Author Topic: Audio phase shift circuit design wanted.  (Read 1752 times)

0 Members and 1 Guest are viewing this topic.

Offline Jim-0000Topic starter

  • Regular Contributor
  • *
  • Posts: 65
  • Country: au
Audio phase shift circuit design wanted.
« on: July 10, 2019, 11:16:48 am »
I understand a variable delay filter can be built around a NTE1641 or a 2399 chip.
If someone can suggest a circuit or a link to a page that might have one, I would much appreciate it..


Jim.
 

Offline frozenfrogz

  • Frequent Contributor
  • **
  • Posts: 936
  • Country: de
  • Having fun with Arduino and Raspberry Pi
Re: Audio phase shift circuit design wanted.
« Reply #1 on: July 10, 2019, 11:38:06 am »
PT2399 is pretty straight forward, datasheet features basic schematics for echo and delay applications.
Also: http://www.diyaudiocircuits.com/tutorials/pt2399-digital-delay-analog-echo/
He’s like a trained ape. Without the training.
 
The following users thanked this post: Jim-0000

Online ledtester

  • Super Contributor
  • ***
  • Posts: 3279
  • Country: us
Re: Audio phase shift circuit design wanted.
« Reply #2 on: July 10, 2019, 11:52:36 am »
Supposedly the NTE1641 is a replacement for the MN3007:

https://www.diystompboxes.com/smfforum/index.php?topic=48790.0

In the datasheet for the MN3007 there is an example application circuit:

http://www.kazumi-kikou.com/pdf/3007.pdf

The example circuit employs a MN3101 as a "BBD clock generator". This thread suggest that more common CMOS logic chips can be used to replace the functionality of a MN3101:

https://www.diystompboxes.com/smfforum/index.php?topic=72447.0
 
The following users thanked this post: Jim-0000

Online SiliconWizard

  • Super Contributor
  • ***
  • Posts: 15797
  • Country: fr
Re: Audio phase shift circuit design wanted.
« Reply #3 on: July 10, 2019, 12:03:27 pm »
Just nitpicking here, but strictly speaking, a fixed delay will only give a fixed phase shift at one specific frequency. So you have to make sure a delay is what you really need.
 
The following users thanked this post: Jim-0000

Offline Jim-0000Topic starter

  • Regular Contributor
  • *
  • Posts: 65
  • Country: au
Re: Audio phase shift circuit design wanted.
« Reply #4 on: July 10, 2019, 01:01:26 pm »
Just nitpicking here, but strictly speaking, a fixed delay will only give a fixed phase shift at one specific frequency. So you have to make sure a delay is what you really need.

Good point and you are right, a delay may not work for what I want.
I want to invert the phase of perceived (or actually heard) frequency of a Tinnitus sufferer.
So, the phase shift (of 180 deg) will need to be adjustable for any frequency over the audible spectrum.
Alternative suggestions most welcome.
 

Offline MagicSmoker

  • Super Contributor
  • ***
  • Posts: 1408
  • Country: us
Re: Audio phase shift circuit design wanted.
« Reply #5 on: July 10, 2019, 02:22:42 pm »
...
I want to invert the phase of perceived (or actually heard) frequency of a Tinnitus sufferer.
...

I'm no doctor, but I don't think applying a real sound will counteract the perceived sound of tinnitus; the latter is a phantom sound that may or may not even be the result of a signal on the auditory nerve (ie - it could be generated in the brain, so truly imaginary). Even if there is a signal on the auditory nerve how would you cancel it out without painstaking scanning of frequency and phase to find the null? This just doesn't seem at all practical to me.
 
The following users thanked this post: Jim-0000

Online SiliconWizard

  • Super Contributor
  • ***
  • Posts: 15797
  • Country: fr
Re: Audio phase shift circuit design wanted.
« Reply #6 on: July 10, 2019, 02:32:29 pm »
Agreed. Some cases of tinnitus can be objectively measured with a microphone inside the ear canal (otoacoustic emissions), but AFAIK, the more common cases are "phantom sounds" as MagicSmoker said and can't be related to anything acoustic. Trying to "counter-act" them on a phase-reversal basis would not make any sense, and estimating the average frequency (es) of tinnitus (unless again they are otoacoustic emissions) is a very tedious and error-prone process.

If, again, you're only interested in otoacoustic emissions, I would still not know how you could "cancel" them out as the measured signal inside the ear canal may have no useful direct phase relationship with the emissions themselves (which are due to the outter hair cells...)
 
The following users thanked this post: Jim-0000

Offline IDEngineer

  • Super Contributor
  • ***
  • Posts: 1950
  • Country: us
Re: Audio phase shift circuit design wanted.
« Reply #7 on: July 10, 2019, 03:59:41 pm »
This is a fascinating idea.

It makes sense that the "source" of the (possibly phantom) sound could be anywhere in the signal chain, from actual stimulation in the cochlea to processing errors in the brain itself. But regardless of the "source", the real question is how would the brain interpret the superposition of a cancelling signal? We understand that acoustic mixing could cancel an actual sound at the eardrum... but electrical cancellation is also a real thing, and once a "signal" makes it to the brain why would the brain magically interpret two different signals differently? So what if one started out as "real" (externally acoustic) while the other started out as "fake" (undesirable stimulation of the cochlea, or even a flaw within the brain). Once both "signals" are within the auditory processing centers they are just that - signals - and aren't tagged as "real" and "fake".

Yes, syncing up the frequency and phase could be challenging. But I'll venture out on this limb a ways with another idea: Biofeedback. There are countless examples of the brain adapting to accommodate things. Some are natural (did you know that each person's neural network is unique, and that an infant's brain learns to map theirs?) while others are external/artificial (such as learning to modulate one's own heartbeat and blood pressure). I wonder if a closed loop system could be created whereby the brain modulates the frequency and phase to minimize the effect of tinnitus. There's an obvious error signal available, and an obvious result of improved control available, so in theory it's possible. Might take time but so does learning to walk with a prosthetic that is controlled by nerve impulses.

Again, a fascinating idea.
 
The following users thanked this post: Jim-0000

Offline Audioguru

  • Super Contributor
  • ***
  • Posts: 1507
  • Country: ca
Re: Audio phase shift circuit design wanted.
« Reply #8 on: July 10, 2019, 04:20:53 pm »
To cancel a sound you need a few accurate things:
1) The exact frequency.
2) The exact amplitude.
3) The exact waveform (including all the harmonics).
4) The exact phase shift.
Who says with tinnitus they always stay the same?
EDIT: At times you might end up increasing the level of the sound.
« Last Edit: July 10, 2019, 04:27:24 pm by Audioguru »
 
The following users thanked this post: Jim-0000

Offline Audioguru

  • Super Contributor
  • ***
  • Posts: 1507
  • Country: ca
Re: Audio phase shift circuit design wanted.
« Reply #9 on: July 10, 2019, 04:33:09 pm »
Thinking about tinnitus, I always hear faint very high audio frequencies. Their frequency and amplitude are random. Try to match that!
 
The following users thanked this post: Jim-0000

Offline magic

  • Super Contributor
  • ***
  • Posts: 7453
  • Country: pl
Re: Audio phase shift circuit design wanted.
« Reply #10 on: July 11, 2019, 08:28:31 am »
So, the phase shift (of 180 deg) will need to be adjustable for any frequency over the audible spectrum.
Then you simply want an inverting opamp stage.

But as others pointed out, you can't cancel a signal that you can't measure in real time.
 
The following users thanked this post: Jim-0000

Online SiliconWizard

  • Super Contributor
  • ***
  • Posts: 15797
  • Country: fr
Re: Audio phase shift circuit design wanted.
« Reply #11 on: July 11, 2019, 01:19:48 pm »
What are still commonly used as a way of relief for tinnitus are masking signals. They are often some kind of broadband noise, such as filtered and/or modulated white noise.
For those interested in knowing a bit more: http://www.tinnitusjournal.com/articles/use-of-masking-for-tinnitus.pdf
( https://www.tinnitus123.com/tinnitus-maskers/ )
 
The following users thanked this post: Jim-0000

Offline ogden

  • Super Contributor
  • ***
  • Posts: 3731
  • Country: lv
Re: Audio phase shift circuit design wanted.
« Reply #12 on: July 11, 2019, 02:20:03 pm »
I want to invert the phase of perceived (or actually heard) frequency of a Tinnitus sufferer.
So, the phase shift (of 180 deg) will need to be adjustable for any frequency over the audible spectrum.

180 degree phase shift for any frequency or just broadband noise is simple DSP task - invert polarity of waveform/signal. Problem in your case is - you do not have waveform nor signal to invert. Basically your idea of creating 180-degree phase shifted sound to cancel Tinnitus sufferer is impossible. If you say that it is possible then why don't you earn money winning lotteries and trading stocks?

[edit] Audiogury pretty much explained why (it is not possible).
« Last Edit: July 11, 2019, 02:28:22 pm by ogden »
 
The following users thanked this post: Jim-0000

Offline Audioguru

  • Super Contributor
  • ***
  • Posts: 1507
  • Country: ca
Re: Audio phase shift circuit design wanted.
« Reply #13 on: July 11, 2019, 02:39:48 pm »
I hate to hear acid rock or Rap because they are broadband noise. Instead of hearing noise to mask tinnitus then maybe good music would be better.
 
The following users thanked this post: Jim-0000

Offline Jim-0000Topic starter

  • Regular Contributor
  • *
  • Posts: 65
  • Country: au
Re: Audio phase shift circuit design wanted.
« Reply #14 on: July 11, 2019, 11:53:26 pm »
What are still commonly used as a way of relief for tinnitus are masking signals. They are often some kind of broadband noise, such as filtered and/or modulated white noise.
For those interested in knowing a bit more: http://www.tinnitusjournal.com/articles/use-of-masking-for-tinnitus.pdf
( https://www.tinnitus123.com/tinnitus-maskers/ )

Thanks for the reply. I will read the link. More later.

Jim.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf