Recapping procedure?

[Axtman]

What is your recapping procedure?

1) Replace all the electrolytic capacitors in the amp.
2) If one or more capacitors test bad then replace all the capacitors.
3) Only replace the capacitors that test bad.
4) Leave the amp alone if it works.

[nickb]

It depends on the customer, the construction and the age.

I seldom go for (1) but the difficulty of doing that in some amps is a factor. I tend to go for (3) but if the caps are 30+ years old then it's 2 or 1.

[The Dude]

Generally, the same here. The customer's budget is also/sometimes a factor.

[Axtman]

What brought about this question is that I am currently working on a Fender SS amp from the early 80's. The filter caps are good quality and test fine. I'm inclined to let sleeping dogs lie (#4) On the other hand if I work on a tube amp from the 70's or earlier, those electrolytic caps get replaced without even testing (#1). Usually one or more are leaking anyway.

My experience has been that caps that get high voltage are under higher stress and tend to fail...except for multi-section can caps. (See my thread about that).

[Helmholtz]

Don't forget the strong influence of temperature on ecap lifetime.

[The Dude]

For sure! I've repaired many an SMPS simply by checking caps near heat sinks and large power resistors. I always leave leads slightly longer when replacing so that the cap can be tipped away from the heat source. IMO, putting e-caps next to a heat source is just poor design.

[nickb]

What brought about this question is that I am currently working on a Fender SS amp from the early 80's. The filter caps are good quality and test fine. I'm inclined to let sleeping dogs lie (#4) On the other hand if I work on a tube amp from the 70's or earlier, those electrolytic caps get replaced without even testing (#1). Usually one or more are leaking anyway.

My experience has been that caps that get high voltage are under higher stress and tend to fail...except for multi-section can caps. (See my thread about that).

It's dangerous to generalize, especially as bad batches can and do happen, but ecaps got a lot better after around 1975 or so. For that reason most of the total recaps I've done are from before that time. OTOH I just did a a 1991 Vox AC30 in which the low voltage caps had leaked so badly they had eaten right through a couple of PCB traces.

[Axtman]

I have also worked on younger amps that had bad leaky filter caps probably due to the capacitor plague.