My Cry Baby mods

[Goodrat]

Working on this for a while and this is where I'm at:
http://rickviola.com/images/Wah5.jpg

[joegagan]

looks great, thank you for putting that up.

i like where you ended up so far, really nice work.

[Goodrat]

I wanted to not be able to overdrive the wah section but still have plenty of gain. I noticed the top of the signal (with signal generator and scope) clipped way before the bottom.
With my HSS Strat in bridge position, full up, playing a chord hard would always distort with the pedal at certain positions. That was unacceptable to me.
So I carefully tweaked to center the biasing of Q1 with the TI circuit simulator.
I also wanted 3V pk output with 200mV pk input at 1Khz (pedal position max for 1KHZ). That is plenty of gain. Then adjusted the other resistors to have symmetrical clipping when the input went over 1v pk input.
There is also an odd distortion with the pedal full back and at about 350-400 Hz. I took care of that with the added resistor in series with the 2.2uF (4.7uF was never needed), which can be lowered to about 300 to 400 ohms, I just have to try that physically. (I need to find one in my junk box).

[joegagan]

Awesome. Great to see what can be accomplished when careful attention is paid to a very specific goal. While my need for ultra clean wah is minimal, i definitely have learned from what you did. Did you sim the freq plot while assessing the 2.2 or 4.7 uf at inductor? I find that different setups need different values ( usually larger than 4.7) for max bass, but this also needs to be balanced with the need for faster charge/ discharge rate of a smaller value cap.
Also, how did you come up with the lower- than - normal mix resistor at q1 collector? I have found this change increases  Q, as have other crafty wah freaks.  While not a mix resistor in the boomerang, the r at this location is 1.5 m in there, while its clone, the dearmond 1800 has 270k. Hmm.( worth mentioning that the 1800 also adds 100k resistors from base to ground at both q 1&2, which the boomer does not have.)

[Goodrat]

I find no difference in the simulator lowering the cap to 2.2uf and I have a bunch of those.
I just tweaked the values again today when I lowered the added C4 resistor. The drawing is updated. I wanted a little more bass. Not enough Wah in the low E string. Doing that, I had to increase the input resistor again because the C4 resistor will affect the low frequencies, then adjust the Q1 collector resistor higher to center the signal again. The 91K resistor (usually 82K) will affect what happens at the low end also in addition to how the signal is centered.  Not  perfectly centered, just positive a little bit or else I get a glitch on the negative peaks. If the 91K is too high, that weird glitch on low frequencies comes back and I would have to increase the C4 resistor again and that would lower all gain.
Basically, I had the original working in the simulator and looked at the signal at the Q1 collector and the output and started varying values to get the signal at Q1 collector to ride about half the supply voltage and if the input increases, clip evenly.  It was a tug of way with all these parameters and It will never be perfect unless I put a little gain in the output buffer so things are not so critical.
Basically, half educated guesses, half trial and error. I bought a HotPotz 2 for it because my HP 1 that was in it was getting scratchy. I may drop the Q resistor back to 33K.

[Goodrat]

This is a comparison of the output of the wah in a simulation, with the original values and my value choices, driven by 750mV peak 1KHz sine wave.
http://www.rickviola.com/images/WahCompare.jpg

[Goodrat]

Basically, this is the issue:

Input 200mvpk: Original  2vpk out............Mine: 1.4vpk out
Input 400mvpk: Original  3.2vpk top, 3.3vpk bottom ..........Mine: 2.7vpk out symmetrical
Input 500mvpk:  Original 3.2vpk top,  3.5vpk bottom..........Mine: 3.2vpk out symmetrical
Input 1vpk: Original 3.4vpk top,  3.8vpk bottom..................Mine: 4vpk out, starting to clip symmetrically.

So the original has a little more gain on low level input, but limits quickly, while mine (the "Clean Wah") has more headroom and effectively sounds louder and noticeably louder than bypass mode.,
which is needed when you filter a signal with a narrow band pass filter like a wah.

[Goodrat]

I should add, the only other concern if you want very clean after these mods, is to not overdrive the output buffer op amps.
They can only handle 3.5V peak from the wah circuit without any clipping at all.
If that is an issue, just increase input resistor R18 (68K) up towards 100K or even 120K.
Or, increase the Supply voltage to maybe 12V. I'd have to look at my power supply in my Rev E version to see if there are issues there.
I think there is a regulator IC in the Rev E, not the Rev B as shown.
Or, maybe a rail to rail op amp.

[matt239]

This is so cool! Goodrat, are you still around? I appreciated your diagram, & your posts on the "Crybaby "Classic" thread.

- I'd like to know more about how you biased the transistors, & eliminated distortion.
Do we need to test for individual trannies, or are your values likely to work for most modern, medium gain transistors?
What is the function of the new resistor in series with the 4.7uF, (or 2.2 in your case) to ground? How did you determine it was necessary?

... Also, how did you come up with the lower- than - normal mix resistor at q1 collector? I have found this change increases  Q, as have other crafty wah freaks.  While not a mix resistor in the boomerang, the r at this location is 1.5 m in there, while its clone, the dearmond 1800 has 270k. Hmm.( worth mentioning that the 1800 also adds 100k resistors from base to ground at both q 1&2, which the boomer does not have.)

Hi Joe! I had always guessed that the 1.5Meg resistor was important to the sound of the Boomerang. Doesn't it cause less bass to be attenuated?
I can see that it would affect the Q. Yet Goodrat has made it smaller than standard crybaby value. - Was this important to the biasing?
I had been thinking of increasing this resistor to 1Meg, or 1.5, as in the Boomer.
The change of Q1 collector resistor from 22k to 10k is a Boomerang value. - Do we think this should generally be an improvement?

WHAT VOLTAGES DO WE WANT TO SEE WHERE ON THE TRANSISTORS?
Much has been made about using certain low gain transistors, but shouldn't almost any decent transistor sound good if we get it biased right?
(Though I get that if we don't want to change the resistors too much because of how they affect the rest of the circuit, then you might want to use lower HFE transistors..)
- I don't currently have a way to measure them..