Testing capacitors with a 'Megger'
 

Hi everyone, does anyone else test capacitors with a Megger insulation tester? I have a dedicated capacitor meter that gives me the value, an ESR in circuit cap checker which tells me that the ESR is within tolerance, but I have found that by far the best tool I have for indicating that a capacitor is servicable or not is the Megger insulation tester. Mine tests at 250, 500 and 1000volts and when set to the value closest to the capacitor rating indicates UNDER LOAD whether or not there is any leakage. A modern new component never shows any leakage (in excess of 200 Meg ohms) but those horrible waxies regularly show leakage sometimes as low as 50kohm or lower! Its certainly handy for checking NOS components before going to the trouble of installing them only to find that they are duff. Your observations are welcome
Bill

Re: Testing capacitors with a 'Megger'
 

I've used a Megger in series with an AVO to good effect. Start on the 1A range whilst cranking, then reduce to the uA range to check for leakage. Maybe it would be better to insert a series limiting resistor if the cap suddenly went short, but I've got away with it thus far... Disconnect the cap smartish or use a diode, though, or the cap will discharge through the meter the other way!

I sometimes check larger caps with an AVO in series with a limiting lamp or resistor (if using a transformer derived supply instead of mains). The voltage across the cap and the current through it can be measured and hence the capacitance. I've known, odd times, a variation of readings when compared to sticking the cap on the bridge or puff-ometer, and I feel that sometimes it's better to check as near its working voltage as poss.

Re: Testing capacitors with a 'Megger'
 

As much as I would like to use one of my (many) vintage insulation testers for this job I always go for a digital one because it frees up an extra limb for wielding test prods etc. Also the output voltage is stable enough that you can compute the leakage current accurately from the displayed resistance. And as Bill says you get a choice of test voltages without having to change instruments. Although if I want to test at 2.5kV I'm back to cranking the Big Megger, or at 10kV when I have to plug the Very Big Megger into the mains and stand well clear. But yes, any sort of Megger is the tool for the job.

I picked up an interesting instrument at Gerry's garden party this year that might be useful for capacitor testing. This is a high voltage, ultra-high resistance meter using valve circuitry, capable of applying a test voltage between 10 and 1000V, i.e. a laboratory-grade megger. I can't remember what the highest range is but it didn't seem possible using conventional electronic stuff mounted on tagstrips, PVC wiring, sockets without guard rings etc. 10^12Ω comes to mind. I will have to take a closer look and see if it is worth getting working. It was made for the National Physical Laboratory so it's unlikely to be just wishful thinking.

Lucien

Re: Testing capacitors with a 'Megger'
 

I might have added: 'or pressing...' Our 'works' Meggers are P/B (apart from the 250V Series II bridge machine) but my own is a 250-500-1000V switchable hand-cranker.

We also test HV caps on a HV megger, up to 12kV, which has appropriate leakage instrumentation fitted. You are correct about cranking, though - it can become a bit of a handful!

Re: Testing capacitors with a 'Megger'
 

Yes, it's inportant to measure paper capacitor leakage at the working voltage because what I have found is that they do not behave like resistors.

The leakage current increases with voltage faster than ohms law would calculate so if you just measure with an AVO on its ohms range you will get a very optimistic result.

The other factor that makes a huge difference is temperature so if you really want to know then warm them up a little. This is also why you can get run-away problems once a little leakage starts.

Re: Testing capacitors with a 'Megger'
 

Thanks so far for the interesting replies. The 'Megger' I use is an import but of excellent quality, its not the old 'hand crank' version its very much akin to a modern multimeter, has the usual Vac, Vdc, resistance, continuity and the powered HV test ranges at 250,500 and 1000volts. I have found it absolutely invaluable and it certainly does find leaky caps and you would be surprised just how leaky some of them are! I mainly restore old radios and test equipment and I really would not be without it. I welcome all observations,
Bill

Re: Testing capacitors with a 'Megger'
 

I guess any old NE555-or transistor multivibrator derived / stepped-up / rectified and controlled voltage source would do (plus an AVO or similar, of course), but the Megger is probably handiest. And although preferring 'hand-crankers' myself, the P/B digital versions obviate the need for an external microammeter.

Re: Testing capacitors with a 'Megger'
 

I use a megger but find it always claims electrolytics leak, even new ones. These get tested via a diode, light bulb and AVO.

It's surprising how good a guide my home-made continuity tester is though.

If nothing else were available an HT battery and ammeter would be better than just a straight ohmmeter. 3V button cells are now so stupidly cheap - a hundred for six quid - and as the current required is minute one could easily and simply be made out of those just for this purpose.

Whatever method is used seems to be a waste of time as so few suspicious capacitors pass and if they've been disconnected to test I feel I may as well replace them anyway. In the past I've left apparently good ones alone only to have them fail within a few weeks.

- Joe

Re: Testing capacitors with a 'Megger'
 

Hi Joe, yes I quite agree, those electrolytics are a different matter, they always show a leak, because they do leak! It may only be a few microamps if you're lucky. For electrolytics, I rely on the capacitor checker to ensure its upto tolerance and then the ESR to make sure it tests good. If they havent been charged up for a while I'll then run them up on the variac over half an hour or so and monitor them closely for a while. If they show any sign of distress, warming or allow too much ripple I dump them or refill them depending on type and age of the set.
Bill

Re: Testing capacitors with a 'Megger'
 

I also use a hand-cranked megger for paper capacitor checks. It delivers 500V if you run it at full speed but a very gentle turn or two is all that is normally needed to show the paper cap is beyond hope. An attraction of hand-cranked machines is you can control the power output.

Re: Testing capacitors with a 'Megger'
 

Up to a point, yes, but they have a 'slipping clutch' to maintain a stable output above a certain speed. The biggest atraction to my mind is they require no batteries. Nothing worse than having to grub about for handfuls of AA cells when a job needs doing :(

Re: Testing capacitors with a 'Megger'
 

I use a Taylor Model 130C insulation tester, this has 250 and 500 volt ranges and a "Push to Charge" button and reads up to 1000 Megaohms and is mains powered. I would guess it dates from the sixties.
Most waxed paper caps are bad, measuring between 50k and 1M, American types are sometimes better than UK ones. The Sprague metal can types from just after WW2 are better, these seem to be around 50Mohms which is OK in most circuit positions. Modern types after charging read infinity.
Regards, Mike.

Re: Testing capacitors with a 'Megger'
 

Interesting replies thanks to everyone. One word of warning though for anyone not conversant with this method of checking leaky caps, DON'T FORGET TO DISCHARGE THE CAPACITOR YOU HAVE JUST TESTED! It will have 500v or so across its terminals ready to bite!
Bill

Re: Testing capacitors with a 'Megger'
 

I use a 250 volt Megger to test waxies etc. for leakage. You can even use them for re forming electrolytics.

Re: Testing capacitors with a 'Megger'
 

This is the Megger I use, a BM8 Mk2. Tests at 50, 100, 250, 500 and 1000V.

Re: Testing capacitors with a 'Megger'
 

I've just bought some 0.47mF 630 volt caps and have access to a Megger BM21 tester, at work, so I'll give them a go. I've also ordered some PIO caps 0.5mF 500 volts, so they'll get the same test.

Re: Testing capacitors with a 'Megger'
 

Hi Richard, if the caps are new you shouldn't have a problem, they will be fine. If 'new old stock' check them and you may be surprised how bad they are. Don't forget to discharge them when you've tested them. Leaky old caps will discharge themselves after a short while but modern ones will retain a charge for a long time. Its the best way of testing them.
Bill

Re: Testing capacitors with a 'Megger'
 

Thanks Bill.
The Megger BM21 takes them up to the set voltage and back down again slowly.
Well, I'm impressed. They all measure over 6Gig ohms except one, which is just over 1Gig, all measured at 500volts. (They are rated at 630v) Capacitance varies between devices, lowest value is 0.448mF, highest is 0.471mF. I'm building some new amps based on the Mullard 5-20 so the components MUST be very good but I'm not willing to pay Audiofool prices. Most of the components will be Russian Military origin on Ebay.
I also have loads of Allen Bradley "Morganite" Hi-Stab 1% resistors of all sorts of values and ratings. These, too, will have to be tested.

Re: Testing capacitors with a 'Megger'
 

Just a silly question I guess but is a megger a multimeter?

Re: Testing capacitors with a 'Megger'
 

No, its a meter designed to measure very high resistances, many megohms, using a much higher voltage than say an AVO8, hence the name 'megger', and although looking like a traditional multimeter it works on a different principle, but rather than trying to describe it in more detail there is this link,

http://www.richardsradios.co.uk/megger.html

which does a far better job than I could do, this shows the traditional hand driven type, which I still use, the latest ones are all electronic.
I've never used a modern one, but with the hand cranked ones, they always looks a bit strange because when not being used, the needle just stops anywhere on the scale and gives the appearance of being broken.

John