frige ration SK REESE LIBRARY OF THK UNIVERSITY OF CALIFORNIA. Deceived , 190* . Accession No. 87602 . Class No. i ne is tne most sirnpie,~ uuiduic duu eco- nomical of ice machines. IF YOU wish to know more about this efficient machine send for our 1900 Catalog-. DO YOU USE Ammonia Fittings? We manufacture them suitable for every machine. Our Ammonia Condensers are especially noted for their economy. OUR FLANGE UNION. DIRECT EXPANSION PIPING. WE CAN HELP YOU. THE FRED W. WOLF CO 139 REES STREET, CHICAGO, U.S. A; Australian Representatives Linde Australian Refrigeration Co., 97 PITT ST., SYDNEY, AUSTRALIA. Cable Address, "WOLFTON, CHICAGO." A. B. C., A. 1, Lieber's, and Manu- facturers' Export Codes used. The Ammonia Company of Australia OFFICE: No. 52 MARGARET STREET, SYDNEY, N. S.W. WORKS: CLYDE, NEAR SYDNEY. MANUFACTURERS OF ABSOLUTELY PURE AND DRY LIQUID ANHYDROUS AMMONIA AND PURE AQUA AMMONIA AFTER PROCESSES AND UNDER LICENSE OF THE NATIONAL AMMONIA COMPANY, U.S.A., For operators of Ice Making- and Refrigerating- Machines and the trade of Australasia and the Orient. Correspondence and orders solicited. THE NATIONAL AMMONIA Co. Main Office, ST. Louis. Eastern and Export Office, 90 William St. NEW YORK. CABLE ADDRESSES AMMONIA" ST. LOUIS. AMMONIA" NEW YORK. CODES USED WESTERN UNION ATLANTIC CABLE DIRECTORY A. B. C., LAST EDITION. Largest Manufacturers and Exporters in the World of ABSOLUTELY PURE AND DRY LIQUID ANHYDROUS AMMONIA AND PURE AQUA AMMONIA FOR REFRIGERATING AND ICE MAKING. European Representatives, LIVERPOOL, ENG., JAS. SIMPSON & Co. Oriental " SYDNEY, N. S. W., THE AMMONIA Co. OF AUSTRALIA. Western SAN FRANCISCO, PACIFIC AMMONIA & CHEMICAL Co. Correspondence and Orders Solicited. THE TRIUMPH ICE MACHINE Co. CIXCLN'N'ATI, OHIO, IT. s. A. INVESTIGATE. IT FAYS. ERECTED MJr;i898. DESIGNED BY P.W.N IEBLING SEND FOR OUR ILLUSTRATED CATALOGUE, FITTINGS LIST, ETC. J. C. HORART, MANAOBK. F. W. NIEHLINO, strrT. MACHINERY FOR REFRIGERATION BEING SUNDRY OBSERVATIONS WITH REGARD TO THE PRINCIPAL APPLIANCES EMPLOYED IN ICE MAKING AND REFRIG- ERATION, AND UPON THE LAWS RELATING TO THE EXPANSION AND COMPRESSION OF GASES. PRINCIPALLY FROM AN AUSTRALIAN STANDPOINT BY NORMAN SELFE LATE CHAIRMAN OF THE BOARD OF TECHNICAL EDUCATION, NEW SOUTH WALES, AUSTRALIA. PAST PRESIDENT ENGINEERING ASSOCIATION OF NEW SOUTH WALES, AUSTRALIA. MEMBER OF THE INSTITUTE MECHANICAL ENGINEERS, ENGLAND. MEMBER OF THE INSTITUTE CIVIL ENGINEERS, ENGLAND. HON. MEMBER SOUTHERN ICE EXCHANGE, U.S.A. ETC., ETC. AUTHOR OF "COMPRESSED AIR AND ITS APPLICATIONS" ETC., ETC. H. S. RICH & CO. 1900 Copyrighted 1899-1900, by H. S. RICH & CO. ALL RIGHTS RESERVED. Press of ICE AND REFRIGERATION* CHICAGO. INTRODUCTION. Among the many marvelous strides which the nineteenth century has witnessed in connection with the arts and sciences, those which have been made in the commercial pro- duction of cold hold a very important place. The evolution of artificial refrigeration from the theoretical and experi- mental to the practical stage hardly dates back forty years, and its present vast proportions have only been approached dur- ing the last quarter of the century. The production of ice was probably the chief incentive to the work of the early inventors, but there can be no doubt that the preservation and transportation of food products, and the requirements of industries connected with cold storage, are largely respon- sible for the remarkable development of artificial refrigera- tion in later days. The mechanical processes carried out in an ordinary refrigerating establishment are, when compared with many others in which machinery is employed, exceedingly simple, but they are dependent upon principles which are not so easy to comprehend; and perhaps no branch of engineering has been less understood in the past, by those who use machinery, than that which is connected with ice making and refrigera- tion. The only books, at one time, which threw any light on the subject, dealt with it simply from the thermodynamic aspect, and for their due comprehension required the reader to be a mathematician rather than a refrigerating engineer. There are now many trade catalogues, issued by makers of refrigerating machinery, which give useful information, both 87602 iv INTRODUCTION. as to theory and practice, and some are of exceeding- merit in the scope and accuracy of the information which they furnish. The establishment of a journal like Ice and Refrigeration not only evidences the importance of the refrigeration busi- ness, but it forms a means of communication between refrig- erating engineers all over the world, and disseminates the knowledge of every improvement to the five quarters of the globe five, because Australia, where it is largely read, is not included in the orthodox four. Apart from this, the pro- prietors of that journal have published their "Compend," which to-day is the rule of faith to thousands of persons who have charge of, or are interested in, refrigerating machinery. More recently the same publishers have issued another book by "The Boy" (Mr. Skinkle), and there are a number of English works dealing with the history and progress of refrigeration, all supplying information under one or more of the many aspects which the subject presents. The author commenced his connection with refrigerating machinery in the year 1858, and with the exception of the years 1884 and 1885, when he studied its progress and improvement in the United States and Europe, he has been in Australia ever since. He should thus look at American and European rival refrigerating machines with an unprejudiced eye. His first writings on the subject were penned in the endeavor to do justice to some of the Australian pioneers in refrigeration, such as Harrison, Mort and Nicolle, whose important labors seemed to be ignored in American and European works. Other papers by him have since then been read before the Royal Society of New South Wales, and the Southern Ice Exchange of the United States, in connection with the same subjects, and have been so kindly received outside the colony that he has now been induced to attempt to write a whole book. In the following pages the reader must not expect to INTRODUCTION. v find anything" new from the theoretical side. First principles never alter, and there are many books available for those who wish to dive into the thermodynamic principles involved in the operation of the machines employed for artificial refrigera- tion, but it is believed that a great many matters relating- to the construction and practical working 1 of such machinery, as well as to the distinctive characteristics of different refrig- erating systems, are now presented, either in a new shape, or for the first time. To the average ice or cold storage man who wants to produce the greatest amount of cold, with the least primary investment of capital, the smallest cost of maintenance, and the lowest working- expenses, this little work may possibly be of some se-rvice; and if the author should at any future time learn that brother engineers like himself, more practical than literary have been helped by what follows to a fuller understanding of the requirements and possibilities of a modern refrigerating plant, it will give him the satisfaction of knowing that his efforts in this con- nection have not been altogether misapplied. No KM AN SELFE. SYDNEY, N. S. W., AUSTRALIA. TABLE OF CONTENTS. PAGE. INTRODUCTION, - - - iii_ v LIST OF ILLUSTRATIONS, ix-xv CHAPTER I. HISTORICAL, 17-25 CHAPTER II. ON HEAT AND COLD, - 26-29 CHAPTER III. THE PRACTICAL WORK OF ARTIFICIAL REFRIGERA- TION, 30-31 CHAPTER IV. COLD AIR MACHINES, - - 32-37 CHAPTER V. THE USE OF GAS WHICH LIQUEFIES UNDER PRES- SURE, 38-40 CHAPTER VI. THE LATENT HEAT OF LIQUEFACTION IN ITS AP- PLICATION TO REFRIGERATION, - 41-43 CHAPTER VII. WHY AMMONIA Is So LARGELY USED IN REFRIGER- ATING MACHINES, - 44-47 CHAPTER VIII. THE ABSORPTION SYSTEM, 48-53 CHAPTER IX. THE COMPRESSION SYSTEM REVERTED TO, - - 54-57 viii TABLE OF CONTENTS. CHAPTER X. IN THE LIQUEFACTION OF A GAS THE WORK OF THE COMPRESSOR OR PUMP is SUPPLEMENTED BY THE ACTION OF A CONDENSER OR COOLER, 58-67 CHAPTER XL THE REFRIGERATOR, - 68-70 CHAPTER XII. THE SURFACE REQUIRED FOR EXCHANGE OF TEM- PER ATURES IN CONDENSERS AND REFRIGERATORS, 71-76 CHAPTER XIII. COCKS, VALVES, PIPES AND JOINTS, 77-83 CHAPTER XIV. THE USE OF OIL IN REFRIGERATING SYSTEMS, 84-93 CHAPTER XV. THE STEAM ENGINE AND THE COMPRESSOR, - 94-173 CHAPTER XVI. ON THE LAWS RELATING TO THE EXPANSION AND COMPRESSION OF GASES, - - 174-203 CHAPTER XVII. STEAM BOILERS FOR COLD STORAGE AND ICE MAKING, - 204-230 CHAPTER XVIII. ICE PER TON OF COAL, - 231-251 CHAPTER XIX. PURE DISTILLED WATER FOR ICE MAKING, - 252-260 CHAPTER XX. SUPPLEMENTARY AND FINAL, - - 261-332 APPENDIX I. TABLES, - 333-350 APPENDIX II. REFERENCES TO LITERATURE ON REFRIGERATION AND ALLIED SUBJECTS, - - 351-358 LIST OF ILLUSTRATIONS. FIG. 1. Perkins' patent, 1834, 18 2. Harrison's ether machine, table pattern, 1860, - 20 3. Harrison's ether machine, horizontal pattern, 1861, six views, 22 4. Australian cold air machine (by the author), 1881, - 23 5. Diagram illustrating- compression and expansion of air, 32 6. Haslam cold air machine, - 34 7. Diagram of compound tandem cold air machine, 35 8. Combined air and ether machine (design), 1880, 36 9. Diagram of vapor tensions, 40 10. Latent heat diagram, - 42 11. Carbonic acid machine, Hall's patent, - 45 12. Section of carbonic acid machine, - 46 13. Nicolle's cold storage for shipboard, 1867, 49 14. Diagrammatic plan of absorption plant, - 50 15. English absorption machine, 51 16. Am moniacal liquor pumps (Australian), 17. A modern compressor and ice making plant, 18. Submerged condenser (English pattern), - 59 19. Atmospheric condenser (Frick Co. pattern), - 60 20. De La Vergne general arrangement, with special condenser, 61 21. Double submerged condenser (right way), - 62 22. Double submerged condenser (wrong way), - 63 23. Two-story atmospheric condenser, Australian, 65 24. Condensers for steam and ammonia, with water cooling tower, - 66 25. Ice box and expansion valve, 69 26. Eclipse expansion valve, - 77 27. De La Vergne expansion cock, - 77 28. Expansion valve, with long taper, - - 77 29. Solid steel manifold valve, 78 30. Solid steel manifold valve, with by-pass, - - 79 x LIST OF ILLUSTRATIONS. FIG. PAGE. 31. Double or return bend of cast metal with gland and bolts, - 79 32. Double or return bend screwed on the pipe with gland, 79 33. Straight coupling-, with double flanges, and double glands, - 79 34. Hudson Brothers' patent ammonia joint (Austra- lian), 80 35. Hudson Brothers' patent ammonia joint (Austra- lian). Another view, - 80 36. Single bends, with bolted glands, and screwed pipe, 80 37. Auldjo's patent joints for ammonia pipes, - 80 38. Auldjo's patent joints for ammonia pipes, 80 39. Auldjo's patent joints for ammonia pipes, - 80 40. Boyle joint (Chicago), 81 41. "Tight" joint (Patent), - 81 42. Old Australian ammonia pipe joint, with tongued and grooved flanges screwed and soldered to pipes (Nicolle), 81 43. Same screwed and soldered to pipes (Nicolle), - 81 44. Valves and lantern bush of horizontal double-acting compressor, 86 45. Hand oil pump, and lantern bushes to stuffing-box, 86 46. Lever oil pump with glass body, 87 47. Oil interceptor for piston rod, - 88 48. Oil separator with baffles, 89 49. Oil separator with wire screen, - 90 50. Liquid ammonia receiver, with welded end, - 91 51. Liquid ammonia receiver, with cast solid end, - 91 52. Dirt interceptor, with gauze wire screen, 92 53. Five diagrams illustrating loss by clearance, - 98 54. Compressor and engine designed in 1881 (Austra- lian), 99 55. Plan of same, - 100 56. Case compressor for ammonia. American, - 100 57. Westinghouse compressor for ammonia. American, 101 58. Antarctic single-acting compressor for ammonia. Australian, 102 59. Hercules compressor for ammonia. American, - 104 60. Auldjo compressor for ammonia. Australian, 104 LIST OF ILLUSTRATIONS. xi FIG. PAGE 61. Antarctic compound compressor for ammonia. Aus- tralian, - 104 62. De La Vergne single-acting- compressor for ammo- nia. American, 105 63. De La Verg-ne double-acting compressor for am- monia. American, - 105 64. Frick single-acting- compressor for ammonia. American, 106 65. Consolidated sing-le-acting compressor for ammonia. American, - - 107 66. York compound compressor for ammonia. Ameri- can, 108 67. Lawrence horizontal compressor for ammonia. English, - 112 68. Selfe's single-acting-compressor for-ammonia (1880). Australian, - 114 69. Selfe's compound compressor for ammonia. Aus- tralian, 114 70. Perspective of York machine, - 115 71. Linde machine plan, 116 72. Hercules beam diagram, - 116 73. Plan of Hercules beam diagram, 116 74. Card from oil injected compressor, - 118 75. Straight-line air compressor, general view, - 119 76. Pair of cards from steam and compressor cylinders of Fig. 75, 119 77. Pair of cards from Corliss engine and ammonia compressor - 120 78. Elevation of Frick machine, 125 79. Elevation of De La Vergne machine, - 126 80. Horizontal engine and two vertical compressors, diagram of elevation, 127 81. Plan of engine, two cranks and one fly-wheel, De La Vergne pattern, - 128 82. Plan of engine, two cranks and one fly-wheel, Frick pattern, 83. Plan of engine, three cranks and inside fly-wheels, 129 84. Plan of engine, three cranks and outside flv-wheels, 130 85. Plan of engine, straight shaft and discs, - - 130 86. Vertical engine and two vertical compressors, 131 xii LIST OF ILLUSTRATIONS. FIG. PAGE. 87. David Boyle's machine, general view (old style), 132 88. Boyle machine, general view (modern pattern), 133 89. Vertical engine and two vertical compressors with outside fly-wheels, 134 90. Cards from same engine and compressor as Fig. 74, but arranged at right angles, - 135 91. Small dairy machine, one engine, one single-acting compressor, 136 92. Diagrams from machine, as Fig. 91, with cranks in line, - 137 93. Diagrams from machine, as Fig. 91, with cranks at right angles to one another, - 137 94. Graphic method of ascertaining the resolution of forces in a compressor, - - 138 Perspective view Antarctic machine, beam pat- tern, 10-ton, - 142 95. Longitudinal section of same, - 143 96. Diagrams showing work of engine and resistance of compressors, as Fig. 95, - 144 97. Geared compressor, Pulsometer company (England), 145 98. Diagram of work with single-acting belted com- pressor, 148 99. Diagram of work with double-acting belted com- pressor, - 149 100. Belted single compressor and condenser combined, 150 101. Belted double compressor, - 152 102. Belted Antarctic compressor, enclosed type of com- pound compressor, 153 103. Belted Antarctic compressor, in perspective, - 154 104. English Kilbourn machine for shipboard (enclosed type), - 155 105. Boiler, compressor and condenser, all on one foun- dation, - 157 106. Linde compound arrangement, - 160 107. Antarctic compound, horizontal pattern, 160 108. Antarctic compound, section of cylinders, - 162 109. 110, 111, 112. Diagrams illustrating the compara- tive strains on the piston rods of a compressor performing the same work when single-acting, when double-acting, and compound, 165, 166, 167 LIST OF ILLUSTRATIONS. xiii FIG. PAGE. 113, 114. Actual low pressure and hig-h pressure indi- cator cards from a compound compressor, - 168 115. Diagram from the two cards, 113 and 114, to a uni- form scale, - 170 116. Diagram of belt strains with machine, Fig-. 100, 171 117. Diagram of belt strains with machine, Fig. 101, 172 118. Cylinder and piston to illustrate the relation of temperature, volume and pressure in gases, 183 119. Diagram of isothermal compression, - 187 120. Diagram illustrating accession of heat and increase of pressure by compression, 193 121. Diagram of adiabatic compression, - 196 122. Water tube with scale inside, - 206 123. Ordinary boiler tube with external scale, - - 206 124. Colonial boiler, Australian pattern, 209 125. Multitubular boiler with regenerative setting, - 210 126. Plan of same, 210 127. Front elevation of same, - 211 128. Transverse section of same, - 211 129. Sections of strengthened furnaces or flues, - 214 130. Cornish boiler front with automatic stoker, - 215 131. Longitudinal section of Lancashire boiler, - 217 132. Front elevation of Lancashire boiler, 218 133. Section of Galloway boiler, - 219 134. Cornish tubular boiler, with regenerative setting, 220 135. Plan of tubular boiler, - 221 136. Front elevation of tubular boiler, 223 137. Cross-section of tubular boiler, - 223 138. Locomotive type of boiler for fixed service, - 224 139. Bjornstad's patent blow-off cock, - - 225 140. Section of same, - 141. Diagram, thermal efficiency of steam engines, - 229 142. Diagram, horse power of compressor for two extremes of climate, - 239 143. Ice factory making distilled water from the engine exhaust, - 240 144. Exhaust steam feed heater for very bad water, by the author, 244 145. Plan of same showing exhaust and feed branches, 245 xiv LIST OF ILLUSTRATIONS. FIG. PAGE. 146. Live steam feed water heater for the deposit of impurities, - - 246 147. Plan of same, - 246 148. Can filler with sliding- float to close automatic valve, - 250 149. Can filler floating- bodily to close automatic valve, - 250 150. Can filler with telescope extension and float to close cock, - - 250 151. Ice factory with triple effect distilled water plant, - 252 152. Illustration of heat transfers in triple-eif ect plant, 256 153. Small sextuple-effect distilling- plant for pure water, - 259 154. Diagram Vog-t absorption machine, - 262 155. Generator or still, Vog-t absorption plant, - - 263 156. Equalizer or exchanger, Vogt absorption plant, 264 157. Diagram Ball absorption machine, - - 266 158. Cochran Co.'s carbonic anhydride machine, - 268 159. Kroeschell Bros.' carbonic acid machine, - - 269 160. Triplex ammonia condenser, Vogt absorption machine, 270 161. Ball ammonia condenser, - 271 162. Frick Co.'s atmospheric ammonia condenser, 272 163. Fred W. Wolf Co.'s atmospheric ammonia conden- ser, - - 273 164. Westerlin & Campbell's double-pipe condenser, 274 165. Ball discharge valve, - 277 166. Fred W. Wolf Co.'s ammonia valve, - 277 167. Frick Co.'s ammonia valve, - - 278 168. Frick Co.'s exhibit National Export Exposition, Philadelphia, 1899, 279 169. Frick Co.'s ammonia compressor cylinder, - 280 170. Section Frick Co.'s ammonia compressor cylinder, 281 171. York Co.'s mammoth machine, - 282 172. Section York compressor, 283 173. Penney's double-acting compressor, - 284 174. Latest Remington machine, 284 175. Section Remington machine, - 285 176. Linde machine, American type, - 286 LIST OF ILLUSTRATIONS. xv FIG. PAGE. 177. Same, tangye frame, 287 178. Linde compressor section, American type, - 288 179. Vilter type, ammonia compressor, 290 180. Section same, - 292 181. Triumph ice machine, - 294 182. Section same, - 295 183. Hercules machine in New South Wales. Fresh Food and Ice Co.'s works, Sydney, - 296 184. Same, latest design, - - 297 185. Same, steamship pattern, 297 186. Ball compression machine, - - 298 187. Buffalo Refrigerating- Machine Co.'s compressor, 300 188. Section same, - 301 189. Boyle compressor cylinder, 303 190. Boyle single-acting- machine with vertical engine, 304 191. Same, with horizontal engine, - 305 192. Arctic compression machine, 1879, - 306 193. Section same, 1900, 307 194. Section Case compressor cylinder, 1 - 309 195. Barber compression machine, --' 310 196. Challoner triple cylinder single-acting compressor, 312 197. Section same, 313 198. End view, same, - 314 199. Ideal refrigerating machine section, - 315 200. Diagram illustrating toggle-joint motion in same, 316 201. Vulcan compressor, section, - 317 202. Vulcan compressor, class "A," - 318 203. Same, class "B," - 319 204. Stallman compressor, - 321 205. Cross-section same, 206. Water tube boiler " Premier," - 324 207. Fire tube affected by soot and dirt, - 325 208. Water tube, same, - - 325 209. Evaporator for water heavily charged with min- erals, - 326 210. Scotch boiler, end elevation, - 326 211. Same, section, 326 212. Munroe water tube boiler, - 213. Same, vertical type, 329 CHAPTER I. HISTORICAL. Over 300 years are supposed to have elapsed since it was first discovered that artificial cold is produced by the chemi- cal action which takes place when certain salts are dissolved, but it is not known how far back the system of making- ice has been practiced which is still in use in India, where shal- low trays of porous material are filled with water and exposed to the nig-ht air, so that the heat may be abstracted by the natural evaporation which takes place. The use of frigx>rific mixtures for the abstraction of heat (many forms of which are still set out in works on chemistry) was known as far back as the year 1607, and the most common combination, that of ice and salt (which is said to have been used by Fahrenheit in 1762, when he placed the freezing- point of water at 32 as the limit of neg-ative temperature), is still in every day use for such purposes as ice cream freezing-. The production of cold by what may be termed mechani- cal means (that is by the use of a refrig-erating- machine as distinguished from chemical action) is of much more recent date. Dr. Cullen is said to have made a machine for evapo- rating- water under a vacuum in 1755, and Lavoisier experi- mented with ether in France, but the next important steps appear to come well into the present century. In the year 1810 Leslie experimented with a machine using- sulphuric acid and water. In 1824 a machine was patented by Vallance, who probably got his idea from the evaporative system so long" used in India. Under this patent, dry air was circulated over shallow trays of water when evaporation took place and heat was abstracted. In 1858 Mr. Georg-e Bevan Sloper patented a similar system in New South Wales.* Under this invention the water to be frozen was contained in canvas bag's, so that the *N. S. W. L. R. t No. 14, 1858. (2) 18 MACHINERY FOR REFRIGERATION. whole surfaces of such vessels were exposed to the evapora- tive effect of the surrounding- air as well as the surface of the water itself. The machine to work this process was designed by the author to carry out the ideas of the patentee just forty-one years ago. It was constructed in Sydney by Messrs. P. N. Russell & Co., then the leading- engineering firm in Australia, and tried in Margaret street, Sydney. No commercial success, however, did or could attend any such system of producing- artificial cold, owing to the excessive r?:^^^*--^^- 1 ^ <^' '^-^?^^:%^ FIG. 1. JACOB PERKINS' ICE MACHINE, PATENTED IN 1834. amount of power required to produce a given result; and in this particular case, as the air delivered into the chamber under partial vacuum was not made to perform work on its way from the atmosphere, it did not part with the equivalent heat beforehand, and therefore did not reduce the tempera- ture of the water, as it might have been made to do, had the knowledge of thermodynamic laws at that time been as widely extended as it is now. In 1834 Hagen used the volatile spirit of caoutchouc, and in the same year Jacob Perkins, of London, constructed what appears to have been the first ice making machine which MACHINERY FOR REFRIGERATION, 19 really worked successfully with a volatile liquid. In this machine of Perkins' ether was vaporized and expanded under the reduced pressure maintained by the suction of a pump; and the heat required for such vaporization was abstracted from the substance to be cooled. The resulting- vapor was then compressed by the same pump into a vessel cooled by water, until under the influence of the increased pressure the vapor parting- with heat to the cooling- water ag-ain condensed to a liquid, and this liquefied medium was then ready to be evaporated and expanded over ag-ain. Fig-. 1 is taken from Jacob Perkins' English patent, No. 6,662, of Aug-ust, 1834, and shows clearly that his invention included the four principal features still in use in all modern compression systems, viz.: The evaporator (1), the com- pressor (2), the condenser (3), and the expansion or regu- lating- valve (4) between the condenser and the evaporator. Althoug-h his machine was the forerunner of all the com- pression systems of the present day, Perkins does not appear to have had any more success in introducing- it for commer- cial uses than Vallance had. Dr. Gorrie, in 1845, seems to have taken the steps which led to the invention of the cold air machine, with which the names of Windhausen, Bell, Cole- man, Haslam, Lig-htfoot, Hall, Giffard and others are asso- ciated, and which were the first class of machines that were successful in carrying- meat from Australia to Europe. In 1850 Carre invented the ammonia absorption process. Between the years 1850 and 1860, Professor Twining- in America, and Mr. James Harrison, of Geelong-, in Australia, devoted themselves to the improvement of Perkins' ether machine, probably without either inventor knowing- what the other was doing-, as there was not much communication between the two countries in those days. Twining- is said to have had a machine at work between 1855 and 1857 in the state of Ohio, and Harrison, in the year 1855, was at work in Victoria when he actually produced ice with fish frozen in the block. In the year 1859 the Harrison machines were intro- duced into New South Wales, and manufactured by Messrs. P. N. Russell & Co.; the author, who at that time was in the drawing- office of the firm, was connected with this work from its initiation. 20 MACHINERY FOR REFRIGERATION. FIG. 2. HARRISON'S ETHER MACHINK TABLE PATTERN. 1859. MACHINERY FOR REFRIGERATION. 21 The original drawing- of these machines is now in his pos- session, and Fig-. 2 is a reproduction from it. As will be seen from the fig-ure, they were made as a double-table engine with four slide valves to the ether pump, a separate inlet and outlet valve on the top and bottom covers being- worked by cams and an eccentric. One of them, when completed, was set to work at the rear of the Royal hotel, Georg-e street, Sydney, and supplied ice to a reg-ular list of customers; another and simi- lar machine was sent to Melbourne. In the same year (1860) P. N. Russell & Co. made more Harrison machines to a horizontal design prepared by the author, who was then their chief draftsman. These worked for many years in New South Wales and Victoria, and were illustrated in Ice and Refrigeration for February, 1895. A large double-cylinder machine, desig-ned by the author also in 1861, is shown by Fig-. 3, on the following- pag-e. Messrs. Siebe, of London, had introduced the Harrison into Eng-land about this time, and it is g-enerally admitted in both America and Eng-land that the very first ice machine ever adopted successfully for manufacturing- purposes was one of Harrison's Australian ether machines, applied to the extraction of paraffine from shale-oil in 1861. The Engi- neer for April 12, 1861, has an illustration of Harrison's machine as made by Siebe. Dr. Kirk invented a sort of regenerative air machine in 1862, which was also used for the cooling of paraffine oil in Scotland. From the years 1861 to 1870 Mr. E. D. Nicolle, of Sydney, worked at the development of the ammonia absorp- tion system, first introduced into France by Carre, the latter years in conjunction with the late Mr. T. S. Mort. In 1863-64 he made a pump to compress anhydrous ammonia to the liquid condition, which proved tight at 30 atmospheres. He, however, considered the absorption system as the more economical in fuel, and his machines at Darlinghurst quite supplanted the Harrison ether machine in George street. Many thousands of pounds were spent by Mr. Mort in experi- ments not only with the ordinary absorption system (many practical improvements in which were patented), but on a compressed air system, L. R.,No. 181, of 1868, on an absorption or "affinity" system operated by a pump, L. R., 216, of 1869 22 MACHINERY FOR REFRIGERATION. MACHINERY FOR REFRIGERATION. 23 (see Ice and Refrigeration, April, 1899, page 298), and also on a system of using- nitrate of ammonia, which was fitted up in the ship Xortham, all under the direction of Mr. Nicolle. FIG. 4. COLD AIR MACHINE, WITH COMPOUND EXPANSION. The first practical compression machine designed in New South Wales, for the use of anhydrous ammonia as ji refrigerating medium, was patented by the author (No. 887, of 1880), and was called the "Colonial Freezing Machine." It 24 MACHINERY FOR REFRIGERATION. embodied many devices which are now in general use. (See Fig-. 68.) In 1885 the late Mr. W. G. Lock, chief engineer to the Fresh Food and Ice Co., of Sydney, patented a compound compressor for ammonia ( L. R., No. 1,729). This consisted of two single-acting high and low pressure pumps, side by side, very similar to the machines now being- made by the York Manufacturing- Co., of York, Pa., U. S. A. In 1881 the author designed the compressed air machine illustrated by Fig-. 4 fora bacon curing- business in a country district where there were only untrained men to work it and water was very scarce. Special condensers were adopted, and the water was used over and over ag-ain, less the waste by evaporation. This machine has worked successfully up to the present time and, though recently supplemented by an ammonia ma- chine, will still deliver air at 50 below zero F. As will be noted from the illustration, the expansion is compounded and, although the high and low pressure valves are worked from one eccentric, thev are connected to separate blocks in a double link motion so as to allow for different grades of expansion being adjusted to each. By this means the temperature of the intermediate chamber in the sole plate can be so regulated as to secure the deposition of mois- ture there, without affecting the cut-off and expansion in the other cylinder. Great numbers of patents have since been issued in New South Wales to local engineers for compressors of more or less originality, and for other details of refrigerating machinery, and it must not be forgotten that Mr. J. D. Postle, by his New South Wales patent, No. 180, of August 24, 1868, was one of the first persons in the world to under- stand and patent the use of an expansion cylinder in a cold air machine. By this means some of the heat held by the air is converted into work, and a lower temperature is produced. It will be seen from this short account that New South Wales has in the past done a large share of the work by which the refrigerating machinery of the world has been brought to its present perfection. It is probable that in the United States the development of the ice machine has been due more to its use in the brew- ery and to the national taste for iced water than to other MACHINERY FOR REFRIGERATION. 25 applications, and that in New South Wales the idea of freez- ing- food products for export, first suggested in 1860 by the late Aug-ustus Morris when he offered to contribute ,1,000 toward the experiment of sending- frozen meat to England was the main factor which induced the late Mr. T. S. Mort to devote his energies, and probably a quarter of a million pounds sterling-, toward the economic production of arti- ficial cold. For more particulars as to Mr. Mort's great work the author would refer those interested to an article he contributed to Ice and Refrigeration for Aug-ust, 1895. 26 MACHINERY FOR REFRIGERATION. CHAPTER II. ON HEAT AND COLD. The words u hot " and " cold " taken by themselves con- vey no definite ideas of temperature; they are merely relative terms. To any one coming in from a snow storm, water at 60 would feel warm, but to a person in a steamer's stoke- hold, at 120 or more, the very same water might feel refresh- ingly cool. If sentient being's exist under the atmosphere of the sun, then the temperature of molten iron and g^old, or the great heat to us of 2,000, may be, relatively, colder to them than we find ice cream in this world. We really know nothing- about a limit to possible heat, that is temperature in the direction of its increase, althoug-h we may admit that there is a condition of thing's in the universe where all matter exists as vapor. We have, however, throug-h modern research, the knowledge that there is a limit to the dis- appearance of heat, and that a condition is possible below which there could be no further reduction of temperature; and, as the production of cold means the abstraction of heat, at this theoretical foundationer zero point all heat must have disappeared and absolute cold be reached. From the thermometric base thus set up the properties and effects of heat can be measured and compared, and abso- lute cold being- the bottom of the scale, all degrees of tempera- ture are simply relative. It is better not to make a practice of speaking- of degrees of heat, because the word heat is applied in several different ways; and it will be well before going- further to refer to four separate expressions insepar- able from our subject, in which the word is used, viz., sensible heat, latent heat, heat unit and specific heat. SENSIBLE HEAT. Sensible heat or temperature is that heat or "hotness" MACHINERY FOR REFRIGERATION. 27 which is apparent to our senses and which we can measure with a thermometer. There are three differently scaled instruments principally used for this purpose. The French or Centigrade thermometer is considered by many the best for laboratory work and scientific research; the Reaumur is used in Germany and in some breweries; but in ice making- and refrig-erating- establishments where the English languag-e is spoken, the Fahrenheit thermometer is the one in universal use; therefore, whenever temperatures are referred to in these pag-es, unless specially excepted, they apply to Fahren- heit's scale. The zero mark or on this instrument is placed 32 below the freezing- point of water; such freezing- point is marked 32 C . The boiling- point of water at sea level is 212, and the absolute zero of temperature is 460 below zero, or 492 below the freezing; point of water. What the intensity of this cold amounts to will perhaps be better understood and appreciated when it is considered that just as much as ice is colder than molten solder, so much is absolute zero colder than melting- ice. HEAT UNIT. A heat unit is a standard of measurement by which is expressed the capacity of a given weig-ht of any body to absorb and retain heat or energy, under a given increase of sensible heat or temperature. As water possesses a greater capacity for heat than almost any other known body, its prop- erties have been adopted for such standard. In scientific circles the French unit or "calorie" is used, that being- the amount of heat required to raise one kilogramme of water 1 (from 4 to 5) Centigrade; but the British thermal unit, written B. T. U., which represents the heat required to raise one pound of water l c (from 39 to 40) F., is the uni versally recog-nized standard of heat measurements among- ice men in America, England and Australia. SPECIFIC HEAT. The specific heat of any body or substance is the capac- ity for heat which any given w r eig-ht of that body possesses as compared with an equal weig-ht of the standard, water. As the specific heat of water is expressed by unity, the spe- 28 MACHINERY FOR REFRIGERATION. cific heat of any other body is a fraction, which represents the proportion of a British thermal unit that is required to raise the temperature of one pound of any such body 1 F. The specific heat of different bodies is found to vary slightly with change of temperature, but, although this amount has to be allowed for in calculations of scientific accuracy, it is of no importance in the practical work of refrigeration. The specific heat of gases, however, varies very much under the two different conditions of their pressure in the one and their volume in the other case, being affected by the addition or abstraction of heat. This will be referred to more fully when dealing with the compression and expansion of gases. LATENT HEAT. The term latent heat is applied to the heat which is taken up by any body without causing a change of temperature, when it passes from the solid to the liquid state, or from liquid to vapor ; in the former case it is termed the latent heat of liquefaction or fusion, and in the latter case the latent heat of vaporization. When either ice or a metal is melted from the solid state at its respective melting point, or water at 212 is converted into steam at 212, then heat is taken up without a change of temperature or sensible heat, and such heat is said to be latent; and, similarly, when the reverse process takes place this latent heat is again given out. ILLUSTRATIONS OF HEAT TERMS. From the table of specific heats which follows it will be seen that the specific heat of ice is .504. The latent heat of liquefaction of water is 142 B. T. U., and the latent heat of its vaporization is 966 B. T. U. From this we are able to see that it will not suffice to abstract 212 units from steam at 212 to bring it down to ice at 0, but that 1,304.5 units must be removed as the quantity of measurement of thermal units by which one pound of water differs as regards its storage of heat under the two conditions. Thus, to reduce: A pound of steam at 212 to water at 212 represents 966. B. T. U. of water at 212 to " at 32 " 180. of " at 32 to ice at 32 " 142.4 of ice at 32 to " at " 16.1 Total 1,304.5 B. T. U. MACHINERY FOR REFRIGERATION, 29 The specific heat of iron is only between .11 and .13 (say only one-eighth part) that of water; and thus rubber bags or metal cylinders filled with hot water are used instead of hot metal, when it is required to store heat for such pur- poses as personal comfort. In old fashioned tea-urns it was customary to use an iron heater to keep the water warm, but owing- to the low specific heat of iron, any given weight of that material cooling* from 560 to 160, or through 400, would only give out the same amount of heat as an equal weight of water cooling- to 160 from the boiling- point. THE HEAT TO BE ABSTRACTED IN THE WORK OF REFRIGERATION. It is evident, therefore, that the work which has to be accomplished in cooling-, chilling- or freezing- food, or other materials, and in reducing- the temperature of the chambers in which they may be stored, is (apart from leakage and con- duction) directly dependent upon the specific heat of the various substances involved, namely, those of which the chamber is constructed, of the air which it contains, and of the goods stored therein. The following table shows the specific heats, as generally accepted, of a few of the more common materials connected with the construction of refrigerators and the substances which are usually stored therein to be refrigerated, it being noted that the specific heat of food products is largely governed by the percentage of water which they contain. Water being 1.00 Marble is .209 to .215 Ice is .504 Air is .238 Turpentine is .467 Oil is .310 Oak is .570 Alcohol is .659 Pine is .650 Strong brine, .is .700 Wrought iron ..is .113 to. 125 Vinegar is .920 Cast iron is .130 Cream is .680 Tin is .056 Milk is .90 Zinc is .095 Fat beef is .60 Copper is . 095 Lean beef is .77 Lead is .031 Fat pork is .51 Coal is .241 Veal is .70 Coke is .203 Fish is .70 to .85 Charcoal is .241 Chicken is .80 Brickwork (Bankint) . .is .200 Fruit or wgrtabta. . .is .50 to .93 Glass is .197 Eggs is .76 Stone.. ..is .270 3() MACHINERY FOR REFRIGERATION. CHAPTER III. THE PRACTICAL WORK OF ARTIFICIAL REFRIGERATION. HOW CAN A MACHINE PRODUCE COLD? Seeing* that all machines work with more or less friction, and that the power thus lost reappears in another form of energy as heat which is sensible and apparent there is some excuse for the difficulty felt by the ordinary lay mind in comprehending- the production of cold by machinery. It may be said at once that no combination of mechanism, even with unlimited power to drive it, could alone make ice from water; and that an ice machine is simply an instrument for dealing 1 with some substance which operates as a medium in such a way that it (the medium) is enabled to take up heat from the body to be cooled and transfer it to another body. Except under very special circumstances, which will be referred to later on, this heat is transferred to the water which is used for the purpose of condensation and g-oes to waste. TWO CLASSES OF MEDIUM USED. There are two distinct systems of mechanical refrigera- tion in use, both operating- by means of a medium. Under the more simple system this medium is a permanent g-as which is alternately compressed and expanded, but not lique- fied under such compression. In actual practice atmospheric air is alone used for this purpose, and the machines are termed compressed air machines. Under a more complex system of mechanical refrig-era- tion a more volatile medium is employed, and in the operation of the machinery there is alternate liquefaction and volatiliza- MACHINERY FOR REFRIGERATION. 31 tion. Althoug-h many different media have been tried, each of which has some special quality to recommend it, the prin- cipal ones to which reference will here be made are sulphuric ether, sulphurous acid, ammonia and carbonic acid. In the system introduced by Carre a solution of ammonia in water is employed, the gas is driven off by the direct application of heat, and is ag-ain reabsorbed by the water after fulfilling- its functions in the circuit of the apparatus; this is known as an absorption system." Althoug-h there are many authorities who still advocate the advantag-es of the absorption process, the greater number of refrig-erating- eng-ineers now adopt the compression system. 32 MACHINERY FOR REFRIGERATION. CHAPTER IV. COLD AIR MACHINES. These machines, coming- under the first or simpler sys- tem already referred to, operate by virtue of the law that all mechanical work has a thermal equivalent. The diagram (Fig-. 5) illustrates the action of such a machine in dealing- with a pound weig-ht of air. At atmospheric density, or 14.7 FlG. 5. DIAGRAM ILLUSTRATING THEORY OF COLD AIR MACHINES. pounds per square inch, and a temperature of 62, one pound of air possesses the intrinsic energy due to its specific heat multiplied by its absolute temperature, /. c., 62+461 c =523 ; and it occupies a volume of 13.141 cubic feet, which is repre- sented by the horizontal leng-th of the diagram. If such a MACHINERY FOR REFRIGERATION. 33 volume of air is compressed to a density of four atmospheres, then between 47,000 and 48,000 foot-pounds of energy will be required to perform the work, and if we assume a f rictionless piston and a non-conducting- cylinder, the air will not follow Mariotte's law, and by an isothermal compression occupy one-fourth or 25 per cent of its original volume at the orig- inal temperature, but will rise to a temperature of 320~, and fill 37.3 per cent instead of 25 per cent of the original volume, the difference representing the work performed by the engine in the operation of compression. Now while the medium is under this increased tension, which with cold air machines seldom exceeds five atmos- pheres, the compressed air may be passed through a con- denser or cooler and have its temperature again brought to 62, in which case the heat or energy expended upon it by the engine will be communicated to the condensing water, and for all practical purposes be lost. The air then only possesses the same intrinsic energy which it did before com- pression, but it is in a physical or mechanical condition which enables it to perform work by expanding again to atmos- pheric pressure. This expansion in practice is carried out in an engine similar to a steam engine, which assists the working of the whole refrigerating machine, and the final temperature of the air is found by simple proportion, thus: As compressed absolute temperature before condensa- tion is to compressed absolute temperature after condensa- tion that is, as 461-+ 320-= 781 C is to 461 C + 62 C = 523- so is original temperature before compression to final temperature after expansion that is 461 C + 62 C = 523 to 348 absolute = - 113 Or, to make a simple proportion sum of it 781 : 523 C : : 523 : 348 C and 348 - - 461 - 113 C In actual practice this theoretical low temperature is never reached, about 80 being the minimum, and 50 an ordinary temperature, the losses from friction and con- duction being proportionately much less in larger machines, as would be supposed. The results with these machines are also much affected by the moisture in the air and other causes. (3) 34 MACHINERY FOR REFRIGERATION. The shipment of chilled and frozen meat from Australia, the Argentine Republic, and the Cape, to England, led to a great number of ships being- fitted with air machines, and several eminent firms in England made a specialty of their manufacture. Owing- to the confined space available between decks on shipboard, these machines were often so placed as to make access to their working- parts very difficult, and when they reached the largest size the whole transverse space available in the ship was filled. Fig-. 6 represents one of the larg-est cold air machines as made by the Haslam Co., of Derby, England, and by its complication suggests that the engineer in charge would likely be glad to see it replaced by a modern ammonia machine. From an inspection of the illustration it will be seen that in this typical cold air machine FlG. 6. HASLAM COLD AIR MACHINE FOR SHIPBOARD. (still greatly used) the high and low pressure cylinders of the compound steam engine are at the crank shaft end, the two compression cylinders in the middle, and the two expan- sion cylinders, with their special slide valves, at the tail end. The surface condenser, cooler and snow-box are all arranged in the bed plate, the latter below the expansion cylinders. However complicated this machine may look, it is really a simple affair when compared with some cold air refriger- ators that have been constructed. Fig. 7 represents the cylinders forming one side of a machine which is fitted w r ith two compound tandem steam engines, two air compressors and two air expansion cylinders, all coupled to one crank shaft, and with steam condensers, air coolers and snow- MACHINERY FOR REFRIGERATION. 35 boxes in the sole plate. Such a machine was fitted in a ship to carry 80,000 carcasses of mutton from Australia, and it broke down at sea. If there is any ammonia man who thinks he has a hard row to hoe in looking- after his own machine on land, perhaps he will just consider what is the first thing- he would try to do if he was in charg-e of such a freezing- machine as this, running- eig-hty-five revolutions per minute with 160 pounds of steam pressure, the ship rolling- heavily, the tem- perature in the holds g-oing- up, $80,000 worth of meat at stake, and his compressor pistons so much in trouble that it is certain they must be g-ot out and be replaced by spare ones. In such a case, which actually occurred, the engineer in charg-e grappled successfully with it, and then had his sole L. P. STEAM //./? STEAM COMPRESSION EXPANSION FlG. 7. COMPOUND STEAM ENGINE, WITH COMPRESSION AND EXPAN- SION AIR CYLINDERS, FORMING ONE SIDE OF REFRIGERATING MACHINE FOR SHIPBOARD SERVICE. Steam engines, 14" and 24" diameters. Compressor, 30". Expansion cylinder, 23" diameter. Stroke 30. Steam pressure, 160 Ibs. Revolutions, 60 per minute. plate break by the working- of the ship and throw all out of line. This necessitated the transshipment of the whole carg-o to another vessel in Sydney harbor after a voyag-e from Queensland had been completed. GREAT POWER REQUIRED BY COMPRESSED AIR MACHINES. Both in theory and in practice compressed air machines require very much more power for a g-iven abstraction of heat, amounting- to from four to six times as much as some other machines do. They are therefore rapidly g-oing- out of use except for special purposes. It is possible in compressing- air to reach very hig-h and low relative temperatures without much difficulty, and it occurred to the author, in the early days of refrig-eration, that some of the heat or energ-y which is dissipated to the condensing- water in these machines, and -vhich is equivalent to the -whole amount of the engine power, might be utilized by combining- a compressed air refrigerator 36 MACHINERY FOR REFRIGERATION, with a modification of the Du Tremblay ether engine; and he took out a New South Wales patent in April, 1880 (No. 812), for a refrigerating- machine which had an ether engine as well as a steam engine to supply the power. FlG. 8. REFERENCE TO COMBINED AIR AND ETHER MACHINE. A 1 Low pressure compression cylinder. A 2 High B ] High " expansion B 2 Low C Steam engine D Ether E Ether pump. F ) f Condensers or exchangers to transfer the 'heat of com- G ) \ pressed air to vaporize ether. H Ordinary surface condenser for water. J Condenser or exchanger for liquefying ether vapors by ex- panded and cooled air. K Receptacle for liquid ether. L Crank shaft. M Fly wheel. N Slide valve eccentrics. O Expansion valve eccentrics. P Slipper guides. a Inlet to first compressor from chill room (through desiccator or exchanger if used). b Compressed air delivery to exchanger F. c Inlet to second compressor from exchanger F. d Compressed air delivery to exchanger G. e Inlet to air expansion high pressure cylinder. f Exhaust to exchanger J from high pressure cylinder. g Inlet to low pressure air expansion cylinder from J. // Expanded air exhaust to cold chamber. / Suction pipe to ether pump E from vessel K. k Delivery pipe from ether pump E to exchanger F. / Pipe conveying heated ether from F to G for further heating. m Pipe conveying ether vapor from G to ether engine F. n Exhaust pipe of ether engine to condenser J. o Pipe conveying liquid and condensed ether to vessel K. p \ [ ( Inlet and outlet pipes for condensing or circulating water 1 to condenser H. - Direction of air. _._._._._._._._._ Direction of ether. MACHINERY FOR REFRIGERATION. 37 In this machine the first heat was to be abstracted from the compressed air in a primary condenser or exchang-er by means of ether sprays on the condenser tubes, and the vapor thus produced was to be utilized in the ether engine to assist the steam engine and reduce the steam power necessary for the work. The machine has never been made, and it is cer- tain that in actual practice a very larg-e percentag-e of the power thus saved would be required to overcome the extra friction resulting- from the additional number of parts; still it appears absolutely certain that it is only by some such method of utilizing- the heat which is now thrown away in the condensers of refrig-erating- machines that any great fuel economy in the future of artificial refrig-eration is possible. 38 MACHINERY FOR REFRIGERATION. CHAPTER V. THE USE OF A GAS WHICH LIQUEFIES UNDER PRESSURE. In referring to the second or more complex system of mechanical refrigeration it was stated that a volatile medium such as ether, sulphurous acid, ammonia and carbonic acid was employed instead of a permanent gas, as in the air machines. Before considering- the construction of machines used with these gases it will be well to consider some of the properties of the g-ases themselves. PROPERTIES OF GASES MOST CONCERNED IN THE OPERATION OF REFRIGERATING MACHINES. It is not many years since the liquefaction of carbonic acid and ammonia, now so much used in refrigerating machines, was confined to laboratory experiments; but since it has been understood that pressure and cold were the factors necessary to liquefy them, other g-ases, which it was for a long time considered impossible to deal with, including hydrogen, have been liquefied also. Condensed liquid oxygen is now sold as an ordinary commercial product, and air, after being liquefied by the gallon, has been frozen solid. It is, therefore, possible that in the future there may be refrigerating machines operating with liquid air under the enormous pressure of, say, 2,000 pounds per square inch, with a primary condenser at, say, 90 temperature, and a secondary or tertiary condenser at 250. At the present time (omitting methylic ether, used under the Tellier system in France) the principal media used in refrigeration machines are restricted to sulphuric ether, sulphur dioxide, ammonia and carbonic acid. Now, as with the conversion of water into steam, all the substances just referred to require to take up heat to change MACHINERY FOR REFRIGERATION. 39 them from the liquid to the gaseous condition. It makes no difference to this property that the boiling- point of three of them is below the ordinary temperature of the atmosphere, so that their normal condition at atmospheric pressure is the gaseous one. As with water and steam, the boiling point of these and other gases means the temperature at which such gases will liquefy, as well as that at which their liquids will pass again to the gaseous condition; in fact a temperature under which a given weight of the material may be either entirely liquid, entirely gaseous, or partly in one state and partly in the other, depending for its condition upon the number of heat units contained in or held by it; and such temperature, as with steam, depends upon the actual pressure to which it is subjected at the time. Conversely the pressure under which any gas can be liquefied depends upon its tem- perature.* At atmospheric pressure the boiling points of these four gases are as follows: SULPHURIC ETHER. SULPHUR DIOXIDE. AMMONIA. CARBONIC ACID. + 96 -J-14 29 124 For the practical purposes of artificial refrigeration the lowest temperature to which heated gases under pressure can be reduced is limited by the temperature of the water used for condensation. f This water may be as low as 45 or 50 in temperate countries, and in hot climates may exceed 90. The diagram, Fig. 9, shows in graphic form the vapor tensions of carbonic acid, ammonia, sulphurous acid, ether and water under the temperatures met with in practical work, or the boiling points of these media under widely vary- ing conditions as to pressure. For instance, it will be seen that carbonic acid, which under atmospheric pressure will boil at 124 J below zero, requires about 1,080 pounds per square inch to liquefy it at 96, affording a great contrast to water, the boiling point of which at 14.7 pounds or one atmos- *The critical temperature of a gas is that temperature above which no increase of pressure will produce liquefaction and the g-as remains permanent. fFor experimental purposes to produce very low temperatures the condensed gas may be cooled by a second refrigeration, and a step-by- step process adopted for attaining the lowest extreme possible. 40 MACHINERY FOR REFRIGERATION. phereis212 c , and which requires the pressure reduced down to 0.089 of one pound per square inch (or a very hig-h vacuum) to enable it to evaporate at 32. Again, sulphuric ether, which boils at 96 C under atmospheric pressure, must be attenuated to at least twelve pounds below atmospheric TEMPERATURE AT PRINCIPAL MEDIA USED BOILING POINTS oS/t>e REFRIGERATING MACHINES Pourtdc I ! 1 Square inch UHIS TEN Py ERE4 Ct VAPOURS OF WATER *MO SULPHURIC ETHER ATER ETHER IA 7 IbS. 7-31 100 60 u - 96lbs. 7-2 3-6 E-6 /^o sma.1/ fo tie react FlG. 9. DIAGRAM ILLUSTRATING BOILING POINTS OF REFRIGERATING MEDIA. pressure before it will evaporate at the freezing 1 point of water. From these figures it will be noted that machines for making- ice by the evaporation of either water or ether must work with a partial vacuum, their pumps exhausting- their refrigerators to pressures below that of the atmosphere. MACHINERY FOR REFRIGERATION. 41 CHAPTER VI. THE LATENT HEAT OF LIQUEFACTION IN ITS APPLICATION TO REFRIGERATION. Although the temperature at which a volatile medium may be made to boil in the coils of a refrigerator has a very important bearing- on the production of cold, as other things being equal the lower the degree of cold produced the greater the amount of heat that can be taken up, yet there is another property of these volatile substances which has a great deal to do with the results that can be attained by their use in a refrigerating machine, and that is their latent heat of liquefaction, or the number of heat units that any given weight of such medium will take up in passing from the liquid to the gaseous condition. To make the import- ance of this property clearer, we may suppose that a pound of one medium in evaporating will abstract heat enough to bring two pounds of the substance to be refrigerated down 100, while one pound of another medium will, under similar conditions, lower the temperature of ten pounds of the same substance, but only by 50; still the medium in the second case would, other things being equal, be two and a half times as efficient for the purpose of refrigeration, because it would, in its conversion into vapor, abstract two and a half times as many thermal units from its surroundings as that in the former one. Supposing liquid, such as wort in a brewery, is the substance to be cooled, then two pounds lowered 100" represents the abstraction of 200 thermal units, while ten pounds lowered 50 would be equivalent to 500 B. T. U. Therefore, before we can ascertain the relative efficiency of two or more different media for abstracting heat in a refrigerator, we must ascertain their respective latent heats 42 MACHINERY FOR REFRIGERATION, of liquefaction under the conditions which accompany their practical application. Fig. 10 is a diagram which shows in thermal units the latent heat of one pound of each of the four principal media 60 -10 LATENT HEAT OF VAPORIZATION ; Per Pound of Medium IN BRITISH THERMAL UNITS With PKi'ncihal Media Used m RefvigeKaftnq Machines FlG. 10. DIAGRAM SHOWING LATENT HEAT OF REFRIGERATING MEDIA. before referred to, and under a range of temperature which covers their ordinary use for refrigerating purposes. USELESS WORK PERFORMED IN THE REFRIGERATOR. When a gas is liquefied under the influence of pressure (whether produced by a pump or through the direct appli- cation of heat), and the abstraction of heat by cooling it in a condenser, the resulting liquid is necessarily at a tempera- ture something above that of the condensing water, and is still under the pressure at which it was condensed; but it is in a position to change its condition again directly that influence is removed. In actual practice the pressure is retained in the condenser, or liquid receiver, by an "expan- MACHINERY FOR REFRIGERATION. 43 sion" cock or "flash " valve, which regulates the passage of the liquid refrigerant into the coils of the refrigerator, releasing- its pressure at the same time from that in the condenser to that of the refrigerator. Under these condi- tions the liquid on its release immediately boils and evapo- rates, or, in the words often used, flashes into vapor, hence the name of the valve. In so doing- it abstracts heat from the metal of the coils and the air or liquid surrounding- such coils; but it must be particularly noted that this gaseous medium has to be cooled down itself before it can cool the refrigerator to any given or required temperature, and that therefore a certain proportion of its actual cooling power is not effective for external refrigeration. The amount of heat, or the number of thermal units, that is thus lost before any useful refrigeration is done is the product of the specific heat of such medium, multiplied by the number of degrees it is lowered in temperature. All this cooling power is absolutely lost so far as any useful effect is concerned, because the medium has to be heated up again by the expenditure of more energy at every circuit it makes through the machine. THREE PROPERTIES OF A GAS CONCERNED IN FORMING AN EFFICIENT REFRIGERATING MEDIUM. From the foregoing remarks it will be understood that the relative efficiency of different gases for refrigerating purposes is mainly dependent upon three properties pos- sessed by them, and not upon any one special characteristic, and these are: 1. A low temperature of vaporization upon which depends the degree of cold that can be produced by such evaporation. 2. A high latent heat upon which depends the total number of heat units which will be abstracted by the evapo- ration of a given weight of the medium. 3. A low specific heat upon which depends the net per- centage of the heat taken up by (2), or, in other words, the proportion of the gross amount of cold produced which can be actually utilized. 44 MACHINERY FOR REFRIGERATION. CHAPTER VII. WHY AMMONIA IS SO LARGELY USED IN REFRIG- ERATING MACHINES. Although ether, chloride of methyl and several other media have been used in refrigerating- machines besides those already referred to, and some are still advocated under special conditions, yet ammonia is now used more than all the rest of them put together, experience having proved the many advantages it possesses. The principal reason why ammonia has supplanted the use of other liquids as the circulating medium in refrigerating machinery is because it has such a high latent heat of vaporization, being 555 B. T. U. at zero, against 123 for carbonic acid. That is to say, one pound of ammonia at zero in passing from the liquid to the gaseous condition would take up 555 thermal units, while the other liquids before referred to would take up less than a third and less than a fourth, respectively, of that amount. There are some compensating advantages iji the case of carbonic acid on account of its high specific gravity, which makes its heat of vaporization for a given volume very much greater than ammonia. The relative volumes at zero of equal weights of ammonia and carbonic acid are about -1 : 32.4, and thus the relative dimensions of the compressors for equal 12^ 2 V ^2 4 refrigerating effects are as ' is to 1, which equals ooo , nearly 7.2 for ammonia to 1 for carbonic acid. This quality would be an ad vantage if all other things were equal, but car- bonic acid reaches a critical condition at 88 F., and its effi- ciency rapidly falls off when the condensing water is above that temperature. Many carbonic acid machines have their refrigerator and condenser placed the one inside the other for the sake of compactness, as shown in Fig. 12, and although MACHINERY FOR REFRIGERATION. 45 not so intended by the makers, such device enables power to be expended to cool the condensing water if desired. Such an expedient is totally unnecessary with ammonia, and machines using- that material often work with the condensing water at 90 or over without any great falling- off in efficiency. FIG. 11. HALL'S CARBONIC ACID MACHINE. On shipboard there have unfortunately been many incidents to create a prejudice against ammonia, which there is little doubt were largely the result of inferior workmanship and want of care, and as a consequence carbonic acid machines are now in great favor for vessels at sea. The saving of fuel in 46 MACHINERY FOR REFRIGERATION. such cases is very large when compared with the consumption by the old cold air machines, and therefore the still greater sav- ing- that could be effected with the use of ammonia plants under like conditions is not at present receiving much attention. No doubt when the suitability of ammonia machines for sea- going ships is better understood they will supplant the car- bonic acid machine to some extent on account of their greater economy; but, owing to carbonic acid permitting the use of FlG. 12. SECTION OF SMALL CARBONIC ACID MACHINE. copper pipes for condensers, the many advantages of copper over iron when subjected to the action of sea-water will always be a heavy handicap for ammonia machines at sea. In a paper read before the Ipswich (England) meeting of the British association on "Carbonic Anhydride Machines," by Mr. Hesketh, one of the directors of Messrs. J. & E. Hall, Ltd., of Dartford, a firm that has introduced these machines all over the world, it is clearly shown that with a machine MACHINERY FOR REFRIGERATION. 47 producing- 9,360 pounds of ice per twenty-four hours from water at the various temperatures tabulated below, the inlet water for the condenser being- the same, the indicated horse power varied as follows: Temperature of water 52 75 85 90 100 I. H. P. of ensrine . . 15.62 20.03 27.2 28.2 42.10 From a series of experiments made by Messrs. L. A. Riedinger & Co., of Ansburg-, the following- results were deduced: Temperature condensing- water Ice production per hour 55 to 69 ! 95 to 97.7 485 pounds. 257 pounds. Other tests of cooling brine show that with its temper- ature reduced to 50 C and to zero, and with condensing water at 60^ and 100, the efficiency was reduced in one case 40 per cent and in the other 60 per cent by the use of the warmer condensing- water. As a contrast to these results the relative efficiency of a cubic foot of ammonia gas under different temperatures between 65 and 105 is shown by the following- table, the figures representing the refrigerating effect in thermal units, as given by Professor Siebel in the " Compend of Mechan- ical Refrigeration": Gauge suction pres- f sure in pounds,.. J Corresponding- tern. [ in refrig-erator. . . } 4 9 20 10 16 24 33 45 +30 Temp. Fahr. 650 75 850 95 1050 Gauge con- denser pres- sure in pounds. 103 127 153 184 218 Refrigerating- effect of a cubic foot of ammonia g-as in British thermal units. 33.74 33.04 32.34 31.64 30.94 42.28 41.41 40.54 39.67 38.80 54.88 53.76 52.64 51.52 50.40 68.66 67.27 65.88 64.49 63.10 85.15 83.44 81.73 80.02 78.31 106.21 104.09 101.97 92.85 97.73 Showing that with back pressure from four to forty-five pounds the increase of condenser temperature from 65 to 105 only reduces the efficiency of a cubic foot of g-as about 9 per cent. 48 MACHINERY FOR REFRIGERATION, CHAPTER VIIL THE ABSORPTION SYSTEM. Although compression machines now largely out-number those working" on the absorption principle, it must be remem- bered that the latter led the way and for a long- time carried all before them. Introduced in 1858 by Ferdinand Carre, of France, and in 1861 into Australia by Mr. E. D. Nicolle, this system was largely developed by the skill of that gentleman and the munificence of the late Mr. T. S. Mort. By the erection of ice works at Darlinghurst in 1863-64 the ammonia system supplanted the ether machines of Harrison in New South Wales at about the same time as Reece and others were working out the same problems in Europe. At the Darlinghurst works food was kept in cold storage for fifteen months, from the end of 1865 to 1867, when the plan shown by Fig. 13 was prepared by Mr. Nicolle, and seems to be the first authenticated proposal ever made for the purpose of refrigerating on shipboard. Mr. Nicolle is now seventy-five years of age, and has for many years retired from active business; he still, however, has a great disbelief in compressors, and has been for three years past working at his beautiful country home, on Lake Illawarra, developing a new process which he hopes soon to make public, with the result to use his own words to the author "of leading this interesting art into its proper chan- nel again." Under the absorption system an aqueous solution of ammonia is the medium used, instead of pure anhydrous ammonia. Taking a solution of twenty-five parts of ammonia in seventy-five parts of water, in a boiler or still, the applica- tion of heat will cause both gas and aqueous vapor (steam) to be given off in the proportion of, say, 90 per cent of ammonia MACHINERY FOR REFRIGERATION. 49 gas to 10 per cent of steam or vapor. This combined vapor is passed into a condenser under the pressure main- tained in the boiler or still, and such pressure is mainly dependent upon the temperature and volume of the condens- ing- water. As an effect of this pressure and the transfer of heat to the condensing- water the ammonia is liquefied. This liquid ammonia is then allowed to expand in the coils of the refrig- FlG. 13. SECTION OF SHIP FITTED FOR COLD STORAGE. Planned by Mr. Nicolle, 1867, for the shipment of meat from Australia to England. erator, where it either freezes or cools the substance it is employed to refrigerate. The gas being driven out of the boiler or still by the pressure generated, the solution left called the weak liquor is then drawn out and cooled in another condenser, after which the ammonia from the refrigerator and the weak mother liquor are allowed to re-unite and form strong liquor in a vessel termed the (4) 50 MACHINERY FOR REFRIGERATION, kliiiiiiijui MACHINERY FOR REFRIGERATION. 51 absorber, from which the system takes its name. After this the strong- liquor can be returned to the boiler to go through the same cycle of operations, which may be repeated over and over again. Fig-. 14 is an illustration not drawn to scale which will enable the whole process to be comprehended; from this the great importance of the desiccator and exchanger will be FlG. 15. AMMONIA ABSORPTION PLANT ENGLISH PATTERN. understood the former, by its separation of watery vapor or steam from the hot gas, saves fuel and condensing water directly; and the latter, by transferring the heat from the weak liquor (which has to be cooled before it again absorbs the gas) to the strong liquor (which is coming back to the boiler to have the ammonia driven off again), saves fuel indi- rectly as well as condensing water. 52 MACHINERY FOR REFRIGERATION. The absorption system involves a comparatively simple process, because the apparatus required consists mainly of the several vessels, pipe coils and valves, and there is no motive power or moving- machinery required except the pump to return the liquor from the absorber to the boiler. Even the pump can be dispensed with, as under several ingenious arrangements a vessel like the "Monte jus," used in sugar mills, is employed, by which the strong- liquor is lifted by the pressure of the gas to an elevated receiver and descends to the still by gravity. Fig-. 15 is a perspective sketch of an English absorption machine which has been larg-ely used in breweries. FlG. 16. PUMP TO FILL BY GRAVITY FOR AQUA AMMONIA. Under one of the patents taken out by Messrs. Mort and Nicolle in New South Wales there were two re-absorbers used which worked under pressure and vacuum respectively, and in order to overcome the difficulty of withdrawing- the liquor from the vacuum chamber the pump shown by Fig. 16 was specially designed by the author. The class of machinery used being- relatively cheap as compared with compressors, and the process being- a simple one, absorption machines are still made and used under cer- tain conditions. Since its first adoption many elaborations MACHINERY FOR REFRIGERATION. 53 have been made to the original elements in order to secure fractional distillation and desiccation of the gas, and also by means of larger exchangers to utilize more of the waste heat; but perhaps the greater amount of condensing water required rather than the greater quantity of fuel practically, if not theoretically, wasted in the absorption machine is the reason that the compression system has taken the lead in popular favor. It is found that water at atmospheric pressure and 60 F. will absorb about 700 times its volume of ammoniacal gas, and that watery vapor will often distill over with the gas, which largely discounts the efficiency of the machine, because this vapor not only requires fuel to raise it, but a supply of cold water to condense it, and although the increased amount of fuel required might not condemn the use of the absorption system where fuel is cheap, yet in most parts of Australia, having to supply double the quantity of condensing water would be a serious drawback, and has led to an increased demand for compression machines. Many changes and improvements have been made in the construction and mode of operation of absorption machines in America recently, some of the latest types of which are illustrated and described in Chapter XX of this book. 54 MACHINERY FOR REFRIGERATION. CHAPTER IX. THE COMPRESSION SYSTEM REVERTED TO. As soon as the defects of the absorption system were understood inventors reverted to the work of Perkins, Har- rison and Twining-, but it was found to be a very different matter to compress a subtle gas like ammonia up to twelve or more atmospheres than it had been to deal with ether vapor at a comparatively low tension; and the results now attained have only been reached by a long series of experi- ments which had for their object the improvement of the compressor. Toward this work English, American, Conti- nental and Australian inventors have all contributed. When we come to compare the machines of different makers, we shall find that great diversity of opinion exists with regard to details, and that many of them keep certain special points in view to the neglect of others which are not, in their opin- ion, of so much importance; hence we have a large choice of ammonia compressors in the market, some of most admir- able design and workmanship, and nearly every one of their respective agents claims for his machine that it is the best in the world. As it is hardly possible that they can all be the best absolutely, seeing how widely they differ from one another, it will be instructive to take a few of the leading types and, comparing one with the other, examine into their construction, method of operation and relative efficiency. It must be admitted that theory and natural laws have no favorites, and that the conditions which result from com- pression and expansion are the same for every one; but theory alone is of little avail in the work of the mechanical engineer, and some of the biggest failures in practice have resulted from hugging one main central theory so closely that all the little attendant theories were forgotten. The MACHINERY FOR REFRIGERATION. 55 practical experience which ensures success generally carries with it a knowledge of many little theories which the ordinary theoretical man or mathematical expert has no opportunity of making acquaintance with, and it has been mechanical FlG. 17. PERSPECTIVE VIEW OF AUSTRALIAN ICE MAKING PLANT. engineers rather than mathematicians who have brought the ammonia compressor to its present improved state. Fig. 17 shows an ice making plant recently built in Aus- tralia, where an endeavor has been made to produce a com- 56 MACHINERY FOR REFRIGERATION. pressor that should embody as many good features as pos- sible by profiting- from the experience gained with many of the well tried designs already in use there, which have been made in America, England and Germany respectively. ALL COMPRESSION SYSTEMS EMBODY THE SAME GENERAL PRINCIPLES. Reference has been made to the many admirable books which are published as trade catalogues by makers of refrig- erating- machinery, and there are undoubtedly among- those which refer to compression plants for ammonia many which are noticeable for the excellence of their illustrations and the amount of information which they make public. Some of these works, however, speak of our system and ^^^ principles on which our machines operate" in a way that'might be taken to imply that such systems and such principles were special and uncommon, whereas they are generally exactly the same as those adopted by other manufacturers in the same line. The special characteristics of the leading- makers of machin- ery are now g-enerally confined to improvements in details. The use of anhydrous ammonia and of apparatus for the liquefaction of its g-as is common property, and a compression plant of the present day embodies exactly the same four fundamental sections which Jacob Perkins showed in his 1834 patent (see Fig-. 1); that is : (1) The refrig-erator, where the medium is vaporized by the heat given up; (2) the pump to withdraw the vapor or g-as from the refrig-erator and compress it into (3) the condenser, where it is cooled and liquefied; and, lastly (4) the regulating- cock or valve, by which the admission of the liquefied medium into the refrig- erator again is regulated. These four leading features are amplified in modern plants by appliances for forcing oil into the compressor cylinder, or to the stuffing-box of the piston rod; by special devices for separating oil and foreign matters from the medium; by the use of vessels for storing the liquid refrigerant, and so on. While there is great diversity to be found in the practical construction of condensers, refrigerators and other appurte- nances that will be referred to in their place, all of these appli- ances put together do not seem to have afforded so much MACHINERY FOR REFRIGERATION. 57 scope for originality of design, as well as diversity of arrange- ment and construction, as the compressor itself. RELATION OF THE SEVERAL PARTS OF A REFRIGERATING MACHINE TO ONE ANOTHER. The relation which the refrigerator, the compressor and the condenser of a refrigerating plant occupy with regard to one another is much the same as that which exists between a steam boiler, an expansion steam engine and a surface condenser each section of the apparatus in either series begins and completes its work upon the medium employed without its efficiency being dependent upon either of the others. As the efficiency of any steam engine is entirely independent of the kind of boiler which supplies it with steam, and the efficiency of the boiler is not measured by the engine, so in a refrigerating plant no particular form or arrangement of condenser or refrigerator is necessarily coupled with any special design of compressor. Individual makers of machinery, however no doubt for sufficie/it reasons often appear to prefer and certainly do adopt cer- tain special combinations of details as their own, but this does not affect the argument that such is not indispensable for successful work. Given ample surface for the conduction of heat, plenty of section for the gas to pass without friction, a free get- away for the liquefied ammonia, or other medium, as cold as possible, and absolutely tight joints, it is more a matter of cost and convenience rather than of efficiency whether the tubes of a condenser are of small or large diameter, straight or coiled, horizontal or vertical, or even whether the conden- ser and refrigerator are made of tubes at all. The first ammonia refrigerators, made in Sydney about the year 1860, were flat boxes constructed of boiler plate, closely stayed like the walls of a locomotive fire-box, and they w r ere effective; but a coil of tubes electrically welded, such as is now procurable, would not only be more convenient, but, for a given surface, would cost only a small fraction of the amount that the stayed boxes did. 58 MACHINERY FOR REFRIGERATION. CHAPTER X. IN THE LIQUEFACTION OF A GAS THE WORK OF THE COMPRESSOR OR PUMP IS SUPPLE- MENTED BY THE ACTION OF A CONDENSER OR COOLER. THREE KINDS OF CONDKNSEKS ARE USED. Refrigerating- condensers may be divided under three separate heads. First, The " submerged," having- coils gen- erally arranged spirally and immersed in a tank of water. Second, The "atmospheric," having the coils more com- monly made of straight lengths of tube with return bends, all exposed to the air, with a trickling of water constantly flowing over them; and Third, The "evaporative," similar to the atmospheric in general arrangement, but with the addition of devices to promote the rapid evaporation of a smaller water supply from the external surfaces. Submerged Condenser. When there is an unlimited sup- ply of water the submerged condenser has certain advant- ages, one of which is that the cold water can enter its tank near the exit of the condensed gas at the bottom, rising as it becomes warmer to where it overflows, and thus, by having the gas delivered into the top ends of the tubes, its down- ward flow is in the opposite direction to that of the water, and the exit of the liquefied gas is in the coldest part of the condenser at the bottom. Besides this, in most waters the pipes keep clean longer if fully immersed. To make a sub- merged condenser thoroughly efficient the water should be kept mechanically agitated all the time it is at work, other- wise a film of warm water forms around the pipes and prevents the full transference of heat to the gradually rising body of water which overflows at the top of the condenser tank. Fig. 18 shows a condenser of this description of Eng- MACHINERY FOR REFRIGERATION. 59 lish make (it is the left hand vessel, which is in section), the four spiral coils for the ammonia and the helical-bladed agitator with its driving wheel being- clearly indicated. The right hand vessel is similar in general construction, but is a refrigerator or cooler. In condensers or refrigerators of this description the agitation should only be sufficient to keep the water moving past the surface of the coils, and should not break up the zones of temperature by setting up vertical FlG. 18. SUBMERGED CONDENSER ENGLISH PATTERN. currents, because in such a case, by making the temperature more uniform throughout, the liquid ammonia would not be cooled so much and the water would go away cooler. Atmospheric Condenser. An ordinary form of atmos- pheric condenser is seen in Fig. 19, which shows a stack of fifty-six tubes in four lines, with cast return bends and heads, and having four water distributors at the head. 60 MACHINERY FOR REFRIGERATION. With a condenser of this description the evaporation of the water flowing- over it may be so great in very dry climates and under certain conditions as to enable it to be used over and over ag-ain less the loss due to evaporation. On the dry plains of Riverina* during- a g-entle breeze, water at 90 flow- ing- on to an ammonia condenser has been known to leave the bottom coils several degrees lower in temperature, the cool- ing- effect of atmospheric evaporation more than compensat- ing- for the heat taken up from the ammonia. As the w T ater must have a downward flow over the pipes in this class of condenser, the g-as must enter at the bottom and ascend as it is being- condensed if it is to travel in the FlG. 19. ATMOSPHERIC CONDENSER AMERICAN PATTERN. opposite direction to the water. This arrangement, of course complicates the collection of the liquefied ammonia, and in the condensers of some eminent makers who adopt this plan they provide for the interception of the condensed medium by con- necting- small pipes at every alternate bend of the condenser which carry off the ammonia directly it is liquefied to the liquid storag-e tank, as shown in the g-eneral arrangement of a De La Verg-ne plant, Fig-. 20. In most atmospheric condensers of moderate size, how- ever, the g-as enters at the top, and with the condensed liquid has a continuous downward flow. In order to g-et the benefit * The country between the Murray and Murrumbidg-ee rivers in Australia. MACHINERY FOR REFRIGERATION. 61 62 MACHINERY FOR REFRIGERATION. FlG. 21. COMPOUND SUBMERGED CONDENSER CORRECT PRINCIPLP:. MACHINERY FOR REFRIGERATION. 63 FlG. 22. COMPOUND SUBMERGED CONDENSER WRONG PRINCIPLE. 64 MACHINERY FOR REFRIGERATION, of cold water to the later stages of condensation, and thus reduce the liquid medium as low as possible, while both the external and internal fluids have a downward flow with regard to the coils, several devices have been adopted. In one of these a primary condenser is submerged in a tank which is fed by the overflow from an atmospheric condenser above, and the medium, after being cooled in the lower coils, passes up again to the top of the colder condenser overhead. The advantages of such an arrangement when the scale of the plant warrants it are obvious. In other cases builders adopt two or more stages of submerged condensers, some- times as in Fig. 21, and at other times as in Fig. 22; but it is not quite clear how any gain can result from the increased complication in the latter case, where each tank has a separ- ate water supply in parallel, and the proper arrangement to save water is as Fig. 21, which shows the same condensers with a water supply in series. Fig. 23 shows a two-story at- mospheric condenser designed by the author for hot climate and scarcity of water, in which the gas flows down through the lower coils first and then passes from the bottom right up to the top of the upper coils, the liquid being drawn off separ- ately at the bottom of each coil. In such an arrangement the upper coils may be of smaller tubes than the lower ones. Evaporative Condensers. If the coils of an atmospheric condenser are covered with a light fabric which is kept wet, while an artificial current of air, propelled by a fan, is passed through them, so that a powerful evapora- tion is set up, it is possible for the water to be as cool at the bottom as at the top, just as in the instance occur- ring naturally in a specially dry climate before referred to. In such cases the gaseous and liquid medium may flow downward and still have its final cooling effected by the minimum temperature of the condensing water. The econ- omy or otherwise of such an arrangement depends entirely upon the cost per gallon of the water and its initial tempera- ture, as compared with the cost of the power required to drive the fan. THE KE-USK OF CONDENSING WATER. Although submerged condensers require a large sup- ply of water they are often used where water is costly, MACHINERY FOR REFRIGERATION. 65 FlG. 23. TWO-STORY WATER SAVING ATMOSPHERIC CONDENSER AUSTRALIAN PATTERN. (5) 66 MACHINERY FOR REFRIGERATION, MACHINERY FOR REFRIGERATION. 57 but under an arrangement by which the water is used over and over again. There are many arrangements differing in detail by which this may be effected, but they all turn upon the transfer of the heat taken up by the water, partly to the atmosphere and partly through the evaporation of a portion of the water itself. The simplest arrangement, and the one in most common use, is probably a louvred tower through which the air circulates while the water descends in a rain or spray, the shower in some cases being broken up by baffles consisting of layers of foliage or by screens of special mechanical construction. In other cases the water is played upward from innumer- able fine jets over a water-tight floor, and the diffusion into the finest spray brings every particle into contact with the air. In special cases evaporation is accelerated and cooling is effected by an upward current or blast of air from a fan or air propeller through a tower with closed sides, which is stacked in some cases with porous pottery or metal pipes and in other cases with sheets of woven wire cloth. The water is sprayed at the top of the tower bv a Barker's mill arrangement or by perforated pipes, and is divided and sub- divided at every separate layer by the obstructions placed for the purpose. These processes of cooling are used in other industries than those connected with artificial refriger- ation, notably for condensing steam engines which require a continuous supply of cool water. A very full description of them would, therefore, be rather beyond the scope of this work. Fig. 24 is an arrangement of an evaporative condenser in a cooling tower erected at such an elevation that the water flows direct to the surface condenser of the steam engine, and the whole circulation is maintained by one pump coupled to the air pump, the engine for which also drives the fan. 68 MACHINERY FOR REFRIGERATION. CHAPTER XL THE REFRIGERATOR. The refrigerator, which corresponds to the boiler in a steam engine system, generally consists of a series of tubes, through the metal of which the heat abstracted from the sub- stance being cooled is conducted to the medium which flows through them, and this heat is transferred to the medium under two distinctive systems of practical refrigeration. Under the first or brine system, as it is termed, there are coils of tubes arranged in a way similar to those of the con- denser (see left hand .vessel of Fig. 18) which are immersed in a tank of non-congealable liquid, generally a solution of ordinary salt (chloride of sodium) or chloride of calcium; from this liquid heat is taken up by the refrigerating medium. This brine derives its heat either from vessels which are immersed in it, as when ice is to be made, or from the atmos phere which surrounds it, as when chambers are to be refrig- erated. In the latter case tubes or troughs are placed in the air of the chamber through which the cold brine flows, and this cold brine abstracts the heat from the room. Fig. 25 represents an ice making tank as filled with brine in which the ice molds are inserted. The centrifugal pump at the right hand side draws the brine from under the false bottom and delivers it over the top of the end diaphragm, and so creates a perfect circulation, which can be controlled by regulating the openings in the false floor. The expansion coil is shown as in one length, the only joints being the flange connections to the manifold expansion and return valves. Instead of having a centrifugal pump for circulation within the tank, it is evident that a force pump could be employed to cir- culate the brine through a series of pipe coils on the walls or MACHINERY FOR REFRIGERATION. 69 ceiling" of a chamber, in order to withdraw the heat from the same and its contents. Under the second or direct expansion system the gas in the refrigerator coils takes up heat by direct conduction from the air of the rooms to be cooled. This transference of heat may take place in the cold chamber itself, over the walls or ceilings of which the expansion pipes may be laid. Fig. 20 shows how the De La Vergne Company increase the surfaces of these pipes by stringing on them a series of discs to act on the same principle as the "gills" of heating apparatus. Air may also be cooled by direct contact with the surfaces of the refrigerator in a separate chamber, and then be made to flow into the rooms to be cooled by a natural or forced current. This was the subject of a long since expired patent by the author. FlG. 25. BRINE REFRIGERATING TANK AND CENTRIFUGAL AGITATOR. In another system of cooling chambers, which is exten- sively adopted by the Linde company, the refrigerator cools the brine, and the brine cools a series of iron plates, alter- nately immersed in and withdrawn from it, which are arranged as revolving discs. These metallic surfaces cool the air which circulates between them, and transfers to the brine from the air the heat which it has abstracted from the gpods to be refrigerated. In this latter case there is a four- fold transference of heat and consequent loss of power, besides a great drying action, but there are compensating advantages arising from the ease with which the circulation of the air can be controlled and directed bv channels 70 MACHINERY FOR REFRIGERATION. wherever required. With direct expansion there is only a double transfer of heat, that is, from the goods to the air and from the air to the refrigerator; and on these grounds it is the more economical arrangement, especially as regards expenditure of power for a given amount of heat abstracted. The brine system has its own advantages in special cases, one of which is the great facility which it affords for storing negative energy by having large tanks of cold brine in reserve. Such reservoirs, by their capacity for taking up heat, can for more or less time be used in case of stoppage of the machine. Another advantage claimed for the brine system is the absence of danger by the escape of ammonia into the cold chambers. With the very perfect system of jointing pipes now adopted by the best refrigerating engineers, however, there is probably more of sentiment than reality underlying the fear of danger from leakage of ammonia which is felt by some persons with direct expansion. The brine system affords greater facilities for subdividing the cooling power of a large machine among a great number of separate opera- tions by reducing the care and attention required at the expansion valves, but an expert would hardly decide whether to use brine or ammonia circulation, or both combined, in any particular industry involving the use of artificial cold until he had the whole of the requirements before him. MACHINERY FOR REFRIGERATION. 71 CHAPTER XII. THE SURFACE REQUIRED FOR EXCHANGE OF TEMPERATURES IN CONDENSERS AND REFRIGERATORS. There may be scope for a great deal of personal pre- dilection in connection with the various patterns of condensers and refrigerators thus far referred to, and which manufact- urers avail themselves of to the fullest extent, often influ- enced, perhaps, by a reputation attached to their "system," and also by their available tools and appliances. Further than this any firm having- once settled on a particular pattern or method of construction would be loath to change it, if the advantages of doing 1 so were at all open to question. When, however, it comes to the amount of surface to be provided for the conduction of heat that has to be trans- ferred, and the sectional pipe area for the passage of the gas, we still find that personal fancy and u rules-of-thumb" largely prevail, although the pros and cons admit of more exact calculation, and the effect of any variation in the proportions of such parts is more easily seen and understood. Now the function of all condensers and refrigerators is to transmit heat from one substance to another, generally from a gas or vapor to a liquid, or vice versa, and through the walls of the apparatus. The amount of heat so trans- mitted is principally dependent upon two conditions, which are: First, The difference in temperature of the two sub- stances; and, Second, The superficial area of the surfaces of transmission. To a large extent also the result is depend- ent upon the velocity at which the gases, vapors or liquids move over the condensing surfaces, in a lesser degree on the relative conductivity of the substances themselves, and to some extent on the character of the metal walls of the appa- 72 MACHINERY FOR REFRIGERATION. ratus. If it were not that in practice these metal walls are relatively very thin the conducting- power of the metal would be of much greater importance than is actually the case. It may be taken as an axiom that the amount of surface necessarv in any condenser or refrigerator is directly as the number of units of heat to be transmitted, and inversely as the difference of temperature which is permissible between the two substances. The importance of having- sufficient surface thus becomes apparent if it is desired to cool as low as possible, and to utilize the maximum amount of the ma- chine's work. The walls of the condensers and refrigerators are almost invariably now of metal tube, copper being- "taboo" for use with ammonia. The relative conducting powers of the principal metals, taking gold as a standard, are as fol- ^OWS. Relative con- Relative con- Metal, ducting- power. Metal. ducting- power. Gold 1,000 Cast iron 562 Platinum 981 Wrought iron 374 Silver 973 Zinc 363 Copper 892 Tin 304 Brass 749 Lead 180 It will be noted from the above that wrought iron, the material usually employed for ammonia, is at a considerable disadvantage when compared with copper, which can be used with carbonic anhydride and sulphur dioxide machines. This difference of 374 to 892 is, relatively, very great, and would require serious consideration if large masses of metal were concerned. Actually it is of small importance in tubu- lar condensers, owing to the metal being so thin that it is practically at the same temperature on both sides of the tube. With regard to the conducting power of the gases them- selves, there do not seem at present to be any records avail- able that have been obtained by means of actual trials with working machinery, and carried out with exact instru- ments in the hands of careful observers. Laboratory experiments have been carried out by Pro- fessor Magnus upon the four following gases: Atmospheric air, hydrogen, carbonic acid and ammonia. A large tube was inserted in a glass flask containing water at the boiling point, -a delicate thermometer was fitted in the center of this large MACHINERY FOR REFRIGERATION. 73 tube, and smaller tubes enabled the larger one to be filled with the several gases. The time was noted which was required for heat to be transmitted through the several media, with the following- results: Name of Gas. Rise of Temperature. Atmospheric air From 20 to 80 From 20 to 90 3.5 minutes 1.0 2.25 3.5 " 5.25 minutes 1.4 6.3 5.5 Hydrogen ... Carbonic acid Ammonia It will be seen from the above that hydrogen shows an extraordinary power of conduction, and that carbonic acid is sluggish, while the conditions appertaining- to ammonia seem to correspond so closely to those of air, that the tables which have been obtained from experiments made on heating- air by hot water pipes may possibly be sufficiently accurate for all practical purposes, if applied to the parallel operation of heat- ing-ammonia g-as in the coils of a refrig-erator. According- to Box the loss of heat from the contact of air with cylinders two inches in diameter is .728 units per square foot for one degree of difference, the efficiency falling- with larger pipe and rising as the difference of temperature increases. When the difference of temperature reaches 150, more than two units, instead of .72 of a unit, is trans- mitted for every square foot of surface and degree of dif- ference. On page 128 (third edition) of the "Compend of Mechanical Refrigeration," this factor (M) is given as .5 unit without any reason being assigned, and if initial cost is of less importance than permanent efficiency, it is certainly taking the safe side to make it so low in figuring out for either a condenser or refrigerator. When hot, dry gas is cooled down and becomes a satur- ated vapor one would suppose that the data obtained from the surface condenser of a steam engine would be most applicable to the case of proportioning the condensing sur- face for refrigerating machines. From some experiments made by Mr. Nichols, recorded in D. K. Clark's large manual, the following results appear and show the heat transmitted both with horizontal and vertical tubes, and also with differ- 74 MACHINERY FOR REFRIGERATION. ent velocities of condensing" water flowing- over their surfaces: Vertical Tubes. Horizontal Tubes. Velocity of condens- ing- water in feet per minute 81 279 390 78 307 415 Heat transmitted per hour per sq. foot for each degree dif- ference in T. U. . . 295 383 401 422 530 | 600 The radiating- or absorbing- power of iron, according- to Peclet, equals .56 of a B. T. U. per square foot for each degree Fahrenheit difference in temperature, but it is evident this g-eneral statement is of no value for practical application. The following* table shows how the conducting- power of cylinders falls off as they increase in diameter from two inches to eig-ht inches, the units being- the number transferred per square foot for each degree difference in temperature: Diameter in Inches. 2 3 4 5 Heat in Units. .7280 .6256 .5747 .5440 Diameter in Inches. 6 7 8 Heat in Units. .5230 .5087 .4978 As this table does not include cylinders less than one inch diameter, the ratios actually given have been utilized in constructing- a curve, from which it appears that a cylinder one inch diameter, or say three-fourths inch iron pipe, would probably transmit .84 or .85 of a unit per square foot for 1 difference. The data given all go to show that the preference for small pipe is established on bed-rock truths, and they f urther sug-g-est that possibly the liquid ammonia is often withdrawn from the condenser at the temperature of liquefaction througii being- run off at once to the liquid vessel as soon as it is con- densed, when it mig-ht have been cooled a few degrees lower with advantag-e if left in the condenser long-er. Refrig-erating- authorities have deprecated the use of the bottom coils of the condenser as the permanent and only liquid vessel, and with good reason, but it is possible that if a very short extra coil of small pipe between the g-as condenser and the liquid bottle was so arrang-ed as to be always kept full of the running- liquid, either by means of a siphon or some MACHINERY FOR REFRIGERATION. 75 other device, it would enable the liquid to be brought down to within l c or 2 of the temperature of the condensing- water. The importance of this is not relatively great because after it is once liquefied there is no more latent heat to remove; but if we take the specific heat of liquid ammonia at 1.2 even then six units would be removed from every pound of liquid pass- ing- for a reduction in temperature of only 5. A point established by the steam condenser experiments is the great superiority of horizontal as compared with verti- cal tubes and the importance of velocity in the movement of the condensing 1 water. Makers of vertical tubular ammonia condensers appear to be very few in number, and results from their practice would be very interesting- for comparison if exact tests had been made and were available. The following table shows the effective surface of stand- ard pipe used in the construction of condensers and refrig- erating coils: Inside Diam. Outside Diam. in inches. External Cir- cumference in inches. i Length requir'd for a sq. ft. Surface in sq. ft. of 1 ft. in length. 1 1.315 4.134 2.903 .344 IX 1.66 5.215 2.301 .434 V/2 1.90 5.969 2.201 .497 2 2.375 7.461 1.611 .612 2# 2.875 9.032 1.382 .752 3 3.50 10.966 1.091 .911 3# 4.0 12.566 0.955 1.074 4 4.5 14.137 0.849 1.178 . The following table shows the number of thermal units to be abstracted to be equivalent to one ton refrigeration in twenty-four hours : Per Day. Per Hour. Per Minute. American Ton English Ton 284,000 312,080 11,833 26,060 197.2 216.7 From the information contained in the preceding tables it is possible to calculate the length of pipe required for any given amount of refrigeration when the temperatures of the two substances on the inside and out are known or assumed. As a practical supplement to this part of the whole refrigera- 76 MACHINERY FOR REFRIGERATION. tion question, the actual proportions of a number of con- densers by different makers have been collected and are given in tabular form for easy reference and comparison, as follows: Different condensers. Lineal feet of pipe. Size of pipe. Superficial feet per ton ice making-. Superficial feet per ton refrig-- eration (or equivalent) . ATMOSPHERIC CON- DENSERS. * "Antarctic," Sydney * " India . . . 218 { 133 '/ 133 IK IK 94.6 66.1 \ 57. 7 f (47.3) Buffalo Co., specially made for Australia. * "Hercules," Sydney. *"De La Vergne".*. *"Frick" (proposed). "Consolidated" (from printed reports) As recommended in ' ' Compend of Me- chanical Refriger- ation. " 75 114 40 62 100 115 IK 2 IK i IK 123.8 99 8 (61.9) 37.2 49.4 24.8 27.1 34.0 49 9 From experience of E. T. Skinkle 58 to 160 36 to 99 Preference of E. T. Skinkle. 150 i 516 E. T. Skinkle, aver- age of four plants from 25 to 100 tons, tabled by E.T. Skin- kle 142 i 01.0 48 8 Average of three pi ants from 75 to 150 tons, tabled by E.T. Skin- kle 99 49 9 SUBMERGED CON- DENSERS. Recommended by E.T. Skinkle. 100 i 34 4 Recorded by E. T. Skinkle as average of eight machines from 10 to 140 tons. . PIPE REQUIRED IN FREEZING TANKS. Average of twelve plants from 2 to 60 tons. (E.T. Skinkle) ' ' Con sol i d ated , ' ' from records as printed. . "Antarctic," Sydney. 89 \ 327 (272 320 292 i i IK i IK 112. ) 118. f 110. 126. 30.6 Machines made for Australian use. Other authorities borrowed. MACHINERY FOR REFRIGERATION. 77 CHAPTER XIII. COCKS, VALVES, PIPES AND JOINTS. COCKS VERSUS VALVES. Some makers pride themselves on the construction of their cocks, while others are thankful that, unlike their neig-hbors, they use nothing- but valves. Every refrig-era- tion plant requires cocks or stop-valves in great numbers besides the most important one which controls the connec- tion from the condenser to the refrig-erator and constitutes the last of the four principal features of the whole plant. These cocks or valves are required for both the forward and FIG. 26. FIG. 27. FIG. 28. back pressures and much importance is attached to their construction. During- the early days of refrig-eration, cocks were g-ener- ally adopted which were made of cast iron with steel plug's, and most careful workmanship was required to secure a per- fectly tig-ht job. A similar arrangement is still used by some leading- makers, but the preference on the whole seems at present to be given to the use of valves. Of the many well 78 MACHINERY FOR REFRIGERATION. known patterns now made for regulating- the supply of ammonia to the refrigerator the most notable perhaps are the Frick valve, as Fig-. 26, and the De La Verg-ne cock, Fig-. 27. The most simple and reliable arrang-ement of expan- sion device for ordinary purposes, however, appears to be a hard steel valve with a long- taper in a casing- of iron or steel, as Fig-. 28. (For latest Frick valve see Chapter XX.) In Australia ammonia valves were formerly made from a solid block of hammered steel, and were in fact a well known rfn i rrn FlG. 29. FORGED STEEL AMMONIA MANIFOLD VALVE. hydraulic fitting- modified to adapt it for ammonia. One of these as made in manifold for connecting- up the return ends to four coils in a refrigerator is shown by Fig-. 29. Such valves are of course more expensive to make than those with cast or malleable cast shells, but on account of their intrinsic merits they were larg-ely adopted in hig-h class work. Fig 1 . 30 shows a solid steel main stop-valve with a by-pass valve, as used for the inlet and outlet of the compressor. MACHINERY FOR REFRIGERATION. 79 PIPKS AND JOINTS. The four principal factors in the constitution of a refrig- erating- plant so far referred to would be useless if they were not connected tog-ether bv conduits or pipes. Owing- to the FlG. 30. FORGED STEEL MAIN VALVE WITH BY-PASS. action of ammonia on copper and its alloys, as already referred to, iron or steel must be employed for 'ammonia fit- ting's. Lap-welded tubes are preferred to cast iron pipes FIG. 31. FIG. 32. FIG. 33. for this purpose owing- to the risk of leakag-e throug-h poros- ity or sponginess in the casting's. Mention has before been made of the necessity for absolutely tight joints, and great 80 MACHINERY FOR REFRIGERATION. ingenuity has been expended in devising- every conceivable arrangement of joint possible for connecting- the separate leng-ths of wroug-ht iron pipes, often no doubt in order that makers mig-ht either have a patent or a claim for a system FIG. 34. HUDSON'S PATENT. FIG. 35. FIG. 36. of their own. Figs. 31 to 43 show a number of these devices which almost explain themselves; some of them have been invented and patented more than once.* FlG. 37. AULDJO'S PATENT JOINT. More loss and trouble are often caused by cheap joints than would pay for the highest class of fittings in the first instance, and for a refrigerating plant, it is safe to sav, iioth- FIG. 38. AULDJO'S PATENT OTHER FORMS. FIG. 39. ing is likely to be so dear as so-called cheap joints. Perhaps the very best all-round joint yet introduced for welded tubes, *The joint, Fig-. 40, has recently been patented in New South Wales by Doug-las Kyle. It was invented years ag-o by the late David Boyle, and known as the Boyle joint, being- patented in 1876. But strange to say it was illustrated in the German Der Constructur in 1868. MACHINERY FOR REFRIGERATION. 81 though not the cheapest in first cost, is that shown by Fig-s. 42 and 43, where the pipe is secured to the flange by sweating it with solder, as well as the screw thread, and the flanges are tongued and grooved together. This joint has been made in Sydney for over thirty years, having been introduced by Mr. E. D. Nicolle, and has FIG. 40. FIG. 41. since been found to be the best by very large American builders of refrigerating machinery, who adopt it as their own, with a slightly modified shape of flange, but with the same male and female joint and recess for a metallic or other grommet. FIG. 42. AX AUSTRALIAN AMMONIA JOINT. FlG. 43. ELECTRIC WELDING. The introduction of electric welding by which pipes can now be made up into long continuous coils has been a great boon to makers of refrigerating machinery, and has enabled joints to be largely dispensed with in out-of-the-way places, where a leak would be difficult to detect and stop. (6) 82 MACHINERY FOR REFRIGERATION. In shops where the amount of coil work turned out does not warrant the outlay for an electric welding plant, wrought iron pipes of good quality may be successfully welded in an ordinary fire, after the two ends have been machined so as to make a male and female cone. A special mandril should be introduced during the swaging. Long lengths so treated, and afterward bent on the welds have stood the test press- ure of 1,500 pounds per square inch, as well as electrically welded tubes. SEVERAL DESCRIPTIONS OF COILS EMPLOYED. When all the separate lengths of tube required for one section of a refrigerator or condenser are welded up into a continuous length they can be easily bent into the kind of coil required, whether a plain helix or spiral, as in Figs. 18, 21 and 22, or an oblong spiral, as in Fig. 23, and the several turns of the coils can be laid as closely together vertically as desired; but the horizontal distance of the two sides apart must be greater, being regulated by the radius of the bends at the ends. When, however, a zigzag arrangement with ver- tical returns is desired, then the several lengths must be spaced wider apart vertically on account of these bends in the tube; and in order to get the greatest number of lengths in the space available the inclined arrangement, as shown in Fig. 25, is often adopted. This design is in some respects objectionable because the liquid must be all evaporated in the first length and bend unless the pressure is sufficient to drive it up-hill to the next bend. The same objection does not apply to these coils laid horizontally, and condensers are sometimes made with vertical headers for the main inlet and outlet, connected by a number of zigzag coils placed hori- zontally one over the other, with or without valves. These condensers have the advantage of giving a short run, a large sectional area of passage and a slow velocity for the gas, hence they cause very little increase of pressure by friction. When a zigzag coil is made of straight tubes and separ- ate returns, as in Fig. 19, instead of with the bends in the tubes themselves, a condenser or refrigerator with a given number of lengths above one another can be kept much lower than otherwise, because the return ends may be cast MACHINERY FOR REFRIGERATION. 83 much closer than the wrought pipes could safely be bent to. Zigzag- coils made in both ways are much used for the floors and sides of refrigerating- chambers, those made with the bends in the pipe itself requiring much fewer connections. When built up from separate lengths they are generally connected at the returns in one or other of the following ways : 1. Cast metal returns, and the screwed and soldered male and female flanges with metallic grommet, as in Fig. 31, all connected up by bolts. 2. Cast metal returns, with screwed socket and addi- tional recess for packing, the ends of the pipes screwed hard into the sockets, and followed up by a packing ring and a gland running on the thread of the pipe, as shown in Fig. 32. 3. Similar to 2, with the ends of the tubes screwed into socket, but with the gland to compress the packing running on the plain body of the tube, and drawn up by two bolts, as shown in Fig. 31. 4. Cast returns screwed right and left-hand alternately and formed with a recess containing soft metal packing that can be closed up by a set screw, and the pipes screwed right and left handed at opposite ends, as shown in Fig. 41. (The author has no personal experience with this joint, but it is claimed as a great advantage that any pipe or return can be easily changed under this system, and the whole kept easily tight.) 5. But when the returns are bent on the separate tubes themselves, then the joints on the straight portion of the tubes may be made with any form of flange or socket as used in any other position. Figs. 34 and 35 show the joint pat- ented and used by a large firm of Sydney engineers, the ends of the pipes being machined to fit into a double-grooved socket. 6. With continuously welded coils the connection to manifolds or headers is frequently made by an Australian flange, as shown in Figs. 42 and 43. 84 MACHINERY FOR REFRIGERATION. CHAPTER XIV. THE USE OF OIL IN REFRIGERATING SYSTEMS. SUPPLY OF OIL TO THE COMPRESSOR. In the early days of ammonia compression, and before the accurate mechanical construction now possible and usual was put into such machines, compressors would not deliver so large a percentage of the cylinder's total volume as they do now. The pistons were, no doubt, not so accurately fitted that the ammonia itself would furnish all the lubrica- tion required, and the clearance was excessive. With a view to the expulsion of the whole cylinder's contents a system of compression was adopted for ammonia similar to that used with wet air compressors, and in some very high class machines now made the cylinder at every stroke receives an injection of liquid; and as this requires a substance which will not saponify under the action of ammonia, special grades of hydrocarbon or mineral oil are prepared for the purpose. The advocates of such an arrangement contend that the oil not only fills all the interstices resulting from bad design, and reduces the effective clearance to ;///, however great the mechanical clearance may be, but that it takes up a great deal of heat from the gas; and thus bv reducing the volume of the same reduces the power required for the work of com- pression. No doubt all this is true in a degree, but it is at the expense of a reduced piston speed and therefore a reduced compressor capacity, because oil cannot be banged about as gas may be. Besides this, the plant must be pro- vided with a complete system of pumps, separators and con- densers for circulating and cooling such oil and restoring it to a reservoir, freed from ammonia, to be used over again. All of these special features have to be taken into account when comparing the first cost of plant and working expenses under this system with the cost of equal results obtained MACHINERY FOR REFRIGERATION. 85 from others. An inspection of Fig-. 20, and comparison of the same with Figs. 17 and 24, will enable the much greater complexity of the oil system to be better understood. NO OIL NECESSARY IN SOME COMPRESSORS. Some makers of high class modern machinery claim that no oil at all is required for the pistons of their compressors, and only use it as a seal to the piston rod. In single-acting vertical compressors a little oil lying in the bottom of the cylinder around the neck bush must necessarily prevent the passage of gas through the packing, and it is only subjected in such cases to the back or expansion pressure. In ordinary double-acting compressors, however, the piston rod and its packing are subjected to the full forward or condensed pres- sure. In an ordinary double-acting compressor working horizontally, as in the Linde system, no body of oil can lie round the neck ring, and it is usual in such cases to have a very long stuffing-box, with a lantern bush separating two sets of packing, as shown by Fig. 44. A small oil pump, generally driven by the machine, or a lubricator, keeps up a supply of oil to this intermediate space, and under such an arrangement any ammonia that escapes is absorbed by the oil, which is carried to a special vessel, where it is separated and then used over again. A certain amount of oil is also carried on the piston rod into the cylinder to lubricate the piston at every stroke, which necessarily requires it more than a vertical machine. In vertical single-acting compres- sors an oil vessel is often attached which has a small hand- pump fitted to it by which the attendant can force oil into the bottom of the cylinder to seal the piston rod as required. See Figs. 45 and 46. The latter is a special design by the author, and has a glass bottom to show the quantity of oil in the well. The oil which escapes through the packing in vertical compressors naturally runs down the piston rod, and to catch the same and keep the machine clean the piston rod often runs through a bowl or cup on the crosshead, as seen in Figs. 11 and 17. In Fig. 11, where glycerine is used as a lubricant, which is forced in between the double leathers of the piston and 86 MACHINERY FOR REFRIGERATION. FlG. 44. SECTION OF LINDF STUFFING-BOX. FlG. 45. OIL PUMP TO LANTERN PACKING. MACHINERY FOR REFRIGERATION. 87 packing-, the cup has an overflow pipe into a portable receiver, so arranged as to be emptied by hand. Fig-. 47 shows a device specially desig-ned by the author to intercept this oil by a second and lighter packing- in a lower stuffing-box, and a circular trough with pipe to carry it to a receiver. FlG. 46. LUBRICATING PUMP WITH GLASS BODY. It will be noticed that any oil which passes the upper or main packing can escape through openings above the lower or "swab" packing and run over a "drip" into an annular 88 MACHINERY FOR REFRIGERATION, channel; a pipe leads the oil from this channel to a reservoir, either cast in the frame or attached, whence it can run down to the glass reservoir of the pump seen in Fig-. 46, to be ag-ain returned to the compressor. FIG. 47. OIL INTB:RCEPTOR FOR PISTON ROD BY THE AUTHOR. In machines of the class shown by Fig-. 11, where the pressure often runs up to 1,100 pounds to the inch, an extremely simple system of automatic lubrication of the MACHINERY FOR REFRIGERATION. 89 FlG. 48. SECTION OF OIL SEPARATOR WITH "BAFFLES. 90 MACHINERY FOR REFRIGERATION, FlG. 49. SHCTION OK OIL SEPARATOR WITH WIRE SCREEN. MACHINERY FOR REFRIGERATION. 91 piston rod is adopted. The vessel shown at the side of the machine to hold the lubricant has a small pipe with regu- lating- valve to adjust the flow to the packing, and also has a pipe which puts it in communication with the full forward pressure. After being filled with glycerine from the upper vessel the filling valve is closed and the pressure valve opened; it is then only necessary to adjust the small valve, seen on the pipe to the stuffing-box, to the flow required. Owing to the catches provided on the crosshead this can be used over and over again. FlG. 50. LIQUID AMMONIA RECEIVERS. FlG. 51. As an escape of gas takes place every time the lubri- cating vessel has to be filled, this system is not so well adapted for ammonia machines, but the small quantity of car- bonic acid which escapes would not be noticed. SEPARATION OF OIL FROM THE AMMONIA. Seeing that oil is almost invariably used in refrigerating compressors, it becomes necessary to interpose certain ves- 92 MACHINERY FOR REFRIGERATION. sels in the course of a refrigerator system to prevent it being carried into the pipe coils of the condenser and refrig- erator, where it would materially reduce the efficiency of the pipe surface as a conductor of heat. The principal oil sepa- rator in a system is usually fitted on the main pipe between FlG. 52. SECTION AND PLAN OF INTERCEPTOR. the compressor and the condenser, and some experts attach great importance to this vessel being very large. Fig. 48 shows such a vessel as made for fixing to a wall, and pro- vided with baffle plates to facilitate the deposition of the oil by the hot vapor. This deposition is facilitated if the vessel MACHINERY FOR REFRIGERATION. 93 is kept comparatively cool, which is difficult to do if it is too small. In some cases the outlet and inlet pipes to a separator are simply placed vertically through the top cover without anything- to baffle or arrest the oil suspended in the hot vapor, and in others wire screens are introduced, as in Fig-. 49. Opinions appear to differ greatly as to what is the best arrang-ement and proportion of parts for effectively keeping- oil out of the condenser. LIQUID AMMONIA RECKIVER. Two separate forms of vessels for containing- the liquid ammonia are shown by Figs. 50 and 51. This vessel is always placed below the condenser, and from it the supply pipe is led to the refrig-erator, which is regulated by the expansion cock or valve, which is sometimes called the "flashing" or flash valve, the name no doubt suggested by the idea of liquid flashing into vapor as its pressure is removed when it passes into the refrigerator. INTERCEPTOR OR TRAP. Another vessel, to act as an interceptor or trap, is often placed on the expansion or low pressure side of the refriger- ator, near to the inlet to the compressor, in order to inter- cept any foreign matter such as scale or dirt that may accumulate in or be carried from the pipes, and prevent the same from entering the cylinder of the compressor, where it might injure the piston or valves. All these vessels may be made and jointed in many ways so long as they are absolutely gas tight, but the general preference is for wrought iron or steel bodies welded up at one or both ends. 94 MACHINERY FOR REFRIGERATION. CHAPTER XV. THE STEAM ENGINE AND THE COMPRESSOR -THEIR FUNCTIONS CONTRASTED. In a steam engine high efficiency demands the produc- tion of a given power with the minimum weight of steam supplied from the boiler, but with a refrigerating plant high efficiency means passing the maximum weight of gas through the cylinder of the compressor with a given expenditure of power. Again, a steam boiler shows its efficiency by the evaporation of the maximum weight of water per pound of fuel burnt, while the efficiency of a refrigerator boiler or vaporizer is measured by the evaporation of the minimum weight of the liquid medium per unit of heat abstracted. In the steam engine and the refrigerating machine the work done for a given expenditure of power is largely modi- fied, and the efficiencies of both are discounted by dispropor- tion of parts, clearance, leakage and friction ; thus, while the theories which are involved in the compression and expan- sion of gases and vapors are the same for everybody, yet the practical results attained with compressors, as with steam engines, differ widely, in accordance with the design and construction of the machines by their respective makers. THE MECHANICAL OPERATION OF COMPRESSING A GAS. In compressing any gas the design and construction of the compressor cylinder with its piston and valves is of very first importance, as they are the primary instruments con- cerned. The shafts, cranks, connecting rods, fly-wheels, steam cylinders or other portions of the prime movers which supply the power to the piston of such a cylinder, occupy, as accessories, a secondary though important part. Almost any form of compressing cylinder, good, bad or indifferent in MACHINERY FOR REFRIGERATION. 95 design or construction, may have its piston driven by almost any mechanical arrangement of cranks, rods or levers, also either ill or well designed, and may also receive its motion from steam, water or any other power, economical or waste- ful, without at all affecting its quality or efficiency as a com- pressor. It is therefore desirable in instituting a comparison between different types of refrigerating machinery to classify their various functions, so that they may be sep- arately and properly compared, and the following appears to be a convenient division to adopt in considering the questions involved: Firstly. The construction of the compression pump itself, with its pistons and valves, and its efficiency for the work it has to do. Secondly. The connection between the motor piston of the engine and the driven piston of the compressor as affect- ing the simplicity and efficiency of the transfer of power from one to the other, and the first cost of the whole machine. Thirdly. The provision for minimizing wear and tear, reducing cost of maintenance, and simplifying access to working parts for inspection and repair. THE QUALITIES THAT ARE DESIRABLE, OR THE CONDITIONS THAT SHOULD BE FULFILLED, IN AN IDEAL COMPRESSION MACHINE. Under the first head just referred to may be placed the following characteristics, which are directly concerned with the work done on the gas: 1. On the in, or suction stroke, the cylinder should fill with gas at a pressure as little below that in the expansion coils as possible, and the outlet valve should be tight. 2. The piston and its rod should work with the maxi- mum of tightness in order to prevent leakage, and with the minimum of friction, which (as it generates heat and requires extra power to overcome it) involves a two-fold loss. 3. On the out-stroke the inlet valve should not permit any leakage back, and the whole contents of the cylinder, less the minimum of clearance, should be discharged through the outlet valve at a pressure as little above that in the con- denser as possible. 96 MACHINERY FOR REFRIGERATION. Under the second head: Dealing- with the general design and construction of the whole, and noting that the very mas- sive foundations which are required by some compressors and their steam engines must be taken into account when comparing the cost of the same in working order 4. The machine other things being equal should be self-contained on one sole plate so as to be easily and cheaply erected on the minimum of necessary foundations. Seeing that with single and double-acting cylinders of equal capacity and piston speed, single-acting machines must have double the piston area of double-acting ones, and there- fore transmit double the stress to the connecting rods and cranks, then 5. The work of the compressor with its crank, rods and crossheads should be double-acting instead of single- acting, and the ratio of compression should be as small as possible during both strokes, in order to distribute the work over as large a portion of the crank pin's path as possible. If it is required to minimize the strain on the crank pins, shafts and connecting rods, and keep down the weight, cost, friction and wear of those parts, and high mechanical efficiency with low working expenses are aimed at, then 6. In order to minimize the friction in the bearings and prevent the loss of power which results from indirect action the connection of the engine piston to the compressor piston should be as direct as possible, and the crank shaft with the crank pins and connecting rods should only be required to take up and transmit the difference between the power exerted by the steam and that required by compressor pis- tons, respectively, at any given position, instead of having to carry the work and friction due to the sum of those powers. 7. The pistons and valves should be easily accessible for examination and renewal. Under the third head, and connected with the mainte- nance of the whole of machine in working order 8. All covers or bonnets should be made with a simple joint, and to insure perfect absence of leakage, such things as double or treble connections, with bridges under one joint face, should be avoided. MACHINERY FOR REFRIGERATION. 97 Lastly, all wearing- surfaces should be adjustable and easily adjusted. THE RESISTANCE TO A COMPRESSOR PISTON IS NOT UNIFORM THROUGHOUT THE WHOLE STROKE. The curves in Fig-. 5 show how the pressure in a cylinder increases as air or gas is compressed and its volume reduced. Leaving- for the present -the question of the difference be- tween adiabatic and isothermal lines, it may be assumed that in practice, the actual curve of compression is always some- where between the two, and that such curve can be ascer- tained at any time when a compressor is fitted with a suitable indicator. This instrument takes a diagram which shows the work done by the piston of a compressor, just as a dia- gram from a steam cylinder shows the work done on the piston of an engine. An engine piston commences its stroke with the maximum pressure acting upon it, which continues until the steam is shut off, when the force or power of the same diminishes by the ratio of expansion to the end of its travel; but the piston of a compressor commences its stroke with the minimum of resistance, or without having- any resistance to meet at all apart from friction, because the g-as is then, or should be, of equal pressure on both sides of it. The resistance to the piston, however, commences with its movement, and the pressure of the g-as in front rises until the condenser pressure is reached, and then it continues uniform as it passes the outlet valve to the end of the stroke. It is not all expelled, however, in practice, because a certain amount, more or less, is left in the space between the piston and cylinder head, called the "clearance." Now this question of clearance has been the bete-noir or bug-bear of g-enerations of compressor builders, and its importance is sometimes forcibly broug-ht home to machine men when they see a cylinder head fly clear of the studs through having- too little clearance. In other cases a very small effective result is obtained through the machine having- too much clearance. It is easy to understand that as the ratio of compres- sion becomes greater, so much the shorter is the latter part of the stroke during- which actual delivery of gas takes place; (7) MACHINERY FOR REFRIGERATION. MACHINERY FOR REFRIGERATION. 99 and, therefore, the greater the ratio of compression the greater is the loss with a given amount of clearance. Fig. 53 shows five diagrams of a compressor, each one with the piston in a different position. In the first one the pis- ton is at the bottom and before com pression commences, and the cylinder is supposed to be full at normal pressure; the others show the respective positions at wh ich the piston arrives before FlG. 54. COMPOUND TANDEM ENGINE AND COMPRESSOR. Designed by the author in 1881. the gas is compressed into one-half, one fourth, one-sixth or one-eighth of its original volume; or, if it is an air compressor, then to two, four, six or eight atmospheres respectively. (The effect of the heat of compression is omitted in all these cases.) The whole of the parallelogram between the piston and cylinder head in each instance represents the volume 100 MACHINERY FOR REFRIGERATION. FlG. 55. PLAN OF 1881 MACHINE BY THE AUTHOR. FlG. 56. SECTION OF CASE COMPRESSOR, BUFFALO, N. Y. MACHINERY FOR REFRIGERATION. 101 at the increased pressure which would be delivered through the outlet valve if the piston was to strike the cylinder head at the end of the stroke. The space between the head and the upper dotted line represents an amount of clearance equal in all cases. The space between the cylinder head and the lower dotted line represents the volume to which the enclosed gas would re-expand and the line to which the piston would return before the cylinder could commence to refill on the return stroke. If this clearance is as much as one-eighth of an inch, then the waste or lost spaces would be one-quarter, one-half, three-quarters and one inch respectively, which would be deducted from the effective FlG. 57. SECTION OF WESTINGHOUSE ENCLOSED COMPRESSOR. stroke in the several cases. This shows that there would be a very large percentage of loss with high ratios of compres- sion that would be intensified with short-stroke pistons. This elementary explanation is no doubt unnecessary to many readers, but it paves the way for the proper considera- tion of the design and construction of compressors as actu- ally built, and of their methods of meeting the conditions required for high efficiency. SOME METHODS ADOPTED IN THE CONSTRUCTION OF REFRIGERAT- ING COMPRESSORS TO MEET THE FOREGOING CONDITIONS. In Figs. 54 to 70 there will be found sections of a number of compressor cylinders including well known and widely 102 MACHINERY FOR REFRIGERATION, FlG. 58. SECTION OF ANTARCTIC SINGLK-ACTING COMPRESSOR, Designed by the author. MACHINERY FOR REFRIGERATION. 103 different types. An examination into their construction will enable us to see how they secure the several requirements which have been considered important in previous chapters. FIRST. The cv Under should Jill with gas as little below the pressure in the expansion coils as possible, or, in other -words, exhaust the maximum -weight from the refrigerator. Fig's. 44 and 56 are sections of two double-acting" com- pressors, the former working- horizontally and the other ver- tically, but in both cases the inlet and delivery valves are placed with their axes lying- horizontal. Such valves will of course not close by gravity alone. Fig-. 57 represents a dif- ferent type of compressor with two single-act ing- horizontal cylinders, and it also has horizontal valves. As such valves have no tendency to close by themselves, they require strong- spring-s to insure their action being- prompt and decisive, and therefore their cylinders never can fill to the full back pressure, because it is evident that during- the admission of the gas there must always be a sufficient difference between the inside and outside pressure to overpower the resistance of the springs and open the inlet valves. Figs. 54 and 55 show a compressor desig-ned by the author some years ag-o with spherical ends to the cylinder and piston, so as to provide a larg-er area for the inlet and outlet valves; this is similar to the arrang-ement adopted in the well known Linde machines, Figs. 44 and 71, and seems to be the best possible arrang-ement for ordinary horizontal compressors with horizontal valves. Neither of these, how- ever, can provide a perfectly free inlet for gas. In Fig-. 58, a desig-n by the author (Sydney), Fig-. 59, the Hercules (American), and Fig-. 60, the Auldjo (Australian), all single-acting- vertical compressors, it will be seen that special devices are in all cases provided whereby free com- munication is established between the inlet branch from their refrigerators and the interior of the cylinders when their pistons are right down. In Fig*. 61 Antarctic compound a similar arrangement is shown in the primary or low r pressure cylinder. In several of these compressors the pistons when on the bottom center uncover the ports shown, which open right through their cylinder walls, and in the case of the Auldjo 104 MACHINERY FOR REFRIGERATION. FlG. 59. SECTION OF HERCULES COMPRESSOR. FlG. 60. AULDJO COMPRESSOR. FlG. 61. ANTARCTIC COMPRESSOR. MACHINERY FOR REFRIGERATION. 105 machine the piston passes the end of flutes or grooves cut in the walls of the cylinder. All of these cylinders can therefore fill with gas without any restriction, because an DE LA VERGNE COMPRESSORS. FlG. 62. SINGLE-ACTING. FlG. 63. DOUBLE-ACTING. equilibrium is insured between the two sides of their pis- tons, whatever the pressure on the spring's of the inlet valves may be. This idea, borrowed no doubt from the old fash- ioned air g-un pumps, is supplemented in the Auldjo com- 106 MACHINERY FOR REFRIGERATION. pressor by an arrangement for opening- the inlet valve auto- matically; this is effected by having- the piston itself loose on the piston rod, and the valve itself fast on the rod in such a way that it opens on the down and closes on the up stroke. This makes a double (and what would almost appear to be an unnecessary) provision for securing- a full cylinder of g"as. FlG. 64. SECTION OF FKICK CO. \S COMPRESSOR. In the cryog-en machines small ammonia dairy refrigera- tors, made in Queensland and other small compressors there are no inlet valves at all, and the inlet is entirely pro- vided for by the piston passing- the end of grooves machined in the bottom part of the cylinder, as in the Auldjo com- pressor. In the Hercules machine there is a belt or passag-e MACHINERY FOR REFRIGERATION. 107 cast around the bottom of the cylinder which is in connection with the inlet branch, and into this belt holes are cored (not bored) through the walls of the barrel. Some of these holes FIG. 65. SECTION OK "CONSOLIDATED" COMPRESSOR. are above and some are below the piston when it is down, and the gas has thus free access quite apart from the valves 108 MACHINERY FOR REFRIGERATION. before the return stroke. This arrangement involves a rather complicated cylinder casting-, but the holes compen- sate for the necessarily restricted size of the inlet valve and secure the full back pressure of gas above the piston before compression is commenced. In the compressors, Figs. 58 and 61, similar holes for admitting gas are provided, but instead of being cored, as in FlG. 66. SECTION OF YORK CO.'S COMPOUND COMPRESSOR. the previous case, they are drilled from the outside. This is an easy process with these machines because the working cylinders in both cases are made as plain barrel castings. In the widely used De La Vergne compressors, Figs. 62 and 63, one of which is single-acting and the other double- MACHINERY FOR REFRIGERATION. 109 acting-, the weight of oil would appear to affect the free admis- sion of the gas, and the small valves in the double-acting piston of Fig. 63 probably reduce somewhat the effective pressure in the cylinder. As these machines run at a com- paratively low piston speed, however, the actual loss may not be so serious as would otherwise be the case. A broad contrast to the last example is seen in the Frick or "Eclipse "compressor,Fig. 64, which has the inlet valve in the piston made so large and so nicely balanced on springs, that when it has completed its down stroke there can be scarcely any difference between the pressure in the cylinder above and below the piston, and thus the filling of its cylinder is insured. In Fig. 65, the "Consolidated" compressor, and Fig. 66, the York compound compressor, all the gas has to be drawn in through the suction valves, which have to share the space on the heads of their cylinders along with the delivery valves, and are thus restricted as to size. Looking at all these details and comparing their relative effects, it may be said that the first condition is more perfectly met (although by different methods) in such machines as the Auldjo, Antarctic, Hercules and Frick. SECONDLY. The piston and rod should work gas-tight with the minimum of friction. With compressors, such as are shown by Figs. 62 and 63, the oil in the bottom of the cylinders must prevent any gas from escaping through the piston rod packing, although in the double-acting one it is subjected to the full forward pres- sure of the gas. The oil used in both these cases is intended to be carried right through the system very rapidly, in order to take up some of the heat of compression, and it is supplied at every stroke by means of a special pump. There is no need therefore for heavy packing and great friction in these machines. It is claimed for the Frick machine, Fig. 64, that specially good workmanship enables oil to be dispensed with altogether for lubricating the piston, except so far as it is carried in by the rod, and it is used only in a lantern bush, which is interposed between two separate packings in the stuffing-box, where it is forced in by a hand-pump. There must, however, be extra friction here, due to the exces- 110 MACHINERY FOR REFRIGERATION. sive length of the two packing's, and as a matter of fact this lantern bush is not at all necessary for single-acting vertical types of compressors with accurate workmanship in the boring and turning of stuffing-box, glands and piston rods, while with the double-acting horizontal compressors, such as the Linde and those shown by Figs. 44 and 54, they are almost indispensable. A pump driven by the engine is used in the Linde ma- chines to eject the oil continuously between the two packings to prevent the escape of gas, and some of this is carried into the cylinder at every stroke. This of course does not apply to such Linde machines as are constructed with a lubricator on the stuffing-box instead of a pump. It will be noticed that in Fig. 58 the oil to seal the rod lies well belowthe inlet passage, and there is thus no tendency for the flow of gas to carry it up in quantity through the valve in the piston. In the com- pound compressor, Fig. 61, it will be further noticed that there are no piston rods proper passing into the cylinders at all, and that a depth of several inches of oil can lie in the bottom of the casing around the rods. In this case the tendency of the oil to pass through the system is minimized while full lubrication and sealing of the rods is secured. In order that the piston of a compressor should work g-as-tight, and yet with the least amount of friction and wear, it is imperative that the metal in the cylinder should be of a very hard and uniform texture. In order to better secure these qualities it is desirable that the cylinder itself should be made as a simple barrel or as plain a casting as possible. Any complication of cores, passages, flanges or projections upon a cylinder casting has a tendency to cause the metal to " draw " or become spongy, and make it very difficult to produce a sound, solid casting from specially hard iron. What is still worse perhaps is that an irregular casting has a tendency to alter its shape with every change of temperature, and as a compressor cylinder is subject to more changes of temperature than a steam cylinder, the desirability of having a casting that will be cylindrical at all temperatures and which will expand and contract equally all over is very evi- dent. This characteristic is most strongly shown in Figs. 57, MACHINERY FOR REFRIGERATION. m 58 and 61, where the working- cylinders are either separate bushes or quite plain barrels, and also in the Frick com- pressor, Fig-. 64, where the working- portion of the cylinder is quite plain. It is in a less degree in the Linde cylinder, which is g-enerally made with the feet cast on. The most complicated cylinders to cast, owing- to cores and passag-es, are probably the De La Verg-ne, Fig-. 63, and the Hercules, Fig-. 59, where the designs are such as to require great skill on the part of the molder to obtain sound and homo- geneous castings which will wear uniformly all over. It will be noted that Fig-. 61 represents a double-acting- compressor in which the piston rod and its packing- are never subjected to the forward pressure of the gas. It must be within the knowledge of every one accus- tomed to compressors that cylinders often want reboring after a single season's work, and that pistons sometimes leak after being started only a few weeks, even if they were tight at first. The power of the engine has probably been employed to wear out the machine through undue friction. The remedy for this is to have cylinders made as plain cast- ings of hard, homogeneous metal, accurately bored and lapped, and pistons that will work satisfactorily even if there are no rings in their grooves. Piston rings are extremely useful and necessary adjuncts, but, as often made, with a very strong spring to atone for a bad fitting piston, they are sim- ply devices to wear out the cylinder and make the fit worse. An inspection and comparison of the several sections will show which are the types most likely to secure hard and absolutely sound castings. THIRDLY. The ivhole contents of the cylinder, less the minimum deduction for clearance, should be discharged at the minimum of pressure. In the machines shown in section by Figs. 62 and 63 the presence of oil insures the full expulsion of the gas. In those shown in Figs. 58, 60 and 64, with movable heads to their cylinders, the pistons on the up-stroke may be so adjusted as absolutely to touch them, and thus the clearance is mini- mized in these cylinders. If smaller pilot valves are placed in the center of these valvular heads, then the compressor cylinders may be made as large as desired. In types, such 112 MACHINERY FOR REFRIGERATION, as those shown by Fig's. 59 and 65, however, the size of the outlet valves is necessarily restricted, because there are two valves both the inlet and the outlet made in the one cover. These machines follow very closely in this feature the design ^^-^^^ FlG. 67. HIGH PRESSURE CYLINDER OF COMPOUND COMPRESSOR. of some of the ether compressors of thirty-five years ag-o, and owing- to such restriction in the delivery orifice require more clearance than is necessary for safety with larg-er out- let valves. The contracted size of the valves also increases the pressure to be overcome and reduces the piston speed. MACHINERY FOR REFRIGERATION. H3 These remarks apply in a modified way to the compressors shown in Figs. 54, 57 and 66. The pros and cons of oil injection have been the subject of several interesting- wordy wars which it is not necessary to touch upon here. Whatever he may once have thought of it, the writer does not now believe in the system. Apart, however, from the question whether the oil used in some of them absorbs and again gives out gas in their cylinders, the De La Vergne, Frick, Auldjo, Antarctic and others of that type are certainly the best fitted of all that have been so far illustrated for fulfilling this third function of fully expelling all the gas at the end of the stroke. The amount of efficiency lost by a given amount of clearance in a compressor has already been shown to be dependent upon the ratio of compression carried out. Thus one-sixteenth of an inch clearance with a two-fold compression would not cause so large a percentage of loss as one-thirty-second of an inch clearance with a five-fold com- pression in the same cylinder. It follows from this that when compression is carried out in stages, as in the " Lock " or St. Clair system, as in Fig. 66, or by the Antarctic system, as in Fig. 61, it is possible to get a very full discharge with- out a minimum of clearance ; for let us suppose in a com- pound machine the high pressure cylinder to be only one- third of the area that a single compression one would require to be, then a given clearance in the same stroke would only waste one-third the volume otherwise lost. Fig. 67 is the back end of the high pressure cylinder of a compound com- pressor made for the author in 1884. It will be noted that, although the compressor is horizontal, the valves are vertical. Although the clearance is relatively large in this design, it is but of small comparative importance, as the ratio of second compression is only about 2:1. In order to still further secure the maximum efficiency in preventing leakage past the pistons, the builders of some high class machines not only bore out their cylinders, but they lap them out afterward perfectly true, and then grind in their pistons. This is due to an advanced idea that the ordinary wear and leakage of cylinders and pistons is almost entirely due to defective material and workmanship, and that 114 MACHINERY FOR REFRIGERATION, the ideal piston that would never leak is the one that fits the cylinder so loosely as not to touch it, and yet so closely as not to permit the passage of gas. This is of course a ques- tion of workmanship; we know that a Whitworth gauge can be made so true that it cannot be passed through its collar with oil on it as there is no room for oil but will drop easily through it when dry polished with a silk handkerchief. In such a case there is evidence of good work. In the compe- tition for business and the demand for cheap machinery of FlG. 68. SINGLE-ACTING COMPRESSOR. Patented in 1880 by the author. FlG. 69 COMPOUND COMPRESSOR. Patented in 1880 by the author. all kinds such high class work is perhaps not common in the construction of refrigerating compressors, and as a matter of fact, the best surfaces of ordinary piston and cylinder walls as they are left by the turning tools are like the ridges and furrows of a plowed field on a small scale, and they are often not so microscopic as to want more than an ordinary eye or finger to detect their inequalities. It is quite certain that a piston may be a very tight fit in a cylinder one day and MACHINERY FOR REFRIGERATION. 115 yet work easily enough to rattle about shortly afterward when the tops of the hills have been worn off the two metallic surfaces. The author is an advocate for a true cylinder that will be equally true whether it is hot or cold, however much it may expand, and a piston which fits it and has such a thick- ness of metal as to heat and expand equally with the cylinder, and he does not like strong" spring- piston ring's of hard steel, which are continually destroying- g-ood cylinders. It is better FlG. 70. YORK CO. 'S COMPOUND COMPRESSOR AND ENGINE. to get a new piston than to spoil your cylinder with hard rings. To those who have never before seen an ammonia cylinder lapped out after being bored, it will come as a reve- lation when they first see it done and realize how imperfect is the surface of the ordinary cylinder that is turned out by the best lathe or boring mill alone. Figs. 68 and 69 represent two designs, one of which is for a single-acting and the other for a compound ammonia 116 MACHINERY FOR REFRIGERATION. FlG. 71. PLAN OF LINDE COMPRESSOR AND ENGINE. FlG. 72. ELEVATION DIAGRAM OF HERCULES MACHINE FlG. 73. PLAN OF HERCULES MACHINE AND STEAM ENGINE. MACHINERY FOR REFRIGERATION. H7 compressor, which were patented by the author as far back as 1880. It will be noted that they both work with valves in their pistons, and that provision is made in both of them to insure that the piston rod packing- is only subjected to the back pressure an arrangement which has since come into very general use. Before leaving- the subject of compound compressing cylinders for the present, it will be well to note that, althoug-h it is not important in small machines and plants, yet with compound compressors of larg-e size it is desirable to pass the g-as throug-h a condenser between the two stag-es of com- pression in order to remove some of the heat, reduce the vol- ume and save power. This is done with the York compressor, shown in section by Fig-. 66, where the connecting- pipes are seen at the top, and by Fig-. 70, illustrating- the machine com- plete. The larg-e compound compressor made for the author in 1884 of which Fig-. 67 is part section of the h. p. cylin- der had a tubular condenser interposed between the two stag-es of compression, and it gave most excellent results, the diagrams showing- nearly isothermal lines. In the low pres- sure cylinder of Fig-. 66 it will be noted that the valve in the piston is annular, and thus it requires only one-half the lift of an ordinary mitre valve to give the same area of discharg-e. In the Auldjo compressor, Fig-. 60, owing to the valve in the piston being fast on the piston rod itself, and the piston being loose on the rod, the amount of its opening will have to be deducted from the nominal stroke, to arrive at the effective length, because the actual stroke of the piston will be so much less than the stroke of its rod. It would seem at first sight a self-evident proposition that a compressor and its steam engine should be combined as one machine. But as a matter of fact such well known and largely used types of refrigerating plants as the "Linde" and " Hercules " are built up from the two machines made separately (numbers have come to Australia with their engine and compressor built by two entirely different makers) see Figs. 71 to 73 and in such case they of course require double foundations and extra careful erection. There must be some reason therefore for this separation which should repay our investigation, to do which we must 118 MACHINERY FOR REFRIGERATION. go back a little, and again consider the work to be done by the piston of a compressor in its relation to the work of a steam engine. THE WORK TO BE DONE BY A COMPRESSOR PISTON. Fig-. 74 represents a diagram or indicator card as taken from an ammonia compressor by an eminent firm of refrig- erating machine builders, who claim to obtain nearly isother- mal compression under their system of injecting oil at every stroke of the machine. From this diagram it will be seen that there is no effective work performed by the piston at the commencement of its stroke when the pressure on both FlG. 74. INDICATOR CARD FROM AMMONIA COMPRESSOR. sides is the same; at quarter stroke the pressure against it is equivalent to about ten pounds per square inch, at half stroke about thirty pounds, at three-quarters stroke, 100 pounds, and the maximum pressure, about 180 pounds, is reached at about five-sixths of the stroke, when the delivery valve opens, and the pressure thence continues uniform during expulsion to the end of the piston's journey. A diagram of the work performed against the piston of an expansive steam engine is, of course, just the reverse of such a compressor diagram, because in the engine the maxi- mum work is at the commencement of the stroke, whence it continues practically uniform until the steam is cut off, and MACHINERY FOR REFRIGERATION. 119 then it diminishes gradually toward the end, in accordance with the grade of expansion at which the steam is worked. Air compressors, such as Fig. 75, do not usually work to as high ratios of compression as ammonia machines do, and FlG. 75. STRAIGHT LINE ENGINE AND COMPRESSOR. most builders of them stick fast to this "straight-line" system. Fig. 76 represents two indicator diagrams taken one each from the engine and compressor cylinders of the same straight-line machine and superimposed. By this it is FlG. 76. INDICATOR CARDS FROM STEAM AND AIR CYLINDERS. shown clearly how unequal is the relative effort and resist- ance at different parts of the stroke. The small portion covered by crossed lines represents the whole portion of the work which is transferred direct from the piston of the 120 MACHINERY FOR REFRIGERATION. engine to that of the compressor, although the engine appears to have a slide valve and carries the ste.am well past half stroke. These discrepancies are greatly intensified if we take the compressor card, Fig. 74, and superimpose upon it the card from a Corliss engine, cutting off at from one-fifth to one-quarter stroke, as in Fig. 77. ABCDEAisa diagram from a Corliss steam cylin- der hatched with horizontal lines, and F G H D F is the diagram from the compressor just referred to hatched with vertical lines. That part of the figure which is covered by FlG. 77. INDICATOR CARDS, CORLISS ENGINE AND COMPRESSOR. the intersected lines represents the very small portion of the whole work that would be communicated directly from the piston of the engine to the piston of the compressor if the two were coupled up in a straight line; that proportion of the work which is shown by plain horizontal lines would have to be delivered into the fly-wheel at the early part of the stroke, and the work represented by the area covered by plain verti- cal lines would have to be given up again by the fly-wheel to the compressor piston at the latter end of the stroke. All these points have to be considered before we can properly investigate the construction of the whole machine, steam engine and ammonia compressor combined. MACHINERY FOR REFRIGERATION. 131 GENERAL DESIGN AND CONSTRUCTION OF THE WHOLE MACHINE. FOURTHLY. Other things being equal, the machine should be all self-contained, be easily erected, and require the minimum of foundations. FIFTHLY. As double-acting compressors require only one- half of the stress on their connecting rods, bearings and cranks, that is necessary -with single-acting ones of equal capacity and stroke to do the same amount of work, then all compressors should be double-acting, unless there are insuperable disadvan- tages connected with such an arrangement. And, as the smaller the ratio of compression, the more equable is the work of the piston, there may be manifest advantages in compressing b\ stages a few ratios at a time, if not accompanied by increased complication and cost in other directions. The writer, from a life's experience of machine builders and machinery users, is inclined to the belief that purchasers often think that they can do without old-fashioned advice, and imagine that they are keen buyers, when they make a saving- by paying- a few dollars or pounds less for one machine than they are asked for another, which to their ideas is a similar one. Such people often find out afterward that they made a mistake, because they did not sufficiently value some old-timer's experience, or take into consideration what the relative cost erected complete and upon their foundations, ready for work, of the different machines offered to them would come to, and understand that annual up-keep and wear and tear are important factors. SIXTHLY. The engine piston should be connected directly to the piston of the compressor, and the cranks, connecting rods and bearing's of the machine should only transmit the DIFFER- ENCE between the engine force and the compressor resistance instead of the SUM of the work represented by the two. This would appear as a self-evident proposition to be universally followed, as it is done in straig-ht-line com- pressors, were it not for the teaching- of preceding- para- graphs, which show how the great want of correspondence between the power of the engine and the resistance of the compressor, during- the cycle or revolution of the crank shaft, necessitates enormous fly-wheels, and increases the frictional losses. 122 MACHINERY FOR REFRIGERATION. SEVENTHLY. The pistons and valves should be easily acces- sible for examination and removal. Horizontal valves are generally easily accessible, but they want looking- to so much oftener than vertical ones, that the makers of the machine shown by Fig. 63 go to great expense to fit vertical valves into cages, which are again fitted into horizontal recesses in the cylinder; but in the same machine there must always be a little bit of a picnic if one of the piston valves gets stuck. In Fig. 61 the difficulty is only apparent and not real, as three valves can be withdrawn by removing the top cover, and the low pressure inlet valve is accessible from the bottom door. The low pressure deliv- ery valve is made so as to withdraw right through the trunk and high pressure piston. A thoughtful inspection of some of the several com- pressor cylinders illustrated reveals interesting features, which suggest a number of questions for instance: Why do makers of some compressors put their outlet pipes on to the covers or heads of their machines in such a way that neither the piston nor the valves can be got at without break- ing a great number of joints and taking down what should be permanent connections? The writer was once shipmate with a steam crane built by makers who were very eminent for certain classes of machinery, but were apparently starting a new line of "grab" cranes; well, this crane had the steam pipes screwed into the doors or bonnets of the slide valve chests, but, owing to the absence of flanges, it was necessary to take off eleven separate pieces of pipe to get to those slides. If compressors of this class are not expected to call forth expressions, at times, which are more forcible than poetic, they will have to be placed in charge of very good men in more senses than one. EIGHTHLY. Covers or bonnets should be made ivith simple joints and no bridges. This, like the seventh condition, will be best illustrated, perhaps, by instances in which the condition is not fulfilled. Fig. 57 is a section of a compressor which has a great num- ber of extremely good points as a machine, but it also has MACHINERY FOR REFRIGERATION. 123 triple face joints under the heads on the lines A B. This arrangement necessitates most accurate workmanship in the fitting", and extreme care when making* the joints at the two bridges, to insure that they do not blow through. A leak in such a case may be going on for a long-time before it is found out. The author's two compressors, Figs. 68 and 69, are sin- ners on this point, but as he is now twenty years older than he was when he committed the offense, he has lived long enough since to see the error of his ways. Fig-s. 59 and 65 show similar joints on their compressor heads; these, like the pipes direct on to the heads, are not necessary at all, unless it is desired that the man who has to make the joint in a hurry and be responsible for it afterward should become an adept in profane language. A joint is of course a relatively simple matter to make, now that flang-es are faced by hig-h-class tools, to what it was formerly. The author has made joints of curious material in his time, such as, for instance, tinfoil and blotting- paper on ether machines, well kneaded doug-h from wheat flour for cold kerosene, fire- clay and red lead for hot oil, and all such nostrums, which were the best thing's known for their respective purposes at one time, before the present great army of patent packing- people made life easier. With all these to hand, he has found nothing- better for a compressor head or any other ammonia joints, than a thin lead gasket placed in a recess where it cannot g-et away. To make a sure success, all bon- nets and flang-es should be plain circular, and turned for the ring-s. If people will make simple flat surfaces and use jointing- material which will squeeze out and g-et in the way of their valves or pistons, they must expect trouble some- times; but such old-time rough-and-ready methods are not good practice now. The jointing material, whether metallic, fiber, rubber or insertion, should be inclosed where it cannot spread. For examples of joints see the covers in Figs. 58 and 62, which show two separate ways of keeping the jointing from spreading. ENORMOUS FLY-WHEELS. It is evident from the foregoing illustrations that builders of compressors have good reasons for the employment of 124 MACHINERY FOR REFRIGERATION. the extremely heavy fly-wheels which often distinguish this class of machinery. Such wheels require heavy shafts and journals, and therefore greatly increase the friction in the bearing's and the power necessary to drive a given sized machine, and also add to the first cost and maintenance when at work. This being- well understood, there has in conse- quence been plenty of inventive skill displayed in devising compressing machinery with all sorts of arrangements to enable the work performed by the steam piston to coincide more nearly with the work required by the compressor's piston at every part of the shaft's revolution. There is a great deal of popular misconception with regard to the power wasted in driving fly-wheels, it being often stated that such power is only required at first start- ing them into motion. The actual horse power continu- ously expended is represented by the formula: . Horsepower: ff= Wherey represents the co-efficient of friction from .03 to .25 in wrought iron upon gun metal lubricated, it cannot safely be taken at less than .05 in actual continuous work; TFthe weight of fly-wheel, S the speed or revolutions per minute, and .26d the circumference of the journal in feet when d= the diameter in inches. Take for example a 5-ton fly-wheel making 90 revolu- tions per minute with 9-inch journals, then 9"X.26 = 2.34 ft. cir. of journal X 90 revs. = 210.6 feet per minute. Five long tons = 11,200 Ibs., which multiplied by .05, =560. 560 X 210 Then ^000" : = 3-3 horse power. With fuel evaporating 8 Ibs. of water per minute and an engine using 30 Ibs. of steam per horse power per hour i nc x 24 3.5X30=105, and =315 Ibs. of coal wasted every o twenty-four hours simply to drive the wheel. RIGHT-ANGLED AKKANGEMKNT OI^ ENGINE AND COMPRESSOR. In order that the continuously varying power of the engine during the course of a stroke or revolution, may be MACHINERY FOR REFRIGERATION, 125 applied in such a way as to correspond letter with the work to be done, and be more effective at the time when the com- pressor piston offers the greatest resistance to it, great numbers of refrig-erating- machines are now built in such a way that the effective axis of the steam cylinder with reg-ard to the crank shaft is at rig-ht angles to the axis and stroke of FlG. 78. SECTION OF FRICK CO. 'S ENGINE AND COMPRESSOR. the compressor, and this is generally carried put under one or the other of the following- arrangements : Under the first one, the two connecting- rods from the crossheads of the eng-ine and compressor respectively, are connected to the same crank pin, and thus transmit the power without any torsion on the shaft, as seen in Fig's. 78 (Eclipse) and 79 (De La Verg-ne), which represent American machines of the very hig-hest class, having- horizontal eng-ines and verti- cal compression cylinders. Examples of the other arrang-e- ment are shown by the Australian compressors, Fig-s. 4 .and 126 MACHINERY FOR REFRIGERATION. 54, where the steam engine is vertical and the compressor cylinder horizontal. Under the second plan the compressor is set parallel to its engine, which is often on an entirely independent founda- tion, especially when the two machines are both horizontal. Two separate cranks are provided, one for the engine and the other for the compressor, which are keyed on to the opposite ends of the fly-wheel shaft at an angle of 90 or thereabouts. See Fig-. 71 for a typical example which should FlG. 79. SECTION OF DE LA VERGNE ENGINE AND COMPRESSOR. be carefully compared with Fig-. 54, because the compressor cylinders are practically the same in the two cases. The operation of the engine on the compressor is nearly the same in both these machines; but necessarily there is in Fig-. 71, besides the torsion on the shaft, more main bearing- friction, additional first cost, and double foundations to be provided. An examination of the double-acting- compressors, Figs. 54 and 71, will show that in both cases two singie-acting" hori- zontal cylinders could be substituted for the double-acting- one, without in any way affecting- the relation of the steam MACHINERY FOR REFRIGERATION. 127 engine piston to the motion and effective power of the com- pressor pistons. HORIZONTAL ENGINE AND TWO VERTICAL COMPRESSORS. Two vertical single-acting 1 compressors operated by a horizontal engine require at least two cranks, generally set opposite to one another or at an angle of 180 C , in which case one compressor only is driven by torsion of the shaft. Three cranks, however, are often adopted, and entail a great deal of additional complication and expense, which of course the designers of the machines consider justified by compen- FlG. 80 HORIZONTAL ENGINE AND VERTICAL COMPRESSORS ELEVATION. sating advantages, and at least five different arrangements of this type are in common use, all of which have their respect- ive advocates. Fig. 80 represents the end elevation of such a machine, five different plans of which follow, some having inside and others outside fly-wheels. The advantage of a large fly- wheel is obvious, because if 5,000 pounds weight of wheel can be made as effective as one of 10,000 pounds in a smaller compass, it will only require, as has already been shown, one- half the loss of power to keep it in motion. 128 MACHINERY FOR REFRIGERATION. To have large inside fly-wheels means very large and heavy sole plates, and therefore some machine builders over- hang- their wheels at the opposite end to the engine, as in Fig. 81, occasionally extending the shaft for a fourth and outer bearing. FlC. 81. HORIZONTAL ENGINE, VERTICAL COMPRESSOR PLAN. In Fig. 81 is seen the plan adopted by a very eminent firm of builders. One crank pin, it will be noticed, is of double length, to take the big ends of the two connecting rods. The work in such machines is necessarily very severe on the middle bearing, and, although the shafts are I ---- J FlG. 82. PLAN OF FlG. 80 WITH INSIDE FLY-WHEEL. made enormously strong as compared with steam engine practice, they occasionally fail as a result of the special strains to which compressors are liable. MACHINERY FOR REFRIGERATION. 129 Fig-. 82 shows the arrangement adopted by another firm of world-wide reputation, who put a large fly-wheel between the two compressors and carry the separate portions of the sole plate on massive girders below the floor line. The solid crank, as before, carries two connecting- rods, but the outer compressor has a disc crank overhung-. If the girders, shown in dotted lines, and the bottom of the separate sole plates are accurately planed, as is no doubt the case, this arrang-ement is a much better one from a practical mechanic's point of view than the one preceding it. Fig-. 83 shows one single solid crank for the engine and two disc cranks for the two compressors and needs a very large sole plate, as the fly-wheels are inside. With large discs the weight of each compressor piston and its connect- FlG. 83. MACHINE WITH TWO INSIDE FLY-WHEELS. ing rod can be balanced separately, instead of in the fly-wheel, and much steadier and smoother running can be assured. The fault of this arrangement is that it requires four bear- ings to be kept accurately in line; as the bushes wear, the unequal wear which is nearly certain to take place, tends to throw strains upon the shaft and break the crank. For Fig. 84, the only thing to be said in its favor is that large fly-wheels can be used with a small sole plate; the downward wear of the two outer bearings, however, offers a premium for breakage of the expensive triple crank shaft. (9) 130 MACHINERY FOR REFRIGERATION. In marine engine practice it is now customary to "build up" these crank shafts, and they are frequently made in short sections with flanges. With triple or quadruple expan- sion a long- series of solid or double cranks and a line of FlG. 84. MACHINE WITH THRICE CRANKS AND OUTSIDE FLY-WHKKLS. bearing's are absolutely necessary; but there is no necessity whatever for such complication with a refrigerating com- pressor. Every experienced engineer knows the advantage of having only two bearings to a shaft, and of that shaft being FlG. 85. MACHINE WITH TWO BEARINGS AND CRANKS, ONE FLY-WHEEL. a plain one without solid cranks that require the crank pins to be the same size as the shaft itself. Fig. 85 shows an arrangement of this kind with two engines, preferably cross-over compound cylinders; there is only one fly-wheel between two bearings, and those bearings MACHINERY FOR REFRIGERATION. 131 should be sufficiently wide apart to prevent the pressure and friction upon them being" materially increased as the effect of the leverage due to the overhang- of the crank pin. The adoption of a larger diameter of the crank pins for the com- pressors is optional, but it is mechanically correct. With such an arrangement the steam engine can be made of longer stroke than the compressor by having- the two pins eccentric to one another. If space can be afforded inthe machine house to g-ive a decently wide spread, there can be no question as to the simplicity and efficiencv of this plan of machine. LJ rr FlG. 86. VERTICAL ENGINE AND TWO VERTICAL COMPRESSORS. With vertical or "inverted" engines and two vertical compressors, the adoption of three cranks is imperatively necessary to secure the right-angled action of the engine. Fig-. 86 illustrates one of the most common designs, with two disc cranks for the compressors, four bearing-s, and two fly-wheels; it would make a better job of it, perhaps, if the outer bearing-s were larger, the shaft strengthened, and the two inner bearing-s dispensed with. This type of machine may be modified by making- three solid forg-ed cranks with outside fly-wheels put on in halves, or still further varied by putting- overhung- fly-wheels, making the plan almost a counterpart of Fig. 84, and shown by Fig. 89. 132 MACHINERY FOR REFRIGERATION. This vertical pattern seems to have been first favored by the great American father of ice making- machinery, the late David Boyle, one of whose machines is shown by Fig-. 87. As a contrast to this work of only twenty years ag-o, and to illustrate by comparison the great advance made since . 87. AMMONIA COMPRESSOR ORIGINAL BOYLE PATTERN. that time, the mag-nificent machine built by his successors is shown by Fig-. 88. Fig-. 89 shows the arrang-ement with a vertical engine, modified by overhang-ing- the fly-wheels. MACHINERY FOR REFRIGERATION. 133 134 MACHINERY FOR REFRIGERATION, In all the accompanying* illustrations where the effective axis of the engine is at rig-ht angles with that of the compres- sor, the eng-ine is on its dead centers (and therefore exert- ing- no power directly from its piston) at the time when the compressor piston is just below half stroke; so that the motive power in such positions must come from the fly- wheel. A little examination will also show, that as the crank comes toward either the top or bottom centers, and, with the compressor connecting-rod, approaches the vertical position, then, the centers of the crosshead pin, the crank pin, and the shaft are coming- into line tog-ether, which constitutes a togg-le joint of the crank and connecting- rod. The action of FlG. 89. VERTICAL MACHINE OVERHUNG FLY-WHEELS. the engine on the central pin of this tog-g-le is to create a gradually increasing- force, which approaches the theoreti- cally infinite at the two compressor centers, just when the compressor pistons offer the greatest resistance. DIAGRAMS ILLUSTRATING HIGHT-ANGLIOD CONNECTION. The actual effect which is produced on the distribution of power from the piston of the eng-ine to that of the com- pressor, whether arranged in one or other of the ways shown by the several machines illustrated, is graphically and effect- ively shown by Fig-. 90, a diagram from a Corliss eng-ine and MACHINERY FOR REFRIGERATION. 135 ammonia compressor, which is merely a transposition of what is seen on Fig-. 77. In this diagram, Fig-. 90, the length of the base line rep- resents the travel of the piston, or the stroke of the machine; and the vertical heights from any points on the base to the curved lines, the relative pressure on the pistons in such positions. The compressor diagram hatched with vertical lines is identical with that on Fig-. 77, but the transposition of the varying pressures shown by the steam engine card is FlG. 90. DIAGRAMS, CORLISS ENGINE AND COMPRESSOR (TRANSPOSED). so radical, that it would not be recognized without explana- tion. The portion batched by horizontal lines, however, represents the equivalent in energy of the engine power, as on Fig-. 77, but so transferred as to correspond with the motion of the compressor crosshead instead of its own. The horizontal base line from left to right represents the stroke of the compressor piston from the bottom to the top center. The power of the engine on the compressor connect- ing- rod and piston is at its maximum at the commencement of the stroke, as the engine is then a little past half stroke; 136 MACHINERY FOR REFRIGERATION. but this power comes down to nothing- at about half com- pressor stroke, when the engine arrives on either of its own centers. It will be noted that this center point of the eng-ine is not exactly at midstroke of the compressor, but is nearer to the left side ; this is owing to the angle of the compressor's connecting rod shortening the height of its crosshead. At the right hand side of the figure the engine is again out a little past half stroke, due, as before, to the angle of its own connecting- rod, and the compressor is then on its top center. In this figure, it will be seen, nearly all the compressor's dia- gram is overlapped, and covered by the crossed lines, and FIG. 91. ENGINE AND SINGLE-ACTING COMPRESSOR. the beautiful effect of the right-ang-led connection is made very clear. Sufficient bare horizontally hatched space is left to represent the surplus power which is required to cover the frictional losses, and it is evident that, other things being equal, any arrangement in which the work to be given and taken as is shown in Fig. 90, will only require a small frac- tion of the fly-wheel power storag-e which would be necessary in a case such as is indicated in Fig. 77. The wide adoption of a machine in which a horizontal Corliss engine is combined with vertical compressors is thus seen to be fully warranted by theory as well as by the result of practical work. MACHINERY FOR REFRIGERATION. 137 In the case of small machines made for dairies, and for butchers' use, as in Fig-. 91, there is often only one com- pressor, and that single-acting-, combined with a slide valve engine. In such case one-half of the work of the engine at least must be put into the fly-wheel. FlG. 92. DIAGRAM FROM MACHINE LIKE FlG. 91. Fig. 92 shows the application of the power of the eng-ine to such a compressor; the power as before being- hatched with horizontal lines, shows the engine cutting- off at three- quarters stroke. The compressor work is covered by ver- tical lines. FlG. 93. DIAGRAMS FROM FlG. 91, WITH RIGHT-ANGLED CRANKS. Fig-. 93 gives the diagrams of the same machine's work, but with the cranks at right angles. HOW TO PLOT DIAGRAMS OF A COMPRESSOR'S WORK. As it may not be clear to every reader how the preced- ing diagrams have been constructed, and as the graphic method adopted may be used for other purposes, such as for 138 MACHINERY FOR REFRIGERATION. ascertaining the loss by friction in a complex machine, and as such a method of investigation will settle scientifically many questions, the answers to which are often only g^uessed at, the larg-e diagram, Fig. 94, is introduced to illustrate the work of a Corliss eng-ine and pair of compressors. DIAGRAM SHOWING THE RESULTANT FORCE AVAILABLE BEHIND THE PISTON or A VERTICAL COMPRESSOR WHEN CON HORIZONTAL CORLISS ENGINE FlG. 94. DIAGRAM ILLUSTRATING WORK OF CORLISS KNGINK AND TWO COMPRESSORS. Ill this diagram sixteen positions are taken in the path of the crank pin, besides the engine centers, and the top center of the compressor. To save space the connecting" rods are centered direct from the crank pin on to the piston centers. The length A B represents the stroke of the engine and C D the stroke of the compressor. To prevent confusion which would result from showing 1 the maze of lines necessary to work out the whole of the nineteen positions, only those MACHINERY FOR REFRIGERATION, 139 are given which have reference to one position (No. 3), although the same work has been done for the whole nine- teen positions of the pistons. The arrow on the crank pin circle shows that the engine runs "overhand." The compression diagram on the left side which is hatched belongs to the compressor working off the engine crank, and which compresses while such crank is passing from position 13 to the top center. The com- pressor diagram, in double line, on the right side, belongs to the other and opposite crank, and the compression there takes place while the engine crank pin is passing from the upper to the lower center. When the compressors are on their two centers one at the top and the other at the bottom the engine crosshead, owing to the angle of its connecting rod, is not at half stroke, but is considerably nearer to the crank shaft. This angle of the connecting rod causes a good deal of inequality, in the work directly available for the two compressors, and makes work for the fly-wheel if the cut-off is the same at the two ends of the engine. The compressor on the engine crank is clearly seen to get more engine power than its fellow, because the portions of the two engine cards which are covered with horizontal lines are much larger in area than the plain portions belonging to the other com- pressor. The space between the lines marked "engine half stroke," and "compressor centers," is added to one and taken from the other, by the inclination of the connecting rod. The position taken for full illustration is No. 3, where the crank is at about an angle of 45 C , and the engine piston is under full pressure, having completed one-sixth of its out- stroke. The height of the upper diagram at 3.3 in the out- stroke or back-end card, represents the pressure on the pis- ton, the area of which is assumed to be unity. This measure- ment representing the force acting against the piston is trans- ferred to a b, on the line of the piston rod, and by the con- struction of the parallelogram a b c g gives b c as the thrust on the engine connecting rod, and b g as the pres- sure on the guides. (By taking the pressure on the guides in all positions an estimate can be made of the frictional losses due to the varying angle of the connecting rods. ) The 140 MACHINERY FOR REFRIGERATION, length b c at the crosshead end of the connecting- rod is transferred to b c at the crank end, c being the center of the crank pin. By the construction of the parallelogram b c e d with d e in line with the crank centers, then the length of c e represents the amount of force or thrust on the compressor connecting rod, and c d the direct down- ward angular thrust on the main bearing ; the latter being the resultant of the separate stresses on the two parts of the crank pin. By drawing e f vertically from the point e, with c f horizontal, the length of the former line, e f gives the amount of the direct vertical force of the engine available for the work of the compressor, and c f represents the pressure of the crosshead against the compressor guides. At the position 3 on the compressor diagram, where the top end of the connecting rod is centered, a line equal in length to f e is set up as representative of the pressure or force available to move the compressor piston in that position. By drawing a similar series of parallelograms to every one of the other positions the corresponding lengths of line have been found which enable the complete diagrams to be con- structed. No allowance or deduction has been made in any of these cases for friction; but it is evident that if the co-effi- cient of friction is known, then the actual loss, and the mechanical efficiency of the whole machine, can easily be ascertained by the same method of investigation. DIAGONAL CONNECTION. It has been shown that a steam engine with a very early cut-off is specially applicable for a right-angled connection with a compressor; but a comparison of Fig. 90 with Fig. 93 makes it clear that a later cut-off is not so well fitted for the purpose, because there is a much greater proportion of power than required at the beginning of the compressor's stroke. This can be rectified by making the connection diagonally at some other angle than 90 degrees. If, as is no doubt the case, there are still many refriger- ating engineers who question the advantages that are claimed for compound compression, there are but few, and certainly none among those who have lengthened experience with compound engines, who fail to duly appreciate the MACHINERY FOR REFRIGERATION. 141 effects of compound expansion. Where the load is steady, as in a refrigerating machine, a tandem compound has many advantages over a single cylinder Corliss engine; it gives more even running with smaller fly-wheel, and requires less working- expenses to make good the wear and tear. It is well known that no form of engine has less loss by friction than a beam engine; and when the connecting rod big-end moves in the arc of a circle, with a versed sine of only an inch or two, instead of in the circle of the crank pin path, then friction on crosshead guides is reduced to a minimum. Having been the first engineer to introduce and design tandem compounds in Australia, the author may (without knowing it) be a little prejudiced in his preference for them; be this as it may, he thought some short time since that it might be possible to arrange a pair of single-acting com- pressors with a single slide valve engine either simple or compound under a new design which should by the adop- tion of levers unite all the best features of a modern machine in a simple and effective combination, in which the engine and the two compressors should be all in line with one another, and erected on one compact sole plate and foun- dation. The machine as designed, for better or worse, is shown in perspective on following page, and in sectional elevation by Fig. 95, and is now open to the free comments of machine builders and machine users, whose criticisms, however harsh, will be gladly welcomed if genuine. No machine is perfect, and this one has many points to which exception will be taken; still, it is only by the gfadual elimination of faults that any machine approaches that perfection to which it can never arrive. An inspection of the two figures will show that there is a single horizontal engine by preference for large machines a tandem compound which is made with a high foundation plate, so as to afford space in which to carry a lever, rocking beam, or bell crank, centered right under the guides. There is only one single bent crank on the shaft, but with an extra long crank pin, this crank shaft may have a fly-wheel on one or both sides of the machine, but is never subjected to tor- 142 MACHINERY FOR REFRIGERATION. sion other than that due to the work of the fly-wheel, which, as will be seen later on, is extremely small. The interven- PERSPECTIVE VIEW ANTARCTIC REFRIGERATING MACHINE BEAM PATTERN TEN TON. tion of the rocking- levers at different angles to the main center, and the different angles of the two connecting- rods MACHINERY FOR REFRIGERATION. 143 with regard to the crank pin, puts the steam piston so far behind that of the compressor, that when the engine is at half stroke the compressor piston has completed six-sev- enths of its journey. The combination of the connecting- rods and cranks forms a most effective toggle joint, and they operate on both compressors without any torsion on the shaft. As the strains are all on the one center line, and the FlG. 95. SECTION OF BEAM PATTERN ANTARCTIC COMPRESSOR. machine is self-contained, hardly any foundation is neces- sary. As the connecting links to the compressor crossheads hardly move an inch out of the compressor's vertical line, the friction of the guides is only nominal. The adjustment of clearance is easily effected by lining under the bushes at the ends of the main lever, thus dispensing with the nuisance of screws and nuts. The piston rod and connect- 144 MACHINERY FOR REFRIGERATION, ing" rod of the engine work between the compressor links. The compressor cylinders themselves are shown on a larger scale in Fig-. 58, and it will be noted that they are plain bar- rel casting's, the belts or chambers around the bottom ends being 1 cast separate with the stuffing-boxes. The resolution of the forces in this machine has been worked out on the same principle as those shown by Fig-. 94, and the diagram produced is given by Fig-. 96. In this the same compressor diagram is used as before, but it will be FlG. 96. DIAGRAMS OF ENGINE AND COMPRESSOR FROM BEAM MACHINE. noted that the point where the engine is on the center, and where there is no power to be given off to the compressor, is much nearer to the commencement of the compressor's stroke than in the right-angled arrangement. The engine power in this diagram is drawn a little excessive, perhaps, for comparison with the work of the compressor, but it will be noted that the compressor card is all but entirely covered by the horizontal lines. This design can be modified by putting a high and low pressure steam cylinder side-by-side instead of in tandem, MACHINERY FOR REFRIGERATION. 145 and for very large machines the levers would obviously be well below the floor line. GEARED COMPRESSORS. Some English builders of refrigerating- machines favor the use of gears, and build double-acting horizontal compres- sors driven by means of spur gear from a horizontal engine running at a higher speed. The saving by this arrange- ment, if any, arises from being able to use a small steam engine, running at a greater number of revolutions per minute than the compressor. Against this has to be set all FlG. 97. HORIZONTAL COMPOUND CONDENSING ENGINE GEARED TO HORIZONTAL COMPRESSOR. the complication of extra shafting, and the noise and friction of the gearing. It is extremely doubtful whether this form of compressor can show a lower consumption of steam for the same weight of ammonia compressed than the best directly driven machines. Such an arrangement must take up an immense floor space, as seen by Fig. 97, and for obvious reasons it is not likely to have many imitators, the more so, as later developments in machine design permit of a much higher piston speed for compressors than was pos- 146 MACHINERY FOR REFRIGERATION, sible in old forms, which are restricted in their delivery through having- both the inlet and outlet valves in the top covers or heads of the cylinders. In other arrangements horizontal compressors are geared to a vertical engine, and vertical compressors to horizontal engines, but they appear to be principally confined to English practice. In a large London brewery there are three pairs of compressors set all in a line, each pair having a mortise-wheel gearing into an iron pinion on the main driv- ing", or extended engine, shaft. As the compressors are of the old fashioned type with two valves in the head, the maxi- mum speed is fifty-five revolutions, and the gearing" is as 2 to 1. As the stroke is only fifteen inches there is no doubt that with more modern valve arrangements these compressors could be driven direct from the engine. It must not be forgotten, however, that there is an advantage in being" able to put one, two or three pairs to work as the demand for cold arises, and that the risk from break-down of a compressor is minimized if your engine is never to be laid up. With machines from experienced builders there does not seem to be any reason why the compressors should not be as reliable as the engine. BELTED COMPRESSORS. No account of refrigerating machines would be complete without a chapter on belted compressors, for while sepa- rately they may be of comparative insignificance when com- pared with the giant steam machines, running up to as high as 500 tons capacity, they are in the aggregate of immense importance, owing to their more widely extended use. The development of modern creameries and dairies, with their steam driven separators and churns, has necessitated in the majority of instances, the addition of a refrigerating machine of proportionate power. In the case of advanced retail butchers who employ steam choppers and other machines, and who, like the dairy men, need refrigerators to keep pace with the times, the line shafting generally fitted up on the premises enables a small refrigerator to be simply driven by means of pulleys and belts. To meet the demand which has thus sprung up, there has been a great increase in the MACHINERY FOR REFRIGERATION. 147 designs for small plants, and their makers now may be reckoned by hundreds. Belt driven compressors may be broadly classified under two divisions, namely, the "open "and "inclosed." About the former class very little need be said, as any of the types of compressing- cylinders already referred to may be fitted up with pulleys on their crank shafts, instead of having- a steam eng-ine directly coupled to the same crank or a sepa- rate one. Such machines need to differ in no other way from an ordinary steam driven compressor, but it is obvious that with only one single-acting- compressing- cylinder, a very heavy fly-wheel is necessary, because in such case the work is all concentrated in about the sixth part of a revolution. The work of the piston shown by the indicator card from the compressor, when transformed by the action of the connect- ing- rod and crank, and bent round in the circle of the crank pin, would appear somewhat as Fig-. 98, where the radial lines in the lower diagram correspond to the work of the indicator diagram above, a rectangle equal to the distance between the two outer circles, D E, multiplied by the length of the inner or crank pin circle, representing- the mean work of the belt. The uncrossed circular lines cover the area representing the work which has to be put into the fly-wheel, while the uncrossed radial lines show the work that must be taken from the fly-wheel, if the work of the belt is uniform. The mean leng-th of these radial lines multiplied by the leng-th of the arc of the circle they stand upon (or the area of the cam shaped fig-ure, if the circle was opened out to a straight line) is exactly equal to the area of the compressor diagram. The diameter of the crank pin path, C D, is exactly the stroke of the compressor, A B. As the power transmission capacity of a belt is uniform it is evident from the above illustration that in the absence of a fly-wheel, such a machine would require belt power to be provided about six times as great as would be necessary if there were uniform resistance. With two single-acting compressors combined, or with one that is double-acting, the working is more regular, as there are two cycles in a revolution, and appears as in Fig. 99, to which the explanation of Fig. 98 applies, it being noted that the circular lines now 148 MACHINERY FOR REFRIGERATION. cover twice as wide a space as before, representing- double the belt power, that is, the rectangle formed by the length of the crank pin circle multiplied by the distance D E. The leng-ths of the arcs on which the radial lines stand, multiplied FlG. 98. DIAGRAM SINGLE-ACTING BELTED COMPRESSOR. by the mean lengths of such lines, represent, as before, areas exactly equivalent to the area of the two compressor cards above, and the maximum resistance of the compressor piston only a little over three times, instead of six times, the mean belt power. MACHINERY FOR REFRIGERATION. 149 From these diagrams it would appear that with belt-driv- ing- a small single-acting- machine requires more fly-wheel 2 8 % FlG. 99. DIAGRAM TWO SINGLE-ACTING BELTED COMPRESSORS. than one of twice the capacity, if it is double-acting- and has double the belt power. 150 MACHINERY FOR REFRIGERATION, Fig. 100 shows a belted compressor of the open type, as designed by the author for small power, where a submerged FlG. 100. BELTED COMPRESSOR ON SUBMERGED CONDENSER. condenser is preferable. This is an extremely simple machine, although the compressor being compound, the MACHINERY FOR REFRIGERATION. 151 framework and condenser tank are in one casting-, and carry the single crank with overhung- belt pulleys. Very little or no fly-wheel is necessary with this machine on account of the equable turning- moments, which is described fully with dia- grams under the head of compound compressors. Fig-. 101 represents two similar compressors coupled tog-ether and driven by disc cranks on a straig-ht shaft; each of these is double the power of that shown by Fig-. 100. By "closed" machines, is meant all those in which the crank and connecting rods, which give motion to the com- pressor pistons, are inclosed in a chamber connected with the gas inlet, and so subjected to the back pressure. There are no piston rod packings required in such cases, and the main stuffing-box is around the crank shaft, where the pack- ing is subjected to a slow rotary wear, which is continuous and in one direction, instead of to the more rapid and recipro- cating wear of the ordinary piston rod. If the oil level is maintained above the top of the shaft in these machines, all escape of gas is prevented through the packing and glands. The Westinghouse machine, Fig. 57, is one of the finest examples of this class of machine that is built; the builders say they believe in putting their eggs in a number of baskets, and prefer small units as more economical where the work varies \vith the season. Although the jointing of the back cover of these compressors with a simple flat sur- face, which requires ports to be cut in the jointing material, has been referred to as an undesirable feature, and although horizontal valves are not so trustworthy as vertical ones, yet there are in this machine a number of points which com- mend themselves to the experienced engineer, and evidence careful thought. Among these are the plain barrels or liners to the compressor cylinders; these enable hard and homogeneous metal to be used, and permit of simple renew- als. The disposition of the centers of the cylinders above and below the center of the yoke so that when the crank shaft revolves in the proper direction the twisting strain on the yoke is practically neutralized is a sound mechanical device. There have been some bad imitations of this machine seen by the author, where the true spirit of the original was quite lost. 152 MACHINERY FOR REFRIGERATION, LONGITUDINAL SECTION - END ELEVATION ' PLAN _ FlG. 101. COMPOUND COMPRESSOR, TWO-TON ICE MAKING PLANT MACHINERY FOR REFRIGERATION. 153 The Remington machine is representative of numbers of different makers' closed compressors which only differ from one another in small details; in all these cases there are one or two open-mouthed cylinders arranged over the crank casing, to which the return gas is led. The improved Rem- ington machines are built with both suction and discharge valves in top head of the cylinder. For latest compressor see Chapter XX. Fig. 102 is a section, and Fig. 103 a perspective view of an inclosed machine, with compound compression, having FlG. 102. SECTION OF ENCLOSED ANTARCTIC COMPRESSOR. both cylinders opening directly into the crank chamber, and taking power on both the up and down strokes. The two connecting rods at their small ends are coupled direct to a crosshead on the trunk between the pistons. The details of this machine are described herein fully under compound com- pressors. In place of an internal crank shaft, some inclosed machines work by means of levers or beams inside the casing, and when the main center or vibrating axis which gives the motion is kept down in the bottom of the casing, a very small quantity of oil is sufficient to seal the shaft at the stuffing-box. 154 MACHINERY FOR REFRIGERATION. FlO. 103. ANTARCTIC COMPOUND COMPRESSOR, PERSPECTIVE OF ENCLOSED TYPE MACHINE. MACHINERY FOR REFRIGERATION. 155 COMBINED MACHINES. Small refrig-e rating- machines are sometimes made not only with their condensers combined on one sole plate as in Fig-. 100, but with both condenser and refrig-erator all com- bined, as in the carbonic acid machines, Fig's. 11 and 12. Fig-. 104 shows two Eng-lish machines of the Reming-ton type with inclosed cranks, which are driven by a vertical intermediate engine; and with the condenser, refrig-erator, and circulating- brine pumps, are all erected complete on one sole plate. A FlG. 104. ENGLISH MACHINE, KILBOURN ENCLOSED TYPE. number of these have been made by the Kilbourn company for shipboard use. Fig 1 . 105 shows a step further than the last fig-ure, and represents a small machine of one and one- half tons refrigerating- capacity, with its boiler as well as a submerged condenser all fitted up complete including- its feed injector on to one cast iron sole plate. This machine is made by the Clyde Engineering- Co., Ltd., of Sydney, for special requirements. COMPOUND AMMONIA COMPRESSORS. The subject of compound compression has already been referred to once or twice on previous pages, but it is one 156 MACHINERY FOR REFRIGERATION. which should have fuller consideration given to it, because there is no doubt the system is making- headway in connec- tion with mechanical refrigeration; and it is possible that in the near future compound ammonia compressors will be the rule, as they are now the exception. If it be said that compound compression complicates the machinery, and that the ice manufacturer or cold storage proprietor wants thing's as simple as possible, it will be well to show that there is not necessarily any more complication and there need be no greater number of parts with a com- pound compressor than there is with a pair of ordinary single-acting 1 compressors; and that under some arrange- ments the compound machine is really much the cheaper, simpler, and better in the matters of first cost, multiplicity of parts, and the attention required when at work. Admitting- for the sake of arg-ument that a compound compressor actually has the same number of working- parts as an ordinary double-acting-, or a pair of single-acting- com- pressors, let us ask: What are the inducements to lead to its adoption? The answer to this is : First, a great equalizing of the turning- moments, which lessens the loads or strains on the cranks, connecting- rods, and pins. This enables these parts to be made of less strength, and so reduces the cost of the machine, while the lessened friction of the wearing- sur- faces economizes the power required to drive it. Secondly, the ability to cool the g-as in the intermediate stag-e, which by reducing- its volume enables the work of the engine to be lessened to a most important extent in large installations; and, Thirdly, the much smoother working, and reduced wear and tear in the whole machine, which is secured by the altered conditions. The idea of compounding a compressor appears to be more than thirty years old, as there were patents granted in connection with it in 1867. The first compound ammonia compressors in Australia were built in 1885 for the Fresh Food and Ice Co., of Sydney, to the designs of their chief en- gineer, the late W. G. Lock, who patented a special invention to maintain the space below the pistons, in both high and low compressors, at the back, or refrigerator, gas pressure. Among the notable builders of compound ammonia com- MACHINERY FOR REFRIGERATION. 157 FIG. 105. SMALL ICE MACHINE CLYDE ENGINEERING CO., LTD., NEW SOUTH WALES, AUSTRALIA. 158 MACHINERY FOR REFRIGERATION. pressors at the present day there are The York Co., of York, Pa., U. S. A. ; The Linde Co., and The Haslam Co., in Eng- land; Clyde Engineering- Co., Ltd., in Sydney, N. S. W., and Humble & Nicholson, of Geelong, Victoria. Fig-. 70, on pag-e 115, represents a 35-ton ice machine by the York Co., and Fig-. 66 is a section of the two cylinders of the compressor by the same makers, who always arrang-e them vertically, while the steam engine may be either verti- cal or horizontal. In the case of only one high and one low pressure cylinder, the arrangement of the whole machine may be the same as in either of the Figures 81 to 85, with horizontal engines ; or as 86 and 89, with vertical engines. Larger machines, as shown by Fig. 70, are made with two low pressure cylinders, which are on the outside, with the one high pressure cylinder between them; in this case the crank pins of the low pressure pistons are in line together, and the high pressure crank pin, on which the horizontal engine works, is at 180 degrees or opposite. Under a design for a still more powerful machine, the builders place four compressor cylin- ders in a row, operated by four cranks; two being high pressure, on the outside, with the low pressure cylinders be- tween them ; and the two connecting rods, from a cross-over compound engine, operate one each on the two outer cranks. There are here Jire main bearings on the shaft.yiwr crank pins, and four crosshead pins to look after ; the fly-wheels are overhung, and therefore must wear down the outer bearings more than the inner ones. (All that is effected here can be carried out with two bearings, two cranks and one fly-wheel. ) The pipes from the cylinder heads connecting the high and low pressure cylinders, through the intervention of the in- termediate condenser or cooler, are seen at the top of the machine in Fig. 70. (For latest design see Chapter XX.) Messrs. Humble & Nicholson, of Geelong, make great numbers of small compound ammonia machines for the butter and cheese factories of Victoria, which have two single-acting compressors arranged side-by-side, driven by a pair of cranks set at 180 degrees; but their machines are all horizontal instead of vertical like the York machine. There is a peculiar feature about all these single-acting side-by-side compound compressors, in that the smaller or MACHINERY FOR REFRIGERATION. 159 high pressure piston is actually a motor on the "down" or 'out" stroke. When the gas is being- expelled from the large, cylinder into the small one, it is of course compressed into the smaller volume at an increasing- pressure, and this pressure acting- on the smaller piston constitutes it a motor, which, acting- on the hig-h pressure crank, assists the rota- tion of the shaft, and therefore assists the engine to force up the larg-e piston against its increasingload. If there was no friction this would mean that the work which the engine had to perform would be equivalent on that stroke that is the low pressure or first compression simply to the pressure of the gas on the difference in the areas of the two pistons. Unfortunately, however, in such cases there is a great deal of friction, and with such machines, the work which the small piston is theoretically able to do is materially discounted, because it has to be transmitted through two pistons, rod packings, two crosshead pins and guide blocks, two crank pins, and the main bearings of the shaft; with friction upon friction, causing increased wear and tear, demanding more attention, and resulting in loss of power at every transfer. In the compound ammonia compressors made by the Linde and Haslam companies, this loss of power and wear and tear are avoided, because the high and low pressure pis- tons are both coupled on to one piston rod, and the cylinders are connected by an intermediate chamber in connection with the suction or back pressure side. Under this arrangement the pressure of the gas in its intermediate stage is conveyed by a pipe from the front of the low pressure piston to the back end of the smaller cylinder, where, acting on the smaller piston in the opposite direction, it directly, and not indirectly, balances an equivalent area on the large piston. In such case, of course, the transference of power is without any fric- tion or wear and tear due to journals and brasses, as in the other plan; and it certainly is a much better mechanical arrangement from an engineering standpoint. Fig. 106 shows such a Linde compound ammonia com- pressor, of European make, combined with a compound steam engine. In this machine the whole engine power is applied to one crank, the whole of the power to work the 160 MACHINERY FOR REFRIGERATION, compressor being- taken off the other crank. In this there is a considerable amount of friction, and a relatively very strong shaft is required, with bearing's to correspond, to withstand the combined strains, or the sum of the working- stresses of the two machines. L.RCYL* FlG. 106. COMPOUND KNGINE AND COMPOUND COMPRESSOR (LINPE). The "Antarctic" compound compressor is so desig-ned that the pressure of the g-as during- the filling- of the smaller cylinder acts upon its piston and directly balances an equiv- alent area of the large piston, just the same as in the Linde machine (Fig-. 106); but the whole arraiig-ement is simplified by the device of casting- the two pistons tog-ether, and then passing- the g-as throug-h the center of them both, from the low to the hig-h pressure cylinder, instead of conveying* it around by a circuitous route of pipes and passag-es. FlG. 107. COMPOUND ENGINE AND COMPOUND COMPRESSOR (ANTARCTIC). Fig-. 107 shows one of these machines combined with a compound engine, where nearly all the work is communi- cated directly from the engine to the compressor, and the crank shaft and cranks only take up the difference, instead MACHINERY FOR REFRIGERATION. 161 of the sum, of the loads. If this arrangement is compared carefully with Fig-. 106 it will be seen that in the latter case the work of the connecting- rods and crank shaft is very much less. Fig-. 108 is a section throug-h the enclosing casing and two cylinders of the "Antarctic" Australian compressor (see following page), with the following explanatory references : A. Main casing- enclosing- the two cylinders. B. Low pressure cylinder. C. Low pressure piston. D. High pressure cylinder. E. High pressure piston. F. Low pressure inlet valve. G. Low pressure outlet valve. H. Passage from low to high pressure cylinder. J. High pressure inlet valve. K. High pressure outlet valve. L. Main inlet branch. M. Main delivery branch. N N. Piston rods. O. Water jacket to h. p. cylinder. P. Crosshead to piston trunk. Q. Openings to insure the filling of cylinder at full back pressure. As the two cylinders both open into the casing, any leak- age past the pistons is intercepted, and as there are no pis- ton rods attached to the centers of the pistons, very large central valves can be fitted in. The top cover or bonnet is secured by only four large bolts, and when the four nuts are off, three of the valves are accessible, as the valve in the low pressure piston is so made as to come right up through the trunk. The lower valve is accessible by means of the special door, which also enables the casing to be cleaned. As the rods do not work through the cylinder bottom, they can be sealed with a considerable depth of oil in this casing without any risk of it being drawn into the system. The cylinders are plain barrels or pipes, and thus they can be easily cast, bored, lapped and renewed. As the valves in the pistons open during the down stroke, and insure practically equal pressure in the two cylinders, the work to be done on the down stroke is found by simply taking the mean intermediate pressure, less the back or casing pressure from the refrigera- tor, and multiplying it by the area of the annulus of the large piston. The upper annulus is of course always exposed to the casing pressure. With the relative areas of the pistons (ii) 162 MACHINERY FOR REFRIGERATION, at 3 : 1 the resistance or load on the down stroke is then the mean pressure as above, multiplied by two-thirds the area of the low pressure piston, and for the up stroke, the load is the forward pressure less the casing pressure, multiplied by only one-third the area of the large piston. It will be as well in order to facilitate a proper compari- son between an ordinary single-acting- compressor, and one MACHINERY FOR REFRIGERATION. 163 of the type illustrated by Figs. 106, 107 and 108, to assume a certain size of machine and then calculate the loads on the two pistons, and the strains or stresses on their respective crank pins, connecting- rods, and other parts. Taking- therefore as a very common size a 20-ton refrig-erating- machine, we may assume a compressor diameter of a little over eleven inches, or say for round numbers, 100 square inches, as the area of the piston, with a back pressure of twenty pounds by the g-aug-e, or thirty-five pounds absolute, and a condenser press- ure of 160 pounds by the g-aug-e, equal to 175 pounds absolute, then the ratio of compression would be 5 : 1. In the ordinary compressor, the load on the piston after the full pressure is reached will be 17535=140 pounds X 100 =14,000 pounds. This pressure continues to the end of the stroke, and all the parts of the machine must be propor- tioned for this amount of stress or load. In the case of the compound compressor, as Fig-. 108, with the small piston one-third the area of the large one, the area of 100 square inches would have one-third or 33.3 square inches of it neutralized by the pressure on the piston above, as it is manifestly the same pressure in both cylinders during- the down stroke. Hence the effective area of the larg-e piston acting- on the gas being- compressed would be 66.6 square inches only, instead of 100 square inches. If the g-as is humid, or is compressed in accordance with Mar- iotte's law, isothermally, into one-third of its original vol- ume, the pressure will rise to 35X3=105 pounds absolute, in the first compression, and if compressed without loss of heat, or in accordance with the adiabatic law, it will reach to about 114.7 pounds absolute. Assuming- then a dry compres- sion, the maximum resistance to the low pressure compound piston will be 114.7 pounds, for round fig-ures, say 115 35=80 X 66.6=5, 328 pounds, which is the greatest stress or load on the machine, instead of 14,000 pounds, as in the other case. Truly a wonderful reduction in the strains to be provided for, in designing- shafts, rods and bearings. In the final compression or up stroke the load cannot be more than 175 35=140 pounds by 33.3 square inches=4,662 pounds. This is a less final pressure than the down stroke, but the point of expulsion is reached much earlier, so that in 164 MACHINERY FOR REFRIGERATION, practice the horse powers of the up and down strokes corre- spond very closely. It is abundantly clear from the foregoing" calculation that the load or stress on the motion gearing of a simple compressor is from two and one-half to three times as great as it need be in one of these compound compressors of equal capacity, quite apart from the disadvantages of unequal running, trouble about clearance, and limited piston speed possible, which do not apply with the same force to com- pound compressors. If this has no more significance than the fact that the same weight of crank shaft, crank connect- ing rods, and such gearing, which is necessary for an ordi- nary compressor of twenty tons capacity, will answer for a compound compressor of fifty tons refrigerating power, it is even then sufficiently startling to inspire the inquiry : If this is true why are compound compressors not more com- monly used? No doubt the scoffer will say, "I could make my com- pressor double-acting and then I would only have 7,000 pounds, not much more than your 5,328 pounds," but he would have full pressure at both ends, loss by clearance and short period of expulsion all absent from the compound machine. Incidentally, there is another feature in the compound compressor which has advantages, in that it produces a more continuous flow of gas from the refrigerating coils, approxi- mating to ordinary double action. The effect of the large piston on the down stroke is to draw into the casing two-thirds of the low pressure cylinder-full from the refrigerator, owing to the enlargement of the capacity of the casing chamber by that volume. On the up stroke this two-thirds is put back into the chamber, and three-thirds, or full volume, is drawn in at the bottom inlet valve; consequently the balance, or one- third of cylinder's volume, is drawn into the casing on the upstroke. This double flow of gas into the casing reduces the friction on the inlet or suction pipe. As the result of the equable distribution of the work throughout both the up and down strokes, these compressors run very steadily, and the author saw one making 140 revo- lutions a minute, as it stood on the fitting shop floor, without MACHINERY FOR REFRIGERATION. 165 a single holding-down bolt. It was found to be quite steady at that speed. In simple compression, with the forward and backward gauge pressures at twenty pounds and 140 pounds, or say thirty five pounds and 155 pounds absolute, the ratio of com- pression would be about 4^ : 1, and the gas would all have to be expelled through the delivery valve into the condenser during the short period of, say, one-fourth of the piston's stroke. With the compound machine a 3 : 1 compression would already exist when the second compression began, so it is evident, as f xf 4/^, that when the high pressure piston DOWN STROKE ATMOSPHERIC FlG. 109. THEORETICAL DIAGRAM SIXGLE-ACTIXG COMPRESSOR. has traveled one-third of its stroke the terminal pressure will be reached, and therefore the expulsion of the gas would be distributed over two-thirds of the stroke instead of being all concentrated on the last quarter of the stroke. This pro- portion is as eight to three, therefore the compound delivery would extend over more than two and a half times as much of the piston's stroke as the other one would do. With this free get-away and more uniform delivery, there is less bank- ing up of pressure and oscillation of the pressure gauge indi- cator, by the friction of the pipes and the jerky supply to the condenser. 166 MACHINERY FOR REFRIGERATION. Fig's. 109 to 112 illustrate graphically the stresses that have just been described, and some of the special features of compound compression by theoretical diagrams. Fig-. 109 is an indicator card from an ammonia compressor, working- between thirty pounds and 160 pounds pressure absolute. From A to B, is the down stroke 011 which no work is done, from B to C shows the work of compression, and C to D the period of expulsion. Both isothermal and adiabatic lines are shown, and the actual curve is taken for the purpose of com- parison, half way between the two. In Fig-. 110 the diagram of the first compression to one- third of the original volume, shows the isothermal line at SECOND COMPRESSION UP STROKE FlG. 110. THEORETICAL DIAGRAM ANTARCTIC COMPRESSOR. 30X3 = 90 pounds, the adiabatic curve rising to over 120 pounds, and the mean pressure at the point O, about 107 pounds. In the second stage of compression, carried on in the smaller cylinder, the curve reaches expulsion pressure at the point P, or about one-third of the stroke. In order to ascertain the effect on the piston rods and crossheads of these different gas pressures the full piston area for the down stroke must be considered as reduced by one-third, and for the up stroke by two-thirds, which gives the two points G and K oil the card as the result of first com- pression. The point L corresponds with P, and thus while the line E O P in Fig. 110 corresponds with B C in Fig. 109 so far as gas pressure on the square inch goes, the spaces MACHINERY FOR REFRIGERATION. 167 which are hatched with vertical lines represent in proper proportion the actual relative amount of work performed by the several pistons, and the different lengths of the vertical lines, the relative stresses or resistances to which the piston rods are subjected under the two svstems, with the actual work of same in both cases. In these diagrams the greater proportion of the stroke L M during- which expulsion takes place in the small hig-h pressure cylinder is very noticeable when compared with C D in the simple compressor. It must be understood that J K FIG. ill. FIG. 112. and H G both represent the same or intermediate pressure as the greater length J O. The area of the simple cylinder being- taken as unity, is represented by J O. The low pres- sure compound cylinder's effective area being- two-thirds of unity, H G is two-thirds of J O, and similarly the hig-h pres- sure cylinder being- one-third the area, J K is drawn one- third of the height, to show graphically the absolute pressure on the whole piston, instead of the pressure per square inch. Fig-s. Ill and 112 almost explain themselves. They are constructed by simply curving Fig's. 109 and 110 round into a circle ; they exhibit the almost steam hammer action of the 168 MACHINERY FOR REFRIGERATION. one case, as compared with the even distribution of the load in the other and compound one, Fig 1 . 112. Figs. 113 and 114 are copies of actual indicator cards taken with separate springs from an Antarctic compressor F/RST COMPRESSION ANTARCTIC REFRIGERATOR. 88 REVOLUTIONS PER MINUTE . 4-8 -5 pa INC. tI 111 3 ^4 -*-' ui S f 1 , ._ -- - -*" 1 " _- ' ' " * d 3 us fcl I*J fcJ vJ CM-* * ' -'- J L ^ 5 *o K i; iu o K v *Oj VI vf 10 <0| N .-? 'k "* i i u ' i : tf ^ 1 ll 1 * 3 5 $ i 1 5 ^J hs * 1 N ^ W * f | f ' 6, ^ in O ^ SECGHO COMPRESSION /JNTARCT/G 1 $ 5 REFRIGERATOR. k V) a? Q 8& REVOLUTIONS PER MINUTE . ll C E 1 n ^ 80 SPRING. 5 t r "- \, i o ^ \^ >o * ^ x t* "^ o " " <5 i> s; \ -- -. U s . '\ --- - -, -.. . --- ' -- -" ,- - - ll THE RELATIVE AREA of CYLINDERS BEING AS 3 ." / THE MEAN PRESSURE ON L.P. PISTON = 27 -O AS /IBOVE ^- ON H.P. PISTON Bo '/M =26-5 f FlG. 113. INDICATOR CARDS FROM AMMONIA COMPRESSOR. running- at eighty-eight revolutions, and up to 171 pounds pressure absolute. In constructing- the theoretical cards, Fig. 110, no account was taken of the chamber H, shown in MACHINERY FOR REFRIGERATION. 159 Fig-. 108, as it does not affect the ultimate results, but its effect on the actual indicator card is clearly seen. With isothermal compression directly from a low pres- sure to a high pressure cylinder, and the pistons moving- uni- formly tog-ether, the low pressure diagram would be a tri- angle, and the line of pressures would be straig-ht instead of a hyperbolic curve. With a chamber like H, in Fig-. 108, which on the completion of the down stroke is filled with g-as of the same pressure as that in the high pressure cylin- der, the lines are considerably altered, because no delivery from the low pressure cylinder will take place on the down stroke until equilibrium is established on both sides of its outlet valve; and this is kept shut by the greater pressure above it at the commencement of the stroke. The inclosed gas, however, is free to pass through the upper valve into the small cylinder, and consequently when the pistons descend it enters freely into the same, the pressure falling above and rising below until the tw r o cylinders and the connecting trunk are in equilibrium; when such is the case, the lower valve opens, and the high and low pressure cylinders are then in direct communication with one another. In Fig. 115 the two cards are reduced to a common scale and plotted together like the cards from compound engines, such as the Westinghouse, and show as follows: Commencing the down stroke at A, from A to B the gas is being com- pressed in the one cylinder alone, and the line is so far the ordinary curve, during which time the pressure in the upper cylinder has been descending from D to E by the expansion of the entrapped gas in the chamber into the small cylinder, when equilibrium is established. This equilibrium is shown at the points B, on the low pressure card, and E, on the high pressure one. From B to C the gas is passing from the low to the high pressure cylinder, and the line E to F is practi- cally straight, corresponding with the line B to C, except so far as it is affected by the resistance of the valves or their spring's. The curve F to G is the line of final com- pression, and the distance G to H the period of expul- sion to the condenser. The curve at J is due to the delay in the opening of the admission valve to the low pressure cylinder. 170 MACHINERY FOR REFRIGERATION. Notwithstanding- the much higher pressure say 1,400 to 1,500 pounds to the inch at which carbonic acid machines are worked, and although air is now compressed to thou- sands of pounds pressure every day (for torpedo work, and so on), the author has not yet met with an example of a com- D I A G RAMS From ANTARCTIC Compressor Taken cuith the same SPRIHG For both Hi^h and Loco Pressure Cylinders . The narrow space between the tcuo diagram* from E. to F. shotus the slight extra pressure on small PISTON due to the SPRINGS on the VALVES . H. G ' \ ' ) o N b rt n w C Q iO Ss x s ft S 8 o o j M] 1C O "6 lo ^ Ml (? in O - n * N tfl r 5 .i col ct BS M r; n 6 ^ ^ ^} 5 C? ^r ? 5 ?' J ^_- " L W P R E & B u R F: A - ,_-- D w N s T R o K E A T M o 5 p H E R l C L 1 N F FlG. 115. INDICATOR CARDS FROM AMMONIA COMPRESSOR. pound carbonic acid machine. This is probably because owing* to the high back pressure the ratio of compression in such machines is less than with ammonia compressors, but there certainly seems no reason why they should not be used, if only to save the trouble and loss occasioned by leakage at MACHINERY FOR REFRIGERATION. 171 172 MACHINERY FOR REFRIGERATION. the piston rod packing- which has to stand the full forward pressure. Fig-. 116 shows, by contrast with Fig's. 98 and 99, the great advantag-es possessed by compound compressors in securing- equable turning- moments, particularly when they are belt driven. The upper diagrams on the Fig-. 116 are identically the same as those shown in Fig-. 115; but cor- rected for relative areas for the pistons, so as to show rela- tive absolute pressures on the whole piston's areas, instead of pressures per square inch. To prevent the confusion of a great number of small overlapping- lines, the graphic con- struction is g-iven for four positions only of the connecting FlG. 117. TURNING MOMENTS WITH TWO COMPOUND COMPRESSORS. rod, which is made two-and-a-half times the leng-th of the stroke. By following- the several parallelograms of force, it will be seen that the heig-ht of the ordinate representing" the pressure on the piston, is first resolved into the horizontal pressure on the compressor guide, and the force acting- in the direction of the center line of the connecting- rod. This force acting- on the connecting- rod, is then resolved into radial thrust or pressure on the crank pin and shaft, and the tang-ential force acting- on the crank pin. The radial ordi- nates set up outside the circle are the same leng-th as those on the tang-ents. The force or pull of the belt which directly MACHINERY FOR REFRIGERATION. 173 corresponds with the power represented in the cylinder dia- grams is, therefore, shown in all positions of the crank or pistons by the curved outer line joining- the radial ordi- nates, and its distance from the crank pin circle. This illus- tration applies specially to the machine shown by Fig-. 100. When two compound compressors are coupled with their cranks, at right angles, the resulting- turning- moments as shown by Fig-. 117 are so uniform, that it is evident no fly- wheel whatever would be required if there was any margin of belt and pulley power. This diagram applies particularly to compound compressors of the g-eneral type shown by Fig-. 101, modifications of which are also made by the Linde and Haslam companies. 174 MACHINERY FOR REFRIGERATION. CHAPTER XVI. ON THE LAWS RELATING TO THE EXPANSION AND COMPRESSION OF GASES. The acquirement of a thorough familiarity with the laws which govern the compression and expansion of gases, can hardly present any difficulty, at the present day, to the col- lege trained youth or university student. The education of such persons should put them in touch with mathematical text books, which now make more or less reference to that branch of knowledge, and whole volumes are to be found, which have been written for their instruction in thermodyna- mic lore. The average refrigeration engineer, however, is, for many reasons, not always able to follow the intricate formulae and equations with which such works abound. This chapter therefore has been written by a practical man (who knows more of the drawing office, machine shop and factory, than he does of the college class room), as an attempt to present to other practical men like himself, some informa- tion connected with the laws of gases, in a more simple form. It is hoped that this will not only assist such persons to investigate the operation of a compressor, but will also enable them to construct theoretical diagrams of the work that should be performed by its piston, so that they may compare them with the actual results, as given by the indicator. If before entering upon this subject it is asked: What is meant by a gas? The reply is: The most distinguishing characteristic of any gas is its elasticity, or its capacity for infinite expansion. Not many years since there were many gases called permanent, which were supposed to admit also MACHINERY FOR REFRIGERATION. 175 of practically indefinite compression; but, although they are now easily liquefied by pressure, when cooled below their critical temperatures, no practical limit yet exists to their expansion. It is found that as the pressure on any gas is diminished, so its volume increases, and that, before all the pressure could be removed, the volume would become so great, that no vessel could be found large enough to con- tain it.* As a consequent result, when the temperature of any gas is increased, either the pressure or the volume, or both pressure and volume, will increase also, and these opera- tions follow more or less closely certain laws, which are gen- erally known as the laws of gases. Our knowledge of these laws is based on the researches of the past two hundred years, and the greatest advances, or those which have led to their comprehension on mechanical, as distinguished from mathematical grounds, have been made during the present century. The establishment of the mechanical equivalent of heat by Joule (under which 772 foot-pounds are accepted as the equivalent of a thermal unit), has enabled the deductions from the earlier discoveries to be corroborated by a sepa- rate process of investigation. BOYLE'S LAW. The first law to be discovered and given to the world in connection with gases, is known indifferently as "Boyle's" law, or "Mariotte's" law. It was published by Robert Boyle in 1662, and Mariotte fourteen years later, in 1675, set it forth carefully verified in his treatise "De la Nature de 1'Air." As French and other continental writers generallv give the credit to Mariotte, English speaking people only do justice to the original discoverer, when they know it as Boyle's law. Under this law, w r ith any mass of gas at constant tem- perature, the product of its volume and pressure is a con- stant. Put in other words, in whatever proportion the pres- *This may be better realized perhaps, if it is borne in mind, that the atmosphere, owing- to this expansive property, extends hundreds of miles out into space, from the surface of the earth. 176 MACHINERY FOR REFRIGERATION. sure of a gas is to be increased, in just such proportion must its volume be diminished, or vice versa, the temperature in both cases remaining- constant. CHARLES' LAW. The second law is called the law of Charles, after M. Charles, who was prof essor of physics at Paris, and who died in 1823. He is said to have been the first discoverer, al- though particulars of it were published by Dalton in 1801, and by Gay-Lussac in 1802. Under this law, with a unit mass of gas under constant pressure, the volume increases from the freezing to the boil- ing temperature of water, directly as the temperature in- creases. It has been found by careful experiment that air under constant pressure increases in volume, as it is raised in tem- perature from the freezing to the boiling points of water, in the ratio of 1:1. 3665 ; or in other words, that 30 cubic inches or feet will increase to about 41 inches or feet, with such an accession of temperature. It follows from this that if Charles' law is a correct one, within the limited range of his experiments, and Boyle's law is good for all temperatures, then Charles' law must also be true for other temperatures and pressures. Because, if Volume is denoted by V\ Pressure by P t Temperature by 7] Then Boyle's law says V P is constant when T is constant; but Charles says when P is constant, and V increases from 1 to 1.3665, then T rises 180; therefore V P is increased at that particular pressure. But Boyle's V P does not depend on any particular pressure, and is true for all pressures. Hence, whatever the pressure of a gas may be, the product of the volume and the pressure, that is V P, will be in- creased in the proportion from 1 to 1.3665, by an increase of 180 Fahrenheit starting from the freezing point of water. Experiments have been carried out with a great number of gases, and their expansion has been measured through the 180 degrees from the freezing to the boiling point, with the MACHINERY FOR REFRIGERATION. 177 result, that the maximum variation is found to range between 1 to 1.367 with air, and 1 to 1.390 with sulphurous acid. As a result of the researches of MM. Regnaultand Rudberg, the ratio of expansion for the average gas, when raised from the freezing to the boiling point of water, may be taken as from 1 to 1.365. That is the volume increases 0.366, or 36.5 per cent, for an increase of temperature of 100 Centigrade, or 180 Fahrenheit; and, as the expansion or contraction is uni- form with each degree, throughout the whole 180 degrees, then it is evident that the expansion for one degree will be ^ which is the same as ^ .,, '. This means that any volume of air at 32, will expand or contract through ^V.* part of its volume, for every Fahren- heit degree that its temperature is increased or reduced, if under uniform pressure throughout. Experimentally, this has been verified up to 700 above, and the law still been found to be true. Inferentially then it is assumed that air would necessarily contract in volume in the same way with a corresponding reduction of temperature, until arriving at 493.2 below the freezing point, or 461.2 below Fahrenheit zero, where it would be in a state of collapse without any remaining elasticity. This temperature of 461.2 is there- fore called the absolute zero of temperature, and Fahrenheit zero is 461.2 of absolute temperature. It will be understood from this that in order to double the volume of a given weight of air at by the thermo- meter, it would have to be heated to 461; and in order to treble its volume, to raise its temperature to 922, and so on. It must not however be inferred also, that these condi- tions apply absolutely and exactly to all gases, and under all conditions. Up to pressures as high as say 100 atmos- pheres, they appear to apply to the more permanent gases, such as oxygen and hydrogen, but not to the gases most used in refrigerating machines, such as ammonia, sulphur- ous acid, and carbonic acid; these are proved to be sensibly more compressible than air. Carbonic acid under five atmospheres does not occupy more than 97 per cent of the volume which air would do under the same pressure, and under forty atmospheres, (12) 178 MACHINERY FOR REFRIGERATION. near the condensing- point, only 74 per cent, or less than three-quarters of the volume it should do on the basis given above. Tables of the progressive pressures required to com- press different g-ases have been published, one of which follows: COMPRESSION OF GASES UNDER A CONSTANT TEMPERATURE. (NOTE. A meter of mercury equals 19.34 pounds per square inch.) PRESSURES IN METERS OF MERCURY. Ratio of the original to the reduced volume. Air. Meters. Nitrogen. Meters. C0 2 Meters. Hydrogen. Meters. 1 1.000 1.000 1.000 1.000 2 1.998 1.997 1.983 2.000 4 3.987 3.992 3.897 4.007 6 5.970 5.980 5.743 6.018 8 7.946 7.964 7.519 8.034 10 9.916 9.944 9.226 10.056 12 11.882 11.919 10.863 12.084 14 13.844 13.891 12.430 14.119 16 15.804 15.860 13.926 16.162 18 17.763 17.825 15.351 18.211 20 19.720 19.789 16.705 20.269 For the purpose of the practical calculations that are required in connection with every-day refrig-eration, it should be sufficiently accurate to estimate on the assumption that the different substances which are used for the medium will behave as if they were perfect g~ases. We can omit the decimal for convenience in ordinary calculations, and admit that a gas will increase 4^3- part of its volume at the freezing- point, or T J T part of its volume at zero, for each degree increase of temperature. If then we have to deal with a given weig-ht of g-as at ordinary atmos- pheric temperature, say 65, and desire to double its vol- ume, it will not be sufficient to increase its temperature by 65, and raise it to 130. Such an addition is altogether beside the mark the actual operation is as under: In order to double 65, first add 461, which gives 526 absolute; and this multiplied by 2, equals 1,052, absolute temperature. Deducting- 461 gives 591, as the thermome- ter temperature of the gas when its volume is doubled. Ag-ain 65 C +461 C = 526X3 equals 1,578 absolute; deduct 461 C , MACHINERY FOR REFRIGERATION. 179 gives 1,117, as the Fahrenheit temperature, when its volume is trebled. Therefore any current of air at 65, such as is ordinarily supplied to the furnace of a steam boiler (where the pres- sure is practically constant), will occupy double volume at 591, and treble volume at 1,117 of temperature. Summing- up all these several laws into a few brief sen- tences, it is found with gases: A. The pressure varies inversely as the volume when the temperature is constant (Boyle). B. The pressure varies directly as the absolute tem- perature when the volume is constant (Charles). C. The volume varies as the absolute temperature when the pressure is constant. D. The product of the pressure and volume is propor- tional to the absolute temperature. The pressure in all these cases is absolute pressure measured from a vacuum; so that the atmospheric pressure must be added to that shown by an ordinary gauge, before making- calculations, and the same must be deducted from calculated results, to g-ive the gauge pressure. The following simple rules based on the foregoing laws may be passed over by the advanced student: 1. With a known volume of a gas at any temperature (the pressure being constant), to find the volume at any other temperature. The sum is a simple one of proportion, V 1 , P 1 , and T l , as before, standing for the volume, pressure and temperature unknown or required. V : V 1 :: 7^+461 : 7^+461, and therefore T + 461 2. With a known volume at a given pressure and con- stant temperature, to find the volume at any other pressure, then F 1 : V: : P : P 1 or F 1 = V-~ 3. With a known volume, at a given pressure and tem- perature, to find the volume at any other pressure and tern- 180 MACHINERY FOR REFRIGERATION. perature. Here the operation is one of double or compound proportion, the result being* in the compound ratio of the absolute temperature direct, and the pressure inversely; thus V: V 1 :: P l (T+461) : P(7' 1 +461), or _ P CP+461) /".(7H-461) 4. With a known volume at a given pressure and tem- perature, to find the pressure at any volume and tempera- ture P-.P-.: F 1 (F+461) : F(7^ + 461) F (2"* +461) F'(7'+461) If the above equations are correct, and the volume of one pound of any particular gas at a given temperature and pressure is known, then it is evidently possible to find a co-efficient for such gas, which will save a great deal of trouble in making calculations connected with it. For instance, take air : The volume of one pound of air at atmospheric density, or 14.7 pounds pressure to the square inch, and at 32, is 12.387 cubic feet. The absolute temperature is 32 + 461 = 493, and hence 12.387 X 14.7 1 = .36935 or 493 2.7074 This fraction, 0.36935, is therefore a constant, which, when multiplied by the weight in pounds, and temperature of the gas in degrees absolute, and divided by the pressure in pounds per square inch, will give the volume in cubic feet; or conversely, the pressure at any volume in cubic feet, of one pound of air. Thus The following table gives the value of this co-efficient (a) for six different 'gases, and any one of these values, multi- MACHINERY FOR REFRIGERATION. 181 plied by 144, gives the co-efficient (a) for pounds per square foot: VOLUME, PRESSURE AND TEMPERATURE OF GASES. Constants (a] for the equation V P = a (7^-461). Name of ga.s. Volume of one pound of gas at 32 F. under one atmosphere. Value of the co- efficient (a). Sulphuric ether 4.79 0.1424 or t r5 Sulphurous acid 5.513 . 1643 or ff o'ss Carbonic acid 8 101 245 or - 1 - Air - 12.387 \j ,&-r*j \ji 4 . j 39g . 3693 or * 4^ Ammonia 21 017 6266 or 1 Gaseous steam 19 913 5937 or A The volume of one pound of, say, ammoniacalg-as, within ordinary working- temperatures and pressures, is found as follows by the use of this co-efficient: ^H61 " 1.596 P That is, take the weig-ht of ammonia in pounds, multi- ply it by the absolute temperature, and divide it by 1.598 times the absolute pressure per square inch, to give you the volume in cubic feet. To find the pressure of any volume of one pound of am- monia: 7H-461 1.596 V To find the density or weight in pounds of a cubic foot of ammonia at a given temperature and pressure : 1.596 P ' 7H-461 THE SPECIFIC HEAT OF GASES. 4 In the earlier part of this volume, there is a table of the specific heats of a number of solid substances ; these in all cases may be taken as constant quantities. M. Reg-nault is the authority for the assumption that the specific heat of a given volume of any one of the permanent gases is also prac- tically constant for all temperatures and pressures, inasmuch as the variation through 360 degrees is not more than 0.2377. But g-as has the property, which solids have not, of altering- its volume considerably; and the specific heat of a gas 182 MACHINERY FOR REFRIGERATION. has to be considered from two separate points, not only when it is under constant volume, but also when it is under constant pressure. The capacity for heat under constant pressure is much greater than under constant volume, and the comprehension of the reason for these two separate attributes of a gas will be much facilitated by the following- diagram, Fig. 118. Here we have a cylinder 35.7 inches diameter, or with an area of 1,001 inches, and a piston rod equal to one square inch in area, so that the effective area of the piston is equal to 1,000 square inches. Now if one pound of ammonia is introduced below the piston at a temperature of 32, or 493 absolute, and a vac- uum is maintained above it, while the weight of 14,700 pounds (equal to the atmospheric pressure of 14.7 pounds on 1,000 square inches) is supported from it, then it will be found that the volume of the pound of gas at such pres- sure and temperature is equal to about twenty-one cubic feet, or 36,288 cubic inches, and that the piston will consequently stand 36.28 inches up from the bottom. Of course this is assuming an absolutely frictionless piston and piston rod for the purpose of the experiment. If more heat is now allowed to pass into the gas until its temperature is doubled, that is raised to 493X2=986 abso- lute, or 525 by the thermometer, and at the same time weights are gradually added in such a way as to maintain the piston continuously in the same (or No. 1) position, then it will be found that when the temperature has been doubled the pressure has doubled also, and that the weights that can be supported in such original, or No. 1, position will amount to 29,400 pounds. If, further, the temperature is trebled, then the piston in No. 1 position will support 44,100 pounds weight, and so on. If the heat that is required to be communicated to the ammonia, to enable it to carry the double load, and to raise its temperature 493, is measured, it will be found to amount to 192.8 thermal units, and if 192.8 is divided by 493, it gives .3911 unit, as the amount of heat which must be communi- cated to the gas for each degree rise of temperature. Therefore .3911 of a unit is said to be the specific heat of MACHINERY FOR REFRIGERATION. 183 PRESSURE. FlG. 118. DIAGRAM TO ILLUSTRATE THE EX- PANSION OF GASES. ONE ATMOSPMES PRESSURE EQUAL. TO I 147OO IBS, 184 MACHINERY FOR REFRIGERATION. ammoniacal gas at constant volume. Similarly a further accession of 192.8 thermal units is again required to be added, when 44,100 pounds is supported in the first position with the same volume of gas, but with its pressure trebled. Let it be assumed that the cylinder is an absolute non- conductor, and that all this additional heat communicated is retained, the first impression of a student of the subject would be that the same amount of heat as is necessary to double the pressure of the gas and carry double the original weight, would raise the original weight alone to the second position; and further, that trebling the heat of the gas would treble its volume, and enable it to raise the 14,700 pounds to the third position. The vacuum, of course, being understood to be maintained above the piston throughout the whole experiment. Such is not the case however. If now a second experiment is attempted, and when the piston is in the first position supporting 29,400 pounds, the additional weights are taken off (leaving only the original 14,700 pounds suspended), in the expectation that the gas at initial volume, and double its initial pressure and tempera- ture, will expand to double its original volume and its initial pressure, it will be found that although the piston will cer- tainly rise and lift the load as the weights are reduced to 14,700 pounds, it will stop a long way short of the double height indicated by the piston in position No. 2. To con- tinue the experiment until the piston is raised to the double height it will be necessary to communicate additional heat to the gas, to the amount of 493 X .1169 unit, which = 57.63 thermal units. With such an amount of additional heat, the piston will raise the original weight the whole of the three feet to the second position, by doubling the volume of the gas beneath it. This additional heat, that is .1169 B. T. U. per pound of gas, is called the latent heat of expansion. When that amount of heat is added to the .3911 unit which repre- sents specific heat at constant volume, it gives a total of .5080 thermal unit, or the specific heat of ammonia gas at constant pressure. In carrying out the first part of the experiments it will be found that as the weights are taken off, the temperature of the gas will fall as the piston rises, although the cylinder MACHINERY FOR REFRIGERATION. 185 is non-conducting; and this fall of temperature represents a loss of thermal units exactly equivalent to the amount of mechanical work done in lifting- the load, as it is reduced in weight. The reason why more thermal units must be imparted to the gas in the second operation (although in both cases only one and the same pound of it is raised in sensible tempera- ture, by exactly the same number of degrees) is easily com- prehended in the light of the mechanical equivalent of heat; because in the second case there is external work performed in raising the 14,700 pounds of weight over three feet. It was from the consideration of these two different aspects of specific heat, in experimenting with gases, that Dr. Mayer is said to have first approximately deduced the value of the mechanical equivalent of heat, which was afterward more accurately determined by Joule, about the year 1842. If we multiply 14,700 pounds by 3.0264 feet, the length of stroke of the piston, we obtain 44,488 foot-pounds as the amount of work done, and if this is divided by 493, then * j*|8 =90.3. This number is the latent heat of expansion for ammonia, expressed in foot-pounds; and when it is added to 301.9, which is the specific heat of ammonia in foot-pounds at constant volume, it gives 392.2 as the specific heat in foot- pounds at constant pressure. This value, it will be recog- nized, is simply the specific heat in thermal units, viz., 0.5080, multiplied by the mechanical equivalent 772, any slight dis- crepancies in the fraction arising from the omission of small decimals. To Mayer is due the credit that he arrived by abstract reasoning at results very close to those which Joule after- ward confirmed by mechanical experiments. The mechan- ical equivalent of heat is now generally termed a "Joule, "and designated by the letter J., as the equivalent of 772 foot- pounds, or 1 B. T. U. The latent heat of expansion expressed in foot-pounds, for any gas, may very easily be found directly, when we know the volume of a given weight of the same at any tem- perature, or have the constants or co-efficients, such as is given in a table on page 181. For instance, take a pound of air at 32 C having a volume 186 MACHINERY FOR REFRIGERATION. of .12,387 cubic feet. It is evident that if such air was con- tained in a flexible bag-, and the volume of the same was doubled by doubling- the absolute temperature, then the whole weig-ht of the atmosphere on one square foot would have to be lifted 12-387 feet. The atmospheric pressure of 14.7 pounds X 144 gives 2,116.8 pounds as the pressure per square foot; and this multiplied by 12.387 gives 26,220.8 foot- pounds, as the amount of work involved in the operation. Then 26,220 divided by 493 (which represents the number of degrees the temperature has to be raised) gives 53.18 as the latent heat of expansion for air, expressed in foot-pounds. The specific heat of air expressed in foot-pounds is, therefore Foot-pounds. At constant pressure 183 . 45 At constant volume 130 . 3 Difference 53 . 15 The ratio of 183.45 to 130.3 is the same as that between .2777 and .1688, which is 1.408 to 1. This ratio has been confirmed by experiments con- ducted by M. Masson, who liberated compressed air, and allowed it to expand back to its original temperature and deduced therefrom the ratio of 1 to 1.41, or 1 to V 2. This ratio is generally represented in the text books by the sig-n y (Gamma, one of the letters of the Greek alpha- bet). The accepted value of y for air is 1.401. ON EXPANSION AND COMPRESSION. In the work of a steam engine expanding a saturated vapor, and in a compressor, such as the Linde machine com- pressing what its makers term "humid" gas, any change of temperature which would be due either to alteration of volume of, or to the work performed by it, or upon it, is modi- fied by the liberation or absorption of heat, that would not affect the operation with a perfect gas. In the steam engine, this arises from the setting free or liberation of heat from the entrained or suspended liquid, on the reduction of pres- sure causing re-evaporation during expansion, and in the compressor, by the absorption of the heat taken up to vapor- MACHINERY FOR REFRIGERATION. 187 ize the liquid held in the gas, which vaporization results from the increase of pressure and temperature. In such cases, the compression and expansion appear more or less closely to follow Boyle's law, and in actual prac- tice with the steam engine, it is generally considered to do so. Under that law, as has been shown, V P is a constant; the curve which represents the variation of the pressure throughout the stroke of the piston, is in such case a hyper- bola, and the operation is termed " isothermal " compression or expansion. To illustrate this graphically let it be assumed that we have gas at two atmospheres of pressure, or fifteen pounds ABSOLUTE PRESSURES. 180 Ibs. 50" B 40" C jo" 20' 15 IO" D E P G INCHES PROM THE END OF THE STROKE. OR RELATIVE VOLUMES. OR VACUUM . FlG. 119. DIAGRAM OF ISOTHERMAL COMPRESSION AND EXPANSION OK A GAS. by the gauge (the atmospheric pressure being taken at fifteen pounds, for the sake of round numbers), let the ratio of compression be 6 to 1, and instead of the cylinder being absolutely non-conducting as was assumed with Fig. 118, let it be a perfect conductor, and through external influences let the gas be maintained at a uniform temperature through- out the whole stroke of the piston. Let the base line of Fig. 119 represent the zero of pres- sure or a vacuum, and its length, A H, sixty inches; this cor- responding with the stroke of the piston, in a compressor or engine working with an expansion or compression of 6:1. It is of course apparent that if the diagram is to represent 188 MACHINERY FOR REFRIGERATION. expansion in an engine, that the stroke of the piston would be from right to left, the cylinder being- filled at initial pres- sure for ten inches out of the full sixty inches before expan- sion begins, and then expanding- through fifty inches to the end of the stroke. As the purpose at present is to consider the action of a compressor, the journey must be made from the left to the rig-ht. With the piston commencing- its stroke at a, the initial volume, or full cylinder, is represented by unit area of piston multiplied by sixty inches, or V=6Q. The initial pres- sure is thirty pounds, or P=30. Then P V equals 1,800, which is graphically illustrated by the lower parallelogram, sixty inches long- multiplied by thirty pounds hig-h. When the piston has moved along- the one-sixth of its stroke the volume of the cylinder will be reduced from sixty to fifty, and if-jj- g-ives thirty-six as the value of P for such volume. At forty inches, /^becomes forty-five; at thirty inches, or half stroke, /^has doubled its original value, and becomes sixty. Similarly, at e and f, P rises to ninety and 120, respectively; while at ten inches from the end, when the gas is confined to one-sixth of its original volume, /^has risen to 180 pounds, or six times its initial pressure. It is evident that the several parallelograms represent- ing P V in all these different positions of the piston, are all of equal area; and as this corresponds with the construction of the hyperbola, a line which joins the points a b g will be a hyperbolic curve. When a diagram of this character has been obtained directly from a cylinder by means of an indicator, the line A H is usually divided into a number of equal parts, say ten or more, by a set of parallel dividing rulers, and ten ordinates or heights are taken in the centers of each of those divisions. The mean height or mean pressure may then be ascertained by adding these ten values together, and dividing them by ten. When, however, there are no means of taking a diagram by an instrument, but the point of cut-off, and the initial and terminal pressures are known, then the mean pressure may be ascertained (without requiring special mathematical knowledge, or the construction of a diagram) by the use of hyperbolic logarithms as given in the table on opposite page. MACHINERY FOR REFRIGERATION, 189 Let/? represent the ratio of compression and expan- sion. H the hyperbolic logarithm of R. P the mean pressure. C the initial pressure before compression. E the initial pressure before expansion. Then for compression P= Cx(l+/7) j? For expansion P= X (1 -f H) K The following- table gives hyperbolic logarithms for a number of different ratios of compression, but it must be understood that they only apply to the compression of any gas under the special circumstances of uniform temperature throughout the stroke with no allowance for clearance. HYPERBOLIC LOGARITHMS FOR CALCULATING EXPANSION AND COMPRESSION OF GASES. Portion of the Portion of the stroke dur- Ratio ing which no of compr es- expansion of si on the gas takes Hyperbolic 1 logarithm. stroke dur- ing- which no expansion of the g-as takes Ratio of compres- sion. Hyperbolic logarithm . place. place. I 9 d 1.11 .104 ft 3.33 1.203 ft 1.14 1.25 .131 .223 I 4.0 5.0 1.386 1.609 1 1.33 .285 6.0 1.7917 /o 1.42 .351 i 7.0 1.9459 i 1.6 .470 i 8.0 2.079 h 1.66 .507 i 9.0 2.1972 I 2.0 2.5 .693 .916 ft h 10.0 12.0 2.302 2.489 1 2.66 .978 ADIABATIC COMPRESSION AND EXPANSION. Reference has already been made to the effect which the humidity of the gas has in its effect on the operation of com- pressing ammonia, it being a special feature of some com- pression plants. It must not however be understood from this, that the refrigerating medium does more than approach to a perfect gas, without actually reaching that condition, even in those plants where the expansion coils of the refrig- 190 MACHINERY FOR REFRIGERATION. erator are of such ample surface (in proportion to the weight of ammonia to be evaporated) that the gas as supplied to the inlet of the compressor is technically "dry." In every-day practice, the line of the diagram, as taken by an indicator, from so called dry compressors, will not follow an exact adiabatic curve, because the walls of the cylinder must trans- mit some of the heat resulting- from the compression, and this heat will be carried away by the jacket of water that nearly always surrounds the cylinders of dry compressors. Every unit of heat thus carried away, by reducing- the pres- sure, of course reduces the amount of power necessary to work the compressor. Notwithstanding- this, if the engineer in charg-e of a com- pressor can set up the true adiabatic line, as well as the isothermal line of compression, upon the actual indicator cards which he takes from his cylinders, he will get then a better idea of the real work which his machine is doing-, and also be able to judg-e whether improvements to it are either desirable or possible. A well fitted compressor should not only have indicator attachments and pressure g-aug-es con- nected as closely as possible to the inlet and outlet branches, but should also have mercury wells for the insertion of ther- mometers close to the same connections. Direct readings of the gauges will give the initial and final pressures, /^and P-\- P 1 , from which 7?, the ratio of compression, can be deduced, and the relation of initial and final volume T^and V 1 . The thermometers will give the initial and final tem- peratures, which by the addition of 461 to each, gives T and T+ T l . When instead of isothermal, it is adiabatic compression which takes place, then instead of P V being constant, it is (P X F)r, or P multiplied by V raised to such power (Gamma) as is appropriate to the special gas under consider- ation, which is constant. These values are given in one of the columns of the table on the opposite page, and it is possible to prove, in accordance w r ith the principles of logarithms, that the numeric ratio which the specific heat of a gas at constant pressure bears to the specific heat of such gas at constant volume (and which in the case of air is 1.408) corresponds with the index of the power to which PxV MACHINERY FOR REFRIGERATION. 191 must be raised to give the true results of adiabatic compres- (V \r -pi I In the case of ammonia, instead of the pressure under compression or expansion varying inversely as the volume, PROPERTIES OF GASES USED FOR ARTIFICIAL REFRIGERATION. GASES. Temperature, 32 F. Pressure, 1 Atmosphere, or 14.7 Pounds Per Square Inch. Sulphuric Ether. Sulphurous Acid. Carbonic Acid. c < Ammonia. Gaseous Steam. 1 Cubic feet in one pound 4.97 5.513 8.101 12.387 21.017 19.913 2 Pounds in one cubic foot 0.209 0.181 0.123 0.080 0.047 0.0502 3 Specific gravity, air being 1 1 2.586 2.247 1.529 1.000 0.589 0.622 4 Co-efficient (a) 0.1424 or 1 0.1643 or 1 0.2415 or 1 0.3693 or 1 0.6266 or 1 0.5937 or 1 V P - a (* + 461) 7.019 6.089 4.1399 2.7074 1.59 1.684 5 Specific heat at constant pressure, In Thermal Units. 0.4810 0.1553 0.2164 0.2377 0.5080 0.4750 6 Specific heat at constant volume, s 0.3411 0.1246 0.1714 0.1688 0.3911 0.3700 7 Latent heat of expansion or A'S = L 0.1399 0.0307 0.0450 0.0689 0.1169 0.1050 8 K k Ratio of specific heats, or or K ^ s 1.41 1.246 1.262 1.408 1.298 1.283 9 Specific heat at constant pressure, | i _= 371.3 119.89 167.06 183.504 392.17 366.7 W ' Specific heat at constant volume, 263.3 96.19 132.32 130.31 301.92 285.64 11 Latent heat of expansion K s = L, 108.028 23.7004 34.74 53.19 90.25 81.06 it varies inversely as the volume raised to the 1.298 power. The following- table gives the ratios of volumes and tem- peratures for air under twenty-five different grades of com- 192 MACHINERY FOR REFRIGERATION. pression, and has columns of differences, to enable interme- diate grades to be dealt with it is taken from a French work: ADIABATIC COMPRESSION OR EXPANSION OF AIR. INVKKSE OF INVERSE OF THESE THESE Ratio of Greater to Ratio of Greater to Less Absolute Temperatures. RATIOS. Ratio of Greater to Less Volumes. RATIOS. Ratio of Less to Greater Ratio of Less to Greater Less Pressures Absolute Temperatures. Volumes. Num- Dif- Num- Dif- Num- Dif- Num- Dif- bers. fer. bers. fer. bers. fer. bers. fer. 1.2 1.054 48 .948 41 1.138 132 .879 91 1.4 1.102 44 .907 34 1.270 126 .788 73 1.6 1.146 40 .873 30 1.396 122 .716 57 1.8 1.186 36 .843 25 1.518 118 .659 48 2 1.222 35 .818 22 1.636 114 .611 40 2.2 1.257 32 .796 20 1.750 112 .571 34 2.4 1.289 30 .776 18 1.862 109 .537 30 2.6 1.319 29 .758 16 1.971 106 .507 26 2.8 1.348 27 .742 15 2.077 105 .481 23 3 1.375 26 .727 13 2.182 102 .458 20 3.2 1.401 25 .714 13 2.284 100 .438 19 3.4 1.436 24 .701 11 2.384 99 .419 16 3.6 1.450 23 .690 11 2.483 97 .403 15 3.8 1.473 22 .679 10 2.580 96 .388 14 4 1.495 21 .669 9 2.676 94 .374 13 4.2 1.516 20 .660 9 2.770 93 .361 12 4.4 1.537 20 .651 9 2.863 93 .349 11 4.6 1.559 19 .642 7 2.955 91 .338 10 4.8 .576 19 .635 8 3.046 89 .328 9 5 .595 86 .627 32 3.135 434 .319 39 6 .691 77 .595 26 3.569 412 .280 29 7 .758 70 .569 22 3.981 396 .251 23 8 .828 63 .547 18 4.377 382 .228 18 9 .891 59 .529 16 4.759 370 .210 15 10 1.950 .513 5.129 ... .195 1 2 3 4 5 After what has been said it must be clear, that in the compression of any gas, the work which has to be done at every successive step or stage, to effect such compression, must add to the pressure, which would result from the sim- ple reduction of volume under Boyle's law, by the addition of the heat units which are equivalent to such work. That being so, the next stage must start with a higher pressure than that which is simply due to P V divided by F 1 . The pressure at the end of each separate stage is dependent upon the work which is necessary to overcome the ever varying MACHINERY FOR REFRIGERATION. 193 pressure during- such stage, and the equation in consequence involves the use of logarithms. It is however only neces- sary to make the steps or stages of the compression rela- tively small to be enabled to arrive at the adiabatic result, with a little more labor, by simple arithmetical calculation alone. As the author is not aware that the method has ever been suggested before, an example may be given, in which some of the stages will be worked by a series of decreasing incre- ments, and others by a system of trials; the proof of the re- sult in all cases will be that the pressure arrived at is directly as the intrinsic energy in the gas, and inversely as its volume. So far as experiments have gone, the specific heat of gases is not seriously affected by difference of pressure and volume. V. V.' O. FlG. 120. DIAGRAM ILLUSTRATING ACCESSION OF HEAT AND INCREASE OF PRESSURE BY COMPRESSION. Let there be a cylinder of known area of piston, filled with unit weight of gas, of known temperature, pressure, and specific heat. Then the volume will be that which is due to the weight, at such temperature and pressure; and if contained in a full working cylinder, the volume divided by the area of the cylinder will give the length of stroke. The intrinsic energy of the contents (which may be called E) will be the product of the weight of the gas multiplied by the number of degrees of absolute temperature, and by its spe- cific heat. In Fig. 120, let the length of the horizontal line V O rep- resent the initial volume of such weight of gas, and the (13) 194 MACHINERY FOR REFRIGERATION. height P O its initial absolute pressure. Then when the volume is reduced to V 1 Q, without accession of heat, the pressure will be increased to P 1 O, and the two parallelo- grams a P O V and b P 1 O V 1 will be of equal area. If the interval from V to V 1 is relatively small, the curve extending from a to b, and representing the increase of pressure during such compression, will approach so closely to a straight line that the mean pressure of the gas during its compression between the two volumes V and V 1 will practically be equal to PO+P 1 O 2 If the mean pressure thus ascertained is multiplied by the area of the piston, it will give the mean resistance to it, or the mean force in pounds exerted by the piston of the machine during the operation. This force multiplied by the distance V to V 1 , in feet (represented in the diagram by the area V a b V 1 ),will give the foot-pounds of work, or the amount of energy, exerted by the piston in effecting that stage of the compression. The number of foot-pounds thus arrived at, if divided by 772, will give the value of such energy in thermal units. If the energy in the gas before compression, and with the piston at V, equals E, and the additional energy which is involved in compressing it from V to V 1 equals E^ , then the total energy in the gas at V 1 will be E -+- E v instead of E, and the temperature at b will be that due to E -f- E 1 , and not that due to E. But if this is the case, and the pressure is directly as the temperature, the pressure at V 1 will not be that first assumed, and represented by the height of P 1 , above O at the point b, but will necessarily be increased in the ratio of E to E 1 , as the effect of the piston's work on the p\ x (E-\-E^^) gas between V and V 1 , and - ^ will give a value P 2 as the pressure due at such volume to the energy in the gas. But if P 2 is the real pressure after compression, then P-LJP* the mean pressure during compression would be - p\ p\ 2 instead of - which has just been assumed to be the case. MACHINERY FOR REFRIGERATION. 195 It is therefore necessary to proceed further, and ascertain the value of the work or energy, E 2 , due to the area and stroke multiplied by the small difference of pressure repre- />2 _ pi" sented by -- - -- , and dealing- with it in the same way as before (by adding- to the already accumulated energy the additional energy represented by the triang-le a b c), find a position P 3 , from which the energy and consequent increase of pressure represented by the triangle a c d could be deduced and added to the g-as. Before this is done, however, the quantities will have become so numerically small that P 3 will be found to coincide very closely with the value obtained by the use of log-arithms. If still greater accuracy is desired, however, then, as the data are established for the area of the triang-le a c d, the heat represented by the addi- tional pressure P 3 may be added, and a fourth value P 4 be found, and so on to the infinitesimal. From this it will be seen that the pressure due to adiabatic compression may be practically arrived at by a series of simple arithmetical additions; it may also be obtained by means of trials, in which the terminal pressure to each stag-e of compression is assumed. In the latter case, the heat or energy, necessary to do the work of compressing- the g"as to such assumed terminal pressure and tempera- ture, is added to the initial energy; and the additional pres- sure due to such work, is then added to the increased pres- sure which is due simply to change of volume. If when this is done, the terminal pressure arrived at by the calculation corresponds with that which was assumed, it may be taken for granted that it is the correct one. If the result is hig-her or lower, then an indication will be given for further trial, which can be repeated until the result is sufficiently close for the purpose. If the interval from V to V 1 in Fig-. 120 is so relatively large that the line a b would have a sensible curve in it, then it is certain that the mean pressure during- such stage of compression would be sensibly less than - , and there- 2 fore any results which mig-ht be obtained by considering- it as straig-ht, would be too hig-h. 196 MACHINERY FOR REFRIGERATION. It may be said that such methods of calculation are use- less, because a table of logarithms will give the same results in much less time than is required for the more lengthy and laborious calculation; but as a great many refrigerating engineers may not think so, and may prefer the simple to the more abstruse operation, it will perhaps be well to go further, and as an example apply these methods of calculation to a compression cylinder of a definite size, and a gas in every day use. Let Fig. 121 represent a cylinder containing one pound of ammonia gas, at a temperature of 32 C or 493 absolute, and a little over two atmospheres, or thirty pounds absolute ABSOLUTE PRESSURES p' ,v\'*s CONSTANT. -=(71) CUACE PRESSURES 165 Ibs. 165 ' 130 120 106 90 > 75 60 45 3O " . t*t " / ISO 135 - ISO " 105 " 9O 75 60 45 " 30 " 15 " O " / / 7 / IMM, / ^/ X / x 93J X x x ^^ x x ^p. '*''? V SOTS, zz^'l'*** *" V ' 38, ' ATMOSPHERIC O " 7 / LINE . 2" 60" 48" .36" >30"- 24"2l" 18" 12" 0" OR VACUUM. ^. B. C. D E. F C. H. J. INCHES FROM THE END OF THE STROKE OR RELATIVE VOLUMES. FlG. 121. ADIABATIC COMPRESSION AND EXPANSION OF AMMONIA. pressure; then according to table on page 191 the volume in cubic feet or Fis equal to ^ Q r~p The pressure P is -L^yo JL^ thirty, and that multiplied by 1.598=47.94. Whence ^fVV gives 10.29 as the volume of the gas in cubic feet, under the conditions stated. If the working length of the cylinder, or the stroke, is six feet, then, --g- 9 - gives 1.715 square feet for the cross section of cylinder, which, multiplied by 144, gives 246.9 (say 247) square inches, as the area of the piston. The initial weight of the gas being one pound, the temper- ature 493, and the specific heat .391, then the initial intrin- MACHINERY FOR REFRIGERATION. 197 sic energy of the gas must be 493 X .391 = 192.76 thermal units - (1) If the compression of the gas from the full six feet length of the cylinder, be calculated through a series of stages, under one or other of the methods just suggested, and the results be compared with those obtained by the use of logarithms, it will possibly lead to a clearer compre- hension of the specific heat of gases under different con- ditions. In the diagram Fig. 121 there are seven stages of com- pression illustrated in the six feet stroke; viz., three of one foot each, two of six inches, and two of three inches each; until the gas is reduced to eighteen inches of the cylinder, or to one-quarter of its initial volume. It may be noted here, that with the accession of heat, a higher pressure is seen to be reached at four-fold compression, than is shown in Fig. 119, with six volumes compressed into one under constant temperature. The total length of the cylinder in this case being six feet, the proportion occupied after the several stages of com- pression will be as follows: Stage of Compression. 1 2 3 4 5 6 7 The length of the cyl- inder occupied by the gas 5' 4' 2' 6" 2' 1' 9" 1' 6" The ratio of original to new volume 1.2 The ratio raised to the power y, which for ammonia=1.298 1 . 267 The initial pressure of thirty pounds ab- solute being multi- plied by these log- arithmic ratios giv- ing adiabatic pres- sure 38.01 1.5 1.692 50 76 2. 2.458 73.74 2.4 3.116 93 48 3. 4.162 124 86 3.42 4.948 148 44 4. 6.04 181 2 The isothermal pres- sures being . 36. 45 60 72. 90. 102.6 120. To calculate the adiabatic pressures for these same stages of compression in the absence of tables : Commencing with the initial pressure of thirty pounds, let it be considered that at the end of the first stage the pis- 198 MACHINERY FOR REFRIGERATION. ton has moved one foot, reducing" the volume in the ratio of 6 : 5; and that the terminal pressure, by increasing- in the ratio 5 : 6, would be thirty-six pounds from alteration of vol- ume alone; then in such case the mean pressure on the piston during its movement would be about thirty-three pounds, because 2 This pressure, thirty-three pounds by 247 inches the area of the piston and by one foot stroke, gives 8,151 foot- pounds as the work of compression; which is equal to 10.43 heat units. Now, as the original energy in the gas (see 1) was 192.76 units, the accession of 10.43 more units would raise the energy of the mass to 203.18 units (2) The pressure for constant volume is directly as the tem- perature or energy; and therefore the pressure of the g-as, when the effect of this extra 10.43 units is taken account of, will not be thirty-six pounds as already arrived at, but ~36 X 203.18 _ 192.76 - But if the terminal pressure is 37.9 pounds it must upset the data on which the previous work was based, because if the terminal pressure is 37.9 instead of 36, then the mean pressure would be 33.95 instead of 33; and the amount of heat or thermal units which should be added for the work done, must be increased in like proportion which is about 2.85 per cent. The 10.43 units when increased by 2.85 per cent amount to 10.72 units, and we can now start the calculation afresh, with 10.72 units as the measure of the additional heat due to the work of compression, the total energy in the gas being 192.76 + 10.72 = 203.48 units - (3) . 36 X 203.48 and ^ - 38 pounds pressure. This amount it will be seen is only one-tenth, or 0.1 of a pound, more than was taken as the basis of the second trial; and although the increment would be too small to have any practical value, still it is evident that by performing- another operation to ascertain the increase of pressure that would be due to the additional temperature that would result from the additional mean pressure of .05 pound to the inch on the MACHINERY FOR REFRIGERATION. 199 compressor piston, the pressure of thirty-eight pounds would be actually increased by a small fraction. The result already attained, however, is so close (within .01, or the one- hundredth part of one pound pressure) to the pressure as calculated by logarithms, as to answer perfectly well for all practical purposes. In connection with the operation of a compressor, where ordinary pressure gauges and thermome- ters are used, the calculation of the pressure to several places of decimals would be useless, because such accuracy would be nullified by the conditions of actual work, and by the relative imperfections of the instruments employed. Commencing the second stage with gas at thirty-eight pounds pressure, and an intrinsic energy of 203.48 units, the movement of the piston through the second foot would re- duce the volume in the ratio of 5 : 4, and the pressure due to 38 X 5 such reduction of volume alone would be =47.5 pounds. For this stage let it be assumed for the purpose of trial that the terminal pressure will be fifty pounds instead of 47.5, then as JL =44, the mean pressure must be taken as forty-four pounds, instead of 42.75 pounds to the inch. A pressure of forty-four pounds on a piston area of 247 square inches, through twelve inches of space, gives 10,868 foot-pounds, = 14.07 thermal units. The total energy at the end of the former stage (3) was 203.48 T. U., and 17.07 added to this, gives 217.55 units total energy (4) The temperature being as the amount of energy, and the pressure as the temperature, 47.5 X 217.55 then - =^TT5 - =o0.77 pounds as the terminal pressure. This differs by only one-tenth of 1 per cent from that calculated by the logarithm ratio, viz., 50.76 pounds. For the third step, reducing the volume in the ratio of 50.77 X 4 from 4 to 3, and initial pressure 50.76 Ibs., - ^-= -- =67.6 o pounds as the pressure due to reduction of volume alone. 200 MACHINERY FOR REFRIGERATION. For trial, as to the energy or work required for the actual compression, assume 72.25 pounds to be the ultimate or terminal pressure. Then - =61.5, which will be assumed as the mean pressure during the operation. The area of piston in square inches, 247, multiplied by 61.5 pounds through one foot, gives 151,905 foot-pounds, or 19.6 thermal units, as the equivalent of the work of compres- sion for this stage. The energy in the gas at the end of the previous stage (4) was 217. 55 units, and adding to this 19.6 additional units as above, gives a total energy of 237.15 thermal units - (5) 66.6 Ibs. X 237.15 Then = 73.6 Ibs. pressure for half stroke. ^4 - / ^O This is over the pressure assumed, and indicates that the assumption was too low ; it is therefore slightly below the pressure found by means of logarithms, viz., 73.74 pounds. If another trial is made and the pressure is assumed to be 73.50, instead of 72.25, then the result will come out prac- tically correct. Having so far compressed the gas to one-half of its original volume, or into three feet length of the cylinder, let the next foot be made by two stages of six inches each. The initial pressure for the stage is 73.7 pounds. The energy in the gas, from (5), is 237.15 thermal units. The compression from three feet to two feet six inches is in the ratio of 6 : 5. Then - 1 = 88. 44 pounds as the pres- o sure due to change of volume only. The mean pressure on the piston for such an increase would be - ' =81.07 pounds. As the stroke is only six inches, with area of piston as 81.07 pounds X 247 inches area ^, . before, - ~~2~~ = 10 > 01 2.14. That is 10,012.14 foot-pounds, is equal to 12.97 thermal units. The initial energy of the gas for the stage was 237.15 units as (5) MACHINERY FOR REFRIGERATION. 201 Therefore the terminal energy is 237.15 + 12.97 = 250.12 - (6) 88.44 Ibs. X 250.12 . - gives 93.27 Ibs. terminal pressure. It is here evident that much too low a figure was assumed for the terminal pressure in taking- 88.44 pounds, because the calculated pressure so far has reached 93.27 pounds, and therefore the amount of energy added to the gas as equiva- lent to the work done on it was too small in at least the same proportion. The actual increase in the work done may be approached closer by a sum in proportion, and 12.97 units X 93.27 . - gives 13.67 units as more nearly the equivalent of the work done than 12.97 units. Trying again with this additional energy allowed for, 237.15 + 13.67 = 250.82 units total energy (7) , 88.44 X 250.8 Then - - gives 93.5 pounds as terminal pressure. The logarithm pressure is 93.42 pounds, and the slight excess probably arises from the mean pressure being some- what less than an arithmetical mean between the initial and terminal pressures. Commencing the second half of the fourth foot of the piston's stroke with a pressure of 93.5 pounds, the volume will be reduced from two feet six inches to two feet, or in the ratio of 5:4; and the pressure for change of volume will be raised proportionately. = 116.87 Ibs. pressure for change of volume alone. ^ =105. 18 arithmetical mean pressure during com- pression. The stroke being six inches only ^l|X247_ =12>9 g 9 7 f 00 t-pounds, = 16.82 units. The heat energy before compression was 250.82 units and 250.82+16.82 = 267.64, total units - - (8) 116.87 Ibs. X 267.64 ^5082" - = 124 ' 61bs - But the pressure on which the accession of heat was based was only 116.87, instead of 124.6 pounds. We have 202 MACHINERY FOR REFRIGERATION. still therefore to allow for 7.73 pounds pressure, and conse- quently the accession of heat due to compression instead of being- 16.82 units will approximate closely to 16.82X124.6 -TT6^- -17.93 units. Commencing- ag-ain, 250.82+17.93 = 268.75 units - (9) as the total energy in the g-as at the end of the stag-e. 116.8X268.75 250.82 ' The log-arithm calculation g-ives 124.86, showing- that the mean was taken a little too hig-h. As the curve in the fig-ure becomes more pronounced at the . hig-h ratios of compression, greater accuracy will be secured by taking- two intervals of three inches each, when reducing- the intervals from two feet to one foot six inches. First, taking- from two feet to one foot nine inches, the ratio is as 8 : 7. 125 X 8 ;- = 142.85 Ibs., due to chang-e of volume alone. Assuming- an ultimate pressure of 148 pounds, take the mean pressure at 136.5 pounds, then the stroke being- the fourth part of a foot 247 v 1 ^6 ^ ^-^ = 8,428.8 foot-pounds, = 10. 91 units. The energy of the g-as at twenty-four inches was 268.75 units (9) and 268.75+10.91 = 279.66 (10) rp, 142.85 Ibs. X 279.66 " 26875 = po^ds. The result as given by log-arithms = 148.4 pounds. Take next step, from one foot nine inches to one foot six inches of cylinder, the ratio being- 7 : 6, = 173.25 pressure due to volume alone. 173.25+148.5 - = 160.8 mean pressure. 1 ^_ L?1Z = 9,929.4 foot-pounds=12.84 units. The heat energy before compression was 279.66 - (10) MACHINERY FOR REFRIGERATION. 203 Then 279.66+12.84 = 292.50 total units - (11) 173.25 X 292.5 And - - =181.2 Ibs. to the inch. 279. ob The result as given by logarithms 181.2. If the condenser pressure is taken for the terminal pres- sure in any compressor, and it is required to ascertain the volume of the gas as expelled, or the point where expulsion commences, it can be found, by working up step-by-step from the back pressure and temperatures, until it is reached. If the elementary methods thus far explained, for the benefit of weak mathematicians like the writer, should give the reader a taste for deeper research into the subject, there are plenty of advanced works on thermodynamics now available for him to wade into. It is not generally found, however, that deep academic research, and great practical skill and experience in the operation of machinery, go hand in hand; the author at any rate has never yet met them combined in the one engineer. Life is too short, and the world is too full of trouble, for a single individual to be able to know everything even about the machinery of refrigeration, although, perchance, you may occasionally meet a man who thinks he fills the order. 204 MACHINERY FOR REFRIGERATION. CHAPTER XVII. STEAM BOILERS FOR COLD STORAGE AND ICE MAKING. Except in the small minority of cases where ample water power is at hand, artificial refrigeration, through the instru- mentality of a compressor, is absolutely dependent upon the boiler as the mainspring- of its operations. Its efficiency and economy become therefore of vital importance in such con- nection. The subjects connected with boilers are however so varied and extensive, and they have already such a consid- erable literature of their own wherein design, construction, use and maintenance are fully dealt with that it may be considered not only rash but futile to attempt to compress any useful information connected with them into the com- pass of a single chapter. On the other hand it is possible that a few things may be said which, without going- too fully into details, are pertinent to the interests of those who are connected with refrigeration and ice making- machinery. Like every other steam user the owner of a compressor is sure to be full of cares in connection with his machinery, and when it comes to the boiler, there are several points about which -he may fairly be anxious. First, That his boiler should be economical in initial cost. Secondly, That he shall obtain from it as many pounds of steam as possible, for every pound of coal he pays for. And Thirdly, That it should cost the minimum amount for attention, maintenance and repairs. Sometimes it is desirable as with other industries that the boiler of an ice factory should occupy as little space as possible, and be independent of brick setting-. In other cases, as when the boiler has to be set up in the close neig-h- MACHINERY FOR REFRIGERATION. 205 borhood of refrigerating- tanks or cold chambers, the heat radiated from the boiler and its setting is not only a direct loss of power, but an indirect loss also, because by heating the surroundings, it increases the work of the compressor, which has to pump such heat out again. THE WATER TUBE BOILER. When we come to investigate the relative cost of different types of boilers, we first note that the evaporation of water into steam is very largely a question of having such water contained in a vessel, and in contact with one side of a metal plate, which plate has its other side exposed to direct radia- tion from the combustion of fuel, or to the heated gases resulting therefrom. It then becomes evident, as metal is generally sold by the pound, that prima facie, the thinnest boiler will be the cheapest. A ton weight of metallic water- vessels, in the form of small tubes, will certainly afford two or three times the heating surface that a ton of plates in an ordinary big boiler shell will do, and therefore, other things being equal, water tube boilers should be the cheapest form to construct for any given power. There is no doubt, more- over, as to their possession of other good qualities, although such are often exaggerated by persons interested in the sale of them. On its average merits, however, the water tube boiler has undoubtedly come to stay. The conditions are not so favorable to water tubes when the steam user has a supply of water impregnated with minerals, which lines them up with a casing or coating almost like marble. Such scale seriously obstructs the con- duction of heat, so that the coal bill may easily be doubled, or the owner may have to pay more for keeping the boiler tubes clean than the interest on the boiler itself comes to, if it is not done in a proper and scientific manner. There are now, however, many special appliances pro- vided for boring out the deposit in water tubes, some of which, operated by tube cleaner companies, such as the Union Boiler Tube Cleaner Co., of Pittsburg, Pa., U. S. A., are so efficient that the removal of the scale becomes a compara- tively simple affair. The necessity for this cleaning is forcibly shown by certificates that boilers after being cleaned had risen from 24.8 per cent to 100 per cent evaporative 206 MACHINERY FOR REFRIGERATION. efficiency ; or, in other words, had by fouling- lost 75.2 per cent, which was restored by the application of a cleaner for a few minutes to each tube. If the water is of an unimpeachable character, or if the deposit can be thrown down in separate vessels either by heating- it in an exhaust steam feed heater, as Fig-s. 144 and 145, or one to heat it to full boiler temperature before feed- ing- it into the boiler itself, as Figs. 146 and 147, then the water tube boiler should give satisfaction. They are specially adapted for high pressure, look well from the outside, and FlG. 122. WATER TUBE WITH FlG. 123. FIRE TUBE WITH SCALE INSIDE. SCALE OUTSIDE. will evaporate as much water as any other well proportioned boiler as long as they are kept clean; but, unfortunately with no other type are there such difficulties in the way of speedily removing the hard scale, or deposit, which rapidl}^ lowers the evaporative efficiency. As this may be thought a strong thing to say, Figs. 122 and 123 should be studied; they show that the deposit in the water tube is absolutely bound in like an arch, and cannot be moved until the key is forcibly broken. The scale however on the outside of the fire tube will crack and drop off with a tap of a hammer, or fly by the sud- den expansion of the tube, when a red hot heater is passed through it. With ordinary hand appliances and hard scale it is no uncommon thing for three men to be half an hour cleaning one water tube; and this costly process had con- siderable weight in restricting the use of water tube boilers in many localities before the resources of inventors provided MACHINERY FOR REFRIGERATION. 207 means for the simple and effective removal of the deposit. So many fine water tube boilers are now made that it would be invidious to mention any by name. Purchasers should look for rapid circulation over the heating- surfaces, and quiet water in the mud drums, as well as simple arrange- ments for cleaning- and removing- the tubes. MULTITUBULAR BOILER. The most formidable all-round rival to the water tube class, seems to be the ordinary underfired multitubular boiler, which appears to hold the premier place in the ice factory, not only in the United States, but in Australia also. When this boiler is properly proportioned and properly set, and is sup- plied with good water, it is, in the author's opinion, the best, the cheapest and the simplest, for its efficiency, of any boiler made. If such a boiler is worked with liquid mud, instead of water, then it should cause no surprise to see carbuncles form on the shell over fire ; such things have happened through carelessness or ignorance, and will probably occur again. Further it will not do to blow these boilers off half an hour after shutting down at week's-end on Saturday, and then fill them up again the same afternoon, when part of the bottom has become red hot from the incandescent furnace walls. A new $4,000 boiler was thus ruined in one act, by a fireman's inexperience, to the author's personal knowledge. Again these boilers often as much as five-eighths of an inch thick over the fire are very susceptible to the action of cylinder oil, when it is returned with the feed water. The oil is apt to form a leathery skin, which keeps the water from direct contact with the plates, and is highly non-conducting. One battery of water works boilers in Australia, at any rate, are known to have become burnt in their bottoms through this cause. Perfect filtration and separation of the oil is ab- solutely necessary for the underfired boilers if the condensed water is returned as feed. Mr. Blechynden, who made ex- haustive experiments on the transmission of heat through plates, has shown that the slightest deposit of grease or dirt on the plates causes a large fall off in the transmission of heat through them. As made in the United States, and illustrated in the cata- logues of refrigeration and other engineers, multitubular 208 MACHINERY FOR REFRIGERATION. boilers for a given horse power, seem to be smaller in diame- ter and to be crammed much more closely with tubes, than is customary in Australia. The pages of Ice and Refrigeration show, that at least one boiler explosion at an ice factory has been attributable to this close packing of tubes, which caused a bad circulation, and a jamming of dirt, in the narrow space between the tubes and the shell. It seems to be often forgotten that the length and diameter of the tubes in a boiler should be determined by the pressure of the draft the chimney can produce. If the draft is light, and the tubes are small and long*, what wonder the} 7 soon foul up and require constant brushing or steam blasting? Australian boiler builders appear generally to favor larger tubes than either English or American makers, four inches diameter being a common size, and it is usual to set them with space down the middle of the boiler wide enough for a lad to get down. This allows for easy scaling, and assists circulation. It is an undoubted fact that numbers of these boilers, which were originally stuck as full of tubes as they could hold, have had their evaporative effi- ciency improved by taking out a row or two in the wings, or down the center of the barrel. The suspension of multitubular boilers over their fur- naces has formed the theme of several engineering papers, and in the discussions thereon great differences of opinion as to small details have been manifested; but there is a general consensus of opinion, that such boilers should be sus- pended from above, and not rest on the side walls of the fur- nace. The long projecting brackets sometimes attached to the shell, to rest on the top of the side wall, must throw a great wrenching strain on the plates, and an angle iron extending the whole length of the boiler is preferable for this purpose if the boiler is not to be hung from girders over- head. In the mounting and setting of multitubular boilers, there is room for great differences of opinion, every maker more or less following his own ideas. For country use in Australia, and in sizes up to about twenty nominal horse power, these boilers are made with a sheet iron casing lined with fire brick, and are known as colo- MACHINERY FOR REFRIGERATION. 209 nial boilers. They are very handy and portable, as will be seen by Fig-. 124, but as they waste fuel by diffusing- the heat of the furnace around them through the four and one-half inches thick of fire brick walls, they can hardly be recom- FlG. 124. UNDER-FIRED BOILER COLONIAL TYPE. mended in connection with refrigeration, except for very small plants. SPECIAL MULTITUBULAR BOILER FOR ICE FACTORY. As a direct contrast to the colonial boiler just referred to, Figs. 125, 126, 127 and 128 show four views of a multi- tubular boiler designed by the author, with a brick work set- ting specially suited for refrigerating- houses. A number of these are working- in Sydney, and are giving" great satisfac- tion, although not in connection with refrig-eration. As will be seen, the air for combustion is taken through the hollow side walls to the ash pit; and all the radiant heat thus intercepted is returned in the heated air supplied to the furnace. This is of course a double advantage, because (14) 210 MACHINERY FOR REFRIGERATION. T7 g^^l^ffisffiff^^ FlG. 125 MULT1TUBULAR BOILER LONGITUDINAL SECTION. FlG. 126 MULTITUBULAR BOILER PLAN. MACHINERY FOR REFRIGERATION. 211 FlG. 127 MULTITUBULAR BOILER DOUBLE FLUE SETTING FRONT ELEVATION. FlG. 128 MULTITUBULAR BOILER DOUBLE FLUE SETTING SECTIONAL VIEW. 212 MACHINERY FOR REFRIGERATION. there is first, better combustion from the heated air delivered to the fuel; and secondly, by the interception of the radiant heat, the walls on the outside of the brick setting- are kept cool. A third advantage is, that the boiler can be shut down from six o'clock in the evening- to six o'clock next morning without losing- more than a few pounds of steam. If Figs. 125 and 127 are examined, it will be seen that the ash pit doors have no hinges; but have a planed groove at bot- tom, which slides on a V-shaped rail. It is very hard to understand why so many boilers should be made with their ash doors hinged, so that when they stand open there is a direct inducement for the fireman to break his shins over them. Apart from the slovenly appearance, when they open at all angles on the floor plates, it is really much easier to regulate the draft when such doors slide than it is when they are hinged. In this particular setting the doors are always kept closed (except when cleaning out the ashes), because the air for combustion enters the regenerative cas- ing by the regulator at the back, and leaves for the ash pit by the openings under the fire bars see Figs. 125 and 126. With these underfilled boilers, plenty of space should be left at the rear for a large combustion chamber, to permit the thorough admixture of the gases; otherwise many of the tubes may have a defective supply of oxygen, and fire will show at the front end when the tube doors are opened, or even at the top of the chimney a sure sign of something very wrong. THK CORNISH BOILER. Human ingenuity has been at work for nearly a centurv designing new patterns of steam boilers. Their number is now so great as to pass any one man's knowledge, and their complicated construction, any one man's power of under- standing. For all that, the most fearful and wonderful designs are still being continually evolved from inventors' brains. Some of these get so far as to be made and tested, while a few reach the advertising pages of the engineering journals. The very best advice that can be offered to any steam user, who is not himself an expert, is to have nothing to do with any revolutionary invention; simplicity is the great MACHINERY FOR REFRIGERATION. 213 desideratum in a boiler, and complication should be shunned. There appears after all these years to be only one man who is entitled to immortality in connection with this branch of engineering-, and that is the father of the high pressure steam boiler, and of the locomotive Richard Trevitbick. One hundred years ago, in 1799-1800, this great Cornish- man was bringing- his high pressure ** puffing" engines into competition with Boulton & Watt's condensers. The "hearse" or "wagon'' boiler of his rivals had superseded Xewcomen's "pot" boilers; but however good the wagon boilers might be for one or two pounds pressure of steam, they were utterly useless for the twenty-five or thirty pounds which the "puffers" worked at. This led Trevi- thick to introduce the horizontal cylindrical boiler, with a tubular furnace and flue, which is now, after a whole century of use, absolutely the same as Trevithick left it so far as form is concerned; and it is still known as the Cornish boiler. These pages are hardly the place in which to pay a tribute to this great inventor of engines, boilers, pumps, steam whims, etc., and also of the locomotive, which anticipated Stephen- son by many years. Fortune favored Watt and Stephenson however, and public opinion has almost made gods of them, while Trevithick's fame seems fair to be forgotten. Neither Watt nor Stephenson appears to have had the mechanical genius of Trevithick, and it is doubtful if the world's real debt to the two together, is as great as it is to the rugged Cornishman. Trevithick however did not possess that faculty o>l generalship,, which is at the present day just as it was in his own time a greater factor than either genius or mechanical skill in securing honors and pecuniary rewards. Trevithick's Cornish boiler is still as good a one as can be obtained for the work of refrigeration, where there is plenty of ground space, and where first cost is not so important as ultimate economy. Such boilers are made by thousands every year, and are used all over the world, as they will run for lengthened periods with less attention than some of the modern patent boilers require every week. In the best boiler builder's work, all angle irons are dispensed with, and the boiler ends are deeply flanged from steel plate circles. The furnaces are all welded up in lengths, without rivets or 214 MACHINERY FOR REFRIGERATION. longitudinal seams, and should be made with the Adamson joint, or be corrugated after one of the patents shown by fox's. FlG. '129 VARIOUS PATENTED SYSTEMS FOR STRENGTHENING BOILER FLUES TO RESIST COLLAPSING PRESSURE. Fig-. 129. The flue behind the furnace should be made the same way, and is further strengthened g-enerally by the MACHINERY FOR REFRIGERATION. 215 insertion of water tubes. Fig-. 130 shows the front of modern Cornish boiler fitted with an automatic stoker. FlG. 130. CORNISH BOILER WITH AUTOMATIC STOKER. The following- table gives the weight and evaporation efficiency of three sizes of modern Cornish boilers by Eng- lish makers: b dj i FOR 160 LBS. PRESSURE. FOR 140 LBS. PRESSURE. p QO ii !* to ! Wl w* Weight. Boiler. Weight. Fittings. Weight. Boiler. Weight. Fittings. 5' 6" 3' 0" 16' 5" 2,2251bs. 14,000 Ibs. 6,720 Ibs. 17,248 Ibs. 6,950 Ibs. 5' 6" 3' 0" 20' 6" 2,950 " 16,352 " 7,050 " 20,160 " 7,286 " 6' 3" 3' 6" 20' 6" 3,700 " 22,960 " 8,170 " 26,880 " 8,406 " If sixteen pounds consumption of steam per horse power per hour is allowed for the 140 pounds pressure boilers, then their horse power comes out at 139, 184, and 231. Allowing- fourteen pounds consumption for the 160 pounds pressure, it makes the horse powers 159, 210, and 264, respectively. 216 MACHINERY FOR REFRIGERATION. THE LANCASHIRE BOILER. When Trevithick boilers are made six feet or more in diameter, they are generally fitted with two furnaces instead of one, and are then called Lancashire boilers. Fig-. 131 is a longitudinal section of a modern Lancashire boiler, suitable for 140 pounds pressure to the square inch, fitted with Gallo- way tubes in the flues. It is a common thing- to hear any ordinary water tubes, which cross a horizontal or vertical furnace, called "Galloways"; the essence of the Galloway tube however, is its conical form, so made in order that the flange at the small end may go through the hole cut for the large end. The small flange is thus fitted inside the flue, as seen in the section, while the large flange fits on the outside. The Galloway company of Manchester, England, are among the most celebrated makers of land boilers in the world, and their special "Galloway boiler" is a modification of the Lan- cashire form; in this system the two furnaces merge into a single kidney-shaped tube or flue, which is filled with taper water tubes, vertical and inclined. The following table gives particulars of eight sizes of Lancashire boilers (two flues) for 105 pounds working pres- sure. From eighteen to twenty-six pounds of coal may be effectively burnt on each square foot of grate per hour, with a good chimney draft: Diam. of boiler. Length of boiler. Diam. of flues. Length of grates. Grate surface. Effective heating- surface. Approximate weight of boiler and mountings for 105 Ibs. working- pressure. Ft. In. Ft. Ft. In. Ft. In. Sq. Ft. Sq. Ft. Tons. Cwt. Pounds. 6 6 18 2 6 4 6 25.5 420 11 9 25,648 6 6 27 2 6 6 30 633 15 33,600 7 21 2 9 5 27.5 541 13 12 30,464 7 30 2 9 6 33 775 18 2 40,544 7 6 21 3 5 30 585 15 12 34,944 7 6 30 3 6 36 839 20 11 46,032 8 21 3 3 5 32.5 626 17 7 38,864 8 30 3 3 6 39 898 22 18 51,296 Fig. 132 shows a front view of Fig. 131. The right hand furnace front being removed, allows the crossed Galloway tubes to be seen, and Fig. 133 is a section of the Galloway MACHINERY FOR REFRIGERATION. 217 218 MACHINERY FOR REFRIGERATION. patent boiler, with two furnaces uniting* in one wide flue, filled with their special water tubes. It will be noted that in Fig 1 . 131 there is a perforated feed pipe, an anti-primer for FlG. 132. LANCASHIRE BOILER FRONT ELEVATION. taking dry steam, a hig^h and low water alarm, dead weight safety valves, and a corrug-ated man hole door, all important. CORNISH TUBULAR BOILEK. A modification of the Cornish boiler, which is daily grow- ing 1 in favor, is shown in the four illustrations Fig's 134 to 137, MACHINERY FOR REFRIGERATION. 219 which represent a boiler specially designed by the author, for using- water which makes a very hard deposit. There are several wide departures in it from common practice, the prin- cipal of which is the placing- of the furnace to one side, in- stead of in the center of the shell. This arrangement g-ives great facilities for the exami- nation and cleaning- of the inside, and also promotes better circulation. The multitubular arrang-ement of the back breaks up the g-ases, and by the increase of heating surface enables the whole boiler to be materially shortened. Two FlG. 133. SECTION OF GALLOWAY BOILER. eminent English authorities have certified that a boiler of this design in use at the office of a London daily paper had an efficiency equal to the evaporation of 10.15 pounds of water from 212 C per pound of coal consumed. By the adoption of one large four-feet furnace instead of having two furnaces each two feet six inches diameter, as is common with seven-feet shells, a much better combustion of the fuel is possible; but with a high pressure like 120 pounds it requires a special construction of the furnace, on one of the systems shown by Fig. 129, to withstand the 220 MACHINERY FOR REFRIGERATION. MACHINERY FOR REFRIGERATION, 221 222 MACHINERY FOR REFRIGERATION. collapsing- strain with metal of a reasonable thickness. It is the Adamson joint which is shown and adopted, principally for the reason that nearly all first-class boiler shops have now a flanging machine, whereas the various systems of corrug-ated furnaces require a special plant to produce them. The "Morrison," which has to a large extent superseded Sampson Foxe's original corrug-ated furnace, is perhaps the most popular one on shipboard now. THK RBGENERATIVB SETTING. In this boiler, Fig. 134 as well as in the multitubular boiler, Fig-. 125, the " setting- " is arrang-ed with double flues; those next the boiler itself are traversed by the hot gases of combustion on their way to the chimney while the outer passages in the brick work serve to bring the cold air to the furnace. After several years' experience with boilers set in this way, the author is able to say with confidence that the arrangement is a most successful one. It is really pos- sible to be for some time close at hand to the boiler without knowing it is at work, as the outer brick work keeps perfectly cool. For this reason the walls do not crack and let in the cold air, or require buck staffs to keep them together. For an ice factory where the water supply is brackish, or has a heavy impregnation of other mineral substances, and space is available, the Lancashire and Cornish tubular boilers may be relied upon for giving- satisfaction. Where the water is good the underfired boiler, as Figs. 124, 125, would be the most economical in the long- run. Where it is absolutely necessary to make the most steam in the least space, no doubt the locomotive boiler, like Fig. 138, would best answer the requirements. A boiler of this type may be made with tubes as small as one and one-half inches or one and one-quarter inches diameter and by having- a forced draft will burn five times as much fuel per square foot of grate sur- face as would be economical or desirable with either the underfired boiler or the Cornish one. THE GENERAL CONSTRUCTION AND MOUNTINGS OF BOILERS FOR ICE PLANTS AND COLD STORES. Leaving- for the present the old argument that there is no advantag-e in having an economical engine to operate the MACHINERY FOR REFRIGERATION. 223 FlG. 136 CORNISH TUBULAR BOILER FRONT ELEVATION. v J FlG. 137 CORNISH TUBULAR BOILER SECTION. 224 MACHINERY FOR REFRIGERATION. machinery in an ice factory, because you must evaporate a greater weight of water to supply the distillate for the ice cans than even a wasteful engine requires, it would be a fair thing to assume a working steam pressure of at least 120 pounds to the inch, if coal costs as much as $2.50, or ten shil- lings, a ton. If economy based upon the best practice is desired, owing to more costly fuel, then 160 to 200 pounds may be employed. Now in order to carry 120 pounds work- FlG. 138. LOCOMOTIVE TYPE STATIONARY BOILER. ing pressure, year in and year out, with satisfaction, the con- ditions demand first-class material and workmanship, and a cheap boiler will surely prove in the long" run a most costly investment; the highest bid however does not necessarily guarantee the highest quality in the article supplied. Among the many important points to be looked for in a good boiler, the most essential perhaps are among the follow- ing: Material, mild ductile steel of moderate say twenty- eighttons, and nothigh, say thirty-two tons tensional strength MACHINERY FOR REFRIGERATION. 225 under test, for shells, furnaces and flues. All plates to be planed on their edges. All rivet holes to be drilled in place, after the plates are bent. All longitudinal seams to be at least double riveted, or double strapped. Manholes to be strengthened with special reinforcement rings, and have stamped steel corrugated manhole doors. No valves or cocks to be bolted directly on to the boiler shell, but be secured either to solid blocks, as in Figs. 124 and 134, or to short welded steel stand-pipes riveted on, as in Fig-. 131. The very best gauge-glass mounting's procurable, preferably asbestos packed, should only be used ; and where they require pipes as in Pig. 125, these connections should be of copper with screw FIG. 139. BJORNSTAD'S BLOW-OFF COCK. FIG. 140. unions. All underfired boilers to have their bottom set with a fall of two or three inches to a chamber to receive deposit for the blow-off cock or valve. If perforated pipes lying- on the bottom are used for blow-off, then a frequent use is required, especially before raising- steam, to remove deposit thrown down. Spring- or dead weight safety valves to be used in preference to those with levers, which latter often vibrate with the pulsation of the steam flow to the engine. All the stop valves to have external screws. The best blow- off cock yet invented appears to be the one shown by Figs. 139 and 140, which has the following characteristics: There are no "ground" surfaces exposed to the deposit when the (15) 226 MACHINERY FOR REFRIGERATION. cock is shut, consequently when it is opened there is no scor- ing* caused to make it leak. It can be packed under steam. The key cannot be withdrawn until the cock is completely closed. Plenty of space should be left around the tubes for their examination and cleaning*. Louvres or regulators should be fitted to the doors, to regulate the supply of air both below and above the fire. The fire bars to be specially suited for the grade of coal used and the rate of consumption. All in- ternal furnaces or flues to be strengthened on one of the systems shown in Fig-. 129, so as to avoid the necessity for heavy plates, and the risk of burning- them. Above all, let the intending- purchaser beware of the 44 great economy " fiend ; and (although it is an old chestnut) it \vill be well for him not to forg-et the story of the steam user who adopted all the latest improvements offered to him, and when he had paid all the bills and totted up what had been promised (as is promised every day), he obtained the following as the result of the gross saving- to be expected : By con- torted tubular boiler, 20 per cent ; acrobatic fire bars, 10 per cent; steam dryer, 5 per cent; automatic damper regulator, 5 per cent; patent cut-off, 15 per cent; waterless condenser, 20 per cent ; economizer and feed heater, 25 per cent ; purifier and softener, 10 per cent, or a total saving of 110 per cent. He therefore thought he should be burning 10 per cent less than nothing, and his coal heap should be getting larger ; but somehow or other he found the coal went away just about the same as before. Lying open on the table as this is being written, is an advertisement, in a highly reputable journal, which boldly undertakes to increase the efficiency of the boiler up to 55 per cent by the adoption of the one particular device offered. Now where things are so bad that a 55 per cent improvement is possible, it may in most cases be taken for granted, that all the saving will not be effected by one piece of apparatus, but will probably require the whole steam plant remodeled by a competent expert. The greater economy which increased steam pressures, and higher grades of expansion will effect, are shown in the following table, which gives the relative quantity of coal required for the same horse power, under UNIVERSITY MACHINERY FOR REFRIGERATION. 227 different steam pressures up to 300 pounds to the inch, and with grades of expansion up to eight fold. It must not be forgotten that 180 pounds of steam is now a pressure in common use, both on land and at sea: COMPARATIVE WEIGHT OF COAL REQUIRED PER HORSE POWER PER HOUR, WITH STEAM PRESSURES FROM THIRTY TO 300 POUNDS PER SQUARE INCH, AND GRADES OF EXPANSION FROM TO VH. e Grade of Expansion. If Oi/ i/ 74 /3 3/3 X % % X K Steam Pounds Inch. Weight of Coal in Pounds. 30 5.6 4.93 1 3.95 3.81 3.30 2.84 2.69 2.35 1.82 35 5.51 4.84 3.86 : 3.72 3.21 2.74 2.60 2.26 1.73 40 5.46 4.79 3.81 | 3.67 3.16 2.70 2.55 2.21 1.68 45 5.41 4.73 3.75 3.62 3.11 2.65 2.50 2.16 1.62 50 5.36 4.68 3.71 1 3.57 3.06 2.60 2.45 2.11 1.58 55 5.31 4.63 3.66 3.51 3.01 2.55 2.40 2.06 1.53 60 5.26 4.59 3.60 3.47 2.97 2.50 2.35 2.02 1.49 65 5.20 4.55 3.57 3.43 2.93 2.46 2.31 1.98 1.45 70 5.19 4.52 3.54 3.40 2.90 2.43 2.28 1.94 1.41 75 5.16 4.49 3.51 3.37 2.87 3.40 2.25 1.91 1.39 80 5.12 4.45 3.47 3.33 2.83 2.36 2.21 1.88 1.35 85 5.09 4.42 3.44 3.30 2.80 2.33 2.18 1.85 1.32 90 5.07 4.39 3.41 3.28 2.'77 2.31 2.16 1.82 1.29 95 5.04 4.37 3.39 3.25 2.74 2.28 2.13 1.79 1.26 100 5.01 4.34 3.36 3.23 2.72 2.26 2.10 1.77 1.23 105 5.00 4.32 3.35 3.21 2.70 2.24 2.09 1.75 1.22 115 4.98 4.31 3.33 3.19 2.69 2.22 2.07 1.73 1.20 125 4.94 4.27 3.29 3.15 2.65 2.19 2.03 1.70 1.17 150 4.81 4.14 3.16 3.02 2.52 2.05 1.90 1.57 1.04 200 4.70 4.03 3.05 I 2.91 2.41 1.94 1.79 1.46 0.92 250 4.69 3.93 3.01 2.81 2.31 1.85 1.70 1.36 0.83 300 4.54 3.87 2.89 2.75 2.24 1.78 1.62 1.29 0.75 This table shows that with the low pressure of thirty pounds steam, and no expansion, as was common many years ago, the consumption of coal would be double that required with eighty-five pounds pressure and a cut-off at half stroke; and further, that more economy can be obtained by increas- ing 1 the expansion and raising 1 the pressure, until the con- sumption is only one-seventh of that given under the lowest conditions. After having- secured a good boiler, the next thing is to have it properly set, with the sectional area of the flues so 228 MACHINERY FOR REFRIGERATION. proportioned for the volume of the gases to be carried to the chimney, as to get the best results from the fuel. Many arguments are being put forward in favor of a mechanical draft, urged by fans, instead of having the natural draft of a chimney, and some of them are very specious. It would be going outside the general scope of this work to discuss this question in detail, and it may be left with the remark that a tall chimney at any rate carries the heated waste gases away well clear of the factory, which the short stumpy outlets much advocated by some engineers certainly do not, and a chimney certainly wants no attention in comparison with a fan or exhauster. Having the boiler set with double side walls, and the top above the brick work encased with at least two inches of good non-conducting composition, the whole setting and casing, as well as the house, should be kept scrupulously clean and white; then the radiation of heat will be reduced to a min- imum. Black, dirty boilers, and settings smothered in dust, with dark and dirty surroundings, all greatly favor the radia- tion and conduction of heat, which as before shown, is specially objectionable in an ice factory. In connection with the efficiency of engines and boilers, no work has probably ever been done of such service to the general steam user to enable him to see where the losses really occur as the report and diagram on the Louisville pumping engines recently issued by a committee of the Insti- tute of Civil Engineers. This celebrated Leavitt engine, at Louisville, has been described in the transactions of the American Society of Mechanical Engineers. Its operations have since been investigated by a committee of the English society appointed in 1896 to establish a standard for comparing and judging the thermal efficiency of steam engines, and has resulted in a report, and the diagram reproduced in Fig. 141. This figure illustrates the flow of heat, in British ther- mal units, from the furnace to the actual brake power exerted. The various losses or leakages by radiation, con- densation, and so on, are clearly shown; and also the saving of heat again picked up, as by the economizer, and the return of hot water from the jackets. MACHINERY FOR REFRIGERATION. 229 230 MACHINERY FOR REFRIGERATION. The heat put into the water from the furnace per minute is 133,600 units, and that represented by the brake power is only 25,990 units, or say 19 per cent; that lost or thrown away in the condenser alone being- 110,240 units, or over 58 per cent. The radiation from the boiler, the steam pipes and the engine is comparatively small, and the flue and other losses are so relatively insignificant, that when an inventor comes along with his offer of 50 per cent saving, the steam user having this diagram in hand, may possibly be able to tell his would-be benefactor that he is professing to save a great deal more heat or power than is actually lost. In the Leavitt engine, 221 units per minute are required for an indi- cated horse power, which in an ideal engine are reduced to 148 units. MACHINERY FOR REFRIGERATION. 231 CHAPTER XVIII. ICE PER TON OF COAL. Looked at as commercial operations, the success, or otherwise, of both ice manufacture and refrigeration is largely a question of coal consumption. It is no doubt true that instances are common where it is advisable to expend a little extra money on fuel rather than incur the additional first cost and subsequent up-keep which would be involved in the change to more economical machinery and highly refined appliances. At the same time it really does seem if the records are true that many ice factories are altogether more wasteful in the use of fuel, and show poorer results, than there is any necessity for. At the annual meeting of the Southern Ice Exchange of the United States held at St. Louis in 1898, a paper was read in which the author, Mr. Sneddon, gave the results obtained by him from twenty-seven different ice factories. These are reproduced in the table on the following page, and show that the water evaporated, or ice made, per pound of coal, ranged from 8.22 pounds in the best, to only 2.25 pounds in the worst case. This in itself appears a very wide range of rela- tive efficiencies, the better results being more than three and one-half times as much as the poorer ones, and it is made the more singular from the fact that the thermal efficiency of the coal used for the smallest evaporation was fully equal to that used for the highest. The best of these tabulated cases, however, compares very poorly with the result of some tests which were made at a Bavarian brewery twelve years earlier, and are recorded in a paper read by the managing director of the British Linde Company before the Institute of Mechanical Engineers in 232 MACHINERY FOR REFRIGERATION. 1886. It is there stated, that as much as 26.3 tons of ice have been made for the ton of coal. As this is more than ten times as great as the results in some of the factories referred to by Mr. Sneddon, and as it is impossible that the difference in climate, and temperature of condensing- water, can be responsible for the whole of such great discrepancies, it will perhaps be worth while to look a little deeper into this ques- tion. There is no serious reason, on the face of it, why some of the factories in the list should not at least treble their effi- ciency, with fair averag-e plants and modern methods. TABLE OF ICE PLANT EFFICIENCIES COLLECTED FROM TWENTY-SEVEN EXISTING AND OPERATING PLANTS ( SNEDDON) 3 i jtffcri _L d 3 ^ 3i2^ o ^^15 'C a a | IS I|I1 -si's '"8 ^1 sill's i d rr, 2 fe 9* s 2 w-c a il 1 H S S. ?3a^ 1 o-~ s d P^ rt & v."* fco'g, P ri -i 0, - S 5 H 5 1 ^ 5.4 4,800 10,800 2.25 13,400 2,261 17.6 75. 5.7 4,800 11,400 2.37 13,400 2,381 17.7 74.8 7.25 4,000 14,500 3.62 12,200 3,638 29.8 57.5 7.5 4,000 15,000 3.75 12,200 3,768 30.8 56.0 10.33 5,000 20,660 4.13 14,858 4,150 27.9 60.2 11. 5,000 22,000 3.93 14,858 3,949 19.9 71.6 14. 12,000 28,000 2.33 12,700 2,341 18.4 74.3 14.5 9,000 29,000 3.22 11,900 3,236 27.2 61.2 16.5 9,500 33, 000 3.47 11,900 3,488 29.3 58.0 15.6 9,600 31,200 3.25 12,300 3,266 26.5 62.2 16.5 11,200 33,000 2.94 12,300 2,954 24. 65.8 20. 12,000 40,000 3.33 12,600 3,346 26.5 62.2 19. 8,000 38,000 4.75 12,200 4,773 39.1 44.2 14.5 6,000 29,000 4.83 12.200 4,854 39.8 43.2 17.5 10,000 35,000 3.5 12,000 3,517 29.3 58.0 17.66 10,000 35,320 3.53 12, 600 3,547 28.1 60.0 27.5 13,500 55.000 4.07 13,000 4,090 31.3 55.3 19. 7,000 38,000 5.42 12,200 5,447 44.6 36.3 20. 6,000 40,000 6.66 13,000 6.693 51.4 26.6 23. 6,800 46,000 6.76 13,000 6,793 52.2 25.5 24. 7,000 48,000 6.85 13,000 6,884 53.0 25.8 29. 14,000 58,000 4.14 12,000 4,160 34.6 50.6 25. 18,000 50,000 2.77 12,000 2,783 23.2 66.9 32. 22,000 64,000 2.90 12,000 3,045 25.3 62.9 31. 14,000 62,000 4.42 10,500 4,442 42.2 39.8 82. 22,500 164,000 7.28 13,100 7,286 55.6 20.6 85. 20,740 170,000 8.22 13,100 8,261 63.0 10.0 It is evident that the plea, "There is no advantage in having an expansive eng-ine, because you would have to con- MACHINERY FOR REFRIGERATION. 233 dense live steam to make the distilled water for the cans," does not apply in these cases. It certainly does not require specially good boilers to evaporate six and three-fourths pounds of water per pound of coal, considering- that eight to nine pounds is easily attainable. Taking- the moderate evaporation of six and three-fourths pounds only, and allow- ing one-third of it, or two and one-fourth pounds, for waste in condensation, drainage, etc., there would still be four and one-half pounds left to make ice from, or double the amount actually yielded by the plant. In making ice there are so many minor losses from radia- tion, conduction, thawing out, and so on, which aifect the ultimate result, that it makes it exceedingly difficult to calcu- late from theoretical data before-hand, what the production of a new plant will come up to. The practical man who has rule- of-thumb notes, deduced from the working of similar plants under varying conditions, will probably get nearer to the mark than the engineer who calculates everything on a scientific basis alone. The main factors which are concerned in the question of maximum ice for minimum coal versus minimum ice for max- imum coal, are as follows: First. There is the thermal efficiency of the coal itself, which may range from 10,000 to 15,000 units in a pound, and the efficiency of the boiler as a machine. Although the best coals are theoretically equivalent to fifteen pounds of water from 212, the highest evaporation in actual practice does not much exceed ten pounds of water per pound of coal,* while seven pounds is a low result. It will be a fair thing to as- sume eight and one-half pounds as a fair average evaporation attainable in an ordinary factory. Secondly. There is the efficiency of the steam engine in terms of the weight of steam consumed. The very highest result so far published, appears to have been attained by an experimental quadruple-expansion engine at the Cor- nell University, with a boiler pressure of 500 pounds to the square inch, and a record of ten pounds of steam per horse- *See reference to boiler, Fig-. 135, page 219, result of trial. 234 MACHINERY FOR REFRIGERATION, power-hour. Nothing- like this is possible in actual work at present, and the very best marine engines probably do not use less than thirteen pounds, even when working- with triple expansion and an initial pressure of 200 pounds of steam. The following- is perhaps a fair averag-e of steam consumption in commercial, as disting-uished from experi- mental, engines: Pounds of steam per horse power per hour. Condensing-, quadruple and triple expansion.. 14 to 16 Ibs. Condensing-, compound 16 to 24 Ibs. Non-condensing, compound or expansion 20 to 30 Ibs. Condensing-, low pressure 30 to 40 Ibs. Non-condensing, low pressure 40 to 60 Ibs. If a plant of machinery is intended for refrigeration only, and distilled water is not required, there are no absolute reasons why the steam engine supplying- the power should not have a surface condenser, and if on shipboard, be worked at the same pressure, and with the same degree of economy, as the main engines. In such case an indicated horse power might be obtained by the expenditure of from 1.5 to 1.7 pounds of coal. When however distilled water must be had in order to make clear, crystal, can ice, it will be better to work the engines non-condensing, and with a backpressure, under one of the two systems to be presently described, and to use a high initial steam pressure with a high grade of expansion, preferably in a compound or triple expansion engine. Although the increased range of temperatures due to a vacuum will be sacrificed so far as the engine is concerned, by not expanding down below atmospheric pressure, there will be no difficulty even then in getting an indicated horse power with twenty-five pounds of steam per hour. The con- denser and vacuum, as will be seen later on, can be turned to* better account than simply to increase the power of the engine. Thirdly. There is the efficiency of the compressor as a complete machine, or the ratio which the indicated horse powers of the steam cylinder, and the compressor, bear to one another. The following table has been collated from the several examples therein quoted, and it shows that the frac- MACHINERY FOR REFRIGERATION. 235 tional losses in such machines range between 12 per cent and 33 per cent of the total engine power: Authority or source of the information. 1 fc II W Horse power of the compressor. Ratio of compressor to engine power, per centum. Friction or loss in terms of the enjrine power, per centum. Friction or loss in terms of the com- pressor, per centum. Mr. A. Siebert in Ice and 25 % 33 % RcfTi&efcition for Janu- to to ary, 1899 33 % 50 % Diagrams illustrating") their machines from the ' De La Vergne cata- f logue . J 63.0 48.0 76.1# 23. 9 # 31.4$, Bavarian brewery in 1886. 53 38 71.7 28.3 39.4 Comparative trials in 1890, Linde and Pictet ma- chines Average of four Pictet. 79.3 20.7 26.1 Best Pictet trial 81 1 12 9 14 8 Average of four Linde 83 4 16 6 19 9 Best Linde trial 87 9 12 1 13 7 "Eclipse" machines Frick Co. 's Red Book ) illustrations C 60.3 51.6 83 17 20.4 "Case" machine from the \ company's book f 63.9 56.5 88 12 13.6 Case Co. 's guarantee 87 13.04 15 Fair average to assume ) with a good design and > the best workmanship. ) 100 83.3 83.3 16.6 20 It will be seen from the foregoing table that the highest efficiency is obtained with a machine which has its steam and ammonia cylinders connected up in a straight line, fully supporting what was said in previous chapters, as to fric- tional losses by round-about connections. The makers of 236 MACHINERY FOR REFRIGERATION. this machine guarantee that their engine power will not exceed the compressor power by more than 15 per cent. It will leave a considerable margin, if in considering" the whole question of efficiency, we assume the indicated engine horse power in a new plant at 20 per cent in excess of that of the compressor. Fourthly. There is the efficiency of the compressor itself considered as a pump, which may vary between very wide limits. In the year 1878, the writer designed the compress- ors, reservoirs, and reducing valves, that have been success- fully used ever since that date, for lighting the cars on the New South Wales railways with gas. The original machinery was all made in Sydney, and the pressure was intended to range between 120 and 180 pounds. A large imported compressor was subsequently put to work, which, when tested by him, was found to deliver only about one- half of its theoretical capacity, the defects being due prob- ably to small valves, too large clearance, and the great heat generated. Although some makers claim 98 per cent efficiency for their own manufactures, such machines will be very effective, and have small clearance, if they can be kept so cool as to pump 95 per cent of their theoretical volume from the refrig- erator. Unless frozen well back, 90 per cent would probably be nearer to the average effect obtained. Then there is Fifthly. The height which the abstracted heat has to be lifted from the temperature in the refrigerator, in order that it may be carried away by the condensing water. From Munich to Central Australia is a far cry, and these places present very different conditions for the ice maker to study. In the records of the Munich experiments the tem- peratures of the condensing water are given as : At entrance 49 49 48 48 Fah. At exit 67 67 49 67 In a machine designed by the writer for an East Indian city it was stipulated among other conditions of the trial, that the average atmospheric temperature was to be 95 and the water 90. In some towns in Australia, such as Bourke, out west (where the crust of the earth is said to be very thin between the people and the place below where there is no MACHINERY FOR REFRIGERATION. 237 ice), the summer heat is often 120 in the shade for long- periods tog-ether. As far as the time required for freezing- the blocks of ice is concerned, the brine mig-ht be kept at the same tempera- ture at these two places having such extremes of climate, but with 40 difference in the surrounding's, the greater leakag-e of heat through the insulated walls of the tank would cause a much more serious loss in the hot climate. Further, the greater tendency of the ice to thaw when drawn from the cans might be an inducement to freeze colder in the hot than in the cool city. Such conditions would widen the disparity in the relative efficiencies of the two plants, when measured by the weight of ice produced for sale at such widely sepa- rated localities, from a given weight of fuel. In order to make a comparison of the relative work required to make ice in the two cases, we may omit the latter considerations and take a back pressure of twentv-four pounds (gauge) or thirty-nine pounds (absolute) in both cli- mates, with a condenser temperature of 65 C in Bavaria and of 105 C in Central Australia, the gauge pressures being 103 pounds and 218 pounds respectively. The table on following page gives the volume of gas required to be pumped per minute in cubic feet to produce one ton of refrigeration, and for purposes of comparison these quantities will be doubled as is usual to give the amount per ton of ice. This leaves a considerable margin for waste, and much more in the case of the cold climate, because forty more thermal units have to be abstracted to make ice from water at 90 than from water at 50, while the melting of a ton of ice represents an absolute quantity of heat or work in any climate. The table under column headed 24 (as the gauge suction pressure) shows, that with a terminal pressure of 103 pounds to the inch, the gas required to be with- drawn from the refrigerator will be 2.87 cubic feet per minute per ton of refrigeration; and with 218 pounds ter- minal pressure, then 3.12 cubic feet per minute must be withdrawn. Now by plotting the isothermal and adiabatic lines of compression, from thirty-nine pounds to 118 pounds abso- lute pressures, to represent the work in a cool climate, and 238 MACHINERY FOR REFRIGERATION. fc <: 5 s fa-*, v as W P 2 w W fa W P^ 5 IS H W < * CO pj ^ fa H O H W rH 10 o (U 8 s g g 5 rH CO $ G iO 1 g t- 8 So ON 00 rH to ON ON 8 C Cu $ o 03 0) ./ s 8 O rH CM rH rH CM* rH IN ci CM fO to o || rH CM to CM CM rH CM ci .1 ON CM' CM M rH & CM C^ rH CM CO \D CM* rH CM ci ci CM' S CM rH '&8) So CM ON ! to $ rH CO 5 ON * * 10 8 o 10 rH 1 O 3 3 y 3 Tj- t * 88 3 o uj uj to X 10 * 1 "o CO 5- 5 fO 1 rH to 2 to \6 rH O 7 E 1 s to fO t t> ^ CN g g 88 C & || jmperature in } n deg. Fahr. . f Corresponding- temperatures in degrees Fahr. o o g o o 1 O o o 8 rH o rH i!! B.8, 1.1 r pressure aug-e) juare inch. 3 rH a . . . . g X rH Q OQ tfc o O Condense (byg Ibs. per s( MACHINERY FOR REFRIGERATION. 239 from thirty-nine to 233 pounds, for the requirements of nearly tropical surroundings, the following- Fig-. No. 142 is the result. We find from the above that the mean pressures during compression are forty-seven pounds and 78.5 pounds under the two different conditions. Now 78.5 poundsX144 inches= 11,304 foot-pounds as the amount of work required to com- press one cubic foot of gas, and 11,304X3.12 cubic feet gives 35,268 foot-pounds per minute as the work for one ton of re- Cfnfrj/ aro/ie dnaf Ce/ifrj/ di/ DIAGRAM or COMPARATIVE PRESSURES o WORK i/ncfer xtremes of C/imf>c*A runt eoj / Z HA ruat /SI i ft Tir/ 1 'AL 'UA A 4*9, fS '[ FlG. 152. DIAGRAM OF HEAT TRANSFERS IN TRIPLE EFFECT. The supply of water fed into the first vessel may be heated by means of coils in the chimney or flues, or by other appliances for the transfer of heat, with increase of economy. If however it be assumed that its temperature is only 120, then the first operation will be to raise the 4,500 pounds of water from 120- to 203, the latter being- the temperature of vaporization in No. 1 vessel: 203 120=83. 4,500 pounds X 83 373,500 thermal units. MACHINERY FOR REFRIGERATION. 257 Taking the latent ^heat of steam at five pounds gauge 373 500 pressure to be 952 units, then ^r- =391 pounds steam con- densed as the equivalent of raising 4,500 pounds of water 83. The steam passing from No. 1 vessel is marked 1,437 pounds, therefore that weight of water has to be evaporated at a temperature of 203, the latent heat at such temperature being 972. The latent heat of the steam in calandria is 952. 1 437 X 972 Then - L %2Uatentheat) =1 ' 467 P Unds &S the Weight f steam condensed equivalent to the evaporation. Adding this 1,467 pounds to the 393 pounds above, gives 1,860 pounds weight of condensed water to be drawn from the first calandria, which is of course the same as the exhaust steam introduced. This water may be returned as feed to the boiler direct, or be filtered and heated in an economizer. Deducting this weight of 1,437 pounds evaporated in the first vessel from the total of 4,500 pounds supplied to it, 4,500 1,437 = 3,063 pounds of water passing to second vessel. This water passes in at a temperature of 203, but as the temperature of the second vessel due to the better vacuum is only 181, it will, in falling the difference, 203181=22, give off vapor as follows: 3,063X22 ~l920atentheat) = ' P UndS (nearly) f eva P ratlon ' As the vapor from the top of the first vessel amounting to 1,437 pounds is condensed in the second calandria it will- being assisted by the better vacuum and lower temperature evaporate an equal weight, or nearly so. Adding 1,437 to 70 gives a total of 1,507 pounds evaporated from the top of the second vessel. Deducting again this weight of 1,507 pounds from 3,063 passing in at the bottom, 3,063 1,507 gives 1,556 of water to supply the third vessel. This being in direct communica- tion with a surface condenser, and having a vacuum of twenty-four inches, the corresponding -temperature will be lowered to 145 C , and the water, in dropping from 181 C , will part with 181 3 145^=36 units per pound. Then *' 556 X 36 _ =55 Ibs. (full) of vapor. 1,012 (latent heat) (17) 258 MACHINERY FOR REFRIGERATION. As before, taking* the evaporation in the third vessel, due to the condensation in its calandria of the vapor from the second one, to be equal in weight, or 1,507 pounds, the total will be 1,507 + 55=1,562 pounds evaporated from the top of the third vessel. The slight discrepancy between 1,556 pounds entering- the third vessel and 1,562 pounds leaving- which should be of course equal is due to slight differences in the latent heat allowed for. The sum of the different weights of vapor passing out of the three vessels to be condensed for the supply of the ice cans is 1,437 + 1,507 + 1,562=4,506 pounds, slig-htly in excess of what was supplied to the first vessel. The weight of steam from the engine exhaust was 1,854 pounds, therefore ^-^~ A = 2.43 pounds of distilled water for each pound of exhaust steam. It will easily be understood, that by putting- an exchanger on the last vessel's outlet to the condenser, where the tem- perature is 145, more initial heat could be given to the water supply of 4,500 pounds weig'ht above 120, with improved results. If the supply is fed into No. 1 vessel at 203 then = 3.08 pounds of water per pound of steam. The condensed vapor from the third vessel will be deliv- ered by the main air pump from the surface condenser, and in order to take the w r ater from the calandrias of Nos. 2 and 3, small voiding pumps or supplementary air pumps, as before described, are necessary. Such an apparatus as that described is found in practice to require about one square foot of heating surface for six pounds of water to be evaporated per hour, therefore - = 750 square feet, or 250 feet for each vessel. The tubes would be about thirty inches long-, one and one-half inches in diameter and No. 16 or 17 gauge in thick- ness. It will be noticed in Fig-. 151 that the vapor pipes differ in size. This is to make the fall of temperature between the vessels as slight as possible. The velocity of the vapor is not greater than 3,500 feet per minute into the first calandria, MACHINERY FOR REFRIGERATION. 259 4,000 feet to the second; 5,000 to the third, and 7,000 feet to the condenser. The Colonial Sugar Refining- Co., of Sydney, have num- bers of these plants some of enormous size working at FlG. 153. SIX-FOLD EFFECT FOR DISTILLED WATER. their mills in New South Wales, in Queensland, and in the South Sea Islands, quadruple and quintuple as well as triple, and by successive stages they have much reduced the com- plication so that one-half of the cocks and fittings as used in Europe are now done away with. They have also, by the use 260 MACHINERY FOR REFRIGERATION. of large pipes giving a low velocity to the vapor, reduced the friction and loss of pressure, and largely increased the efficiency of the plant. The author is much indebted to his friend, Mr. Hector Kidd, member Institute Mechanical Engineers, for much reliable information derived from a very wide experience with these evaporating plants, and for the information that with the company's quintuple effects as much as six pounds of water per pound of steam is evaporated, or say fifty pounds to one pound of very ordinary fuel. A paper by Mr. Kidd on this subject will be found in the third volume of the trans- actions of the engineering association of New South Wales. Fig. 153 shows a sextuple effect plant suitable for such places as the dry uplands of Western Australia, where the water supply is so salt or brackish as to.be unfit -for potable uses. It is not so powerful or economical as the plant in Fig. 151, but is differently arranged to enable the salt deposit to be easily removed. The distilling condenser and cooler lie horizontal and communicate with the sextuple effects coupled up to the vertical column of separators. Such machines are capable with six effects of producing four and one-half pounds of fresh water from sea water for every pound of steam raised in the boiler; and being generally independent of the exhaust steam of an engine, are worked at a much higher initial pressure and temperature than under the system shown in the larger plan No. 151. MACHINERY FOR REFRIGERATION. 261 CHAPTER XX. SUPPLEMENTARY AND FINAL. The loss of time necessarily involved through this work, written in Australia, being- printed and published in Chicago, has sufficed for a progressive art like mechanical refrigeration to move perceptibly forward in the interval. This would seem to warrant the inclusion of the additional illustrations and remarks regarding same which follow. A large part of the matter comprising this chapter is merely an outline of the principal distinctive features of each of the machines illustrated, with comparatively little analyti- cal comment upon same, except in a few cases. As intimated above, the time necessary to accomplish such work would unduly advance the date of publication. It is proposed, how- ever, to prepare for a second edition of this work an exhaust- ive analysis of all features of machinery and systems herein illustrated, the principles of which have not been thoroughly explained and described in this edition. LATE TYPES AMERICAN ABSORPTION MACHINERY. The absorption system is briefly described in Chapter VIII, and the process diagrammatically explained by Fig. 14, but no details or illustrations are there given of the vari- ous parts of the plant. In the Vogt type of absorption machine, illustrated by Fig. 154, on following page, the noticeable feature is the ab- sence of round coils and bent pipes throughout the entire system. Fig. 155 shows three views of the improved generator or still of an absorption plant, as made by the Henry Vogt Machine Co., Louisville, Ky., U. S. A., including the rectify- ing and analyzing devices. By the system of fractional dis- 262 MACHINERY FOR REFRIGERATION. MACHINERY FOR REFRIGERATION. 263 tillation thus carried out, it is claimed that practically anhy- drous ammonia is obtained. The strong- liquor enters by the side connection on the top of the stand pipe. The gas dissolved in such liquor is evaporated and driven off as it passes through the successive stages involved in flowing through A, B, C, D, E and F. The FlG. 155. GENERATOR OR STILL FOR VOGT ABSORPTION PLANT. liquor is left very weak by the time it reaches the compart- ment O. An examination of the mechanical construction of this generator shows that it consists of a main casting, divided into four compartments, communicating with each other ; and four horizontal pipes, connected to main casting, which contain the steam heating coils. The upper compart- 264 MACHINERY FOR REFRIGERATION. ment of the main casting- is connected to a stand pipe containing- an analyzer and rectifying- coil for drying the g-as before leaving- the still. The strong liquor is admitted at top of stand pipe, passes through the rectifying coils and analyzer to the upper compartment of the main casting-, flow- ing thence over the steam coil in the horizontal pipes from MACHINERY FOR REFRIGERATION. 265 one to the other until the lower compartment is reached. The gas generated passes through the opening in each com- partment to the stand pipe, where the moisture is deposited, and the dry gas passes to the condenser. Fig. 156 is a modern type of heat exchanger or economizer. It is made with straight concentric pipes, and is of a most mechanical and trustworthy design. It will be seen that the outer tubes are connected at the alternate ends by H pieces, and that the internal pipes are coupled by external bends, which also act as glands to the jointing. This method of con- struction makes what should be a thoroughly reliable job. The strong liquor on its way to the still enters the ex- changer at the bottom, leaving at the top. The weak liquor from the still enters the exchanger at the top and leaves same at the bottom. The ammonia pump used is of the double-acting hori- zontal fly-wheel pattern. The special feature of this pump is the ammonia stuffing box and the water chamber sur- rounding it, which latter acts as a lubricator for the piston rod. The speed of the pump is twenty-five revolutions per minute. The absorber is constructed like an upright tubular boiler open at the top. Tubes are distributed uniformly and arranged in such manner that they can be cleaned while the machine is in operation. The cooling water enters at the bottom and discharges at the top. The return gas from the expansion coils enters at the bottom and the weak liquor at the top, the flow of the latter being controlled by an auto- matic regulator. The Ball American absorption machine, made by the Ice and Cold Machine Co., St. Louis, Mo., U. S. A., as originally constructed in 1878, and of five tons daily ice making capacity, was a slight modification of the Carre machine. The ice tank was eight feet square and twenty- four inches deep, and the ice cans four inches thick by eight inches wide, also eight inches square by twenty inches deep, making ice weighing twenty-five and fifty pounds each, respectively. The cans were made of galvanized iron, some of them of copper. The original cost of building machine was $14,000. 266 MACHINERY FOR REFRIGERATION. $ * Hiitl&i* ^W^^ ^S^F)!^ w o o w H g "" o p MACHINERY FOR REFRIGERATION. 267 After twenty-two years the machine is still of the Carre type, enlarged and made to meet the American idea of large units and expansion. See Fig. 157. The tank of eight feet square has developed into tanks 30 X 90 feet, and from one to four attached to one machine. Blocks of ice no longer weigh twenty-five pounds, but from 100 to 400 pounds, and, if you talked copper cans, you would be thought crazy. The generator is a vertical cylinder of marine steel, with removable top head to same, heated with steam coil, and with drying pans in the gas dome. The condenser is of the open air or submerged type, depending upon the water. . The poor liquor upon leaving the generator goes into the shell of ex- changer or equalizer, which is a cylinder with removable heads containing tubes. The poor liquor, from the shell of this exchanger, goes to the poor liquor cooler coils (either of submerged or open air type), and from there to the absorber. The gas being liquefied in the condenser goes through expansion valves to expansion coils in freezing tank, and returns from freezing tank to absorber. The absorber is a cylindrical vessel with vertical tubes, the water passing up through the tubes, cooling the ammonia and carrying off the heat generated by absorption. The ammonia pump consists of two single-acting vertical pumps driven by direct connected vertical engine, pumping the now enriched ammonia from absorber through the tubes of the exchanger into the top of generator, completing the cycle. The separation of moisture in retort is exceedingly good, an air blast of 14 below zero F., being obtained under ordi- nary working conditions, and a temperature in the ice tank of from zero to 2 above, F., being maintained for months at a time. In experimental machines, absorbers built of straight pipe and injecting the poor liquor into the gas returning from the expansion coils have been made with very satisfac- tory results, especially so where this absorber is placed from twenty-five to thirty feet above the expansion coils, and the weight of the rich liquor coming down to the rich liquor tank reducing the back pressure in the expansion coils some six or seven pounds, and allowing a very low temperature to be 268 MACHINERY FOR REFRIGERATION. carried. An absorber of this type works after the same manner as a Bulkley siphon steam condenser. LATE AMERICAN CARBONIC ACID MACHINES. When discussing- the use of carbonic acid as a refrigerat- ing agent or medium, the only examples illustrated were by Messrs. Hall, of Dartford, in England. Such machines are now made in Germany, in Australia, by Mephan Ferguson, of Melbourne, and in the United States. FlG. 158. SECTIONS OF THE COCHRAN CO. 'S CARBONIC ANHYDRIDE MACHINE, LORAIN, OHIO, U. S. A. Figs. 158 and 159 are two American machines of recent introduction, which show that notwithstanding the greater power required for a given amount of refrigeration, carbonic acid has more than compensating advantages for small units, and in special circumstances. This is owing to its innocuous character, and the absence of danger in the case of an escape of gas from the machine. Messrs. Kroeschell Bros.' machine has an extremely neat and mechanical looking appearance. That by the Cochran Co. is made more complete and portable by having the con- MACHINERY FOR REFRIGERATION, 269 denser combined on one sole plate with it. The illustration, Fig-. 158, shows a cross and transverse section of their simple motor driven compressor, and is one of their latest designed machines. SOME LATE AMERICAN CONDENSERS. Fig. 160 is an atmospheric condenser for an absorption plant, made by the Henry Vogt Machine Co., Louisville, Ky., U. S. A., and is of the type (described on page 82 ante) where vertical headers are connected by zigzag coils laid horizon- tally. These zigzag coils form a two-storied condenser, as they are built in two separate sections. The gas condenser FlG. 159. KROESCHELL BROS. CARBONIC ACID MACHINE. CHICAGO, ILL., U. S. A. headers are set over those for the weak liquor, and this saves condensing water, as the weak liquor on its way to the ab- sorber is cooled by the waste water from the condenser proper. Besides these two 'exchangers forecooler coils are shown in the water tray below, and the arrangement is such that the hot gas first enters these submerged coils, where it is partially cooled by the water from the coils above. The gas then passes to the top of the first condenser headers, flowing horizontally through the coils until the anhydrous liquor is drawn off at B. From D to C is the weak liquor cooler, the liquid entering at D and flowing upward in the reverse direction to the 270 MACHINERY FOR REFRIGERATION. cooling- water to its exit at C. Practically this is a three- story or triplex condenser, and it is as such (and as an illus- LJ VOTeS SUPPLY 1 FlG. 160. TRIPLEX AMMONIA CONDENSER FOR VOGT AMERICAN ABSORPTION MACHINE. tration of a modern detail to secure economy in the use of condensing- water) that it is introduced. Submerg-ed condensers are used with this machine, especially where the water is impure and contains much lime. MACHINERY FOR REFRIGERATION. 271 Fig. 161 is an illustration of an ammonia condenser of the Ball type. It is especially effective where the cooling- water is warm or scarce, and was originally constructed in 1890. Since then it has been extensively copied by other builders. As seen, the hot gas from compressor is admitted in the lower pipes and after having- the sensible heat taken out of same it is piped to the top of condenser where the fresh water comes in contact with the cool gas, instead of being heated by hot g-as, as in the old type of condensers formerly con- structed. The saving in actual expense is claimed to be about M 4 . Sprinkling- Trough. 15 per cent in the water bill, or about 10 per cent in the con- denser pressure. Condensing- liquid has the same direction of flow as the gas, and not an opposite flow, as in some con- densers. Fig-. 162 is a section of Frick Co.'s latest design of atmos- pheric ammonia condenser. In the construction of this con- denser the manufacturers have aimed to have the cold water come in contact with the coldest g-as, which, becoming- warmer as it meets the warmest g-as, flows off over the hot gas pipes from compressors, finally passing off through the overflow. The ammonia condensers designed and constructed by the Fred W. Wolf Co., Chicago, are of the atmospheric type 272 MACHINERY FOR REFRIGERATION. MACHINERY FOR REFRIGERATION. 273 (IX) 274 MACHINERY FOR REFRIGERATION. MACHINERY FOR REFRIGERATION. 275 (see Fig-. 163), the standard size of each section being- twenty- four 2-inch pipes twenty feet long-. These pipes are manufact- ured from selected skelp, and the drop forg-e Bessemer steel flang-es are screwed on to same while hot, thereby allowing- the flange to shrink on as it cools. These condensers are supplied with galvanized iron water troughs with patent level- ing device, and between the pipes is fastened a perforated steel strip, thereby allowing- a free circulation of air. Each section of these condensers is supplied with an inlet and outlet valve, thereby allowing each section to be evacuated of ammonia without interfering- with the operation of the re- maining- sections when connected with the suction of the ma- chine. Fig-. 164 shows a side and end elevation of the Westerlin & Campbell patent double-pipe ammonia condenser, an inven- tion of recent date, constructed with a view of incorporating all of the best and most practical features of both the sub- merg-ed and atmospheric types of ammonia condensers. Reference to the cut will show that the condenser is con- structed with a small pipe encased within a larg-er pipe; usually the internal pipe is one and one-fourth inches and the external pipe two inches. The g-as inlet is located in the center of the coil, and the hot ammonia g-as enters the space between the l^(-inch and 2-inch pipes, spreading- both ways from the center toward the two ends. At the ends the gas travels down to the next space between the pipes below, where it travels from both ends toward the center, and again spreads tow r ard the two ends in the next succeeding- space, the object being- to film the gas out to the greatest possible extent, bring-ing all of the gas in direct contact with both the internal and the external cooling- surfaces. The water en- ters the 1^-inch pipe at the foot of the condenser and travels back and forth upward until it overflows into the manifold at the top and end of the condenser. It will be noticed that while the travel of the g-as is downward the travel of the water is upward, making an interchange of temperature that results in the warmest water meeting with the current of the warmest gas, and the gas is gradually cooled down and condensed into liquid as it travels along-, meeting with the cooler water, until finally the ammonia liquid is discharged at 276 MACHINERY FOR REFRIGERATION, the bottom of the condenser at a temperature as low as the temperature of the initial water in the internal pipe. At the foot of the condenser connections are provided for conveying- the liquid ammonia to the liquid receiver, and also for con- necting- to the suction pipe of the machine, or to the absorber, in case the condenser is used in connection with an absorp- tion machine, so that the g-as inlet and liquid outlet can be closed and all of the gas in the condenser can be drawn out in case of necessity for repairs without interfering- with the operation of the balance of the plant. The condensers are usually erected in nests of several stands, and it is always possible to cut out one stand for repairs without shutting- down the plant. The water connections are cross-connected in such a manner that the water current can be reversed when it is desired to wash out the internal pipe. It has been asserted by engineers that such a construction of condenser could not be used in connection with waters badly impreg- nated with scale forming- properties, but a considerable experience with the worst waters in America has demon- strated that the scale will not form in the pipes at all, even after more than a year of continuous operation, the scouring of the rapid current of water through the internal pipe posi- tively preventing deposit of scale, sand or mud. No water is used over the outside of the condenser, consequently it can be located at any desired point about the plant, without necessity for water pans or tight floors. The condenser can also be placed at any desired level, as the water can be deliv- ered to any height above the coils. COOLING TOWERS. On pages 64 to 67, ante, reference is made to the re-use of condensing water and the different arrangements by which this may be effected. Fig-. 24 shows an evaporative conden- ser in a cooling- tower. Recently several new devices for ac- complishing- the same result have been devised, notably that embodied in the patents of John Stocker, St. Louis, Mo., U. S. A., which shows a hig-h degree of efficiency. A very ingenious method for distributing the water over the tower is adopted, the cooling- surfaces being so arranged that a perfectly even discharg-e of air over the water is accomplished by means of two fans instead of one. MACHINERY FOR REFRIGERATION. 277 SOME RECENT AMERICAN VALVES. Fig-. 165 is a section of the Ball valve used by the Ice and Cold Machine Co., of St. Louis, Mo. Its construction is re- FlG. 165. BALL DISCHARGE VALVE. FlG. 166. FRED W. WOLF CO. 'S AMMONIA VALVE. f erred to in the description of the Ball compression machine found on page 303. 278 MACHINERY FOR REFRIGERATION. The ammonia globe valves, manufactured by the Fred W. Wolf Co. (see Fig-. 166), each contain the soft metal seat at B. These valves are well proportioned, of good weight, and are made to stand a pressure of 500 pounds; further- FlG. 167. FRICK CO. 'S AMMONIA VALVE. more, standing the strain of expansion, contraction and the weight of pipe and settling. They are also so constructed that they need not be closed to repack stuffing boxes, as in having the valve entirely open any leak through the stuffing box is entirely obviated. MACHINERY FOR REFRIGERATION. 279 280 MACHINERY FOR REFRIGERATION. The latest construction of the Frick ammonia valve, re- ferred to in Chapter XIII of this work, is shown in section by Fig-. 167, on page 278. THB LATEST DESIGNS OF AMERICAN AMMONIA COMPRESSION MACHINERY. Fig-. 168 is a half-tone illustration of Frick Co.'s exhibit at the National Export Exposition, Philadelphia, 1899. FlG. 169. ELEVATION FRICK CO. 'S LATEST AMMONIA COMPRESSOR CYLINDER, WAYNESBORO, PA., U. S. A. Figs. 169 and 170 are the elevation and section respect- ively, of the most recent pattern of "Eclipse" pump in other words, the ammonia compressor cylinder as now made by the Frick Co., of Waynesboro, Pa. This new design is worth careful study by both the student and hard-shell engineer, as it is a good example of MACHINERY FOR REFRIGERATION. 281 how efficiency may be combined with simplicity. It will also be noticed that it embodies those special qualities upon which PURGING VALVE FlG. 170. SECTION FRICK CO. 'S LATEST AMMONIA COMPRESSOR CYLINDER, WAYNESBORO, PA., U. S. A. so much stress was laid (as being- desirable in such cylin- ders) on pages 110, 111, 123 and 151, ante. It was there argued that plain barrel cylinders, without attached feet or encir- 282 MACHINERY FOR REFRIGERATION. cling 1 passages, favored homogeneous casting's of sound and solid metal; and this is effected in the pump under notice, by the simple but elegant device of detaching the delivery pass- age from the main body of the casting, for the whole length of the piston's travel in the cylinder. The connection of the pipes to the inlet and outlet branches is simplified, by bringing them both below the water jacket ; this leaves the FlG. 171. YORK MFG. CO. 'S MAMMOTH MACHINE, 400 TONS REFRIGERATING CAPACITY. head quite clear, and makes inspection of the valves an easy matter. No lantern bushes are shown in the piston rod packing, but the stuffing box is still longer than many engineers con- sider necessary or even desirable ; and oil is fed below the packing. This pump should be compared with that shown by Fig. 58, page 102; both aim high, but seek perfection in MACHINERY FOR REFRIGERATION. 283 different -ways, and both are better than the one illustrated by Fig-. 64, to which, nevertheless, the indebtedness of Fig. 58 for some ideas is gratefully acknowledged. The York Co., of York, Pa., have only so far been represented in this work as manufacturers of compound ammonia -compressors, and by Figs. 66 and 70. In Fig. 171 FlG. 172. SECTION YORK MFG. CO. 'S COMPRESSOR. there is a perspective view of a modern mammoth machine made by the same builders, and equal to 400 tons refrigera- tion. It has two single-acting compressors, thirty inches diam- eter, forty- eight-inch stroke, fitted with cross-compound con- densing steam engine; high pressure cylinder, thirty-inch bore; low pressure cylinder, fifty-eight-inch bore, forty-eight- inch stroke. The crank shaft has two throws and four bear- ings. The machine is fitted with one fly-wheel in the center 284 MACHINERY FOR REFRIGERATION. 1 IG. 173. PENNEY'S HORIZONTAL DOUBLE-ACTING COMPRESSOR, NEWBURGH ICE MACHINE AND ENGINE CO., NEWBURGH, N. Y. , U. S. A. FlG. 174. LATE REMINGTON MACHINE, WILMINGTON, DEL., U. S. A. MACHINERY FOR REFRIGERATION. 285 of the bed plate, between the two cranks. The weight of this machine when completed was 400,000 pounds. Fig-. 172 shows a section of the York vertical machine, late design. FlG. 175. SECTION OF REMINGTON MACHINE, WILMINGTON, DEL., U. S. A. For the reasons given on pag-es 119 and 120, straig-ht-line ammonia compressors have not been greatly favored in the past, althoug-h it is common enough for compressed air. In Fig-. 173, the Penney machine, made by the Newburg-h Ice Machine and Eng-ine Co., Newburg-h, N. Y., U. S. A., is a 286 MACHINERY FOR REFRIGERATION, MACHINERY FOR REFRIGERATION. 287 288 MACHINERY FOR REFRIGERATION. modern example of this type, and the heavy character of the fly-wheels, supporting- what is said on pages 123 and 124, is very clearly apparent. The Reming-ton vertical compressor, as shown in Fig's. 174 and 175, is of the single-acting-, inclosed crank type, and has but one stuffing- box, that on the revolving- shaft. The ordinary type of trunk piston is used, and the crank shaft is supplied with a center bearing- in order to provide for a rig-id construction, and at all times runs in oil. FlG. 178. SECTION AMERICAN LINDE COMPRESSOR CYLINDER, FRED W. WOLF CO., CHICAGO, U. S. A. There are two cylinders made in one casting- provided with heads in which are located the suction and discharge cag-es and valves. These cag-es and valves are readily acces- sible by removing- the cross-bars on top of the heads, without breaking- any other joints than those directly over the cag-e. The heads of the two cylinders are connected on the suc- tion side to a common strainer box for catching- the dirt and MACHINERY FOR REFRIGERATION. 289 sediment, and the discharge side to a throttle valve common to both cylinders. The American type of the Linde machine, as manufact- ured by the Fred W. Wolf Co., Chicago, shows many varia- tions in construction from the original Linde machine, as de- signed by Prof. Carl Linde, in 1875, the construction and operation of which have been thoroughly described and ex- plained in Chapter XV of this work. Illustrations of the latest type American Linde machine are inserted here to show these variations in construction. . These machines are of the ammonia compression type, operating on the humid gas system. Fig-s. 176 and 177 represent the "Standard" and "Tangye" styles of frames, the working parts of each, how- ever, being of the same general design and construction. Fig. 178 is a sectional view of the latest Wolf design of the Linde compressor cylinder. The refrigerating machine as built by the Vilter Manu- facturing Co., Milwaukee, Wis., U. S. A., illustrated by Figs. 179 and 180, consists of one or two horizontal double-acting ammonia compressors driven by one horizontal engine, gen- erally of the Corliss type, built also by the same firm. The engine and compressor cranks are keyed on the ends of the shaft at angles to each other, bringing the highest gas pressure in the compressor at a point where the engine gets the highest steam pressure. The ammonia compressor, as shown partly in section and partly in perspective, is generally cast with slides and pillow block in one piece. After the guides and frame that is, the water jacket of the compressor are bored, a cylindrical bush- ing is forced into the water jacket, forming the compressor wearing surface proper, and then a finishing cut is taken in one setting of the entire frame through the guides and com- pressor, for the purpose of making the guides absolutely true with the compressor. The four ammonia compressor valves are placed in the two circular heads. The heads fit into a recess, and are packed with a metallic packing. The suction and discharge valves are readily accessible, which, together with the stems, are made of forged steel, and (19) 290 MACHINERY FOR REFRIGERATION. MACHINERY FOR REFRIGERATION, 291 are provided with gas cushions, to avoid crystallization and noise in the working- of the valve. The valve seats are of cast steel, turned true, and fit with a ground -joint in the compressor head, making- the use of packing- unnecessary. The compressor plung-er is provided with self-adjust- ing- packing- ring's having- bull ring's besides, which can be replaced easily. The piston and follower are turned to a cir- cle to fit exactly into the front and back heads respectively of the compressor. The clearance between the plung-er and the head is thereby reduced to a minimum. The leng-th of the plung-er rod can be adjusted, so as to divide the clearance equally at both ends, and so take up the wear of the crank and cross-head boxes. The stuffing- box consists of a metallic packing- in the head, which is held in position by a long- hollow sleeve, throug-h which oil is circulated by means of an automatic oil pump, and this oil is used for lubrication of the plung-er rod, as well as for forming- a seal ag-ainst the escape of ammonia. The outer end of this-hollow sleeve is held in position by a separate sup- port, which is bolted to the compressor frame proper, and at the outer end of this support a packing- is provided for retaining- the oil. By this arrangement of the stuffing- box, it is claimed by the manufacturers that they are enabled to oper- ate the compressor with a very much hig-her discharg-e pres- sure than could otherwise be effected. Proper by-pass, or cross-connections, are placed between the suction and discharge pipes close to the compressor, so that the valves can be operated for pumping- out the con- denser without leaving- the engine room. The cross-heads are provided with adjustable shoes by wedge adjustment; the connecting- rods have solid heads, and the crank pin is provided with a brass box lined with babbit metal, the cross-head box being- of solid brass. The wear of both boxes is taken up by wedg-e adjustment. If two ammonia compressors are driven by one eng-ine, they are g-enerally arrang-ed tandem, and both connecting- rods are coupled to one crank pin. Larger compressors are often driven by compound non-condensing- or compound con- densing- engines, and if there is a scarcity of water, cooling- towers may be added, so that water can be cooled and re-used. 292 MACHINERY FOR REFRIGERATION. MACHINERY FOR REFRIGERATION. 293 The compressors may also be driven by belt or rope transmission from a line shaft operated by an engine, also doing- other work, or by electric motor, or by water power. Fig-. 181 shows a perspective view and Fig-. 182 a section of the compressor of the Triumph Ice Machine Co., of Cin- cinnati, Ohio, U. S. A. This compressor is of the horizontal double-acting type, and is fitted with five valves, three suc- tion and two discharge. The third, or auxiliary suction valve, is perfectly balanced, and is much lighter than the main suction valves. The main suction valves must of neces- sity be of sufficient size to admit the charge of gas quickly at the beginning of each stroke. The springs controlling them must therefore have an appreciable tension, and it can be readily seen that in consequence the pressure of the gas in the cylinder during admission is less in the suction pipe by just the tension of these springs. The construction of the suction valves is as follows: A guard is screwed on to the stem, fitted inside of the cage, and is ribbed so as to reduce the port area, the stem being made larger at the bottom for this purpose. Both suction and dis- charge valves have a stem leading to them through the stuff- ing box, and can be handled from the outside, thereby allow- ing any tension to be brought on the springs at any time. This, it is claimed, is necessary on account of machines being worked at different pressures and their relative tempera- tures. For instance, one side of a machine may be called upon to work on a temperature of 10 to 15 below zero, the other side to 10 or 15 C above, consequently the springs on one of them would have to be changed so as to make it oper- ate properly, and also enable the engineer to know by obser- vation whether they are opening and closing at the proper time, and whether they have the proper amount of lift, etc. The stuffing box has three compartments for packing, and is fitted with a relief valve, which leads into the suction. The piston is shrunk onto the piston rod, making it a perfect fit. The heads are concave, and of such a radius as to obtain a larger valve area. Every part of the compressor is accessible. The main shut-off valves are so constructed that they can be packed while the machine is in operation. 294 MACHINERY FOR REFRIGERATION. MACHINERY FOR REFRIGERATION. 295 296 MACHINERY FOR REFRIGERATION, An inspection of the great works of the Fresh Food and Ice Co., of Sydney, N. S. W., affords a splendid example of the progress that has been made during- the past few years in improving- and perfecting- refrig-erating- machinery in its various applications to the needs of commerce. The great FlG. 183. HERCULES MACHINE IN NEW SOUTH WALES FRESH FOOD AND ICE CO. 'S WORKS, SYDNEY, N. S. W. work of the late Mr. T. S. Mort (referred to in the historical chapter), is still in evidence, and is continually expanding. This company, of which two of Mr. Mort's sons are direc- tors, is the largest company of the kind in the southern hemisphere, and in its way unique. It runs refreshment rooms, and it has a large railway and shipping business, both MACHINERY FOR REFRIGERATION. 297 FlG. 184. LATEST DESIGN HERCULES LARGE COMPRESSOR. FlG. 185. LATEST DESIGN HERCULES MACHINE, STEAMSHIP PATTERN. 298 MACHINERY FOR REFRIGERATION. MACHINERY FOR REFRIGERATION. 299 domestic and foreign, in ice, fish, poultry, rabbits and hares. The New South Wales railways run into its premises, carry- ing- its own refrigerating vans. Its milk tanks, at head- quarters alone, have a capacity of over 30,000 gallons, and in one month it has shipped 80,000 frozen sheep to London. The freezing machinery of this company's principal works includes two compound compressing plants made under the Lock patents, one De La Vergne machine and one Auldjo machine, all four machines being built by the Morts Dock Co., of Sydney. There is also one De La Vergne ma- chine, made in New York, besides the latest addition to the plant, which is a 70-ton Hercules machine, with compound tandem engine. Fig. 183 is a view from one angle of the principal engine room. The Hercules machine is the promi- nent feature, while the De La Vergne machines will be noticed in the rear. Since the World's Fair at Chicago, Australia and New Zealand have been so well exploited in the interests of the Hercules machine, that it is now running in much greater numbers than other makes of refrigerating plants throughout the colonies, and as Figs. 72 and 73 are only diagrams, more justice is done to its importance by this later illustration and Figs. 184 and 185, the former showing the latest design of large machines, the latter the steamship pattern, as made by C. A. MacDonald, Chicago and Sydney. The illustration No. 186 deserves notice for several rea- sons. It is a perspective of a gigantic machine, as shown by the comparative size of the men alongside, and is rated as equal to 725 tons refrigeration. The builders are the Ice and Cold Machine Co., of St. Louis, Mo., and their design pre- sents differences in detail from any machine so far referred to. It is a straight line machine, but it embodies the arrangement advocated on pages 130 and 131 with a right-angled connec- tion in having a straight shaft and only two bearings; the fly-wheel being in the center, and a crank at each end. Where it differs from the average straight line machine is in the adoption of cross-heads, guides and connecting rods, to both the steam and ammonia ends; two connecting rods, side by side, being connected to the same crank pin. This, of course, adds to the length of the machine, and increases the frictional losses; but there are, no doubt, good and 300 MACHINERY FOR REFRIGERATION. FlG. 187. BUFFALO REFRIGERATING MACHINE CO. 'S COMPRESSOR, BUFFALO, N. Y., U. S. A. MACHINERY FOR REFRIGERATION. 301 weighty reasons for adopting- this arrangement, instead of the more usual one of connecting up the steam and compres- sor pistons to one cross-head only, and with one connecting rod to each side. One of these reasons is obvious, and that is, it enables either of the ammonia cylinders to be discon- FlG. 188. SECTION BUFFALO REFRIGERATING MACHINE CO. 'S COMPRESSOR CYLINDER, BUFFALO, N. Y., U. S. A. nected without requiring the engine on the same side to be stopped also. A further reference to the cut will show that the valve is located on the cylinder, being a gravity valve without springs. It works over a plunger, the cushion of gas for closing same being regulated by a needle valve, which regu- 302 MACHINERY FOR REFRIGERATION. lates the compression or vacuum in the chamber formed by the valve and plunger, the suction valve being- directly oppo- site to the discharge valve (see Fig. 165). Fig. 187 is a perspective view of the latest design 25-ton vertical straight line machine, manufactured by the Buffalo Refrigerating Machine Co., Buffalo, N. Y. It is double-act- ing. The ammonia compressor and steam cylinder are in alignment and bolted to a rigid cast iron frame, mounted on a heavy and substantial bed plate, in one piece. The machine is therefore self-contained. The form of construction, as illustrated by this machine, has been exhaustively treated in Chapter XV of this work. This compressor may be operated by an engine of either the slide valve, automatic cut-off or Corliss pattern. The clearance in cylinder is reduced to a minimum. The piston is provided with patented self-adjusting packing rings, one at the top and one at the bottom end. The pressure of the ammonia gas acting upon the conical sur- face of the ring expands the same in all directions outward against the wall of the cylinder, forming a perfectly tight joint. The pressure and suction valves are of ample area to handle the gas without wire-drawing, and their construction is such that they leave but little or no useless space inside of the cylinder, in which the compressed gas can collect. The valves are made of forged steel, case-hardened on seats, and are ground to a perfect seat. They hava long guiding sur- faces and are arranged with cushioning chambers to relieve them from undue strain, prevent slamming and bring them gently and noiselessly to their seat. The stem of suction valve is provided at the bottom with a collar, which prevents the valve from dropping into the cylinder in case the nut on top of the valve stem should get loose. The cages in which the valves work are made of cast steel, and are so arranged, as will be seen in the section Fig. 189, that they can quickly and easily be removed and replaced without disturbing any other connection. The stuffing box is long and is so arranged that between the upper and lower packing an oil chamber is provided, which is automatically supplied with oil from the oil tank, as shown MACHINERY FOR REFRIGERATION. 303 FlG. 189. BOYLE COMPRESSOR, PENNSYLVANIA IRON WORKS CO. PHILADELPHIA, U. S. A. 304 MACHINERY FOR REFRIGERATION. in cut. The operation of the oil arrangement is as follows: The oil tank is supplied as often as necessary with oil by the hand pump attached to the tank. The lower end of the tank is connected to the lower part of oil chamber in the stuffing FlG. 190. BOYLE SINGLE-ACTING MACHINE WITH VERTICAL ENGINE, PENNSYLVANIA IRON WORKS CO., PHILADELPHIA, U. S. A. box, and the upper end of chamber is connected to the upper end of oil tank; a connection also is made from upper end of oil tank to suction valve of machine. By this means the oil in oil tank will be under suction pressure of the ammonia gas MACHINERY FOR REFRIGERATION. 305 on the top and bottom side, and as the oil tank is placed above the oil chamber in stuffing- box, the oil will flow into the latter by its own gravity, and any leakag-e of ammonia from the ammonia cylinder throug-h the first layers of packing- into the FlG. 191. BOYLE SINGLE-ACTING MACHINE WITH HORIZONTAL ENGINE, PENNSYLVANIA IRON WORKS CO., PHILADELPHIA, U. S. A. oil chamber of the stuffing- box will be drawn into the suction pipe of the machine, and consequently the pressure in stuffing- box, it is claimed, can never exceed the suction pressure under which the machine is working-. The quantity of the (20) 306 MACHINERY FOR REFRIGERATION. oil fed to stuffing- box is regulated by the valves on pipes communicating' with the oil tank. The oil, adhering 1 to the piston rod , finds its way into the cylinder in sufficient quantity to lubricate same. The gas cylinder and its top and bottom head is sur- rounded by a water jacket for removing the heat of com- FlG. 192. TYPE OF ARCTIC AMERICAN MACHINE BUILT IN 1879. pression as far as possible. The clearance space between the cylinder heads and piston is reduced to a minimum, only the thickness of the sheet packing being allowed. Reference is made on page 132 to the Boyle vertical ma- chine, as built some twenty years ago, in comparison with same machine as built by the Pennsylvania Iron Works Co., Philadelphia, U. S. A., and illustrated on page 133. MACHINERY FOR REFRIGERATION. 307 The distinctive features of this modern type of the Boyle machine lie in two vertical single-acting- compressors in com- bination with either a vertical or horizontal eng-ine. See Fig-. 189, being- a section of the cylinder, and Figs. 190 and 191, showing- single-acting- compressors, with vertical and hori- zontal engines, respectively. The compressor valves are inclosed in removable cag-es, both suction and discharg-e of which are located in the upper FlG. 193. SECTION ARCTIC AMERICAN MACHINE AS BUILT IN 1900. head, being- held in position by cross-bars and a single set screw r on top of each, the head having- a division in the center, the g-as entering- throug-h its pipe, which is screwed into a pocket extending- from the body of the cylinder, and commu- nicating- with the inlet chamber, and throug-h its valve enter- ing- the cylinder during- the downward stroke of the piston. Upon the return stroke the g-as is compressed until it equals 308 MACHINERY FOR REFRIGERATION. the pressure within the condenser, the discharge valve in the opposite side of the head then lifting- and allowing the discharge of the gas through the valve and communicating chamber to the discharge pipe from the compressor. The suction chamber also communicates with the lower end of the compressor cylinder, filling the same with gas during the upward stroke of the piston, and allowing its exit during the downward stroke thereof. The upper portion of the cylinder is surrounded by water jacket, having inlet and outlet open- ings provided for the flow of water from the same, taking up a portion of the heat due to the compression of the gas, and keeping the compression valves and different parts at a tem- perature to not interfere with their proper operation. The compressor piston is of the solid type, having a number of snap rings, the tension of which makes them tight enough to prevent the leakage of gas past them. The stuffing box has an evaporating pressure only upon it, is easily kept tight, and consequently there is little wear or tear on the rod. Owing to the single-acting feature, the clearance can be reduced to a minimum, the compressor piston traveling as close to the head as possible without touching same. The piston and valves being perfectly balanced also, exert no side wear upon the cylinder or other parts, and present the most desirable features for continued service. The Arctic machine, built at Cleveland, Ohio, U. S. A., has been on the market since 1879, and machines of that date are still in use. It was one of the first of this class of ma- chines to come into general use in America. Fig. 192 is a per- spective of an Arctic machine, as built in 1879; Fig. 193 is a section of machine, as built in 1900, by the Arctic Machine Co. The machine is of the double-acting ammonia com- pressor type, built either with a vertical steam engine and vertical compressor or with a horizontal steam cylinder and two vertical compressors. The construction of this machine has changed in many ways, /. ., the large fly-wheel has given place to one more in proportion, and usually placed between the columns; when placed on the outside the shaft has an outside bearing. The compressor valves are now fitted in cages; formerly the head MACHINERY FOR REFRIGERATION. 309 of compressors had to be removed to get at them. The stuff- ing- box of compressor is deeper and fitted with oil sleeves. Corliss valve motion has taken place of the slide valve, while connecting- rods, cross-heads, piston, etc., have been made to conform to modern practice. FlG. 194. SECTION OF CASE COMPRESSOR, CASE REFRIGERATING MACHINE CO., BUFFALO, N. Y. Fig. 194 is an illustration of a refrigerating- machine built by the Case Refrigerating Machine Co., of Buffalo, N. Y. These machines are of heavy build, and occupy comparatively small floor space. The peculiarity of the construction of the 310 MACHINERY FOR REFRIGERATION. machine is that both the steam and compression cylinder piston rods are connected to the same cross-head, which works between the two cylinders. The steam cylinder is below and in a direct line with the compression cylinder. This allows a direct push and pull on the piston rods, calcu- lated to remove all strain from the crank shaft and connect- ing- rods, and thus reduce the friction to a minimum. A water jacket surrounds the compression cylinders, where a small stream of water is kept running- to cool the compressor when in motion. FlG. 195. THE BARBER TYPE AMERICAN COMPRESSION MACHINE. The compressor suction and discharg-e valves work hori- zontally, which allows a very small pocket for compressed g-as, and reduces the clearance to a minimum. The machine illustrated by Fig-. 195 is manufactured by the A. H. Barber Manufacturing- Co., Chicag-o, U. S. A. It is of the horizontal type, with a double-acting compressor. It is built with a box frame, with a center crank for those run by belt, and a tang-ye frame or side crank when directly con- nected to Corliss or slide valve eng-ine. The shaft, pulley and fly-wheels are all in proportion to the size of the compressor. The cylinders are let down into the frame. A flat locomotive g-uide is used, thereby g-iving- the machine a deep and rig-id frame, so that it is impossible for the cylinder to g-et out of line. MACHINERY FOR REFRIGERATION. 311 The cylinder and valv 7 es are entirely surrounded by water. The valves and seats are made of tool steel, and both are hardened so as to prevent pitting" and to increase their wear- ing- efficiency to the utmost. The valves are easily removed for inspection, without breaking- or disturbing- any other joint. The piston is made as lig-ht as possible, and provided with metallic packing- ring's. The stuffing- box is perfectly sealed, having- a double packing-, with an oil chamber in the center. The lubricator is so arrang-ed that it oils the cylin- der, valves and piston rod. In the suction conduit, close to the compressor, is placed a strainer, which prevents scales from the system getting into the compressor. The clearance is reduced to a mini- mum, and the connecting- rod is so arranged that any wearing on the crank shaft or g-uide can be easily adjusted. The Challoner machine, illustrated by Fig's. 196, 197 and 198, belong-s to the inclosed type described in a former chapter. The frame is cast in one piece, having- two heavy ribbed flang-es, and secured to a bed plate cast in one piece, strongly arched where frame rests upon it. Each end of the frame is provided with heavy circular removable flanges containing long- babbitted bearings for the crank shaft to rest in, and extra long stuffing boxes with glands and nuts to prevent any leakage of gas or oil around shaft. Within the frame the bearings for the crank shaft are bolted in place so as to be readily removable. The case is supplied with a charge of oil, so that all working parts run in same, to insure lubrication without exterior oilers or lubricators. The top of the case is faced off true and bored out to receive the compressor cyl- inders, which are sleeve castings and can be removed and replaced in case of necessity without renewing the case or frame. The crank shaft for the small machines is a solid steel casting, while for the larger sized machines it is a solid steel casting for each outboard end, with center crank of cast steel, all put together with turned faced male and female joints, securely bolted with turned bolts in reamed holes. The connected parts of shaft form a very large bearing sur- face in the journals inside the case, to insure permanent 312 MACHINERY FOR REFRIGERATION. alignment of the shaft and minimum wear of the journals. The larger sizes of the machines are provided with two and the smaller sizes with one extra heavy large band fly-wheels. The connecting rods are made of hammered iron, the upper ends being bored out and provided with hardened steel sleeves for bearings on the piston pins. The lower FlG. 196. THE CHALLONER TRIPLE CYLINDER, SINGLE-ACTING COMPRESSOR, GEO. CHALLONER 'S SONS CO., OSHKOSH, WIS., U. S. A. ends of the rods are threaded and passed through the yokes of the crank stubs, having keys in the stub yokes to prevent turning of the rod out of line with the crank and piston pins. The larger size of machines are provided with safety heads, the action of which, should the pistons be set too close to the heads, would be to lift and prevent possibility of knocking out a head. MACHINERY FOR REFRIGERATION. 313 The suction valves are placed in the pistons, and the dis- charg-e valves are placed in the safety heads or in the false heads. Both the suction and discharge valves may be re- moved by taking- off the pump heads and without disconnect- ing- pipe connections. FlG. 197. THE CHALLONER TRIPLE CYLINDER, SINGLE-ACTING COMPRESSOR SECTION. The suction connection is made to the case below the cylinders so that the case is kept cool by the return of the low temperature g"as. The discharge connections are made to the cylinders above the safety heads. Both connections are provided with suitable stop valves, and by-pass connection is arrang-ed so that the machine can readily be reversed to pump the g-as from the high pressure to the low pressure 314 MACHINERY FOR REFRIGERATION. side of system. A purge valve is also placed on the discharge connection so that the case may be pumped out for opening and examining, merely with a few turns of the crank shaft, and all air entering the case when opened can be discharged to the atmosphere, thereby preventing accumulation of per- manent gases in the system. FlG. 198. THE CHALLONER TRIPLE CYLINDER, SINGLE-ACTING COMPRESSOR, END VIEW. On page 124 reference is made to the many devices that have been brought forward by the inventive skill of engineers, for the purpose of distributing the work of a compressor piston more evenly, in relation to the motive power. In Fig. 199 there is an elevation of a quite recent design, known as the "Ideal" machine, and built by the Ideal Refrigerating and Manufacturing Co., of Chicago. It will MACHINERY FOR REFRIGERATION. 315 be seen that the diameter of the crank pin circle is much greater than the stroke of the piston; and that, as the con- FlG. 199. SECTION OF IDEAL REFRIGERATING AND MANUFACTURING CO. 'S MACHINE, CHICAGO, U. S. A. necting- rod pulls the two members of the toggle joint toward the vertical position (where the joint pins are in a straig-ht line) the force available to move the piston gradually 316 MACHINERY FOR REFRIGERATION. approaches the infinite, apart from the toggle action of the crank itself. Fig-. 200 shows graphically how the moments of resistance to the turning of the crank shaft differ from an ordinary direct connection; the cam-shaped diagram drawn in full line representing, in the radial lengths from the crank pin circle, the resultants from the theoretical compressor diagram above. The dotted figure corresponds with that resulting from a direct connection from the crank to the piston rod cross-head, like Figs. 98 and 99. As an effect of this toggle action it is apparent that the first half of the compressing stroke, from position to posi- tion 3, is made by the piston, while the crank pin travels /Ing. le A.refiresent5 morion of crank ftrfa/f stroke of Piston Angle B represents motion of crank /or one sufti part of Piston sfrote . 200. DIAGRAM SHOWING EFFECT OF THE MOTION IN ' ' IDEAL ' ' MACHINE. through less than one-sixth of a revolution; but that during the second half of the piston's stroke, from position 3 to 6 (where most of the work is concentrated), the crank pin moves through rather more than two-sixths of its course. The manufacturers of this machine claim to know from actual experience that the effect the intermitting motion of the cam-head on piston has on the valve is to prolong the life of same more than double, compared to that in an ordinary crank motion machine. This is due, it is argued, to the pro- longed stop caused by the crank passing over the dead center, and the toggle being in a straight line with the piston rod at the same time. The advantage claimed is that it gives the valve ample time to get seated, and all the gas is discharged MACHINERY FOR REFRIGERATION, 317 from the cylinder, and not drawn back into it on the return stroke, thereby developing- a very high efficiency. Owing to the throw of the crank being- greater than the stroke of the piston, the stress, and, therefore, the friction and wear on the crank pin, is reduced to that extent by the tog-gle device. Fig-s. 201, 202 and 203 illustrate the class "A" and class "B" "Vulcan " refrigerating and ice making machines, man- 318 MACHINERY FOR REFRIGERATION. ufactured by the Vulcan Iron Works, San Francisco, Cal. These machines are furnished in sizes up to ten tons re- FIG. 202. "VULCAN" COMPRESSOR, CLASS "A." frig-crating- capacity. Machines of larg-er sizes are of the horizontal double-acting- type (class U D"). MACHINERY FOR REFRIGERATION. 319 These special styles were designed to meet the de- mands for small machines that would embody simplicity of desig-n, construction and operation, and include a number of distinctive features. The compressor is vertical, single-acting-, and of the in- closed type, the working- parts being- automatically lubricated FIG. 203. "VULCAN" COMPRESSOR, CLASS "B," WITH STEAM ENGINE. by the oil in body of machine. (See sectional view, Fig-. 201.) The cylinder opens into crank chamber, the sides of which form the supporting- frame, thereby bringing- the cylinder and shaft close tog-ether, and doing- away with the long- con- nections otherwise made necessary by piston rods and cross- heads, and making- this compressor a compact, strong- and 320 MACHINERY FOR REFRIGERATION. accessible machine. The crank is forged on end of heavy steel shaft, which passes through stuffing box in side of crank chamber. The crank pin is of special construction, having hardened steel sleeve held in place by collar. The piston is operated by a crank working in box that slides in a yoke that is made part of the piston, the yoke having a guide at bottom, and being guided by the piston at the top. The crank cham- ber is provided with a removable cover. When machine is in operation the body of machine is filled with oil to a point just above stuffing box of crank shaft, the height of the oil being shown by a gauge glass, the oil acting as a lubricant for the moving parts and also as a seal for stuffing box of crank shaft. The body of machine is separated from the ammonia cylinder by a dividing or packing ring (R), through which the trunk or piston works, in such manner as to admit only sufficient oil to lubricate cylinder. The suction valve, which is fitted with a safety cage, is placed in center of piston (the ammonia gas enters body of machine below piston) and the discharge valve is placed in cylinder head. A dirt trap (U) is attached to each compressor body, into which the ammonia suction pipe discharges, intended to prevent the passage into ammonia cylinder, of any scale, dirt, etc. The relief valve (S) is for convenience in starting the machine. Wearing parts are supplied with removable bushings. A class A machine has only one ammonia cylinder. (See Fig. 201.) A class B machine has duplex ammonia cylinders of the class A type. The class A machines are self-contained, /. ., the ammo- nia compressor, ammonia condenser, steam engine, ammonia and oil receivers, ammonia gauges and pipe connections (also brine pump, if required,) are all placed on one bed plate, thus making a very compact arrangement, and especially suitable for use on steamships. Figs. 204 and 205 show in perspective and cross-section the construction of the Stallman compressor, manufactured by the Creamery Package Manufacturing Co., of Chicago, 111. This machine is of the vertical single-acting, water-jacketed type, and made in sizes from two to ten tons refrigerating ca- pacity. The lower parts of the cylinders are cored out, so as MACHINERY FOR REFRIGERATION. 321 to form a series of ports leading- from the suction inlet around the piston and into the cylinders, when the pistons are at the bottom or limit of their downward stroke. The filling- of the cylinders having- been partially effected by the passing- of the g-as throug-h the suction valves in the pistons during- their downward stroke, is thus at the very end of the stroke fully FlG. 204. PERSPECTIVE VIEW STALLMAN COMPRESSOR, CREAMERY PACKAGE MFG. CO., CHICAGO, U. S. A. completed, and the full evaporating- pressure secured in the cylinders by the unobstructed passing 1 of the g-as throug-h these ports. The upper part of the cylinder is enlarg-ed, and upon the shoulder thus formed rests the discharge valve seat, which is made of tool steel and is pressed into position before the finishing- cut is taken. It is then bored out with (21) 322 MACHINERY FOR REFRIGERATION. the cylinder and forms a part of the cylinder wall. Immedi- ately above the valve seat, connected with the enlarged part of the cylinder and branching- off at rig-ht angles, is the out- let port, which receives the discharge pipe. The discharg-e valve is made of steel. It is turned up from the solid, with a disc-like bottom larger than the bore of the cylinder, thus extending over and resting- upon the tool steel seat above described. On its upward stroke the piston passes through the dis- charge valve seat and comes into metallic contact with the valve itself, discharging completely the contents of the cylin- der past the valve, and leaving no gas to re-expand. There is therefore absolutely no clearance and consequently no loss of efficiency from this source. The valves, being large, have but slight movements and practically instantaneous action, and at the same time give very large areas of openings that permit the rapid passing of large volumes of gas. The valve and cylinder construction of this compressor should give the maximum results for power expended. Attached to and forming part of the discharge valve is a band-like extension that takes the place of a valve stem, the enlarged portion of the cylinder forming the guide for the valve. In the center of the discharge valve is a boss or cen- ter, around which is placed a spiral spring. This spring is provided with a screw, passing through the cylinder head, for adjusting its tension, not shown by cut. The piston is fitted with cast iron snap rings, turned to bore of cylinders. In the shell of the piston is the suction valve guide, held in position by the steel valve seat, which is threaded to and surrounds the upper part of the piston shell. The cylinders are mounted upon frames containing the shaft bearings and guides to bring all strains directly upon the frames and not upon bearings in a separate bed plate; in this construction the rigidity of the alignment is assured. A heavy box pattern bed plate securely ties the frames in position, making a compact and yet convenient arrangement throughout. The construction permits of operating the compressors independent of each other where conditions of varying tern- MACHINERY FOR REFRIGERATION. 323 peratures and consequent varying back pressures prevail, such as in plants for both ice making- and refrigerating-, and FlG. 205. CROSS-SECTION STAT.LMAN COMPRESSOR, CREAMERY PACKAGE MFG. CO., CHICAGO, U. S. A. where freezing rooms are used in connection with ordinary cold storage. Independent suction connections can be made to the compressors under such circumstances. 324 MACHINERY FOR REFRIGERATION. MACHINERY FOR REFRIGERATION. 325 FURTHER REMARKS CONCERNING WATER TUBE BOILERS. It would be difficult to decide whether the water tube boiler is at present creating- a greater revolution at sea or on shore; possibly it is more so in connection with steam vessels. One of the many lines of mail steamers trading- to Sydney (now said to be the fourth most important port for shipping in the world) has been carrying- the Belleville boilers for years, and they have obtained a footing- in the En g-lish, Amer- ican and foreig-n navies. For present purposes we are more concerned with land types, and as no illustrations appear in the sub-section com- FlG. 207. FIRE TUBE AS AFFECTED BY SOOT AND DIRT. FlG. 208. WATER TUBE AS AF- FECTED BY SOOT AND DIRT. mencing- on pag-e 205, Fig-. 206 will g-ive a good idea of how nearly every inch of such boilers is utilized for heating sur- face. In this figure the removable covers for scaling the tubes are clearly seen, as well as the doors in the brick work to enable the sooty deposit to be removed from their external surfaces. As Figs. 122 and 123 illustrate the advantages of fire tubes in connection with the deposit of scale, it is only fair to g-ive an illustration by which the advocates of water tube boilers show their great advantages with regard to the de- posit of soot and dirt from the fire. It is maintained, and is no doubt true, that if the draft is weak and the ordinary tubes are not attended to, they 326 MACHINERY FOR REFRIGERATION. h * FlG. 209. EVAPORATOR FOR WATER HEAVILY CHARGED WITH MINERALS. FIG. 210. FIG. 211. END ELEVATION AND SECTION OF (SO CALLED) SCOTCH BOILER. MAgHINERY FOR REFRIGERATION. 327 will practically soot up completely in time; while only a lim- ited amount of soot will lodge on the water tube, whether it is looked after or not. (See Fig's. 207 and 208.) In fitting- up a large ice factory with water tube boilers, and modern distilling plant, there would be no difficulty (with apparatus such as that described in Chapter XIX) in insuring the tubes being kept absolutely free from deposit, by first evaporating all the water supplied as feed. At sea special provision is now made for supplying the " make up " as it is termed, and one of the evaporators-used would be ap- plicable for smaller installations. Fig. 209 is an evaporator, so constructed that when the copper steam coils are coated up on the outside, by the salts of lime, magnesia or other mineral removed from the water, they can be swung right clear out of the casing for easy cleaning. Internally fired boilers with return tubes have many ad- vocates, and although primarily a marine type, they are much appreciated in many factories on land. In the United States they seem to have gotten the name of "Scotch" boilers why is not very clear, as they are not so called in England or Aus- tralia. Figs. 210 and 211 represent two views of this type, designed for land use with a good chimney draft (for sea the proportion is generally much shorter in relation to the dia- meter). It is evident that with such boilers in an ice factory, near to brine tanks or cold rooms, their shells should be well covered with the best non-conducting composition, to prevent the radiation of heat. Figs. 212 and 213 are sections of R. Munroe & Sons' safety and vertical water tube boilers, made at Pittsburg, Pa., U. S. A., and widely used in America. All the plates used in the construction of this boiler are made of open hearth homogeneous flange steel. There are two water chambers made in exact duplicate of each other, the outer heads of which are dished. The outer head of the front chamber contains from two to six patented eclipse man- holes according to the size of the boiler. These manheads permit ,of free access to all of the horizontal tubes. The thickness of the material used in these water chambers varies from five-sixteenths to seven-sixteenths, according to the size of the boiler. The chambers are made extra strong- by being 328 MACHINERY FOR REFRIGERATION. MACHINERY FOR REFRIGERATION. 329 FIG. 213. MUNROE'S VERTICAL WATER TUBE BOILER, AMERICAN TYPE. 330 MACHINERY FOR REFRIGERATION. double or triple staggered, riveted at the point where the sheets are lapped. The tube sheets of water chambers are five-eighths of an inch thick, and made of homogeneous flange steel. Riveted on to the outer heads of the water chambers are from four to eight angle braces, according to the size of the boiler, and connected to the braces are connecting or tension rods. These rods are made of soft iron, and they are connected to the braces by pins varying from three-fourths of an inch to one and one-quarter inches in diameter. The tension rods are from one inch to one and one-half inches in diameter; they have a swivel on one end, so that they can be easily removed at any time for cleaning or repairs. The steam and water drum is made of homogeneous flange steel, same as the water chambers, the outer- heads of the drum being dished, the front or rear head containing a manhole, placed in the head opposite to stack. The water legs are made of the same material as the drum and chambers, and they are of ample size to meet the requirements of the various boilers, the front leg being larger than the rear, so as to allow a large liberating surface. The water legs are double riveted at their flanges to the water chambers and steam and water drums. The horizontal tubes are four inches in diameter, and vary in length according to the size of the boiler. They are placed in a staggered position in the tube sheets and ex- panded with a Dudgeon expander, then turned over. The boiler is set up on brick work and suspended bv heavy cast iron lugs riveted at their proper angles on to the water chambers. Each boiler has four lugs, two on each water chamber. Over the grate bars and under front water chamber is placed an arch, and immediately over the tubes there is another arch made of fire brick; it is built from side wall to side wall of the boiler, and it runs from the front water leg to a point, changing according to the size of the boiler, so as to allow a sufficient draft area. Resting on the top row of drop or circulating tubes is the tile arch to form a cover for the draft area; the other end of the arch resting on the top of the side walls. The pipes in side walls are of a diameter to admit sufficient oxygen to facilitate combustion. As the tubes MACHINERY FOR REFRIGERATION. 331 are on an angle of one and one-fourth inches to the foot, good results must be obtained, as the heat units impinge directly on the tubes, and as the tubes are staggered, the heat units are distributed over the entire heating surface. Ten square feet of heating surface is allowed per horse power. The water is fed through a pipe which is connected to the steam and water drum at a point directly opposite the center of the rear water leg, and which extends almost to the center of the rear water chamber, and steam is taken out of the opposite end. THE HOLDEN ICE MAKING SYSTEM. Although nothing has heretofore been said as to the relative merits of the can and plate systems of ice making, as they form quite a separate question from those that have been discussed, .the opportunity may be here taken to describe an entirelv different system of ice making, which has been recently introduced by Mr. D. L. Holden, of Phila- delphia, U. S. A., with what ultimate success remains to be seen. Reference was made on pages 71 and 74 to the effect of velocity and thickness of material in its effect on the conduc- tion of heat. The effect of the low conductivity of ice in reducing the ratio at which it forms, either in the can or on the plate, is well known and exercises a great influence in restricting the thickness of the blocks as made in actual practice. Mr. Holden makes a wide departure from the usual method (although his ideas are not new as applied on a small scale for ice cream freezing), and constructs his refrig- erator as a hollow cvlinder, which rotates in the water to be frozen. The liquid ammonia is carried in through one of the trunnions, and the other one is connected either to the re-absorber in the case of an absorption plant, or to the suc- tion side of the compressor in a compression system; and the evaporation of the ammonia in this cylinder freezes the water in contact with its metallic surface very rapidly. According to accounts appearing in contemporary jour- nals, this rate of freezing is so fast that it would incrust the cylinder at the rate of a quarter of an inch per minute; but as soon as it is formed, it is cut or shaved off by a set of rotary 332 MACHINERY FOR REFRIGERATION. knives, and these ice shaving's are carried, by a creeper or conveyor to hydraulic presses. When inclosed by brass molds and under a pressure of between 300 and 400 pounds to the inch, this mush ice is so compressed that all the water and air is got rid of, and regulation is brought about. This regulation produces blocks of compact and solid ice, but whether by sufficient pressure crystal clear ice can be thus made is not stated. It is claimed, however, that ice can be made cheaper in this manner. If the process really turns out to be cheaper, then it must be so by the saving- effected in the plant labor and accessories connected with the actual ice making-. It certainly cannot make more ice with a given amount of refrigerating effect as produced by a machine, than an ordinary refrigerating 1 tank and metal molds can do, if there is careful insulation and absence of waste in thawing out. IN CONCLUSION. In taking- leave of the reader the author would say that, in commencing this task he had no idea that his original paper would expand to the dimensions this work has now assumed, but he is now quite aware that there are sufficient interesting- matters omitted in connection with the machinery of refrig-eration to make another volume. He cannot let the opportunity pass without expressing- his obligation to the publishers for the handsome dress they have put him into, and for their artistic reproduction of his original drawings. To builders of refrigerating machinery who have kindly forwarded him catalogues and information he here expresses his obligation, and as gratitude is said to be "a lively sense of favors to come," he hopes to be the recipient of any new editions that may be published. To such builders or mana- gers of refrigerating machinerv as may be numbered among the readers of the ideas herein set forth, he would say that any information or suggestions in connection with that department of engineering, with which they may favor him, will be much appreciated and duly acknowledged. APPENDIX I TABLES. 334 MACHINERY FOR REFRIGERATION. TABLE SHOWING THE MEAN PRESSURE OF STEAM IN CYLINDERS OF COMPOUND AND TRIPLE- EXPANSION ENGINES. WITH VARIOUS INITIAL STEAM PRESSURES, EXPANDING DOWN TO A NOMINAL TERMINAL PRESSURE OF 15 I.BS. PER SQUARE INCH. Points ABSOLUTE PRESSURE. of cut-off Mean Ratio of Expansion, or number of times steam is expanded. Hyperbolic Logarithms of the Ratio of Expansion. of the Stroke, reckoned from the commence- ment. pressure dur- i.ig- the stroke, the initial pressure being- taken as = 1. Initial pressure in Ib. per square suitable for given ratio of Mean pressure in Ib. per square inch. expansion. 6 1.7918 j 0.4653 90 41.8 6M 1.8326 2 4 5 0.4532 93.75 42.4 6% 1.8718 1 2 3 . 4418 97.5 43 1.9095 2 4 7 0.4310 101.25 43.4 7 1.9459 1 0.4208 105 44 7M 1.9810 2 4 9 4111 108.75 44.6 7% 2.0149 A 0.4002 112.5 . 45 7% 2.0477 0.3932 116.25 45.6 8 2.0794 1 8 0.3849 120 46 8M 2.1102 0.3779 122.75 46.3 8% 2.1401 1 2 7 0.3694 127.5 47 2.1691 3 4 5 0.3621 131.25 47.5 9 2.1972 9 0.3552 135 47.9 9M 2.2246 0.3486 138.75 48.3 9% 2.2513 1 2 9 0.3122 142.5 48.7 9% 2.2773 0.3361 146.25 49 10 2.3026 0.3302 150 49.5 ioM 2.3279 --T 0.3246 153.75 49.8 10% 2.3513 2\ 0.3191 157.5 50.2 10% 2.3749 0.3139 161.25 50.6 11 2.3979 A 0.3089 165 50.9 HM 2.4201 4 4 S . 3010 168.75 51.2 11% 2.4430 0.2993 172.5 51.6 11% 2 . 4636 ? 0.2947 176.25 51.9 12 2.4849 0.2904 180 52.2 12M 2.5052 * 0.2861 183.75 52.3 12% 2.5262 0.2821 187.5 52.8 12% 2.5455 J\ 0.2780 191.25 53 13 2.5649 jig 0.2742 195 53.4 1334 2.5840 JL 0.2704 198.75 53.8 13\4 2.6027 2 2 7 0.2668 202.5 54 13% 2.6211 5 4 5 0.2633 206.25 54.2 14 2 . 6391 1 0.2599 210 54.5 14M 2.6567 5 4 7 0.2566 213.75 54.8 14% 2.6740 2 2 9 0.2533 217.5 55 14% 2.6913 5 4 9 0.2502 221.25 55.3 15 2.7081 0.2472 225 55.6 15% 2.7408 g 2j 0.2412 232.5 56 16 2.7726 A 0.2358 240 56.5 MACHINERY FOR REFRIGERATION. 335 ui ^ O CI u~, l^ O sO 01 X uc xo sO O l^ c iO t^ ^ >C ^ LC ^ LC O T ^ ^ iC ^, iC X 1C O X i/3 !> <*; C Cl 1C 1^. O C) ir, t^ c O C) C: K ^ ri C t^ X a >- Cl ro u; C t^ X M iC ^: c -^ l^ i-J ^if-ix r crifOTi-oxo s ccr-r-i r^ i-< LC 01 v Ol Cirir*5rCf r 3-*'*iCi LC o vc o ic c 01 c i> LC 01 ic X^ONOHfO _C1 Cl 01 Cl \J^ O LO O Ol LC I> O 01 rf tt 01 Ol Cl Cl 3 o o o o Cl TJ- C X o o o o o o o o M C -*- C X M ^* t^XXO s OWr r 3^ rt I ' O H rO LC olclolol re Tf TJ- TJ- r. C 2 i^ S3S S**'^'^^***' ^' ^^ **^o **^ St>XO v OT-OlTj-LCvOXCCl'+sCXC;>COOlX^l-OOCl ^ ^H rl ^ rH rl rH Cl Cl Ol Ol Ol r^ f*3 ^ "t TJ- Uj vC sC t- saqoui oi 93^0 j;s 336 MACHINERY FOR REFRIGERATION. PERCENTAGE OF SAVING OF FUEL BY HEATING FEED WATER. The following- table gives the percentage of saving- for a steam pres- sure of sixty pounds per square inch, with various initial and final temperatures of the feed: CTs < 1> ^5 10 rj- ro Cl H rH ri O ON GO r^ MD VO 10 O rH c GO O^ O^ O O rH* C^ CO O\ o M M ^^iO C jo OrH dM OrHC^ C'IOn NOrHc s i rinrl rl- ro 'Of s lCia\l^iOrOrHGOvOr r 5C5 rH O Cl O\ t^ & if) ^t CO rH O ON GO OOrHC-l nOlClC) <<* 10 v ^ J> 00 O\ O rH O rH riri '* Cq ooooooooooooriooooooooooo rHrHC^CO^l-lOv^t^GOO>OrHrHr)r<0-^-O'Or^X<^OrHr^ MACHINERY FOR REFRIGERATION. 337 WEIGHT OF CAST IRON PIPES IN POUNDS PER LINEAL FOOT. si fcl THICK .NESS OF METAL. y ~Q M inch. ^inch. V* inch. H inch. Kinch. y inch. 1 inch. \y& inch. 1% inch. Inch. 1 Pounds. 3.06 Pounds. 5.06 Pounds. Pounds. Pounds. Pounds. Pounds. Pounds. Pounds. 1W 3.68 5.98 1^ 4.29 6.90 9.82 15< 4.91 7.83 11.05 2 5.53 8.75 12.27 16.11 2^ 6.14 9.66 13.05 17.64 2^ 6.74 10.58 14.72 19.17 24 7.36 11.50 15.95 20.70 7.98 12.43 17.18 22.19 27.62 3^ 8.59 13.34 18.35 23.78 29.45 3iJ 9.20 14.21 19.64 25.31 31.03 37.53 3%r 9.76 15.19 20.86 26.85 33.13 39.73 4 10.44 16.11 22.10 28.38 34.98 41.88 49 09 4\i 11.10 17.08 23.37 29.97 36.87 44 08 51 60 & 11.66 17.94 24.54 31.44 38.65 46.17 53.99 62.12 4^4 12.27 18.89 .25.77 32.98 40.50 48.32 56.45 64.89 5 s% S l /2 s% 6 6% 6K 6M 7M 7K 7# 8 8M 8H 8M 9 9M 9^ 12 88 13.50 14.11 14.73 15.34 15.95 16.57 17.18 17.79 18.41 19.03 19.64 20.02 20.86 21.69 22.09 22.71 19.78 20.71 21.63 22.55 23.47 24.39 25.31 26.23 27.15 28.08 29.00 29.69 30.83 31.74 32.90 33.59 34.52 35.43 36 36 26.99 28.23 29.45 30.68 31.91 33.13 34.36 35.59 36.82 38.05 39.05 40.50 41.71 42.95 44.40 45.40 46.64 47.86 49 09 34.51 36.05 37.58 39.12 40.65 42.18 43.72 45.26 46.79 48.10 49.86 51.38 52.92 54.45 56.21 57.52 59.07 60.59 6" > 13 42.33 44.18 46.02 47.86 49.70 51.54 53.39 55.23 56.84 58.91 60.74 62.59 64.42 66.26 68.33 69.95 71.80 73.63 75 47 50.46 52.62 54.76 56.91 59.06 61.21 63.36 65.28 67.65 69.79 71.95 74.09 76.23 78.38 80.76 82.68 84.84 86.97 89 13 58.90 61.36 63.81 66.27 68.73 71.18 73.41 76.09 78.53 81.00 83.45 85.90 88.35 90.81 93.49 95.72 98.18 100.63 1f\T. 09 67.64 70.41 73.17 75.94 78.70 81.23 84.22 86.97 89.74 92.50 95.26 : 98.02 100.78 103.54 1 106.53 109.06 111.84 114.59 mis 76.69 79.77 82.84 85.91 88.75 92.04 95.10 98.18 101.24 104.31 107.38 110.45 113.51 116.58 119.87 122.72 125.80 128.85 m9A 9% 10 ioM lOfc 10^ 37.28 38.20 50.32 51.54 52.77 54.00 55 22 63.66 65.20 66.73 68.26 69 80 77.32 79.16 80.99 82.84 84 67 91.28 93.42 95.57 97.71 99 86 105.54 108.00 110.44 112.90 115 35 120.12 122.87 125.63 128.39 ' 131 15 134.99 138.06 141 . 12 144.19 147 26 11 56 46 71 33 86 5^ 10 7 01 117 81 133 9^ 150 33 11M 72 86 88 35 104 15 1 9 9 6 136 67 153 40 1m 74 39 90 19 106 30 ' IT* 71 139 44 156 44 \\\ 75 93 92 04 108 45 125 18 142 18 159 54 12 77 46 93 60 110 60 127 60 144 96 162 60 (22) 338 MACHINERY FOR REFRIGERATION, USEFUL TABLES AND MEMORANDA RELATING TO PRIME MOVERS. AREAS OF CIRCLES ADVANCING BY EIGHTHS OF AN INCH. Diam. y* X % l /2 % M '% Diam. .0000 .0122 .0490 .1104 .1963 .3068 .4417 .6013 1 .7854 .9940 1.227 1.484 1.767 2.073 2.405 2.761 1 2 3.141 3.546 3.976 4.430 4.908 5.411 5.939 6.491 2 3 7.068 7.669 8.295 8.946 9.621 10.32 11.04 11.79 3 4 12.76 13.36 14.18 15.03 ! 15.90 16.80 17.72 18.76 4 5 19.63 20.62 21.64 22.69 23.75 24.85 25.96 27.10 5 6 28.27 29.46 30.67 31.91 33.18 34.47 35.78 37.12 6 7 38.48 39.87 41.28 42.71 44.17 45.66 47.17 48.70 7 8 50.26 51.84 53.45 55.08 56.74 58.42 60.13 61.86 8 9 63.61 65.39 67.20 69.02 70.88 72.75 74.76 76.58 9 10 78.54 80.51 82.51 84.54 86.59 88.66 90.76 92.88 10 11 95.03 97.20 99.40 101.6 103.8 106.1 108.4 110 7 11 12 113.0 115.4 117.8 120.2 122.7 125.1 127.6 130.1 12 13 132.7 135.2 137.8 140.5 143.1 145.8 148.4 151.2 13 14 153.9 156.6 159.4 162.2 165.1 167.9 170.8 173.7 14 15 176.7 179.6 182.6 185.6 188.6 191.7 194.8 197.9 15 16 201.0 204.2 207.3 210.5 213.8 217.0 220.3 223.6 16 17 226.9 230.3 233.7 237.1 240.5 243.9 247.4 250.9 17 18 254.4 258.0 261.5 265.1 268.8 272.4 276.1 279.8 18 19 283.5 287.2 291.0 294.8 298.6 302.4 306.3 310.2 19 20 314.1 318.1 322.0 326.0 330.0 334.1 338.1 342.2 20 21 346.3 350.4 354.6 358.8 363.0 367.2 371.5 375.8 21 22 380.1 384.4 388.8 393.2 397.6 402.0 406.4 410.9 22 23 415.4 420.0 424.5 429.1 433.7 438.3 443.0 447.6 23 24 452.3 457.1 461.8 466.6 471.4 476.2 481.1 485.9 24 25 490. 8 495.7 500.7 505.7 510.7 515.7 520.7 525.8 25 26 530.9 536.0 541.1 546.3 551.5 556.7 562.0 567.2 26 27 572.5 577.8 583.2 588.5 593.9 599.3 604.8 610.2 27 28 615.7 621.2 625.7 632.3 637.9 643.5 649.1 654.8 28 29 660.5 666.2 671.9 677.7 683.4 689.2 695.1 700.9 .29 30 706.8 712.7 718.6 724.6 730.6 736.6 742.6 748.6 30 31 754.8 760.9 767.0 773.1 779.3 785.5 791.7 798.0 31 32 804.2 810.5 816.9 823.2 829.6 836.0 842.4 848.8 32 33 855.3 861.8 868.3 874.8 881.4 888.0 894.6 901.3 33 34 907.9 914.6 921.3 928.1 934.8 941.6 948.4 955.3 34 35 962.1 969.0 975.9 982.8 989.8 996.8 1003 1010 35 36 1017 1025 1032 1039 1046 1053 1060 1068 36 37 1075 1082 1089 1097 1104 1111 1119 1126 37 38 1134 1141 1149 1156 1164 1171 1179 1186 38 39 1194 1202 1210 1217 1225 1233 1241 1248 39 40 1256 1264 1272 1280 1288 1296 1304 1312 40 41 1320 1328 1336 1344 1352 1360 1369 1377 41 42 1385 1393 1402 1410 1418 1427 1435 1443 42 43 1452 1460 1469 1477 ! 1486 1494 1503 1511 43 44 1520 1529 1537 1546 j 1555 1564 1572 1581 44 45 1590 1599 1608 1617 ! 1626 1634 1643 1652 45 46 1661 1671 1680 1689 1698 1707 1716 1725 46 47 1734 1744 1753 1762 1772 1781 1790 1800 47 48 1809 1819 1828 1837 1847 1854 1868 1876 48 49 1885 1895 1905 1914 1924 1934 1943 1953 49 50 1963 1973 1983 1993 2003 2012 2022 2032 50 D=Diameter A=Area. C=Circumference. S=Contents of Sphere B=Contentsof Cylinder. D = 3.14159 or V A -f- .7854 or C X. 31831. A=D*X.7854 or (C -=- 3. 5446) 2 . C=D X 3. 14159 or 3.5446 \' AT S=D 3 X.5236. B = AX lenglh. (A being the area of one end.) MACHINERY FOR REFRIGERATION. 339 USEFUL TABLES AND MEMORANDA RELATING TO PRIME MOVERS. CIRCUMFERENCES OF CIRCLES ADVANCING BY EIGHTHS OF AN INCH. Diam. % M ' % % % M % Diam. .0 .3927 .7854 1.178 1.570 1.963 2.356 2.748 1 3.141 3.534 3.927 ! 4.319 4.712 5,105 5.497 5.890 1 2 6.283 6.675 7.068 7.461 7.854 8.246 8.639 9.032 2 3 9.424 9.817 10.21 10.60 10.99 11.38 11.78 12.17 3 4 12.56 12.95 13.35 13.74 14.13 14.52 14.92 15.31 4 5 15.70 16.10 16.49 16.88 17.27 17.67 18.06 18.45 5 6 18.88 19.24 19.63 20.02 20.42 20.81 21.20 21.59 6 7 21.99 22.38 22.77 23.16 23.56 23.95 24.34 24.78 7 8 25.13 25.52 25.91 26.31 26.70 27.09 27.48 27.88 8 9 28.27 28.66 29.05 29.45 29.84 30.23 30.63 31.02 9 10 31.41 31.80 32.20 32.59 32.98 33.37 33.77 34.16 10 11 34.55 34.95 35.34 35.73 36.12 36.52 36.91 37.30 11 12 37.69 38.09 38.48 38.87 39.27 39.66 40.05 40.44 12 13 40.84 41.23 41.62 42.01 42.41 42.80 43.19 43.58 13 14 43.98 44.35 44.76 45.16 45.55 45.94 46.33 46.73 14 15 47.12 47.51 47.90 48.30 48.59 49.08 49.48 49.87 15 16 50.26 50.65 51.05 51.44 51.83 52.22 52.62 53.01 16 17 53.40 53.79 54.19 54.58 54.97 55.37 55.76 56.15 17 18 56.54 56.94 57.33 57.72 58.11 58.51 58.90 59.29 18 19 59.69 60.08 60.47 ; 60.86 61.26 61.65 62.04 62.43 19 20 62.83 63.22 63.61 64.01 64.40 64.79 65.18 65.58 20 21 65.97 66.36 66.75 67.15 67.54 67.93 68.32 68.72 21 22 69.11 69.50 69.90 70.29 70.68 71.07 71.47 71.86 22 23 72.25 72.64 73.04 73.43 73.82 74.22 74.61 75.00 23 24 75.39 75.79 76.18 76.57 76.96 77.36 77.75 78.14 24 25 78.54 78.93 79.32 79.71 80.10 80.50 80.89 81.28 25 26 81.68 82.07 82.46 8285 83.25 83.64 84.03 84.43 26 27 84.82 85.21 85.60 86.00 86.39 86.78 86.17 87.57 27 28 87.% 88.35 88.75 89.14 89.53 89.92 90.32 90.71 28 29 91.10 91.49 91.89 92.28 92.67 93.06 93.46 93.85 29 30 94.24 94.64 95.03 95.42 95.81 %.21 96.60 96.99 30 31 97.39 97.78 98.17 98.56 98.% 99.35 99.74 100.1 31 32 100.5 100.9 101.3 101.7 102.1 102.5 102.9 103.3 32 33 103.7 104.1 104.5 104.9 105.2 105.6 106.0 106.4 33 34 106.8 107.2 107.6 108.0 108.4 108.8 109.2 109.6 34 35 110.0 110.3 110.7 111.1 111.5 111.9 112.3 112.7 35 36 113.1 113.5 113.9 114.3 114.7 115.1 115.5 115.8 36 37 116.2 116.6 117.0 117.4 117.8 118.2 118.6 119.0 37 38 119.4 119.8 120.2 120.6 121.0 121.3 121.7 122.1 38 39 122.5 122.9 123.3 123.7 124.1 124.5 124.9 125.3 39 40 125.7 126.1 126.4 126.8 127.2 127.6 128.0 128.4 40 41 128.8 129.2 129.6 130.0 130.4 130.8 131.2 131.6 41 42 131.9 132.3 132.7 133.1 133.5 133.9 134.3 134.7 42 43 135.1 135.5 135.9 136.3 136.7 137.1 137.4 137.8 43 44 138.2 138.6 139.0 139.4 139.8 140.2 140.6 141.0 44 45 141.4 141.8 142.2 142.6 142.9 143.3 143.7 144.1 45 46 144.5 144.9 145.3 145.7 146.1 146.5 146.9 147.3 46 47 147.7 148.0 148.4 148.8 149.2 149.6 150.0 150.4 47 48 150.8 151.2 151.6 152.0 152.4 152.8 153.2 153.5 48 49 153.9 154.3 154.7 155.1 155.5 155.9 156.3 156.7 49 50 157.1 157.5 157.9 158.3 158.7 159.0 159.4 159.8 50 D=Diameter. A=Area. C=Circumference. S=Contents of Sphere. B=Contents of Cylinder. D = 3. 14159 or V A -*- - 7854 or C X. 31831. A=D 2 X.7854 or (C -f- 3. 5446) *. C=D X 3.14159 or 3.5446 V A. S=D 8 X.5236. B=AX length. (A being- the area of one end.) 340 MACHINERY FOR REFRIGERATION. TABLE OF COMPRESSOR CAPACITY IN CUBIC INCHES. FROM 1 TO 36 INCHES DIAMETER OF CYLINDER, AND FROM 1 TO 24 INCHES STROKE. The tabular number multiplied by strokes per minute and divided by 1,728 gives cubic feet per minute theoretical capacity of the cylinder. Cylinder diam. in inches. LENGTH OF STROKE IN INCHES. Cylinder diam. in inches. 1 2 3 4 5 6 7 8 9 10 C. Ins. C. Ins. C. Ins. C. Ins. C. Ins. C. Ins. C. Ins. C. Ins. C. Ins. C. Ins. 1 .785 1.571 2.356 3.141 3.927 4.712 5.498 6. ; 283 7.068 7.854 1 1* 1.767 3.534 5.301 7.068 8.83510.602 12.370 14.137 15.90517.672 li 2 3.141 6.283 9.425 12.566 15.70518.84921.991 25.132 28.27431.416 2 21 4.908 9.817 14.726 19.634 24.54329.452 34.360 39.269 44.17849.087 2i 3 7.068 14.137 21.206 28.274 35.34342.41149.480 56.549 63.61770.686 3 4 12.566 25.132 37.698 50.265 62.830 75.39687.962 100.53 113.09125.66 4 5 19.635 39.270 58.905 78.540 98.175117.81 137.44 157.08 176. 71196. 35 5 6 28.274 56.548 84.822 113.09 141.37169.64 197.92 226.19 254.46282.74 6 7 38.484 76.968 115.45 153.93 192.42 230.90 269.39 307.87 346.35384.84 7 8 50.265 100.53 150.79 201.06 251.32 301.59351.85 402.12 452.38 ! 502.65 8 9 63.617 127.23 190.85 254.47 318.08 381.70445.32 508.93 572.55636.17 9 10 78.540 157.08 235.62 314.16 392.70 471.24549.78 628.32 706.86785.40 10 11 95.033 190.06 285.09 380.13 475.16 570.19665.23 760.26 855.29950.33 11 12 113.09 226.18 339.27452.36 565.45 678. 541791. 63 904.72 1007.81120.9 12 13 132.73 265.46 398.19 530.92 663.65 796.38l929.ll 1061.8 1194.51327.2 13 14 153.93307.86 461.79 615.72 769.65 923.581077.5 1231.4 1385.31539.3 14 15 176. 71 1353. 42 530.13 706.84 883.55 1060.21236.9 1413.6 1590.3 1767.1 15 16 201.06 402.12 603.18 804.24 1005.3 1206.31407.4 1608.4 1809.5 2010.6 16 17 226.98 453.96 680.94 907.92 1134.9 1361.8|1588.8 1815.8 2042.8 2269.8 17 18 254.46 508.92 763.38 1017.8 1272.3 1526.71781.2 2035.62290.1 2544.6 18 19 283.52 567.04 850.56 1134.0 1417.6 1701.11984.6 2268. 12551. 6 2835.2 19 20 314.16628.32 942.48 1256.6 1570.8 1884.92199.1 2513.22827.4 3141.6 20 22 380.13760.26 1140.4 1520.5 1900.6 2280.82660.9 3041.0i3421.1 3801.3 22 24 452.39 904. 78 1357.1 1809.5 2261.9 2714.33166.7 3619.14071.5 4523.9 24 26 530.93 1061.8 1592.7 2123.7 2654.6 3185.53716.5 4247.44778.3 5309.3 26 28 615.75 1231.5 1847.2 2463. 3078.7 3694.5 4310.2 4926. 5541.7 6157.5 28 30 706.861413.7 2120.5 2827.4 3534.3 4241.14948. 5654.86361.7 7068.6 30 32 804.241608.4 2412.7 3216.9 4021.2 4825.45629.6 6433.97238.18042.4 32 34 1907.921815.8 2723.7 3631.64539.6 5447.56355.4 7263.38171.29079.2 34 36 1017.8 2034.1 3051.2 4068.3 5085.4 6102.47119.5 8136.69153.7 1017.0 36 MACHINERY FOR REFRIGERATION. 341 TABLE OF COMPRESSOR CAPACITY IN CUBIC INCHES. FROM 1 TO 36 INCHES DIAMETER OF CYLINDER, AND FROM 1 TO 24 INCHES STROKE. The tabular number multiplied by strokes per minute and divided by 1,728 g-ives cubic feet per minute theoretical capacity of the cylinder. Cylinder diam. in inches. LENGTH OF STROKE IN INCHES. Cylinder diam. in inches. 11 12 13 14 15 16 18 20 22 24 C. Ins. C. Ins. C. Ins. C. Ins. C. Ins. C. Ins. C. Ins. C. Ins. C. Ins. C. Ins. 1 8.639 9.42510.110 10.995 11.781 12.566 14.137 15.708 17.279 18.849 1 tt 9.43921.20622.973 24.740 26.507 28.274 31.809 35.34338.877 42.411 li 2 34.557 37.699 40.841:43.982 47.124 50.265 56.549 62.83269.113 75.399 2 2i '53. 995 58. 904 63. 813 68. 721 73.63078.539 88.356 98.174 107.99117.81 2i 3 77.754|84.82391.892 98. 960 106.03 113.09 127.23 141.37 155.51 169.64 3 4 138.22150.79163.36 175.92 188.49 201.05 226.19 251.32 276.55 301.58 4 5 215.98235.62255.25 274.89 294.52 314.16353.43 392.70431.97471.24 5 6 311.01'339.29367.56395.83 424.11 452.38J508.93 565. 48^ 622. 03 678. 57 6 7 |423. 32 461. 81 500. 29 538. 77 577.26 615. 74 692.71 769.68846.65923.61 7 8 552.91 603.18653.44703.71 753.97 804.24 904.77 1005.31105.81206.3 Q 9 699.78 763.40 827.02 890.63 954.25 1017.8 1145.1 1272.31399.5 1526.8 9 10 ! 863. 94 942. 48 1021.0 1099.5 1178.1 1256.6 1413.7 1570.8:1727.8 1884.9 10 11 1045.31140.31235.4 1330.41425.4 1520.5 1710.5!l900.6i2090.7 2280.7 11 12 1233.9|1357.1 1470.2 1583.31696.4 1789.52035.7 2261.9 2488.12714.3 12 13 1459.911592.7 1725.5 1858.2 1980.9 2123.7 2389.1 2654.6 2920.1 3185,5 13 14 1693.2|l847.2;2001.1 2155.1 2309.0 2463.0 2770.8 3078.7 3386.6 3694.5 14 15 1943.8:2120.5,2297.2 2474.0 2650.72827.4 3180.8 3534.3 3887.74241.1 15 16 2211.612412.7 17 2496.8|2723.7 18 J2799.03053.6 2613.8 2650.7 3308.0 2814.8 3177.7 3562.5 3015.9 3404.7 3817.0 3216.9 3631.6 4071.5 3619.1 4085.6 4580.4 4021.2 4539.6 5089.3 4423.34825.4 4993.55447.5 5598.36107.2 16 17 18 19 3118.7 13402.3 3685.8 3969.4 4252.9]4536.4 5103.5 5670.5 6237.66804.7 19 20 3455.7 3769.9i4084.0 4398.2 4712.45026.5 5654.86283.2 6911.5 7539.8 20 22 1 4181.4J4561.5|4941.75321.8 5701.96082.1 6842.37602.6 8362.99123.1 22 24 4976.25428.65881.0 6333.4 6785.87238.2 8143.09047.8 9952.5 10857 24 26 5840.2i6371.1 6902.0 7433.0 7963.98494.8 9556.7 10618 11680 12742 26 28 6773.27389.0 8004.7 8620.59236.39852.0 11083 12315 13546 14778 28 30 i 7775. 4 8482. 3 9189.1 98%. 010602 11309 12723 14137 15550 16964 30 32 8846.6 ! 9650.9 10455 11259 12063 12868 14576 16085 17693 19301 32 34 9987.1 10895 11802 12710 13618 14526 13642 18158 19974 21790 34 36 11187 12214 13232 14240 15258 16275 18311 20347 22383 24418 36 342 MACHINERY FOR REFRIGERATION. NEATH GIVE THE W m < CO M J5 S P S iJ o E O ^ w M pq H- O HH O O iH rH rH cO(Noioj machines 155 with boiler 157 with condenser 150 Combustion improved by heating- air 153 of fuels, table 348 " Compend of Mechanical Refrigeration," importance of iv Composition of fuels, table 349 Compound ammonia compressors, Antarctic 104 " Clyde Engineering Co 157 Haslam 158-160 " Humble & Nicholson, of Geelong. .158 Linde Co 159 Lock 24 York Co 24, 108 Compound ammonia machines, Antarctic, duplex 152 Antarctic, enclosed 153, 154, 160-168 Compound compressor by theoretical diagrams . . 166 Compound compressor cylinders, use of condensers in 117 Compound compressor system 15G Compound expansion 141 Compound submerged condensers, correct principle 62 wrong principle 6& Compressed air, literature on, appendix 354 Compressed air machine 23, 24 by author 123- Compressed air machines, power required by 35 Compression, adiabatic 189, 196 of air 192 Compression and expansion, diagram 32. Compression, compound 156 curves of 97 isothermal 187 heat and pressure by, diagram 19& hyperbolic logarithms for, table 189 Compression machine, Arctic 308 Ball 298 Barber 310 Boyle 306 Buffalo 302 Challoner 312-314 first in New South Wales 23 Hercules, latest design 297 Ideal, effect of motion, diagram 316 Ideal, section 315 Stallman 321 relation of parts 57 Vulcan 317, 318, 319 Compression of gas 94 Compression of gases 1 74, 186 table 178 Compression plant, leading features 56 Compression system, forerunner of 19 Compression systems, general principles of all 56 Compressor, accessibility of parts 122 and engine pistons, mean speed, table . .335 and steam engine 117 arrangement of engine and 124 capacity, table , 340, 341 MACHINERY FOR REFRIGERATION. 363 PAGE. Compressor, construction of 101 Compressor cylinder, Buffalo, section 301 Frick Co.'s 280 " wear of Ill " desirable qualities of 95 efficiency of 234 functions of 95 horizontal arrangement of engine and 126 how to plot diagrams of work of 137 importance of 94 indicating- the 97 " lubrication of 109 piston, work to be done by 118 single-acting and compound, comparison between 163 single-acting and engine 136, 137 " single-acting, diagram 165 supply of oil to 84 " vertical, horizontal engine 128 vertical, overhanging fly wheels 134 with two engines 130 Compressors, air 119 belted 146 " compound ammonia 155 geared 145 no oil necessary in some 85 single versus double-acting 121 " two vertical and horizontal engine 127 vertical, with inside fly wheels 129 " vertical, with outside fly wheels 130 with vertical engine 131 Condenser pressure, atmospheric 60, 61, 65 " evaporative 66 hot and cold climates 239 or cooler 58 submerged 59, 62. 63 Condenser water, velocity of 74 Condensers 58, 243, 269 " actual proportions of 76 atmospheric 59, 65 double pipe ammonia 275 evaporative 64-67 for absorption plants 269 functions of 71 pipe required for 76 submerged 59 triplex 270 use of, in compound compressor cylinders 117 Vogt, for absorption machines 270 Condensing water, re-use of 64 Condensing water temperatures 236 Conducting power of cylinders 74 Conductivity of metal 72 Conductors of heat, tables 72 ' ' Consolidated ' ' compressor 109 Construction of boilers 222-230 Construction of compressor 101 Construction of refrigerating machinery 121 Consumption of coal 227, 231 Conversion factors, table 346 Cooling processes 67 364 MACHINERY FOR REFRIGERATION. PAGE. Cooling- towers 67, 278 Copper, conductivity of 72 Cornish boiler 212-215 tubular boiler 218-223 Cost, first, not the most importance 121 Cost of boilers 205 Creamery Package Manufacturing- Co 320 Cryogen machine 106 Cubic feet of g-as per ton refrigeration, table 238 Cullsn, Dr 17 Curves of compression 97 Cylinder, a true 114, 115 Cylinders, conducting- power of 74 D De-aerated water for ice molds 247 Definition of heat terms 26 De La Vergne compressor 108 De La Vergne plant, g-eneral arrangement of 61 Design and construction of refrigerating machinery 121 Desiccator, importance of 51 Diagonal connection of engine and compressor 140 Diagram illustrating work of Corliss engine and two compressors. .138 how to plot, of compressor's work 137 " indicator, from belt-driven machines 148, 149 " indicator, from compound compressor 166-174 " indicator, from diagonal connection 144 indicator from toggle machines 316 indicator, right-angle connection 135 138 " indicator, small, straight line compressor .137 indicator, single-acting compressor 118-120 indicator, straight line compressor 119 Direct expansion system, advantages of 69 Discharge from compressor 111-113, 165 Discharge valve 277 Discs on expansion pipes 69 Distillation for can ice from exhaust and live steam 141 by triple effect 253 Distilled water 253-260 " triple effect, diagram 252 Distilling apparatus 252 Double-acting compressors, air 23, 112, 119 ammonia 99,101,104,105,116,145 ether 22 Double pipe ammonia condensers 275 Dry compression, drying action, with cold air 69 duplex ammonia compressor 152 DuTremblay ether engine 36 E Eclipse pump 280 Economizer 246 Economy of high pressure steam, table 227 Effect of climate 236 Efficiency as affected by clearance 113 Efficiency of boilers, tables 215. 216 coal 233 " compressor 95, 2o4, 236 compressor and engine, table 235 MACHINERY FOR REFRIGERATION, 365 PAGE. Efficiency of engine 23& fuel, boiler and engine. 233 ice plants 231-251 refrigerating plant 94 " thermal, steam engines 228-230 Electrical and mechanical unit equivalents, table 345 Electric welding 81 Elevation, process of 67 Energy of gas at different stages of compression, calculations . .196-203 Engine and compressor pistons, mean speed of, table 335- Engine, Leavitt, pumping 228-230 Engines, mean pressure of steam, and cylinders of, table 334 " steam, efficiency of 233 thermal efficiency of 228-230 Equivalent, Joule's 185 Equivalent measures of volume 350 Equivalents, electrical and mechanical unit, table 345 Equalizer, Vogt absorption plant 264 Ether, experiments with 17 Ether machines, Harrison 19- 22 Siebe & Gorman 21 Twining 19- Ether, vapor tension of 39 Evaporated value of fuels, table 348 Evaporating water, machine for 17 Evaporation condensers 64, 67 in dry climates 60 per pound of coal 219, 23 to produce cold 17, 19, 41, 43, 49 Evaporator, water 32T Exchanger, importance of 51 Vogt absorption plant 264 Exhaust steam for distilled water 240 Exhaust steam purifier 241 Expansion, adiabatic 189- of air 192 by stages 23, 24, 36 Expansion, latent heat of 184, 185 Expansion of gases 174, 186 " diagram 183 steam, table 227 Expansion valve, purpose of 43- Explosions, boiler, causes of 208 F Factors, conversion, table 346 Feed water heaters 243-245 Filter for exhaust steam distilled water 247 Filters . ." 243-249 Flashing valve 49, 77 Flues, strengthening of 214 Fly wheel, enormous 123 power required by 124 overhung 128, 130 Food products, first proposal to refrigerate for shipment 25, 4& specific heat of, table 29 Forecooler 240, 248 Freezing point of water 27 Freezing tank 68- " functions of 71 366 MACHINERY FOR REFRIGERATION. PAGE. Fresh Food and Ice Co 156, 296 Frick Co. 's ammonia condensers 272 " ammoni a val ve 278 " compressor 109 " " engine and compressor, section of 125 " latest ammonia compressor cylinder 280 " machine, latest design 279 Friction of fly wheel 1 24 machine 95, 96 Fuels, average composition of, table 349 " evaporated value of, table 348 " heat of combustion of, table 348 " saving of, table 336 value of wood, table 349 " weight and combustion of, table 349 G Galloway boiler 216-219 tubes 216 Gas, ammonia, saturated, properties of, table 347 " carbonic acid, saturated, properties of, table .' . . .347 " compression of 94 " cubic feet of, per ton refrigeration, table 238 " discharge Ill " liquefied under pressure 30 " nature of 124 " pressure 103 Gas pressures, table 178-181 Gases, compression of 94, 174, 186 table 178 conductive power of 72 " critical temperature of 39 " expansion of 174, 186 " expansion of, diagram 183 " formula for pressure, weight and volume 181 " hyperbolic logarithms for calculating 189 " properties of 38 tables 191,346 " relative efficiency of 43 " specific heat of 28, 181-186 " volume of 179-181 in cylinder at different stages of compression, table. 197 Gas, sulphur dioxide, saturated, properties of, table 347 Gas volumes and pressures at different stages of compression, cal- culations 196-203 Gauge pressures 179 Gauges, tables 344 Gay-Lussac, researches of 176 Geared compressors 145, 146 Generator, Vogt absorption plant 263 Gifford machine ;: 19 Glycerine as a lubricant 85, 88, 91 Gorrie, Dr. John, experiments of 19 Grease separators 242 H Hagen's experiments 18 Hall's carbonic acid machine 45 Harrison, James, in Australia > 19 Harrison's duplex (ether) ice machine, diagram of 22 MACHINERY FOR REFRIGERATION, 367 PAGE. Harrison's ether machine, diagram of 20 Haslam cold air machine 34 " compound compressor 159, 160 Heat abstracted equivalent to one ton refrigeration, table of 75 " Heat" and ** Cold," relative terms 26 Heat and its applications, literature on, appendix 352 Heat and pressure by compression, diagram 193 Heaters, feed water 242-245 Heat, latent 28 of air 186 " " of expansion 184, 185 " " of liquefaction 28, 41 of vaporization 28, 42 mechanical equivalent of 175, 185 of combustion of fuels, table 348 sensible 26 specific 27 of air 186 " of ammonia 1 84, 185 of brine , 29 of ice 28 of gases 181-186 Heat terms, illustrations of 28 " to be abstracted in the work of refrigeration 29 " transmission of, through various media 73, 74 " unit 27 Hercules machine 103. 104, 116. 296, 297 High pressure cylinder, section of 112 Holden ice making system 330 Horizontal arrangement of engine and compressor 126 compressors 99, 101, 112 engine and two vertical compressors 127 engines 115, 116, 119, 125, 126, 133, 143 " and compressors 34- 36 " versus vertical tubes in condenser 75 Humble & Nicholson, of Geelong, compound compressors 158 Humid gas 186 Hydrogen, conduction of heat by 72, 73 Hyperbolic logarithms, table 189 I " Ice and Refrigeration " established iv Ice and salt 17, 26 Ice cream 17, 26 Ice freezing tank 68 Ice, from distilled water 234 " from impure steam 242 Ice machine, first successful, for manufacturing purposes 21 Ice made in Australia in 1860 21 Ice making in America, beginning of 19 " " Australia, beginning of 19 India 17 " plant, Australian 55 " system, Holden 330 Ice per ton of coal 231 Ice plant efficiencies 231-251 Ice and Cold Machine Co. 's absorption machine 2H5 " compression machine 298 Ideal compression machine, effect of motion, diagram 315, 316 Inclosed compressors 101, 153, 155 368 MACHINERY FOR REFRIGERATION, PAGE, Incrustation of boilers 206, 208 India, natural ice making- in 17 Indicating- the compressor 97 Indicator cards from compressor 118 " Antarctic compressor 168-170 " from Corliss engine and compressor 120 " from steam and air cylinder 119 Indicators for compressors 166, 170, 174, 190 Interceptors of dirt 92 of oil 89, 90 Iiiterchang-er of temperature 50, 52 Intrinsic energy 32, 33, 193, 199 Iron, conductivity of 72 Isothermal compression and expansion of g"as, diagram 187 J Joints, materials for 123 Joule's equivalent 185 K Kidd, Hector 260 Kilburn inclosed machine 153 Kirk, Dr., regenerative air machine 21 Kroeschell Bros, carbonic acid machine 269 L Lancashire boiler 216-219 Lantern bushes 85, 86, 106, 109 Latent heat 28 " of air 186 " " of expansion 184, 185 " " of liquefaction 28. 41 " of vaporization 28, 42 Lavoisier's experiments 17 Laws of g-ases 174 Leavitt pumping- engine 228-230 Leslie's experiments with sulphuric acid and water 17 Linde compound compressors 159 " compressor, American type, section 288 " and engine, plan of 116 " system for cooling- 69 " oiling- apparatus 85 Liquefaction, latent heat of 41 Liquefied air machine 38 Literature on compressed air, appendix. 354 heat and its application, appendix 352 refrig-erating- machinery, appendix 353 " refrigeration and allied subjects, appendix 351 thermo-dynamics, appendix 353 Lock's compound compressor 24, 158 Locomotive type boiler 130' Lubricating- pump 87 Lubrication of compressor 109 piston 1 09 M MacDonald, C. A., compression machine 29T Mag-nus, experiments by Prof 72 Marine installations 34, 35, 45, 49- Mariotte's law 175 MACHINERY FOR REFRIGERATION. 369 PAGE. Measures of volume 350 Mechanical and electrical unit equivalents, table 345 Mercury wells 190 Metals, conductivity of 72 relative weights of, table 350 Modern ice machine 55 Mort, T. S 296 " experiments of 21, 25 Mountings of boilers t 222-230 Multitubular boilers 207-212 Munroe, R., & Sons boilers 327 N Newburgh Ice Machine and Engine Co. 's compressor 284 Nicolle, Ed., development of ammonia absorption system 21 " " personality of 48 Nitrate of ammonia process for shipboard 23 o Oil about compressor 84 Oil and grease separators 242 Oil, effect of, in boilers 207 " effect of, upon clearance 84 Oiling apparatus, Linde system 85 devices 108 Oil injection in compressor 109, 113 Oil interceptor 87 Oil, methods of injecting, into compressor 84-99 Oil pump 85 " " section 86 Oil, separation of, from ammonia 91 Oil separators 92 Oil separator with "baffles, " section of 89 wire screens, section of 90 Oil, unnecessary in some compressors 85 " use of, in refrigerating systems 84 P Pamphlets on refrigeration, appendix 352 Pasteur filter 247 Penney horizontal double-acting machine 285 Pennsylvania Iron Works machine 133, 306 Perkins, Jacob, machine 18, 19 Pioneers in refrigeration iv Pipe bending methods 82, 83 Pipe coils 82 Pipe, effective surface, table of 75 " length of, required 75 Pipes and joints 79 " weight of cast iron, table 337 Pipe welding 81 Piston, ideal 114 lubrication of 109 Piston rings Ill Piston, work of 109 " work to be done by 118 Pistons of engine and compressor 121 Postle's cold air machine, 1868 .24 Pots 254 Premier water tube boiler 324 (24) 370 MACHINERY FOR REFRIGERATION. PAGE. Pressure of gas at different stages of compression, calculations. 196-203 of g-ases, table 178-18] of steam in cylinders of engines, table 334 Prime movers, tables 338, 339 Properties of gases 38 " tables 191, 346 of saturated ammonia gas, table 347 of saturated carbonic acid gas, table 347 of saturated sulphur dioxide gas, table 347 Pulley, work of 171-173 Pump, Eclips'e 280 for oil 85, 87, 102 " for vacuum re-absorber 52 Pumping engine, Leavitt 228-230 Purifying exhaust steam, process 241 Q Quadruple expansion engine 233 Queensland machines 106 R Reboiler 247 Reboring of cylinders Ill Receiver, liquid ammonia 93 Refrigerating machine, relation of parts 57 Refrigerating machinery, design and construction of 121 " literature on, appendix 353 Refrigerating machine, small 146-155 Refrigerating media, boiling points of 39, 40 " diagram showing latent heat of 42 Refrigerating on shipboard, first proposal for 48 Refrigerating plant, efficiency of 94 Refrigerating systems, use of oil in 84 Refrigeration and allied subjects, literature on, appendix 51 artificial, first beginnings of 17 books on, appendix 352 business, importance of iv " by nitrate of ammonia 22 " cubic feet of gas per ton of, table 238 " heat to be abstracted in the work of 29 " pamphlets on, appendix 352 " pioneers of iv systems of, in use 30 treatises, appendix 352 Refrigerator 68 Refrigerators, first ammonia, in Sydney 57 functions of 71 Regenerative boiler settings 222 Regnault, researches of 177 Remington compressor, latest design, description 288 machine 153 Resistance of compressor piston 97 Right-angled connection of machine 125-136 " " " diagram 135 Rudberg, researches of 177 s Saving of fuel, table 336 Scale 205 Scotch boilers . . 327 MACHINERY FOR REFRIGERATION. 371 PAGE. Sensible heat 26 Separator, oil, with " baffles, " section of 89 " with wire screens, section of 90 Separators, oil and grease 92, 242 Settings, boiler 208, 227-230 regenerative, boiler 222 Sextuple effect distilling- plant 259 Shipboard machines 155 Shipboard refrigeration 36, 37, 49 Single versus double-acting compressors 121 Skimmer 246 Sloper's, Geo. Beven, patent 17 Specific heat 27 " of air 186 " of ammonia 184, 185 " of brine 29 " of ice 28 " of gases 28, 181-186 " of solids 29 Speed of compressor and engine pistons, table 335 Stallman compression machine 321, 323 Steam engine, efficiency of 233 Steam per horse power 215, 234 " pressure of, in cylinders of engines, table 334 Steam purifiers 241-243 Still, Vogt absorption plant 263 Stocker's, John, water cooling tower 276 Straight line compressors 119, 120 Submerged condensers 58 compound, correct principle 62 wrong principle 63 Sulphur dioxide gas, saturated, properties of, table 347 Sulphuric acid, vapor tension of 39 Sulphuric dioxide, boiling point of 39 Sulphuric ether, boiling point of 39 Systems of refrigeration, different, in use 30, 31 T Tandem compound engines 99, 133, 141 compound machine 141 Technical connection 140 Temperature, critical, of gases . . 39 effect of, on compressor castings 110 " of aqua ammonia, correction of, table 342, 343 of gases 179-181 Temperatures, condensing water 236 exchange of, in condensers and freezing tanks 71 Theory of compressed air machines 32, 33 Thermal efficiency of coal 233 steam engines 228-230 Thermodynamics, literature on, appendix 353 Thermometers, different, in use 27 Towers, cooling 276 Trap 93 Treatises on refrigeration, appendix 352 Trevithick, Richard 213 Triplex condensers 270 Triple effect distilled water, diagram. 252 process 253-260 Triumph American compressor machine 293-295 372 MACHINERY FOR REFRIGERATION, PAGE. Tubes, cleaning- 205 " Galloway 216 " horizontal versus vertical in condenser 75 " water, advantag-es of 325 Twining- in America 19 V Vallance machine 17 Valve, expansion, purpose of 43 Valves, Australian, ammonia 78 " for reg-ulation 77 " horizontal to compressor 99, 101 " in pistons 104, 105, 108, 114, 135, 143, 152, 162 " manifold 78 " recent inventions 277 " slide to compressor 20, 28 " with by-pass 79 Vaporization, latent heat of 42 Vapor tension of g-ases and vapors 39 " of water 39 Velocity of condensing- water 74 Vertical compressor, horizontal engine 128 Vilter ammonia compressor machine 289 Vogt absorption system, diagram 262-264 " condensers for absorption machines 270 " type of absorption machine 261 Volume, equivalent measures of 350 Volume of gas at different stages of compression, calculations. . .196-203 Volume of gases 179-181 Vulcan compression machine 317, 318, 319 W Water cooling towers 276 Water, distilled 253-260 Water evaporator 327 Water tube boilers 205-215, 325, 327 Water tubes, advantages of 325 Water, vapor tension of 39 Weight and comparative fuel value of wood, table 349 " of cast iron pipes, table 337 " of steam per horse power 234 Weights of metals, table 350 Welding pipe 82 Wells for thermometers 190 Westerlin & Campbell's condenser 274 Westinghouse machine 151 Wolf, Fred W., Co. 's American valve 277 " " ammonia condensers 273 " latest designs American Linde machine., 289 Wood, weight and comparative fuel value of, table 349 Worms for distillation 243 Wrought iron, conductivity of 72 Y York Co.'s compression machine .24, 108, 109, 115, 117, 158, 282, 284 Z Zero point 26 MACHINERY FOR REFRIGERATION, 373 CLASSIFIED TRADE INDEX TO ADVERTISERS. PAGE. AMMONIA. Ammonia Co., of Australia opposite inside front cover Barrett Manufacturing Co 383 Herf & Frerichs Chemical Co 416 Linde Australian Refrigeration Co 409 National Ammonia Co opposite inside front cover AMMONIA FITTINGS. (SEE FITTINGS, AMMONIA.) AMMONIA PACKING. Garlock Packing Co 3% ARCHITECTS AND ENGINEERS. Brubaker, Samuel H., & Co 393 Clyde Engineering Co., Ltd 394 Mort's Dock and Engineering Co 388 BOILERS. Clyde Engineering Co., Ltd 394 Frick Co 414, 415 Mort's Dock and Engineering Co 388 Munroe, R., & Sons 393 Newburgh Ice Machine and Engine Co. 412 Pennsylvania Iron Works Co inside back cover and page opposite Remington Machine Co 381 Vogt, Henry, Machine Co 376 York Manufacturing Co 410, 411 BOILER TUBE CLEANERS. Union Boiler Tube Cleaner Co 387 BRINE AND FR EEZI NG TAN KS. Frick Co 414, 415 Marlin & Co., Inc 385 Munroe, R., & Sons 393 Newburgh Ice Machine and Engine Co. 412 Pennsylvania Iron Works Co . . . inside back cover and page opposite Scaife, Wm. B., & Sons 400 Vogt, Henry, Machine Co 376 York Manufacturing Co 410, 411 CENTRIFUGAL PUMPS. Morris Machine Works ... ... 399 PAGE. COILS AND BENDS. Barber, A. H., Manufacturing Co 379 Clyde Engineering Co., Ltd 394 Farrell & Rempe Co 378 Harrisburg Pipe and Pipe Bending Co. 382 Philadelphia Pipe Bending Works 379 Whitlock Coil Pipe Co 408 CONDENSERS. Arctic Machine Co 397 Barber, A. H., Manufacturing Co 379 Challoner's, George, Sons Co 398 Clyde Engineering Co., Ltd 394 Cochran Co 377 Creamery PackageiManufacturing Co.380 Farrell & Rempe Co 378 Frick Co 414, 415 Ice and Cold Machine Co 407 Ideal Refrigerating and Manufactur- ing Co 395 Linde Australian Refrigeration Co 409 MacDonald, C. A 406 Mort's Dock and Engineering Co 388 Newburgh Ice Machine and Engine Co. 412 Pennsylvania Iron Works Co inside back cover and page opposite Remington Machine Co 381 Triumph Ice Machine Co third page advertisements in front Vilter Manufacturing Co 417 Vogt, Henry, Machine Co 376 Vulcan Iron Works : 384 Westerlin & Campbell 401 Wolf, Fred W., Co inside front cover York Manufacturing Co 410, 411 CONSULTING ENGINEERS. Selfe, Norman 389 Westerlin & Campbell 401 COOLING TOWERS. Stocker, Geo. J .403 DOORS. (Refrigerator and Cold Storage.) Stevenson Co., Ltd 413 374 MACHINERY FOR REFRIGERATION. PAGE. ENGINES. Clyde Engineering Co., Ltd 394 Frick Co 414, 415 Mort's Dock and Engineering- Co 388 Newburgh Ice Machine and Engine Co. 412 Pennsylvania Iron 'Works Co inside back cover and pag-e opposite Vilter Manufacturing Co .... 417 Vulcan Iron Works 384 York Manufacturing Co 410, 411 ENGRAVERS. Illinois Engraving Co Stromberg, Allen & Co. .. .394 395 FEED WATER HEATERS AND PURIFIERS. Harrisburg Pipe and Pipe Bending Co. 332 Robertson, Jas. L., & Sons 399 Whitlock Coil Pipe Co 408 FILTERS. (Water.) Marlin & Co., Inc Scaife, Wm. B., & Sons. . . 385 ... 400 FITTINGS. (Ammonia.) Arctic Machine Co 397 Barber, A. H., Manufacturing Co. . ..379 Challoner's, George, Sons Co 398 Clyde Engineering Co., Ltd 394 Cochran Co 377 Creamery Package Manufacturing- Co. 380 Frick Co 414, 415 Harrisburg- Pipe and Pipe Bending Co. 382 Ice and Cold Machine Co 407 Ideal Refrigerating and Manufactur- ing-Co 395 Linde Australian Refrigeration Co. ..409 MacDonald, C. A 406 Mort's Dock and Engineering Co 388 Newburgh Ice Machine and Engine Co. 412 Pennsylvania Iron Works Co . . .inside back cover and page opposite Remington Machine Co 381 Triumph Ice Machine Co third page advertisements in front Vilter Manufacturing Co 417 Vogt, Henry, Machine Co 376 Vulcan Iron Works 384 Westerlin & Campbell 401 Wolf, Fred W., Co inside front cover York Manufacturing Co 410, 411 FLUE CLEANERS. Union Boiler Tube Cleaner Co 387 FREEZING ESTABLISHMENT. New South Wales Fresh Food and Ice Co... ...388 PAGE. ICE AND REFRIGERATING MACHINERY. Arctic Machine Co 397 Barber, A. H., Manufacturing Co 379 Challoner's, George, Sons Co 398 Clyde Engineering Co., Ltd 394 Cochran Co 377 Creamery Package Manufacturing Co. 380 Frick Co 414, 415 Ice and Cold Machine Co 407 Ideal Refrigerating and Manufactur- ing Co 395 Linde Australian Refrig-eration Co 409 MacDonald, C. A 406 Mort's Dock and Engineering Co 388 Newburgh Ice Machine and Engine Co. 412 Pennsylvania Iron Works Co . . .inside back cover and page opposite Remington Machine Co 381 Spiers, James, Jr 396 Triumph Ice Machine Co third page advertisements in front Vilter Manufacturing Co 417 Vogt, Henry, Machine Co 376 Vulcan Iron Works 384 Westerlin & Campbell 401 Wolf, Fred W. Co inside front cover York Manufacturing- Co 410,411. ICE CAN FILLERS. Burns, Jas. F 393 Sauls Bros... ...389 ICE CANS. Marlin & Co., Inc 385 Scaife, Wm. B. & Sons 400 INDICATORS. Robertson, Jas. L., & Sons 399 INSULATING MATERIALS. Barrett Manufacturing Co 383 Bird, F. W.,& Son 399 Cabot, Samuel 391 Gilmour, R. M., Co 391 Johns, H. W., Manufacturing Co 391 Nonpareil Cork Co 384 U. S. Mineral Wool Co 391 PACKINGS. Garlock Packing Co 396 PRINTERS AND STATIONERS. Stromberg, Allen & Co 395 REFRIGERATING MACHINERY. (SEE ICE AND REFRIGERATING MA- CHINERY.) SEPARATORS. (Oil and Water.) Robertson, Jas. L., & Sons . .399 ADVERTISEMENTS. 376 MACHINERY FOR REFRIGERATION. Our Mighty Midget Ice and Refrigerating Machine OCCUPIES LITTLE SPACE. DOES GREAT WORK . DESIGNED ESPECIALLY FOR PACKING HOUSES, CREAMERIES, SMALL REFRIGERATING PLANTS, THREE TO FIVE TONS CAPACITY. ::::::: HENRY VOGT MACHINE CO, LOUISVILLE, KY. New Catalogue on Application MACHINERY FOR REFRIGERATION. 377 MACHINERY FOR REFRIGERATION FOR STEAM, WATER OR ELECTRIC POWER. Designed particularly for the Cold Storage of Food Products, Fish Freezing, Ice Cream Manufacture and the like, where safety, efficiency and the absence of offensive odors is indispensable IS THE SPECIALTY OF THE COCHRAN COMPANY, LORAIN, OHIO. 378 MACHINERY FOR REFRIGERATION. Farrell & Rempe Co MANUFACTURERS OF WROUGHT IRON COILS FOR ICE AND REFRIGERATING MACHINES PIPE COILS A\ADE ELECTRICITY All Ammonia Coils made of the very finest quality of Pipe (in any desired continuous length) and tested to 400 pounds air pressure. Coils of all descriptions for Heaters, Soap Makers, Blast Furnaces, etc. Manufacturers of Ammonia Pipe and Fittings Pipe Bending of all kinds a Specialty. PRICES FURNISHED ON APPLICATION. OFFICE AND WORKS: Cor. Sacramento and Carroll Avenues, Chicago MACHINERY FOR REFRIGERATION. 379 C. BAILE. H. LEIDY. PHILADELPHIA PIPE BENDING WORKS PHIL ADELPHI A BARBER COMPRESSOR BUILT BY A. H. BARBER MFG. CO. M A XTJ FA C TIT H E COILS OP ALL KIXDS AMMONIA VALVES AJTD FITTINGS REFRIGERATING AX D ICE MAKING MACHINERY Built in twenty different sizes, from 1% to 50 tons. The first machine built in 1894. Over 500 in successful opera- tion January 1, 1900. Catalogue sent on application. 229-231 SOUTH WATER STREET, CHICAGO 380 MACHINERY FOR REFRIGERATION. Our Refrigerating Machinery IS SUBSTANTIALLY AND SCIENTIFICALLY CONSTRUCTED. The lines of our compressor will be found to combine SYMMETRY with STRENGTH and DURABILITY. Ours is the only small machine of the duplex type on the market. We do not cater to cheapness, but furnish an outfit which has STAYING QUALITIES Write for Catalogue and full information. Estimates promptly made. CREAMERY PACKAGE MFG. Co, 1-3-5 WEST WASHINGTON STREET, CHICAGO, ILL. MACHINERY FOR REFRIGERATION. 381 Remington Machine Co WILMINGTON, DELAWARE, U.S.A. BUILDERS OF REFRIGERATING AND ICE MACHINERYgOB BAKER & HAMILTON, Pacific Coast Agents, San Francisco, Cal. Vertical single acting- Ammonia Compressors, with" engines direct connected, and with Fly Wheels for belt, from % to 12 tons refrigerat- ing capacity. Horizontal double acting- Ammonia Compressors, with Corliss engines, 16 to 100 tons refrigerating capacity. Complete plants installed and guaranteed. ICE MAKING: Can and Plate Systems. REFRIGERATING : Direct Expansion and Brine Systems. 382 MACHINERY FOR REFRIGERATION. Iron, Copper and Brass PIPE COILS FOR ICE MAKING AND REFRIGERATION. Bends and Manifolds for all purposes. Harrisburg Pipe and Pipe Bending Co. HARRISBURQ, PA. Wrought Iron Ammonia Cocks, Ammonia Valves and Fittings. MANUFACTURERS OF WROUGHT IRON PIPE FOR ALL PURPOSES. SPECIAL QUALITY REWORKED PIPE FOR AMMONIA WORK. Carbonic Acid Gas Cylinders, Ammonia Bottles or Flasks, Stills and Absorbers for Absorption Machines. HARRISBURG FEED WATER HEATERS Strictly high-grade, made of pure seamless copper coils. MACHINERY FOR REFRIGERATION. 383 290 BROADWAY, NEW YORK CITY MANUFACTURERS OF ABSOLUTELY PURE AND DRY ANHYDROUS AND AQUA 26 GUARANTEED FULL STRENGTH. ALL KINDS OF ROOFING AND BUILDING PAPERS Write to nearest branch for latest samples. Barrett's Rope Insulating Paper WATERPROOF AND ODORLESS. New York 290 Broadway. Philadelphia 1205 Land Title Bldg-. Chicago 909 Stock Exchange Bldg-. St. Louis 109 North 9th St. Cleveland 29 Euclid Ave. Cincinnati 639 West Front St. Allegheny 160 Rebecca St. Columbus, Ohio. Louisville Clay and Franklin Sts. Kansas City 1st and Campbell Sts. Minneapolis, Minn. New Orleans 508 Hennen Bldg-. and WARREN EHRET CO., 1210 Land Title Bldg., PHILADELPHIA. 384 MACHINERY FOR REFRIGERATION. \7lTf ^ A iv T Ice Making and \ Ul^CAIN Refrigerating Machines We carry in stock Ammonia Piping, Condenser Coils, Ammonia Fittings, Chapman Valves, Mineral Wool, Insulating Paper. REFERENCES: 10O MACHINES IN California, Oregon, Washington, Arizona, New Mexico, British Columbia, Mexico, Central Amer., i; South America, Hawaii, ~- Philippines. Pacific Mail S. S. Co. Pacific Coast S. S. Co Oceanic S. S. Co., U. S. Transports, of an3 r desired capacity, ON THE SIMPLEST AND MOST ECONOMICAL SYSTEM. _,,_ SEND FOR CATALOGUE. VULCAN IRON WORKS, s an Francisco, NONPAREIL CORK PATENTED PERFECT SECTIONAL BRINE PIPE COVERING. INSULATION IN SHEETS. SAMPLES, CIRCULARS, ETC. GLADLY FURNISHED. THE NONPAREIL CORK MFG. Co. LONDON OFFICE, 28 QUEEN ST. 90 WEST BROADWAY, NEW YORK MACHINERY FOR REFRIGERATION. 385 GALVANIZED STEEL ICE CANS OF EVERY DESCRIPTION AND ALL OTHER WORK OF GALVAN- IZED IRON IN CONNECTION WITH ICE MANUFACTURING ALSO MANUFACTURERS OF Exhaust Heads and Pipe, Portable Tanks for Storage of Oil, Filters, Reboilers, Skimmers and Storage Tanks, Cornices and Skylights, Crestings and Finials, Conductor Pipe and Fittings, Have Troughs, MARLIN & CO., me 23d and Smallman Streets, PITTSBURGH, Pa., U.S.A. <25) 386 MACHINERY FOR REFRIGERATION. The Western Brewer and Journal of the Barley, Malt and Hop Trades ILLUSTRATED CHICAGO 177 LASALLE ST. COR. MONROE NEW YORK 206 BROADWAY COR. FULTON ST. The Largest Paper in the World devoted to the interests of the Brewer and Maltster, and the Recognized American Authority in the Trade THE WESTERN BREWER Has been the most notable success ever achieved in journalism. It is SUPERBLY ILLUSTRATED with portraits of prominent brewers, plans of new breweries, detailed drawing's for the construction of breweries and malt houses, and of brew- ery plants, engraving's of all new inventions in the trade, results of microscopical examina- tions, etc., etc. Its REPORTS of the BARLEY, HALT and HOP fl ARKETS are the most elaborate and exhaustive published, and its quotations are standard in the trade. Its EDITORIALS are original, carefully and ably written, and cover the field of tech- nical research and study, the temperance question, and all subjects of interest to brewers, maltsters, or those engaged in the supply trades. Its BREWERY NEWS is g-athered from all quarters of the world, and keeps the trade constantly posted on current events and happening's. This feature of THE WESTERN BREWER is peculiarly its own, and no other trade journal possesses such a complete organi- zation for news collecting-. THE WESTERN BREWER carries, besides its engravings and all letterpress matter, upwards of ONE HUNDRED AND FIFTY PAGES OF ADVERTISEMENTS, covering every article of manufacture entering into the brewery economy. Its advertising pages are a complete directory of the trade. Subscription $5.00 a Year in Advance which includes a copy of the Brewers' Hand-Book of the U. S. and Canada, and all the Illustrated Supplements H. S. RICH & CO., Publishers Address either the New York or Chicago Office as above. MACHINERY FOR REFRIGERATION. 387 The I Jnion B O ILER TUBE me union CLEANER co 272 PENN AVE., PITTSBURGH, PA., U.S.A. HAS set the acknowledged standard for the world for removing by POWER driven MECHANICAL devices all conditions of scale from all makes of Water Tube boilers, whether horizontally inclined, such as the Babcock & Wilcox type, or vertical straight tubes, such as the Cahall type, or those having- single or double curved tubes, such as the Climax, Stirling, Thornycroft or Firmenich types. We sell or lease our device, which Is more in the nature of a royalty rather than the sale of machinery, or clean boilers by contract at a fixed price per tube. After nearly- five years of phenomenal success in an unique and exclusive industry in cleaning boilers all over the United States, in Eng- land and Scotland, we offer the results of our experience to those we have not been able so far to reach. Will be pleased to furnish free descriptive illustrated circulars with reference list comprising the largest firms in the United States, England and Scotland, copies of tests, and other particulars. 'We call attention to page 205 of this book. WE HAVE THE ONLY FLEXIBLE SHAFT of remarkable strength an d phenomenal durability under great stress, which we were compelled to design, owing to the fact that other makes required more time to keep them in repair than it did to do our part of the work that of cleaning Curved Tube Boilers. 388 MACHINERY FOR REFRIGERATION, Founded by the late Thomas Sutcliffe Mort, A. D. 1861. Incorporated under the Acts of New South Wales, 1874. N.S.W. FRESH FOOD & ICE Co. Ltd SYDNEY, N. S.W., AUSTRALIA. LARGEST AND MOST COMPLETE FREEZING ESTABLISH MENT IN THE SOUTHERN HEMISPHERE. Mutton and Beef Freezing for Export. Ice Making. Makers and Exporters of Pasteurized and Creamery Butter. Wholesale and Retail Purveyors of MILK, ICE, FISH, BUTTER and ALL PERISHABLE FOOD PRODUCTS. Depots, 92 and 135 King- street; 23 Royal Arcade. Branches at Summer Hill and North Sydney; also at Fremantle, W. A. Central Butter Factory at Grafton. Creameries on the Clarence River, and Illawarra District. Head Office and Works: 25 Harbour Street, SYDNEY. H. PATESON, Manager. FOUNDED BY THE LATE THOMAS S. MORT, A. D. 1854. Mort's Dock & Engineering Co., Ltd. BALMAIN, NEW SOUTH WALES. Engineers, Ship Builders and Machinists. Have Four Dry Docks and Three Patent Slips, Docking Vessels up to 10,000 Tons Burden, With the Most Extensive Engineering Works IN AUSTRALIA. THEY ARE MANUFACTURERS OF Refrigerating Plants for Land and Shipboard. ALSO BUILDERS OF MACHINERY FOR WATERWORKS, MINING, LOCOMOTION, CRANES AND LIFTS, ELECTRIC LIGHTING, QUARTZ CRUSHING, GOLD DREDGING. J. P. FRANKl, General Manager. MACHINERY FOR REFRIGERATION. 389 NORMAN SELFE MEMBER INSTITUTE CIVIL ENGINEERS. ENGLAND. MEMBER INSTITUTE MECHAN- ICAL ENGINEERS, ENGLAND. ASSOCIATE AUSTRALIAN INSTITUTE PATENT AGENTS, ETC., ETC., ETC. :::::::::: CONSULTING ENGINEER SYDNEY, AUSTRALIA MR. SELFE has the oldest established business in the Australian colonies as a Consulting- Engineer, Mechanical and Patent Expert and Arbitrator in Technical matters. He will be happy to be of service to home, American or foreign friends in connection with the Refrig-erating, Hydraulic, Pneumatic and Electrical branches of Engineering', or the design and construction of Machine^' to meet Australian requirements. s AULS' PATENT AUTOMATIC ICE CAN FILLER We present this filler to you in its improved form, and we have a filler that is not theoretical in any way, but is built for hard use and will stand the tankman's thump- ing", and at the same time be accurate and thoroughly reliable. It will save one man's work on a large machine, and makes all blocks of ice weigh exactly alike; prevents waste of distilled water and weakening- of brine; is ad- justable, and made of the best material; threads are standard and repairs are easy. You will not regret fit- ting out your factory with these fillers, and we guaran- tee satisfaction. All the best factories and manufacturers of ice machinery use them. We have made them since 1889, so you see it is no experiment. We solicit your order; have a large stock, and can ship "at once." SAULS BROTHERS, MANUFACTURERS PATTERNS, CASTINGS, MODELS, DRAWINGS AND LIGHT MACHINE WORK. COLUMBUS, GEORGIA. 390 MACHINERY FOR REFRIGERATION. THIRD EDITION Revised and Enlarged Compend of Mechanical Refrigeration A Book of over 400 pages By PROF. J. E. SIEBEL DIRECTOR ZYMOTECHNIC INSTITUTE, CHICAGO 'T'HIS work presents, in a convenient form, the rules, tables, * formulae and directions which are needed by refrigerating" machinery engineers, ice manufacturers, cald storage men, brewers, meat freezing- establishments, packers, contractors and all others interested in the practical application of refrigeration. It is, in fact, designed to give ready and plain answers to most of such ques- tions as are daily occurring in any one of the different branches of practical refrigeration. The most popular book yet written on Mechanical Refrigeration. PRICE Bound in Cloth $3.00 Bound in Morocco 3.50 Sent prepaid to any address on receipt of price. H. 5. RICH & CO...Publishers 177 LA SALLE STREET, CHICAGO 206 BROADWAY, NEW YORK MACHINERY FOR REFRIGERATION. 391 COVERINGS KOR Ammonia, Brine, Cold Water and Steam Pipes SURE AND POSITIVE INSULATION. FURNISHED AND APPLIED IN ANY PART OF THE UNITED STATES. H. W. JOHNS M'FQ CO., ioo WILLIAM ST., NEW YORK SEND FOR PRICKS AND PARTICULARS. COLD STORAGE Sectional and Combination COVERINGS INS L7L A 1 TV. . . for Brine, Ammonia, Water and Steam Pipes. SPECIAL, SECTIONAL AND SHEET MATERIALS, For Pipes, Tanks, Walls, Floors, Refrigerator Cars, etc. T-J AID T7T7T T'Q WATERPROOF SHEATHING PAPERS, ll..r\.iIV ^ m^ i O PAINTS AND CEMENTS, Asbestos Materials, Roofings, Coatings and Coverings. R. M. GILMOUR COMPANY, 84 John Street, NEW YORK. CABOT'S INSULATING- QUILT A cushion of dead-air spaces, absolutely preventing conduction by circulation. Impervious to decay or vermin, and uninflammable. The scientifically perfect insulator. : : : : "An A No. 1 insulating medium.'" Syracuse C. S. it Warehouse Co. " The best insulating medium in use." Express Refrigerator Car Co. SOLE MANUFACTURER . Samples and full details sent on request. SAMUEL, CABOT, TO KILBY STREET, BOSTOX, MAS S FOR COLD ATT IV T S .T RAGE M [\ CHEAP AND AND -.. ^ EASILY ICE HOUSE APPLIED INSULATION \ A/ 1 If 1 USE \\ \^J \^J I j SAMPLES FREE United States Mineral Wool Co. 2 COURTLANDT STREET NEW YORK 392 MACHINERY FOR REFRIGERATION. INDICATING THE REFRIGERATING MACHINE fjt THE APPLICATION OF THE INDICATOR TO THE AMMONIA COMPRESSOR AND STEAM ENGINE, WITH PRACTICAL INSTRUCTIONS RELATING TO THE CONSTRUCTION AND USE OF THE INDI- CATOR AND READING AND COMPUTING INDI- CATOR CARDS. BY ...GARDNER T. VOORHEES THIS book gives simple, practical tables for applying- the true compression curve to the indicator card. So simple are these tables that any man competent to run a compressor can use them. It is as important for the owner of a compression machine to have these tables as it is to know whether he has a hole in his money pocket. Full instruction for attaching- the indicating- apparatus, with numerous cuts of indicator cards, showing- and explaining- faulty running- of the compressor. Also full descriptions, with cuts, of the standard makes of indicators, planimeters, reducing- motions, etc., etc.; together with a practical treatise on the action of the compres- sor; with a full appendix of ammonia tables, etc., etc., and all data necessary to work up indicator cards from the ammonia compressor or steam engine. J Bound in Cloth ..... $1.00 MofOCCO . . . . J.5Q Sent prepaid to any address on receipt of price. H. S. RICH & CO., PUBLISHERS 177 LA SAL.LE STREET, CHICAGO 206 BROADWAY, NEW YORK MACHINERY FOR REFRIGERATION. 393 SAMUEL, H. BRUBAKER & CO. COLD STORAQE ARCHITECTS A^D ENGINEERS AUTOMATIC ICE CAN FILLERS Are necessary adjuncts lo all Ice Factories using the ( Can" System of Freezing. Burns' Can Fillers will put the water up to the proper heig-ht in every can, ever\' time, keep on doing- it, and are so g-uaranteed. MANUFACTURED SOLELY BY JAS. F. BURNS (Can Filler Maker to the Ice Baron) 811-813-815 Fair-mount Ave. PHILADELPHIA, PA. Correspondence Solicited. WEST POINT... W BOILER WORKS MANUFACTURERS OF STEAM BOILERS, STEEL SMOKE STACKS, FREEZING TANKS and BRINE TANKS For Modern Ice Plants R. MUNROE & SON, 23d St., PITTSBURG, PA. 394 MACHINERY FOR REFRIGERATION. ^^^^MM^ pictures tell more than ttfords WE MAKE FINE ILLUSTRATIONS FOR ALL PURP05ES CONSULT U5 ABOUT YOUR NEXT AD. OR CATALOGUE. 3tlinots 346-356 DEARPORN ST. CHICAGO. NOTE THE CUTS IN THIS BOOK MADE BY US FOR FRED W. WOLF CO. PEMN. IRON WKS.CO. HERCULES ICE MCH. CO. THE ICE & COLD CO. OF ST. LOUIS. SAML.H.BRUBAKER. CEO. CHALLONERS SONS CO. MEW SOUTH WALES FRESH FOOD & ICE CO. THE BUFFALO REFRIC. MCH. CO. AMD OTHERS. PHONE HARRISON" CLYDE WELDED CONTINUOUS COILS A SPECIALTY THE CLYDE ENGINEERING COMPANY LJ? Cranville and Sydney, N.S.W REFRIGERATING ENGINE IRS MAKERS OF LINDE, AULDJO.ANTARCTI LYDE, AND OTHER TYPES OF MACHINES. MACHINERY FOR REFRIGERATION. 395 HIGH-CLASS CATALOGUES AND PAMPHLETS A SPECIALTY TELEPHONE MAIN 157O PRINTERS AND STATIONERS BLANK BOOK MANUFACTURERS 337 AND 339 DEARBORN ST. 74 AND 76 PL.TMOUTH CT. PKIXTERS OF THIS WORK. CHICAGO, The Ideal Refrigerating & Mfg. Co* SHEFFIELD and NORTH AVES., CHICAGO. Builders of SMALL MACHINES EXCLUSIVELY One-half to Ten Ton Refrigerating Capacity OUR TOGGLE MOVEMENT, 396 MACHINERY FOR REFRIGERATION. TO OBTAIN THE BEST BUY YOUR AMMONIA, STEAM, GAS OR WATER PACKING OF THE GARLOCK PACKING CO PALMYRA, N. Y. NEW YORK CITY, BOSTON, CHICAGO, ROME, CA. , PHILADELPHIA, SAN FRANCISCO, DENVER, CLEVELAND, ST. LOUIS, PITTSBURGH. JAMES SPIERS, JR. NO. 15 FIRST ST., SAN FRANCISCO, CAL. AGENT FOR THE ANTARCTIC REFRIGERATING MACHINERY MR. SPIERS is prepared to dispose of the Patent Rights in the United States connected with the Ant- arctic Machines, or to furnish working drawings and issue licenses to REFRIGERATING MACHINE BUILDERS, covering the whole or any part of the improvements secured by the patents. :::::: MACHINERY FOR REFRIGERATION. 397 ARCTIC MACHINES INSTALLED IN 1879 STILL IN CONTINUOUS SERVICE COMPLETE ICE MAKINGand REFRIGERATING PLANTS OF ANY SIZE. Pipe. ..Fittings Ammonia Cans. ..Tanks Supplies STYLE B. Correspondence Solicited. Send for Catalogue. THE ARCTIC MACHINE CO, CLEVELAND, OHIO, U.S.A. 398 MACHINERY FOR REFRIGERATION. "CHALLONER" IMPROVED SINGLE ACTING COMPRESSOR. For strength and durability, simplicity in construction and operation they have no equal. FOR SECTIONAL VIEWS AND DESCRIPTION SEE PAGES 311 TO 314. EITHER BELT DRIVE OR DIRECT CONNECTED TO STEAM ENGINE. OUR SPECIALTIES : From J to 30 tons Refrigerating Capacity For Ice Factories, Breweries, Cold Storage Warehouses, Candy Factories, Restaurants, Hotels, Creameries, etc. GEO. CHALLQNER'S SONS CO, We solicit your correspondence and will cheerfully furnish estimates. OSHKOSH, WIS. MACHINERY FOR REFRIGERATION. 399 " Neponset " Insulating Paper From the first this has been the standard among- the leading- Cold Storage experts, for best insulation work here and abroad. "Laminoid" Insulating Paper Of all insulating- papers yet made, this one stands the highest on tests for transmission and absorption of moisture. A POSTAL F. W. BIRD & SON, BRINGS SAMPLES. PAPER MAKERS. Western Office: 1434 Monadnock Bldg., Chicago, 111. Eastern Office and Mills: East Walpole, Mass. CENTRIFUGAL, PUMPS Made of brass, or iron with brass working parts, for cir- culating purposes in con- nection with Refrigerating Plants and Ice Making Ma- chinery. Directly connected pumps and engines, com- pact and substantial : : : : : MORRIS MACHINE WORKS BALDWINSVILLE, N. Y. New York Office, 39-41 Cortlandt St. HENION & HUBBELL AGENTS 61 North Jefferson St., Chicago, III. IMPROVED Robertson-Thompson INDICATOR. FEED WATER HEATERS, HTC. FOOD FOR THOUGHT An INDICATOR will at all times tell you if your engine is working- economically. An ELIMINATOR either on the steam line to separate water or exhaust line to extract oil, is as g-ood as an insurance policy. There are none better than ours, and the prices are very low. _ JAS. L. ROBERTSON & 5ONS NEW YORK. BOSTON. PHILADELPHIA. HINE ELIMINATOR. REDUCING WHEELS. PLANIMETERS. 400 MACHINERY FOR REFRIGERATION. WE HAVE THE LARGEST AND BEST EQUIPPED FACTORY IN THE UNITED STATES FOR THE MANUFACTURE OF GALVANIZED IRON WORK. T TsJ X Brine Tanks Exhaust Steam Filters, Water Filters, Reboilers, Cooling Tanks, Tanks and Vats. And all Sheet Iron Work required in Ice Factories, Breweries and Cold Storages. WM. B. SCAIFE & SONS ESTABLISHED 1802. Office: 221 FIRST AVENUE, Pittsburgh, Pa. MACHINERY FOR REFRIGERATION. 401 WESTERLIN & CAMPBELL CHICAGO, ILL. CONSULTING ENGINEERS AND CONTRACTORS FOR Ice Making and Refrigerating Machinery PATENTEES AND MANUFACTURERS OF THE IMPROVED WESTERLIN & CAMPBELL DOUBLE PIPE AMMONIA CONDENSERS BRINE COOLERS, BEER COOLERS DISTILLED WATER COOLING COILS, ETC. SEND FOR CIRCULARS We own the patents in the United States and Great Britain for these Condensers and Coolers. Patents pending- in other countries. (26) 402 MACHINERY FOR REFRIGERATION. Practical Ice Making and Refrigerating A practical, common sense treatise on the construction and operation of Ice Making' and Refrigerating' Machinery and Apparatus BY EUGENE T. SKINKLE "THE BOY" Every branch of ice making" and refrigerating 1 is handled with a view to setting out the best and most economical practice in the construction and operation of the plant. The benefit of years of experience in the construction of ice making and refrigerating plants and the erection of their machinery, as well as study of their operation from the practical side, is given to the trade in plain language, free from technicalities, and will be found of great practical value to owners and operators alike, and of exceptional value to those about to erect new plants or to rearrange or overhaul old establishments. ^ Bound in Cloth, .... $1.50 c "/ Bound in Morocco, .... 2.00 SENT PREPAID TO ANY ADDRESS ON RECEIPT OF PRICE. H.S. RICH &CO. PUBLISHERS 177 La Salle St., Chicago 206 Broadway, New York MACHINERY FOR REFRIGERATION. 403 : GEO. J. STOCKER MANUFACTURER OF COOLING TOWERS (PATKXT, JOHN STOCKER.) Apparatus for the Re-Cooling of Ammonia and Steam Condenser Water. Saves from 90 to 95 per cent of the water required for Condensing and Cooling Purposes. Owing to the superior construction of the cooling: surfaces (about 40 per cent larger than with the Gradirworks, patent Klein) and the most perfect methods of distributing the water, the efficiency of this Cooling Tower is greater than with any other in the market, and the temperatures obtained considerably lower. References from leading firms all over the United States. Information and estimates, etc., cheerfully furnished. 2831 Victor Street, St. Louis, Mo. 404 MACHINERY FOR REFRIGERATION. The Recognized Authority In all matters pertaining to Mechanical Refrigeration A MONTHLY REVIEW OF THE ICE, ICE MAKING, REFRIGERATING, COLD STORAGE AND KINDRED TRADES. 'P'HE oldest publication of its kind in the world, and the only medium through which can be obtained all the reliable technical and practical information relat- ing- to the science of mechan- ical ice making and refriger- ation. ICE AND REFRIGERA- TION is invaluable to any one owning, operating, or in any way interested in ice making or refrigerating ma- chinery. It has won the confidence of all classes of the trade throughout the world by its absolute independence and impartiality. It aims, to be a thoroughly representative paper, catering to no particular class, but striving to become indispensable to all. It is not shackled by any pet theories, and no man or class of men has any private pull with it. Its columns are open to the entire trade: to any one who has anything of interest or value to say. SUBSCRIPTION PRICE. In United States, Canada and Mexico, . In all other countries, Payable in Advance. Remit by postoffice or express money orders, or by bank draft on Chicago or New York. $2.00 per year. 3.00 per year. H. S. RICH & CO., Publishers NEW YORK: 206 Broadway Corner Fulton CHICAGO: J77 La Salle Street Corner Monroe MACHINERY FOR REFRIGERATION. 405 Ice and Refrigeration (ILLUSTRATED) A Monthly Review of the Ice, Ice Making-, Refrigerating, Cold Storage and Kindred Trades OFFICIAL ORGAN OF THE SOUTHERN ICE EXCHANGE, THE NORTHERN ICE MANUFACTURERS' ASSOCIATION, THE SOUTHWESTERN ICE MANUFACTURERS' ASSOCIATION, THE INDIANA ICE MANUFACTURERS' ASSOCIATION, THE FLORIDA ICE MANUFACTURERS' AS- SOCIATION, THE TRI-STATE ICE MANUFACTURERS' ASSOCIATION, THE WESTERN ICE MANUFACTURERS' ASSOCIATION, AND ILLINOIS ICE MANUFACTURERS' ASSOCIATION. It gives the earliest reliable information of improvements in machinery and appliances for handling or making ice or for producing cold. The department of "Answers to Correspondents" is one of the most valuable features, and is open to every subscriber for the presentation of the problems encountered in daily practice of making ice or operating cold storage or other refrigerating plants, or for the elu- cidation of scientific and theoretical questions. Every legitimate inquiry is fully answered by experts; we have personal knowledge of scores of cases where the use of this department by subscribers has been the direct means of saving them large sums of mone3 r , as well as of enabling them to save still more by its suggestions of better methods for constructing plants and operating their machinery, based on scientific investigation as well as practical experience. You cannot Afford to Be without It. SUBSCRIPTION PRICE. IN U. S., CANADA AND MEXICO, . . . $2.00 PER YEAR IN ALL OTHER COUNTRIES, .... 3.0O PER YEAR PAYABLE IN ADVANCE. H. S. Rich & Co. PUBLISHERS 206 Broadway, New York 177 LaSalle St., Chicago 406 MACHINERY FOR REFRIGERATION, ICE MAKING AND REFRIGERATING MACHINES LOWEST FUEL CONSUMPTION. SURPASSES ALL FOR STABILITY, DURABILITY AND ECONOMY. Write for Circulars and List of Customers. C. A. MAcDONALD, MONADNOCK BLOCK, CHICAGO, ILL. MACHINERY FOR REFRIGERATION. 407 THE BALL MACHINE 500 TOXS FOR LAHGE INSTALLATIONS. THE BEST DESIGNED, MOST ECONOMICAL AND GENERALLY SATISFACTORY MACHINE ON THE MARKET. FRICTION LOAD 6/0 % ICE & COLD MACHINE Co ST. LOUIS, MO., U.S.A. 408 MACHINERY FOR REFRIGERATION. Wrought ...Iron Pipe COILS OF EVERY DESCRIPTION. BENDS AND MANIFOLDS FOR si- and Refrigerating Machinery Whitlock Coil Pipe Company The Cable and Telegraph Address MAIN OFFICE AND WORKS ..WHITLOCK?' H A ELMWOOD, CONN, Director}' Code U. S. A. Iron, Brass and Copper COILS OF ALL KINDS FOR HEATING AND COOLING. efe MACHINERY FOR REFRIGERATION. 409 LINDE REFRIGERATING AND =ICE MACHINES OVER 4,100 MACHINES SOLD. Improved Air=Cooling Apparatus Refrigerating and Freezing Machines For Meat Freezing and Chilling Establish- f or a11 purposes, ments. and for Cold Stores, giving a per- Ice Making Machines feet circulation of cold dry air at any For the economical production of white, temperature. clear or crystal ice in blocks of any size. Pure Anhydrous Liquid Ammonia and Chloride of Calcium in Stock. THE LINDE AUSTRALIAN REFRIGERATION CO. Offices 11 A. M. P. Buildings, BRISBANE. 97 Pitt St., SYDNEY, AUSTRALIA. The Brewers Hand-Book THE ORIGINAL DIRECTORY OF THE BREWING AND MALTING TRADES. THE BOOK GIVES Published Annually. A complete list of all Brewers in the United States. A complete list of all Brewers in Canada. A complete list of all Brewers in Mexico. A complete list of all Brewers in Central America. A complete list of all Brewers in South America. A complete list of all Brewers in the West Indies. A complete list of all Brewers in Australasia. A complete list of all Brewers in India. A complete list of all Brewers in China. A complete list of all Brewers in Japan. A complete li.st of all Brewers in Africa. A complete list of Brewmasters in the United States. A complete list of Brewmasters in Canada. A complete list of all Maltsters in the United States. A complete list of all Maltsters in Canada. A complete list of all Maltsters in Australasia. A complete list of all Grain Distillers in the United States. A complete list of all Grain Distillers in Canada. All Brewers that Bottle are designated. All Brewers that make Malt are designated. The kind or kinds of Malt Liquors brewed by each Brewer is shown. Also a vast amount of miscellaneous trade information. PRICE, 85.00 A. YEAR. H. S. RICH & CO., Publishers, 177 LA SALLE ST., 206 BROADWAY, CHICAGO, ILL., U. S. A. NEW YORK, U. S. A. 410 MACHINERY FOR REFRIGERATION, York Manufacturing Co YORK, PA. DESIGN OF OUR MEDIUM SIZE MACHINE. Manufacturers of ICE MAKING AND REFRIGERATING MACHINERY ALSO ENGINES AND BOILERS. 400 TON REFRIGERATING MACHINE. TWO SINGLE-ACTING AMMONIA COMPRESSORS. 30" DIAMETER BY 48" STROKE. CROSS COMPOUND CONDENSING STEAM ENGINE. HIGH PRESSURE CYLINDER 30" X 48" LOW PRESSURE CYLINDER 58" X 48" MACHINERY FOR REFRIGERATION. 411 WE ARE PREPARED TO FURNISH TO THE TRADE ANY APPARATUS OR FITTINGS USED IN MACHINERY FOR THE MANUFACTURE OF ICE QR FOR REFRIGERATING PURPOSES PARTIAL VIEW OF OUR WORKS. Our Works are Conceded to be the Most Modern in Existence XT! h Our Own Foundries ANY KIND OF AMMONIA FITTINGS AND CASTINGS MADE OF Charcoal Iron, Malleable Iron, Gun Metal or Semi-Steel. YORK MANUFACTURING Co YORK, PA. CAPITAL, : : : Sl,OOO,OOO. 412 MACHINERY FOR REFRIGERATION. PENNEY'S TWIN CONNECTED ICE MACHINE WITH CORLISS ENGINE. EDGAK PKNNEY, President and Manager. ROBERT WHITEHILL, Sec. and Treas. A. B. WHITNEY, Vice-President. GEORGE B. SALISBURY, Auditor. Newburgh Ice Machine and Engine Co* ICE MAKING AND REFRIGERATING MACHINERY Using Ammonia or Sulphurous Oxide. Corliss Steam Engines, Simple or Compound, for any duty. Steam Boilers and Steam Power Equipments. Iron and Brass Casting's. Address. NEWBURGH, N. Y. PENNEY'S TANDEM HORIZONTAL, DOUBLE ACTING, CENTER CRANK COMPRESSOR, CORLISS STYLE. MACHINERY FOR REFRIGERATION. STEVENSON'S DOORS FOR COLD STORAGE AND AIR-TIGHT STORAGE. SMOKE-TIGHT FIREPROOF DOORS. These doors have won for themselves a reputation as the highest standard. All the finest Cold Storages in the United States are fitted up with them. They are made of the best of everything used, and their appearance is neat and elegant. They include our hardware and our Adjustable Flexible door frame, all fitted up complete and adjusted, ready to push in place, screw fast and use. Where truck- ing is done they have our improved beveled threshold, avoiding faulty sealing, binding on floor, and constant sweeping up, jolting, splinters, etc. For cement or as- phalt floors, the lower ends of frame are con- nected by angle irons- bedded in the floor be- low the surface. For overhead track they have a tight fitting trap, opened and closed by our new, positive acting cam device. No- Cord Pulley or spring hinge, and are all complete in one structure. Freezer doors. Metal covered, smoke-tight, fireproof doors. Combined self closing door and chute to pass ice in or out of stor- age. Nothing perishable about it. No rush of air. No trouble with careless help. Will count the passing blocks. Full information, illustrations, diagrams, order forms and long lists of patrons in all lines of business, in our circulars. 414 MACHINERY FOR REFRIGERATION. MACHINERY FOR REFRIGERATION. 415 FIFTEEN TO TWO HUNDRED TON ICE MAKING. 416 MACHINERY FOR REFRIGERATION. PURE SULPHATE MADE Anhydrous ABSOLUTELY DRY. PURE SULPHATE MADE 26 Aqua SPECIALLY PURIFIED FOR USE IN ICE AND REFRIGERATING MACHINES MANUFACTURED BY ^ prerichs (Chemical ST. LOUIS, U. S. A. The Vilter Manufacturing Co. 860-870 CLINTON ST., MILWAUKEE, WIS., U.S.A. BUILDERS OF IMPROVED HORIZONTAL, DOUBLE- ACTING COMPRESSION REFRIGERATINGCtlCE MAKING MACHINERY DOUBLE-ACTING AMMONIA COMPRESSOR, DRIVEN BY "VILTER" CORLISS ENGINE. For "DIRECT EXPANSION" or "BRINE CIRCU- LATION" SYSTEMS. COMPLETE LINE OF AMMONIA FITTINGS. CORLISS ENGINES, A , OVERDUE. nd AMMONIA Pennsylvania Iron Works Co* Refrigerating Machinery SEND FOR ILLUSTRATED DESCRIPTIVE CATALOGUE, UNIVERSITY OF CALIFORNIA LIBRARY BERKELEY THIS BOOK IS DUE ON THE LAST DATE STAMPED BELOW Books not returned on time are subject to a fine of 50c per volume after the third day overdue, increasing to $1.00 per volume after the sixth day. Books not in demand may be renewed if application is made before expiration of loan period. DEC 2. URI NOV19 1 SEP 3EC 6 I ICAL ENGINE. ia, Pa."' Broadway, New York, U.S. A.