[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference tekvax::electro_hobby

Title:Electro_Hobby
Notice:TEKVAX may be up-and-down 'till we get settled in..
Moderator:TEKVAX::KOPEC
Created:Thu Jan 15 1987
Last Modified:Tue May 20 1997
Last Successful Update:Fri Jun 06 1997
Number of topics:1831
Total number of notes:14297

1827.0. "Mystery filter design ... help" by EVTAI1::SAAD () Thu Feb 06 1997 14:11

    Hi,
    
    Wah-Wah pedals use, I believe, a high Q passband filter with a
    sliding center frequency which is controlled with a pedal.
    
    Can someone explain the details of the filter design in 
    the schematic FILTER.GIF at EVTAI1::.
    
    In particular, two things "bother" me : 
    - the lack of symmetry (One inductor) in what seems otherwise to be
      to be a Twin-T type filter (but I'm only guessing - is it a PI type?)
    - the fact that a variable resitor is able to modify the center frequency 
      in what seems to be a L (and hence F depends on L and C) based circuit.
     
    Any ideas ?
    
    Regards
    Bill
    
    
T.RTitleUserPersonal
Name
DateLines
1827.1mysteryWRKSYS::INGRAHAMAndyFri Feb 07 1997 21:2636
>   Wah-Wah pedals use, I believe, a high Q passband filter with a
>   sliding center frequency which is controlled with a pedal.

I would have guessed a variable lowpass filter.  Maybe some peaking
around the cutoff point might have an interesting effect too.  But I'm
not a musician.


>   In particular, two things "bother" me : 
>   - the lack of symmetry (One inductor) in what seems otherwise to be
>     to be a Twin-T type filter (but I'm only guessing - is it a PI type?)

There's more to this circuit than meets the eye.

The filter doesn't seem to be a twin-T.  Re-draw it.  The input to the
filter is through the 470K resistor.  I think maybe it's more like a
low-pass filter??  Or maybe bandpass?  It's part of the feedback
branch around the first transistor.

Then there's another feedback branch by way of the second transistor
(emitter-follower), which seems to have that potentiometer in series
with its base, perhaps as a sort of crude gain control?

>   - the fact that a variable resitor is able to modify the center frequency 
>     in what seems to be a L (and hence F depends on L and C) based circuit.

The transistors have a lot to do with how this circuit works; it's
more than just a passive filter with transistors added for buffering
or gain.  By combining signals having different phase shifts and
amplitudes, and varying their relative amounts, that is apparently how
they achieve the desired effect (F depends on R).

I haven't figured out how the circuit works, these are just some
guesses.

Andy
1827.2REGENT::POWERSMon Feb 10 1997 11:5517
>>   Wah-Wah pedals use, I believe, a high Q passband filter with a
>>   sliding center frequency which is controlled with a pedal.

In general, no.  

>I would have guessed a variable lowpass filter.  Maybe some peaking
>around the cutoff point might have an interesting effect too.  But I'm

In general, yes.  Though newer effects pedals may do the more complex
filtering of variable Q, variable center frequency, the traditional
Wah-wah pedal was just a foot operated version of the passive tone control
that would be mounted on the guitar itself.  This is a variable cut-off
low-pass filter.

Try the NAPALM::GUITAR notes file for more current informed opinion.

- tom]
1827.3Bandpass it seems !EVTAI1::SAADMon Feb 10 1997 13:4710
    
    The NAPALM::GUITAR guys seem to indicate that Wah's are sliding
    frequency bandpass filters.
    
    I guess I just have to build one and test what the circuit 
    actually does !
    
    Thanks
    Bill
    
1827.4I've always seen notch filters.EVMS::PIRULO::LEDERMANB. Z. LedermanTue Feb 11 1997 12:354
    The circuits I've seen for WahWahs (many years ago) was that it was a
    variable frequency notch filter, cutting out a portion of the frequency
    band rather than enhancing it.  I suppose the opposite would do
    something interesting as well, though.
1827.5makes senseEVTAI1::SAADTue Feb 11 1997 13:205
    That makes sense actually. Building it will certainly help 
    solve the mystery ...
    
    Thanks
    Bill
1827.6REGENT::POWERSWed Feb 12 1997 12:1515
If you've ever shopped for a guitar special effects device, you will be aware
that if a transformation can be applied to a sound signal, someone will
have tried it.
I built my first fake wah-wah pedal in college by copying the passive 
tone control circuitry in my guitar and connecting it to a foot pedal (made 
from a bread loaf tin and a drilled out piece of wood).
I talked through a design with a colleague that would have linked
variable Q and variable center frequency that would have used two
pots on a single foot pedal (one up and down, the other left and right).
We never built it.
So you can call it a "wah-wah" if it changes the tone,
but it's up to you whether it's a notch filter or a low pass
or whatever.

- tom]
1827.7I think I've cracked it ?EVTAI1::SAADThu Feb 20 1997 09:5215
    Hi,
    
    I think .1 and .6 are right.
    
    I ran a PSPICE simulation (I can provide the source) of an RLC 
    passive filter in which I managed to change the "apparent" value of L
    (and hence the center frequency) by re-injecting into the L part of the 
    signal with a phase change (RC circuit with variable R).
           
    Thanks a lot everybody (and PSPICE !)