From the Rolls-Royce experimental archive: a quarter of a million communications from Rolls-Royce, 1906 to 1960's. Documents from the Sir Henry Royce Memorial Foundation (SHRMF).
Article by Professor W. Morgan about the essentials of practically comparing and testing different fuels for engines.
| Identifier | ExFiles\Box 32\1\ Scan012 | |
| Date | 28th January 1911 | |
| THE AUTOCAR, January 28th, 1911. 107 Fuel Tests. The Essentials of Practical Comparisons. By Professor W. Morgan. THE value of a fuel to the engineer depends primarily on the heat it can develop, for in a suitable engine this heat can be converted into mechanical energy. Not the whole of the heat obtained by burning a fuel in air can be thus converted, but only a certain fraction, the amount of which depends on many things. In the case of a petrol engine the proportion of the heat developed in the cylinder and delivered to the flywheel as mechanical energy will depend on such things as the compression and the general condition of the engine, the type and timing of the ignition, the position of the throttle, and the composition and character of the mixture, these being some of the chief factors. The first step in testing two fuels for the ordinary road engine should be to determine the thermal values of each, i.e., the amount of heat generated by the complete combustion of one pound of each fuel. This determination requires a trained man, but is not a very difficult operation. One method is steadily to burn a weighed quantity of the fuel, and observe the rise in temperature which is thereby caused in a weighed quantity of water. Then the weight of the water in pounds multiplied by the rise in temperature in degrees Fahrenheit gives the units of heat developed by burning the weighed quantity of fuel. A simple calculation will then give the heat which would be obtained from one pound of fuel, i.e., the thermal value. The thermal values of two fuels enable a forecast of their relative values in an engine to be obtained, for, other things being equal, the same percentage of the heat will be converted into mechanical energy in both cases, and as a consequence the respective consumptions will be inversely proportional to the thermal values. For example, two fuels A and B have thermal values of 20,000 and 15,000. If A can give a consumption of 0.5 lb. per b.h.p. hour, under the same conditions B should give 0.5 x 20,000/15,000 lb. per b.h.p. hour. If on trial a result differing from this be obtained, the reservation mentioned above—"other things being equal"—must be remembered, and the conditions examined to see if other things were equal. In the actual engine test, into which a fuel trial must resolve itself, it should be reasonably easy to arrange that the same engine in almost identical condition should be used, equal speeds maintained, like conditions of ignition and throttle obtained, and a direct reading dynamometer to absorb the power. Preferably a standard type of engine should be used, for then the results can be confirmed by reference to current practice. But now we come to the most important condition and one most difficult to govern—the composition and quality of the mixture. It is clear that there must be definite evidence that the combustion is complete, and that no partially burnt products, such as carbon monoxide, be produced. Further, the strength of the mixture must be the same in both cases, for with full charges it has been definitely established that weak mixtures are more economical than correct mixtures, meaning mixtures which exactly balance fuel and air. More than that, the charge must enter the engine with the fuel completely vaporized, and the whole mixture be homogenous. To fulfil these requirements will possibly necessitate a special carburetter for one or both of the fuels. Granted the thorough vaporisation of the fuel and the complete mixing of the charge, the point arises, how shall the composition of the charge be determined? Clark, Watson, and Hopkinson have shown that exhaust gas analysis will serve exactly, provided care be taken that no contamination of the exhaust gases with outside air be possible. For a fuel trial an exhaust containing a definite percentage of oxygen, say two per cent., must be obtained with each fuel. The engine is then run with one fuel under trial conditions of speed, ignition, and throttle, and the carburetter adjusted until an exhaust of the required quality is obtained; the necessary measurements of load, speed, and the amount of the fuel used being noted, the consumption per b.h.p. hour is calculated from the figures resulting. As already stated, it will be reasonable to expect the consumptions so obtained to be inversely proportional to the thermal values of the fuels. If this be not the result the test must be closely scrutinised, and an attempt made to discover what property of the fuel causes the discrepancy. In general it will be the method of experimenting that is at fault, the most probable source of error being the composition of the charge, and after that the physical state in which the fuel enters the cylinder. Accurate information on work of this type is very scarce, but there is every reason to believe that if the b.h.p. per hour consumption of two chemically similar fuels be not inversely proportional to their calorific values the cause lies in some condition of the test. Conceive that one brand of petrol of 0.72 specific gravity is being compared with another of 0.76 specific gravity. Now, the thermal values of these fuels are within the limits of experimental error the same, so that if the experimenter, as the result of his work, should conclude that the brake h.p. hour consumptions were 0.6 lb. and 0.66 lb. respectively, we should be justified in demanding more rigorous conditions of trial, and, indeed, in flatly refusing to accept the results. With dissimilar fuels, such, for example, as carbon monoxide and hydrogen, the physical qualities of the gaseous charges and the resulting exhaust gases will differ from each other mainly in their specific heats, thereby affecting the possible fraction of heat convertible into work, and somewhat modifying the expectation previously discussed. In such a trial many other things of importance with reference to the fuels would emerge—the ease of evaporation, production of deleterious exhaust products, and the general properties of the substance will be determined, showing what carburation difficulties can be expected. The range of combustible mixture obtainable from the fuel could be judged by noting how the engine ran on throttled, weak, and rich charges. In all cases it would be necessary to determine the composition of the charge by analysis of the exhaust gas, even to form a general opinion, for no carburetter has yet been devised which can be relied on to give a constant mixture over any considerable range of conditions, or even to act quite consistently from moment to moment. Another important point is the degree of compression which can be used with a given fuel. The higher the compression the higher the percentage of the heat which can be converted into useful work. Handwritten note: X 1527 | ||