The Most Expressive Wah I Have Played...

[R O Tiree]

Here ya go...

[Gus]

R7 is the resistor to make a fixed resistor and pot for biasing in the colorsound.  Also I don't think you will need to change the 47uf to 10uf C2 if converting a crybaby like wha it should not matter.

FWIW  I do think a buffer before the wha circuit is a good idea, something like a high input resistance EF.

[MohiZ]

Hey, I'm not arguing, it's just debating.  icon_wink

I was expecting someone to bring up the absence of a resistor on the emitter of Q1 and that being a "big factor". It does have an effect on the circuit because if there is not enough gain, it won't sound good. But it's not going to radically affect the sweep range of any wah circuit. It might alter the frequency response slightly, but it's not going to be anything big.

No, I just didn't want YOU to think that I was arguing  

What I found out, having a 100k pot in place of the collector resistor of Q1, is that it affected the sweep range in a major way. The toe position frequency did not change, but the heel position frequency decreased with a higher collector resistance setting (expanding the sweep more into the bass range). A lower value collector resistor diminished the bass range of the sweep. By setting the collector resistor and the "sweep cap" just right, you could set the highest and lowest frequencies of the sweep pretty much anywhere you want. There might be some difference between changing the collector resistor and the emitter resistor. I didn't try changing the emitter resistor at all. They both affect gain, but the collector resistor doesn't change the amount of current flowing through the transistor, changing the emitter resistor does.

But, maybe this doesn't work so well if you have the stock bumpers in place and are only able to use part of the pot's rotation. Who knows? It's kind of tedious to try out every single possibility with these things.

About the Q resistor, since the Colorsound wah doesn't have it, it's the same as having a resistor with infinite resistance in its place. So it would have a very high Q, just as you thought it sounded.

Do I believe that a stock crybaby has a narrower range than a colorsound wah? Oh, undoubtedly. Do I think the range of a Crybaby can be modified to exceed the range of a Colorsound wah? Absolutely. Do I think it'll sound the same? No. Not unless you have a pot that's tapered _exactly_ right to match the mechanics of the Colorsound, which is highly unlikely. Guess I was just making a point that the amount of "expression" in a CryBaby can be extended. Talking about stock pedals, I'll shut up now.  

Oh, and R O Tiree: thanks for pointing that out.

[R O Tiree]

R7 is the resistor to make a fixed resistor and pot for biasing in the colorsound.  Also I don't think you will need to change the 47uf to 10uf C2 if converting a crybaby like wha it should not matter.

FWIW  I do think a buffer before the wha circuit is a good idea, something like a high input resistance EF.

IIRC from RG's "Technology of Wah Pedals", C2 is made to look like a variable cap by virtue of the way it's connected (via various other components) to Q2 and the inductor. So I'd think it would be wise to go with the flow and make it 10µF, not 4µ7F.

Wouldn't be hard to adapt this concept to later CryBaby boards that have the input buffer already fitted.

[MohiZ]

IRC from RG's "Technology of Wah Pedals", C2 is made to look like a variable cap by virtue of the way it's connected (via various other components) to Q2 and the inductor. So I'd think it would be wise to go with the flow and make it 10µF, not 4µ7F.

It's C5 that RG means.

[Gus]

C5 is the cap made to look variable. C2 is to cause an "AC ground" at the node of R5, C2, L1, R7. Emitter of Q2 drives via C5  L1 and R2 and the base of Q1.  Q2 is an emitter follower with a gain of just less than 1.  

Paul if you have a scope and a triangle wave generator build a grounded emitter gain stage, try a 22k C to 9VDC a Si transistor a 10K base to ground and select  or use a pot and fixed resistor base to +9 to bias it at different collector voltages, cap couple in and out use a signal gen set to triangle and watch the scope as you increase the drive to the circuit.

EDIT I see MohiZ posted about C5 when I was writing

[R O Tiree]

Ahhh - light dawns

I went back and re-read it. The last cap RG mentions specifically in that para about "the secret" is the 4.7µ and then he just says "that capacitor" several times in the next few sentences... I remember reading it yonks ago and getting disconfabulated, thinking it was some kind of weird Thevenin thing that I couldn't get my head around. Makes much more sense now.

[MohiZ]

I went back and re-read it. The last cap RG mentions specifically in that para about "the secret" is the 4.7µ and then he just says "that capacitor" several times in the next few sentences... I remember reading it yonks ago and getting disconfabulated, thinking it was some kind of weird Thevenin thing that I couldn't get my head around. Makes much more sense now.

I know, I had to read it a few times over too after you mentioned the 4.7uF cap. 

[Paul Marossy]

Paul if you have a scope and a triangle wave generator build a grounded emitter gain stage, try a 22k C to 9VDC a Si transistor a 10K base to ground and select  or use a pot and fixed resistor base to +9 to bias it at different collector voltages, cap couple in and out use a signal gen set to triangle and watch the scope as you increase the drive to the circuit.

Thanks, I will try that.

What I found out, having a 100k pot in place of the collector resistor of Q1, is that it affected the sweep range in a major way. The toe position frequency did not change, but the heel position frequency decreased with a higher collector resistance setting (expanding the sweep more into the bass range). A lower value collector resistor diminished the bass range of the sweep. By setting the collector resistor and the "sweep cap" just right, you could set the highest and lowest frequencies of the sweep pretty much anywhere you want. There might be some difference between changing the collector resistor and the emitter resistor. I didn't try changing the emitter resistor at all. They both affect gain, but the collector resistor doesn't change the amount of current flowing through the transistor, changing the emitter resistor does.

The Colorsound wah does go quite deep in the bass range. That's where it starts to sound synthy...

[DougH]

I don't know if this is helpful or not. I found this old file on my computer and uploaded it to my gallery folder. The thing I noticed is that some of it follows some of the things being discussed- input buffer, more gain in Q1, etc. This fattened up my crybaby pretty nice. Would be fun to try the Colorsound values.

[Pedal love]

Its very very nice looking, just gets hard to care about anything anymore.

...which is one way of telling me to drop it, I suppose.

However, in the absence of a gear/pulley system with a better than 1:1 ratio, or another lever arm or 2, or some other system, there is no way that the pot shaft can be rotating further than the actuator arm (aka "plastic gizmo" in Paul's post above).

The older/wiser/more knowledgeable heads on this site are always encouraging people not just to be cooks (copying others' work) but learn to be chefs (paraphrasing a thread from some months ago). Some, like RG, Mark Hammer, etc, spend an inordinate amount of time explaining again and again, from different angles, for which I for one am most grateful. But, that does not absolve anyone from taking time and effort to prove things for yourself, learn the equations and how to manipulate them, researching other resources, so you really understand why things happen the way they do, and not to just accept something just because someone else says so. And we all (should) know that, to make decent pedals, we have to be a combination of artist, elec engineer and mech engineer.

So, if Colorsound managed to design a way to get an extra 60+ degrees out of a simple 2-lever arrangement that only appears to rotate through 120 degrees or so, then I'd really like to know how they did it. If there's extra gearing in there of some description, then that explains all. If not... then it doesn't.

Sorry Mike, I was just lamenting my own sad life. You didn't do anything.

[Pedal love]

Ok guys I'll tell you what I did. I got tired of all these mechanisms and just got a 100k lin. fader. Same crybaby circuit, but I ripped the fader out of the housing so there was only the resist element and the phenolic base. I loosened the straight gear from the plastic holder/guide and glued the element along side of the straight gear, with the element facing the holder/guide. I imbedded and glued a piece of the wiper's metal, in the plastic holder/guide and put a wire through a small hole behind the wiper, connected all wires accordingly and presto.  I 'm still working on it.  A few major problems - 1) no metal surrounding it, not allowing a small faraday shield which helps pots with rf. 2)The wires need to be strong but light to move with the straight gear.  3)you still need an (n.c.) pot and round gear to give opposing physical resistance. The thing I'm trying to say is it does make a huge difference what you do mechanically. I'm going to try to work these things out and give you a demo on you tube. Its worth a shot.

[R O Tiree]

Sorry Mike, I was just lamenting my own sad life. You didn't do anything.

No worries Thanks.

Using a linear pot - keep us posted on progress? Sounds like it might have promise. Dust and fluff getting gummed up in the works might be a player, though?

[Paul Marossy]

While we are on the subject of pots, out of curiousity... does anyone know what the taper is on those Dunlop Hot Potz II pots? I know the pot taper also is a factor in how the sweep is on a wah pedal, besides the actual rotation of the pot.

[R O Tiree]

Well, it turns out that the taper is seriously weird... I still had the base off mine from earlier, so I de-soldered the "Output" wire.

Looking at the cog and with the pot fully clockwise (which gives min resistance measuring from "Output" to the wiper):-

0    - 120 gives a steady reading of around 100 ohms.
120 - 135 increases to about 1k
135 - 180 rapid increase to 58k!
180 - 225 slightly less rapid increase to 91k
225 - 270 gentle increase to 97k
270 - 315 top reading of 99k (fully counter-clockwise)

So, given that the setup procedure is to put the treadle toe down, set the pot fully CCW, then lay the toothed bar up against the pinion, then rotate the pinion by one tooth CW and re-engage, then place the plastic spring up against the bar and tighten the screw... and there are 14 teeth on the pinion... that's retarded by about 25 degrees which gives 98k with the treadle fully toe down and the pot at 290 degrees.

For the next 20 degrees, we only get a 1k drop...
For the next 45 degrees, we drop 6k
For the next 45 degrees, we drop 33k

That's 110 degrees, so far, and another 25 degrees will take us down to about 30k.

That's a total of a 70k drop over the 135 degrees of "normal" travel I showed in my video earlier but it is not, as I've shown, linear over the range. Instead, it's quite rapid from fully heel down to about 80% of travel, then it slows right down over the last bit.

Remember from your video, Paul, where the first bit of travel of the treadle the cam turns quite rapidly, then it slows right down at the end? Well...

...this weird taper wouldn't be ANYTHING to do with trying to emulate the rotational behaviour of the cam in your Colorsound, would it? A small tweak would give us an almost identical R vs relative treadle-throw curve as the Coloursound, and would result in a resistance change in the order of 40k or so, again, just like the Coloursound. OK, the snag is that the treadle on a CryBaby shell doesn't have to travel as far to get the same change as the Coloursound does, which makes it more of a blunt-edged sword and harder to play subtly, but it's k-i-i-i-i-i-nd of trying to do it.

Every day's a learning day, isn't it?

[R O Tiree]

Update - I just tweaked the pot angle as I reassembled it, and got it going from 99k down to 45k. That's probably as close as I can get it without breaking the pot as the switch is engaged (that's why they advise to knock it back from fully CCW by one tooth of the pinion, I assume). Following the procedure exactly gave a minimum of 28k.

So, if you do the Coloursound mods on a standard CryBaby, you can expect it to be even more bassy and synthy down the bottom end, I guess, extrapolating from Paul's descriptions right at the beginning of this saga.

Here's the curve:

[Paul Marossy]

Huh, except for that really flat spot at the beginning, the Hot Potz II pot taper looks pretty linear right up to the end. Thanks for doing that! After all of this input from everyone, I think some of the difference must be to do with the linear pot taper vs. the taper on those CryBaby pots.

Just for fun, I drew up the Colorsound wah shell in AutoCAD just to show how that cam works. It's not 100% accurate, but I did take the actual measurements and stuff to make the drawing, so it should be pretty close to the real thing (let's say 95% accurate) - http://www.diyguitarist.com/PDF_Files/ColorsoundWahShell.pdf

This was a very interesting discussion, and I think we got to the bottom of what is going on in this wah pedal of mine. I guess I jumped to conclusions a bit too quick...

[juse]

This has evolved into quite the wah thread. Thanks for the info, guys.

Nice autocad drawing of the Colorsound, too Paul. Very nice as a matter of fact. 

[MohiZ]

Thanks for the graph and the AutoCAD drawing folks! I guess the Hot Potz is trying to be an "S" taper pot, which I've heard is sometimes used in wahs.

...this weird taper wouldn't be ANYTHING to do with trying to emulate the rotational behaviour of the cam in your Colorsound, would it? A small tweak would give us an almost identical R vs relative treadle-throw curve as the Coloursound, and would result in a resistance change in the order of 40k or so, again, just like the Coloursound.

That's what I'm talking about! Since the Crybaby's pot turns in a linear fashion with the treadle movement, the taper is used to compensate. The Colorsound mechanism naturally emphasizes the bottom and high end of the sweep, because the pot turns slower in those positions.

The Colorsound mechanism looks similar to the one found on my Bespeco cheap-o wah, except that that one really doesn't work too well. It used special pots that only had 90 degrees of rotation available.

[R O Tiree]

Thanks for the graph and the AutoCAD drawing folks! I guess the Hot Potz is trying to be an "S" taper pot, which I've heard is sometimes used in wahs.

...this weird taper wouldn't be ANYTHING to do with trying to emulate the rotational behaviour of the cam in your Colorsound, would it? A small tweak would give us an almost identical R vs relative treadle-throw curve as the Coloursound, and would result in a resistance change in the order of 40k or so, again, just like the Coloursound.

That's what I'm talking about! Since the Crybaby's pot turns in a linear fashion with the treadle movement, the taper is used to compensate. The Colorsound mechanism naturally emphasizes the bottom and high end of the sweep, because the pot turns slower in those positions.

The Colorsound mechanism looks similar to the one found on my Bespeco cheap-o wah, except that that one really doesn't work too well. It used special pots that only had 90 degrees of rotation available.

It's funny you should mention S-taper, MohiZ...

I took Paul's AutoCad drawing (nice one, BTW ) and then pulled it apart in PaintShop Pro, ending up with images of the shell, the treadle and the cam and then set about laying the cam and treadle at various angles  to construct the relationship.

The first big surprise is that the treadle only moves 16.5 degrees!!! It looks as though it should be a lot more, but that's the rotation angle I ended up using to get from heel down to toe down on Paul's AutoCad pic.



Having got these numbers, I popped into Excel and then scaled the cam angle (which turned out to be 123 degrees travel... ) into Resistance, based upon the info provided by slacker about 2 pages ago - linear pot, max 91k, 41k change across the whole range:



OK, it is a lot smoother than the CryBaby graph, and doesn't tail off as sharply as I though it would. Mind you, the extra throw on the treadle compared with the Crybaby provides for finer control.

The end result is that we now have a direct comparison between treadle position and R for both pedals. I'd say we all know a hell of a lot more about wahs today than yesterday. Possibly more than we ever wanted to