Signal dips when pot is adjusted

[DoofusMaximus]

[Hey all, new to the forum and very new to circuit design... definitely only know enough to be dangerous at this point... but working very hard to learn!]

I have been modifying a fuzz circuit by adding a 5K pot (functioning as a rheostat) to the emitter leg of the transistor. The problem is, when I adjust the pot (up or down) while passing signal through the circuit, the signal dips momentarily - it comes back quickly, but obviously I'd prefer that it didn't do this... I've been troubleshooting it, and it's not a faulty pot, and I can't use a pot with less resistance (the pot doesn't do much of anything if I use one with less resistance)...

Can anyone tell me what is happening, and how I can fix this? Thanks for your help!

[R.G.]

What is happening is that the capacitors on the input and output of the transistor are charged up to some voltage across them. When you change the emitter resistor, you are forcing a change in the DC bias on the transistor, and thus a change in its dc conditions. This usually causes the overall gain to change for a short time, and the dip then fades out as the transistor reasserts a stable bias condition and the caps charge to some new level.

How to fix it depends on what you're trying to do by changing it, which you unfortunately don't say. If you're trying to change the transistor gain, there are ways to do that without changing the overall bias.

You're not alone in this problem. Z Vex shipped many pedals with a knob that changed gain that was labeled "Crackle OK". His circuit used MOSFETs, whose nature doesn't all for the volume dip you see with bipolars, but it amplified every little roughness on the pot surface.

[DoofusMaximus]

Thanks R.G.!!

Yeah I was trying to change the gain on the transistor, so you can control how fuzzy this fuzz circuit sounds - it's pretty nice right now because you can get an overdrive-y sound when the pot is low and a wild buzz when the pot is dimed...

Do you have suggestions for a better way to do this?

[PRR]

Try a smaller cap.

The "dip" will happen quicker.

The bass is weakened, but this is often a good thing in a distorter.

If the cap is already "Carefully Considered", then you don't want to change it, or not much. But if you just put out your hand and grabbed whatever cap was handy, take another smaller grab.

[DoofusMaximus]

Thx PRR - so helpful. Actually... the caps took searching and kinda are the reason this circuit sounds so amazing... I guess I will live with the dip!!

[R.G.]

Eliminate the DC offset problem. Use a fixed resistor for the emitter, and a large capacitor from the emitter to another pot to ground. The pot to ground is wired as a variable resistance (that is, only the wiper and one end lug is used) and it makes the >>>AC<<< impedance the signal sees at the emitter be the parallel combination of the DC resistor to ground and the paralleled but DC blocked pot-as-resistor to ground. Turning down the pot can make the AC resistance to ground go to nearly zero, giving high gain.

There's more juggling to do, but you'll learn a lot doing it.

[PRR]

> I will live with the dip!!

No, follow where R.G. is leading you. The problem is certainly solvable, even if my cheaper approach isn't your answer.

And post a schematic. Now I am not sure if you are hanging on a Base cap or Emitter or Collector cap.

[thermionix]

Z Vex shipped many pedals with a knob that changed gain that was labeled "Crackle OK". His circuit used MOSFETs, whose nature doesn't all for the volume dip you see with bipolars, but it amplified every little roughness on the pot surface.

Not knowing the circuit, I had assumed it was DC across the pot, like a Rangemaster.

[R.G.]

Not knowing the circuit, I had assumed it was DC across the pot, like a Rangemaster.

It was DC across the pot. It's just that the MOSFETs don't have current flow through the gate to cause the volume dip. This bipolar circuit would probably have crackle from the DC too, but I think the volume dip from readjusting the DC conditions is covering it up.

It's just a theory...

[DoofusMaximus]

Hey, here's the schematic - it's based on Christian's Buzz Box circuit:



R.G.: thank you for the suggestion to fix the DC offset problem, but I don't quite understand how to wire that. The transistor emitter to cap to pot to ground makes sense, but the fixed resistor also coming off the emitter... what do I connect the other end of that to?

Thanks for your help guys!!

[DoofusMaximus]

*bump*

[anotherjim]

Did you try replacing an emitter resistor with a similar pot (5k) and then a cap to ground from the pot wiper?

[R.G.]

Here's what I was talking about:



The idea is to leave a DC resistor from the emitter to ground, and add a capacitor-isolated pot that can go from zero ohms to maximum. The pot and the emitter resistor parallel to 5K as shown, and the AC impedance through the cap will go down to 0(ish)  ohms when the pot is turned down.

[DoofusMaximus]

Thanks R.G.! I will try that ASAP and let you know how it goes.

anotherJim: thank you! you're suggesting the capacitor come AFTER the pot, on its way to ground, right? It looks like R.G. is suggesting the cap come before...

I can try both... it's also very possible I am misreading or misunderstanding...

[anotherjim]

The cap could fit in at either end of the variable, makes no difference. That should be as noise free as you can get it. However, if C is polarized, it can still be noisy/dippy if the cap is leaky or you fit it reversed - should be negative towards ground even if it isn't directly to ground.

To make things simpler, you could replace the 10k emitter resistor with a 10k pot terminals 1 & 3. Now the wiper terminal 2 connects the C to ground. This works like the version drawn above, saves one resistor, but will have a slightly different sweep response over the turn of the pot.
If it turns out that the DC bias is bad with 10k, then you just use a 5k pot instead, so bias stays as designed.

[R.G.]

A.J. is correct. The cap must have the DC across it in the correct direction if it is polarized. If it's non-polar (NP) or bi-polar (BP) or film type cap, it is not polarity sensitive.  Also, you can use the two "outside" terminals of the pot as the emitter resistor and tie the cap to ground. He's also correct that this will have some noise from having DC across the pot body and having the wiper move up and down the DC levels across the two ends of the pot. This will cause some crackling and also some "thump" or "heaving" as the capacitor momentarily has to change to a new DC level and this affects the inatantaneous bias of the transistor.

The version I drew sidesteps these issues by never letting the pot wiper go to a different DC voltage; the capacitor blocks DC from the pot wiper; the pot in turn keeps the capacitor pulled down to 0V on one end and so the capacitor's DC voltage does not change. The cap and the pot help each other stay quiet.