View Single Post
Old 01-19-2022, 12:02 AM   #9
master hoarder
momaka's Avatar
Join Date: May 2008
City & State: VA (NoVA)
My Country: U.S.A.
Line Voltage: 120 VAC, 60 Hz
I'm a: Hobbyist Tech
Posts: 10,861
Default Re: sancon china caps

Sorry I didn't put enough details in that post.
By "better ESR and ripple current", I meant that the replacement cap should be rated to have same or lower ESR and same or higher ripple current rating.

The ripple current rating of a capacitor specifies the maximum "AC" the cap can take before overheating internally and getting damage... and usually that goes together with the endurance rating, specifying for how many hours you can run the cap at maximum ripple current (and usually max voltage and temperature) before non-reversible damage/degradation occurs.

So for example, if a switching circuit is producing about 1000 mA of ripple current and the cap is rated for 2000 mA of ripple current, then the cap will be able to handle the filtering of this circuit continuously without any issues.
Now let's say the same circuit with the same ripple current (1000 mA) is filtered by a cap that is only rated for 700 mA of ripple current. The cap *may* still be able to handle the filtering of this circuit continuously, but only if the cap is cooled well enough, such that the internal temperature of the cap does not exceed its rated temperature.
In both of the above cases, the two caps will work fine. However, the first case with the cap being rated for 2000 mA - that cap should last much longer, because the cap will run cooler internally, due to much lower ripple current presented by the circuit relative to what it can handle. On the other hand, the cap rated for 700 mA may not last very long because it would likely be running at a higher internal temperature. And for electrolytic capacitors, the lifetime of the capacitor approximately doubles for a drop in every 10 degrees Celsius (e.g. a 105C-rated cap with 2000 hour endurance rating should last at least 4000 hours at 95C, 8000 hours at 85C, and etc.)

As for ESR - this directly dictates how well the cap will be able to function as a "true" capacitor in the circuit, and therefore also directly affects the output ripple and noise of the power supply. Lower ESR = cap being closer to an "ideal" capacitor, and therefore filtering better and reducing the ripple and noise on the PSU output maximally. But this is only true to a point, because in real life circuits, extremely low ESR can also be a problem for PSU! This comes from the fact that any inductors in the PSU will form a resonant L-C circuit with the filtering capacitors, and that can cause "ringing" and oscillations on the PSU output, thus making it possible to end up with more ripple and noise (in addition to also upsetting the voltage feedback / monitoring circuitry.) And this is why polymer capacitors should not be used for recapping power supplies, unless the power supply was actually designed to be filtered by ultra-low ESR capacitors.
So when it comes to ESR, it's a good idea to keep the ESR of the replacement capacitors close to what the original caps were rated for.
In my experience, most power supplies will be OK with +/-50% change in the ESR (i.e. 1.5x lower or higher ESR.) But I don't suggest going past 5x higher ESR or 2-3x lower ESR - that's usually the tolerable limit for most power supplies.

Last edited by momaka; 01-19-2022 at 12:20 AM..
momaka is offline   Reply With Quote