It was discovered that at some point, between the extremes of temperature, the oil was subjected to the proper degree of heat, a good quality of gas being pro duced with efficient results. But this point was quickly reached and more quickly passed, and the results ob tained were lost in the aggregate. By a process of elimination, we may assume that if one temperature is right for reforming the oil, all the rest must be wrong. Working on this theory, the effort has been toward the maintenance of a uniform temperature, thus treating all the oil alike. The greatest mechanical development tending to unify the heat to which the oil is subjected was the improvement of the two-shell machine with the gas offtake located between the point of combustion of the oil used for heating the generator and the stack valve. It has been definitely proved not only by ex perience, but by careful tests, that the heat of the checker brick or heat reservoirs immediately adjacent to the stack valve and also the point where the initial combustion of the heat oil takes place is "false heat" or "surface heat," and is not the substantial heat which we find in the checker brick in the central portion of the generator. This is due to the fact that when oil is injected into the machine together with a forced blast of air for heating the apparatus, combustion is not immediately complete, but the heat of the checker brick is utilized in a gradually lessening degree to convert oil into a gaseous or more easily combustible state until the combustion is complete. The following three sam ples of combustion products will show more clearly : C, N, Sample No. 1 14.2 0.0 85.8 Sample No. 2 15.0 0.0 85.0 Sample No. 3 15.4 0.0 84.6 These samples were taken simultaneously at in creasing distances from the combustion chamber, the increasing percentage of CO2 showing the combustion to be more nearly complete some distance from the initial point of combusiton. In order to minimize the loss of heat at the stack, the combustion should be reg ulated so that the maximum liberation of heat is in the central portion of the checker work and the checker work nearest the stack gets only the end of the com bustion, so that while the brick there may attain a high temperature, they do not contain any great quantity of heat.
A result of these peculiar combustion conditions is the inverse variation of the temperatures in the two ends of a generator if the combustion end is excess ively hot, the stack end will be proportionately cool. From this fact, the two-shell machine, with the gas offtake located at an intermediate position, derives its undisputed advantage over all other forms of gener ators, the natural balance of heat in the two shells maintaining a uniform quality and production of gas at all times.
In the two-shell sets where oil is injected with steam into the top chamber of both the long and short shell simultaneously, and the gas offtake is located in the lower half of the long shell, the first temperature to which the oil is subjected is not the maximum temperature. The checker brick in the upper por tion of each shell, with which the oil first comes in contact, do the greatest amount of work and are the regulators which limit the quantity of oil per run, but the temperature of these bricks does not con stitute the fixing or superheating temperature to which the gas is subjected in the central portion of the long shell near the offtake. While the temperatures in the upper portion of the checker brick in each shell, to which the oil and steam are first subjected, vary several hundred degrees Fahrenheit during a run, the super heating or fixing temperatures remain fairly constant.
Mr. L. B. Jones of the Pacific Gas & Electric Com pany recently conducted at San Francisco a series of experiments in order to reduce the amount of lamp black per 1000 cu. ft. of gas made. He worked along the line of combining into one process what was for merly accomplished in two ; with two spare double shell sets he was able to develop what is now known as the "Improved Jones Oil Gas Set." It will be noticed that this set has a long and short shell connected at the bottom by a large throat piece, the top of the short shell being the blast inlet, and the top of the long shell the stack valve. The common gas offtake is located on the middle of the side of the long shell. The bricking and checker-work are so arranged as to form double chambers in the upper end of each shell. Short piers in the bottom of each shell support the main checker up about two-thirds of the height of •the short shell and three-fourths the height of the long shell. At this point, open arches are sprung across the shell forming the tops of the lower cham bers and supporting the upper sets of checker-work. About twelve courses of checker-work rest on each of these arches, and the top of this checker, and the corbel work, forms the top chamber of each shell.
Into the top chamber of the primary or short shell extend the oil burners for heating the apparatus, con nected to coils of pipe encircling the shell. In the same manner, the injectors for admitting oil for gas-making are connected into the lower primary chamber. To the top chambers of each shell are connected gas lines for the admission of gas under pressure for producing the catalyptic atmosphere. This supply is regulated by valves controlled from the operating floor. Into these chambers also are steam connections for supplying steam for gas-making, and also for purging. After the machine is properly heated and ready for a gas making period, the stack valve and blast valve are closed, and the gas and steam under pressure are ad mitted into the top chambers of each shell.
During the first minute, no oil is admitted, and all the products of combustion from the previous heating period are purged from the machine. Thus at the end of the first minute, when oil is admitted into the lower chambers of each shell, it comes into contact with an active atmosphere of gas and steam, highly superheated by passing through the upper section of checker brick. The decomposition or destructive distillation of the oil therefore begins and is continued in an active atmos phere, and when the excess carbon is freed by the re formation of the hydro-carbons, it is surrounded by steam in a highly superheated condition ideal for dis sociation and combination with this carbon.
With this new process a much better gas is made and the lampblack has been reduced to about 5 lb. per 1000 cubic feet of gas. It has not only effected a sav ing in the oil required, but it has eliminated a great deal of the smoke that was a nuisance around gas plants. It may seem strange to note that the catalyptic atmos phere is produced by using gas direct from the storage holders, but the results that were obtained proved that it was the proper thing to do.
The following report was taken from a Jones set at the Metropolitan works in San Francisco :