Weird Cap in Mesa Lead Channel

[chipaudette]

Hi All,

I've been looking at (and trying in my amp) the lead circuit used in the Mesa Boogie MK-IIC through MK-IV. Overall, I think that I understand the purpose of most of the elements that were used to shape the tone and minimize oscillation.

But, there is one cap that I don't know the purpose of. Can any here help me out?

Check out the schematic at the link here. The bottom half of the picture is the lead circuit for the MK IV. Prior to this circuit is the standard Fender pre-amp that was stolen for use in all the MK series (Input->Gain Stage->Tone Stack->Gain Stage->Lead Circuit).

My question is, what is the purpose of C3? It is the little cap (120pF) that connects the grid of V3A to the cathode of V3A. Why is it there? What does it do?

Thanks,

Chip

[chipaudette]

I forgot to mention that, yes, I did implement this in my amp. So, I could cut it out of the circuit and listen to the results. That would be the best course of action rather than theorize here.

But, when I put it in my amp, I didn't do a great job with my layout. As a resultthat cap is tucked underneath some other stuff and is hard to get to. I could sneak my snippers in there and cut one leg of the cap to take it out of the circuit, but I'd have a heck of a time getting the soldering iron in there to put it back in the circuit. I don't really want to do that. So, we're stuck with talking about it instead.

Thoughts on its function?

Chip

[pdf64]

As the cathode is bypasssed, the cathode end of the 120pF cap is effectively grounded. So it's acting as a low pass filter, in conjunction with whatever series resistance there is between the grid and the voltage source feeding it, which will be at least ~40k plate circuit resistance, + 680k series resistor // 1M // 470k = ~228k, therefore the filter break point will be 5k8Hz or higher, depending on the setting of the gain pot. Most likely there to stop oscillation. Peter.

[chipaudette]

Thanks, Peter. I agree with your interpretation. I just have one lingering doubt with that interpretation...

If the intention was to cut the highs (oscillation) by shunting to "effectively" ground, why didn't they just shunto to real ground? Why not put that cap to ground instead of to the cathode?

My only thought was that maybe something funny happens when the tube is overdriven. In that case, the grid might want to go negative and the cathode might not be able to. In that case, a weird disjoint might happen between the grid and cathode and the cap in between might mediate that disjoint in some favorable way.

Any thoughts?

Chip

[Robert M. Martinelli]

Thanks, Peter. I agree with your interpretation. I just have one lingering doubt with that interpretation...

If the intention was to cut the highs (oscillation) by shunting to "effectively" ground, why didn't they just shunto to real ground? Why not put that cap to ground instead of to the cathode?

My only thought was that maybe something funny happens when the tube is overdriven. In that case, the grid might want to go negative and the cathode might not be able to. In that case, a weird disjoint might happen between the grid and cathode and the cap in between might mediate that disjoint in some favorable way.

Any thoughts?

Chip

Hi Chip,
I agree with your and Peter's interpretation, the cap is surely there to stop oscillation, but I don't understand what you mean with "the grid might want to go negative and the cathode might not...".

The grid is made negative vs. the cathode by the bias resistor, which lifts cathode from GND thanks to the voltage drop across it, and the electrolytic bypass cap is there to make this cathode potential stable ( to achieve the maximum voltage gain over freq. ), so the grid starts from the designed center point ( bias ) and then it either goes more negative ( during the negative half of the signal ) or it goes less negative ( during the positive half of the signal ). Should the grid go positive it would cause the grid to attract electrons and a current to flow between the grid and cathode so input resistance would fall and so would do the signal at the grid, so AFAIK the grid is already kept more or less negative ( with respect to the cathode ). Am I missing something?
Best regards
Bob

[stokes]

Positive feedback loop.

[Daver]

Positive feedback loop.

That's what I was thinking because the grid and cathode are in phase. But, since the cathode is bypassed to AC ground is there anything to feedback? I guess I'll have to try it!
Dave

[Enzo]

Don't over-analyze it. Perhaps the cap went to the cathode because it was closer than the nearest ground point - easier to get to. If the cathode is as good as ground, on a 12AX7, pins 2 and 3 or 7 and 8 are right next to each other, a perfect spot for a small cap. I am not concerned with the cap's proximitry to the circuits, just thinking that as much stuff as Mesa crams into small spaces, the less trace length they can use the better for layout.

[Robert M. Martinelli]

Positive feedback loop.

To me that cap simply seems to be acting as an LPF shunting higher freqs to GND ( oh, well, to cathode ) to avoid self-oscillation - if it' s something different, then I am INDEED missing something. I could think about a PFB if the signal was getting back to another point in phase with the grid and with an AC potential different from zero ( signal-wise ), but here the cathode is completely bypassed so signal-wise it actually is at zero potential, this is why I think the cap to be a simple LPF.

If I am wrong please enlighten me, I'm here to learn as much as I can.....

Best regards

Bob

[chipaudette]

So, if it only used to have some high freq roll-off to minimize oscilation, I think that it is interesting that they use this particular technique on this stage, but on the next gain stage they use a cap across the plate resistor to achieve (I assume) the same function.

So, cap from grid to (AC ground) cathode in one stage and cap across plate resistor in the next stage. That difference seems odd to me. Why not use the same technique in both places?

Maybe it's as Enzo says...don't overthink it. Maybe it just turned out that way by accident when prototyping the first prototype Mark IIc and they stayed with it.

Odd.

Chip