Super-hot bias.....please don't say it's a bad OT.....

[Paultergeist]

I am really lost on this one......

Rocket-inspired build.....quad of EL84s cathode-biased via a single 50 Ohm cathode resistor. About 300V DC out of the rectifier with no power tubes installed, but that voltage is pulled way down to around 240V once the tubes go in. Each tube plate dissipation comes in at 23 watts -- super high!

I have swapped out all the power tubes -- same result, so it is not the tubes.

Compared against a couple of friend's very similar builds, the most significant difference is that I have a different OT. Could the OT cause something like this?

Or should I just raise the value of that cathode resistor until the tubes are in spec?

Any help appreciated!

[M Fowler]

Try 62 ohm cathode bias resistor.

[Paultergeist]

Thanks Mark.....

I did not have a 62 Ohm value in the parts bin......I had a 150 Ohm 10-watt resistor, so I expected a very cold bias when I removed the (stock) 50 Ohm value cathode resistor and replaced it with the 150 Ohm resistor.

To my surprise, however, the 150 Ohm cathode resistor resulted in a plate bias of just under 22 watts per tube --- WAY too hot for an EL84, and surprisingly almost the same value as that which was measured with the 50 Ohm cathode resistor in place!!!

I am totally lost......

[Paultergeist]

Other things......

I lifted the 220 uF power tubes' cathode by-pass capacitor (runs in parallel with the cathode resistor) -- no difference.

I also pulled the phase inverter tube (looking for an oscillation, possibly?) and again, no difference on the really hot bias at the power tubes.

There is just so much heat coming off of those tubes (it can readily be felt) and I just do not know how long this thing is going to play before it has a melt down.

The one big difference between my build and those of my colleagues (all of whom have functional amps which bias nicely) is that they have the stock Ceriatone OT while I went boutique with mine......maybe?

[dorrisant]

Sounds like some sort of short between the cathode pine of one of the tube sockets and ground maybe. Post some pics and or voltages for us...
How are you measuring the bias?

[Paultergeist]

I can do voltages; regrettably, I cannot post pictures. I do not know why, but neither my home nor work computer will allow me to access TAG, as some kind of malware detection trigger is getting tripped. I can only access this forum via my iPad currently, and I cannot post photos over. I'll work on getting pin voltages.

To bias: well here is what I do:

Measure power tube plate voltage referenced to ground.
Measure voltage drop across cathode resistor, then -- using resistor value (50 Ohms) -- calculate current for the four power tubes (I = V/R). Divide by four -- this equals the cathode current for each power tube.

Measure voltage drop across screen resistor. I have an individual screen resistor on each power tube in this circuit. Figure out current for the screen using the Ohm's Law calculation as above, then subtract this from the per-tube cathode current. That which remain is the PLATE current. (Note: the screen current is typically a little less than 10% of the cathode current; some folks just deduct 10%).

Knowing the plate current and the plate voltage, Watts = volts x amps. That value is what I am reporting as my per-tube plate dissipation.

[Paultergeist]

Problem solved! Thanks to all who replied to help me.

The issue: floating ground-reference resistors in the PI circuit. As a consequence, the input grids were floating or semi-floating. I must have missed that connection during the build.

Surprisingly, the amp has a number of playing hours on it in that grid-floating configuration, and still sounded really good. I am surprised that it did not melt or destroy something, but I did not even lose a tube during this. To be honest, before this event, I never realized how significant an influence the control grids had upon tube bias -- I had thought the control grids were only involved in feeding the signal -- now I know better.

So, it has been interesting (and frustrating), but it has also been an educational experience. I consider myself really lucky that the amp did not catch on fire, so hot it had gotten.

Thank you to all who offered their assistance to me.

Best regards to all.