Article courtesy of http://RacingSecrets.com
I. THE PROBLEM.
The necessity for muffling the exhaust of airplane engines is hardly, open to argument. The objects in view are the minimizing of noise to delay detection in military service, to protect the general public, particularly those living near aviation fields, and lastly to give the operator a better chance to know what the rest of his power plant is doing. The last point is perhaps even now of little importance, as an exhaust pipe long enough to end behind the operator is quite enough to make the exhaust noise less prominent than some other rackets.
A study of the general problem of silencing the airplane power plant, not only in the laboratory but also by means of observing airplanes in a large number of flights, has led to certain conclusions none of which are, however, new. In the first place the exhaust noise is not the only disturbance to be dealt with, although perhaps the most important, because the staccato barks of open exhaust carry to greater distances than the other attendant noises. It is, however, not a matter of great difficulty to so far suppress these barks that the exhaust noise ceases to be the most prominent in relation to some others. As a matter of fact, a simple pipe of sufficient length will do this for the high-speed multicylinder engines, and we understand that some American and German planes are using this scheme. It serves at least to protect the operator, even if it does not go a great way toward actually suppressing the pulsations as far as an observer at a distance is concerned. The very fact that the impulses follow so rapidly upon one another seems to make the problem of taking off the "bark" easier, for we found it much harder to muffle single-cylinder slow-speed engines.
Assuming however, that a successful device for completely muffling the exhaust can be found we should still have to deal with other noises such as the hum of the propeller, the singing of gears, and the rattle of the valve gear. It will be admitted that all of these sources of noise can be minimized, but elimination does not seem to be in the realm of possibility.
On the 8-cylinder engine used for our last experiments the propeller of the fan brake caused a deep, more or less musical note, which appeared to come from the crank ease. This noise disappeared when the blades were removed from the fan arm-, and the engine was operated at speed under its own power, swinging only the arm. The same sort of humming note can be identified in connection with planes in flight at considerable altitudes and distances. To silence this disturbance presents a problem on which at present the writers have no suggestions, except that slower speed (geared) propellers might help.
Another source of noise is in the valve gear. In the case of the engine under test this consisted in a sort of rattling hiss at high speeds. It can be easily identified when the observer is close by. In planes of flight at some distance from the observers the noise would appear to-be drowned in the exhaust roar and in the hunt of the propeller. In any case this disturbance can be minimized by accurate adjustment. But it is difficult to see how the valve slap can be entirely eliminated.
The last source of noise is in the gears. This can be partly suppressed at least by the use of spiral gearing and by accurate machine work and mounting.
These four sources of noise are the principal ones requiring attention. We would place them in order of importance: (a) Exhaust no(b) propeller noise, (c) valve-gear noise, (d) gear noise. We believe that it is most important to suppress the exhaust noise, because its staccato barks will undoubtedly advertise the rising of a plane sooner than the other three by reason of its greater carrying power. But the problem of the other noises remains; and we are further of the opinion, based on our experience in the past year, that the exhaust noise can be so far suppressed with comparatively simple means that it--forms the smallest source of the disturbance of the four. We will not venture to predict complete suppression. Any muffler construction in connection with its manifold will have to take in some one of its parts the full force of the original blow, and since lightness of construction is one of the requirements calling for thin walls the chances are that there will always be more or less of a pulsating roar, at least near the engine. We have so far not reached the stage of considering this part of the problem.
Confining our attention now to the particular problem in hand, the silencing of the exhaust, a successful device will have to meet three requirements: (a) Satisfactory suppression of noise with least back pressure, (b) lightest possible weight, (c) greatest durability. For the last year we have, confined our attention to the first of these, believing that the other two could be successfully met if the first requirement were satisfied.
