Troubleshooting channel dropout

[Jim W]

I'm working on an ADC B100 preamp and I'm having some problems troubleshooting an intermittent channel. When the preamp works, it sounds great. But the right channel is occasionally dropping out. If I put a scope probe anywhere in the signal path past the volume control, the channel pops right back, which makes it pretty tough to troubleshoot. The problem is in the high level section. I have swapped tubes left to right and it stays in the right channel. Doesn't seem to be a connector or switch problem either.

It's also possible to pop the channel back in by cranking the volume momentarily, and with the volume at 80% it's at the threshold and passes only the peaks - sounds awful.

Any advice from someone that's seen something like this before would be appreciated.
Thanks,
Jim

[HollowState]

Servicing intermittent problems is always a bear. Freeze mist, heat guns and variacs are some of the standard tools. Probing and poking are also helpful. Somehow, I don't envy you.

[kavermei]

Possibly the volume pot, a corroded tube socket, or a bad solder joint.

These things are a real pain to find

[Arnulf]

Originally Posted by Jim W

If I put a scope probe anywhere in the signal path past the volume control, the channel pops right back, which makes it pretty tough to troubleshoot. The problem is in the high level section. I have swapped tubes left to right and it stays in the right channel. Doesn't seem to be a connector or switch problem either.

It's also possible to pop the channel back in by cranking the volume momentarily, and with the volume at 80% it's at the threshold and passes only the peaks - sounds awful.

Bias issue ? Does your potentiometer double as grid resistor ? Schematic please.

When you said "anywhere in the signal path", did that refer to anywhere between the wiper of potentiomer and output terminal of the tube or did that refer to the section up to and including grid terminal ?

Either way, resolder all resistor joints.

[m6tt]

The touchiness might indicate ultrasonic oscillation. Try adjusting lead dress while repairing the cold solders, and try with less or no global feedback if present (temporarily). If you can get it to drop out by tapping/banging on it, it's probably mechanical, whereas is you can't, it may not be. Check all operating points, i.e grid v. cathode voltages during the failure mode. Good luck, if all else fails take a break!

[Jim W]

Thanks for all the suggestions:
Pot - no, still have signal at the wipers. Tube socket - probably not, wiggled and replaced tubes, no effect. Bias - probably not but I'll chase that. Didn't see anything at the output when it's dropped out, so an oscillation would have to be higher than my 10Mhz scope would see. I'm leaning toward a bad connection, open trace, bad solder joint. I got the channel back yesterday by pulling on the front panel. Of course, that sent a shock through the tonearm on top of the rack and it might have been the big low freq signal that popped it back in too. I have checked most of the voltages but I really need to get organized and make a table for both channels.

B100

I put up a quick and dirty website with a link to a schematic. I was going to put up a real website for the B100 anyway since it's a shame these things have faded into near total obscurity.

Jim

[kmtang]

What kind of tubes used in the B100 preamp??


Johnny

[Jim W]

The tubes are all 6DJ8.
Jim

[kmtang]

Hi Jim,

Thanks a lot for your info. I have some 6DJ8 and 7DJ8's in hand. May be it is fun to experiment this cascode amplifier circuit.


Johnny

[Jim W]

Just spent too much of my weekend chasing this gremlin. Making a long story short (freeze spray, heat, beating on things, resoldering, etc.), it appears that the problem was a socketed delayed turnon relay. I won't bet a lot on this, because the preamp would run for hours at a time without a problem so it might come back. If this really is the case, then apparently creating a transient by using a scope probe to look at the signal pretty much anywhere in the preamp would fix the open for a while.

The good news is that I managed to re-socket two other B100's while waiting for the gremlin to pop up. It takes a while to replace fourteen sockets. Thanks for all the help and suggestions.
Jim