Should we test connector @rated voltage AND current (8kv, 200A)? 1

[ramjet]

Hi,
I'm a Mechanical Engineer at a company where make feedthru connector systems to run 3-phase submersible pump motors in oil wells. Our standard product is rated at 5kv and up to 210A. Typically we do dc hi-pot testing (1 minute @24kvdc, 5 microamps max leak) to validate the voltage rating, and also low voltage, no load, closed circuit heat rise testing to validate the current rating. With new upcoming applications requiring 8kv we are now concerned with PD and we're in the process of purchasing partial discharge test equipment. When that comes we'll need to figure out some acceptable initiation and extinction values, but my question right now is regarding a proposal to outsource full load testing. Do you guys think it is of value to test the connector system with a load so that it's at rated current and voltage simultaniously or is that no different than testing voltage and current independently as we currently do?

 

[ScottyUK]

The only reason why you'd need to do a full voltage / full current test would be to prove (or disprove) any thermally-induced effects arising from the passage of current which affect the behaviour of the insulation when under voltage stress.

If you really want or need to do a combined high current / high voltage test then I don't think it would be unreasonable to drive an HV current transformer in reverse as the source of test current and float the CT 'primary' (high current side) at whatever proof voltage you desired. That way you're using relatively low energy sources to develop high current and high voltage without needing an MV substation and a fairly large load bank to achieve it. That said, it would be a test rig with enormous potential to kill the unwary and would have to be very carefully designed to ensure safety.


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

If we learn from our mistakes I'm getting a great education!

 

[ramjet]

Thanks ScottyUK. I have no intention of setting up this test here, we would outsource it to someone who can do it safely. If I understand your response correctly, there would be nothing to learn from a full load test that you couldn't also learn from a hi-pot test performed at an elevated ambient temp equal to that which would result from max rated current (in this case 300F). Is that right?

That's a test we could run for less cost than a full load test.

 

[ScottyUK]

That's my off-the-cuff opinion, yes. You might need to test certain specific conditions - for example cold water spraying over the connector housing while the contact assembly is at 90°C or something similar which could be hard to reproduce in a test chamber without passing current through the connector. Frankly any test involving 200A, 8kV and water all in one place would make me nervous!

I might be overlooking something obvious, but I can't think what. No doubt one of our peers will point it out to me soon!


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

If we learn from our mistakes I'm getting a great education!

 

[electricpete]

I have a hard time coming up with any credible scenario's that aren't checked when testing separately (other than Scotty's thought about temperature affecting insulation).

It seems more of a power question than a motor question to me though.

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

 

[ramjet]

Thanks electricpete, I'll try posting in the Electric Power forum.

 

[waross]

Scotty said pretty much what I was going to say. In regards to current injection testing versus elevated ambient temperature, I once (note only once in a long career) had a faulty exciter that checked out fine at room temperature. It would fail under load but check out fine by the time I was able to get it to a winding shop. When heated in a bake oven it tested fine. As it turned out, the relative thermal expansion of the windings versus the core was such that the windings shorted when the exciter was heated internally from load current but not when it was heated externally in the bake oven. I devised a test that allowed me to test the windings in situ when in failure mode. When I described my test to the tech in the motor shop, his response was that despite what his tests in the oven may have indicated I had definitely proven a failing rotor condition.
It probably won't make a difference in your instance but inadvertent poor connections may perform differently depending on whether the heat is applied externally or internally.
BTW, Using CTs to test at high current while floating at high voltage may not be much more hazardous than testing at high voltage.
And, testing at a full 200 Amps and 8kV with a load bank may end up with the kWHr charges being a major portion of the testing expense.

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

 

[ScottyUK]

"Using CTs to test at high current while floating at high voltage may not be much more hazardous than testing at high voltage."

I agree Bill, but the electricity bill will be much lower, say a 10kW load compared to 1.6MW. Lower arc flash hazard too.


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

If we learn from our mistakes I'm getting a great education!

 

[waross]

I'm agreeing with you Scotty.
Yous
Bill

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