# Why don't airliners use noise cancelling?

Some cars have started using noise cancelling in their cabins, through the speaker system. It's a well established technique for dealing with low frequency noise like engine rumble and wind noise.

So why don't modern airliners use it?

• Welcome! Noise cancelation works by injecting a signal with opposite phase that cancels the original signal. As phase is varying with the distance from the source, this works only in a small area, but in large areas like a cabin, the phase is not constant within the whole volume, so the cancelling phase must be different according to the location in the cabin. This would require using headsets for instance. note that injecting out of phase noise, adds noise heard according to an interference distribution. A lot of noise! – mins Aug 6 '16 at 11:17
• @mins I see your point, but noise canceling has both a theoretical and 1 practical side. Physics can cover the theory, but it can't address the practical issues that are unique to aircraft/cars/buildings etc. That's an engineering question and I think any aviation-specific aspects of noise canceling are certainly on-topic here. – Pondlife Aug 6 '16 at 14:29
• @Pondlife: Good point, I've retracted my vote. – mins Aug 6 '16 at 15:07
• The Bombardier system works because prop noise has a well defined frequency spectrum (as shown in the link) at known frequencies - it's easy to measure the prop RPM accurately - and the source of the noise is localized (the prop wake acting the airframe). Neither of those conditions apply to the OP's "engine rumble and wind noise." – alephzero Aug 6 '16 at 20:06
• @TylerDurden Pay double, and you won't have at least one of the people next to you... – a CVn Sep 27 '17 at 6:39

First, some airliners do use it - the Bombardier Q400 uses a NVS (Noise and Vibration Suppression) system to reduce cabin noise. Basically, it uses devices called Active Tuned Vibration Absorbers (ATVAs) mounted on the fuselage frames to 'cancel' the vibrations from propellers and outside noise, thereby quieting the cabin.

However, there are some issues with using active noise cancellation in aircraft:

• The aircraft cabin is quite big and complex, and it would be very difficult to achieve noise cancellation over the entire volume; as active noise cancellation depends on the cancellation of sound waves using opposite phase, it works best in small, confined spaces. If the waves turn out to be in phase in a region, the sound would be doubled - not a good result.

• A significant contributor to the noise inside the cabin is the wind outside; this is usually random and the frequency spectrum is also quite wide, resulting in difficulties in noise cancellation.

• Noise cancellation is effective mostly when the frequency of the source is constant, which is not usually the case (the props are an exception - they lend themselves well to noise suppression)

Instead of active noise cancellation, aircraft manufacturers are putting more energy in reducing the noise itself:

• Better engines, which has lesser noise through the use of chevrons, etc.

• Better modelling of aircraft, so that the fuselage design minimizes the noise due to wind.

• Better damping and cabin design, so the noise produced (for example, due to air conditioning and ventilation) is reduced.

“It's a well established technique for dealing with low frequency noise”

– exactly. Noise cancellation works precisely if, and only if, the area you try to shield from the noise is significantly smaller than the wavelength of the sound you try to cancel, because then you can ensure that the anti-sound signal will in fact interfere everywhere destructively with the environment noise. Wavelength scales as $\lambda = \tfrac{c_{\mathrm{s}}}{\nu}$, so it gets ever smaller as the frequency $\nu$ increases.
($c_{\mathrm{s}} \approx 343\:\mathrm{\tfrac{m}s}$ is the speed of sound.)

As soon as the size of the room you try to apply this to is larger than the wavelength, it is inevitable that the artificial signal will in fact interfere constructively on half the space, i.e. it will in many spots actually make the problem worse than it was by itself!

In a car, you can go quite far – the space is really small, and LF rumble has a generous wavelength – up to ~100 Hz you can be sure that it at least won't interfere constructively anywhere in the cabin.

Quite different in case of an airliner – not only is the cabin much bigger, also, jet noise has much more high-pitched components. Hence it is completely hopeless trying to cancel this sound with open speakers. It is possible to cancel it directly at the ear, with noise-cancelling headphones. But these are already available individually, so it's not something the airline needs to worry about!

Unless you can exactly predict the direction of the interference noise. That is only possible if you know a priori the exact phase of the source noise. As said in the other answer, this is actually possible to some degree with turboprops (which also have more low-frequency components), but not for jets.

• I think the proportionality constant between wavelength and frequency is usually only called c when it's the speed of light. – Peter Cordes Aug 6 '16 at 19:16
• @PeterCordes: it is often used for sound speed too, but I agree that this probably isn't a very clear symbol choice. – leftaroundabout Aug 6 '16 at 19:37
• @SteveV. Note for future authors: "iff" is correct but it's not an abbreviation that I'd expect people outside mathematics and maybe engineering of physics to understand. In the interests of comprehension, I've expanded it to "if, and only if,". – David Richerby Aug 7 '16 at 9:23
• @DavidRicherby It's a very useful word. Iff we use it more often, more people will understand it. Especially on a site like aviation. – Peter Aug 7 '16 at 12:22

The main problem in a large cabin is, "What do you cancel?"

The noise in an aircraft is different depending upon where you sit. The engine noise is different whether you are in front of the engine or behind it.

Canceling the noise in one place in the cabin would increase the noise in another.