Replacing Capacitors - Results?

[tdiaz]

On some older boards, of the SE/30, IIci era, you can see capacitors leaking all over the place - and that's pretty obvious .. "replace them", but when you've got boards of similar vintage that are inoperable, and showing no distinct sign of leakage .. would you still shoot for the capacitors first?

Does this have a good return rate, of returning boards to operation, statistic wise?

 

[Andropov]

The fact that there's no visible leak doesn't mean that there isn't any. Capacitors can start leaking by its bottom, so the leak wouldn't be visible until it was bigger. And sometimes you can't see a leak, but the capacitors are swollen, or they have cracks.

If a working old board dies, it's probably due to capacitors, at least if we're talking about a motherboard (power supply boards can fail because of a wider variety of things). If you aren't the original owner of the board... then you know if it failed by its own or by a human error (water, high voltage peak, overheating, to if it was hit...). But, anyway, an electrolytic capacitor will leak, later or sooner, so the more old it was when it failed, the more likely is a capacitor.

If you know soldering, or know someone who does, it worth a shot. Capacitors are really cheap.

 

[tdiaz]

I've got a couple IIgs prototypes that used to work quite well, and have since gotten to the point where turning it on once in about 100 times it might work, otherwise it's mostly got an offset screen, missing about a third, ADB input does not work, or it may even get a '911 error' (which is common on the older IIgs, not the ROM 3)

The SE/30 like capacitors look reasonably dry, and there's not that many of them.. I'll have to put together a list and try it.

Never tried any of that over the years.

Soldering isn't an issue, I build stuff often.

 

[MacTech68]

You can use the Sony camcorder diagnosis in some instances.

Years ago, when camcorders shrunk in size, Sony used the same pesky surface mount electrolytic capacitors. When units began failing, a bulletin was released advising to use a hairdryer to gently heat the capacitors and re-test. If the unit began working, the diagnosis was that the capacitors were beginning to fail.

That's why when a Mac Classic first goes "checkerboard", leaving it on for about 20 minutes and then switching it off and back on immediately would often make the machine spring back to life.

Eventually, the capacitors would completely fail and the gentle warmth trick would no longer work and tracks would begin to be etched by the leaked electrolyte.

Of course, a gentle heat can also reveal cracked/dry solder joints, but these are unlikely to occur on the logic board. However, they do occur on the analog board.

As for statistics, it's not that easy, especially if tracks have been etched through. The difficulty is finding them, especially if they are under an IC. My personal experience is about a 75% success rate.

Also, replacing the surface mount capacitors is IMHO, is a priority. Especially on boards that don't appear to have leaking caps. These horrible caps will eventually leak & etch tracks which makes repair even more difficult.

Replacing them with surface mount tantalum capacitors is my recomendation. They don't leak and if they fail, they generally short and blow a pin size hole in themselves, making them easy to find.