Preamp negative grid voltage, oscillation?

[Michael Allen]

Chuck: Yeah exactly. Each individual stage is fine, couple together it gets jacked up.

MWJB: The schematic tube labels are irrelevant, that's just what eagle labeled them. Stage 1 and 2 are V1. Stage 3 and follower are V2, PI is V3. Part values were what I had available, I tested the tone stack in Duncan and it works fine. At this point those parts and stages are all stripped out. Nothing is in the amp except the first two preamp stages just to eliminate any other possibilities. I'll try to take a picture of how I've got it going right now. The frustrating part is that it was a working amp that I decided to "re-do" and now I've got issues. Layout as far as tubes and pots/jacks/transformers is still the same.

jmaf: I'll check that, but each stage individually works normally and the part values are correct.

Picture to come

[MWJB]

Well the values that you have at the input triode on the schem are pretty common...there's not much that we can glean from the schem in that respect, so it would be useful to see how the wires are physically routed.

Plus it's worth looking into that glitch with the cathode follower, apparently it's dissipating 0.6mA accross the 100K, rather than more like 1.8mA. Sure you don't have a 10K in there by accident?

When you inject direct into the second stage are you injecting the same VAC as the 2nd stage would see if you were in the input stage (e.g. does it look normal because you are injecting a x50 smaller signal)?

[Michael Allen]

I don't have the cathode follower in the amp anymore. It has been gutted except for the first two triode stages.

Yes i'm injecting the same small VAC signal into the second stage, but it doesn't matter which stage is in which order. I can inject into the second stage and it's fine, then connect the second stage volume pot to the first stage grid and the same phenomena happens.

[MWJB]

OK, so do you get any clean W at the output/speaker?

We really need to see the amp.

[Chuck H]

It's all a little confusing. The amp is gutted except for the first two stages. Does that mean there is nothing at all connected in the signal path beyond the plate output of the second gain stage? Also, try to follow me on this, if you inject a signal at the input of the second stage and it reads clean output, then take that same signal and inject it at the input of the first stage and it reads clean, then, with the signal going into the first stage you connect the voltage divider to the second stage you will have an amplified version of the signal now going to the input of the second stage. So unless you adjust the level of your signal generator your not getting information consistent with actual operation. Try this:

Inject the signal into stage one and measure the output of that stage. Now adjust your signal generator to that voltage and inject it at the input of stage two. Does the stage two grid go negative at this drive level?

[Michael Allen]

Chuck: I see what you're saying. My signal generator is only a simple two transistor sine generator at ~1k so my output voltage is about 500mA VAC. I can't insert a signal on the same magnitude as would come from an amplified stage. These are normal amplification stage gains though and at fairly standard voltages so I wouldn't imagine there is too much drive going into the second stage.

Here's some pictures. It's a mess right now since I've torn everything up and have individual ground wires going back to the star. I have the OT completely disconnected at pulled aside to the right. Closest to the OT is the first stage, then the second stage. On the left/middle of the board is a third stage I added on the second triode to make sure it wasn't just happening with that first socket.

[jmaf]

The voltage you're finding is a ghost voltage between grid and cathode.

Tie a 1M resistor from each grid to ground then tell us what you read please.

Edit: on the direct-coupled stage ahead(cathode follower) do not tie a 1M resistor to ground. Every capacitor-coupled stage will need a tie-down resistor from grid to ground.

[MWJB]

You don't appear to have the PI wired up?

It's not a good idea to have that gridstopper at V1 pin 2 lying against the heater wires.

I can't see a ground reference for the volume pot.

OT placement is bad.

If you are injecting 500mV at the input (which is a lot), you could be seeing 10-15 (or more) whole volts RMS at the 2nd stage grid. It would make more sense to build up the circuit (following best practices in layout/grounding/shielding) then measure clean output at the speaker & work back to the preamp noting AC voltages.

I think you are making life hard for yourself by sticking with that chassis. If the transformers are OK, why not buy a more conventional chassis & utilize them on that?

[Michael Allen]

MWJB, no there is no PI or output section wired up. There is a ground reference for the second stage that i've been resoldering and using for the first stage depending on what stage i'm experimenting with. The grid stopper on the heater gets moved around as I put the input signal on stage 1, 2 or 3 so it's not permanently against that heater wire.

I'm a little stubborn using this chassis I know, but I already built a combo cab to its exact dimensions because this at one point was a working amp. I'll try and wire up the rest of the amp and then work backward again and see what happens.

Jmaf: have yet to get to wiring in those grid references yet, still to come

[MWJB]

Then, I would say wire up as much as you can point to point, not on the eyelet board (use a couple of terminal strips to mount plate resistors), move the input & pots down away from The OT as best you can, shield the input wire.

...or see if you can find another chasis with the same dimensions.