If I put 200V across a string of two 100k resistors, then put an electrolytic capacitor across one of the resistors and wait for the cap to charge. If the voltage across it is 95V, does this mean that the leakage current is equivalent to that of a 1M resistor, and can I assume for all intents and purposes that this cap has such a resistor across it?
Yes, but due to the chemical nature of these capacitors, leakage current depends on things such as temperature and aging, and may not follow a linear relationship to applied voltage. In other words, don't rely on a particular leakage current value for electrolytic capacitors. Furthermore, leakage current changes also from unit to unit.
Hi,
Since becoming a member I have learned a few things about caps.
The leakage you have found @ 200V (about 50uA) will probably be at least trebled by the time you raise the applied voltage to 400Vdc.
If the cap was suffering excessive leakage then you need to reform it to at least it's final working voltage plus a margin for voltage variations. Otherwise heating due to excessive leakage will be cumulative with ripple and other heating effects.
Finally, the change in ESR seems to be the best guage of life remaining in the cap. About a doubling from the spec value is a guide to end of useful life. I suspect that you could get some half decent performance when ESR is double and maybe a bit further.
Some manufacturers test the ESR soon after a quick version of reforming. There appears to be some kind of relation here.
I hope I have got this right.
Since becoming a member I have learned a few things about caps.
The leakage you have found @ 200V (about 50uA) will probably be at least trebled by the time you raise the applied voltage to 400Vdc.
If the cap was suffering excessive leakage then you need to reform it to at least it's final working voltage plus a margin for voltage variations. Otherwise heating due to excessive leakage will be cumulative with ripple and other heating effects.
Finally, the change in ESR seems to be the best guage of life remaining in the cap. About a doubling from the spec value is a guide to end of useful life. I suspect that you could get some half decent performance when ESR is double and maybe a bit further.
Some manufacturers test the ESR soon after a quick version of reforming. There appears to be some kind of relation here.
I hope I have got this right.
Hi,
the leakage is often quoted as 0.01CV or 0.02CV
You can calculate this @ 0.02*180*450*10^-6=1620uA=1.6mA
Can someone jump in and confirm that the C in the formula is Farads?
I have checked a few lower voltage caps, I have reformed, and they often leak at less than 5% of the formula projection.
the leakage is often quoted as 0.01CV or 0.02CV
You can calculate this @ 0.02*180*450*10^-6=1620uA=1.6mA
Can someone jump in and confirm that the C in the formula is Farads?
I have checked a few lower voltage caps, I have reformed, and they often leak at less than 5% of the formula projection.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.