Peavey 400 amp missing part.

[g1]

The amp seems to go between working and loading down everything. Probably because I have a lightbulb is series now.

Can you give more detail about this? There are certain things that will give false indicators on the bulb, like trying to run signal.

[Enzo]

They don't need adhesive, you are assembling it by hand. They are held in place between the part and heat sink. Why are you clinging to them as the problem? I swear, if they poke through you blow fuses, because it grounds out your main power rail. They don't "fail", they can tear.

WHAT elevated voltage? It runs on only 50 volt rails.

Frankly the most suspicious part in the thing is that speaker jack with the extra contacts.

[capndenny1]

Getting warmer!

I took the power transistors off. I inspected the insulators. Some had more than 1 insulator, which I'm sure is not a good idea. I was also concerned about the temp sense diode on the power transistor board. It has a clamp on the back panel that holds it down. If it would slide a little to the side it would short out the the back panel. I have some thermal glue, I may make sure it is centered, and glue it in place just in case.

So I put just the driver transistor and one of the set of 3 main power transistors. Same thing, light bulb on. So I measured the resistance of the power amp rails and found a shorted driver transistor? Perhaps I did that moving the heat sense diode? I replaced the shorted (12 ohms) transistor and out it together, and it seems to work fine. The owner gave a me a bunch of transistors. Unfortunately I already found 2 that were bad. And I don't have any more that are good.

I have some of the oem Motorola SJ6357's but not enough. He used MJ15003G transistors. I checked them all and the Hfe is all over the map on them. Shouldn't I try to match those three as much as possible? The 4th one on that side probably doesn't matter so much, but I would think the 3 that are parallel should be close. What about the 3 on the other side of the push pull? Do I need to match the + side amp to the - side amp? This a guitar amp, not a hifi amp, but if one tranisitor is doing all the work, when you try to get to 400 watts, poof! I guess I could get like 20 and find 6 out of the batch that are close?

[Enzo]

DOn't fret the small details, they rarely matter. No, the plus and minus sides need not be matched. In fact, none of them need matching. Those ballast resistors, the 0.33 ohm guys for each power transistor, make the currents even out.

Your hand meter does not measure gain at current. Your transistors have their gain spec'd at several amps, not the couple milliamps of your meter. The SJ6357/MJ15003 is spec'd by Peavey at 4 amps, for example. Look at the curves. Your gains will even out a whole lot at that current, plus the ballasts.


You have THREE outputs on each side, the fourth one is a driver. That is Q3, Q13 are drivers. Note they drive the bases of the other three. Q2, Q12 are predrivers.

ANy time any of the large transistors fail we always replace the predrivers as well.

[J M Fahey]

As a side note, the Silpads are considered disposable, for the very good reason that they are made out of rubber,l which crystallizes so loses all elasticity along time.

When you first apply them they flex and flow and fill any air gap available ...... good.
In a year they crystallize or at least become way more rigid, no big deal because "they are already where you need them" so they do their job.

BUT whan you replace transistors, they will NOT flow to fill the new/different air pockets so they provide false security, they are not doing their job of lowering thermal resistance at a very critical point.

So you either use new ones or switch to grease + mica, in both cases cleaning well mating surfaces.

In that old amp, keeping old silpads and even worse, stacking them, is a bad move to be corrected.

[capndenny1]

The pads are all still very flexible. I will order more.

I cleaned and retensioned the TO-3 sockets. I don’t think it did much.

The light bulb was glowing, but it was only drawing 1/4 amp. So I took the lamp out of theac line.

I was able to turn up theoltage all the way. I go therails up to about 50v, the +/- 15v. It sits there for a while, then suddenly the current jumps way up. So I turn the voltage way down, then when I try to trn the voltage back up the current goes up and the +/- 50v won’t come up. I put the scope on it and the +/- 50v has a bunch of ripple voltage. The output voltage at the speaker out is right at zero.

It seems to work for a while then it goes crazy. Once I let it rest it’s ok again for a while. It almost sounds like. Thermal issue.

I don’t understand why it draws current from both sides? It’s like the idle current goes up, but the center stays correct?

[capndenny1]

So what would fail that way. I’ve watched to voltages on the scope. It sit’s there ok, then all of a sudden something happens, and now you have a high ripple voltage on the +/- 53V rails. Like 8v.

Cap, power transistor?

It will cost me $50 to buy new transistors and caps. Then if it is still doig it, I am out the $50. Itwouldn’t be fair to charge the customer.

[The Dude]

Posting a schematic might get you more replies.

[capndenny1]

There are 3 or 4 schematics on page 1.

[The Dude]

Gotcha. My bad and my apologies. I completely missed page 1!....... More coffee is in order.

[The Dude]

I think you might have a bias issue and it's running away. What voltage do you have across the output emitter resistors? Another thing I like to check: When you replace a driver transistor, check any associated resistors in the path. If a resistor has gone towards open, it can radically change bias. If you can get it to stay on, even at lower voltages, look for abnormal voltages on transistors (a B-E junction reverse biased, missing collector voltage, etc.)

[Enzo]

Is the connector for the heat diode loose?

[capndenny1]

No, i don’t thin so. The connector seems tight. The clamp that holds it down to the heatsink is questonable, but nothing heats up, so I can’t see how it matters. It can’t go into thermal runaway just sitting there idle can it?

I fixed a Woodson (Kustom) amp that had the thermal protect keep tripping, but that was after 5-10 minutes of playing. The temp sensor clamp was loose, so Iused thermal epoxy, and that fixed it.

I guess if the silpads aren’t working and transistor starts to heat up, but the heat desn’t get transferred to the heatsink, perhapsit might run away before the sensor kicks in, and because the heatsink isn’t keeping it cool?

I tried watching various places with scope. Everything goes south at once.

I've blown 4 or 5 power transistors.

I removed all 3 of the parallel transistors and jus left the first one that has the 1 ohm emitter resistor. That seems perfecly happy. But when I add one of the ones with the 0.33 ohms it starts to act up.

I ordered more mj15003g power transistors.

[J M Fahey]

Mesure emitter resistor voltage drop at idle,cold. - Play 2 minutes LOUD - Stop - remeasure - rinse and repeat.

Then draw a graph (pencil on school type graded paper is enough) showing voltage drop (idle current) creep up.

Post results.

[Enzo]

but nothing heats up, so I can’t see how it matters. It can’t go into thermal runaway just sitting there idle can it?

I didn't expect it to be thermalling. But if that diode circuit opens - as it would if the connector was intermittant - then the outputs would slam on full. THAT is why it matters.