Dynaco ST-35 bias voltage skyrocket

[DanGuitarMan]

A few years ago I rebuilt a Dynaco ST-35 using Dave Gillespie's boards and I've encountered a problem that I haven't been able to figure out. I've reached out to Dave and he has been very helpful so far, but I thought it couldn't hurt asking the rest of the AK community.

A few weeks ago a tube shorted out and took out the fuse and a resistor with it. I replaced the resistor, tube, and fuse, and went to re-bias the amp. The bias voltage for this amplifier should be around .27VDC.

A minute after I powered it up, the tubes began to red-plate and the bias voltage was high at .90VDC. Turning the bias potentiometer was not affecting the voltage at all. I immediately shut the amp off and started looking for the cause.

I thought it might be the voltage regulator, but when I replaced it the problem persisted. I measured all resistors shown here in red, and they were ok.

Could it be the potentiometer, or a coupling cap? I'm not sure where to look next. I've attached the scematic below.

Thank you!

 

[primosounds]

Well, it could be any of those things but usually, when you "lose" bias it is because a resistor has gone open and the bias voltage has dropped down causing the tube to red plate. The pot has a contact point, the wiper, to the resistive surface and if it has been damaged it may not make contact which would also be no bias voltage. You can easily measure the pot with a meter, but you might have to disconnect it from the circuit to make a meaningful test. My experience is that the SS device is most likely to fail and cause similar problems. If you 100% sure you have ruled out those items, did you try switching tubes in the bad channel? It may show a bad tube.

 

[Shadowdog]

Assuming the pot is good maybe the output tubes are not matched well anymore or a faulty one is dragging everything down as you have just the one pot to control all 4.

 

[gadget73]

Is it going up on all 4 tubes, or just one? If its just one, I'd say coupling cap. All four and I'll say its probably going to be a component in the EFB circuit somewhere.

Pull the tubes and take some measurements directly on the EFB regulator. If you're looking at the number side, the left-most pin is the adjustment one. That should vary with the bias trimmer, guessing somewhere in the 12-20 volt range. Middle is ground, it should read 0 volts. Right-most is the output pin. It should pretty nearly mirror the adjustment pin, but be about a volt lower.

I did fry an LM337 recently by reversing two of the connections. The output didn't work like it should have, and the voltage across the 10 ohm resistors was way high like you're experiencing. Not sure how tolerant those things are of output tube shorts but generally solid state devices tend to die pretty quickly when abused like that.

 

[daveshel]

Probably not your issue - but I know that if the input grounds are isolated from the chassis ground your bias voltage will float and rise. I happen to know this because when I rebuilt mine several years ago, I accidentally isolated the input ground by using the insulated fiber spacers that came with my new input jacks. I powered up to set bias and saw that it was up around 1 volt and I cut the power.

 

[davidpg]

I imagine that Dave has already said this, but the voltage that you are measuring is not the bias voltage itself. It's a voltage drop across resistors in the ground circuit to the cathodes. The voltage drop represents the current flowing through the tube. The bias is established by the difference between the voltages at the cathode and the control grid. The high voltage drop and the red plating that you see are basically telling you the same thing: There is an inadequate bias voltage between the cathode and control grid of one or more tubes . If you see the same high voltage at both of the test points, it must be something in the bias control circuit.

An additional test that might be useful is to measure the voltages at the bias test points relative to the chassis ground. I'm guessing that voltage will turn out to be less than the 15 v specified in the manual for the EFB board. This is the regulated voltage, and you can test this without the tubes. You can then check that the potentiometer does, in fact, control the regulator, and look for components that might be bad and causing the problem.

I hope this helps.
David