Hoffman Amplifiers Tube Amplifier Forum

Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: silverfox on January 31, 2016, 08:40:42 pm

Title: Testing for Oscillations
Post by: silverfox on January 31, 2016, 08:40:42 pm
Got to thinking today regarding how to track down high frequency oscillations in a tube amp. One of the problems could be the scope trace being washed out by lower frequency signals that are of a much larger amplitude. Is the floor noise going to be larger than the idle state amplitude of a high frequency oscillation?

How about the use of a filter to screen out anything under 15khz to 20khz so only the higher frequencies would appear. I know that is supposed to be the function of the sweep knob but somehow I don't think that works for tracking down oscillations in the amp. Would the filter work?

silverfox.
Title: Re: Testing for Oscillations
Post by: HotBluePlates on January 31, 2016, 09:10:41 pm
Usually people apply an audio test signal, and look for a sharp, clean trace. Any fuzziness might be an oscillation riding on the test signal.

You'd zoom way in on your test signal by adjusting the time base to check if fuzziness is just the scope not making a sharp trace, or if you see a low-level high frequency signal riding on the lower frequency test signal.
Title: Re: Testing for Oscillations
Post by: silverfox on January 31, 2016, 09:33:28 pm
Ah. Now I know what to look for in the future. I've seen that kind of signal surfing before but not on any of the amps I've built so far.

Thanks,

silverfox.
Title: Re: Testing for Oscillations
Post by: jjasilli on February 01, 2016, 08:14:11 am
The recommended test can show the presence of oscillations.  I'm not sure it will help to specifically identify and cure them.  There's lot's of info if you Google: "how to test for oscillation in an amp".


The existence of hi-frequency, inaudible, oscillations in a guitar amp may not matter as a practical concern.  They may slightly alter waveform so long as they don't ruin tone.  The extra energy of the oscillations is easily absorbed by a guitar speaker.  But in hi-fi, 2 problems:  distortion of waveform is not OK; and the extra energy can burnout tweeters which are far more delicate than instrument speakers.
Title: Re: Testing for Oscillations
Post by: shooter on February 01, 2016, 10:41:59 am
Quote
You'd zoom way in
I've found mine just this way, 1 needed no input signal at all, it was just lurking in the *grass*.

The hard part is when they lurk on the ground plane and you use that as scope reference, AC coupling helps, but you might have to *float the scope*, which WILL cause catastrophe if you're not on your game.