1000KVA Rotor Winding IR Test

[Andrew247]

Good day,

I have a question about an IR Test (insulation resistance) on the rotor of a 1000KVA Leroy Somer LSA 50.1 alternator. I use the Megger MIT1525 for the IR test and I'm doing a rotor to earth insulation test.

As I know from the past when performing resistance tests on other alternators & field windings , the resistance to earth should be very high (M/Ohm) or even infinity (if brand new) between any winding and earth. In the case of my Megger MIT1525, when conducting a test such as this, a rising test voltage is introduced across the tested points for a 60 second period (0VAC to 1000VAC), and at the end of the test period it will indicate to what voltage the winding can perform, breakdown resistance, and what the general resistance is between the windings and earth.

In my case, when conducting the IR test on the suspect rotor's disconnected leads and the shaft (earth), I get a warning beep and an indication on the meter display of a fault after barely initiating the 60 second test, maybe 5 seconds in, as soon as the test voltage reaches around 15VAC, that does not rise again until the conclusion of the test, and the final insulation resistance is very low and does not even register on the meter display (minimum reading is 100 Kilo Ohm)

So, my question is: Is the rotor shorting to earth and do I need to have the alternator refurbished/replaced?

At the moment I'm second guessing myself and don't want to make an incorrect judgement call.

 

[waross]

Did you disconnect the rotating diodes and surge suppression?


Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[Andrew247]

Yes sure! I'm only testing between the rotor windings and earth. The two rotor windings have been completely disconnected and in air. I'm testing between the disconnected windings and earth (the alternator shaft)

 

[edison123]

Is the rotor winding relatively clean and dry? A dirty, dusty, oily and moist winding will show low IR value. Cleaning and drying out will normally improve the IR.

Muthu

http://www.edison.co.in

 

[flexoprinting]

Sounds to me like you have a problem, I would second guess myself as well. I would recheck the Megger on a known good winding different motor or simply leave the leads open and check for infinity. Try a simple ohm test see if you get a reading. Still in doubt borrow a Megger from a local sparky shop.

Chuck

 

[thermionic1]

I once was commissioning a 345kV, 300MVA transformer. The first test I did was core ground megger. With my 500V megger, I had similar results as you have on your alternator. I then used my Fluke 87 and measured the same. I believe I got in the neighborhood of 25 Ohms. I later found that the impact recorders were not checked and once inspected, indicated a large impact somewhere in shipping.

 

[waross]

When the first electronic meggers came out, we didn't trust them. Early electronic Voltmeters and Ammeters had some real problems. Some were almost unusable for trouble shooting.
My first electronic megger had a meg-ohm range resistor taped to it and I would check it before each use.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[Andrew247]

OK,
Thanks everyone. I'm wrapping this up now. I've made my decision. The alternator is coming out for replacement.

I have used two additional Meggers to test the rotor and the conclusion is that the rotor is shorting down to earth as voltage is introduced. I used another old style crank Megger on 1000Volt setting as well as a similar electronic one with the push button. All three Meggers confirm insulation breakdown to earth.

Additional info: I can put this alternator back together and run it, and it will give me a perfect 400VAC/50HZ (R450M AVR), without load no problem. It's output goes through an 11KV step-up transformer, through an HT vacuum breaker onto a common 11KV bus in a substation, along with another 7 other identical machines. As soon as the Woodward GCP20 synchronizes and closes the HT breaker and gradually loads up the alternator, power factor would start to drop from an initial .99 down to around .16, load will be shed and the generator would shut down, indicating power failure. This is caused by the rotor starting to short to earth when load is introduced, and higher forces is exerted onto it.

 

[dArsonval]

It was never clear to the reader if the apparatus was in a dismantled state during the tests.
[Rotor completely removed from field frame/stator portion of the apparatus.]

From the above "Additional info", It is stated that, "... I can put this alternator back together..."
which suggests the machine was indeed tested while dismantled.

But the above may be implying 'I can reconnect the rotor' etc.

Can the OP more clearly describe if the rotor was completely removed from
the rest of the apparatus during testing?

The reason I'm asking is because an assembled machine can have multiple paths
to ground giving an erroneous test.

If the rotor was tested by itself, completely removed from the machine then it is certainly
a candidate for repair.

If the machine has not been fully dismantled, you may discover upon re-testing the rotor
that it is not grounded, and the initial fault detected... exists somewhere else.

John