3ph Motor Earth Fault

If you have a leakage that is polarity sensitive this is most probably due to copper oxide acting as a semiconductor diode, before the days of silicon and selenium diodes copper oxide diodes were used as rectifiers but were not very efficient.

From what you say the leakage is very small, I suggest you run the motor suitably grounded on a fused circuit for a few hours until it gets good and hot and dried out then check the leakage resistance with a high voltage 'megger' and provided it is above say 1M ohm put it in service.

The leakage may well be in a terminal block and not the winding in which case this could be be replaced rather than a rewind.

.

thankyou for your explanation syhprum, the poarity sensitive fault was the confusing part for me.

bill

Syhprum is absolutely correct about the copper oxide asking as a semiconductor under certain circumstances. I have seen some really strange thins happen because of corroded copper on connections.

I would suggest that you check all the terminals, clean them, make sure there is no contaminants and check it again before running it for a while. If the fault is still there see if you can get hold of a Component_Tester and test the capacitance between and to ground of each winding then check the inductance of each winding, it may give you a better idea of what is going on.

Component_Testers aren't super expensive and can be almost as useful as a millimeter. I bought one, the same as the one in the link, about 10 years ago to diagnose a specific problem and afterwards wondered why I didn't buy one sooner.

Yep - oxidation in strange places always leads to interesting troubleshooting.

Don't know why, but the link Masu provided didn't work for me here in the US, though a search for a component tester at the source site gives me this url (minus some server referral coding riffraff) you may try:
http://www.rsaustralia.com/cgi-bin/bv/rswww/searchBrowseAction.do?D=component%20tester&N=0&forwardingPage=line&R=0219642

And for our US browsers, they may wish to Google or check out B&K's similar 815 unit - or look here: http://www.testequipmentdepot.com/b+k%20precision/815.htm

Have fun!

Sorry about the link folks, I'm not quiet sure why it doesn't work but the link Sandman has provided is to the page I was trying to link to so alls well that ends well. The second component tester in Sandman's post isn't quiet as good as the first one because it can't test inductors and transistors like the first one but then again it's only half the price.

Even so they are well worth the expenditure and believe me once you have one you will wonder how you survived without one. They can be particularly handy when using cheap transistors that have a wide variation in forward bias current gain, you just plug them into the test port and you get the gain to three decimal places.

It does not matter which way you connect your test leads, the result should be the same. Ignoring this "phenomena" - a 1.3 Megohm reading is not a fault; any resistance to earth above 1Megohm is quite OK to power up - assuming you are connecting up to 440 volts.. Obviously if it is a high voltage motor you will need to be testing with a better instrument than a multimeter such as a 1KV or better Megger.

is any standard available for earth leakage (ex: for home or for office or for hospital etc., ) measurement or all are same.