Need help with a measuring problem

[jefffay]

Hello all. First post, total newbie.... I have a maintenance manual that gives the following instructions to check a capacitor:

C. Test Capacitor.
Test the electrical properties of the capacitor using the GenRad Digibridge 1157 component analyzer and a megohmmeter, or equivalent test equipment.

1. Capacitance value should be measured at room temperature and charged to 400 ± 10 VDC. The service limit of the capacitor is .315 to .385 microfarad.

2. Test the insulation resistance of the capacitor using a megohmmeter. The resistance measured between the capacitor stud and shell should be 10 Megohms minimum at 135 ± 5 VDC.

What is the best way to accomplish the above and what equipment is necessary? My assumption is that the GenRad system would not be optimum.....

Additional info: I currently own an oscilloscope and several multimeters. The capacitor is removed from the circuit.

Any info would be most welcome. Thanks.

[amspire]

1. Capacitance value should be measured at room temperature and charged to 400 ± 10 VDC. The service limit of the capacitor is .315 to .385 microfarad.

2. Test the insulation resistance of the capacitor using a megohmmeter. The resistance measured between the capacitor stud and shell should be 10 Megohms minimum at 135 ± 5 VDC.[/i][/color]

Technically this is probably right, but no one would check an electrolytic by charging to full voltage.

Make sure it is discharged and measure the capacitance if you have a capacitance range on your multimeter or a LCR meter.

If you don't, no problem. Get a 9V battery and a 10K resistor and measure the amount of time to charge the capacitor to 6V.  400uF equals about 4 seconds. If it is about 3 seconds or lower, get a new capacitor. You do not need much accuracy in measuring electrolytic caps.

You can probably use the multimeter to check the leakage resistance. start with the shell and stud shorted, remove the short and let the meter rise. If it rises over 10MOhms, it is probably OK.

If you have to replace the capacitor, the replacement will probably be a different construction and much smaller. Once you find a way to mount the replacement, you will probably never have to worry about shell to stud leakage again. If you do need to replace, you will probably be getting a 470uF/400VW cap.

[KJDS]

pssst, it's a 330n, or 0.33uF, not 330u

Whilst the factory method may be to test the cap at or above full voltage, it's not necessary, just use an ESR meter

[G7PSK]

I dont know what that capacitor is but is the case one of the connections? I had a micro wave capacitor go short to ground it was not detectable with a DVM but showed up with a megger or when you tried to run the oven it would blow the fuse or throw the trips. The cap tested fine with an esr meter and DVM but as I have already said at high voltage it would break down, so the HV test recommendation might well be for a good reason, if you are in doubt about such a cap just change it.

[rf-loop]

Sometimes DC bias is very important factor.

If it reads in service manual it is good to believe it. Peoples who write service manuals mostly understand best why something is important in some equipment what they have designed.

[amspire]

pssst, it's a 330n, or 0.33uF, not 330u

Whilst the factory method may be to test the cap at or above full voltage, it's not necessary, just use an ESR meter

You are right. If they need to write that much about a single component, just replace it. If the manufacturer doesn't trust the part will work, why should you?

Perhaps it is a paper capacitor in which case definitely replace it. It would be very rare to see a modern capacitor with a resistance of 10MOhms, but it is not that strange for some antique capacitors.

How old is the equipment?

[jefffay]

Thanks for all the replies and input. I should have probably provided additional info....

The capacitor that I need to test is one of the components in a magneto (old school engine ignition system). The test of which I speak should be accomplished every 500hrs of in service time. The replacement component is close to $150 so changing it out regularly is not in the cards.

Let's see if I am smart enough to attach an image...... If not here is an external picture: http://www.skygeek.com/unison-slick-k3984-kit-capacitor.html

How would I charge it to 400v if that was indeed deemed necessary? Also, I noticed that many of the meters out there have narrow slits, presumably to insert the conductors of a typical capacitor. Do the leads do the same thing?

If anyone has any equipment suggestions I'd be much obliged.

[Rufus]

Let's see if I am smart enough to attach an image...... If not here is an external picture: http://www.skygeek.com/unison-slick-k3984-kit-capacitor.html

When a faulty capacitor results aircraft engine failure testing it properly is advisable.

To test properly you are going to need a high voltage adjustable source, very little power is required. Don't know if you can buy anything cheap, you could build a dc-dc converter. With a variable ac source (variac or perhaps signal generator and audio amp) you could use a mains transformer backwards (or two in series) to get a high voltage ac adjustable source to rectify.

Leakage is easy, just charge to 135v with a large resistor in series and measure the voltage across the resistor to determine the capacitor leakage.

Capacitance is harder. If you have a proper meter that measures capacitance with ac stimulus then you can put a larger known capacitance in series with the meter to block the dc bias and measure the series capacitance of the two from which the unknown capacitance can be calculated. Another less accurate but probably valid method would be to discharge the capacitor from 400v through a known resistance using your scope to measure the discharge time from which capacitance can be calculated.

It would probably be helpful to buy a quality 330n 650v capacitor to play with and for sanity checking your methods.

[SeanB]

So this is a paper capacitor for an aircraft engine, so it has to work and needs to be tested to the required high voltage.  The high voltage is easy, look for an insulation tester that does have a 400V range ( most will be probably 500V so close enough) and use it to measure the leakage current ( insulation withstand resistance) of the capacitor under test. Then discharge for around 10 minutes with shorted terminals ( these are rather notorious for having stored charge) then measure capacitance with a multimeter. You may have to use one which has capacitance measurement via the probe leads ( choose one of the Fluke units that do so and which are probably in use in your shop) or one of the multimeters reviewed here. This will simply be clipped to the connections of the capacitor and the value read on the display.

For an ignition capacitor it will have to operate at high voltage, at high temperature and operate reliably over a large temperature, pressure and vibration range, so it has to be very reliable, and this needs to be tested so that you can detect failing units before they degrade to the point of failure well ahead of time.

[KJDS]

If it's on an aircraft engine, get someone else qualified and insured to test it.

[G7PSK]

If you have to have a certified aircraft engineer test that capacitor it is probably better to just change it for a new one, as if it fails the test you need a new one and I expect that a ticketed aircraft engineer will charge quite a bit just to run the test.

[willflyer]

Hi Jefffay,

I understand precisely the basis of your query concerning the testing of certain magneto capacitors.  I, too, am grappling with this issue and I would be very interested to know whether you have made any progress.  All the relevant kit seems to hark back to the dark ages!

I would greatly appreciate your help.  More in hope than expectation ...

Cheers,  Willie