3
\$\begingroup\$

I want to use supercaps in a system for minimal power interruption backup (1-2 seconds). My systems works at - 40 to 85°C.

I have seen that supercaps are usually rated for 1000 - 2000 hours at max temp (say 85°C) and max voltage (say 5.5V).

And from my understanding this rating is the point where the capacitor will loose 30% of its capacitance and its resistance will double.

Now say my system works at average of 45°C (but will peak to 85°C during hottest days for few hours) and instead of using 5.5V I run it from 3V and instead of needing the whole capacitance I am OK with even 50%

How do I calculate the actual lifespan? I don't see charts in the datasheet but only the number of 1000 hrs @85°C with 5.5V.

\$\endgroup\$

2 Answers 2

4
\$\begingroup\$

Generally the estimate that manufacturers recommend is that life doubles for every 10°C below rated. This is the Arrhenius equation with some assumptions.

Some also have a correction factor for slight voltage derating of Vop/Vrated for up to 2x. The factor used is Vop/Vrated.

So at 45°C the life would be 2^4 times the 85°C rating approximately, or 16x the 85°C rating, so 16,000-32,000 hours ignoring any improvement from voltage derating.

n.b. The relevant temperature is the core temperature of the capacitor, not the ambient. There can be a significant difference between the two for power supply capacitors where self-heating is significant. Probably not so much for a supercapacitor in a typical application.

The excursions above 45°C and the time below 45°C will affect the life, so there will be a shelf life (sometimes specified separately).

If you have a known temperature-time curve you could probably write an equation for instantaneous 'wear out' as a function of time and integrate to get an estimate of the life (when wear out passes 1.0, say) for your arbitrary time-temperature curve.

Or, in a simplified form, consider the time spent at 85° as a = 10% and the rest spent at 45°C. We can then write:

Life \$\approx \frac{1}{\frac{a}{2000}+\frac{1-a}{16000}}\approx\$ 9400 hours

\$\endgroup\$
2
  • \$\begingroup\$ What happens after the stated life? It drops by 30% capacitance and resistance doubles. When does it happen again? If I also take more capacitance and very low esr than needed can I also use that? \$\endgroup\$
    – Uriel Katz
    Commented May 14 at 4:24
  • \$\begingroup\$ The wear-out mechanism AFAIUI is that the electrolyte dries out, so ESR increases and capacitance decreases. Since ESR also increase at low temperatures, a worst case would be very low temperatures near end-of-life. Of course if you have more capacitance/lower ESR than your circuit needs to operate your circuit will have longer life since the end-of-life (due to wear-out) point is arbitrary. The opposite is also true. \$\endgroup\$ Commented May 14 at 5:29
1
\$\begingroup\$

Check the application notes. For example, KEMET provides this paper:

Choosing the Right Supercapacitor for your Application

Wurth's ANP012 | How does a Supercapacitor age? Lifetime Model of Electric Double Layer Capacitors also gives curves for temperature, though not a calculation method.

The materials used should be fairly common between manufacturers, but to be sure, check with your manufacturer.

\$\endgroup\$

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.