Design Auto Wah Filter: Learn How to Create Frequency Sweep

Join the discussion
Ask a follow-up here, or get your own question answered by working scientists, mathematicians and engineers — people, not an autocomplete.
Real named experts · corrections over time · the nuance an AI answer skips
5 replies · 5K views
chaoseverlasting
Messages
1,051
Reaction score
3
I'm working on a college project building analog guitar effects. I want to build an auto wah. It's essentially a variable resonant frequency band pass filter. Traditional auto wah's either use a low frequency oscillator for a constant 'sweep' across the filter's frequency band or use an envelop follower to gauge the amplitude of the input signal and produce a proportional sweep across the frequency range of the filter.

The problem is that I can't figure out how to create the frequency sweep. I know I could look at the already existing circuit diagrams and just reproduce the effect, but that defeates the purpose of the exercise. Plus, something like this shouldn't be too difficult to do.

What I have so far is this:
As this effects is a variable frequency band pass filter, if I just change the resonant frequency of the filter, I should, theoretically, get the effect I'm after.

This much I can do. The part that I can't figure out is how to automatically change the resonant frequency depending on the input signal. One way I thought of doing it was to change the inductance or capacitance of an LC filter to change the resonant peak.

What I need is a way to change the capacitance of the capacitive element based on the input voltage or current. So a current or voltage controlled capacitor. What device should I use?

I'll be using TL 084 op amps and 7809 IC's for the power supply. Also, I'll be using a gyrator circuit for the inductive elements, which is why I need to use a voltage or current controlled capacitor.

I hope this makes sense.
 
Engineering news on Phys.org
I think what you want is an RC (not LC) filter where the R can be varied proportional to some input voltage. This used to be -- before all that digital synthesis nonsense came around -- called a Voltage Controlled Filter. I googled that phrase and found this circuit:
http://www.musicfromouterspace.com/analogsynth/STATEVARVCFFEB2006/STATEVARVCFFEB2006.html
which is pretty ugly, but does have a vague explanation of how it works "using LM13700 transconductance amplifiers as voltage-controlled integrators". There also used to be single chip VCFs...ah here: SSM2044 -- obsolete but maybe available from the analog synth underground folks...
 
Last edited by a moderator:
schip666! said:
This used to be -- before all that digital synthesis nonsense came around -- called a Voltage Controlled Filter.
:smile:

I remember them well; having designed and put together a synthesizer (but it did have a little digital too).

Other keywords to search on are "Envelope-Controlled Filters".
 
schip666! said:
http://www.musicfromouterspace.com/analogsynth/STATEVARVCFFEB2006/STATEVARVCFFEB2006.html .

That is exactly what I'm looking for, except that I've got to design the circuit myself. OTA seems to be the right way to go, but all that the OTA is doing is giving an output current proportional to volgate, isn't it?

The interesting part in that circuit is controlling the Q factor and the filter resonance frequency simulatneously. That is, essentially, all that I want from my circuit.
 
Last edited by a moderator:
Most people use the transconductance amps, but in college, I did some effects that used pulse-width-modulated analog switches.

I used the CD4066 analog switches and put them in series with the resistive elements in a biquad filter. As the pulse width changed, you could vary the Q or frequency.

As for the source of the PWM, I wanted it proportional to a control voltage, so I built an oscillator that gave me a nice triangle wave (i.e. 100-200kHz). Then, I compared it to the control voltage. Be careful though. If your PWM goes to zero and you don't have a high value resistor across the analog switch, then the filter circuit will "crash."

That was ages ago, and I didn't record the schematic, but I can tell you that it was probably built around LM339's TL084's and CD4066's. Ocassionally, I would use CD4049's as well for inverters and sometimes as analog parts.

Best of luck,

Mike
 
Mike_In_Plano said:
Most people use the transconductance amps, but in college, I did some effects that used pulse-width-modulated analog switches.

I used the CD4066 analog switches and put them in series with the resistive elements in a biquad filter. As the pulse width changed, you could vary the Q or frequency.

As for the source of the PWM, I wanted it proportional to a control voltage, so I built an oscillator that gave me a nice triangle wave (i.e. 100-200kHz). Then, I compared it to the control voltage. Be careful though. If your PWM goes to zero and you don't have a high value resistor across the analog switch, then the filter circuit will "crash."

That was ages ago, and I didn't record the schematic, but I can tell you that it was probably built around LM339's TL084's and CD4066's. Ocassionally, I would use CD4049's as well for inverters and sometimes as analog parts.

Best of luck,

Mike
Thank you, Ill look into it. I have a couple of days to think something up and there are other parts of the project which are more easily approached. The wah is one of 4 effects blocks I'm building and the only one that has me in a tight spot.