CHAPTER VIII HOUSEHOLD REFRIGERATING MACHINES (Part 2)

in whose presence the ammonia evaporates. The machine as designed refused to work and Geppert has himself seen the drawbacks of a design in which the ammonia has to dittuse through a thick layer of inert gas. This is apparent from other draw- ings later on submitted by him. In the next design he reduced the thickness of this layer and emplo_\ ed a fan to aid the evapo- ration, by having the lower parts of the fan blades dip into the liquid. By doing this he was able to further separate the cold evaporator from the warm absorber therebv' reducing the refrigerating losses. The fan had to be operated by a motor and this was one of the pieces of equi])ment that (je]jpert had set out to eliminate. Another reason \vh}- the machine proved impractical was that when the ammonia \ ap(ir> are absorbed 310 HOUSEHOLD REFRIGERATION by water heat is liberated. The absorber therefore acts as a heater. Cooling water had to be provided in pipes located within the absorber. Notwithstanding this, heat liberated during the absorption rose into the evaporator space above, thus either entirely or partially counteracting the heat with- drawn from the surroundings by the evaporation of the am- monia. Effective refrigeration therefore cannot take place, or only in a very limited degree. In the year 1901, Geppert attempted another design which was somewhat of an improvement over the two previous models. He still maintained the combined vaporizer and ab- sorber. The receptacle was provided with a double. wall, cool- ing water circulating in its hollow space. Into the receptacle is inserted a cylinder at a very slight distance from the inner face of the double wall of the receptacle. The cylinder con- tains salt water. The cylinder does not extend to the bottom of the receptacle. Into the free space flows from the boiler "poor solution." The operation was as follows : The ammonia which had become expelled from the rich solution and liquified in the condenser flows through a small pipe to the outer surface of the wall of the cylinder, which wall is covered with porous material. There the ammonia becomes distributed and evaporates. Heat is therefore with- drawn from the salt water contained in the receptacle and cold is produced. The produced ammonia vapors diffuse through the small intermediate space to the opposite inner surface of the double wall of the receptacle. This surface is sprayed by poor solution which by means of a small pump is continually pumped through the pipes from the lower portion of the receptacle upwards into the space between the double wall of the receptacle and the c\ linder. The surface on which the poor solution flows down is cooled by the cooling water in the hollow space of the double wall. As the poor solution flows down, it absorbs the ammonia gases which are diffused in its direction from the opposite surface and thereby is en- riched with ammonia. Thus the outer surface of the cylinder acts as a vaporizer, and the inner surface of the wall of the receptacle as an absorber. The absorber therefore, is, not like in his previous patent, below the vaporizer, but the absorber ABSORPTION REFRIGERATING MACHINES 311 and vaporizer located in one and the same vessel, at the same level side by side. It will be noted that Geppert had to take recourse to the pump, which is one of the pieces of equipment he started out to eliminate. While he succeeded in producing cold, with his last design, the efficiency was small — also he failed to attain one of his objects, namely to eliminate the pump. After Geppert failed, no trace can be formed of any prac- tical and useful small refrigerating machine of any importance, operating according to the absorption principle in a continuous manner, until the year 1922 when two students, Baltzar Carl Von Platen and Carl George Munters of the Royal Swedish Institute of Technology developed and designed a working model which dispensed with all moving and mechanical parts. This unit was later developed by the Electrolux Aktie- bolaget in Europe and the Electrolux Servel Corporation in the United States, so that today we have a workable and saleable refrigerating unit that is indeed marvelous. In order lo develop the present day product a large laboratory for re- search work was established in Stockholm, Sweden, and in Brooklyn, New York. In these laboratories developments and experiments are taking place so as to develop new types for further commercial applications. How well this unit with refrigerator has been developed was evidenced at the American Gas Association Convention at Atlantic City, where three complete refrigerators and an exposed unit were presented for the inspection of the gas industry. Platen-Munters, independent of Geppert had like him the idea to have in the entire system, by the introduction of a second gas, everywhere the same uniform total pressures and to efifect the pressure difference required for the vaporization of the refrigerating medium. Contrary to Geppert, however, they carried out this idea in a manner which at once resulted in a practical solution. They recognized what has remained concealed to Geppert that in such a system into which is in- troduced a pressure compensating gas there occurs within the system inner fores, i. e., physical actions which can be utilized in order to efifect the circulation required for such a system. Furthermore, they recognized that this peculiar action can be 312 HOUSEHOLD REFRIGERATION still considerably im])r()Vfd upon if the pressure compensating gas possesses special characteristics, for instance, as regards its specific weight differing considerably from that of the vapors of the refrigerating medium. There were ways of avoiding the pitfalls of Geppert. First.- — As regards the puni]). B}- applying heat from a source the solution rich in ammonia, is made to boil in a tube and by the thermo-syhon action thus established, the liquid is raised from the lower level of the absorber to the high level of the generator. Second. — Stagnation and poor circulation, instead of using air as Geppert had done hydrogen gas was used. Absorber and evaporator are placed at about the same level or the latter somewhat higher than the former. When the ammonia vapor has been absorbed in the absorber, pure hydrogen flows through the upper ]3ipe into the evaporator, where it mixes with the vapors from the evaporating liquid ammonia. The mixture of ammonia vapor and hydrogen being specifically lighter the greater its ])ercentage of h}clrogen, it follows that the column of gas in the evaporator will be heavier than that in the absorber. An automatic circulation of gas consec{uentl} takes place, giving an upward fiow in the absorber and a down- ward flow in the eva])orator. If instead (jf hydrogen, the inert gas had been nitrogen the flow would have been reversed. The api)aratus comprises the generator, condenser, evapo- rator, absorber, heat exchanger, the^'mo-syphon, which are interconnected by pipes. In all portions of the completely and tightly closed apparatus exists the same total pressure. The boiler is to a large extent filled with aqua ammonia (so called rich solution) only the upper vapor space of the boiler is free from liquid. Into the Ijottom of the generator is inserted an electric heating element, connected to a source of electrical energy. From the upper free space of the generator leads a pipe to the condenser and said pipe continues on into the top of the evapo- rator. The latter is filled with hydrogen gas. At the bottom as well as at the top there is a connecting pipe to the absorber. The absorber i> surrounded b}- a jacket through which circu- lates cooling water, \hicli then passes to the condenser. From ABSORPTION REFRIGERATING MACHINES 313 the bottom of the abs()rl)ei- a ])ii)e runs to and coils around the heating element. In addition there is a i)ii)e connecting the bot- tom of the generator with tlie top of llie abscjrber. This i)ii)C where it is horizontal surrounds the pipe which i)asses from the absorber to the upper space of the generator. The apparatus is (jperated as follows: Current sui)]jlied to the heating- element heats the rich solution in the generator. The ammonia gas expelled from the ricli solution fills the upper free space of the generator and flows througli the pii)e into tlie water cooled condenser. Because of the cooling the hot ammonia gases are condensed to pure ammonia liquid. This licjuid flows to the eva])orator. There in the presence of h}(lrogen the liquid ammonia exaporatcs. Through the e\a])o- ration heat is withdrawn from the brine tank surrounding the evaporator and consequently cold is produced. The am- monia gases in the evaporator diffuse into the hydrogen, and the mixture, sinks downward, liecause as compared to the gas mixture in the absorber it is lieavy. In its downward move- ment it passes on to the absorber, fn this \essel the gas mix- ture meets with the water (i)oor solution) coming from the generator. The liquid level in the generator is higher than the "poor solution" pipe to the absorber; there is therefore, a continual flow of poov liquid into the absorber. In the ab- sorber the poor solution absorbs from the gas mixture the ammonia gas and collects at the bottom of the absorber as "strong or rich liquid," while the lighter hydrogen free from ammonia ascends and through the pipe connecting the ab- sorber with the evaporator, again enters the top of the evapo- rator. The rich solution is conveyed through the coils of the pipe around the lower end of the heating element and is thereby preheated so that the ammonia gas bubbles around the pipe. These bubbles carry along globules of liquid, which thereby reach the upper portion of the generator, from which we began the cycle of operation. Outside the cycle of the ammonia which takes place in all four vessels (generator, condenser, evaporator and absorber) there occurs in the apparatus still two other cycles. On the one hand the circulation of the water, or poor solution from the 314 HOUSEHOLD REFRIGERATION bottom of the generator, to the top of the absorber down to the bottom of that vessel as strong solution, then to the top of generator, through the thermosyphon pipe. On the other hand, the circulation of the hydrogen gas from the bottom of the evaporator to bottom of absorber, and from top of ab- sorber to top of evaporator. The above description covers the machine as originall\ designed. Rarely, has an invention required less time to per- fect. On August 18, 1922, the first patent was deposited in Sweden and in 1925 a great number of refrigerating machines were in commercial service. The Electrolux Servel Cori)oration has by exhaustive tests and experiments developed a machine somewhat different than the original Swedish design. These changes have practically doubled the "ice melting capacity of the machine" and have greatly increased its efficiency. They have in addition per- fected the machine for gas heat instead of electric heat and have reduced the quantity of cooling water needed to properly operate the unit. In order to do this several changes had to be made to the apparatus. 1. Inner flue placed in generator to permit the use of a gas flame for heat. 2. Rectifier — to catch water that may be carried over with aninionia gas. 3. New type condenser — simplified construction. 4. Gas heat exchanger — placed between rectitier and the absorber and the evaporator, where ,the cold ammonia hydrogen mixture coming from the evaporator is warmed by the hot ammonia coming from the rectifier and the hot hydrogen from the absorber. 5. The liquid heat exchanger. The two pipes located between the absorber and the generator the one being placed inside the other, act as a heat cxcJianger on the counter flow principal, by means of this the hot, weak liquid, which flows from the bottom of the genera- tor into the absorber, is pre-cooled by the comparatively cool strong liquid that flows from the absorber to the thermo-syphon. Thi- solution is at the same time pre-heated before entering the generator. The unit before being charged with ammonia, distilled water and hydrogen is given a careful air and hydraulic test under the most rigid factory supervision and after being charged is heremetically sealed by welding. The original charge does not have to be renewed, as there is no leakage. ABSORPTION REFRIGERATING MACHINES 315 The unit is equipped with a thermostatic safety burner which automatically shuts off the gas supplied if for any rea- son the supply is interrupted. One of the features of the unit is that the operation involves absolutely no danger even if the condenser water supply should be interrupted for any length of time. Inasmuch as there are no moving parts and being rigidly constructed, no serviceing is necessary, and that is saying a lot. The refrigerator is a steel box of approximately 6>4 cubic feet of food space — finished with several coats of duco over baked white lacquer. The cooling section inside the box is of cast aluminum having five trays with a capacity of about fifty cubes. The box is insulated with three inches of high grade corkboard, thus bringing the thermal losses and operating costs down to a minimum. (From address delivered by F. E. Sellmann before the New York Section of the American So- ciety of Refrigerating Engineers in October, 1926.) The original ice melting capacity of the Swedish machine was about forty-five pounds per twenty-four hours while its thermal efficiency was about 18 per cent. The Swedish public were apparently content to utilize a manually controlled ma- chine, the control simultaneously regulating both gas and water. It was found that in order to make the machine salable in this country it would have to be designed so as to operate and give sufficient refrigeration where room temperatures of 100° F. and cooling water of 90° F. were encountered. It further had to be developed so that the machine would have to give desired refrigerating effect automatically, and with controls making. the unit serviceable for use with either manufactured gas, natural gas, electricity or oil. A laboratory was established in Brooklyn where exhaustive developments were made by united effort of engineers of the American and Swedish companies. During the next 3^ear these men were able to redesign the machine so as to bring about an ice melting capacity of seventy-five pounds per twenty-four hours and to raise the efficiency to 32^^ per cent when operated by gas. When operated electrically the efficiency rose to 38 per cent. The machine was also capable of producing sufficient 3]6 HOUSEHOLD REFRIGERATION refrigerating effect to take care of the designed refrigerator under conditions of 100° F. room temperature and 90° F. cooling water. The maximum efficiency of the original Swedish unit was reached when an input of 730 B.t.u.'s per hour was furnished, while the maximum efficiency of the machine developed in America was reached when 1350 B.t.u.'s were used. The capacity reached its maximum at aliout 1300 B.t.u.'s Avith the Swedish machine. l)ut with about 1650 for the American ma- chine. These improxements l)oth as to cajjacity and efficiency were brought about by many developments including a new type of rectifier, improved Thermo-syphon, and the use of a gas heater exchanger. The figures quoted above were fur- nished from tests conducted b}- the Consolidated Gas Com- pany of New York. As the efficiency and cai)acity increased with the increase in B.t.u.'s furnished the unit, it was therefore necessary to con- trol the heat in]:)ut so as to get a i)redetermined refrigerating effect. Using gas of 540 B.t.u. ])er cubic foot heating value the minimum gas required to assure satisfactory i)umping througli the Thermo-syphon was l}^ cubic feet per hour and this flame had to be increased to a maximum of three cul)ic feet per hour when maximum refrigeration was desired. This meant there- fore a development of a burner that would burn satisfactorily between the ranges of iVj cubic feet and three cubic feet per hour and that the burner in addition must be of the safety type so that, if for an" reason the gas flame were extinguished that the gas supply to the burner would be automatically shut off". The first burner developed possessed these characteristics but was designed for a gas pressure of about 2}^ inches of water and 540 B.t.u. gas. U'ith the sending of refrigerating units into districts where gas pressures and B.t.u. values vary con- siderably it was necessary to develop burners suitable for both water coke-oven and natural gases and to test and ap])rove gas pressure regulators. As the minimum gas required at an" time was 1^ cubic feet per hour the gas thermostat was therefore designed so as to always allow that quantity of gas to pass through it. but when the thermostat acted on the gas supply it augmented ABSORPTION REFRIGERATING MACHINES 317 gradually the flow until three cubic feet capacity was reached. The thermostat is of simple construction easily set and ad- justed. It consists of a six inch bulb located within the food chamber. The operating- mechanism of the thermostat is located in the machine compartment and is inter-connected by capillar) tubing. The bulb is partly filled with a licjuid which when expanded int«j a gas, actuates by pressure throug-h the tube a diaphragm located in the body of the thermostat. With operating the machine electrical!} . similar conditions must of course be taken care of so that the machine will con- tinually ])ump. With this in mind a double lieating' element was developed which furnished a minimum wattage to kee]) up pumping but increased the wattage to take care of maxi- mum load. From the consumption curve of the cooling water needed it will be noticed that after a certain amount of water had been used further increase in water consumption becomes unnecessary and wasteful. This therefore clearl- indicated