Megger resistance threshold for 5KV shielded substation feeder cables

[powerpete69]

At what resistance level should 5 KV shielded cables be replaced when meggering?

I've read some things that say if you get below 1500 mega-ohms on any phase, replace cable.

I've read other things they say you can push it to 100 mega-ohms?

I've also read you need 1 mega ohm per thousand volts. So in this case, 5 Mega Ohms should be okay.

Personally, I've seen cable short circuit at 800 Mega ohms.

My actual cable in question has the following megger measurements from phase to ground:

Phase A: 1.8 Giga ohms
Phase B: 40 Mega-ohms
Phase C: 90 Mega-ohms

To me, this is a replace cable ASAP situation.

Also, they do VLF/Tan delta tests as well which requires 7,000 volts compared to 1,000 volts for megger test.
My electrician skipped the VLF/Tan delta test because he said we would likely blow up a cable with less than 100 mega-ohms on it if you put 7,000 volts across it.

Anyways, does anyone have any solid info on this from their experience?

 

[winnie]

I don't have experience with HV cables, but have some experience from the electric motor world.

Do you only have single numbers, or do you have historic values and can look at a trend?

The reason is that you can have a surprising amount of leakage in a perfectly functional system, especially with older insulation materials. So if you have an older system that has been functioning perfectly well at 5 mega-ohm (with a bunch of leakage), that low insulation resistance isn't an indication of impending failure. On the other hand, if you have a system built from modern insulation materials that started with an insulation resistance in the giga-ohm range that suddenly measures in the 5 mega-ohm range, that is an indication of a large change and a problem.

For the system you measured, is that a single 3 phase cable or 3 separate cables forming a single circuit? If the latter was phase A replaced?

IMHO if A, B, and C were supposed to be the same (eg a single 3 phase cable) then the difference in values suggest something is going on. Maybe not in the cable itself, but in the terminations. If I saw those numbers on a motor I'd investigate further.

 

[powerpete69]

Well, when we megger test the big 500 MCM 4160 Volt cables, they are completely disconnected and isolated when testing. In other words, any loads are left out of the equation on purpose.

Back in 2022, cable had these readings:
A phase: 1.1 Giga ohm
B phase: 90 Mega ohm
C phase : 14 giga ohm

A-phase got better
B and C phase dropped considerably from these measurements back in 2022. Especially C phase.

All that being said, when pulling out cable, they always cut the splices and megger both ways. If a section of wire is good, they leave it and professionally splice it to the new one. Then re-megger and VLF/tan delta the entire run.