7
\$\begingroup\$

I am considering using a KZE series (by United Chemi-Con) electrolytic capacitor, which has a specified lifespan of 5,000 hours at 105°C with rated ripple current. Now, for an example assume I am using it at room temperature (25°C) with half the rated ripple current and half the applied working voltage. How long would the capacitor last? In future, how would I calculate it?

\$\endgroup\$
3
  • 1
    \$\begingroup\$ Your wall thermometer may say 25'C, but I guarantee that, in operation, your capacitor is warmer. \$\endgroup\$
    – davidcary
    Apr 8, 2011 at 21:07
  • \$\begingroup\$ @davidcary, you're right. So, let's assume a 470µF 25V cap which according to the KZE datasheet has an ESR of 0.041 ohms (@25°C) and a ripple current maximum of 1,250mA. We can calculate at, let's say 600mA ripple current, the cap will dissipate 14mW (I^2 * R.) So, where does the cap heat up from, as this is tiny? Am I missing something? \$\endgroup\$
    – Thomas O
    Apr 8, 2011 at 22:34
  • 2
    \$\begingroup\$ For some strange reason, people try to hide most capacitors in opaque boxes full of other heat-generating stuff -- fast CPUs, or quickly-discharging batteries, or transistors switching heavy currents, or etc. The temperature of the cap (and everything else in that box) is at least the product of the total heat energy generated by everything inside that box, multiplied by the thermal resistance of that box. I don't know anyone who can look at a schematic and accurately guess the actual interior temperature. In my experience, it's always at least 10'C hotter than anyone expected. \$\endgroup\$
    – davidcary
    Apr 17, 2011 at 0:29

1 Answer 1

10
\$\begingroup\$

The general rule used is that usable lifetime doubles for every 10°C reduction in operating temperature. This should be considered an upper bound on expected life.

As the capacitor ages its capacitance will decrease and ESR will increase. The increase in ESR can result in a temperature increase if running high ripple currents which in turn increases the rate of aging.

Also make sure that the increase in ESR / decrease in capacitance over the cap's life span is acceptable. For instance, make sure that the circuit still operates when the cap hits -20% of baseline capacitance, or whatever its end of life rating is.

Other factors can increase the rate of aging as well. For instance, the more air flow over the cap, the shorter its life expectancy due to evaporation effects. On the other hand, a reduction in temperature from the air flow can more than offset this issue if the cap is running warm.

\$\endgroup\$
3
  • \$\begingroup\$ Here's a resource from UCC on capacitor life, which elaborates on the above. \$\endgroup\$ Apr 6, 2011 at 12:57
  • \$\begingroup\$ Do you have a reference for this air flow effect? \$\endgroup\$
    – davidcary
    Apr 8, 2011 at 21:08
  • \$\begingroup\$ @davidcary I actually can't find the original source for that. I was doing a lot of research on this issue awhile back and it came up in some report I was reading. The reasoning was that AL caps don't have a perfect seal and the air flow increases evaporation of the electrolyte. I believe it was in reference to large snap in caps rather than low voltage caps which generally have much better seals. \$\endgroup\$
    – Mark
    Apr 11, 2011 at 22:27

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.