PD testing - preferences and references 1

[Skogsgurra]

Hello,

Partial Discharge testing in repair shops for motors and transformers is done in several different ways. Mostly RF based, but also Ultrasound and perhaps (never seen that) using ozone detection.

This is a broad question and all having any experience with PD testing and evaluation is invited to share their views and practises.

A few questions to start with:

1 When do you use PD testing?

2 For what objects?

3 At what levels?

4 What detectors do you use?

5 How do you evaluate your results?

6 Tips, tricks, traps?



Gunnar Englund

http://www.gke.org

 

[electricpete]

We do partial discharge testing quarterly (1) for thirty motors ranging in size from 2500hp to 9000hp (2) all 13.8 kv (3). We use Iris Power Engineering bus coupler capacitors (4). We trend the 10pps cutoff level "Qm" and evaluate the phase pattern in this trends up (5). The test is an on-line test and we find it very useful. We found one item which we didn't correct and it progressed to failure; we found one item and inspected and found severely oil-degraded endwinding jumpers; and we had one motor fail to ground that never showed any symptoms. Also we have been monitoring for a few years two motors that have tracking two that have surface partial discharge on the interface between semi-con and grading treatments. This seems to be a fairly benign partial discharge and we gain confidence from the fact that we can see changes in behavior. We don't use this on our 4kv motors as it is generally believed that detectable partial discharge only occurs right before failure at this voltage level (much less warning time).

We have on occasion requested off-line partial discharge tests from our repair shops after rewind.

Our repair shops also use TVA probe to check for partial discharge activity in the slots.

I have used a Drager sampler with disposable tube to check for ozone on our motor's air exhausts in cases where they had high partial discharge indicated by the Iris equipment. Never saw enough to be detectable.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

Here you will find technical papers and case studies

http://www.irispower.com/newsite/index.aspx

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[motorspert]

We have used PD Tech in Switzerland. We jhave fitted them to machines all over the world successfully. They have a good range of sensors amd monitors. Its up to the user to decide whether to have continuous or peiodic monitiring, online of offline. All down to how much you want to spend!

http://www.pdtech.ch/index1.php

 

[eglaude]

I use the Iris capacitive couplers as well, on 4kV motors ranging from 5000HP to 10,000HP. The on-line test done by Iris is simple, and the data gathered is quite extensive. I'm not quite sure however the value of this yet, as the data is very subjective, and based on typical trends from the Iris database.
To say that these tests ever saved one of my motors, I can't, but I believe that there is some value in PD testing motors in the long run. Question is, if the ability to get a unit out of service once every 4 to 8 years, do you trust a +/-Qm plot or do you just get it serviced?
electricpete: Do you have any info to support the value, or lack of, PD testing on 4 kV motors? I always thought that 4kV was a valid testing potential?

 

[electricuwe]

PD testing is used (or should be used)as a test after manufacturing or repair on systems containing solid organic isolation that are subjected to voltages exceeding several 100 V.

e.g.
Cast resin transformers
potential transformers
motors
pulse transformers for thyristor firing
transformers within IGBT-drivers

The PD test level is only slightly higher than the voltage during operation, but PD testing a device which has to withstand 1000 V with 1200 V provides much more reliability that doing highpot-test at several kV.

For details check the standards (e.g. IEC).

 

[electricpete]

eleceric - The Iris folks say that the warning of time-to-failure on 4kv machines is much lower.. perhaps a matter of
hours or days weeks after pd pattern is detected before failure is imminent (because pd does not occur at that voltage until insulation is very close to fully breached).

In contrast for 13.8kv machines you can have months or years.

So to be effective at the lower level, you need continuous monitoring/alarm (maybe that's what you have) in order to have time to react.

At higher voltage level you can do the walkaround checks and still get good benefit.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[ScottyUK]

Skogs,

You might like to explore this website a little. The man who runs it is a one-man-band consultant who specialises in large generating plant. He's done a lot of research into PD measurement on these large machines. And he's also a nice guy on top of that.

http://www.exon1.demon.co.uk/main.htm

----------------------------------

I don't suffer from insanity. I enjoy it...

 

[eglaude]

pete:

Thanks for the post. Currently we do not do continuous monitoring, but that is something to look into for the future. On my 5kV motors, I try to do semi-annual checks, winter and summer to get some idea of temperature variations on the equipment.