Lowering gain by adjusting cathode resistor value or removing cath. bypass cap

[Emetal]

Since I was not satisfied with trying out different preamp tube, I am thinking to lower the gain on channel three by maybe removing cathode bypass cap or by increasing value of a cathode resistor. Am I thinking in the right direction? And should I do that at V1-A (C17 and R42) or at V2-A (C16 and R45) location... or any other thoughts and suggestions on values and ways to do it, please? Talking about my EVH 5150 III 6L6 MkII 50W amp.

[loudthud]

Increasing the cathode resistors will lower gain, but also headroom. There are limits on how much gain you can remove and still sound good. Removing the caps will lower gain, but not as much in the low end. Can't you turn the Three Gain control down far enough ? Inserting a resistor between WJ10 and WJ3 might help, try 1 Meg.

[DrGonz78]

Perhaps change plate resistor R72 to a 100k value. In a few instances I have removed cathode bypass caps on amps that I designed. Those designs were all about cascading gain with a preamp full of too much gain. Removing the cathode bypass caps helped remove a tiny bit of gain. It was enough to tame it down a bit but it was also about the tone too. Since the bypass cap was part of the equalization of that stage of the circuit. In my case it worked okay but it is finite adjustments at that stage of tweeking the circuit.

[Mick Bailey]

Can't you turn the Three Gain control down far enough ?

That's an important point. Because the control goes to zero, is there a point where the gain reduces to where you'd want it (even if it's only just fractionally off zero)? This would give a better insight into where the gain reduction is best placed - whether the signal needs to be attenuated just before the gain control but after the preceding stage, or whether the gain of that stage needs reducing. Or maybe some of each.

[Emetal]

[QUOTE /Can't you turn the Three Gain control down far enough ?/QUOTE]
If I turn Three Gain too low like below 2, I start loosing power, or should I say volume all together. It sounds tiny and sterile.

[Emetal]

That's an important point. Because the control goes to zero, is there a point where the gain reduces to where you'd want it (even if it's only just fractionally off zero)? This would give a better insight into where the gain reduction is best placed - whether the signal needs to be attenuated just before the gain control but after the preceding stage, or whether the gain of that stage needs reducing. Or maybe some of each.

Well I would love somewhere beteeen 1 an 2. Right now if I do that it sounds weak, tiny... So gain pot needs to go on let's say at least 3 but to have less gain

[loudthud]

If I turn Three Gain too low like below 2, I start loosing power, or should I say volume all together. It sounds tiny and sterile.

OK, like I said insert a resistor between WJ10 and WJ3, maybe 1 Meg, then change R14 to something like 100K so you can get the THREE GAIN control up to where it doesn't sound thin. If it sounds too dull and lifeless, you'll need a small cap between WJ10 and WJ3 in parallel with the resistor, maybe something like 100pF to 270pF.

[Chuck H]

I haven't gone back to the old thread to see if this suggestion has already been given or been done, but...

If the Z1/Z2 circuit is clipping at whatever gain setting you use then it seems to me that if you lift either lead of either Z1 or Z2 you would be able to reduce clipping and actually gain a little volume. Allowing you to keep the gain knob in a better spot for balance with CH 1&2 perhaps. This would probably change the character of the channel quite a bit. but maybe for the better.

EDIT: Don't bother with this. See post 10

[Helmholtz]

Z1/Z2 might be necessary to limit the drain-source voltage of Q1.

[Chuck H]

Z1/Z2 might be necessary to limit the drain-source voltage of Q1.

Must be the case. I had suspected by the Zener value and their location that they might not be signal clipping diodes so I simulated the circuit. The most you'll get out of that first triode is going to be about +/- 10V on big chord attacks. So at 16ZV they'll never clip the signal even with the gain pot up full.

[Helmholtz]

Must be the case. I had suspected by the Zener value and their location that they might not be signal clipping diodes so I simulated the circuit. The most you'll get out of that first triode is going to be about +/- 10V on big chord attacks. So at 16ZV they'll never clip the signal even with the gain pot up full.

According to schematic test voltages the gain of the first triode is 47. So a 1Vp input wil cause a 47Vp output.
1Vp PU signal at attack is possible.
V2a grid conduction will limit positive signal but not negative signal.

[Chuck H]

1Vp PU signal at attack is possible.

Maybe. But that's a really hot pickup and really fat strings. I remember when Dimarzio came out with their Super Distortion they claimed half a volt. Not being satisfied they came out with an even hotter pickup. I can't remember the name on that one but I do remember that it had big hex head pole pieces and they claimed as much as a volt of output. I've tried to capture peak voltage output for guitars a few times using needle meters but only ever got what I would have to qualify as questionable results. Usually 250mV to just over 500mV. These are by no means qualified tests but as a result I've always considered maximum guitar output on attack as potentially half a volt.

[Helmholtz]

Zollner measured PU instantaneous peak voltages up to and above +/-1.5V using a LP with Burstbuckers strumming chords.
As these are fast transients measuring requires a scope.

[Chuck H]

Zollner measured PU instantaneous peak voltages up to and above +/-1.5V using a LP with Burstbuckers strumming chords.
As these are fast transients measuring requires a scope.

I don't doubt that result. I was pretty green when I did the test I described above and didn't even own a scope yet. I'm sure a good capture on a scope would give more accurate results than I was able to get with a cheap needle meter. At any rate...

Even at a volt of input it doesn't seem to matter in the simulation because grid conduction from the following stage smashes one end of the wave form before the diode can have any affect and the other end of the wave form is clipped in that stage well before the diode clips what is going in. There is no difference in the clipped wave form or equalization from the second preamp stage with the diodes in or out.

[Helmholtz]

So best leave the zeners alone.