Massive volume drop on a gain stage

[Ripthorn]

Hey all, I am working on putting together my final pcb for my new submini amp ("the kitchen sink") and found something strange when I play with the pcb.  The problem is with the third gain stage.  I have a solid input (everything as normal) on the grid of the third triode, but on the plate, I have a greatly reduced signal.  Still sounds good, but the drop is massive.  The schematic is attached below (the first gain stage in the schem is my third gain stage overall) and I am wondering if anyone could offer some suggestions.  Would a very thin trace (i.e. one that got etched under the resist a little too much) cause this, because that is the only readily apparent thing to me.  Also, what is that 470pF cap actually doing?  Thanks.

[frequencycentral]

Yeah, what is that cap doing? It's looking a little like a cathodyne phase inverter, the signal at the cathode is a quieter in-phase version of the out-of-phase signal at the plate. I would guess that you're getting some phase cancellation, ie volume drop. I've never seen a triode set up like you have that one, with the 470p cap - but I haven't seen everything. What if you remove that cap? My guess is that it's dragging your signal down....maybe if you want to keep the cap in (please explain why) you should hook it up after C4. It just doesn't look right to me though.

[Boogdish]

I will second Rick's suggestion.  Lift one end of that 470pf.  Right now you have a negative feedback loop.

I hope this doesn't seem like a rude question, but if you didn't know why the 470pf was in there, why did you include it in your design?

[Ripthorn]

This circuit is actually based on a schematic from a fellow forum member and that was in there.  I figured I wouldnt' remove it until I had a little idea of what was going on there.  I found the main culprit of the drop: a trace that was actually broken right next to the solder pad so i didn't see it.  I am going to experiment with that cap though, it has me wondering the dreaded question (and the reason I got into physics) "I wonder what happens if..."

[frequencycentral]

This circuit is actually based on a schematic from a fellow forum member and that was in there.  I figured I wouldnt' remove it until I had a little idea of what was going on there.  I found the main culprit of the drop: a trace that was actually broken right next to the solder pad so i didn't see it.  I am going to experiment with that cap though, it has me wondering the dreaded question (and the reason I got into physics) "I wonder what happens if..."

Maybe time to PM Jered. Now I'm really curious, Jered definately knows his stuff. Though you have changed the values of the cathode resistors and caps. I'd love an expanation of whats happening with that 470pf.

[kurtlives]

The cab is a snubber cap.

It will reduce a bit of high end and smooth out the tone.

I use them a lot with no troubles, I would way your issues lie elsewhere.

[Ripthorn]

Hi Rick, that is the schematic.  Though I think the one I posted above is a little outdated (more values changed) but the structure is the same.  Jered seems to know his stuff quite well, thus my use of his design as a base, but you get the idea.  I have never seen snubber caps before, but now I know (and knowing is half the battle).

[frequencycentral]

I've seen snubbers before, there's a resistor/capacitor snubber in the nixie SMPS, the Valvecaster tone control is just a snubber, and they are also the basis for voltage controlled filters. I've never come across one connected to the cathode like that one, all the ones I've seen go directly to ground. Live and learn!

[slacker]

The 470p is essentially going to ground, the bypass caps on the cathode resistors make that point look like ground to AC signals.
My Fender Concert amp has caps from plate to cathode like that, like kurtlives said they are supposed to tame the high end a bit. Some people recommend removing them on the Concert though, to "improve" the tone.