Has anyone a PM Routine for this task

Hi,
Although we used to monitor partial discharge (via a pm) the results were not repeatable. There were so many variables regarding the use of the equipment that, depending who was using the equipment, where they were standing, what component(s) were being looked at, how the kit was set up, etc etc was dependant on the type of results generated. The results themselves were very subjective and open to mis interpretation. However, maybe this is because of the type of kit we used, which was ultrasonic and relied on the user identifying discharge through a set of headphones. Consequently after many attempts at getting the parameters right, it was deemed better if the kit was put in a darkened room and forgotten about!! Maybe others have more successful reports of the use of partial discharge monitoring, but my experience was all negative.

We have several large very critical transformers. We have continuous monitoring of a composite combustible gas using a Syprotec Hydran oil monitor, we have quarterly oil samples for complete dissolved gas analysis as well as moisture, dielectric etc. And we do Doble testing (power factor, TTR, bushing watts loss etc), every 18 months.

Like UK Vibeguy, we do no partial discharge testing on transformers (we do it on large motors using Iris equipment and have had good luck).

Hello David,
there are lots of people all over the world routinely inspecting their equipment. I know that SDT have several application notes on this and include it in some detail in their Level 1 Ultrasound course. The other organisation with a huge amount of experience in the field is EA Technology Limited.

Drop me a line if you need more specifics.
tom@reliabilityteam.com

Tom:

Most of the ultrasonic testing for PD (actually, corona) and arcing on oil filled transformers actually relates to the bushings and not PD and arcing within the transformer itself. The most accurate method is still dissolved gasses.

Howard

[QUOTE]Originally posted by electricpete:
We have several large very critical transformers. We have continuous monitoring of a composite combustible gas using a Syprotec Hydran oil monitor, we have quarterly oil samples for complete dissolved gas analysis as well as moisture, dielectric etc. And we do Doble testing (power factor, TTR, bushing watts loss etc), every 18 months.

Hi, we also do the above tests. But Do you have a good software to help you analyse these test results. Resently a Bushing on a 66kV TX blew up and our inhouse investigation found that capacitance dissipation factor tests on that bushing had been getting bad over a period of time and nobody noticed it untill the accident.We have now concluded that we need a packaage that can store all condition monitoring results and help trending the same results to identify a patern for analysis. find attached pictures of the bushing.

Partial Discharge monitoring for transformers is still in its infancy stage. There are some technologies that are promising but nothing really that could be used in a regular, “route” based way. A few years ago there was a test conducted in the UK involving 14 or so “best PD monitoring technologies” from around the world. The results were that there was noting available on the market that could be widely used, though there were a number of promising technologies. One of them used submersed ultrasonic and electromagnetic sensors for full on-line, real time monitoring. It was quite successful in locating two faults within the transformer to +/- one feet. Very impressive. It was abandoned if I am not mistaken. The one that you can get right now is through the capacitive couplers that are installed on bushings. How good it is? I question the results though there are many who are believers. I do believe in partial discharge monitoring but right now for transformers use what others mentioned – oil analysis and dobble testing. Both tests should give you sufficient information related to transformer condition.

There are really two methods to monitor PD on a transformer. Acoustical and Electrical. These are really complementary technologies. Acoustical can be used to triangulate and localize a discharge. However, the sensitivity is quite low. The defect would need to be quite developed and close to the surface of the winding. If it is deeper in the winding, it will not be detected. Electrical is far more sensitive, but can be subject to external noise from corona, etc near the bushing terminals. use of good noise cancellation techniques can remedy most of the noise concerns and detect down to 50 - 75 pC, which is quite adequate for field measurements.

However, using electrical you can not localize the discharge. Typically it can be determined if it is "In the top half of the HV A Phase coil"

You can visit my website for further info. www.elec-di.com