whisky-dave wrote:
On Thursday, 30 April 2020 20:35:27 UTC+1, Commander Kinsey wrote:
Cost is only 20% higher. Lifetime is many times greater.

Not compared to the types of capacitors check out the data sheets
typically up to 2000 hours.

You have to apply the Arrhenius temperature correction to those
service hours at high temperature.

In chemistry, Arrhenius is taught as if it's a monolithic
thing. Every 7C (10F) rise in temperature, doubles reaction rate.

In electronics, something different happens. Arrhenius
exponential is curve fitted, and the doubling rate
for reliability, differs depending on topic area. The
doubling is around 15C or so for curve fit of
electrolytic capacitor reliability. You use the curve fit data
from the manufacturer, to correct the service hours at
a lower temperature.

If the capacitor operating hours are five temperature-doublings
lower, then the service hours would be 2^5 * 2000 hours or 64000 hours
sitting inside your computer case with the fan running. It's
not that warm in there.

When quoting the 2000 hour number, you also have to include
the temperature at which the number is quoted. A capacitor
quoting 2000 hours at 105C is better than a capacitor
sporting 2000 hours at 85C. You have to adjust them to
a common temperature, and then compare the hours, for
a fair comparison. Maybe the one at 105C is four times
better than the one at 85C. As a rough number pulled
from the air.

When you get those catalog numbers, it would pay then
to find the appnote where the manufacturer has
characterized the devices, and extracted the temperature
necessary for a doubling of reliability. Be it 15C or
whatever.

Paul