Dreaded SVT3 Output Mistake....

[Leo_Gnardo]

The output transistors (IRFP....) are MOSFET transistors and not JFETs. When I said "replace all transistors", we were talking about output transistors and I meant exactly output transistors (power MOSFETs). But... any of the other transistors could have failed as well. It's your responsibility which of the other transistors require to be replaced. Mark

Add to that, you must replace any resistors, diodes, thermistors etc that may have perished. The trick in solid state amp repair is, "everything is important." A defect in any part may result in your amp not working, in fact may cause the amp to self destruct again.

No wonder I try to avoid repairing transistor amps.

Looks like you have a complete inventory of transistors ready to order Mr_b, check the rest of the parts on that board in case there's anything else. And I'll be doing the same here.

[Mr_bibbles]

Tight! Thanks doods.

[Mr_bibbles]

OK So i've checked all the diodes and the transistors in the power board and removed shorted offenders. However, with just the power board plugged in, the current limiter is still brightly lit. There are only three transistors left on the board that all pass the "double diode check" so my question is... what could cause this? The shorted diodes been removed from circuit until the new ones arrive.


1. Would I have to replace all of the components in this amp to see if the current would stop being drawn? Wouldn't removing them from the circuit effectively stop the shorts?

2. Since damage was caused by a transistor shorting to heatsink, we can rule out the big caps and the transformer, yes?

3. We can also rule out rectifier, since I powered it unlatched from the board and the bulb stayed dim... yes?

4. Most of the .47 OT resistors are shorted so I'm going to replace them all, but that wouldn't cause infinite current draw right?

Thanks everyone, I am over my head with this one. SS stuff is not very simple.

[Enzo]

I seriously doubt the 0.47 ohm resistors are shorted. They are only half an ohm to start with, your meter is likely having trouble measuring that low, your probes might have that much resistance. Those resistors go bad by going open, not shorted. I am unsure how they could physically short anyway.

I hope you realize the output transistors are MOSFETs and not bipolar transistors. ANy that measure shorted source to drain need to come out, and yes you can remove them all for testing the board.

Shorting power rail to ground COULD have damaged the Q3 or Q4 circuits.

[The Dude]

I'm not sure which diodes you have removed from the circuit, but it's generally not a good idea to run a solid state amp with components missing. The diodes may set up bias for something, provide a supply voltage that (if missing) may cause a rail imbalance, etc., etc. I would not fire it up until all parts are replaced and checked good. There are, of course, a few exceptions.

[Mr_bibbles]

I seriously doubt the 0.47 ohm resistors are shorted. They are only half an ohm to start with, your meter is likely having trouble measuring that low, your probes might have that much resistance. Those resistors go bad by going open, not shorted. I am unsure how they could physically short anyway.

I hope you realize the output transistors are MOSFETs and not bipolar transistors. ANy that measure shorted source to drain need to come out, and yes you can remove them all for testing the board.

Shorting power rail to ground COULD have damaged the Q3 or Q4 circuits.

Thanks Enzo, excellent information as always.

[Mr_bibbles]

I'm not sure which diodes you have removed from the circuit, but it's generally not a good idea to run a solid state amp with components missing. The diodes may set up bias for something, provide a supply voltage that (if missing) may cause a rail imbalance, etc., etc. I would not fire it up until all parts are replaced and checked good. There are, of course, a few exceptions.

I've only run it on the current limiter, but that makes a lot of sense. I'll get the new ones in and make sure all diodes are %100 before moving on to transistors. Thanks Dude!

[Mr_bibbles]

OK so I've replaced all questionable components except for lm35 (in the mail...) , turned the bias pot all the way to the left and powered it up- Lightbulb stays dim! Fan goes full blast but I'm guessing temp sensor lm35 will fix this.

So, moving on, what is your "brand new OT transistors" checklist before adding a load and then finally removing it from current limiter?

Rail voltages should be + /- 65vdc, OR + / - 100vdc but I'm reading + / - 81 on a limiter FOR BOTH RAILS (measured at r22/r20 r19/r14)

DC offset is... 37 volts! Have not hooked this up to a load yet or removed from current limiter, bias resistor is still all the way clockwise. ALL transistors in this amp are new.

[Mr_bibbles]

Anyone? Is it possible d2 or d3 has become compromised merging the two rails? My theory is pitiful and but I know Zeners can be used to control voltages for different sections, yes?

[Enzo]

If they test OK as diodes, they generally are OK. But if in doubt, they cost pennies apiece, so swap them out to be sure.

[Mr_bibbles]

OK- First I want to thank all of you for your help so far. I can see the light at the end of the tunnel!


The amp is stable without a load now. Rails with current limiter read 84 /-84 55/-55 (Healthy, about 20v +/- which seems like a result of the current limiter.)

NOW the issue is, the INSTANT you plug a load into the back, the lamp limiter spikes bright and I hear a relay click. You can hear a DCish hum from the speaker. This is the first real complicated SS amp I've ever fixed, so to be honest I'm not even sure where I should be looking in the output section. My only thought is that I have the bias set at it's lowest point until I'm sure this thing is stable and working, could that cause the relay to blow with a load???

Help me M.E.F. Kanobi, you're my only hope.

[The Dude]

......DC offset is... 37 volts! Have not hooked this up to a load yet or removed from current limiter.....

If you have 37VDC offset, there is still something wrong with the amp. Don't hook it up to a load until you find the problem.

[Mr_bibbles]

Ok- So I've done a lot of poking around.

1. Removed v2, as it's driving the output section yea? No change in offset.
2. -/+ 84 Voltscoming in at q4 / q3, but as soon as it passes through them I lose the + half of voltages in the output section. -44 on all the flameproof resistors, at the speaker, etc. Replaced q3 and q4, no change.
3. Can anyone help explain to me the functions of d4/d5 and d6/d7? From a guess it separates the - or + rail from the 'merged' rail??? I have no idea though. Replaced them all, no change.
4. I did notice there is -44 present on the trimpot AND the legs of q2. This means the bias circuit is feeding the output section -44 right, so the output may be functioning correctly and the bias may have an issue?

[Enzo]

Repeat: do not connect a speaker or load until there is NO DC on the output.

V2 drives the output in terms of signal,but note it is DC isolated from the outputs by caps C6, C7.

The diodes are back to back zeners in pairs. They keep the voltage between source and gate of the outputs from going over 13 volts. Did you replace them with 13v zeners? DO NOT replace them with plain diodes.

It sounds like your +65 main rail is missing.

[Mr_bibbles]

Thanks for the help Enzo,

Question 1: So C6 and C7 are blocking the DC, and if they're functioning the output should have it's own supply independent from the previous stage? If C6 had failed, could that possibly merge the + rail with the previous one, accounting for the offset?

I did replace them with In4743 13v 1w Zeners. It definitely appears that the +65 is missing, but it is present on the other side (top of schematic) of q3. When Measured at R22, there is +55v present (lightbulb limiter connected).

Question 2. Since all of the output transformers and flameproof resistors have been replaced, as has q3 and q5, what part of the + supply circuit should I be examining for failure? I noticed the voltage across r16 and r17 was asymmetrical- 3v and .5v. Conversely, the voltage drop across r14 /r19 matched up at 2.1v. This means the issue is within the output circuit, and not from the rails themself?

Question 3. What are the functions of q5 and q6 in this circuit?? Is there a name for this type of circuit I can google? To be frank I have no Idea what's going on here and would like to understand how it works.