Ampeg SVT 3 pro question

[Swampy]

Hi everyone,

Can someone outline the functions of the components R25 / D 8 / R24 /Q5 / C9 / D28 in the circuit branch tied to the source of power mosfet Q7 only?
Obviously the circuit is paralled on the - side of the power amp...is this a feedback network of sorts for stability?

SVT PRO 3.pdf

Cheers,
Matt

[Jazz P Bass]

Looks like a protect circuit to me.

[Enzo]

It is a limiter. A pretty common circuit really, found in many solid state power amps both MOSFET like this and in bipolars.

the output transistor current flows through R32, causing a voltage drop across the resistor. The more current, the greater the voltage across the resistor. If that voltage gets high enough, current will start to flow through R25 into the base of Q5, turning Q5 on. WHen Q5 conducts, it tends to shunt the signal going to the gates of the output MOSFET, thus preventing the output current from growing any larger. D28 is ther to protect Q5 from reverse voltages when the signal at the gates drops negative - which is half the time unless you are amplifying weird stuff.

Some amps sample each output transistor, this one samples just the one. I suppose they figure the outputs all work in unison, the gates have a common drive, so whatever the one does, they all are doing, so they sampled the one to limit the gate drive, and that will limit the whole row of them..

[Swampy]

Thanks for your feedback gents.

R25 and Q5 are totally cooked...the unexpected flame took place on a loud note. Once I've replaced the bad parts I'll fire it up on the lightbulb limiter and check voltages, does anyone want to wager a bet on whether the cooked components are cause or effect?

Matt

[Enzo]

Well, figure it out.

Let's say Q5 shorted completely. That would put R25 in parallel with R32. I don't think that 330 ohms parallel to 0.47 ohms is going to do anything to the output transistor. And the fraction of current through the R25 branch would be miniscule. And that would essentially put the diode D28 across Q7s gate, which I would imagine pretty much pinch off any drive. SO I doubt Q5 hurt anything by failing. And R25? All it can do is go open, and that won;t hurt anything either.


SO I'd have to say those are effect, not cause. You didn't say what else might have been bad, so we cannot reconstruct the whole story.

Looking the other way, to burn up those parts, current had to flow through them. The base of Q5 can't handle a lot, so once current ramps up through it, it will short, and that current had to come from somewhere. About the only paths we have are through Q7 or down through D28. What if Q7 shorted? Quite likely takes out R32. And when that happens you have V+ sitting there at the source leg of Q7 looking for a way home. That is 65 volts with a lot of current behind it. Without R32, the path to common is through R25 and Q5. Oh, it might look at D8, but 2700 ohms of R24 would probably make that less attractive than burning out Q5 to the failure current. SO were Q7 and R32 bad? If you had to replace Q7, you sure better check R32.

Alternates? D2 and Q3? I don;t like them,but maybe. Was Q3 bad? If D2 shorted, that would turn Q3 off. If R2 opened, then Q3 would turn on a lot, but R14 should limit the current that coulod get to Q5, plus D4 and D5 ought to help protect it.

[Swampy]

Right, it all makes sense now. Writing things down is a sure is a valuable scientific process huh.

Initially the power section shorted due to overbias (causing a short between the rails). During my troubleshooting process I lifted the gate resistors R28 - 31 and fired up via the lightbulb limiter. After a few seconds R25 smoked a little, so I powered down.

I replaced all 8 power mosfets and source resistors but forgot about R25...then fired up and tested with the above results.
A pack of 330 ohm resistors came in the mail but I couldn't recall what they were for...until I reviewed my notes.

Oh well, live and learn huh. Thanks for your time and knowledge sharing.