Stubborn Oscillation

BRSHiFi · 2012-03-14 3:49 pm

I am almost done with my KT88 push push parallel amplifier. There is a stubborn stability issue I am having with it that only occurs when a speaker load is connected. An 8 ohm dummy load presents no oscillation. One speaker I tried it oscillated at 312k, and at present it is about 125k. Sometimes the frequency and amplitude can vary but it is relatively consistent. Another symptom is a squeal that lasts several seconds as the amp is warming up, while the oscillation is climbing up to maximum amplitude.

I added an RC network after the upfront stage with a 10k POT in an effort to stop the resonance. No cigar after trying a few caps and sweeping the pot.

One person suggested placing an RC network across the OPT primary. I tried that with a 270 ohm and .01uF capacitor, as I do not have picofarad capacitance high voltage capacitor stock yet. No cigar with that.

Another person said to try placing a small cap between the phase splitter plates. I have not yet tried this as I am waiting for my high voltage caps in the mail.

I also tried a few resistors and caps across the output. That did nothing either.

Am I on the right track, or is there something else I should be looking at? The schematics are attached.

Osvaldo de Banfield · 2012-03-14 4:01 pm

As a first try, remove 39pf at the feedback resistor loop. If the oscillation stops, then, the amplifier is not unity gain stable, so you must stabilize it after closing the loop. If it doesn´t stop, the chech abnormal loops like trough power lines, capacitive coupling between in and outs´etc.

Best regards. keep us posted.

BRSHiFi · 2012-03-14 4:30 pm

The cap has been removed. There is still an oscillation.

Osvaldo de Banfield · 2012-03-14 4:55 pm

Re-check the amplifier whit only a single push pull, without the parallel, then only a tube in each half primary. May be some interation between tubes.

BRSHiFi · 2012-03-14 5:55 pm

Nope, half the tubes out still produces the oscillation.

With the scope I checked:

- The output of the SRPP upfront stage.

- The speaker output.

The oscillation is affecting all stages of the amp, not just one.

TheGimp · 2012-03-14 6:30 pm

First I'd break the feedback loop until you eliminate the oscillation.

Try changing your zobel from a single pri-pri to two each pri-screen.

BRSHiFi · 2012-03-14 6:40 pm

The oscillation goes away when the FB loop is removed.

Osvaldo de Banfield · 2012-03-14 7:11 pm

reverse the output transformer leads. The can be wrong wired.

Wavebourn · 2012-03-14 7:31 pm

Play with R12 and C31.

One more option is to add a loop across output tubes, like 240K 2W resistors from anodes of output tubes to anodes of driver tubes.

DF96 · 2012-03-14 7:47 pm

Temporarily remove the 39pF across the 10k feedback resistor, then try increasing the 10k until the amp becomes stable. This will give you an idea of how far off stability you are.

One option to try is splitting the 10k into two resistors, with a small cap across one of them.

Wavebourn · 2012-03-14 10:32 pm

DF96 said:
Temporarily remove the 39pF across the 10k feedback resistor, then try increasing the 10k until the amp becomes stable. This will give you an idea of how far off stability you are.

One option to try is splitting the 10k into two resistors, with a small cap across one of them.

I never do such things, with caps in feedback. They are good for oscilloscope pictures, no more. Otherwise it is harmful, requiring more of compensation for unity gain stability that is not needed. Also this cap feeds signals from cellphones to the amp's input.

BRSHiFi · 2012-03-14 10:52 pm

So you are saying leave out the 39pF cap all together?

gingertube · 2012-03-14 11:05 pm

Suggestions,
Make sure the 10R Anode stoppers are right on the tube socket pins.
Up the screen stoppers to 1K. Ultralinear Mode REQUIRES larger values of screen stoppers.
ASIDE: I had a 4 x EL34 HiFi Amp with 150R screen stoppers. It was fine in triode ande Pentode Mode but oscillated like crazy in Ultralinear Mode - increasing the screen stops from 150R to 1K fixed the oscillation in that amp.
Up the grid stoppers to at least 4K7 or even 10K

Cheers,
Ian

BRSHiFi · 2012-03-14 11:10 pm

Working on it right now.

Installing 1k screen stoppers and 10k grid stoppers.

Wavebourn · 2012-03-14 11:23 pm

kingneb said:
So you are saying leave out the 39pF cap all together?

Yes. Forget about this nonsense.

Aand go this way:

Wavebourn said:
Play with R12 and C31.

One more option is to add a loop across output tubes, like 240K 2W resistors from anodes of output tubes to anodes of driver tubes.

roline · 2012-03-15 12:42 am

The LTP second grid is tied to ground. My circuits has it biased to the other grid with a 1M and a .1u to ground. That ensures proper bias and LTP function.
Is it stable in triode mode?

BRSHiFi · 2012-03-15 1:08 am

Upped grid stoppers do not help.

As far as R12 and C31 go I have a 470pF cap in there right now and a 8.2k ohm resistor.

I tried a 10k ohm pot in there and swept it. The oscillation persisted. I will try a few more caps in there.

Salas · 2012-03-15 1:16 am

What about RC grid to anode on the 12BH7s? Say 27p + 100k.

BRSHiFi · 2012-03-15 1:18 am

I will try and see.

Salas · 2012-03-15 1:19 am

Also U56 can be the culprit? Stick the probe at the cathodes node it sinks and see if there is some oscillation there?

Stubborn Oscillation

Previously known as kingden

Attachments

Member

Previously known as kingden

Member

Previously known as kingden

Member

Previously known as kingden

Member

Member

R.I.P.

Member

Previously known as kingden

Member

Previously known as kingden

Member

Member

Previously known as kingden

diyAudio Chief Moderator

Previously known as kingden

diyAudio Chief Moderator