Deacy amp question...

[digi2t]

Hi all,

I've had the parts kicking around for a while, and this morning I broke out the breadboard to test. I'm using this schematic;



AC125 in Q1, AC126 in Q2, and a pair of matched AC128K's in Q3/Q4. LT44/LT700 transformers. Sooooo, here's the kicker...

I build up the circuit on the board, and right off the bat, I've got a nasty oscillation. Double check the connection points, everything checks good. Double check values, nope, all good. Recheck the transistors, they're giving me decent numbers on my Peak meter as well. The only thing that kills the oscillation is pulling the 100uF cap below Q2. OK... at this point, I lean back, and take a deep breath. What the fu....?

Head back to the net, and I start checking similar schematics. I then come across this schematic of the Pignose;



Not quite exactly the same, but what strikes me is the way the first transformer's primary connections are shown. In the first schematic, the negative line comes straight down to the transformer, and the other lines goes down to the transistor.

In the Pignose schematic, the leads are shown swapped. And guess what... switching the leads on my breadboard layout to match the Pignose, killed the oscillation. Dead.

So, what's going on here? Is the polarity of the primary that important vis-à-vis the secondary? If so, would it be sensible to note this on the schematic? It really isn't an obvious problem, yet makes all the difference between it working, and not.

[teemuk]

Note the feedback loop from output towards earlier stages: Pignose has it on speaker output, (presumed) Deacy at collector of one of the output transistors.

Transformers, as you noticed, have a polarity. The same polarity also determines signal's PHASE.

So, when you swap polarity either in primary or secondary the signal directed to feedback loop will either be seen as positive feedback, which often triggers oscillation, or (in another phase) as negative feedback. Correct phase is important.

Yes, polarity is therefore important as well and yes, sometimes schematics do have little dots indicating transformer polarities. Sometimes they don't. There's no universal rule of what details to include in schematic. In either case, proper polarity is usually very easy to deduct.

It really isn't an obvious problem, yet makes all the difference between it working, and not.

Actually, it is a very obvious problem. Usually when transformer coupled amplifiers oscillate the first advise is trying if reversing transformer polarity corrects the issue. This issue is so common that it is obvious to pretty much everyone who is familiar with transformer coupled designs that include a feedback loop.

The only thing that kills the oscillation is pulling the 100uF cap below Q2

Pulling off the emitter bypass cap reduces gain of the stage, subsequently reducing open loop gain of the entire power amp within the feedback loop. The gain reduction is somewhat counteracting gain increase from positive feedback and may in some cases be enough to prevent oscillation. Needless to say, it's not the proper cure for the problem.

[Kipper4]

I made one and bought the transformers off ebay cheap.
Had the same loud osillation turns out i had the transformer legs wrong. i crossed them over and boom its alive and kicking

[digi2t]

It really isn't an obvious problem, yet makes all the difference between it working, and not.

Actually, it is a very obvious problem. Usually when transformer coupled amplifiers oscillate the first advise is trying if reversing transformer polarity corrects the issue. This issue is so common that it is obvious to pretty much everyone who is familiar with transformer coupled designs that include a feedback loop.

OK, I should have been more precise. I'll rephrase, and state "It really isn't an obvious problem to a noob like me, yet makes all the difference between it working, and not."

Thanks for the science lesson teemuk. I appreciate the illumination. 10-4 on pulling the cap, I didn't think it was a solution either.

@Kipper4; That's exactly what I've faced. It works great now. Just need to find a cool speaker for it now. Think I might try an old 6x9 car speaker. It's 4 ohm, and paper cone, with a whizzer. Right now I have it driving a 12" speaker out of an old Farfisa, but it's 8 ohms.

[Kipper4]

Your welcome Deano

[mac]

Had the same loud osillation turns out i had the transformer legs wrong. i crossed them over and boom its alive and kicking

I had the same problem first time I fired it up. I suspected it was a phasing issue, changed wires and no more oscillations.

mac

[Kipper4]

A friend who was doing a simultanious build made the suggestion after i sent him an mp3.
This was my first build almost exactly a year to the day.
I was greatful of the help and its just nice to be able to give some back. Merry christmas guys.