ImPasse Preamplifier

Yup. Just double checked it again. R8 is 680 K

Here is something else that seems strange to me. When I remove R9 from the circuit, it measures a perfect 1 M. Bands match the other 1 M resistors used elsewhere on the board. All is good and happy. When I measure it in circuit it reads erratically around 60 K.
 
OK, those are the voltages with the ECC88 removed? If so, the voltage divider is working normally. You've got at least two possibilities: the ECC88 is bad, or it's oscillating at RF. If replacing the ECC88 has no effect, install a grid stopper resistor as close to each grid pin as possible. That will take an Exacto knife and a few minutes.
 
Well, I don't have another ECC88 to test with, so I will need to order up another. Do you think I should throw in a grid stopper in before I order up another tube? If so, what value should I try. I read 68 K somewhere is a place to start. I have some 75K in my box of parts.
 
I just probed the grid power with my scope and I saw no noise. I dropped the time resolution down as far as it would go and there was a tiny bit of hash, but nothing that would cause this kind of oscillation. I am guessing that the tube is bad.
 
So, I just ordered up a new Electroharmonics 6922. A bit of a bummer because the tube I was using was a nice vintage Tungsram from Parts Connexion. At least with the new Electroharmonics I will know it is not the tube if I have more issues.
 
I should have stated in my question, that if the devices aren't heat sinked the regulator gets itself into a bit of trouble. I had the voltages wander around owing to this factor.

I was not satisfied with an artificial center-tap on the original transformer I used -- and use the Allied one instead.
 
Thanks for the ideas. Everyone.

Last night I turned the preamp on and immediately probed the gate pin. I watched as the bulbs went out, the voltage rose to 150 V. I think that would rule out heat in the regulator as it hadn't been on for more than a 30 seconds.
 
I'm still working on it -- ran some tests today -- I think it tests a hair better than the author would admit to -- first chart is 20V out, second is 2.83 out:

An externally hosted image should be here but it was not working when we last tested it.


An externally hosted image should be here but it was not working when we last tested it.


I still have a little work to do getting some of the 60/120 Hz a.c. out -- the layout in the PAS chassis might be a little too cramped.
 
Yes, its "balanced to balanced" -- the only load on the preamp is the AP Analyzer. I'm using the Sowter transformers which will add a tad to the distortion. The input impedance needs some work as I have both ALPS volume and balanced pots which would be 100K||100K = 50K. I am using Electro Harmonix bottles.