biasing a SF twin reverb

[JC@]

Is there a reason why I shouldn't bias the output tubes with a 'normal' value?
The schem, at least the one for the 135 watt silverface, seems a bit too cold (maybe it is because of the tubes I'm using..), but at the nominal negative voltage I only get about 20mA per tube.
Thank you.

[Tom Phillips]

The "normal" value is just the original "nominal" reference voltage that Fender picked for the tubes of the day. Just a starting point really. There is no reason that you need to set the referenced bias voltage. I reality we choose the bias voltage to get a bias current which will result in the desired idle plate dissipation.
Tom

[JC@]

problems

When I measure the voltage on the plates it is sometimes 505 volts, then sometimes 520 volts.
The tubes sometimes iddle at 15mA, then 30mA. Exactly half the expected being the second reading the good one I'm trying to set.

I know I have a problem here but I can find the cause.

Any help?

[Amp Kat]

I would try changing the coupling caps which are the .1uf 600v usually blue ones right before the power tube grids on each pair of tubes. If they are leaking your bias will flucuate and eventually it will go into thermal runnaway. If you put your meter on the negative bias you can see it go up and down. A bad bias capacitor can do it to but usually just drops the voltage to a lower set value.

[Bruce / Mission Amps]

How are you measuring this?
What is your setup?
I very very seldom find these to be idling anywhere near 30ma each tube and have to modify the bias supply.
Usually they are painfully cold "ear wise".

[JC@]

Bruce, I have been letting it be for a while but I'm back now.

When I first opened the amp, it was iddling at about 20mA each tube, and about 496 volts on the plates.
I then made the measuring and found it should be set at 31mA or 32mA, and while the amp was there on the bench, while looking for the appropiate resistor to "mod" the bias circuit, I noticed that it went down to 10mA each tube. Then up again to 20mA. Sometimes it is 10mA, sometimes 20mA. Not a fast change: unplug the amp, have a look, turn on again, change. Or leave it on for 20 minutes while at 10mA, the suddenly both tubes go up again to 20mA, then a while after, back to 10mA.

I noticed that the negative bias supply is always about correct, just the voltage on the plates and the current through the tubes change.

The amp is working with just two power tubes, not four, maybe that's why plate voltage is a bit high.

In my first message I mentioned 15mA and 30mA, but that was after the bias "mod". Now it is back to stock: 20mA, and when the problem kicks in, 10mA.

Once, it skyrocketed up to 60mA each tube. Inmediatly cut the mains, let it cool, then turned it on again, and it was back to 20mA.

At this moment I'm checking the .1 coupling caps.

[Euthymia]

Bias current climbing can be due to ultra (or sub) sonic oscillation.

[JC@]

How are you measuring this?
What is your setup?

Well.. I get different readings using the transformer shunt method, almost back to normal.
I'll try some other ways for measuring and monitoring the current tomorrow. Maybe it is just some of my measuring equipment what is failing.