VLF Dead break terminations - 21.1kv test voltage recommendations

[tcoupekyle]

We are out on a solar farm testing 35kv cable with Dead break terminations. The termination is 21.1kv rated but 35kv class. Cable is 35kv xlpe cable.
Neta says test at 44kv VLF for 35kv cable but there seems to be a gray area in the termination.
Any experience or recommendation on test voltage? Or test procedure?
When we try to test at 44kv vlf the set is arcing across the body.
So the plan as of now is to test at 22kv for 15 minutes and step the voltage to 44kv after that 15 minutes.

[SecondGen]

We are out on a solar farm testing 35kv cable with Dead break terminations. The termination is 21.1kv rated but 35kv class. Cable is 35kv xlpe cable.
Neta says test at 44kv VLF for 35kv cable but there seems to be a gray area in the termination.
Any experience or recommendation on test voltage? Or test procedure?
When we try to test at 44kv vlf the set is arcing across the body.
So the plan as of now is to test at 22kv for 15 minutes and step the voltage to 44kv after that 15 minutes.

You may need to reduce the voltage to match the rating of the termination. Where do the cables land? If they connect to a switch that you can isolate you might be able to get some more voltage with the cables terminated.

[tcoupekyle]

You may need to reduce the voltage to match the rating of the termination. Where do the cables land? If they connect to a switch that you can isolate you might be able to get some more voltage with the cables terminated.

So we brought that issue to engineering that the term is only rated for 21.1kv and at first they said test at 26kv. But then (after testing) they said no they wanted to re-test at 44kv and they couldn't find any literature to back up the test voltage one way or the other. So they decided to re-test at 44kv.

The cables terminate on an inverter so we can't jump through.
We are using a dead break connecting cap to input the voltage into the cable. And dead end caps screwed together on the other side.

We have gotten the cable to test in the last hour. We covered the exposed end of the connecting cap with 130c tape. It's also hotter and drier now this morning was 99 percent humidity.
It's probably 95% humidity now. So somewhere between the tape on the connecting cap and the humidity drop we have been able to get a complete test on 1 single conductor.

I am still curious about others opinion on the test voltage though.

[tcoupekyle]

Trying to attach picture

[SecondGen]

Trying to attach picture

That is a sweet looking setup, I'd still be worried about overstressing the term but it sounds like that burden is on engineering now. Is this a new install? Obviously cables should have been tested at 44kV prior to terminating and then re-tested at the term voltage but that's not possible now.

[tcoupekyle]

Yes. You're exactly right. New installation, acceptance testing.

Cable were not tested prior to the termination being built.

And yes, we recommended not testing at full voltage because the term is the weak link in the chain. But got over turned after the fact. So yes the burden of liability is on the engineers in the office.

[tcoupekyle]

So today we came back out and continue testing. The first cable we've tried to test today fails at 44kv within the first minute.
No progress as of today.

[TestTech]

So today we came back out and continue testing. The first cable we've tried to test today fails at 44kv within the first minute.
No progress as of today.

Your technically Testing a "Half a Splice" connected to your cable. Why would you expect it to hold the voltage?

Why not simply finish making the splice correctly, add some cable and test away?

See attached (I hope) picture (Stolen from CBS Field Services post on Linkedin without permission)

It shows a 200A elbow but the concept is the same.

[tcoupekyle]

Hey! Yeah that's what we decided to do build another and a stress cone to try it.
That's exactly what we been looking for. Is that how you normally test these?
I've always just used the half splice and hadn't had issue until now. But maybe in the past it was all lower voltage. In Maintenance situations.

[Kalbi_Rob]

So we brought that issue to engineering that the term is only rated for 21.1kv and at first they said test at 26kv. But then (after testing) they said no they wanted to re-test at 44kv and they couldn't find any literature to back up the test voltage one way or the other. So they decided to re-test at 44kv.

The cables terminate on an inverter so we can't jump through.
We are using a dead break connecting cap to input the voltage into the cable. And dead end caps screwed together on the other side.

We have gotten the cable to test in the last hour. We covered the exposed end of the connecting cap with 130c tape. It's also hotter and drier now this morning was 99 percent humidity.
It's probably 95% humidity now. So somewhere between the tape on the connecting cap and the humidity drop we have been able to get a complete test on 1 single conductor.

I am still curious about others opinion on the test voltage though.

35KV terms are actually rated maximum phase-to-ground 21.1 KV and stamped on the side of the termination. If you read the literature, you will find they are actually rated for up to 38kV class operating voltages, see link:

https://www.eaton.com/content/dam/ea...ca650008en.pdf

I agree with the above comments, since it looks like your trying to test a cable using pass-thru bushings with insufficient insulation and corona suppression (that bushing is technically uninsulated unless installed inside the t-body). This would be possible at lower voltages (still not recommended), but at higher voltages, you must fully insulate the test specimen and suppress the corona. Damage to cables comes from discharge of voltage, which means your non-destructive test has now become a potential destructive test due to insufficient tools. This is common practice, since management typically doesn't want to spend the money to make a adequate test rig, typically or just lack the knowledge to properly test cables.