A dark tale of big cheese debug...

[Gregory Kollins]

The first pedal I ever built was an OLC Big Cheese. I bought the kit, and followed the instructions. I screwed some stuff up but got it to work. However, the pedal has been haunted with intermittent functionality and almost accidental debug resolutions every time (three times to date). The pedal has sort-of died for the fourth time and I am considering getting a new PCB and starting over. I sort of have a bit of an emotional connection to it now though... I've probably spent total around 30 hours on the pedal and I feel like I must fix it as is.

So help me if you can, though I think that my questions will go unanswered. It's very perplexing, and well, just kinda crappy.

Here's the build guide, which contains everything:
http://www.olcircuits.com/documents2008/olc_chunkycheese_buildguide.pdf

So my current issue is this:
It still passes signal, its behavior is just really funky.
I don't know what is happening. I don't know what my name is.

Most of the odd behavior can be conveyed by describing what is does in various switch positions.
In mode 1, the tone is NOT bypassed, and the gate is ON to such a severe degree that past 12 o'clock on the gain knob, the whole signal drops out.
In mode 2 there is still a noticeable mid scoop, and the tone is on, however there is no FUZZ, instead there is a cheesy (ha) super-crunchy 80's-ish distortion.
Mode 3 is much the same. The mid boost is there, the tone is on, but the sound is cheeseball.
Mode 4 has TONE BYPASSED, NO GATE, and NO FUZZ. It's the same distortion sound, but very very trebbly and thin, no bass at all.

Now for voltages:
Q1 -
C - 1.8v
B - 0.5v
E - 0v

Q2 -
C - 1.8v
B - 1.8v
E - 0v

Q3 -
C - 0v
B - 0v
E - 0v

IC1 -
1 - 0.2v
2 - 4v
3 - 0.4v
4 - 0v
5 - 0.2v
6 - 4v
7 - 9v
8 - 0v

IC2 -
1 - 0.2v
2 - 3.8v
3 - 0.4v
4 - 0v
5 - 0.2v
6 - 4v
7 - 9v
8 - 0v

D1 -
I don't know what to properly call each side of the diode so...
Side with stripe... - 0v
Side without stripe... - 0v

This is clearly wrong. Everything is wrong. Help me.

Oh and an extra note - In the various failures of this pedal, these voltage readings have been wrong before, but in different ways. And then I look at it for hours and get exhausted replacing various parts to no avail. I poke it a few times... come back to it an hour later and it works.

[marrstians]

1st step is to double and triple check to make sure you put the right parts in the right spot... then check every connection for continuity... i had a problem with the cheese with wiring the 4 way switch wrong on my 1st try and it was all out of wack... as far as your Q's V's, check them again with the 4 way switch on position 1 and 4... the switch is connected to one of the trannys so it's only on in the right position... hope this gives you a little start till the professionals give you some tips

[Gregory Kollins]

Ah well I took that into account and compared V's for every switch position. The do not change over any of the four positions for any of the transistors.

And - All the transistors are, correctly, 3904s. They are oriented correctly. None of them are "bad". I replaced all three, twice. Both IC's are good. They are TL071s. I tried 81s. Same thing. I also tried another pair of 71s. Nothing new. I tried swapping out the diode, three times. Nope. I DID find THREE broken tacks, two at the pad and one more where the actual break was not found (maybe under the latex-or-something coating on the board). Bypassed those with wires. Checked the switch wiring (though it was working before this). It's good. All to spec.

I think one of the main problems I am having with this is, since I did not do the layout I am having trouble comprehending what's going where and what needs to do what. But that's not the problem, just my difficulty in diagnosing it.

[R.G.]

Oh and an extra note - In the various failures of this pedal, these voltage readings have been wrong before, but in different ways. And then I look at it for hours and get exhausted replacing various parts to no avail. I poke it a few times... come back to it an hour later and it works.

Mother Nature is whispering in your ear. If it suddenly begins working more or less OK, then changes to not work yet a different way, you likely have the right parts in the right places, but you have intermittent contacts somewhere opening and closing erratically. It's not surprising that replacing parts doesn't help - that's a lot like trying to cure athlete's foot by removing teeth.   If the parts aren't the problem, then replacing them doesn't help.

It is essentially impossible to debug something like that, other than to know that works/changes problems are almost always bad contacts, and that there are contacts where there should be solid connections. That is - there are soldering and/or wiring problems keeping solid contacts from being made. With a board that's been debugged at for many sessions, it's also possible that the copper pads are lifting from repeated soldering heat and possibly cracking loose from traces.

If it were mine, I'd resolder the whole thing, one pad at a time, making sure of the resolder as I went. Then I'd check the wiring.

[Gregory Kollins]

Well, my lack of experience has led me to do a lot of fumbling around in the dark.

I'm checking over the whole board right now. I'm also re-soldering. It's fun. I'm going back.

[britpeace]

1st step is to double and triple check to make sure you put the right parts in the right spot... then check every connection for continuity... i had a problem with the cheese with wiring the 4 way switch wrong on my 1st try and it was all out of wack... as far as your Q's V's, check them again with the 4 way switch on position 1 and 4... the switch is connected to one of the trannys so it's only on in the right position... hope this gives you a little start till the professionals give you some tips

I am having a problem with the rotary switch. I know I got the right one, the 3p4t rotary from mouser. But i can't figure out how to get it to only click 3 times (giving it the 4 settings it should have). Sometimes it clicks up to 12 times, sometimes it only clicks once. Is there a way to 'set it' or something?

I would reaaaallly appreciate the input, realllyyy

[R.G.]

I am having a problem with the rotary switch. I know I got the right one, the 3p4t rotary from mouser. But i can't figure out how to get it to only click 3 times (giving it the 4 settings it should have). Sometimes it clicks up to 12 times, sometimes it only clicks once. Is there a way to 'set it' or something?

Yes, there is probably an internal stop just under the top plate. There has been on the ones I've looked at. Google search the manufacturer's site for the data sheet on the switch and it should show you.

On the bad side, some rotary switches do not have stops on the rotation. Can't tell without knowing which switch is which.

[britpeace]

I am having a problem with the rotary switch. I know I got the right one, the 3p4t rotary from mouser. But i can't figure out how to get it to only click 3 times (giving it the 4 settings it should have). Sometimes it clicks up to 12 times, sometimes it only clicks once. Is there a way to 'set it' or something?

Yes, there is probably an internal stop just under the top plate. There has been on the ones I've looked at. Google search the manufacturer's site for the data sheet on the switch and it should show you.

On the bad side, some rotary switches do not have stops on the rotation. Can't tell without knowing which switch is which.

Yup, there was a little washer we didn't notice that you use to set it. Thanks! So before boxing it up, it worked perfect. After boxing it up, it does not work perfect! The only thing that was done different was putting polycrylic on the bottom for protection (which we've done before and it worked) and wired up an LED because there wasn't a spot for it on the board.
Position 1 is loud and a little distorted-not fuzzy and the tone control is working with it, which isn't supposed to
Position 2, 3, and 4 is an extreme volume cut, pop, and an almost octave-like effect.
I don't know what could have happened between before boxing it up and after, but I could reeeaaally use some help. 

[R.G.]

Yup, there was a little washer we didn't notice that you use to set it. Thanks!

Great!

So before boxing it up, it worked perfect. After boxing it up, it does not work perfect! The only thing that was done different was putting polycrylic on the bottom for protection (which we've done before and it worked) and wired up an LED because there wasn't a spot for it on the board.
Position 1 is loud and a little distorted-not fuzzy and the tone control is working with it, which isn't supposed to
Position 2, 3, and 4 is an extreme volume cut, pop, and an almost octave-like effect.
I don't know what could have happened between before boxing it up and after, but I could reeeaaally use some help. 

We've all run into this. The answer is - something changed when you put it in the box. If the polycrylic really is nonconductive, then when you put it in the box, something is either broken or shorted out by the mounting process or position. No easy answers here. You just have to find what putting it in the box changed.

The ones I hated were the little balls of excess solder I sometimes generated. These rolled under boards and shorted things, then rolled away when the board was removed, only to short something else later.

[britpeace]

Yup, there was a little washer we didn't notice that you use to set it. Thanks!

Great!

So before boxing it up, it worked perfect. After boxing it up, it does not work perfect! The only thing that was done different was putting polycrylic on the bottom for protection (which we've done before and it worked) and wired up an LED because there wasn't a spot for it on the board.
Position 1 is loud and a little distorted-not fuzzy and the tone control is working with it, which isn't supposed to
Position 2, 3, and 4 is an extreme volume cut, pop, and an almost octave-like effect.
I don't know what could have happened between before boxing it up and after, but I could reeeaaally use some help. 

We've all run into this. The answer is - something changed when you put it in the box. If the polycrylic really is nonconductive, then when you put it in the box, something is either broken or shorted out by the mounting process or position. No easy answers here. You just have to find what putting it in the box changed.

The ones I hated were the little balls of excess solder I sometimes generated. These rolled under boards and shorted things, then rolled away when the board was removed, only to short something else later.

Duly noted. Thanks so much for the help and advice. Patience and understanding has helped the most with this. Currently rebuilding the whole thing.

[chemosis]

almost every pedal ive ever built with a 3p4t has failed. ive built about 5 or 6 builds with 3p4t and only one worked properly. i never have problems with spdt or 3pdt or dpdt. i refuse to build pedals with 3pdt. 20 pads for 1 switch ?

[ElectricDruid]

The ones I hated were the little balls of excess solder I sometimes generated. These rolled under boards and shorted things, then rolled away when the board was removed, only to short something else later.

I had a similar problem with a synth. Worked with the lid open, didn't work with the lid closed. It had me going mental. Eventually I discovered a switch that was busted, and it just so happened that when you tipped the lid up, all the bits dropped into the right places!
Sometimes you have to be a bit lateral-minded.

Those intermittent problems are always a bitch to debug. Bad solder joints are most likely, followed by lifted or cracked PCB tracks. But you can find them if you're patient.

Tom

[Rob Strand]

almost every pedal ive ever built with a 3p4t has failed. ive built about 5 or 6 builds with 3p4t and only one worked properly. i never have problems with spdt or 3pdt or dpdt. i refuse to build pedals with 3pdt. 20 pads for 1 switch ?

The quality of rotary switches these days is nothing short if disgusting.  There's a point where cheap turns to crap and they have manage to succeed reaching that point!

Dismantle one those Black and White plastic jobs you see a bent aluminum wiper, which also acts as the spring, which scrapes over some rough contacts.  By design it is total crap, it will not last.

The open metal/phenolic styles are another type.  These are a better design but I have also seen these fail too.  Sometimes they are rough to rotate which is fixed with a little lube on the ball.

If you look at old test equipment they used heaps of multipole rotary switches and they are still going strong today!   Those guys new how to design stuff.

These days anything that looks half decent is expensive.

[There can only be good bugs in anything to do with cheese.]