Decreasing volume

[edo1123]

Hi guys. I'm trying to build my own pedal, so I started modifying a big muff (I basically removed the tone stack and the gain recovery). I'm not sure the layout is completely correct, but the only omission are the resistors between the output and input of every stage. Here is the problem. It works. It works for something like 30 seconds (sweet and oxtave-ish) and after these 30 seconds, the volume starts to drop and I have to strike harder and harder to get some sound. After 2 minutes I get no sound at all. I didn't use any filtering cap or polarity protection. What could it be? I know it's hard to say, but maybe someone else experienced the same problem and can share the experience.

[PMowdes]

This is worth a look

http://www.kitrae.net/music/big_muff_guts.html#Circuit

I think the resistors and caps between stages are fairly important.

[edo1123]

I took a look. Great site. Do you think it may be because there's a broken capacitor or weird values?

[PMowdes]

No idea, post a picture of what you've made and maybe a schematic.

[Ciaran Haslett]

Using an audio probe would make quick work of localising the problem.

A schematic and pics of your build would help too.

[edo1123]

I removed that "mod" and I don't have the schematic since I was looking to a big muff circuit.

[Ciaran Haslett]

No base resistors?  These are required to bias the transistor into its "on" region.

Measure your voltage on the collectors.  I expect them to be somewhere in the millivolts when you're looking for 3-4ish volts.

[edo1123]

You're right. Got something like 0.01 on base of first two transistor and 0.54 on the last. How could I fix it without desoldering everything? I'm short on vero supplies, so it would be a shame to waste one. Moreover I'd like to understnd more on how to put schematic on vero. I thought that the 1M resistor would have sat the bias for Q1 and Q2.

[Ciaran Haslett]

Try adding a 100K resistor from the base of each transistor to ground.  Your layout won't make this easy.  Maybe do it on the copper side?

[edo1123]

Tried you're modification but still not sounding right so I tried to listen to every stage separately. First one works completely (with the 100k to ground from base) the second and third doesn't sound. Third one doesn't sound at all with the 100k from base to ground. Could it be something whit the layout? Maybe something in che feedback loop? I'm planning to re-drawing it, so every modification are accepted.

[Ciaran Haslett]

Yeah man...it's probably worth a redraw to be honest.  Your Q1 and Q3 have some tricky problems.  Here's what I see.

The 1meg from gain 2 to ground is constantly in parallel with the gain pot.  Remove it.

The 680K feedback resistor on your Q3 (first clipping stage) isn't actually connected between the collector and base.  You need to somehow get its top leg moved down to the base or solder a wire on the copper side from the base to the top leg.

The 220n next to the 680K off the clipper diodes from the base isn't connected to the collector.   Lift its right leg and move it 2 rows below its left leg.

You Q1 connections should be identical to the fixed Q3 but there's a lot of errors here.

470p not between C and B

No feedback resistor between C and B (can't tell if you meant the the 820K or the 220K to be honest)

If all the clipper diodes are coming from the base, then all 3 need to connect to 1 leg of the 100n cap, and its other leg must go to the collector.

So all those changes need to be made, as well as the base resistors earlier.  You're in for a bitch of a night

[edo1123]

I don't think I'll manage to do it tonight, but I'm getting started. As soon as I finish I'll post the result. I know it won't be really worthy due to the amount of "big muff alike" out there, but it's more to get confidence with vero and schematic.

[edo1123]

Actually it took me less than I expected. Here is the review with your modification. Still not sure about all connection, but I'm a bit more confident or at least it looks like it would work. Tell me if there's still something wrong so I can fix it before cutting the last pice of vero (still waiting the order ).

[Ciaran Haslett]

Nearly there

Q1 missing a 100r emitter to ground.  There's room to squeeze it in between the 100r and 100K

The 1N34's bottom legs need to come up 1 row to the collector.

Gain 1 is missing.  I assume you're wiring it to ground.  There's usually a "minimum" resistor between this and ground to stop the circuit going completely silent when gain is fully CCW.  It's not necessary if you're fine with that operation.

But yip...fix those 2 errors and you'll have yourself a fully working fuzz box

[Ciaran Haslett]

I should also say...Nobody learns anything by getting it right every time!  Every mistake you make is a valuable piece of knowledge gained.  Keep up the good work

[edo1123]

Thanks. Since I would end up with some standing diodes, maybe I'll put them on a dpdt with ge and si. I'll fix the 100r from from the emitter to ground and I'll try to see what happens

[edo1123]

And I completely agree with you. Today I learned something on feedback loop and how biasing a transistor correctly. This forum is giving me so much knowledge