RCA BA-6A distortion during compression

[mrtnasty]

Hi I recently acquired an RCA BA-6A that has had all electrolytic and signal caps replaced. It sounds really good if I don't drive it too hard with an external preamp, but as soon as the compression kicks in, it starts distorting. I tried balancing the A and B pots as suggested by the manual, but that did not fix the problem. Any suggestions on where I should try trouble shooting would be greatly appreciated. Thanks!

 

[mrtnasty]

Try different 6SK7’s. Try a different 6H6.

Thanks for the advice!

 

[mrtnasty]

Im happy to report that the output of the BA-6A was just very large even on the lowest output setting and clipping my audio interface. With an attenuator put in-between the output and the audio interface, the BA-6A sounds great!

 

[thomasdf]

Maybe a better impedance matching too ?

 

[mrtnasty]

Maybe a better impedance matching too ?

Perhaps, I used a Davin 600 ohm T attenuator, but the BA-6A is 600 ohm out, so that shouldn't change the impedance, right?

 

[emrr]

T after a T, both at zero, there's no load at all. Needs a 680-1K load resistor across the output, that'll knock gain down right away, without the second T.

 

[mrtnasty]

T after a T, both at zero, there's no load at all. Needs a 680-1K load resistor across the output, that'll knock gain down right away, without the second T.

Hmm sorry I don't understand

 

[emrr]

Do you have a 600 ohm load resistor on the output? If not, you should. Otherwise the output attenuator does not attenuate as much as it should, and set wide open, there's no load at all, like turning on a guitar amp with no speaker connected. Push-pull 6V6 10W power amp here.

 

[mrtnasty]

Do you have a 600 ohm load resistor on the output? If not, you should. Otherwise the output attenuator does not attenuate as much as it should, and set wide open, there's no load at all, like turning on a guitar amp with no speaker connected. Push-pull 6V6 10W power amp here.

Doesn't a T attenuator always have 600 ohms on the input and output?

 

[scott2000]

Doesn't a T attenuator always have 600 ohms on the input and output?

if it has 600 ohms on it's input and output irrespective of the attenuator itself
At least that's how I understand it...
but I'd be just as happy just following the advice given and not understanding...lol

 

[inkster]

Check the interstage xfrmr, I have had to replace those or have them rebuilt.

 

[emrr]

A real 'T' does not present 600 ohms to anything, it is meant to be BETWEEN 600 ohm devices without loading either inappropriately. If the thing after the 'T' is 10K, then you need a load resistor for the 'T' to work correctly. A 1K is probably close enough, anything between 680-1K. MY POINT IS DITCH THE SECOND T AND TRY A LOAD RESISTOR WITH THE BUILT IN T, see how that is first.

Go find my many posts about the loading and behavior of 'T's, tons of them. Somewhere I address the 'not really a T modern T's, they are functionally similar but always introduce some loading loss because they don't get out of the way like a real T that's all the way up. I think that's in a Collins 356A preamp racking thread.....or a Gates.....

A ladder always presents a 600 load to the source.

 

[mrtnasty]

Check the interstage xfrmr, I have had to replace those or have them rebuilt.

Thanks for the heads up, ill keep that in mind!

 

[mrtnasty]

A real 'T' does not present 600 ohms to anything, it is meant to be BETWEEN 600 ohm devices without loading either inappropriately. If the thing after the 'T' is 10K, then you need a load resistor for the 'T' to work correctly. A 1K is probably close enough, anything between 680-1K. MY POINT IS DITCH THE SECOND T AND TRY A LOAD RESISTOR WITH THE BUILT IN T, see how that is first.

Go find my many posts about the loading and behavior of 'T's, tons of them. Somewhere I address the 'not really a T modern T's, they are functionally similar but always introduce some loading loss because they don't get out of the way like a real T that's all the way up. I think that's in a Collins 356A preamp racking thread.....or a Gates.....

A ladder always presents a 600 load to the source.

Are you basically saying turn the built-in T attenuator into a bridged T attenuator using a load resistor?

 

[emrr]

No

 

[JMan]

Is this just a matter of terminating with a 680-1k resistor across pins 2 & 3 of the output xlr?

 

[mrtnasty]

No

Sorry I am just trying to understand what you are saying, I don't want to do something I don't understand. I couldn't find the Collins 356A racking thread you were talking about. On a separated but probably related note, I looked up the line input impedance of my audio interface (Clarett+ 4pre) and it says 66k ohms.

 

[inkster]

You want to align the circuit with a DMM instead of the VU meter(which I have serviced too, they age and drift too) and null the voltages at the correct locations in circuit...this should only be done by experienced techs using proper safety precautions of course and after 30 mins warm up.
Check the tolerance of all the parts in circuit, some can be done without removal(lifting one leg)...I recommend watching a Mr Carlson youtube video if you're not familiar with the technique.
Interstage needs to to tested at least partially out of circuit
This applies to all vari mu limiters.
Cheers

 

[mrtnasty]

Do you have a 600 ohm load resistor on the output? If not, you should. Otherwise the output attenuator does not attenuate as much as it should, and set wide open, there's no load at all, like turning on a guitar amp with no speaker connected. Push-pull 6V6 10W power amp here.

Thank you very much for this info, I installed a 620 ohm resistor between xlr pins 2 and 3 on the output, and it sounds good and reduces output so that a second t pad is unnecessary. It took me a while to wrap my head around the reason why this was the proper thing to do, but was convinced thanks to ccaudle and others in this thread:

https://groupdiy.com/threads/impeda...-tube-gear-and-modern-audio-interfaces.85606/
What I learned in short, it is not correct to let tube gear that is expecting a 600 ohm load drive modern equipment that has a high input impedance that is designed to be impedance bridged. In addition to it not being safe for the tube gear (as emrr pointed out), if a 600 ohm (or similar) resistor on the output, the tube gear would always think it's driving a 600 (or similar) ohm load, and the modern gear it is feeding would also be happy because its input impedance is way higher than 600 ohms.

I still don't quite understand what to do in the opposite situation, where a piece of modern equipment with low output impedance (~60 ohms) is feeding tube gear with slightly higher (200 - 600 ohms), but still low input impedance. If the tube gear is expecting to have its impedance matched (because it was designed in the 40's or 50's), should I transform the modern output up to the slightly higher impedance, or just do nothing and live with a bridged input that does not quite obey the golden rule of impedance bridging, that the source is 10x lower impedance that what it is feeding?

Thanks!