Medium Voltage test for cables

[a10jp]

I am looking for standards which specify the test volatge and duration for DC hipot test. This is for the 6.6kV cables that we have installed. Which standard specify these? Also, I know for some manufacturers that they use 80% of the dielectric, and is able to reduce the test duration to 10min if the leakages stablizes quickly, which I believe should be the case most of the time. Again, where in the standard did they specify this?

 

[dpc]

I think you're going to hear that field dc hi-pot testing of cables is no longer recommended by IEEE.

http://www.electricalreliability.com/Literature_PDFs/PuttingHipotOuttoPastureArticle.pdf

 

[Marmite]

In the UK the standard is IEC298.
For a system voltage of 6600V the test voltages would be 10kV ph-ph, 6kV Ph-E. Test duration 5 mins.
If the cable is screened you would only test to earth. If it's a mix of screened and belted along the route, test as though it's screened.
The test voltage is derived as follows:-
Ph-E: System Voltage x sqrt 2 x 1.1 / sqrt 3
Ph-Ph: System Voltage x sqrt 2 x 1.1
The leakage current of a healthy cable should be less than 0.2mA, although it could be greater for long lengths or depending on the termination arrangement.
Regards
Marmite

 

[oldfieldguy]

IEEE and NETA might both be pushing the partial discharge test, but the hipot test is far from dead, and done correctly by skilled personnel using proper techniques, it provides useful information for determining cable condition. I have seen it used for exactly that purpose with good result.

However, if the technique consists of grabbing a semi-capable operator and letting him use the old "crank up the voltage and write down a number" technique, the data won't be worth looking at and quite possibly a cable will be reduced from serviceable to scrap in short time.

The partial discharge test for cables has yet to gain a wide acceptance in the areas in which I work. The few testing contractors who've purchased the equipment want to charge premium rates and my former clients who paid for the testing were not impressed, citing inconclusive results and bing at the mercy of the contractor technician for interpretation.

Incidentally, NETA still includes AC and DC hipot test voltages in their testing specifications. Unfortunately, I'm on the wrong computer to access this data.

old field guy

 

[Zogzog]

i dont think NETA is pushing the PD test, it is only as good as the data you already have. VLF and Tan Delta seem to be the best route.

For 8kV cables the NETA ATS DC test voltages are 36kV/44kV (100%/133% Insulation). Of course you need to do it right with the required # of steps and corona suppression to make the data worthwhile, like old field guy stated.

 

[dpc]

Here are Okonite's recommendation for field testing of cable in service for LESS than five years.

Note that this only applies to EPR insulation. XLPE should NOT be dc hi-pot tested, per manufacturer's recommendations.

http://www.okonite.com/engineering/high-voltage-testing.html

 

[alehman]

Last year I inquired directly with Okonite's engineering group. They still recommended DC hipot for new cables.

This topic is very controversial. It has been discussed at length in previous threads...

thread238-125873
thread238-102550

 

[oldfieldguy]

Insuring that I use the term "Properly Applied" up front, a hipot test will not hurt new cable in good condition, i.e, undamaged, stress cones properly applied.

It will not hurt older cable in good condition.

In both cases the test will give information suitable for trending and analysis.

This all depends, of course, on the the qualifier I first mentioned. A comprehensive and rigorously applied test regimen applied by knowledgeable technicians is essential.

old field guy

 

[benlanz]

Here we go again.

We need to be very careful about making recommendations before we understand the application.

a10jp,

What is your cable's insulation type? Paper Oil, EPR, XLPE?
What is the age of your cable. I think I understand that they are new cables?


For example:

Both DC and Tan delta are useless on new EPR and XLPE.

However, a proper DC test can be useful on new or old PILC.

If the cable is aged Tan delta is a reasonable chose but, there are many caveats.









Benjamin Lanz
Vice Chair of IEEE 400
Sr. Application Engineer
IMCORP- Power Cable Reliability Consultants

 

[a10jp]

These are Japanese cables, distribution class 6.6KV, and have the equivalence of XLPE insulation, that were developed to be environmentally friendly. Anyhow, I have also checked the Okonite website, and got the info on HV testing. For XLPE cables, dc tests are not recommended because it can destroy the insulation. Then what test should be used? In Japan, after some research in their testing standards, the AC test voltage for upto 6900V is 1.5 times which gives 10,350V. Then the dc test voltage is 2 times the ac test voltage which gives 20,700V. I thought that was too high but the Okonite webiste shows 20kV for 5kV voltage also. I am not experienced in this field, as such I am trying to look up published standard where these voltages are specified in the US. And the sole purpose of this test is to make sure the dielectric strength of the insulation is intact after installation. There are also splices that connect the new cables to the existing (9~10 years old) feeding the facility. I am sure this will limit the practical voltage to use in this situation.

 

[alehman]

benlanz has a good point. Okonite only makes EPR insulated cable. Their recommendations would not apply to XLPE or any other materials.

 

[benlanz]

a10jp,

There is good news and bad news.

The good news:
you can apply whatever DC voltage you want with in reason (4Uo) and will not harm the new cable.

The bad news:
you may inflict damage on the the aged portion of the cable.
you are not likely to find any defects even if they exist in an extruded system because DC is blind to even gross defects.

OPTION 1 If DC is your only option:
1) hire a manufacturer’s representative to inspect each component installation including the conductor crimp, semiconductor cut back, and installing the cable itself.

2) perform a jacket test to make sure the jacket is not punctured. This will give you the confidence that the cable is most likely in the same condition it was in the factory.

3) perform a 5kV test for as long as it takes to charge up the cable to make sure that a short condition does not exist.


OPTION 2
In addition to the above, if a AC withstand test is available this would be much more meaningful than a DC test. A very low frequency (VLF) withstand is likely to be your cheapest option. Go to at least 2.0 Uo for 30minutes.

OPTION 3
The best option would be to perform an off-line PD diagnostic acceptance test with 5pC sensitivity to 2.5Uo and measure the accessories response at 1.5Uo for 5 seconds. I don’t know if this available in Japan, but I could probably find out if you really need help.




Benjamin Lanz
Vice Chair of IEEE 400
Sr. Application Engineer
IMCORP- Power Cable Reliability Consultants

 

[a10jp]

Hi Benlanz, how does a PD diagnostic test work? I am unfamilair with this testing. So any info you you have will be very helpful. According to the Japanese Electrical code (or Law) here, the majority of the testing uses only AC test on the dielectric insulation, that is regardless whether it is EPR or XLPE... The test method started off with 1000, and then 3000 for leakage current stability, then for a full 10350V AC for 10min. 10 min is standard for AC test in Japan. However, somehow I felt that they might have been blindsighted on any other test methods that can be considered. AC test is all that they have.

 

[StineIng]

Hi

I have to do some test on just installed 6 kV PEX cables.

I have read about the Series Resonant Test, where you put a reactor at the end at the cable, and then adjust the frequency to the circuit go into resonans.

But are these test preformed on HV only, or are the test used on MV too?

- Stine

 

[benlanz]

a10j,

Your test measures the leakage current of an AC test? I have heard of this in laboratory but, it is very rare in my experience.

How a PD test works... That is short conversation but, a long written explanation, depending on what level you want to discuss the topic. The following is an overview from 10,000ft.

PD (a small micro arcing inside insulating material) is associated with the final breakdown mechanism of more than 99% of all extruded dielectric systems. This is why manufacturers have been using PD tests to prove the integrity of extruded dielectric components for the last 40 years. An AC voltage stress is applied to the system and if there is a defect it says ‘ouch’ in the form of PD activity at the defect site. If your detection system is sensitive enough (proven through a sensitivity assessment), the PD signal response can be measured and located. Once he defect is located you can determine if the defect is in the cable, joint, or termination and apply the appropriate component standard (IEEE 404 for joints, IEEE 48 for termination, IEEE 386 for separable connectors, and ICEA S-93-639 for MV cable insulation) to determine if it passes the standard.

StineIng,

For MV class (5k - 35kV) cable system tests we typically use a series resonant transformer (200-500kVA). HV class (69 -400kV) tests we typically use a frequency resonant transformers (1000kVA - 22MVA).

Regards,


Benjamin Lanz
Vice Chair of IEEE 400
Sr. Application Engineer
IMCORP- Power Cable Reliability Consultants