what is the best way to test electrolytic capacitors in circuit?

[EDWARDEFFECT1]

anyone steer me in the best way to check electrolytic capacitors in circuit?esr meter,oscilloscope,capacitance meter,voltage readings,ect.what type of pattern would you look for with an oscilloscope?bypass caps,filter caps,coupling caps.any help appreciated.what is a better capacitor meter the sprague to6a or the sencore c55..what do you reccomend for a meter or testing?or should i say what do you do and use??...thanks....ed!!

[Enzo]

Like any component, I test it by the way it does - or does not do - its job in the circuit.   Does a filter cap leave too much ripple?  Does it not hold up under load?  Does a coupling cap let through DC?  Does a coupler seem to not pass signal?   Does a tone shaping cap not shape tone as it should?

In the repair business, someone is paying me a dollar a minute to work on his stuff.   SOme shops charge twice that even.   How many minutes should I spend testing a 38 cent capacitor?  If I think a cap is bad, I just replace it.   Parts are a lot cheaper than labor.

[EDWARDEFFECT1]

hi enzo could you be more specific what you use for test equipment for these tests.example.ripple(oscilloscope??)bypass capacitor passing dc.coupling cap (multimeter???)also if using an oscilloscope do you put it on ac or dc.just wondering if you could give me the lowdown.thanks! enzo and all others on this site for being able to help us all out with our troubleshooting.i see there are alot of others looking at this post and i think they have the same question ,but are afraid to ask...thanks for all your help everyone.....this is the best site on solid state amps on the net!!! love it.....ed!!!

[DJPhil]

I saw your request in another thread, so I thought I'd try adding more here.

For all caps-
If it's bulging, burnt, or leaking goo, then it's bad. Always remove and replace such caps, as they're dead or dying and just waiting for the right time to take a bunch of stuff with them!

Most multimeters will be able to tell you if a cap has failed as a short circuit. The larger the cap the lower it's resistance when it's good, but it should always be at least a couple megaohms (1000KΩ). For example, I just measured some 2200uF 35V electrolytics that are brand new, and they measure 4.3M/37M depending on the polarity of the ohmmeter. I then measured several small 10nF (0.01uF) polyester caps and they're so high they don't register on my meter (40M max). A cap that's failed short will probably be 1K or less, perhaps even just an ohm or so.

The actual capacitance value is usually pretty unimportant unless the cap is part of a (very, very high) precision filter. Capacitor tolerances are horrid, rarely better than 5% in small values and soaring to -40%/+100% in most electrolytics. Guitar tone stack filters usually use caps with a tolerance of 10%, which is pretty standard for non-electrolytic caps in the nF to uF range. This sloppy tolerance stuff is worrisome, but don't lose any sleep over it. Consider that everything you've ever used uses caps with these tolerances, and it all (probably) works ok! This means that you can usually use a multimeter cap tester to see if the cap is anywhere near it's listed tolerance. I use a lot of salvaged parts and my rule is to toss any cap that's more than about 30% off, but I do test all salvage parts before I assemble them into something. Even for relatively precise filters it's more important that two caps are close to each other than close to a fixed value.

It's really tricky to test the value of a cap in circuit, I'd avoid that if at all possible because you really have to know about the rest of the circuit to be sure you're not getting a bad reading (which is very likely to nearly certain for most situations). What Enzo described above is really the best way, see if the cap is doing it's job when it's working with the rest of the circuit. Keep in mind that it's dangerous to test a live circuit, so please use caution and ask if you're not sure about the proper safety procedures. No one here will think less of anyone asking about safety, and we'd all be glad to point anyone asking to good resources.

Example: AC coupling or DC blocking cap
The purpose of one of these is to remove the DC offset of a signal before it's sent to the next stage (see here for details). Using a scope you probe the 'input' side and see a DC offset, which might be very small. Probing the 'output' side should show a reduction in the DC offset, usually setting it to zero. If you're still getting a DC offset on the output side then the cap's not doing it's job, time for a replacement! If you've grounded your probe and you get very little or no signal on the 'output' side then it's possible that the cap's failed open.

Example: Power Supply Filter cap
These are for shaping the 'bouncing ball' rectified AC from the transformer into a flat DC supply (see this ST appnote for details). Keep in mind that the goal is usually not to completely eradicate ripple, but reduce it to a manageable level. Load is a factor as well, and you may have to test a power supply under load to see the worst case ripple in the supply. It's possible to calculate the expected ripple, though a rough guess is usually good enough. The best way to see ripple on a supply is with a scope (just probe the rail AC coupled) but if you're familiar with your multimeter (i.e. it's true RMS or you can do the necessary correction) you can get a fair idea of the magnitude of the ripple by testing the AC voltage. Anything over a couple of volts is probably suspect, but it really depends on the supply and the load. If you're seeing more than five volts of ripple then it's fairly likely that at least one filter cap is bad.

As Enzo says, it's usually done this way because it's faster and caps are relatively cheap. There are other factors with caps that come into play when designing circuits, like ESR, but I'd argue that it's not worth worrying too much about if you're simply fixing and not modifying or creating. An aged electrolytic with high ESR will be slow to charge and discharge and behave poorly as a filter cap, but you don't need an ESR tester to see this if you can probe for ripple.

I'm no pro, but I hope that helps some. Caps are a confusing mess sometimes, but at least they're not inductors.  :duh

[EDWARDEFFECT1]

thanks fot the reply.i have tested capacitors with a esr meter.cap tester,oscilloscope ,and an ohm meter.i tested a bunch of capacitors in a rolland gr20 they all looked good with a capacitance meter and esr meter.i got a reply from j m fahey and he said to send a sine wave into the analog part of the circuit and trace back from the output till i found a good signal.i did this and found a bad cap that previously tested good with all the other ways.thats why i posted this topic.thanks for your help and concerns....ed!!

[EDWARDEFFECT1]

just fixed a  hp vs17e  computer monitor.it had 3 bad electrolytics in the power supply(monitor was dead). the only way i could tell is they were bulged on their tops.they tested good out of circuit with a $200.00 esr meter.i took a regular capacitance meter and 2 read about 309uf and 1 measured 820uf.i'm guessing out of circuit a capacitance meter gives a pretty good indication if an electrolytic is bad.this was my first monitor repair.guess it's pretty hard to know for sure whats bad and what good in circuit...later...ed

[EDWARDEFFECT1]

one for the gipper! for all you that want to test electrolytic capacitors out of circuit,but don't have money to spend on a expensive cap meter you can make your own cheap.take a 9 volt battery connector and solder a set of alligator leads or ic connectors; whatever you choice may be.then hook a 9 volt battery to the connector.connect the battery leads to the electrolytic cap observing polarity.now hook a dc volt meter across the capacitor also observing polarity again.the meter will read the 9 volt battery voltage.then disconnect one of the battery alligator clips and the voltage will drop off. if the voltage drops to zero in a second or 2 the cap is bad.the voltage should bleed of slowly.this does the same thing as many high dollar cap testers.my sencore cap tester does the same thing.you can make one for abot a buck, so why not make a couple......later....ed

[Enzo]

That will test basic capacitor function, but you cannot rely on it.   Older caps might test fine at 10, 20, 40, 80 volts and leak like a screen door at 200v, to say nothing of 450v.

Of course if they don;t work at 9v they won;t probably work at higher voltages either.

[Jack1962]

Caps are cheap if you think a cap is possible bad replace it , like Enzo said when someone is paying you to repair a amp or anything else , it is cheaper for you and the customer to just replace any cap that you may think is defective , if a cap is leaking tar , bulging or has any other visable defect replace it , it may work now , but that cap will go out.

                                                       Rock On

[EDWARDEFFECT1]

just learned from a tv repair friend of mine.to test for leaky and sorted capacitors.he put my ohm meter on the lowest scale and we read 2.2 ohm on a capacitor in a peavey classic 30, the amp melted 2 resistors due to the shorted cap allowing the power to go to ground.pretty cool. we took it out of circuit and sure enough it was bad...just thought i'd add this note to this subject....ed

[joecool85]

just learned from a tv repair friend of mine.to test for leaky and sorted capacitors.he put my ohm meter on the lowest scale and we read 2.2 ohm on a capacitor in a peavey classic 30, the amp melted 2 resistors due to the shorted cap allowing the power to go to ground.pretty cool. we took it out of circuit and sure enough it was bad...just thought i'd add this note to this subject....ed

You're going to want to make sure to drain the cap you're testing first.  Voltage on a DMM when on a resistance setting can fry it.