FlightGlobal.com
Home
Premium
Archive
Video
Images
Forum
Atlas
Blogs
Jobs
Shop
RSS
Email Newsletters
You are in:
Home
Aviation History
1964
1964 - 0451.PDF
276 FLIGHT International, 20 February 196$ EXPENSIVE NOISE Mr S. L. Bragg's Royal institution Discourse— Concluded Last week we reproduced the first two sections of the talk— "Noise and Oscillations in Jet Engines"—given before the Royal Institution, as one of its Discourses, by Mr S. L. Bragg, chief scientist of Rolls-Royce Ltd. Here, slightly abridged, is the concluding section. WE now turn to the question of the noise produced by aeroengines. This is not particularly destructive—althoughfailures of tailplane panels have been ascribed to it in one or two cases—but is a serious social nuisance that has impor- tant repercussions on airline operating procedures. Types of Noise Source In order to understand the ways in which jet engines produce noise it may be helpful to look first at the simplest principles of sound production. Let us imagine that a flashing light represents a tiny balloon which is rhythmically in- flating and deflating itself. As it inflates it pushes the air out of the way. The surrounding air, however, is locally compressible, so the whole atmosphere does not move instantaneously. A signal in the form of a compression wave travels out in all directions with the message to provide more room. It is just like adding a man to the end of a bench which is already fairly full. Everyone does not make room immediately: but first the man's neighbour feels squashed and moves, then his neighbour, and so on up the line. Similarly, as the balloon deflates, a rarefaction wave travels out, telling the atmosphere to contract again. These waves travel out into the air at about 750 m.p.h., and our ear registers them as sound. Oscil- lating balloons are rare in nature, of course; but one gets roughly the same effect from a siren, which emits air in little puffs at regular intervals. Such a source is called a monopole source (Fig 16). Now, if I put two of these panting balloons together, and arranged things so that one was collapsing while the other ex- panded, I would have a dipole source. Suppose two alternately flashing lights represent the balloons. You will see that any point in the plane at right angles to the line joining the lights, and mid- way between them, is equidistant from the sources. Such a point receives from the two sources signals that are exactly equal and opposite, so that they cancel. On the other hand, a point on the line joining the sources is inevitably closer to one source than to the other. Thus one signal is stronger, and also the two signals are not quite out of phase, since one has travelled further than the other. So they do not cancel. The phase effect obviously becomes more important as the frequency increases. It can be shown that a single sphere oscillating from side to side produces just the same effect at a distance as does a dipole pair. In this case it is evident that the sphere imposes an oscillating force on the air: and indeed it can be shown that at reasonable distances from the source the same sound field would be produced by the same oscillating force, however it was applied. The important new concept introduced with the dipole source is the directional effect—no sound is radiated at right angles, most sound on the axis, and proportional amounts at intermediate posi- tions. We can illustrate this with a little water model: if we arrange a vibrating dipper to oscillate the surface of water in a dish, a regular train of circular waves spreads out, as from a monopole source. If we vibrate the dipper sideways to act as a dipole source the waves spread only in the direction of vibration and are negligible at right angles to it. Most of the sounds we meet in real life come from dipole sources —the oscillating diaphragm of a microphone is an example. A case that is very relevant to the present discussion is the humming of telephone wires in a wind. This sound is caused by the shedding of a regular succession of alternate vortices in the wake of the wire: these produce an oscillating reaction on the wire, which therefore acts as a dipole sound source. Because of the directional effect I have just discussed, the hum should be loudest just below the wire and inaudible upwind or downwind of it. If we go one stage further and add two out-of-phase dipole sources side by side, we get a quadrupole source of noise. This has a rather more complicated pattern, but one that can again be understood by looking at our light pairs. There are now two per- pendicular directions in which any point is equidistant from op- posed sources, and so receives no net signal. But at 45° the signals don't quite cancel, because of the phase and amplitude differences, and some sound is radiated. Cancellation is nearly complete, how- ever, and the quadrupole is a pretty feeble way of producing noise. Little was thought about this relatively complicated quadrupole arrangement until Dr Lighthill, some ten years ago, showed its vital relation to jet noise. The reason for its relevance is this: the oscillating force necessary for dipole radiation can only be applied on the gas by a solid boundary—the telephone wire or microphone diaphragm. But a pair of oscillating forces means an oscillating torque, and this can be produced by eddies in the gas itself, without a solid boundary. When a fast-moving jet is exhausted into the atmosphere, violent shear and mixing take place at its boundaries —think of the rippling edges of the exhaust steam jet from the safety valve of a railway locomotive. And so it is the eddies in this intense shear zone which produce the jet noise. The eddies are not regular like those behind a telephone wire, but rather random, so one does not hear a single discrete tone from a jet, but a roaring noise with no particular frequencies. The quadrupole nature of jet noise also explains its directional effect. Little noise is radiated at right angles to the jet and little along its axis, but plenty in "lobes" of sound in between. Noise Energy As I said earlier, quadrupole sources are in- efficient. In fact, with an exhaust velocity equal to 1,200 m.p.h.— which is a reasonable figure—only one-thousandth of the energy in the jet is actually radiated as noise. But since the jet power is high —of the order of ten million Watts, say—the total noise energy may reach 10.000W. To put this in perspective, it is useful to remember that a wireless set produces sound energy at a fraction of a Watt, and even this modest power is quite capable, if radiated at an in- opportune moment, of causing considerable disturbance. The fraction of noise energy radiated increases as the fifth power of the jet velocity, so that in a large rocket engine, with a jet velo- city of several thousand miles an hour, it approaches 1 per cent of the jet energy. This means that the rocket can radiate several million Watts of noise—the output of every radio and TV set in the Greater London area concentrated.in a single source about as big as a telephone kiosk! You may be surprised to hear that even though rocket-take-offs from this country are pretty rare, we do have an almost identical source of sound that is indigenous. The flow of energy in a rocket jet is comparable to that in a large modern power station. Steam is raised in the latter at a rate and at a pres- sure comparable to those of a big rocket. So if for any reason the power station generators are suddenly thrown off load, and all the Fig 16 Three varieties of sound source: monopole, dipole and quadrupole 1 11 ((.\ \ \ \ \ / / / / I/ / / /
Sign up to
Flight Digital Magazine
Flight Print Magazine
Airline Business Magazine
E-newsletters
RSS
Events
