I have an application for a supercap in a 1 amp sinewave generator. It would be a 10F, 5V supercap. I can not find power ratings for the supercaps of this size available via Digikey or Mouser.

If I subject my supercap to 1 amp sinusoid for an hour, will it be damaged? If the supercap internal resistance is, say, 50mOhm, that is only 50mW. That is not enough to heat a thing the size of that component above its rated temperature (it is certainly larger than a 1/4 watt resistor), so I am below that limit. Are there other limits to the current rating for supercaps?

The frequency of the current runs from Khz to a few cycles per second. The capacitance is high enough that the voltage limits of the capacitor are not exceeded.


For more conventional capacitors, see this answer on SE. There, you will find a link to 'Selection of Capacitors for Pulse Applications', relating to Polypropylene capacitors. It is not limited entirely by thermal dissipation.

This same company (WIMA) offers supercaps for power applications, like cars and wind storage, and they do have current ratings. Here is the WIMA supercap_r page, where the lowest current rating, for a 100F supercap, is 30Amps (r=12mOhm, so 10 Watts). A note on the page implies that current ratings for their supercaps are strongly related to thermal dissipation: "Due to the large case surface of the cells heat being generated by the continuous current flow can better be drawn off."

An attempt to get some sort of metric from their data gives me from 120mW/gram to 250mW/gram for their 90 to 120 gram capacitors.

The little CooperIndustries 3F supercap is 8grams, so (using 100mW/gram) a limit of 800mW is predicted, or 4 Amps.

All that could be worthless info, however. The little cap may not have been designed for anything but a backup battery replacement, with no attention to the details needed for high continuous current use.

  • \$\begingroup\$ Heat dissipation is not a factor of weight, but surface area-- assuming no heat conduction through the pins to the PCB. The cap you reference in the Q says plainly in the datasheet that it is intended for pulse power and bridge/hold-up power. Clearly not intended for backup battery replacement where the currents are much lower. \$\endgroup\$ – user3624 Jan 28 '13 at 0:17
  • \$\begingroup\$ I used watts/gram to target rate of temperature rise. \$\endgroup\$ – Bobbi Bennett Jan 28 '13 at 0:52
  • \$\begingroup\$ Watts/gram is correct for temperature rise, assuming that you have a fixed number of watt-hours. But you don't. Given infinite time, you have infinite watt-hours. If you don't dissipate that heat to the surrounding air then eventually 50 mW is going to become unbearable. The speed at which you can dissipate that heat is roughly proportional to surface area of the cap packages. \$\endgroup\$ – user3624 Jan 28 '13 at 1:19
  • \$\begingroup\$ @DavidKessner I am not convinced equilibrium operating temperature is the limiting number here, but instantaneous internal temperature might be. \$\endgroup\$ – Bobbi Bennett Jan 28 '13 at 1:33

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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