RCA BA-6A problem - SOLVED!

[audiopete]

Hi everyone. I have an original RCA BA-6A here for repair. I've been through every square inch and can't figure out the problem. When signal (100Hz 700mV sine) exceeds threshold, the 6SK7s shut right down from the control voltage.  Current monitoring on the cathodes confirms this.  All voltages are pretty much there, with the 6J7 stage running hot on the meter, but can't see how that would cause this issue. Tubes were changed prior to me getting it with no luck. The bias point on the 6H6 is correct, but today I want to monitor to see if that shifts with signal indicating leaking caps from the 6V6 anodes. Other than the 6SK7s themselves (I am getting in a pair or two of matched from my supplier next week), I am at a loss here.  I've had the kind and generous help of member Dave P, and also I've been consulting with another excellent tech friend of mine, but no luck yet.

Here's a link to a vid showing the problem on the meter at the studio. The buzzing is a synth bass signal fed into the BA-6A.

https://www.dropbox.com/s/wvceix5tcmbwtnm/Video%202018-01-05%2C%205%2035%2051%20PM.mov?dl=0

Thanks for your time.

 

[emrr]

Definitely try another 6H6, I've seen many bad side chain rectifiers in similar comps.  They always test as 'good', but don't work in this position. 

 

[audiopete]

I did try another I had in my stash that tests good on my B&K 747.

 

[scott2000]

I did try another I had in my stash that tests good on my B&K 747.

What are the chances that one may be another that tests good but doesn't work??

 

[audiopete]

There's always that chance with dealing with these old tubes. I am not ruling anything out. Chances are it is fine since it's just a dual diode and I am getting the neg control voltage at my 6SK7s, plus that would be 3 6H6 now that have gone into it with no change. For whatever reason, it's putting the 6SK7s right into cutoff as soon as threshold is crossed. The 6SK7s are my prime suspect at the moment and i'm getting a pair or two next week to try. I researched that the closer plate and screens are, the more neg voltage on the grid to shut it down, but I've tweaked voltages on screen and lowered the entire supply to a 115VAC input with no effect. Either the pair of 6SK7 are no good and can't take the specified control voltage. or the control voltage is coming in too low and shutting the tubes right down. I'm seeing -5ish volts on my Fluke 87V at the grids but I have to look into the manual to see how to capture max reading and see what is firing at the 6SK7s.

 

[emrr]

No behavioral difference in single versus double?

 

[Monte McGuire]

Make sure that the voltage regulator tube is operating properly (OD3) It sets the 'threshold' voltage at the GR diodes. I mention it only because it wasn't mentioned earlier, and it's important to the GR process.

 

[Monte McGuire]

I just added the complete RCA BA-6A manual in two parts to the Documents section. There is more troubleshooting information in the manual, as well as some good discussion of the theory of operation. Hope it gets the old beast going again!

https://groupdiy.com/index.php?action=dlattach;topic=44740.0;attach=61232

https://groupdiy.com/index.php?action=dlattach;topic=44740.0;attach=61233

 

[audiopete]

Thanks everyone. I will have to do some deeper monitoring to see if there is some anomaly during operation. Static voltages are withing spec, but something must be shifting when threshold is crossed. There is about 67V at the 6H6 cathodes.

No change with single vs double beyond the normal change of time constants.

One interesting observation. If I "flick or tap"  the 6SK7 plate resistor leads with my voltmeter probe, it triggers the problem and shuts down the 6SK7. This seems like a momentary voltage spike but again, why would that be just shutting them down? The only thing I can think of is the tubes themselves.

 

[audiopete]

I have a hard copy of the manual thanks. I've been through it forward and backwards

 

[noulou]

Thanks everyone. I will have to do some deeper monitoring to see if there is some anomaly during operation. Static voltages are withing spec, but something must be shifting when threshold is crossed. There is about 67V at the 6H6 cathodes.

No change with single vs double beyond the normal change of time constants.

One interesting observation. If I "flick or tap"  the 6SK7 plate resistor leads with my voltmeter probe, it triggers the problem and shuts down the 6SK7. This seems like a momentary voltage spike but again, why would that be just shutting them down? The only thing I can think of is the tubes themselves.

dirty tube sockets perhaps?

 

[audiopete]

All sockets were replaced just before I got it. The previous tech was hoping that would fix it.

 

[audiopete]

OK more observations:

Swapped out C14 C15 no change.

Took MIN reading of control voltage with Fluke 87V and I'm seeing almost a -49V spike. Surely this is much too much, and checking the datasheet would confirm this.

What could be causing this large negative voltage? The bias voltage on cathodes of 6H6 are almost perfect 67V.

Also, would the 6J7 stage running too hot cause this?  It shows high out of range on the meter - both tubes.

 

[audiopete]

Another observation is the threshold voltage rises to 82V max when this anomaly triggers. Is this voltage supposed to fluctuate or stay stable?

 

[emrr]

That threshold voltage should be stable. 

 

[audiopete]

Hmm so why would it shift?  It shifts different on each cathode. One side rises more than the other.

 

[CJ]

try 500 hz or 1 K C, does it do the same thing?

 

[noulou]

Long time ago I had trouble with a ba6a. It was impossible to match the 6sk7s in order for it to go into gain-reduction without huge thumps.
It turned out to be the interstage transformer.

 

[SPbStan]

Does there seem to be normal balancing of the 6SK7's  with  no signal?  Or with the Bal-Lim switch in "A" position?  If those do not balance, nothing else will behave as expected.

 

[emrr]

Long time ago I had trouble with a ba6a. It was impossible to match the 6sk7s in order for it to go into gain-reduction without huge thumps.
It turned out to be the interstage transformer.

What about it?  I'm having trouble imagining a transformer problem that wouldn't also present as an obvious response problem.