diyAudio (
-   Tubes / Valves (
-   -   Fixing Sneaky Oscillation in P-P Amplifiers (

wrenchone 4th January 2013 11:09 PM

Fixing Sneaky Oscillation in P-P Amplifiers
I have a P-P amp I've been tinkering with that has a snarky burst of oscillation that occurs on only one half-cycle. I recall SY saying in one thread that this could be due to mismatch between transformer primary halves. The problem and its solution are deeply buried in a thread here somewhere, and the problem is frustrating enough that the problem and solution deserve a thread of their own for future frustrated amp builders.
SY, can you restate your suggestion to fix this problem? I think that a series RC across each primary half, or one between plates might be the solution.

astouffer 4th January 2013 11:15 PM

If the primary resistance is off you could add a series resistor to try balancing it.

HollowState 4th January 2013 11:48 PM

I remember something about this. At least one suggestion was to add grid stopper resistors to the output tubes if not already there. Values around 1K to 1.5K are common. If pentode operated, I'd also consider smaller ones on the screens and perhaps even a small inductor at the plate terminal in series with the OPT.

Wavebourn 4th January 2013 11:56 PM

What's the frequency of oscillations?

wrenchone 5th January 2013 12:06 AM

A few MHz...

wrenchone 5th January 2013 03:45 AM

One of the things I might try before anything else is to increase the output tube grid stopper values to a value similar to that of the Red Light District, with the intent of limiting grid current during positive grid excursions. This may sufficiently squelch the output tubes such that no further action is necessary. The grid stoppers are probably a k or two right now, maybe less. I'd have to look at the half-built carcass to tell, and it's down in the basement, where it'll stay until I get to it...

astouffer 5th January 2013 04:56 AM

What does the burst look like?

Wavebourn 5th January 2013 05:44 AM

If few MHz it is not transformer ringing. Either grid stoppers would help, or a cap directly from anode to the feedback input, to bypass feedback through transformer. Can be also result of some parasitic feedback, so shielding may help.
If you remove GNFB, does it stop or gets worse?

20to20 5th January 2013 12:45 PM

What are the lowest and highest test frequencies that it occurs on? Where is the first point in the circuit you can detect it? Any unusual circuit design elements, components? Apart from being a P-P amp, we don't have much to go on.

EDIT: Does it exist at all output power levels? Does it start at the crossover point, the positive or negative half-cycle, top ot bottom of the wave?

LinuksGuru 5th January 2013 02:33 PM

Can you please post a schematic and photo of layout?

All times are GMT. The time now is 08:37 PM.

vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2017 DragonByte Technologies Ltd.
Copyright 1999-2017 diyAudio

Content Relevant URLs by vBSEO 3.3.2