Ampeg SVT 3-PRO output stage repairs

[jonnyeye]

I've worked on small tube amps and pedals, this is the first big solid state amp I've touched. A friend asked me to look at it as it was fuzzy/distorted at low volumes - a sign of bias issues. I thought it would be an simple fix (as there is lots of information here about the amp, thanks!) but as I was doing checks I managed to short one of the output transistors, which blew. I've got it back up and running (the protection circuits did their job thankfully), but I have a couple of questions that relate to long term reliability:

Only one of the output IRF240s was dead (the aforementioned Q7), so I replaced just one with a new one (Vishay manufacture, replacing an IR part) that matches for Vgs. I do have a set of four new transistors that match - should I replace all of them instead? Also, is it safe to reuse the existing sili-pads dry, or should I replace them, or...?

Thanks.

[MarkusBass]

...so I replaced just one with a new one (Vishay manufacture, replacing an IR part) that matches for Vgs.

I'm not sure what you mean by that. You usually need at least two transistors to matched them. How did you verify that the new transistor matches for Vgs? I think that it's very difficult (or very easy ) to match just one transistor. SVT3Pro uses matched transistors but four of them have to be matched. With the new transistor you can check whether the quiescent current of this transistor is simmilar as of the other transistors. If not, my advise is to replace the whole set (assuming that you verified that the set is matched) - not alway transistors from one production batch are matched. I usually buy 8-10 transistors to get 4 of them matched.

Mark

[jonnyeye]

Sorry, I wasn't quite clear. I bought a lot of 8 and used the matching technique Nelson Pass describes - 15V through a resistor (I used 2k2 for 5ma current) to gate and drain, source grounded. I found one new one which matched the three existing good ones (as well as a matched quad of new). After rebiasing, the N-channel side with the new transistor matches better than the P-channel side does (with all stock transistors). I am just concerned for the long term.
Also, what are your thoughts on reusing the existing sili-pads?

Thanks for your help,
-J

[MarkusBass]

I often found out that "factory matched" MOSFETs in Ampeg differ more than 300% one from another (whilst it should be something like20-30%). So you matched the new transistor better than the factory does and you are concerned for the long term. But what exactly you are concerned with? Even with high differences between MOSFETs the amp will work correctly for many years. Regarding reusing the existing sillipads my advise is to look at them with magnifying glass. If you don't see anything wrong, you can use them. But they are also very cheap so if you worry about using old sillipads, you can buy a new ones.

Mark

[jonnyeye]

Ok, thank you very much! My concern is mainly that I don't have the experience to feel that I can make judgement calls - I guess I am just trying to make sure I am on the right page.

-J