Ampeg GS-15R Gemini VI

[mtlbasslad]

Well we don't speak French, so it left here.

I love your humor Enzo

[g1]

Derecho is of Spanish etymology.

[mtlbasslad]

Derecho is of Spanish etymology.

Yes, means 'big wind' & we sure got that - still 50,000 without power. We got ours back after only 2 days - generator still working fine.

[mtlbasslad]

The diode must be there or you would not have the negative voltage at the power tube grids. One end of it will connect to one end of the 56K. Maybe it's under that brown cap next to the copper one.
The resistor added across the 39K is just a tweak of the bias range (will allow more negative voltage at grids). The 39K is shown as 33K at the bias pot in schematic. It will measure something lower now as it has parallel resistance added. Looks like maybe a 68K, which would make the new value measure around 25K ?

So got everything connected, powered up (slowly with the variac) No magic smoke, yay. Turned the lights off in the shop & noticed one tube slightly redplating, hmm. Swapped the tubes & the redplating followed the tube - conclusion: it's the tube not the circuit. These were supposed to be 'matched' tubes - but on one the glass is 1/2 inch longer...

Bias measured -17 at pin 6, range was -14 to -19. Removed the 68k now bias range is -15 to - 25. No more redplating. Strangely, before removing the 68k redplating would only occur at low volume - turning volume up to 6-7 redplating would disappear. It seems to me that the so-called 'matched pair' were far from matched.

Comments welcome.

[mtlbasslad]

Still can't get the -50v bias anywhere I check. And still a little redplating on one tube. Does the missing 100 ohm pot between pins 5 have something to do with this? Or the 39K should be 33K? Or is it a bad tube?

[g1]

Swap the power tubes around and see if the redplating stays with the tube.
If you are referring to the -50V shown on the Piazza drawing, I think it's an error. The factory drawing looks like -22V. What is the max. (-) volts you can get out of the pot?

It's a hybrid bias with cathode resistor as well as negative voltage on the grids. Without the cathode resistor it will need to be a bit more negative, but I think you should have range to accommodate.

[mtlbasslad]

Did the tube swap & redplating stays with the tube. Max is -26v range is -19 to -26. that is after removing the 68k.

See post #34

[g1]

Are the extra screen resistors on the power tube sockets still in place? (removing them may increase red-plating)
At full max. negative bias volts, the red-plating still happens when playing?
It's definitely a tube issue, the question is just whether it can be accommodated.
Can you measure the power tube idle currents? (OT voltage/resistance method)

[mtlbasslad]

Extra 1K resistors still in place. Full max neg bias is now -26v no redplating - but only tested with signal generator. Yes i think the so-called 'matched tubes' are not at all.
Tube idle current measure is beyond my current knowledge...
I think I'll order new tubes - if any are available...

[mtlbasslad]

Besides the bad tube problem, should I be finding a 100 ohm pot for that 'balance' between the power tubes?

[Helmholtz]

Besides the bad tube problem, should I be finding a 100 ohm pot for that 'balance' between the power tubes?

Definitely.

[mtlbasslad]

Definitely.

Thanks, found one & ordering now. To install, just put it between the power tubes with center (wiper?) to ground?

[The Dude]

Yes, tube cathodes to each end of the pot and wiper to ground. It's basically a variable/dual cathode resistor to match tube current. I'm curious, though, what is there now (cathodes to ground)? You must have something or tubes would not be conducting.

[g1]

Definitely.

What kind of wattage for the pot?

[Helmholtz]

Pot power rating should be at least 2W.

Edit: Increased to 2W to accommodate power tube clipping.