Two different plate voltage readings

[67Mopar]

I have a Marshall 6100LM that reads 488VDC on the plates, with no power tubes installed. When I install the tubes, I get a plate voltage reading of 470VDC on outer sockets V8 and V11. The plate voltage on the inner tubes V9 and V10 read 452VDC. The tubes are a matched quad... If I swap the tubes around, the readings are the same. Any idea what could be causing this?

[pdf64]

Your amp may be oscillating.
Attaching a meter to the plate can result in that, as probe lead may happen to fall across the pre-amp circuits, causing coupling between stages.
Much better to take the HT voltage.
If the plate voltages are more consistent with the LTP removed, then that's a strong indication of the above.

[67Mopar]

I noticed that the Marshall put the swamp resisters on the PCB instead of soldering them directly to PIN 5 of the power tube sockets. Could this be a cause of the voltage difference? I was thinking about adding 5.6k resistors to PIN 5 of each socket, or remove the 5.6k resistors from the PCB, and secure them to the sockets.

[pdf64]

Swamp resistor = grid stopper, so the absence of a full complement of adequately implemented grid stoppers may reduce the margin of stability sufficiently for the scenario described.

[Stevem]

How do the 4 screen resistors test out?

[67Mopar]

How do the 4 screen resistors test out?

All four test good at 5.6k ohm. They are roughly 6" away from each #5 terminal. I'm thinking about either adding 4 x 4.7k resistors to pin 5, or removing the resistors from the PCB, and adding 5.6k r's directly to pin 5. Maybe use 10k instead of 5.6k?

[Stevem]

Thanks, but those are not the screen resistors, the 4 screen resistors are 5 watt resistors!

[martin manning]

Check the screen resistors (470R 5W) that connect to pin 4's, and where are they located, btw? I would add 4x 5k1 or 5k6 at pin 5's to see if that changes anything. You needn't remove the ones on the board.

[67Mopar]

Sorry... Yes, the screen resistors checked good, but I will check them again. They are soldered to the PCB, along with the 5.6k grid-stoppers... Will I still get an accurate reading?

Thanks! I didn't want to have to remove the grid-stoppers from the board. I would like to use some 10k Dale resistors that I have on-hand. I assume 15.6k is fine as well?

[martin manning]

Well that's not the best place for the screen (g2) resistors either, but I would start by adding g1 stoppers at the sockets. I'm assuming that you don't have access to a scope to see if indeed you have an oscillation going on here, but for a dollar's worth of parts and a few minutes time it's worth a try.

[67Mopar]

I'm going to re-check the screen resistors. Being soldered to the board, will this effect my readings?

How high can I go with the screen resistors? Can 1.5k screen resistors be used in a 6L6 amplifier? I will just add additional screen resistors (for a 1.5k total) to the socket terminals, if there is no minimal requirement.

[martin manning]

No, not as long as the connections are good, but screen resistors have the dual function of limiting screen dissipation and damping oscillation, and they will be more effective if they are on the socket pins just like the g1 resistors. Btw, are the power supply filters in good shape?

[67Mopar]

The amp is a '96. The filter caps are original. No visible swelling... Have not checked ESR's, nor do I have a good ESR meter.

[martin manning]

20 years old? Probably time for those to go.

[67Mopar]

I added an additional 5.6k grid-stopper directly to pin 5 on each power tube socket. I've had the amp running now for 10 hours without so much as blip or pop from the power section. Even at extreme volumes, the power section is rock-solid! The only mishap was that I bent the corner of the rear panel outward. Pisses me off something fierce! Anyway, it's running like a champ, and sounds incredible. The only remaining tweak I would like to do, is trim some of the top-end off of channel 3.