Insulation tester / megger to test capacitor leakage

[mvno_subscriber]

Hello,

I'm looking around for an insulation tester / megohmmeter to, amongst other things, test capacitors and transformers. Especially, capacitor leakage current.

My question is, is there anything in particular I should be aware of? I try to find meters with high resistance readings and a good selection of voltages, but is there anything else that's important? And do you have any good recommendations within a reasonable budget ($100-$200)?

Buying used is fine.

[Gyro]

You can often pick up used  'Electrician grade' insulation testers for less than that amount. A good example is the Kewtech KT35 which will test at 250/500/1000V at up to 2G . They include a voltage scale too which shows the remaining voltage (decay) after test. They also include high resolution continuity test with lead resistance nulling They are good for testing low cost imported SMPS chargers etc and general PAT testing.

Important warning: It is potentially very dangerous leakage testing capacitors of any size. Although the tester current is limited to 1-2mA, the capacitor can store significant (potentially lethal) charge.


EDIT: Check out this recent thread too... https://www.eevblog.com/forum/testgear/vc60b-megger-for-wire-insulation-testing-decent/

[threephase]

I will second the comment from Gyro regarding charging capacitors with an insulation tester and the stored energy that will be present. Now, the vast majority of modern insulation testers will contain a post test discharge circuit, however the discharge rate will vary a lot across the instruments. Some instruments like those from Sonel have quite a high resistance in the discharge circuit but will not display the reading until that voltage is discharged below the safety level, which is usually 50V. They cannot also lockout completely if the voltage is not discharged within a specified time frame.

I will add an attachment of the insulation testers I have reviewed that shows the discharge circuit resistances.

A lot will depend upon exactly what you are testing and what you want to achieve. Capacitors are very voltage sensitive and if you use too high a test voltage can be easily damaged. The vast majority of insulation testers will have fixed test voltages, common ones are 50V, 100V, 250V, 500V and 1000V. If you have a 400V capacitor, then the 1000V range will definitely damage it, the 500V range might be ok, but you are risking it a bit. The capacitor will survive the 250V range, but you will not be testing it at its working voltage. It will depend whether that aspect matters to you or not. This could be overcome with a tester that has a variable test voltage, such as the Megger 400 series, but they are usually more expensive. The only one I know of with a variable test voltage in your price range would be the Uni-T UT505A.

This raises another point in that a lot of insulation testers will have reduced resistance ranges at the lower voltages. The Kewtech KT35 is slightly different in that it appears to have 2 GOhm across the range of test voltages, a lot of instruments will only get into the GOhm on the 1000V range, and will drop to circa 550 MOhm at 500V, 255 MOhm at 250V and so on, so I would look carefully at the range specifications in the data sheets and not just the general manufacturer's brochure statements. You also suggested measuring leakage currents, not all insulation testers have this function and not all will display the actual voltage being applied either. You may therefore be looking at using a uA meter as well. The IEC standard, specifies that insulation testers should be able to drive into a 2uF load and remain stable, so if you are testing capacitors above this, it may take a while to charge them and get a stable reading.

I will add a further couple of attachments of the meters I have tested, that list their functionality and test ranges, that may be helpful to you.

Some of my blogs of the insulation testers are on here, where they are not they will be on element14, but if you want to know about a specific instrument I have tested and cannot find it, let me know and I will point you in the direction of the blog.

[mvno_subscriber]

@threephase, thank you for this invaluable information! I should elaborate a bit on how I currently test leakage current - I simply hook up the capacitor to my lab supply, and add a multimeter on the µA range in series. However, my supply won't go higher than 30V, and the whole setup can be a bit cumbersome.

I guess I could add some sort of voltage multiplier and be done with it, however I don't consider that particularily safe. I have, multiple times, forgot to properly discharge the capacitor, and 30V does cause some sparks. 250V... I don't have the intelligence to manage that in this manner.

I also hope this insulation tester can be used for other purposes as well, for example testing for damaged insulation in transformers.

The UT505A really seems like a good unit, but the resolution would be limited to about 1µA. I guess that's ok, based on the available voltages. Variable voltage is also a very, very big bonus. However, I see your tested discharge resistance is 8.3Kohm - does it just discharge a set amount of time, or actually check that the voltage is low enough? I have never used an insulation tester, so all of this is new to me. I'll find that review now.

Thank you!

[threephase]

The vast majority of insulation testers will have the discharge resistance applied across the terminals when an insulation test is not being carried out. So as soon as you switch the function switch to insulation test, the discharge resistance is present. When you press the insulation test button, the discharge resistance is removed and then the instrument applies the test voltage. When the test is finished, the test voltage is switched off and the discharge resistance applied and it will stay that way until the next test is started or the function switch moved to a different position. Most testers will alarm until the voltage goes below its safety limit, usually between 30 and 50V, although the discharge resistance will still remain connected after the alarm has gone.

On most insulation testers, the test button is a momentary style, so press it in and hold to keep the voltage nom as son as you let go it switches off. Some have a lock function to keep the test voltage applied until it is stopped manually, and some have a timed function, where you can set the test to run for a fixed duration, anything from 5 seconds to 999 seconds, some even longer.

The insulation tester can be used for testing cables, motors, transformers. Those that have a continuity function can be used to test the resistance of connections, wires and windings. The handheld ones up to 1kV output usually have a short circuit capability between 1 to 2 mA, so are not particularly powerful, but the uncontrolled release of a stored charge can obviously create more significant issues, so I guess an insulation tester in that respect would offer an improvement over your current setup.

[mvno_subscriber]

Closed shops and nothing to do easily gets the best of you in these times.. so I got an offer for a UT-505B at the same price as an A, and went for it. The B not only has 1Gohm at 50V, but also comes with remote control probe and other niceties. Hopefully it will address the shortcomings you pointed out in your UT505A review. It's perhaps overkill for my needs right now, but experience tells me one day I'll be very thankful for having upped my "investment".

Thanks for all the help!

[threephase]

You're welcome. If the UT505B would have been available to me, I would have opted for it over the UT505A.

Hope it serves you well.

[mvno_subscriber]

Just a quick afterthought here - I see it being said that measuring leakage current and leakage resistance are two different things. But if I know the leakage resistance at a given voltage, wouldn't I know the leakage current as well? Or how do these two concepts differ?

[Gyro]

Yes, you can apply ohms law for a given voltage. I think the difference is that leakage current is often non-linear with applied voltage - it certainly is by the time you get near breakdown, any corona discharge effects etc.

[threephase]

For the purposes of your tests, given how small the circuit is and what it is made up from, I would say that the leakage current will be directly related to the insulation resistance using Ohm's law.

In terms of DC insulation testing, an insulation tester displays the total current in the form of a resistance value. This total current comprises of capacitive, absorption and leakage currents. If you apply a DC test voltage for long enough, the capacitive and absorption currents decay down to zero and you are left with just the leakage current. This is the purpose of the polarisation index (PI) test and timed insulation testing in general.

How long you need to apply the test voltage for is dependent upon what it is you are actually testing and the type of insulation and ambient conditions. It may take several minutes on motor windings and cables. The larger generators I test can take up to an hour.

I presume it will only take a few seconds for your test to stabilise as the capacitor charges up, unless you are using a series resistance and are testing particularly large capacitors.