0
\$\begingroup\$

What does the "Ripple current" and the "Life (in Hours)" in electrolytic capacitors specifications mean?
My intuition says it's the highest current it can handle and it can operate for that hours (when it works under that ripple current) before it fails, is that right?

enter image description here
In this picture it's not mentioned the "life" specs but I've seen it in many other (mainly electrolytic) capacitors. It's usually something around 2000 hours.

Let me know if I get this right. This is a theoretical question.
Let's say we have the following specs (after calculations) for a DC to DC converter.

\$C = 350 μF\$
\$ ESR_{min}=10 mΩ\$
\$V_{OUTripple} = 50 mV\$
\$I_{Lripple} = 3,5 A_{peak-to-peak}\$

What would happen if the capacitors can handle lower ripple current (even in parallel)?
For example, if I put 2 capacitors like the specs above does that mean my capacitors can handle up to 1700 mA?
Would it be destructive for the capacitors if they can't handle the specified ripple current?

Could I ignore the ESR if I'm okay with higher voltage ripple (\$V_{OUTripple} = ESR * I_{Lripple}\$)?

\$\endgroup\$
2
\$\begingroup\$

Electrolytic capacitor lifetime is given in hours at the maximum operating temperature, and that's what it depends on: its operating temperature.
For every 10°C reduction in operating temperature, the capacitor lifetime is doubled, so for a 105°C cap rated for 2,000 hours that means that it could last about 32 times longer, or about 64,000 hours at 55°C.

The ripple current rating tells you the manufacturer's recommended maximum so that it doesn't cause a detrimental rise of the capacitor's temperature. You could ignore it and that will shorten the capacitor's lifetime.

You should either use a larger capacitor or multiple capacitors in parallel so that their combined rated ripple current is higher, and their combined ESR is lower than the ones calculated for your circuit. Using higher (than calculated) capacitor values can only benefit both your circuit and your capacitors (and their lifetime), while using lower values will shorten the lifetime of the capacitors and the circuit along with them, and could also impair the circuit's functioning.

\$\endgroup\$
2
  • \$\begingroup\$ I think using very high capacitance than the required one may cause stability problems. But I think you answered my question about capacitor ripple current. \$\endgroup\$
    – MrBit
    Mar 26 '20 at 21:10
  • \$\begingroup\$ In high frequency power converter circuits, the ESR and the ripple current (also the acceptable ripple voltage) ratings are usually the most important factors in determining the size of the capacitors. If you only used capacitance vs. frequency calculations, you would indeed come up with much smaller capacitor values, but we don't live in an ideal world. Larger capacitance on the output actually INCREASES stability in this type of circuits, plus it ensures longer proper operating lifetime as the capacitors dry out and lose their capacitance over time (worse so in hotter, power circuits). \$\endgroup\$ Mar 31 '20 at 14:03
1
\$\begingroup\$

The main failure mechanism of an electrolytic capacitor is time at temperature drying out the electrolyte. The lifetime is usually quoted to an increase in ESR or a reduction in capacitance, often in the 10% to 20% range, rather than total failure.

850 mA rms ripple current in your specified capacitor will cause a heating of \$I^2R\$, so 0.8520.08 = about 60 mW. This doesn't sound a lot, but it's a small case, so there will be a temperature rise above ambient. At lower temperatures you would hope for a longer lifetime. Two of those in parallel would be able to handle 0.85 Arms each, or 1.7 Arms total. You would need five of those in parallel to handle 3.5 Arms, it depends how your ILripple is specified, is it rms, peak, or pk-pk?

\$\endgroup\$
2
  • \$\begingroup\$ Thanks! The current I mentioned is peak-to-peak. Does that mean I need capacitors with ripple current 3.5/2 ? What would happen if the total capacitors can't handle the total ripple current? \$\endgroup\$
    – MrBit
    Mar 26 '20 at 17:09
  • 1
    \$\begingroup\$ I have to correct Neil_UK here: the ripple current through your capacitor in this case will not be the rated ripple current of the capacitor, but the calculated ripple current of the circuit, which means you will get a lot more internal heating and a shorter lifetime. \$\endgroup\$ Mar 26 '20 at 19:01

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.