CARfiURETION in this paper will be confined to the induction principle-that is, one in which the flow of air and fuel is induced and maintained by air displace ment. It will therefore be apparent that the flow must vary from zero when the piston is at rest to max imum veloeity when the viston has attained its m,lxi mum velocity. As soon as the piston begins to move on its induction stroke, air displacement occurs and the pressure in the passages between the carburetor and piston head become

more »

... tive, decreasing gradually until maximum velocity is reached. The simple carburetor consists of a tube of constant diameter, having a fuel jet in communication with the air flow. As the quantity of both air and fuel vary as the square of the tube diameter, it would seem nec essary only to so proportion the tube and fuel duct to attain a proporional mixture for all air velocities. If one were dealing with a perfect gas or were able to so deliver the fuel to the air stream thoroughly atomized, homeogeneously mixing it with the air with out any expenditure of energy of the air stream, a pro portional mixture of the air and fuel for all speeds might be expected. This, however, is not possible as frietion head, viscosity and other retarding influences that vary with the velocity must be contended against. \Vhy does a simple carburetor seemingly deliver an increasingly richer mixture as the air velocities in crease? This in the opinion of the writer is due to inefficient atomization when air velocities are low and lack of thorough saturation of the air stream with the fuel molecules. To offset this condition fuel must he admitted under less retardation, or in other words, the fuel orifice must be increased. As the air velocities inf"rease, so also does the efficiency of atomization. The inf"rease in the fuel orifice, which was necessary for low air velocities now becomes excessive, consequently thp quantity of fuel molecules becomes greater as the air vploeities increase, resulting in an over-rich mixture. Th�> Rimple carburetor must of necessity be limited in its range of capacity and is at best w�steful of fuel. A combination of the mixing valve and the simple tube carburetor becomes the foundation of our present instrument. The range of the mixing valve is far more limited than the simple tube, the tendency being toward rapid impoverishment as the air 'velocities decrease with air-valve opening. It will be noted that its action is the reverse of the simple tube carburetor. The mixing valve delivers an increasingly weaker mixture as the air quantities increase, while the tube delivers a richer mixture under the same conditions. Attempts to correct the tendency to over-richness ex hibited by the simple carburetor led to the early adop tion of the auxiliary air valve. The popular concep· tion of its function seems to be that of correction by diluting with air the over-rich mixture delivered by tllP simple tube portion of the carburetor. This is true, but its dual function is to not only add an extra amount of fresh air, but more particularly effect a modification of the air velocities at the fuel jet; that is, it increase3 the air supply, which decreases the tension on the air flow at the jet and consequently effects a reduction of inspiration, resulting in a weakened mixture. As will be apparent from the foregoing, the auxiliary portion, be it spring or weight, can serve but one pur pose---that is, to correct an over-rich mixture and at one point at a time. It therefore follows that no adjustment of spring tension can do more than slightly modify this ten dency toward impoverishment of the mixture, while the addition of various forms of subsidiary springs becom ing operative only at some point of valve opening can do no more than correct the air at one given point and then start, as it were, merely a new scale of errors. This is not only true when springs are used, but is an una voidable fact when correction is attempted in this manner by any means. The multiple-jet type may be compared to the aux iliary-air type, being an attempt at correction by the addition of jets as needed, and is subject to the same criticism as pointed out in the auxiliary air-valve con struction. Results approa�hing perfection may be at tained by this me thod, but of necessity they must be sensible in construction. ployed, with consequent troubles from water and for eign matter in the fuel. The variable-fuel orifice type, in which a tapered valve is withdrawn from the fuel orifice in relation to the quantity of air paSSing through the carburetor, in principle approaches the ideal. However, when one con siders that the fuel and air ratio by volume is about one to eight thousand, and as this method is a direct graduation of the one part, its sensitiveness will be apparent. The successful carburetor must accomplish as fully as possible atomization of the fuel at all working air velocities and a thorough saturation of the air stream with fuel molecules, consistent with volumetric effi ciency of the motor, and must so deliver the mixture to the manifold. It .must be simple, with no more moving parts than absolutely necessary, and its princi ple must consist in controlling the air pressures which directly effect in spira tion. To successfully utilize kerosene or other fuel oils in an internal-combustion motor of the carburetor type, it becomes necessary to deliver the fuel charge to the combustion space thoroughly saturated with the fuel molecules in a correct proportion for all air velocities. 'l'he accomplishment of this depends upon several fac tors. The correct design of carburetor would be one in which the air velocities must be maintained sufficiently high at all speeds to effect a thorough atomization of the fuel stream from the jet and at the same time not so high as to cause wire drawing, resulting in a partial charge and loss of power. The writer would divide the process of carburetion of air and fuel into three stages, the carburetor being responsible for the first stage, the manifold and inlet llassages the second stage, and the period of compres sion the third and final stage. Assume that the mixture has been delivered to the manifold in a thorough state of saturation, which is the completion of the first stage. '1'hroughout the sec ond stage this thorough admixture of air and fuel not only must be maintained, but must be assisted by llll" tial vallorization and delivered to the third stage. In the third stage occurs gasification due to compression, whic-h accomlllishes a closer commingling of the fuel molecules with the air, resulting in rapid flame propaga tion. This is assisted materially by the rise in tem perature due to compression and heat from the cylinder walls. Summarized, we find: First stage, atomization; second stage, vaporization; third stage, gasification. Where these three stages are accomplished successfully, the use of fuel oils is possible. The temperature of the incoming air is subject to con siderable change in llassing through a carburetor drop ping llractically riO per cent under normal conditions. '1'his is due to expansion after leaving the tube restric tion and loss from evaporation of the fuel. As kerosene begins to give off a vapor at about 80 deg. Fahr., it follows that the temperature must not fall below this figure at any time and should be somewhat higher consistent with volumetric efficiency. To offset this temperature drop, we may increase the tempera ture of the incoming air. However, this is not enough. If we could maintain a suspension of fuel molecules in the air stream without any deposition, this would bl' sufficient. Unfortunately we must have bends and turns in fuel passages, all(: as the fuel molecules have a greater specific gravity than the air, they will be im pinged or thrown aguinst the sides and remain in a liquid state. This means an impoverished mixture. This must be avoided and can be by applying heat to the sides or walls of the manifold, which will materially assist in revaporizing these molecules and sending them back into the air stream. Modern practice seems to point out the advisability of applying heat to assist carburetion. There are at least three practical methods being used successfully: Raising the temperature of the intake air, heat jacket ing the manifold, and applying heat directly to the fuel -supply. The successful burning of the lower-grade distillates depends upon at least the two former means. The third may be emllloyed with good results. In the application of heat we are attempting to main tain a temperature within the carburetor and passages sufficiently high to assist and maintain vaporization. This is pest ac('omplished through the medium of rais- ing the temperature of the intake air. Preheating the fuel will assist in atomization, but owing to the ex tremely small amount entering the air stream, can have but slight effect in maintaining the necessary tempera ture for vaporization. Applying heat to the fuel pas sages is very necessary, especially where the charge has some distance to travel. Deposition of fuel must in evitably take place, and unless this can be got back into th� incoming charge, it must arrive at the combustion space in an impoverished condition. It is therefore very essential that heat be applied to the manifold amI brought as near to the carburetor as possible. As the specific gravity of the fuel increases, so should the temllerature of heat application increase. With present-day gasoline much benefit could be derived by jacketing the manifold with hot water throughout its entire length; but with kerosene, hot water is not suf ficient-exhaust gas must be used. I-Jere arises a situa tion which does not lend itself to direct application when best results are sought. At low motor speeds, when air velocities through the carburetor are low and atomization is incomplete, we should have the highest temperatures, especially around the manifold jacket. The temperature and quantity of exhaust gas is limited at this time. Therefore the sup ply of exhaust gas is inversely proportional to the needs. To offset this condition a governing means might be employed so as to utilize all the exhaust at low speeds, controlling it as much as may be found necessary for the increase in speed. The manifold should be as short as possible con sistent with good diffusion and proper distribution. It should have few bends and large radii, a voiding pockets and change of cross-section to such an extent as shall affect the velocity of the incoming charge. Can Cannon Fire Cause Rain? THE cause of the llrolonged rainy period that pre vails in France is traced by some persons to the CUlI non tiring, and there is a difference of opinion in this regard among the public. In order to learn what com· petent authorities thought about the matter, the .Tourlla/ consulted Prof. A. Angot, director of the Central 1\1et ('oro logical Bureau at Paris. He considers that therp is no relation whatever between the rain and the call nonading that is now going on, and remarks that tlIP rpars 1910 and 1912 were notable for heavy rains, amI in the former year there occurred the inundations of the Seine. On the contrary the year lUll was unURU ally dry. Experiments have been made in this and other countries to bring about rain by firing numerous nnd very heavy ('harges of explosives, but without pro ducing t he desired effect: It should be noted that tllP production of rain is always mnllected with the gplI pml movements of the atmosphere. '1'0 muse rain th? moist air must be drawn Ull by an ascending air cur rent, which is sufficiently rapid and prolonged, and it rains at a certain place because the winds, often com ing from a long distance, being moist air to this poillt and are at the same time strongly ascendant. It is worthy of remark tha t these ascensional movements are made by enormous masses of air comllared with which the air displacements caused by the explosion of shells or firing of cannon are quite negligible. Such effects are so minute in comllarison with the complex llhe n. omena of the atmosphere that man's action would bp quite chimerical should he attempt to control such phe nomena, at least in the llresen t sta te of our knowlellge. Army Service Museum A MUSEUM which presents a special interest at this time was recently inaugurated at the Val-de-G rac-e Hos pital of Paris, which takes a prominent part as regards the scientific treatment of the wounded. The museum, which is in charge of Prof. Jacob, has two large rooms devoted to surgical subjects, among which are remarked the numerous models relating to reconstitU tions of partfl of the human body. Other rooms contain a large col lection of modern scientific means of destruction used in the war and also a series of llrotecting masks for as phyxiating gases. We also remark the well-fitted lah oratory, which is devoted to anti-typhic vaccination and in another place there is installed a miniature rep: resentation of the working of the health service in the army.