5879SE new issue

[firemedic]

Okay, I know I said I was done with this one but....
Just ONE MORE TIME, I went back in to change out the MV for a fatness control, a 250k pot controlling the grid to the parallel 2nd PA stage. Since the wiper was connected to the volume wiper I disconnected the ground to this 2nd pot. Prior to this it was working fine.
I finished up, plugged it in & played for about a minute. I turned up the fatness & volume to see what it would do when it made a scratchy noise then- silence. No hum even.
Changed out power tubes- no change.
OT fuse is OK. I took out the PA tubes so I guess the prob is in the power section.
When I checked the plate voltage, at first it was going all over the place. Then sat at >400 (should be 310-350). Then it gradually went way lower than normal. The cathode voltage, nominal 21v, goes up to over 50V. I changed out the 560R 10W cathode resistor but no change there.
I had the 6V6GT biased kind of hot in there, at 38ma, I think 340v or so plate to cathode, I don't have the numbers in front of me.
So what could be causing these wacky voltage swings & overvoltages?

[firemedic]

OK, fixed it. I'm embarrassed to say what it was.

[Geezer]

Oh com'on...we're all friends here. 

Forgot to solder the new connection? didn't have a tube in the socket? What? 

While I know it is a humbling experience, telling what we did (or didn't do) can help others with similar problems in the future.

Geezer

[darkbluemurder]

Just ONE MORE TIME, I went back in to change out the MV for a fatness control, a 250k pot controlling the grid to the parallel 2nd PA stage. Since the wiper was connected to the volume wiper I disconnected the ground to this 2nd pot. Prior to this it was working fine.

Could it be that the lifted ground connection upset the power tube's bias and made also the plate voltage rise?

[bluesbear]

Yeah, really. When you get experienced and are what I like to call "pretty good" at this stuff, the small "stupid" mistakes are the ones you're gonna make. They're also the hardest ones to find! So... if you've already found yours, you're doing good.
Dave

[firemedic]

Yes, that is exactly what happened. When I took the 250k pot (with its ground reference) out of the signal to the power tube, I forgot to replace it with a 220k grid leak (like a 5f2).
I had NO IDEA it would do those weird things to the plate voltage. It never occurred to me that the little grid signal voltage could have that effect, so of course I wound up chasing my tail for a while. I had to stop & think, retracing my steps, then popped in a 240k gridleak (all I had on hand) and everything is wonderful again. I'm glad there seems to be no permanent damage.
BTW the 250k with the ground connection lifted doesn't really do anything with the sound. I would probably have to use a 1M for it to have any effect.

[PRR]

> It never occurred to me that the little grid signal voltage could have that effect

That's the magic of tubes. A great big bottle able to flow a lot of current, but can be controlled with a teeny grid signal.

With NO steady DC connection to grid, you are not TELLing the tube what steady current to flow. The tube has no real opinion on this, just sorta wanders. Could be zero. (Which would un-sag the PS to a high voltage.) What's this, a 6V6? Could be over 160mA. (PS sags to way-lower.) With a coupling cap from a previous stage, there's a strong tendency for turn-on and leakage to bring the big tube to maximum current. Which in an SE Class A stage is at least twice the designed steady current, and liable to smoke something.

> The cathode voltage, nominal 21v, goes up to over 50V.

We actually want the GRID-to-Cathode voltage. With a 100K-1Meg grid-to-ground resistor, grid "IS" ground (within a volt). WithOUT a grid to ground resistor, grid can go anywhere. In this case it wandered up to +30V or +50V. The cathode does not know the grid has got loose, so it obligingly rises to over 50V. 50V/470 is 100mA cathode current..... something's gotta give sometime.

Is a good idea to check all your grid-to-ground resistances, or at least the Power Tube grids, before turn-on. Maybe I will try that someday.