Bias issues with new 6G2 build... WAY low current draw from power tubes - huh?

[Britt Rossman]

No I changed just the 100k resistor. That value gave ~22mA average for the 5881's. I tried a 120k, which gave ~42mA average. Today I rewired with 110k (2x 220k paralleled) to get somewhere ~32mA to yield 14-15W idle dissipation. The 220k grid leak resistors are still in place and the same value. I can totally see how this could be confused, but I think I'm getting the hang of it.

One problem that still has yet to be figured out is why my screen voltage is constantly 3-5 volts higher than my plate... Maybe I'll move it to the next node. I didnt update the schematic yet to reflect my modified PS... I'll do that and edit.

[g1]

Ok, sorry the 2 x 220K thing threw me off. Doh!
To lower the idle current, you should lower the 100K series resistor. To increase idle current, you would lower the 30K resistor that goes to ground. Raising the value of the 100K should make the idle current increase.

The screen voltage issue if fairly common and no big deal. There is more loss through the OT to the plates than there is through the 1K dropper for the screen supply. You could add series screen resistors at each power tube screen (pin5) to drop a bit more if desired.

[eschertron]

No I changed just the 100k resistor. That value gave ~22mA average for the 5881's. I tried a 120k, which gave ~42mA average. Today I rewired with 110k (2x 220k paralleled) to get somewhere ~32mA to yield 14-15W idle dissipation. The 220k grid leak resistors are still in place and the same value. I can totally see how this could be confused, but I think I'm getting the hang of it.

One problem that still has yet to be figured out is why my screen voltage is constantly 3-5 volts higher than my plate... Maybe I'll move it to the next node. I didnt update the schematic yet to reflect my modified PS... I'll do that and edit.

With the amp at idle (no signal) the DC resistance of the OT may be enough to drop the plates below the screens. Depends on how the PSU is set up. When running at full tilt, the screens will draw enough current to sag their PSU node a bit more than the plate PSU node. And the 3..5v DC drop across the OT will be relatively constant at any operating point, so it tends to disappear relative to the loading. Did I say that well enough for it to make sense?

[Britt Rossman]

Okay guys, I updated the schematic to reflect the changes to the power supply that will (hopefully) clear up a bit of everyone's questions fer me...

[Britt Rossman]

Yes, I understand why I;m losing some voltage, and that's no problem... I'm trying to figure out why the screen is higher than the plate... I'd like it to idle lower, especially with the 6L6GB's. If I switch them for the GC's, I'll be happy running them at more or less the same voltage.

[Enzo]

The screen is higher than the plate because a greater voltage drop occurs in the output transformer winding resistance than in the 1k screen node resistor. And now you replaced that 1k with a choke having even much lower resistance, so even less voltage drop to the screens. Quite normal.

You don;t need a tone stack bypass switch, just tack solder a wire from the wiper of the volume pot to the grid of the next triode, and disconnect the tone stack from those points. We are only doing this as a temporary think merely to discover if indeed the tone stack insertion loss is where your signal is going. Your tone stack circuit has considerably more insertion loss than the original simple circuit. There is no reason to guess or assume, when we can do this simple test in a minute or two.

[Steve A.]

Okay guys, I updated the schematic to reflect the changes to the power supply that will (hopefully) clear up a bit of everyone's questions fer me...

[ATTACH=CONFIG]39345[/ATTACH]

[Britt Rossman]

Okay so here's a new development. I wired in this bypass wire and got some oscillation, so threw a series cap in and got good signal. Now, I said well hell, let's see how she chooches... I blew one of the pair of JBL k110's testing... ouch. seems they weren't right for this anyhow. Guess I didn't think ~30W could do that...

So I move on to testing my other 8 ohm load (a 15" speaker), and as I crank the volume -- this time using the variac to bring node 1 down to 365 from 455 -- it'll go nicely for a few seconds and then spit sputter and get quiet. Same thing that happened with the speaker dying, except no death. It's that typical "I don't have voltage" gargle, but it's not on the power tubes - They are holding bias and plate/screen supply is solid. Works fine again after a few minutes...

Is it possible the preamp tube get saturated and cuts off completely until it can... reset?

EDIT: I thought it'd be funny if both speakers weren't dead after the same gargling death, so I double/triple checked that "dead JBL, and it's still alive... *phew!*

[g1]

If you have the variac on the AC line, what are your heater voltages dropping to when it cuts out?

[Britt Rossman]

Well, I didn't measure that, though it was a consideration. It would theoretically be something like 5.8-6vAC, given that currently it is at 6.7 with 121vAC mains.

However, it cut out before I used the variac.