Optimizing Audio Signals: Tips for Lagging and Synchronizing Electric Waves

[Evgeny]

Hello!
I'm building a circuit which requires a sound wave to catch up to an electric signal of the same wave generated by a microphone (after its gone through amplification, phase correction, etc of course). Can anyone please help me with this, or at least point me in the right direction?
Just if anyone is interested, what I'm trying to do is the classic active noise-cancellation circuit, where a sound wave is electrically inverted and superpositioned on top of the original sound wave - hence, cancelling anything out. While this sounds like a fairly simple circuit, I'm learning electronics from the ground up, and getting the output wave and the original wave in synch is proving to be a tough thing to do.
Thanks in advance,
-Evgeny

 

[chroot]

Are you using analog or digital circuitry?

- Warren

 

[Evgeny]

Analog - like I said, I'm a complete beginner and am just trying to grasp the basics of electronics, so digital is a bit further than where I am right now in my studies.

 

[faust9]

Analog - like I said, I'm a complete beginner and am just trying to grasp the basics of electronics, so digital is a bit further than where I am right now in my studies.

Digital is actually much easier to learn. No dealing with finite levels and gains and the like. Digital boils down to one of two states(you'll see these states defined as on/off or 0/5 mostly but in reality you the designer can call these states anything you like--which is akin to programming languages use of the if statement). This is an over simplification to say the least but it should help you see what digital actually is. Finding the Q-point for a cascaded transistor amp can be a pain in the butt. Thinking in terms of simple numbers can be a little easier IMHO.

A high speed inverting opamp(you'll have to use at least one transistor as part of the gain loop to allow for cancellation and not just overpowering with a higher magnitude sound that is out of phase) could allow for a little bit of noise cancellation. The problem is--unless you are designing a full blown analog computer--analog circuits won't allow for prediction. Digital noise cancellation uses very complex prediction algorithms to improve in the cancellation quality.

Start simple: play around with basic transistor amps--common base, common emitter, and common collector. Learn how to set the Q-point and the characteristics of each amp configuration. Once you can do that (this is the easy part) redesign each amp as a switch. Once you can do that, design an H-Bridge. Make some A, AB, B, and C class amplifiers. This step will help you make your noise cancellation circuit. Once you know how to use basic transistors(go to Jameco or mouser or digi-key and buy metric butt ton of 2n2222A's and 2n2907A's) move on to op amps. There are millions of web sites out there detailing op amp usage as well as transistor circuit usage.

Once you've done the above, try and make some astables or monostables and/or oscillators.

Well, good luck in your endeavor.

 

[chroot]

Here's a design idea for an active analog delay:

http://www.elecdesign.com/Articles/ArticleID/6348/6348.html

See the schematic linked in as a figure.

- Warren

 

[rayjohn01]

only possible answer

Chroots suggestion is not for audio the delays are miniscule as are the delays of coax the problem with audio is the frequency is low even to create an echo is something in Milliseconds not micro or nano secs.
I am sorry to say the only practical way is to convert the signal to digital then store the words in a memory and recall them later if this is not your thing yet -- then you too will have to delay . There is only one thing slow enough and that is sound I believe but have not researched the subject that there are sonic delay generators --- but you will have to research this and it sounds expensive .
Ray.