Replacing electro caps in older gear - when is it time?

[AusTex64]

I have a Manley Elop from the early 90’s that had a 6414 crap out. As I was changing it out, I thought “is it time to replace the electrolytic caps in this thing?” I use it on everything so it has a lot of hours. Yet my top tech and I inspected the electros throughly for bulging tops or sides, found nothing of concern. I suppose we could go through the circuit with the scope looking for DC or other signs of leakage. We opted to do nothing at this time.

Where this is headed - is there a rule of thumb, based upon your experience, that a piece of gear X old gets recapped? I almost always recap Fender amps from 1970 and older if it’s never been done before. What about outboard gear?

 

[beatnik]

An ESR meter like Peak ESR70 provides some information on the health of the cap, to a certain extent.

Sometimes it turns out the cap is leaky or the capacitance drifted even if there are no actual visible signs on the outside.

 

[abbey road d enfer]

An ESR meter is good only when you have a standard to compare with.
Older lytics generally have larger ESR than modern ones, just because nobody really cared at the time. It's only when smps came that ESR became a concern.

 

[beatnik]

the ESR70 manual has a chart that lists the expected ESR for a given capacitance / voltage rating.
it doesn't really tell accurately how much life is left in the cap but it can definitely spot the duff ones.

in the case of the op, if it's a piece of gear that is important and gets used a lot I wouldn't mind replacing the caps with some high quality long life rated at least +105 C , lot of manufacturers cut corners in this department and even Manley most certainly didn't fit the best capacitors available

 

[JohnRoberts]

Even old rack gear can experience elevated ambient temperatures depending on what it is racked with and even internal design. I've seen some early digital stuff that ran firecracker hot inside. Heat is the enemy of electrolytic caps.

Good practice when troubleshooting old gear is to measure everything you can and look for outliers or deviation. A dried out blocking cap will often reveal itself as lacking LF response. For PS decoupling it is harder to measure in circuit.

If you discover one bad soldier, it is good practice to replace all of the same value that likely came from the same production batch and experienced similar stresses.

Of course do no harm...

JR

 

[abbey road d enfer]

the ESR70 manual has a chart that lists the expected ESR for a given capacitance / voltage rating.

This chart is probably relevant to modern capacitors. Electrolytic capacitors technology has evolved so much in about 30 years that a capacitor that was of good quality in t995 would be rejected today by this chart.

it doesn't really tell accurately how much life is left in the cap but it can definitely spot the duff ones.

The important factors are not the same for diverse applications.
For coupling caps, nominal capacitance and leakage are paramount.
For decoupling (bypass), capacitance is paramount, with ESR a close second. Of course it it leaks and heats up, it's no good, but the ESR meter won't tell it, particularly if measured "in-circuit".
For smps, ESR is the number one issue.
There'e not one single, simple answer.

in the case of the op, if it's a piece of gear that is important and gets used a lot I wouldn't mind replacing the caps with some high quality long life rated at least +105 C

I fully agree.

 

[pucho812]

rule of thumb, most people are capacitor crazy and will recap thinking it solves issues, a lot of times it's not that simple, but recaps don't hurt so from what I have experienced first thing people do when there is an issue is to recap even if it is not needed.