0
\$\begingroup\$

I'm finding information that seems to be contradictory in various places, and I think people are talking about different ways of looking at the same spec.

The specification that Digikey uses to search/sort by is called Ripple Current @ Low Frequency.

This is measured at 120hz on many of the capacitors that I've looked at. 120hz is the frequency of the pulsed DC in a linear power supply with full wave rectification.

If the application is ripple smoothing in a linear power supply, specifically on the input side after the bridge rectifier and before the voltage regulator, assuming all other things are the same (voltage, capacitance, life), which capacitor is the better choice?

A. Ripple Current @ Low Frequency = 860 mA @ 120 Hz

-or-

B. Ripple Current @ Low Frequency = 3.4 A @ 120 Hz

\$\endgroup\$
2
  • \$\begingroup\$ All other things can't be the same. For example, higher ripple current implies lower ESR, which means lower power dissipation, which means longer life... \$\endgroup\$ Jul 28 at 22:36
  • \$\begingroup\$ if you want lower ripple, go for lower ESR. One neat way of doing that is using many parallel capacitors instead of one big expensive one, since parallel resistors divide. 3 cheap 330u's will have a lower ESR than most "low ESR" 1000u. The ultimate would be something like 68 cap board made for high-end audio: ebay.com/itm/254524509484 \$\endgroup\$
    – dandavis
    Jul 28 at 22:53
1
\$\begingroup\$

If the application is ripple smoothing in a linear power supply, specifically on the input side after the bridge rectifier and before the voltage regulator, assuming all other things are the same (voltage, capacitance, life), which capacitor is the better choice?

A. Ripple Current @ Low Frequency = 860 mA @ 120 Hz

-or-

B. Ripple Current @ Low Frequency = 3.4 A @ 120 Hz

Either might be, depending on your specifications. B is the better capacitor, but A might be good enough, and the better choice if it's cheaper.

The component chosen for your post-rectifier reservoir capacitor must meet at least two criteria

  • enough capacitance to feed the load with current when the rectifier is not conducting, while keeping the low voltage dip and the ripple amplitude below limits for the load
  • low enough ESR so that it does not overheat due to the ripple current flowing

Your load defines its input current, so also defines the ripple current that the reservoir capacitors see. If your particular load requires a ripple current of 100 mA, then either capacitor would do. If it requires 1 A, then you would have to use capacitor B. If it requires 500 mA ripple, then your choice would depend on your attitude to quality and longevity. It's within the published capability of A but near the maximum, and B will run cooler and last longer, especially important in a hot ambient. You'd have to make a cost/performance judgement, maybe after measurements of actual temperatures in your operating environment.

\$\endgroup\$
1
  • \$\begingroup\$ Often times the capacitor with the higher Ripple Current @ Low Frequency value is cheaper too, which is why I wanted to make sure I was reading the spec correctly. For example UFW1V472MHD vs UKA1V472MHD. The UKA has a lower ripple rating, but costs $2 more. The only other difference is the temperature rating of 105 C vs 85 C, which favors the UKA. \$\endgroup\$
    – Nick
    Jul 30 at 0:04
0
\$\begingroup\$

General Purpose bulk caps are rated in ripple current for 120 Hz to standardize D.F. Are “normal” ESR bulk caps. My Rule of thumb is ESR * C < = 10 us is low ESR. And > 100 us is G.P.

Low ESR Caps will be rated as such and are less than 10% of G.P. ecaps.

The uncommonly used term “crest factor” is the ratio of peak charge to average discharge current is also about equal to the 1 / % V ripple in Diode rectifiers since they charge only just before the leak voltage.

  • so a line f bridge cap that supplies 10A at 10% voltage ripple must support 100Apk or 10x the average outlut current for peak, but heat loss is rated in Irms ^2 ESR and is usually rated by the % Dissipation Factor instead.

SMPS demand low ESR ripple in mV unlike 100/120 Hz caps with 10x more ripple tolerance and can user lower cost caps for the same energy storage.

More details, ESR in Electrolytic capacitor vs Voltage rating

\$\endgroup\$
0
\$\begingroup\$

The short answer is yes. The ripple current is the amount of current passing through the cap. Higher current = more power dissipated in the cap which makes it hotter, hotter cap = shorter life. Caps rated at higher ripple current should last longer, but you will need to take several variables into consideration when comparing.

\$\endgroup\$

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.