Bias issue driving me nuts!!! SOLVED w/ embarrassment

[mlp-mx6]

I have built a 50W ODS. I have separate bias controls for the 2 tubes. I have 2 50K 10-turn pots with the -63V common across the "top" of the pots, and the ground resistor common "below" the bottom of the pots. The wiper for each pot goes to the 220K resistors leading to the grids. If you have TUT Vol. 2, this circuit is on page 3-19, from the top right it is the 2nd one down along the right edge. (SET1 & SET2, 2x50K picture)

Here's the behavior - for one pot, it works normally, perfectly, just as I would expect. The voltage on the wiper varies with the position of the wiper, from more negative at the "top" to less negative at the "bottom". For the other pot, the voltage on the wiper wants to stay at the -63V that is available at the "top" of the pot. if I turn the pot toward the "more idle current" end of the pot the voltage on the wiper will suddenly drop at one point, but maintains the -63V up until that point. This "break point" is where the current is 75+mv, way too much current - which is about -35V.

It gets better, though. If I push on the pot shaft in one direction, the pot will behave properly. It will "engage" and work as I expect. If I release the pressure the voltage will jump back to -63V.

I have quintuple-checked the wiring. I have replaced the pot. The pot I removed checked out perfectly once removed from the amp. The new pot did exactly the same thing! I have NO IDEA what the problem is here. Has ANYONE ever seen this behavior? What have I done incorrectly? It seems so simple, and I have used this exact circuit before. I'm completely stumped, and very frustrated.

If I can't figure this out, I'll end up going to the single pot and have less-than-optimum bias. I have tried this, just to check to make sure that the rest of the wiring FROM the pot to the power tube grids is okay, and that worked too.

Please help! Thanks,
Michael

[Bob-I]

I'd lose the shared tail resistor. My guess is that the pots are interacting, when you push the shaft you're disengaging the wiper.

Try it with 2 tail resistors.

[ayan]

I have built a 50W ODS. I have separate bias controls for the 2 tubes. I have 2 50K 10-turn pots with the -63V common across the "top" of the pots, and the ground resistor common "below" the bottom of the pots. The wiper for each pot goes to the 220K resistors leading to the grids. If you have TUT Vol. 2, this circuit is on page 3-19, from the top right it is the 2nd one down along the right edge. (SET1 & SET2, 2x50K picture)

Here's the behavior - for one pot, it works normally, perfectly, just as I would expect. The voltage on the wiper varies with the position of the wiper, from more negative at the "top" to less negative at the "bottom". For the other pot, the voltage on the wiper wants to stay at the -63V that is available at the "top" of the pot. if I turn the pot toward the "more idle current" end of the pot the voltage on the wiper will suddenly drop at one point, but maintains the -63V up until that point. This "break point" is where the current is 75+mv, way too much current - which is about -35V.

It gets better, though. If I push on the pot shaft in one direction, the pot will behave properly. It will "engage" and work as I expect. If I release the pressure the voltage will jump back to -63V.

I have quintuple-checked the wiring. I have replaced the pot. The pot I removed checked out perfectly once removed from the amp. The new pot did exactly the same thing! I have NO IDEA what the problem is here. Has ANYONE ever seen this behavior? What have I done incorrectly? It seems so simple, and I have used this exact circuit before. I'm completely stumped, and very frustrated.

If I can't figure this out, I'll end up going to the single pot and have less-than-optimum bias. I have tried this, just to check to make sure that the rest of the wiring FROM the pot to the power tube grids is okay, and that worked too.

Please help! Thanks,
Michael

Without a schematic in front of me, it's a bit hard to guess. However, as I understand you correctly, you have a common source voltage at -63V, from there you add two pots in parallel with one another, each with a tail resistor to ground. It would seem that both pots would measure around -35V at the junction with the tail resistor. Finally, the bias is (must be based on your description) derived from the pots' wipers. So far am I right on?

Then, one pot works well while the other and its substitute don't. What kind of pots are these? Can't think of any reason why you would observe this behavior. If both pots measure the same value, and the same voltages at the input and output lugs, then it must be a mechanical problem with one of the pots (or a whole batch of pots you purcdhased somewhere).

One comment, I prefer to wire pots as variable resisotrs instead of voltage dividers in bias cirtcuits to prevent the very same problem with describe. Of course, if you will be using two pots in parallel the way you are, you need to be careful because the circuits will definitely be very interactive.

Gil

[mlp-mx6]

Bob, the thing is it is acting as if the wiper were totally connected at the "source" end unless I push on the shaft. This makes no sense to me. Also, if the wiper were disengaging there would be no bias voltage at all on that power tube and it would be in runaway, if I understand you correctly. This is not the case - it has too much negative voltage until it hits some "magic" spot.

Gil, you are correct except that there is only one "tail" resistor at 47K. Perhaps I'll try 2 tail resistors, as Bob mentioned.

To reiterate, if I push on the pot shaft, it behaves properly - apparently without affecting the other pot. And I checked the pot I removed, it varies the resistance exactly perfectly, just as you would expect. It just makes ZERO sense.

The pots are Amphenol 10-turn 50K pots. Seem to be high quality. Model 4201B, RES +-5%, 50K, LIN +- 0.25% is the labeling. Pictures attached.

Thanks again.

[mlp-mx6]

Gents, I'm very sorry for your trouble. I cannot explain how the pot I first installed went bad, nor how it measured well after I removed it, nor how the 2nd went bad also (checked before I installed it). Nonetheless, another pot was installed (along with the separate "tail" resistor) and all seems well now. I'll continue to check it frequently until I regain my confidence in the part.

[Bob-I]

Bob, the thing is it is acting as if the wiper were totally connected at the "source" end unless I push on the shaft. This makes no sense to me. Also, if the wiper were disengaging there would be no bias voltage at all on that power tube and it would be in runaway, if I understand you correctly. This is not the case - it has too much negative voltage until it hits some "magic" spot.

Ok, since I've never seen the inside of these pots I don't know what's happening with you push the shaft. Possibly the wiper is shorting to the source?

I'd doubt if the pot is faulty since you've tried 2 and have the same issue. It must be the ckt.

Gil, you are correct except that there is only one "tail" resistor at 47K. Perhaps I'll try 2 tail resistors, as Bob mentioned.

I've always used 2 tail resistors so I don't really know what will happen if you use only one. I'd try 2 tails since it'll only take a minute to connect. I'd also isolate the bias ckt from the amp, disconnect the wiper from the grid load resistor, and see what that does. Perhaps you have some current flowing that's changing the voltage divider.

To reiterate, if I push on the pot shaft, it behaves properly - apparently without affecting the other pot. And I checked the pot I removed, it varies the resistance exactly perfectly, just as you would expect. It just makes ZERO sense.

You're right, this makes 0 since so there's some factor we're overlooking. Again, take the bias supply out of the amps ckt and see how the voltages react.

[Bob-I]

Gents, I'm very sorry for your trouble. I cannot explain how the pot I first installed went bad, nor how it measured well after I removed it, nor how the 2nd went bad also (checked before I installed it). Nonetheless, another pot was installed (along with the separate "tail" resistor) and all seems well now. I'll continue to check it frequently until I regain my confidence in the part.

I don't think the pots went bad. I think the shared tail resistor was affecting the voltage divider somehow, injecting a voltage into the bottom of the second pot. I'd bet that if you install one of the "bad" pots all will work fine, except of course for your confidence in the part.

[heisthl]

This is one of life's mysteries - Grid current is almost nothing yet there is major interaction when you share a tail resistor. I have fought this in the past and have learned never to do it.....