Best method(s) for testing PS caps?

[clm811]

I have several large electrolytic caps I would like to try out for power supplies(for power amps), but want to verify their good condition before using, to avoid any problems(explosion, etc).

My capacitance meter only goes up to 2,000 ufd, so it's not good for this.

What are the easiest (and best) ways to check/test large electrolytic caps.

Thanks in advance!

-Chas

[nigelwright7557]

Originally Posted by clm811

I have several large electrolytic caps I would like to try out for power supplies(for power amps), but want to verify their good condition before using, to avoid any problems(explosion, etc).

My capacitance meter only goes up to 2,000 ufd, so it's not good for this.

What are the easiest (and best) ways to check/test large electrolytic caps.

Thanks in advance!

-Chas

Charge them up, then discharge through a resistor while watching on a scope the decay time. From that you can work out the capacitance.

[clm811]

Charge them up, then discharge through a resistor while watching on a scope the decay time.

Any other methods not requiring an oscilloscope?

[brig001]

My old moving coil meter had notes in the manual for how to estimate capacitance using the ohms range. Can't remember if you noted peak needle position or timed it...

Would certainly give you a way to compare some (discharge first).

HTH,
Bri.

[SY]

Voltmeter and stopwatch; using the estimated capacitance, set the time constant for about a minute, charge the caps up, then monitor the decay.

But that's only part of the story. You'll really want to check leakage current- if it's significant, the caps will be less effective, run warm, and die young.

[h_a]

Originally Posted by clm811

to avoid any problems(explosion, etc).

Electrolytics only explode when voltage polarity is reversed (and exceeds 1.6V). When properly used, no need to worry.

Have fun, Hannes

[cbdb]

For leakage current, charge the cap and see how long it takes for the voltage to drop with no load. The ESR is also important for power caps but more complicated to measure.

[kevinkr]

Originally Posted by h_a

Electrolytics only explode when voltage polarity is reversed (and exceeds 1.6V). When properly used, no need to worry.

Have fun, Hannes

Unfortunately, that is absolutely NOT true in my direct experience. I have had more than one do this, and in all cases the capacitor was operated well within its ratings, including voltage, ripple current, and ambient temperature. Not a huge number over the years, but enough to not summarily dismiss it as an impossibility..

The most egregious of these was a brand new CDE brand axial type, it was "violently" defective out of the box. This was about 25yrs ago, running at 65% of rated voltage, room temperature and no ripple. It physically destroyed a channel in my just completed mosfet power amplifier project - I smelled something, but by then it was way too late to prevent what happened next. The capacitor exploded, blowing the end out and destroying everything in its path - the mess extended some several feet beyond the amplifier and bench to the walls and floors of the lab where I was working. IIRC this was a 22uF/35V part with about 24V across it. (Who would have thought such a small cap could make such a big mess, but it did.. Cap had orange jacket and was about 1.25cm wide x 3.7cm long.)

It is always a good idea to test leakage current if you are unsure.

Brand new caps can be defective as received, and older caps can and do explode usually at some point (hopefully well) beyond what was considered useful end of life by the manufacturer. Ever wonder why modern caps are designed to vent "safely" - it is because these sorts of failures are statistically common over the life of the part.

[clm811]

It is always a good idea to test leakage current if you are unsure.

Did a quick search on leakage current. lots of posts, but nothing detailing test procedure(s).

Can anyone suggest a quick and easy way to test leakage current, with no test equipment other than a voltmeter(and transformer with rectifier, of course)?

[SY]

Put a resistor in series with the cap, then apply voltage. Measure the voltage drop across the resistor and that will give you leakage current. You want the applied voltage to be somewhere near the max voltage rating of the caps, and size the resistor to something like 1k, which will give you 1V per milliamp of leakage.