What's the ideal operating temp for capacitors?

[TELVM]

We know heat is bad for them, but there is a such thing as excessive cooling for capacitors?

For instance, could a constant 21C internal case temperature be 'too cold' for the polys/lytics inside a computer?

What's the ideal operating temp for them?



(Just for laughs ): The Ice Cap & The Submarine Vents

[Heihachi_73]

I had a feeling it would be Rodalco or someone like that.

[eccerr0r]

There is a tradeoff of course, actually cooling by peltier is better than fan because fans will encourage evaporation. However if things get too cold where liquid electrolyte freezes, that's a problem. Also cooler temperatures will reduce ion mobility, increasing ESR (that's why some capacitors work better when heated.)

I think there is a happy point but in terms of overall performance, but for lifetime you can't do better than keeping it cool, just don't freeze them.

[tmiha71]

Hard to say, but recommendation is to use capacitors with lower value of Z-25℃ / Z+20℃ (quotient of impedance at -25 degC and impedance at 20 degC ). That will result in lower deviation...

In datasheets usually defined at 120Hz, but (probably) can be applied to higher frequencies - capacitors with higher quotient will have higher impedance than capacitors with lower quotient, probably valid for increase in temperature too...

Some datasheets give information about impedance versus temperature - if there is that info it should be consulted first...

Some datasheets give information about capacitance versus temperature, if we presume higher capacitance leads to lower impedance (always true), it's safe to presume higher capacitance also leads to lower ESR...

To emphasize - deviation is bad, leads to shorter lifespan of electrolytic capacitor...

You should search through this forum, think there was posts about this issue, and some info about other types of caps.