10

Seems like a simple enough question:

Why is a whistle sound emitted when air is pushed through a tight space?

J.Todd
  • 1,811

1 Answers1

11

You describe "a whistle sound when air is pushed through a tight space". That is a familiar phenomenon. Essentially, you hear a self-sustaining resonance - something is driving the air mass in a way that causes a resonance that becomes a recognizable pitch.

When air flows rapidly around a tight opening / sharp lip, you get flow with a very high Reynolds number - meaning that the flow will want to become turbulent. But turbulence is a funny thing - while it appears random, it can be disturbed (and thus shaped) by small external forces. If these external forces are somehow synchronized with the turbulence, a "tone" can appear.

So what usually happens is that an initial turbulence gives rise to a pressure disturbance, and the shape of the cavity in the vicinity of the aperture / lip causes part of this pressure wave to be reflected. The reflected wave returns to the aperture, where it interacts with the turbulence. If the turbulence is "back in phase" with the pressure wave, the action reinforces itself - and a tone is created.

If there are too many different path lengths (or not enough reflections), then there is no particular frequency that is favored, and instead of resonance you get mostly white noise - a "hiss". Anyone who has learnt to play the flute (or who has tried to blow across the opening of a bottle) will be familiar with this - when you first try blowing into the mouthpiece, no sound comes out: only when you blow at just the right angle (namely, when the point where the turbulence is created can interact with the pressure reflections inside the flute) will you produce a sound. When your flute's holes are all closed, the sound travels all the way to the end before returning and a low resonance occurs; as you open holes closer to the mouthpiece, the pitch (frequency of sounds) will increase. With the right combination of holes and technique for blowing, you can create harmonics - instead of a half-wavelength round trip for the sound you end up with $\frac32$ or higher, with correspondingly higher pitch.

This is also briefly described on wikipedia: http://en.wikipedia.org/wiki/Fipple

Floris
  • 118,905
  • 2
    CuriousProgrammer - I think you have a better answer now. . . – HDE 226868 Dec 05 '14 at 00:44
  • 1
    Regarding blowing at the right angle, also jugs and bottles (a far cheaper option to conduct an experiment ;) ). – jpmc26 Dec 05 '14 at 01:57
  • 1
    I'm not sure if I buy the argument that it's due to turbulence... yes, the Reynolds number increases as it goes through the contraction, but it's also not an isotropic process. The vortices/eddies created are being "stretched" in the axial direction and squished down in the transverse direction. A careful reading of the Fipple wikipedia article says that the harmonics are created by turbulent effects, but the original tone I'm not so sure about... Something for me to think about here. – tpg2114 Dec 05 '14 at 02:54
  • @tpg2114 I think fundamental frequency is just the zeroth harmonic... You may be reading too much into the Wikipedia entry? – Floris Dec 05 '14 at 02:56
  • @Floris Check out the page on how a whistle works. They say the air splitter is unsteady causing oscillations which generate pressure waves. The resonance chamber amplifies certain waves. I guess it's splitting technical hairs between the lay-definition of turbulence and the technical definition of turbulence, but oscillations in a fluid aren't technically turbulence unless they have specific characteristics. So I'll concede that unsteady flow features cause the noise, but I don't think it's turbulence. Just unsteady flow... – tpg2114 Dec 05 '14 at 03:00
  • I am OK with unsteady flow. – Floris Dec 05 '14 at 03:01