• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Oscillation in tube amps

Can oscillation sound something like cicadas but slightly more empty glass bottle like in the distance or would that more likely be a blown tweeter? Have been hearing this from my right channel on and off and seams like its more pronounced when my amp (pp) has been on for an hour or two. Guess i need to feed some tones and watch whats happening on the scope. I guess a faster check would be to simply swap right and left.. ugh.. guess its time to get the se 300b off the bench, trying to track an imbalance problem with that one..
Wish i was as capable as many here, i enjoy the learning but now feel pressure to resolve asap... and, im also wondering if a capacitor in my xover could be an issue..

Any advice on a methodical search path would be greatly appreciated!
 
Yes, swap the L and R speaker leads at the amplifier L and R outputs, to see if there is anything related to a speaker problem.

If the problem is still in the same amplifier channel, then swap the L and R signal source outputs to the amplifier inputs to be sure the signal source is not the problem.

By then, you will be well on your way to finding the source of the problem . . .
Signal source,
Amplifier,
or Speaker.
 
Hi guys,
Usually I see that, in order to stabilise an amplifier, negative feedback is used. However, would it make sense to use a Miller cap as in Analog design for integrated circuits? This way we could achieve pole splitting and the system would be a single pole system => no oscillations.
Any thoughts?
Cheers,
Pedro
 
Design an op amp that only has one high frequency pole (I am not talking about some that have a purposeful dominant pole that is formed by a purposeful capacitance).
Then you might be hired by Analog devices.

Vacuum tubes and solid state devices all have capacitance and inductance.

Just my opinions
 
Design an op amp that only has one high frequency pole (I am not talking about some that have a purposeful dominant pole that is formed by a purposeful capacitance).
Then you might be hired by Analog devices.

Vacuum tubes and solid state devices all have capacitance and inductance.

Just my opinions
What I mean is to actually add a cap between the grid and the anode, so one pole goes to very low frequencies and the other to very high frequency. If the high frequency pole only appears after the gain of the circuit is below 1dB, then this would be a single pole system.
 
Design an op amp that only has one high frequency pole (I am not talking about some that have a purposeful dominant pole that is formed by a purposeful capacitance).
Then you might be hired by Analog devices.
If you could do that, you’d be able to make a killing licensing the patent. Each gain stage has two capacitors in it, and it’s really hard to just get rid of poles entirely.
 
If you could do that, you’d be able to make a killing licensing the patent. Each gain stage has two capacitors in it, and it’s really hard to just get rid of poles entirely.
At least, in simulation, it works. But I will try it in one of my pcbs.
I know how to stabilise multi-stage (3 - 4 stages) amplifiers, but in CMOS (analog design) not with valves. In this case, the same concepts can or cannot apply. However, I will give it a try.
Cheers,
Pedro
 
EVERYTHING works in simulation. Realizing an op amp (or even a single gain stage) with one and only one pole is physically impossible. But it works in simulation.
The idea is not to design an amp with one pole, it is to have a dominant pole and the rest of the poles at very high frequency (100s of KHz).
This way, the amplifier will behave like a single pole system.
Cheers,
Pedro
 
I really don't understand this argument. Classic opamp design comes from a single dominant pole model of a simple integrator fed by a transconductance stage and followed by a wideband buffer, both assumed to be negligibly wider bandwidth than the integrator. How is this not a single pole response in any way that matters?

Of course a real vacuum valve audio amplifier cannot be this primitive without very poor practical performance, but that's not what has been suggested.

All good fortune,
Chris
 
  • Like
Reactions: 1 user
Nothing to argue about.

But . . . remember when you were a Newbie in electronics; or pretend that at least you once were a Newbie . . .
In that case, as a Newbie, you might be confused and you might be mislead by Post # 268.
Help the newbies understand that there are multiple poles, versus only a dominant pole, and no other poles.
Just Saying
 
Fair enough. When I argue for a single dominant low frequency pole, folk will usually counter that some split-pole arrangement is better, and it might well be. My argument is that an iron-core OPT causes enough signal level dependent variations that it cannot be trusted, in the general, nobody-has-actually-measured=the-transformer-adequately, sense. So, a brute force solution applies.

In that context, a single dominant high frequency pole is also correct (if not the very best possible, like some split-pole modern design) but will work fine. My disagreement is with post #274, which is of course true in the widest band abstract sense, but not necessarily useful for an audio amp. But I'd be very interested in the details of the argument.

All good fortune,
Chris
 
  • Like
Reactions: 1 user
In college, I did a project on amp design that turned into a project about stray capacitance. I designed a nice little BJT amplifier and constructed it as a “ball of parts” and it worked great. When I created a PCB for the design and stuffed all the parts, it oscillated like hell.. thus the change in project focus. I was able to get it all stable. But had to do it by adding intentional capacitance and throw away bandwidth..