HR Deville Runway Current Problem

[Perkinsman]

Hi all, This weekend I worked on a HRD that had alot of crackle/sizzle. I fixed it by resoldering the PI pins and the amp sounded good but because one of the filter caps was visually leaking, I replaced all 6 main caps. Brought up the amp slowly on the variac (equipped with a volt & current meter). Everything was fine, it sounded good after that point too. Then for some unknown reason I decided to replace a small cap that happened to be in my line of sight at the moment, C38, 22uf 63v. One leg of the trace happened to lift and I repaired carefully with a hardwire from one end to the other. I soldered in a new cap in and plugged into the variac. I brought it up slowly and the current started rising too quickly so I shut it down. Since it was the last thing I had done prior to the current rise problem, I assumed it's the trace repair...... but it showed continuity for the repair so I reflowed all the cap solder joints for the new caps I had installed and still no improvement. Any suggestions how to troubleshoot this since I can't power up past 60vac without the current jumping past 2A?

[The Dude]

Double check to be sure you got cap polarity correct on all the caps you replaced. C38 is a bias voltage filter, so the positive side will go to ground.

[Enzo]

Your new trace showed continuity for its original path, but perhaps it has gained continuity to some other adjacent trace where it ought not be?

C38 is the filter cap for the bias supply. That is a NEGATIVE supply, so any chance you soldered it in backwards? The POSITIVE end is grounded, not the negative.


Ah geez, I had ya beat, but got caught up in Ferris Bueller singing Beatles tunes.

[Perkinsman]

Right, i noticed that positive went to ground and figured it was a bias supply cap when I put a new one in...I suspect it's something to do with the the hardwire trace patch. Be checking that tomorrow.

[Enzo]

OK, well check the neg end for resistance to ground. Is it shorted to ground somehow? I had a guy once bring me a unit he had recapped. He forgot to trim the excess lead wires off one of the ccaps, and they stuck against the metal chassis, shorting out his HV line. No it wasn't me, but I have done bonehead things too.

[Dave H]

Try running it up with the power tubes removed. If it runs up OK check the bias supply voltage.

[Perkinsman]

OK, well check the neg end for resistance to ground. Is it shorted to ground somehow? I had a guy once bring me a unit he had recapped. He forgot to trim the excess lead wires off one of the ccaps, and they stuck against the metal chassis, shorting out his HV line. No it wasn't me, but I have done bonehead things too.

I’m attaching a couple pics of the hardwire patch I ran. It runs from the negative side of C38 to the common legs of R59&R60. I was just following the trace. In the schematic, the negative side of C38 (C-) goes to the common legs of R59&R60. Did I do this correctly?

As I stated, it ran fine after recapping (I pigtailed). The runaway current problem started after I replaced C38 & had to hardwire patch one side due to a lifted trace.

As recommended, I pulled the power tubes & as a result, now there is no runaway current.

[Perkinsman]

Heres the pics

[Enzo]

You bypassed my question. I do not doubt you connected the cap to the resistors correctly. But I asked if that connection was ALSO shorted to ground.

[Perkinsman]

You bypassed my question. I do not doubt you connected the cap to the resistors correctly. But I asked if that connection was ALSO shorted to ground.

The cap was installed from the "bottom" up so there are no leads that would ground to chassis. Is that what you mean? You have more confidence in my connection skill than I do! I posted the pics to have others more experienced tell me whether it's done correctly or not since that's where the problem seems to have started. Testing with continuity shows no shorts between the positive side and ground. What have I missed?

[The Dude]

......Testing with continuity shows no shorts between the positive side and ground. What have I missed?

The positive side of the cap SHOULD be ground, so something is wrong. IMO, you also need to add some solder to your connections. They don't appear to be soldered very well.

[Perkinsman]

The positive side of the cap SHOULD be ground, so something is wrong. IMO, you also need to add some solder to your connections. They don't appear to be soldered very well.

Yes, the positive side has continuity with ground. The negative side has continuity with the 2 resistors. There are no shorts between each of those connections. I added some solder to the joints in question. I put the power tubes back in and the current started rising so I quickly tested the bias voltage at TP30 and read 175mV, 3x over the setting it had been set at yesterday, 62mV. I then started lowering the bias back to 60mV and the current stabilized. I'm starting to think that the bias pot is not working properly. Is there a way to test it?

[nevetslab]

Yes, the positive side has continuity with ground. The negative side has continuity with the 2 resistors. There are no shorts between each of those connections. I added some solder to the joints in question. I put the power tubes back in and the current started rising so I quickly tested the bias voltage at TP30 and read 175mV, 3x over the setting I had set at 62mV. I then started lowering the bias back to 60mV and the current stabilized. I'm starting to think that the bias pot is not working properly. Is there a way to test it?

YES. Remove the power tubes again, then measure the range of negative bias voltage from min to max on the bias pot. Note the bias voltage where you have it set that gave you 62mV at TP30. Using the resistor values in the bias network, I calculate the adjustment range is from -68.8V thru -55V. I'd expect you can go lower than -55V, which would increase plate current.

[Enzo]

I can see the cap has no excess wires, what I meant was, get out the ohm meter and with power off, MEASURE resistance to ground from the supply line to ground. Across C38, does it measure shorted or low resistance? Don't pull the cap, I am not expecting the cap itself to be shorted, just something in teh circuit.

[Perkinsman]

YES. Remove the power tubes again, then measure the range of negative bias voltage from min to max on the bias pot. Note the bias voltage where you have it set that gave you 62mV at TP30. Using the resistor values in the bias network, I calculate the adjustment range is from -68.8V thru -55V. I'd expect you can go lower than -55V, which would increase plate current.

Ok, I removed both power tubes, powered on and with one probe on chassis ground and the other on bias test point, TP30, adjusted the pot, R82 thru it's range. The reading didn't move much, if any. Does that indicate a bad pot. Could it be the pot's solder joints? Would that cause the initial problem of crackling too?