, and this holds true in any proportionate part. Table 1. Table o rH COiHCOCOOOTHTtiCOOOiHCOlOh-OiM^COOOOiMT^COOOOiM'^CDQOOSTH 000iC505OOOOOHHHHIM(NiMC-KlMC0C0C0C0^^^^^^1O HHHHH(MNWW(NNlNW(N(MMNNNNWNN(N(N(N(NNNiM © G0C0C0Wt>O(M^CDfflHC010l>05Ht0l0t>01HC0TtUD00OHC0l0t> t-OOOOOOOodffiflSojgiododoHHHHHCJWC'lWWfOfOCOMM GO t-©rHCO»000©04Tt<©aO©04TH10b-©rHC010l>-aO©rHC010©00©rH 0Ot«Nt-t-OOOOOOQOQOO5ffiaO5O3OO©OOOHHHHHHH(M t— ©> rtH©00rHC0K3©00©04Tt<©00©rHC04O©00©THC0''tf©l>-©©04C0>O K0l010©©©©©t'-l>b-l>l>t~-o6o0a000o6o0©©©©®>©©©©© HHHHHHrlHHrlHHHHHrlHHHHHHHrHrlrlMIMWlN ©C0lOb-©THC0^©00®rHC0^©b-®©b-t^b^t^o6o6aOOC)o6Q6 H ri H -rH HrlTHrHHHrHHHHHrlHrlHHHHHHHrHHHHH 10 t~©rHO4''tl©00©rHO4C0Tt'<©l>®rHO4''tllO©t-©©O4C0C0»O©r^00 (MiMCOCOCOCOCOCO'^rtlTtlThlTtl^ti'^tlOlOlOlOlOJOlO©©©®©©©© HHHriHHHHrHHHrlHTHHTHHHHHHTHrlHHHrHrlrlH 04TtllOt-00©rH04C0lO®r>©©rH04C0lO®00©©©rHO4C0Tf<»O-©a0 THrH-HrHTHrH(M*00©rHO4C0TtllO©t~00©©©r-!O1C0Ttt-00 i^t^t~i>i>ooooQCaoQ6Q6o6o6o6o6o6o6o6®>©©©©j©©©©©©© tH ©©rH0404C0C0THlO.lO©©fc'»t'-00 00©©©©©rHO104C0C0'*f^THlO i0l0l0l0X0IQOl0»0X0l0l0l0l0V0l010i6©®©©©©©©©©©<© rH04COrHiO©t'-00©©rH04eOTH10©t-00©©rH04CO-tflOCOI>00©© COCOCOCOJOCOCOCOCOTfl'^’^ti'iH'^lTti'^r^Tti'^iOOiOlOlOiOlOiOlOlO© © rH rH©00 00l>lO©C0©©04Vi0 C0^K5CDI>06©ci©rHTH04O4C0C0C0'tl-H»OJO4O©©©t>l>l>t-00 06 THrlHrHrHHHHrHrHHHHHHHHrlrlHTHrl © ©Tfl©lOC0rHt-C0©'*ti©''ti©C0b-r-iO®04i0©04l000rH'Htt>- HHHHHHrlHHHHHHHHHHHrlrlrl 00 0004C0rH©©0400C000C0t'-rHO©C0b-©C0©©04K300rHTHb-©0'lTt< 04^lO©COb»o6o6©05©©THrH r^O404C0C0C0C0^t^TtI^^ , lOlClOlCCO© rlHHHHrHHHHHHHHrlHHHHrHH b- t~©©00lOrHt>*04t-O4©©^b-rHlO00rH'*tit-©C0lO00©C0lO00©O4 04'^lQl6©b-t-o6o6©©©©©rHTHrH04 0101COCOCOCOTH'^TjH'^lOlo5 HHHHHTHHHHHHrlHHHHHrHrH © lOrr-©TtH©©iH©iHlO©04©©04»OOOTH^t-©01'^©00©01 , !Ot©00 t>OOo6oO©©®©©©THrHTH0104oio4 0]COCOCOCOCO HrHHHrlHHHrHHHHHHrlrl lO C0'^04® l OOO©C0l-TH^t>©C04O00rHT^©00©04'^©00©04T^© 04 CO ^ ^ W CD © CD b* b“ 00 00 00 © © ©©©©©©rHrHrHrHrH04040' : 10 : l HHHHHHHHHHHrlrH rt< rHrHQ0TfH | 5S^00rHlO00rH^t>®04'^l>©rHO4'^©00©O4^©l>-C5© Oioia5©©©©©6TH t— 1 rH rH rH t— 1 rH rH CO 3Dt-C000 < N®©C0©00rHC0lOI>©04C0»OL'-O:©04'^lO)l>00®rHOlTH rHoicoco^^totoioiDcpocococpb-b-b-b-b-ooooaooooooooo©©© 04 L H ^ l> . T Hrtlfc-©<01rH©l>©04'^lAb-00©rH04C0lO©t-00©rH04C0Tt< rH0404C0MC0TjHT}l>l>l> rH rH'^00r-|C0fc0b-00©rH04^1OC0t-00©©rHO4 04C0'^lOlOCDl>b-00® r -| lHTH C4 ( M ( N0404COCOCOOCiCOCOCOCOCO"+'TtH-HrHTtHrtHT^TtHTtiTti'HHTtiTH rH04C0'^K3 i; £ l b-G0©©rHOlC0THlOC0l>00©©rH04C0TtrHTt(Q0iHTjfQ0THl0i00!N»0i00rHTtC0OrHTHTH (MC\|(M*rHTtt00THTH00TH'^l>OCClC00:-O iOCOCOb-b-b^OOOOOOOSOSOSOOOTHHrHTHCqfNoiffOCOCOCoHnHlO (N(M0(MiO)OOOCO»OI> ^HlOlOlOCOcdcdb-t-b-t^QOOOOOOSOSOSOSOOOiHr-irHr-JcqC^C^Cq 0(M»Ot-Oi(M'^ (N^oocoMrHHHioioioiocococot-t-t-b-odoooooooiosoioiosoo CO rH t-OCOCOO-Oi(M^I>Ci(MTHb-05(MTHCOOOrHCOlor>0(M rH^C'ioioicoeofOTtiHHH»o»oioiocdcocDcot^t-i^t^o6ooo6odoioi Ol *©04' , cHCOGO©04^COOO OHHHHM 01(MCOCOCOfoH^-HTHJf5lO)0>OCOCOCOCOCOt^t-b-b-t- 040404040404 04 04040404040404040404040404 04 040404040404040404 rH rH OOrH^t-05(MlOl>0-Oi(MrHCDO:rHCOlOl-05rHCOCOOOOrHCO»0 OiOOOOr-irHrH(M’oi(M01C'icOCOCOCO^t''+lHH'tl»OlOlO»OCOCOCDCO rH 04 04 04 04 04 04 04 04 04 04 04 04 04 04 04 04 04 04 04 04 04 04 04 04 04 04 04 04 04 o rH 04 CO^H40COt-00<3}©rH04CO^lOCOt-OOCS©TH04COrH»OCOt~OOOS© cooooooocoeofCfOco'^'^ , ^ ,; t | '^ lr ti T ^'^'^ l ^OLoioioioioiOioo»oco O4O4C504CDt-t-CDlOTf-01 '»05OrHClC0'HHl0l0C0t-b-060i0sdorH(d0ic0C0HH'HHlCil0C0 HHHHHHHHHHHHHMNC'IWMWNNNMNNC'I 00 rH OOlO^HHHOOCO^HOO^Ht-COCO^O^Oi^CiOMt-lNCOO^ HcOJ^(»dH^lffOCOrHlOCOCOt-Qdo6o505odrHTH(N(NCOCoHH»OlO t- rH OiOOMt-OOQOOOt-iOiMOb-COOCOOJb-COOOCOOOCOb-fMCOTHlOOSCCt- Ccdl>o6a>OrH(NCOTfiiO»OCOt-t>OOaDOS<35C>©THrH(N(NeOCOCOTl-Oll>-iHC0rHlO©C0b-iH»O© coiot^o6oi©rHoicocoHio»o©b-i>*oooo©ai©©rHrHrH(Nooicoooco rH’iHrHrHr— IrlHHHrtrlrlr 1 rl r 104040404040404040404 lO rH t-lO©TtlOlOrH©rHCO©lO©^QO(M©OTtHt-TH COlOCOQOOj©rHrH01foHH»OCOCOt-b-ODOOOid©©©rHiH-ioeo©t-co©'^ | ©»o©'^o5TtiQooJi©©TjHoo(MiO)o:(M C010COfc'*<350}©rH04COCO'^ , 'tfl40COCDb-b-b-OOCOCS©©©©rHTHrH04 HHHHHHrlHrlHHHHHHHrlWNNNPilNN CO rH U304lOt-t>*CO''+lrH(3}KC>04aOCOC5 , '^©CO -©Tt-t-OOo6oO©oi©©© HriHHHHHHHHHHHHrlrlrHHHHHHNN rH rH (Mt^THrHTHCSCOCOOSlOrHb-lNb-THCOO^aOlNCOCSCOCDOeOCPCKNlO CO^cdt-QOOO©0©rHoi(MCOeOrtH^lOlO»OCO©©l>*l>OOQOQ6oOOi© HrlrlHHrlHrtHrlHHHHHHHHHrlrlHrt rHO©t>Q0©©rHC^C0Tfl»0©b-Q0©OrHTHHOaOCM COrHrHnonO©©tdb-o6oO©©©©drHiHCMCMCMCOCOrHrHrHnondnd© COCOCOCOCOCOCOCOCOCOCOCOCOCOrtirtlrHrH''tl^tlrHrHrtlrH''tfrHrHrH , 'f^ © CM OOH5PHCOOlOO^OiMOOCI©O^OONCOO^OONOOCOCOOM C0C0rHrHion0©©l>ldt>00 00©©©d©rHrHCMciCMC0cdc0rHrHn0n0) COCOCOCOCOCOCOCOCOCOCOCOCOCOCOrHrHrHrHrHrHrH^rHrHrHrHrtf-tfrH 00 CM rH©rH©rH©rHOOCOt-CM©©rH©COb-rHnO©CO©©rHt-rHnOOOCMnO CMCMCOCodrHlOnO©©b*b“QOQOGO © ©©©©rHrHCMCMCMCOCOCOrHrH COCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCO' ,: t rHCMOlCOCOrHrHn0>nd©>©©b-b^o6GOo6©©©ddrHrHrHCMCMCOcdcd COCOCOCOCOCOeOCOCOCOCOCOCOCOCOCOCOCOCO^rHrHrHrHrJlTH^'©COb-©rHb-rHrHOOrHrHOO HHMClC0C0^^T)ik0 1OCDCPt-b-h-00G000CiCJodc>HHH(^(NCq COCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOrHrHrtt'-o6oOo6©©©©©©rHrHrH COCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOrHrHrH^rHrH rH Cl ©nO©''tfOOCMt-rHnO©eOt-rHnOOOCM©©COt~©COl>'©COt-©CO©© ©©©rHrHCMCMcdcdcOrHrtodnonod©©b-ldo6oOo6©>©©>C>dd© COCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOrH'tirHrH CO CM COOOCMb-rHnOdrHOOCMlO^COt-rHTtfOOrHlOOOCMnOOOiHnOOOTHrHt-© ©©©©rHrHCMCMCMCOcdcdrtlrHnonOnO©©©t~b-t^o6oOo6©©©© CMCMCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOrfi Cl CM b-rHnO©©OOCM©©''tlt>rHnO©CM©©CO©©CO©©CMKOOOCMnOOOrH coddooOrHr-icMCMCMcdcocddd^jiOiodcoddt^t-rdooooood CMCMCMCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCO rH ©rHt-T-HrHt-rHrHb-©CO©©CMnOOOrH GCa6aO©©d©©rHrHCMCMCMcdcOCOrHrH^»onOnO©©©>©b-ldl>o6 CMCMCMCMCMCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCO rHCMCO^nO©b-O0©©rHCMCOrHnO©t~O0©©rH 00 © © rH CM CO CO rH no CD CD b- 00 00 © © © tH rH CM CO rH rH rH rH rH rH tH rH CM 00 © © tH rH CM CO rH no no © b- b- 00 © © © © rH rH CM rH rH rH rH rH rH rH rH CM t- 00 © © © © rH rH rH rH rH rH rH rH rH rH tH CM no CM © © CM 00 rH © © CM b- CO oc CO £m rH t- oo o CO to © IH 00 © © rH CM CM co rH rH i£3 © © b- 00 00 © © © © rH rH rH rH tH rH rH rH rH rH CM t-O0 rHrHrHrHrHrHrHrHrHrHrHCMCMCMCMCMCMCMCMCMCMCMCMCMCMCM ClCO©'*HOO©©©©00©COrHOOnOrHOOrti©©rHl>-CMI>CMb-CMt-iH© -H©GC©©CMCOCOrHnO©t-0000©©©rHCMCMCOCOrHrHnonO©©b-t~ rHCMC0THn3©l>00©©THCMC0rHlO©b-00©©rHCMC0rHn0©b-00©© | HHHrt tH rH HHHHNN CM r> rH CD tH ID O rH 05 CO t- TP 00 CO 05 co §< o rH rH rH tH TH CO rH 33 ID rH ID rH 3 rH b- rH 00 rH 00 rH 05 rH 05 rH O ID O ID rH io rH iO CM ID CM ID CM D CO ID CO ID CM 05 ID rH b- CO 00 rH o ID O CD rH CD rH CD rH CD rH CD O ID O rH 05 CO 00 CM CD O CO on on 05 o o rH tH CM CO CO rH rH ID ID CD CD b- b- 00 00 05 05 O O O rH rH CM CM CO CO co CO rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH ID ID ID \o IO D ID ID 00 rH O CD CM CO 05 rH 05 ID O ID O ID O D O rH 05 rH 00 CO b- CM CD rH ID 05 CO eo b- 00 05 05 o o tH rH (M tH CD rH ID 05 rH 00 CM CD O rH 00 CO CD 1- on 00 05 05 o o rH rH CM 01 CO CO rH ID ID CD CD b- b- b- OO 00 05 05 o O o co CO co CO CO CO rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH ID ID ID CO oo rH O CD rH CD CM t- CM t- CM t> rH ID 05 CO t- tH CO CD CD b- 00 00 05 05 o o tH rH Dt-o6o50CMCO'^VDCDb^b-o6o50rHCMCMCOrHrHlDCDb-b- tHtHtHtHtHt— IrHCMCMCMCMCMCMCMCMCMCMCOCOCOCOCOCOCOCOCOCOCO 00 CO rHrHCOCM05rHOOCMrHCDt-OOOOOOOOb-CDrHCOrH05CDrHrHOOlDCMOOlDrH lD00OCMC0iDCD00 05OrHDlC0rHlDCDb-00 05OOrHCMC0C0^»DlDCDb” hhhhhhhcmcmcmcmcmcmcmcmcmcmcococococococococococo b- co C0OCMrH00C0CDOCM^»DlD»DlDrHC0CMrH05b-lDCMOb-rHrHb-rHOb- IDOOOCMCOlDCDGOOSOr-lDlCOrHlDCDt^OOOOOSOrHCMCMCOrHrHlDCDCD tHtHtHtHtHtHtHDICMCMCMCMCMCMCMCMCMCMCOCOCOCOCOCOCOCOCOCO CD CO C0OrH05CDrHlD00OrHCMCMCMCMrHO05b-lDC0rH00lDCM05CDC005CDCM lD00OrHC01DCDI>05OTHCMC0rHiDCDCDb-00 05OOrHCMCMC0rHrHlDCD rHr-iiHiHT— I tHt— ICMCMD4CMCMCMCMCMCMCMCMCOCOCOCOCOCOCOCOCOCO ID CO CM05O500rH05C0VDI>00O505O5C!500CDlOC0r-!O5b-rHrH00lDTH00rH!-lb- lDb“05i— icOrHCDb-000501— ICMC0rHlDCDb"00G005Oi— I t— ICMCOCO rHlDlD tHtH^tHtHtHtHDICMCMCMCMCMCMCMCMCMCMCOCOCOCOCOCOCOCOCO rH CO CMO0QOb-CMb*T— IC0lDCDb-b-CDCDlDC0CMO00lDC0Ob*rHOb-C005CDCM lDb“05rHC0rHCDb”0005Or-lCMC0'rHlDCDb-b-00 05OOr-i0'lDlC0C0rHlD tHt— ItHt-Ht— lr-lr-iCMCMCMCMCMCMCMCMCMCMCMCOCOCOCOCOCOCOCOCO CO CO iHt-t-lDTHJDOOTHCOCOrHrHCOCMrHOOOCDrHrH051DCM05CDCMOOrHrHCD lDb-05rHC0rH»Dl>00 05OrHCMC0rH»DlDCDt-00 00 05OOTHCMCMC0rHrH THrHrHiHrHiHrHCMCMCMCMCMCMCMCMCMCMCMCMCOCOCOCOCOCOCOCO CM CO OCDCDrH05C0CD05OHHTHG05 00CDrHCMOb-rHrH00»D'Hb-C005lDTH lDl>05rHCMrti»DCDo6o50THCMCMCOrHK5CDb-b^o6o505C)rHTHCMCMCOrH rHTHiHTHTHrHTHDlCMO^CMCMCMDlCMCMCMCMCMDlCOCOCOCOCOCOCO rH CO OiDlDCMb-OrHb-GOOOQOOOb-CDlDCOrHOOCDCOOb-COOCDOlOOrHOCD l0b*05HCJr)ilDCDb-0005OH(M"c0rHL0 1CCDb*00 00DOOHHCMC0C0 tHtHtHtHtHtH-HtHCMCMCMCMCMCMCMCMCMCMCMCMCOCOCOCOCOCOCO iHCMCOrH10COb-00050T-ICMCOrHlDCDb»a0050T-ICMCOrH10CDb-G0050 tH tHtHtHt-ItHtHtHtHtHCMCMCMCMCMCMDICMCMCMCO Table 9. Table 10. kD rH©©d©©CO©©dOOrH©©dOOrH©kDrH©Clt>db«COOOCOOOCO COCOrHkDkD©b^OO©©©rHrHoidCOrHrHkDkD©©>b-b-OOo6©©©* rHrHrHrHrHrHrHrHrH''HrHkDkDkDkDkDkDkDkDkDkDkDkDkDkDkDkDkDkD© 3 t-^HQOlOHOO^Ht-e005WHt-MC5^0>OHOHCPlNt-Wb-(Nt- dCOrHrHkD©©b-o6o6©©©rHrHddCOrH-^kDkD©>©b-r>00 00©© rHrHrHrH'b^b^o6o6©©©rHrHdoicOrHrHkDkD©©t~b-0000© rHrHrHrHrH-'HrHrHrHrHrHrHkDkDkDkDkDkDkDkDkDkDkDkDkDkDkDkDlDkD b- rH ©©CO©©d©kDrHb-CO©kDrHb-dOOCO©rH©rH©kD©kD©rH©rH rHdCOcdrHkDkD>©b^b^o6o6©©©rHiHddCOCOrHrHkD>©©©b-b-00 rHrHrHrH''HrHrHrHrHrHrHrHrHkDkDkDkDkDkDkDkDkDkDkDkDkDkDkDkDkD © rH rHrHOOkDrHOOrH©©dOOrH©kDrHb«db-OOOOCOOOCO©COOOCOOOOOOO Hcqdco^^wd©NNQOciQddHHddcooj^Tjiio»ocD(©Nb- rHrHrHTHrHrHrHrHrHrHrHrHrHrHkDkDkDkDkOkDkDkDkDkDkDkOkDkDkOkD kD rH ©fc-rH©©CO©kDrHb-CO©'©rHrH01dCOCOrHTHkDkD©©>l> 'H^''HTHrHrHrHrHrHrHrHrHrH''HrHkDkDkDkDkDkD©kDkDkDkDlDkDkDkD rH rH ©d©kDdO0rH©©dO0rH©kD©lDrH©rH©rH©rH©rH©rHkD©kD ©iHiH01COCOrHkDkD©©b-b^o6©©©©rHrHddCOCOrHrHkDlD©© rHrH''HrHrH-!HrHrHrHrHrHrHrH''HirHrHkDkDkDkDkDkDkDkDkDkDkDkDkDkD CO rH rHOOrHrHb-CO©kDTHb-dOOrH©rH©kD©lD©kD©kD©kD©^©COOO ©©T-ICldCOCOrHkDkD©©b^l>ad©©©©rHrHddCOCOrHrHrHkDlD rHrHrHrHrHrHrH''HrHrHrH"HlTtirH''H©t-b-o6oO©©©©>rHT-lCldCOCOrHrHkD COrHrHrHrH'HirtiTHrHrHrHrHrHrHrHrHrHTHrHIDkDkDkDkDkDkDkDkDkDkD rH M001CHt-0:05WOCDClb-(N00WC0MXM00M00Ml-Mt-HCDOTt< C5010HHoi(NCO^'H1010CDCOt-t-0000010500HHW(MO:CO^T)H COCO-HrHrHrHrHrHrHrHrHrHrH''HrHrHrHrHrHrHkDkDkDkDkDlDkDkDkDkD HCl«rtllO!Ot-OOCJOH(MM^LOCONOOOiOHClCO^lO0t-OOOlO COCOCOCOCOCOCOCOCOrH^rHrHrHrHrHrHrHrHkDkDkDkDkDkDkDkDkDkD© o rH CO 00 CO CO 05 id 05 d k o r> 05 o Cl Cl Cl Cl Cl rH rH © 00 ID CO tH © © rH ic CO 05 rH CO \o r> 00 o i— 1 CO rH kD CO 00 a o rH 00 o rH CO rH kO CO t~ 00 05 rH rH Cl CO rH kO CD b- 00 00 © © rH d rH r-l rH rH rH (M 00 o rH Cl rH kD CD I- 00 C5 O' rH Cl CO rH kO CD CD b- GO © © © rH rH rH rH tH rH d CO t- rH rH CO 00 05 o rH rH rH o o 00 t- CO rH t> 00 00 00 t- CD no rH o o Cl CO -H 1-0 CD OO 05 O rH Cl CO CO rH kD CD r- b- 00 © © rH rH rH tH tH rH Cl Cl Cl Cl Cl Cl bj Cl Cl co CO CO CO © CO CO CO co co CO CO co ro t- CO oo 05 co CO 00 (M ID b- 05 rH Cl Cl CO Cl Cl rH o 00 t- kO CO rH © co CO © 00 rH * •» kD 00 o 00 00 © rH tH 1-1 tH rH rH Cl Cl Cl Cl Cl Cl Cl Cl Cl CO CO CO CO CO 00 CO CO CO CO CO CO CO CO o t- rH rH C0 o CO kD t- 00 05 05 o 05 05 00 CD rH CO rH © b- kD Cl © © CO © id 00 o CO kD rH CO rH © b- ID CO rH 00 ID d © ID -H kD 00 Cl tH to b- (50 rH Cl CO -H lO CD 00 05 © r-l Cl Cl CO -H kD ID CO r- b* 00 rH rH tH rH tH (N Cl Cl Cl Cl Cl Cl Cl Cl CO CO CO co CO co CO CO CO co © CO L 0-1 CO rH kD CO t- 00 05 O rH Cl CO iO 0 t- 00 o © rH Cl CO rH ID CD t- 00 © © rH rH rH T -1 rH rH rH rH rH rH Cl cq Cl Cl Cl Cl Cl Cl d d CO Table 11. Table 12. s ©aOCOCOrHOOCOCOOt-THrHOOtOTHOO'^OL'-COCilOr-lt'-COGO^OtOT-l t-tr— GOOiOOrHOJCOCO^lOlOCOt— tr— 0OciCOOb-C005IO(MOOTtHOCD(Mt>COOi' 06 cia}drHrH 0 idc 0 ’c 0 TH»OlO ''tfThiTtfTFTtHlOtOlOlOlOtOlOlOlOlOlOtOlOlOlOCOCOCOCDCDCDCOCDCOCO 00 l o CO©OOCDCO©OOlOC005CD(MOO'^lCilOrHb-(MOOCOOiTH OCDh-OOOOC50rHrH(MCOCO'«^lodcDCDr>OOOodoiorHrH(N(MCOCOTH ^TfHTjH-tfTtiTHlOlOlOtOlOtOlOlOlOlOlOlOlOlOlOlOCOCOCOCOCOCOCOCD CD O 10COOOOlO(M05CDCOOt-COOCDr>o6o6dddrHT-idoicO '^'^■^■^TiHTtlTjHlOlOlOlOlOlOlOiOlOlOlOiOlOlOlOlOCDCDCDCDCDCDCD to t-1000Ood©©rHrHoicOCO^THlOCDCDt-t-o6ood©©iHr-lCooo6oioi©© '^^^^^•^^'H^^iOlOlOlOOlOLOWmiOlOWlOlOlOlOLOOCpCD s rH-o6ci©©THOieO'^lOCD iHrHrHiH(MC0CD©00©TH^TtlT^ t- to ■^CD00©©©TH(MCO^»0 THr-lrHr-lCOCOl010COo6©©iHrH(NCO'^ tHtHtH'MtH^'^'^^ T* to CDT^(N© CD©o6©©©TH(MCOTH THTHiHrHrH(N(MCDCDCOOOCOCDC505©00©©T-l(MCOTtilO©b-0005©©rH(NCO THrHrHTHrH(Mt-lOr-lt-r-l''fit-©0JC0TH , ''tilOlOlO''tiTHC0tOe0©00 CD©THCOtOt-®>©