Home General Asylum

General audio topics that don't fit into specific categories.

Doppler distortion and sonic holography.

209.86.4.3

Greetings Everyone,

I want to breach this subject carefully, because it flies in the face of the current thinking of at least one highly respected industry guru. If you're in the audiance doc, I hope you'll join in and show me where I've got it wrong.

The doppler effect is simply the frequency shift caused by relative movement between a source and an observer. The whistle of an approaching train sounds higher in pitch because the sound waves are squeezed closer together by the trains motion. After the train passes and begins moving away, the pitch of the whistle seems to drop, because the waves are now being stretched out as the train moves away. You only hear the ture pitch of the whistle if both you and the train are stationary. Now this is just a simple explination, and is intended only for those who have never heard of "Doppler" but hopefully it's sufficient for the following arguments.

For at least 40 years there has been a segment of the audio community that has expressed concern over "Doppler distortion". Originally the arguments centered around the assertion that the motion of a loudspeaker driver producing a low frequency tone would cause an alternating frequency shift in any higher frequencies reproduced by the same driver. The cone is moving, toward you, then away from you, just like the train, and this is stipulated as a source of distortion. More recently it has been asserted that the same thing happens with a microphone because the motion of the diaphram, in following the lower frequencies, modulates the high frequency waves, again due to the Doppler shift caused by the relative motion between the source and mic. And now I understand that a circuit has been developed that will remove this "Doppler distortion" from the sound. Is this a good idea?

Consider this. If a very low note and a very high note are played on a piano (real piano, real sound), the sound that results is produced by a single sounding board. The back-and-forth motion of the sounding board as it produces the lower note modulates the frequency of the higher note, due to the doppler effect. This is not distortion. It's a piano. Now if you go and design a circuit to remove doppler effect from the sound, you've got a little problem. Your circuit must recognize this as a valid shift, and pass it through without change. If you do remove this effect from the sound of the piano I'd guess that each note would seem to have a more independent existance, which it does. I would expect this to make the notes sound more independent and clearer, and I also expect that it would be very easy to mistake this as an improvement in the sound quality. Lower distortion usually improves clearity, and if clearity improves, distortion must be lower. But distortion is not lower! It's higher! You may like the "new" sound of the piano, but all else being equal, it has to sound less like a real piano. I hope this position is self evident, because if it isn't then I have done a poor job of explaining things. It's tough trying to keep this kind of thing short and coherent at the same time and I'm not sure I'm very good at either. But I think people need to be warned about the problems with this "new" technology before they go and upgrade their speakers to benefit from this new distortion reduction technique. It's really a coloration, and you will tire of it over time (OK, I could be wrong here. I know nothing about your tolerance for coloration, but most people will eventually tire of ANY coloration. Then they upgrade, again.). I would be very surprised to see such an approach lead to long term satisfaction. Like Bob Carvers seductive "Sonic Holography", this too will wear thin over time.

With a little more thought it's not hard to see that, in fact, even notes that come from different instruments modulate each other as they mix in the air. It's not a Doppler effect the way we normally think about it, but the end result is exactly the same. While it is often convinient to think of sound as "waves" it is sometimes more accurate to think of sound as compressions and rerifications in the air, which travel at (surprise!) the speed of sound. When a low frequency wave is moving through the room, there are large areas of high compression and of low compression. Any higher frequency sound which is traveling through the same air will add or subtract it's own compression and rarification areas from the air it currently exists in. If the high frequency tone is passing through a low frequency rarification, it's frequency is reduced. Try to visualize the rarification as a "stretching" of the air, and a resultant thinning. The air that is being stretched contains a several of the higher frequency waves, and as the molicules of air are streteched apart (and rarified) the little waves inside get stretched too. And the LF compression wave compresses the air that it occupies, so any higher frequency waves contained therein also get squeezed closer together. The end effect is that the long wavelength compression waves cause a frequency shift upward in any sound occupying the same space, and rarification does just the opposite. If you think of the compression as moving toward you and the rarification as moving away, you may notice that there is something going on here that is very closely related to Doppler shift. The similarity is more than skin deep. All the same equations apply. I'm not a big math fan, but sometimes it can be cool. I think the point I am trying to get at here is pretty self evident by now. The Doppler effect doesn't just happen within a single instrument. Every sound that shares the same air space with other sounds is modulated by the other waves sharing it's space, and the effect is identical to the Doppler effect. It IS the Dopplar effect. This effect is not distortion, but rather a part of all sound. It's true that if you mix a very low frequency with a very high one and play it through your speakers, you can clearly hear the lower tone altering the higher one. But this proves nothing, dispite it's appeal. If you produce each tone on a seperate driver you still get the same interaction. Try it.

Now a disclaimer: This is a very "dark" post, which I would prefer to avoid, but as this issue has been raised in well respected circles it deserves serious consideration. I really would hate to see folks flocking to this new technologh only to have it wear thin in short order. Equipment is too expensive for such short excursions, and I think most of us are better off without them. If you want to play with this technology, and enjoy it's qualities, that's fine with me, because it will most surely be DIFFERENT, but please don't buy off on the argument that it is actually an improvement in accuracy. It isn't. Even if I have most of this wrong the hard fact remains that Doppler shift DOES exist when a single instrument produces multipal tones from a single sounding board. Removing this natural effect from the sound will make it different, but is WILL be less true to the original.

So now it's time for all the really smart guys out there to tell me how I've got this whole thing wrong. I'm all ears. :-)

Be excellent,

Charles




This post is made possible by the generous support of people like you and our sponsors:
  Amplified Parts  


Topic - Doppler distortion and sonic holography. - Charles 01:38:50 08/06/99 (31)


You can not post to an archived thread.