Smoke on the Horizon... oh s**t!

From: Holger Veit <>
Date: Tue Feb 17 04:38:53 2004

On Mon, Feb 16, 2004 at 12:39:12PM -0800, Michael Sokolov wrote:
> So here is my question then: if not tantalum, then what?

Why not tantalum? Take into consideration that nothing is for eternity,
whatever you build in. For this particular application, they are appropriate,
and "The Right Thing"(tm).

> I've always been told that aluminum electrolytic caps were cheap stuff that
> doesn't last and that tantalum caps are much better and (that was the claim)
> eternal. So I guess this is wrong. What is the truth then? Is it the other

This claim has been wrong all the time, and will be in the future as well.
It is by nothing founded on technical issues (and can't be); it is just
opinion and perhaps techno-religion. Don't expect truth.

But BTW, there has been some technical improvements to electronic components
since the days of the NorthStar. More than 20 years *is* somehow
eternity for certain electronic components. Replace the capacitors in question
with modern low-ESR types *designed especially for use in power supplies*.
This will give you some more years - for that particular kind of failure -
but it certainly won't give you any warranty that the system doesn't suffer
another failure.

> way around? Or are they both bad? And why are so many people, including on
> this list, saying "aluminum electrolytics are cheap crap, use tantalum"?

Price shouldn't be an issue for conserving this old stuff. There is
more than the question of aluminum caps vs. tantalum ones - that's why the
manufacturers offer such a large variety of capacitors, and resistors, and

> And what are the *good* caps to use?

Sorry, any recommendation will be useless here, as you might not find the
exact component (and slight variations do matter), and depend on the circuitry,
and the environment. You just touched the area of correct design work of
electrical circuits. If you expect "eternity" you cannot rely on any
cookbook recipe if you have no clue what you do.

Received on Tue Feb 17 2004 - 04:38:53 GMT

This archive was generated by hypermail 2.3.0 : Fri Oct 10 2014 - 23:36:42 BST