Re: [CAD_CAM_EDM_DRO] Re: tantalum caps was emc success

ballendo@... wrote:

> Wally K, Jeff, Jon E, Mariss, List,
>
> I too have used tantalums' where they seemed to make sense. Never
> blown one up. No electrolytics, either. So I gotta ask myself, "Do I
> feel lucky?... Well, Do I...?
>
> What I want to know is:
> Is this a situation (like selenium rectifiers, and OLD paper/wax
> capacitors) where the value, or other operational characteristics (of
> the tantalum cap) change, and LEAD to the "flash, Bang, and white
> smoke"?

The Tantalum dielectric is fragile, and is built into the capacitor at
assembly time by chemical processing. The Aluminum electrolytic
dielectric can be reformed pretty much at will by the chemistry
already inside the capacitor. If you polarize the tantalum cap
in reverse, or charge or discharge too quickly, it can destroy the
dielectric. If you leave it discharged for too long and then apply
power (in a decoupling application) they have a tendency to fail.
An aluminum electrolytic will draw current for a moment until
the dielectric reforms, but it generally won't hurt the capacitor.

When the tantalum dielectric fails, it fails shorted. A shorted
capacitor across the 5V 100 A power supply in one of those
old 'mainframe' computers makes a big mess. In fact, a late
military spec requires a fuse built into every tantalum cap!
If it explodes, you're lucky. That clears the short, and you
don't have a fire. If it just gets real hot, it can burn big holes
in the PC boards.

Jon