Wah Wah Woes

[Jack Collins]

I'm trying to put an old Thomas Organ Crybaby back together.  It was in pieces because some time in the past I used parts from
a couple of wahs to try and make the best one.  I also took the opportunity to install a dpdt switch.

Got the thing all put together using this schematic:

http://www.geocities.com/CollegePark/Library/1355/crybaby.gif

And the diagram from the Fulltone Site for installing a true bypass switch:

http://www.fulltone.com/ThomasCrybaby.jpg

Plugged it in and got nothing out of it.  I found a couple of solder bridges.  Also found that Q1 was dead
(on my multimeters hfe tester it had a gain of 0, while Q2 had a gain in the 250 range).  Fixed the bridges,
and installed a 2N3904.  Tested, still no joy.  I signal traced it.  I get signal up to the base of Q1 and then
it disappears.  Measured the voltages on Q1 legs.  With a fresh battery I got e=0, b=0 and c=8.8 or so.

Clearly something wrong.  I checked all of the components surrounding Q1. A couple of the resistors had
drifted up a bit but nothing horrible.  No shorts or open circuits that I could find.  The 470 ohm resistor
connecting the emitter to ground is new. 

I double checked all the connections between the board, jacks, pot and switch, they seem correct.  Checked
again for solder bridges.  I'm using a signal generator and a scope to trace signal.

I'm stumped, I've been working on this a couple of nights and am not sure what to check next.

Oh, the RC filter and diode connected between the 22k resistor and the 9v source in the schematic above
aren't present in my crybaby.  It also has one of those stack of dimes inductors.

Any suggestions greatly appreciated.  When I get this fixed, I'm going to give it to a buddy who is just getting back into
playing guitar.

[Dai H.]

Q1 pin outs the same (maybe they are different?)?

[Paul Marossy]

Are you sure that the PCB isn't damaged? You might have a broken foil that looks OK, but isn't.

[Jack Collins]

Q1 pin outs the same (maybe they are different?)?

They're different but I bent the transistor leads to accomodate that.

[Jack Collins]

Are you sure that the PCB isn't damaged? You might have a broken foil that looks OK, but isn't.

I checked for continuity all over that area of the board.  Did a visual
inspection with a magnifier. 

I will check again from the leads of the components though.

[Pedal love]

You aren't using the same pot capacitors as on the schematic? That looks very small. Check it.pl

[Jack Collins]

You aren't using the same pot capacitors as on the schematic? That looks very small. Check it.pl

I'm using the original capacitors.

[Paul Marossy]

I assume that you have already checked to see if everything in the circuit is getting power? What about the inductor? Perhaps one of those tiny inductor wires broke at the post(s)?

[Jack Collins]

I assume that you have already checked to see if everything in the circuit is getting power? What about the inductor? Perhaps one of those tiny inductor wires broke at the post(s)?

The voltages on the other trannie look OK. 

I checked the inductor.  Don't remember exactly but r was somewhere in the low 100's of ohms which I think is OK.

How could a bad inductor cause the Q1 voltages to be wacky?

[Paul Marossy]

How could a bad inductor cause the Q1 voltages to be wacky?

Well, I seem remember a fairly recent thread were a bad inductor made the wah go dead on effect engaged mode. I remember that one of the wires in the inductor had broken, and that was the cause of the problem, IIRC. That probably wouldn't have anything to do with wacky Q1 voltages, but you did say that it was dead, so it came to mind as something to check. It sounds like it's OK if you were able to read 100 ohms of DC resistance. Maybe the Q1 transistor is just bad?

[Jack Collins]

How could a bad inductor cause the Q1 voltages to be wacky?

Well, I seem remember a fairly recent thread were a bad inductor made the wah go dead on effect engaged mode. I remember that one of the wires in the inductor had broken, and that was the cause of the problem, IIRC. That probably wouldn't have anything to do with wacky Q1 voltages, but you did say that it was dead, so it came to mind as something to check. It sounds like it's OK if you were able to read 100 ohms of DC resistance. Maybe the Q1 transistor is just bad?

I subbed in another 2n3904 with the same result.  It gave a gain of 250 on my dvm set to hfe.

Spent another hour tonight trying to figure out what's going on here.  Pulled up all the components connected at the b,c and e nodes one by one to see if one of them was pulling down the voltages or something.  Checked for continuity between component leads.

Dang, this is driving me nuts. 

[Paul Marossy]

Yeah, this is weird. I am still a little bit suspicious of that inductor. Are you sure that the inductor is installed correctly? That's in between the 1.5K resistor on the base of Q1 and that 33K resistor to ground, so if it were not on the PCB correctly, it would break the string of components from the base of Q1 all the way to the 100K resistor and the 4.7uF cap. I dunno... it's just a thought.

Or another thought that I had is that maybe that cap between Q1 and Q2 is faulty. If so, it would be a direct connection between the base of Q1 and the emitter of Q2. This is the "sweep cap", and is a feedback path for the AC signal from Q2 and is responsible for the capacitave reactance effect. If this cap were bad, it would then be DC coupled and the operation of the circuit would be changed.

Have you tried feeding the circuit with a test signal and audio probing it? I would try using an audio probe and check everything out thoroughly. If you refer to the patent document, it may be of some help: http://www.diyguitarist.com/Misc/Patent_3530244.pdf

[Paul Marossy]

Any updates? I'm curious about this one...

[Jack Collins]

Any updates? I'm curious about this one...

Didn't work on it last night as I had band practice.  I'm gigging all weekend so I probably won't
get back to it until Sunday.

My plan is to to give it the once over with fresh eyes and then, if I don't find anything, shotgun it.
Desolder all components from the board.  Clean the traces and then put it back together, measuring
the components off the board before I solder back in.  I'm going to build the circuit around Q1 and
see if I can get that conducting first.

[Paul Marossy]

Alright, good luck. I usually do better finding out the source of mysterious problems when I have put it aside for a few days. Maybe it'll be the same for you. 

[Jack Collins]

Figured it out.  It ended up being a torn trace between the 1.5k resistor and the input cap.  I guess that means that
the Q1 wasn't getting any bias voltage.

Here's how I figured it out.  I'll start by saying I've never taken any semiconductor circuit classes.  All I've had is
a semester of basics.  I got my old breadboard out and built the circuit around Q1 up.  I kept adding stuff until
I got it to where it was amplifying.  When I broke the bias part of the circuit (at least I think that's what it is.)
it behaved just like my broken wah's Q1.  I traced the continuity of that part of the circuit and then did a close
visual inspection until I found the break.  Made a bridge across the break and it still didn't work!...

Then I noticed I had just torn the ground lead from the battery snap.  Fixed that and voila!  It works.

Still has a scratchy pot.  I was know there is no fixing those, no matter how much pot cleaner you spray
in there.  Has anyone ever figured out why? 

It only took me a couple of weeks to get this right.  Course I learned a bunch.

Thanks for all the encouragement!

[R.G.]

Still has a scratchy pot.  I was know there is no fixing those, no matter how much pot cleaner you spray
in there.  Has anyone ever figured out why? 

It's because the constant rocking has actually worn through the conductive element in places. See "The Secret Life of Pots" at GEO, and "The Technology of Wah Pedals" as well. (http://www.geofex.com). Lubricating and cleaning the pot with cleaner only makes it slip across the blank spots more easily.

Here's a "secret" well known to computer techs used to working with tiny, fine traces that break easily. If you think that a PCB trace between ...here... and ...there... may be bad, set your ohmmeter to "beep" on the ohms scale and probe the soldered pads at the ends of the trace. If it beeps, there is continuity. If it reads resistance greater than a few milliohms, then other things are conducting, not the copper trace, and it's broken. You can find hairline cracks by holding one probe firmly in a soldered connection on one end of the trace, then scraping the point of the other probe down the trace away from the fixed one. When the beep quits, you've either wandered off the trace or crossed a crack.

[Paul Marossy]

It ended up being a torn trace

Aha! I knew it. That was the only thing that made any sense.

On the pot, just replace it. They get scratchy because they're worn, plain and simple. I like the Dunlop Hot Potz II pots the best - I think they sound good and last a very long time (translation=no scratchy pot for a very long time).

EDIT: RG beat to the punch on the pot.