Cable Sheath Fault

[Neilie01]

Any ideas folks any feedback would be appreciated?

Our Construction Dept. has had several cable sheath faults over the past few weeks. See attachment.
They are on the feeders which connect 11Kv RMU's.
The cables are new, the size is 3c x 300mm* CU/XLPE/SWA/PVC with graphite coating and they are buried.
They have been tested at 10Kv DC between armour and coating which I think may be causing the problem. I believe it should not be greater than 5Kv /1 min.
The other factor which may/may not have a bearing is the test set is powered via 10kva gen set.

Cheers...

 

[cuky2000]

The following information should be helpful to know:
- What standard waw used to perform the HiPot test? Any deviation from the approved standard practice?
- Is this a maintenance test or new installation acceptance test?. Any Preliminary testing step such as ‘Megger’ test. Any low reding?
- Any unusual weather or site conditions such as raining, high moisture, underground water?
- Confirm that prior the test all cable shield, sheath and all conductors, except the one to be tested were grounded (earthed).
- There is any indication of cable damage or mechanical overstreess during installation such as exceeding cable bending radii, over pulling or any questionable workmanship practice?
- Any reel test was performed on the cable?
- There is any cable production test available?

 

[7anoter4]

A direct voltage of 4 kV/mm of specified thickness of extruded oversheath shall be applied with a maximum of 10 kV
between metallic sheath and outer electrode (moist backfill or a graphite layer) for a period of 1 min to demonstrate
the integrity of the cable oversheath after laying (Clause 5 of IEC 60229).
So, if the jacket failed during this test it has to be repaired using a sort of shrinking tube, for instance. The cable manufacturer has to be
informed in order to improve the sheath extrusion-usually by checking the extruder temperature controllers.

 

[dpmac]

Why is a 10KVA power supply required to power the DC Hipot tester. Most standard 0-80kVdc) DC hipots are about 600VA. Small Thumper/burners would be 3-5KVA or so. Its possible your testing at too high a voltage and burning the fault down rather then tripping a hipot off (5MA-10MA max output). Check with the cable manufacturer for proper test voltage, duration and maximum leakage current allowed.

 

[ausphil]

at this voltage, we generally only perform an insulation resistance measurement between the metallic screen of the cable to earth. This should give you a reasonable indication as to the condition of the sheath. It is only when going to the higher system voltages that we will perform hv tests on the outer sheath.

if you are performing hv tests, you should be performing insulation resistance measurements before and after the test, to determine if you have caused any damage from the test voltage being applied.

10 kV seems excessive for test voltage on sheaths on that voltage cable. The IEC 60299 reference is good, but you need to ensure from the manufacturer that the oversheath was manufactured to this standard. If not, then you need to ask the manufacturer what tests you can do, and what (if any) test voltage to use. You can't just take the after laying tests from this standard without the manufacturer designing the sheaths to the standard, and performing the type and routine tests in the factory (which is considerably higher than what should happen in the field after laying).

on the test set size, personally, I like a test set that will produce a decent amount of damage at the point of fault (assuming it is testsed at the correct test voltage as above) because the smaller the test set fault current, the harder the fault is to find once you have one.

ausphil

 

[Neilie01]

Hi Guys

Thank you for all your comments, they are much appreciated and reinforced my thoughts..
Another factor has came to light...
The contractor has used the services of a sub-contractor 'specialist' to help with the location of the fault/s, so it looks like the third party's test gear may be the origin of the track trees in the attachment.

The cable was manufactured in KSA for which we have excellent FATs apart from the sheath test. We await conformation from them if they carried out FAT on the sheath.

The 10kva genset was used because it was the only convenient power supply available.

I cannot confirm what tests were done previously as I was only asked to take a look when all else had failed. Presently when tested between the conductive outer layer of the sheath and the SWA at 5kV, the result is 20K ohms.

I will post and let you know the outcome of this debachle.


Cheers Neil

 

[waross]

If new cables are failing under test, I would suggest either bad quality and the cables are not up to spec, or the test voltage is too high for the cable as specified.
Another possibility may be damage during installation. After all, one reason for post installation testing is to verify that the cable was not damaged during installation. As an example, the installation crews may for some reason be bending the cables at less than the recommended minimum radius during installation resulting in hidden damage and multiple failures. If you have access to the work site, look around for any small radius cable pulling sheaves that may have been misapplied.

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

 

[7anoter4]

Following waross idea, also it is possible, for this large diameter cable[90 mm] the Manufacturer extrusion line pay-off drum diameter is too small.

 

[7anoter4]

Sorry,I meant take-up drum!

 

[Neilie01]

All OK now and we energised this circuit Thursday past.
Construction had to repair several faults so I think they will be more carefull in the future.
We never found out what caused the 'tree tracks' and are unlikely to now.
Thanks for all your helpfull comments.
Cheers Neil