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 from 'The Autocar' magazine discussing the design principles and various types of vehicle exhaust silencers.
| Identifier | ExFiles\Box 152\3\ scan0012 | |
| Date | 13th July 1912 | |
| 70 THE AUTOCAR, July 13th, 1912. Silencers. Since the question is obviously one of reducing speed, in order to obtain a practicable length of pipe it becomes at once clear that the cooling of the exhaust gases is a matter of great consequence, since by cooling them their volume can be greatly reduced. Thus, whereas thousands of cubic feet of gas are entering the engine end of the pipe in unit time, during the same period only hundreds of cubic feet need be dealt with at the other, and at correspondingly reduced velocity. In other words, if the plain pipe have sufficient radiating area the egress orifice can be made quite small. Fig. 1.—The type of silencer used on some racing cars. The first requirement of an effective silencer is, therefore, that its surface be sufficient to cool the gas down to as nearly as possible atmospheric temperature. Its pressure being similarly reduced, it can issue without any impact at all, and, accordingly, noiselessly. We have remarked that the long exhaust pipe must be of thick material. This is likewise a matter of importance, because if thin it will simply vibrate under the periodic exhaust impulses, and so set up a sound, or rather a noise of its own. Visitors to Brooklands will have re-marked that many of the racing cars give out a very metallic sounding exhaust. This is always due to the use of thin exhaust pipes and boxes. At the same time it is quite obvious that the thinner the pipe the better will be its radiating effect, hence a happy medium must be struck. Fig. 2.—The simple type of silencer on Calthorpe cars. Another requisite is that near to the engine, that is to say where the velocity is very high, the exhaust pipe must be as free as possible from sharp bends, and internal irregularities capable of causing surfaces of discontinuous flow, and constrictions. In order to see how necessary it is to observe this precaution, it is enough to consider the speed of the exhaust gases at the moment of their egress. Assuming that the piston speed is 1,000 feet per minute, and that the gases in the cylinder are at normal temperatures and pressure, then if the diameter of the exhaust pipe be half that of the cylinder (and its area consequently one quarter), the initial speed of the gases through, this pipe will be 4,000 feet per minute. But instead of being at atmospheric temperature and pressure, they are very hot (volume is accordingly greatly increased), and also under considerable pressure. Their speed will be, therefore, probably about 8,000 feet per minute, or, roughly, ninety miles per hour. At this high speed sharp bends exercise a great deal of resistance to flow, and even slight irregularities in the pipe section will set up surfaces of discontinuity—eddies, in other words. Any resistance thus set up must immediately react upon the power of the engine, for it not only opposes an opposing effect to the exhaust stroke but allows an unnecessary amount of gas to remain in the cylinder and thus impoverishes and diminishes the volume of drawn in on the inlet stroke. When the velocity Fig. 3.—The type of final outlet which is in fairly general use. of the exhaust gas has been considerably reduced it will offer little resistance to going over sharp corners and through small holes which would well nigh arrest it altogether at a very high speed. This is one reason why nearly all silencers are to be found at the back of a car, or, at all events, some considerable distance from the engine. That this effect of high gas velocities is not generally realised is evidenced by the fact that people will often attempt to gauge the degree of “back pressure,” or resistance of a small silencer, by blowing into it. At the low speed at which one’s lungs are capable of expelling a piece of fine gauze would offer scarcely any resistance, yet to gas at a high velocity it would act like an impenetrable barrier. Where perforated baffle plates or holes are used the size of them is immediately dependent upon their number and the extent to which the gas velocity has been reduced by the time it reaches them. Placed at the end of a long pipe a silencer might give little or no back pressure at all, whereas clamped direct to the engine it would offer a great deal of resistance indeed. Fig. 4.—The arrangement of the baffle plates in the Nest-o’-Cups silencer. For this reason the design of the exhaust ports which are cast into the cylinders themselves is a matter of the highest importance, as no excellence in the silencer device can make up for the resistance which sharp corners, ill-conceived curves, or rough edges will set up at this point. It is, of course, a great advantage to have as much as possible of the exhaust outlet water-cooled so as to accelerate the temperature reduction at the moment the gas commences to leave the cylinder. In motor boats silencing is much simplified by the fact that the whole of the exhaust pipe and box can readily be water cooled. Fig. 5.—The three concentric chambers of the Ford silencer. So far we have considered only the case of a long pipe. Since this cannot be used on a car some modification has to take its place, and accordingly practically every form of silencer in use at the present time is simply a contrivance in which the gas is made to travel a circuitous course, so as to obtain the same effect as it would in a long pipe. Artificial resistances, such as baffle-plates and perforated tubes, are used primarily to put a slight brake on the gas, and, secondarily, to break up the sound waves. It is in this regard that the constricted end of the pipe referred to previously is additionally valuable, as it has exactly this effect. It acts, of course, exactly conversely to a megaphone, which develops the waves of sound, whereas the constriction reduces them. | ||