Capacitor Question (in distortion pedal)

[canman]

Hi everyone,

I'm building a distortion pedal for a friend of mine (Triple Wreck...can't seem to build enough of these) and I've had to do a little debugging.  I was really stoked because I didn't have to come here and ask for help, because I thought I solved the problem, but it only led to a different problem.

So, here's the problem.  My voltages all check out.  Everything is perfect.  Bypass works, engaged circuit kind of works.  I am getting really glitchy distortion.  The tone stack works, the volume works, the gain...well, it kind of works.  Fully open, it works, but when I turn it down, it creates a gating effect.  The distortion is also pretty octave-y.  Clearly not working properly.  I managed to trace the circuit down to where the glitch is happening.  Here's the schematic:



You'll see that 1uf cap directly after lug three of the gain pot.  When I probe the leg of the capacitor that is connected to gain 3, the distortion sounds full, warm...basically how it should sound.  But when I probe the other lug of that cap...the one connecting to the 10k resistor, I get glitchy distortion.

Is this supposed to happen?  Is that the purpose of that capacitor?  I'm not so sure it is.  When I probe pins 8 and 9 of the TL074 and some pins of the TL072, the glitch is still there.  I replaced that capacitor, I reflowed the solder joints on the 10k resistor, and I'm still coming up short.  What am I missing??

[PRR]

> when I probe the other lug

With what? A meter? What kind?

If a low-resistance meter, poking that point throws IC1c all out of bias, screws-up the sound.

> a gating effect

This very often suggests mis-bias, but you have not reported finding any voltage clues.

[GGBB]

What are the DC voltages on either side of the cap? If you used a polarized cap you might have it reversed which will damage it (schematic could be wrong too).

[canman]

Oh, sorry...the probe is an audio probe.  Just a 100nf cap connected to the tip of a 1/4" jack (and grounded) and plugged into the amp.  Not a DMM!

I built it up based on IvIark's layout:



I was hoping to learn more this time around and didn't want to just rely on the layout he did, so I was following the schematic as best I could with the audio probe to see where the signal (initially) went out on me (my first problem was no effect coming through at all).  So then I did the same thing and traced through the schematic to see where the glitch started happening.

As you can see in the layout above, the capacitor used isn't electrolytic, and I've built this before with no problems, so I know it will work with a regular cap.  I am guessing the schematic is either not correct, or the symbol is to indicate what the polarity would be if you used electrolytic caps instead?

Back to the mis-bias point and gating...if the TL074 was mis-biased, would the voltage be off at pins 6, 7, or 9 (where the pot connects to the IC)?  Or would it be at pin 4, where the 9v comes in?

[canman]

After thinking about it a little more, if the bias was not correct, wouldn't I be getting glitchy distortion all over the place?  I only get glitchy distortion on the second leg of that 1uf cap.  Which ties into the first IC.  But pin 1 of the TL074 gives me full on distortion, no weirdness...I'm confused!

Could another component be interacting with this capacitor?

[J0K3RX]

remove the cap and add a jumper see what it sounds like.. should still work. Other thing, try using a different cap like say 440n or 220n.. these should work also. You are not using an electrolytic are you?

[canman]

Nope, not using an electrolytic.  I'll jump it and see what I get.  I guess if it works without the cap, may as well leave that cap out, haha!

My gut feeling is that I'll still have a problem even if it's jumped...I'll cross that bridge when I come to it, I guess.  Thanks for the suggestion, I'll let you know what I find!

[canman]

When I jumper instead of using the 1uf cap, I get the glitch at pin 1 of TL074 and lugs 2 and 3 of the gain pot.  Didn't probe anywhere else because I wanted to see how it sounded with the cap in there again.  Pin 1 and the gain pot both sounded like they should (at least that early in the circuit...before the diode clipping I think) but still got glitch at pins 8 and 9.

So maybe the bias really is off, and somehow the cap changes it?  I dunno, doesn't make sense to me.  But then again, not much does these days!   

Thanks for the help guys, I'm trying to learn from my mistakes!

[J0K3RX]

check (really well) for any shorts and double check all of your resistor values.

[canman]

Shoots, I need to get a new DMM...I only get voltage readings, it won't do resistance anymore.  There's always a problem...

[amptramp]

I am not too sure you can remove C5  - when the gain is high, you have a gain of 101 for the input offset voltage of IC1B.  This can pull the signal out of the allowable input common mode range for the next stage.  Add a signal at full gain and you are almost certain to exceed the input common mode range for IC1C.

IC1B has a pot lead going directly to the inverting input.  A pot usually introduces capacitance to ground, so you can get oscillation if the capacitance to ground exceeds C5 / gain.  Add a resistor in the vertical line going from C3 to R6/P1 to minimize this effect.

I also don't see IC2A doing very well when the mid control is moved to the left - you get 0.0909 µF as the load on the op amp output.  The phase lag caused by this capacitance and the internal output impedance of the op amp can produce enough phase shift that the negative feedback goes positive at some frequency. You need a resistor in series with C19.

I would bump the supply voltage up and see if the distortion goes away.

[canman]

Wow.  Thanks for the response...albeit well over my head!  I didn't design this circuit, I just built it according to IvIark's layout.  I've built it before and it works fine, so I probably won't mess around with the supply voltage.  Especially since I'm not sure what exactly I'd be doing, haha.

I'll sleep on it and see if I can find something tomorrow.  I will keep you guys posted!  Any suggestions in the meantime is certainly appreciated

[antonis]

And PLEASE use the right symbols...

(if somebody follow your R1 value the result will be a HUGE signal loss..)  

[duck_arse]

if you've built it before ..... check you've made the cut under R8 470k.

you're not measuring ohms with the power on, are you? and if yr meter still read volts, can we have the voltages on the ic's, please?

and, if you lift one leg of R9, what sounds at pin 8, still glitchy or just VERY loud?

[canman]

I'll check that cut, thanks for pointing me in that direction.  I will also post the voltages in a couple of hours...I've got a few things to get done this morning. 

And yeah, I'm not measuring ohms with the power applied to the circuit...my meter is going bad on me (it was a cheapie) and I can only get solid voltage readings at this point, unfortunately.  But I will get the voltages on here soon, for sure.

I will also lift one leg of R9 and let you guys know what I find.  Thanks for all the pointers and suggestions!!

[canman]

OK gang, here are my voltages:

TL074:
1: 4.35
2: 4.35
3: 2.95
4: 8.87
5: 2.95
6: 4.36
7: 4.36
8: 4.36
9: 4.36
10: 4.36
11: 0
12: 4.36
13: 4.36
14: 4.36

TL072:
1: 4.35
2: 4.35
3: 2.95
4: 0
5: 4.33
6: 4.35
7: 4.35
8: 8.77

As far as I can tell, none of those voltages look off.  I'm probably missing something though.

I removed one leg of R9 (the 1K resistor in series with the 220nf cap) and I'm still getting the glitchy distortion at pin 8 of the TL074.  So I assume this means I can rule out any component beyond R9 in the signal as causing the problem, right?  Can I safely say that the problem lies somewhere between C5 and R9?

I realize how unlikely this may be, but what are the odds that the TL074 is a dud?

Anyways, thanks for all the help again.  I think I'm learning more this time around with debugging, and specifically how to trace through the layouts with the schematic and locate where things are going wrong.  Like I said...I *think* I'm learning...

[duck_arse]

looks like yr volts is good, looks like the 074 is ok. what about that pot, have you a replacement you can try, or even a couple of fixed resistors? next step would be photos of your board.

[canman]

I'm sure I have some kind of replacement I can toss in there and see what happens.  I wonder though, since when I probe lugs two and three of the gain pot with an audio probe and I get normal distortion (at least, that early in the circuit), and the glitchy stuff happens on the second lug of that 1uf cap, would changing the pot make a difference?

I'm just trying to wrap my head around all of this and make sense of it, I'm not trying to say it's a bad or wrong idea, because we all know you guys are far more experienced than I am!

I'll swap out the pot later today, I need to work until about 3 (pacific time) but I'll let you know what I find.

I will also take some pictures of the board...maybe some fresh eyes will spot what I'm missing!

[amptramp]

That 1 µF cap (C5) may be the problem.  Something suitable for this size of board will have a very thin dielectric and any defect will result in erratic sound.

[canman]

OK, I was swamped at work today and just got back.  With it being dark, I'm not sure how much progress I'll make.  I can take pictures in a little bit and trace my build to the layout at least. 

What are the odds that TWO 1uf caps were bad in this case?  I can't get past how one leg of the cap gave me normal sound, but the other leg was glitchy.  I just don't understand how that could be, assuming the cap was in fact a good cap.  Aaaaagh, so frustrating!!

Pics coming within the hour!  Thanks for everyone's patience!!