Circuital Audio Engineering

Thread Starter


Joined Dec 2, 2008
Hello all. I'm experiencing a little trouble with a project I am working on. The objective is to process a provided audio signal (.wav file), first to remove a high frequency (1khz) noise present, secondly to remove background interference (of which a pure sample is provided in a second .wav file), thirdly to provide a means of variable amplifying the output signal, and finally to provide a means of sampling the output signal for track reconstruction. This is all to be modeled in MultiSim.

I have decided to use a simple low-pass RC filter to remove the high frequency noise (step 1), and have been looking into using a common gate MOSFET amplifier for step 3 (the variable amplification). The second and fourth part, point by point signal comparison and removal of the secondary interference followed by the sampling have me a bit stumped.

Any advice, help, ideas, or suggestions would be appreciated both on what I've done and what I should do. Thank you for your time :).

P.S. Details from the project:

Your circuit schematic must operate correctly and contain the following elements:
-A method of removing the high-frequency noise from the recording
-A method of removing the undesired dolphin chatter from the recording
-A method of amplifying the signal by a variable gain
-A FET based method of buffering the signal current while following the current (current gain = 1); this device should have a low input impedenance and a high output impedance
-A method of sampling the output signal at a particular frequency for track reproduction
Last edited:


Joined Nov 26, 2008
First, if the noise is 1khz, I don't think you want to use a low pass filter to remove it since it will also get rid of any signal below 1khz. You ought to build a bandstop filter at 1khz. Second, if you are given the noise, are you allowed to use that signal? Because if you are, you can invert it and sum it up at the output to remove the injected noise. I'm not sure what you mean by sampling the output for track reconstruction if the signal isn't being converted to a digital format. Perhaps looking up Nyquist Sampling Theorem might help you there?

Thread Starter


Joined Dec 2, 2008
Using a difference amplifier, simple op-amp, would accomplish the same goal of inverting the noise and summing it up at the output correct?

I've been looking into bandstop filters and they might be a little outside the scope of the class but I'll see what I can do.