>=20
> 1. Tantalum capacitors are touchy and should be well derated (probably
> by to 50% of their stated value).

That is a requirement used for derating space flight components - capacitor=
s to be derated to 50% rated voltage. A specific waver needs to be applied =
for if this derating is to be violated.

> 2. Aluminum and ceramic capacitors are less touchy and can be used at
> closer to the stated value but still should be derated (say, to 75%).

Wet aluminium capacitors should be run close to rated voltage to maintain t=
he die-electric at the correct thickness for the capacitance value. If run =
significantly under voltage (the prime example is 1uF 50V ones which come i=
n a titchy little can) the die-electric which is formed by a chemical react=
ion on the surface of the electrode cannot maintain the 'forming' and goes =
funny, thinning out and leading to high leakage and wrong capacitance value=
s.

=20
> So, I'm wondering- what sort of test data exists out there to back this u=
p?
> Has there been any critical study of this? Any articles regarding it?

For high reliability space applications, where a supply bypass capacitor fo=
rms a single point failure node that can take out the whole instrument, it =
is 'considered normal' and often required to have 4 capacitors in a series-=
parallel network with a pair of voltage balancing resistors, so that a sing=
le capacitor failure doesn't short the power supply.
--=20
Scanned by iCritical.

--=20
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.