Johnson777717

Good evening folks!

I came across something interesting in my trig book today in class, and it has intrigued me. Although, it may be something that everyone already knows about--except me.

Anyway, a math problem explains of an ANC system (Active Noise Control) where a mic picks up the sound, and is "processed electronically" to produce the same frequency and amplitude, but out of phase with the original. Then the signal is output through a speaker. The two waves (original and out of phase) collide to result in destructive interference. Thus, the noise is reduced.

Okay, I have a couple questions regarding this:

1. Do you folks know what kind of circuit this would entail? My first thought is to use an op amp.

2. This may be a little off the wall question but--You have the two waves out of phase from each other (I'm assuming 180 degrees out of phase???) and they collide to form destructive interference. So the waves are basically mauling each other. How exactly does this reduce the noise? My guess is that the interference causes the wavelenghts to decrease, thus the waves don't travel as far as they could. So an observer would interpret this as a reduction of noise, because not as many waves are reaching the observer (or the amplitudes are smaller). Am I correct?

This is pretty facinating to me. Does anyone know of any circuits, or an idea of how to go about creating something like this? I'd like to mess around with it, try different phase angles, amplitudes, wavelengths etc.

Thanks for reading, and taking the time to assist! Have a great one!

Oznog

I tried to do that. It's unfortunately more complicated than it sounds, at least to make it work well. I didn't have much success.

There's a number of difficulties:
1. It's limited by phase shift. If there's over 90 deg and less than 270 deg phase shift, it's amplifying the noise and will create unlimited feedback. The mic, speaker, the time delay due to the physically different location of mic & speaker, and any filter circuitry must be managed.
2. The amplitude of the feedback needs to change depending on the microphone & speaker's freq response. The lower the phase shift, the higher the cancellation is possible.
3. Generally, I would think you don't want the mic to catch the antinoise coming out of the speaker, though I have seen commercial products (NoiseBuster headphone box) do this.

All the sets I know only cancel low freq, up to 500 hz, I expect because of the phase shift limitations. The results are impressive enough.

You can get some response without being complicated if you don't need a high degree of cancellation results. Ideally, digital has far, far more potential than analog since digital filters can have complex freq/phase responses with no undesired minimum phase shifts outside of what's in the mic/speaker.

Johnson777717

Thanks for your reply Oznog!

This sounds like where most of the complication is, eh? I think I may leave this to the acoustic engineers I believe managing this is beyond my capabilities at this time.

Yes, to me, your statement makes complete sense. I wonder why the commerical products allow for this? Possibly just design contraints etc.

Well, again, I think this is beyond my capabilities at this time, so I'll leave this as something to think about at idle times. Thank you for the information though, I appreciate it very much!

PS.
What kinds of applications would the ANC be used? For instance, would a common approach be for Air conditioning, aircraft, headphones, etc? I'm just trying to figure out where these ANC units are commonly used.

Thanks again!

gerty

Some factories are using "white noise" as a quieting agent to reduce/offset the noise created by machinery. This is supposed to prevent hearing loss problems. I'm not sure how the correct frequency is arrived at, but the reports are positive as to the results..

__________________
gerty

Johnson777717

that is interesting. Do you know what frequency the white noise band is? I've always seen white noise generator circuits, but had no clue as to what they were for.

Oznog

White noise won't do anything to prevent hearing loss! It will make it worse since it will reduce the perception of loud noises somewhat, thus less avoidance.

As far as I know, "white noise" is noise whose frequency is evenly spread through the spectrum. It is used for many testing purposes, setting equalizers, etc.

Illini

White noise is specifically noise which appears to have a Gaussian distribution.

Nigel Goodwin

No! - you use pink noise for that.

White noise has the same amunt of energy between 100Hz and 1000Hz as between 1000Hz and 1100Hz.

Pink noise has the same amount of energy between 100Hz and 1000Hz as between 1000Hz and 10,000Hz. It matches the logarithmic nature of the audio systems it's used for testing.

__________________
PIC programmer software, and PIC Tutorials at:
http://www.winpicprog.co.uk