Cable termination

[TMD]

When terminating a medium voltage cable, (15 and 35 kV, EPR), any damage (cuts, nicks, slices) to the EPR requires that the cable be cut back and you start again. However, in some situations, this may not be possible due to limited cable length.

Has anyone ever tried to repair damaged insulation? How did you determine if it was a good repair? Seems to me a fix may work for awhile, but is not a long term solution.

On a related note, is there a way to test stress control sleeves after they are installed? We've used PD testing in a test lab to check complete assemblies, but what about field terminations?

(Threads 238-42135, 18808 and 6889 are relevant)

 

[dede61]

Hi TMD,

In case the cable gets to short for termination we mostly cutback for a few meters (if possible, but it usely is) and with a joint attach a new lenght of identical cable (that is long enough of course). If possible the additonal lenght of cable is pulled back a bit to give some spare in case the assembly of the termination fails (again ?).

regards,

Danny

 

[advidana]

I never heard of anyone fixing a damage high voltage cable. Best bet is to replace the cable or splicing it.

 

[DanDel]

Acceptance testing using an overpotential(HiPot) test set is typical for MV cable installations. This will verify the integrity of the stress terminations.

 

[TMD]

In our experience, it is a very delicate operation to terminate a 35 kV cable and then have it pass a PD test at less than 10 pC (Raychem HVT kits). Given field conditions, it must be extremely difficult to do a "perfect" termination. Hence the reason for a test that will identify a poor termination.

Some of the other threads I listed in my first post try to identify a pass/fail criteria for PD testing, but no definitive answer.

Trying to repair a termination will likely just lead to wasted time and another wasted termination kit. When you can, chop the cable off and start again afresh.

 

[DanDel]

benlanz, I'm sure you consider a 3 month operation at 2x rated voltage something that should fail an improperly made stress cone, but it is certainly not a Hipot test.
The standard Acceptance test voltage for a cable Hipot test is between 4 and 5x the rated voltage, which is reached in successive incremental steps, and held for 15 minutes. The leakage is measured at each step, and excessive leakage, or a non-linear increase in the rate of leakage is enough to indicate that the something is wrong with the cable installation. The cable doesn't have to 'fail' during the test; in fact, the idea is to discover the problem and repair it before it goes into service.
My company also performs PD testing, but this is recommended for in-service cable only. Hipot testing is still the best for Acceptance testing of new cable.

 

[benlanz]

DanDel,

Your point is well taken. A careful measure of DC leakage current can certainly tell you something about the cable. I am quite familiar with this technique. I happen to be the vice chair of the IEEE 400 field guide to power cable testing.

Consider this scenario
What if you get a 'bad cable' reading on a new cable. What are you going to do then? Replace the entire phase? Chances are there is defect in one of the joints or terminations. Which one? What if you get a so-so reading, what then? What if your cable has mixed dielectrics what does lossy reading mean? It would require in excess of 1000pC to create enough charge for the most sensitive ammeter to detect a defect in an accessory. IEEE specifications require the accessories pass a 5pC PD test by design. Why not use it in the field?

DC was great 30 years ago when it was the only option, but now with advances in PD technology I believe it is a far better choice. PD is the only technology that allows you to locate defects!!

PD technology is so advanced that utilities are now using fault location crews to perform highly successful PD cable testing!!
Please refer to the following article

http://www.imcorptech.com/TandD.pdf

 

[DanDel]

If a high leakage reading during a Hipot test is encountered, and the problem is not able to found(usually you can 'hear' the leakage if there is a problem at a stress cone, which is the most common issue after inadequate corona suppression), then I expect you could then do a PD test.
Don't forget, it was your description of a cable surviving 2x rated voltage as being the susceptibility of the Hipot test which started this issue. I am merely trying to point out that a Hipot test, done correctly, will not allow a defective installation to pass. (This is why it is sometimes called a 'Go-NoGo' test.)
Usually, it is cheaper to re-do both cable ends than perform a PD test. The investment in a Hipot test (for 99.9% of new installations where there are no problems) is much less expensive than automatically performing PD in every circumstance.

 

[benlanz]

Dandel,

My friend, I completely agree your statement that a DC HIPOT test is much cheaper than a PD test. I also acknowledge that I was not correct to directly compare DC HIPOT to 4 to 5x to AC HIPOT of 2x.

--DC HIPOT is cheaper than PD Test
Have you ever heard the old adage “You get what you pay for.”

--AC vs. DC HIPOT
An AC HIPOT is a much more robust HIPOT. The changing polarity allows PD to show up at a much lower voltage.

I respectfully disagree that a DC HIPOT test performed at any voltage is as good as AC PD test. In our 10,000km of cable testing experience we have seen many cables pass the DC HIPOT test, but completely fail the PD test.

---The human ear can not hear many serious PD sites.
In order to hear a PD the magnitude must be in the 1000s or 10,000s of pC. If one uses an ultrasonic sensor you may be able to 'hear' PD as small as 100pC. IEEE 48 termination specification states for a properly built termination can not have even a few pC at around 1.4Uo. It is not the magnitude of the PD that will brake down the cable/termination, but rather the pressure per unit area! A termination with continuous PD at operating voltage with 50pC of magnitude will fail the cable eventually. If the 50pC is over a small area the termination will fail much quicker than if it is distributed over a large area. If a termination fails 10years later, does anyone fault the workmanship? The contractor’s 1 year warranty is long gone!

--PD Test can find much more DC HIPOT tests
DC HIPOTs detect leakage – tracking/conduction –resistive currents. The defects that a PD systems can see and locate range from tracking (conduction), which DC can find, to defects that will only cause failure after years of material degradation, such as voids. A void in layers of a modern heat shrink or cold shrink termination is not likely to be picked up by a DC HIPOT test because there is very little leakage current in the high impedance defect. The largest component of the current due to PD in a void is capacitive. One has to measure pico Amps to find many serious defects!

--The old days
Back in the old days when terminations were made by hand with semiconducting tapes, tracking/conduction was a much bigger issue. The DC HIPOT was also useful for PILC cables, again because conduction losses are what cause many of the failures.

--New tools for moderns systems
Modern PD systems can easily identify defects with PD from a few pC to greater than 10,000pC. Some modern terminations are resistively graded. This means they will naturally have a leakage current. If one use modern diagnostic techniques for modern systems I sure you will find many more workmanship errors than .1%.

If a cable system owner would like to be sure their system will be reliable for years to come a comprehensive PD test is a small price to pay. Cable, termination, and joint manufactures will tell you that the best test commissioning test is a PD test. This is what they do in the factory. Why not use this baseline in the field?

-Cheers!

 

[DanDel]

First 'Benlanz' truns into 'pdlocator', then most of 'pdlocator' posts disappear! Possibly a commercial venture on this forum being removed?

 

[benlanz]

Dandel,

My handle at other sites is pdlocator. I thought that perhaps I should change my handle to be consistent. To stem the confusion I have changed my handle back to my name, benlanz.

I do work for a testing company that happens to have, with our customers, the largest database on cable testing to date. I just want to share this information, because there is some much confusion regarding the topic of cable testing. Some many people are making claims on the basis of laboratory experiments, or incorrect assumptions that are just not supported by statistically significant field data!

Apparently I referred to our website too many times, and for this I was duly reprimanded. I will be more careful to not include links in the future. I would like to apologize to you and any others who I have offended. I have really enjoyed discussing the issues of cable testing and hope to continue to do so in the future.

Kind regards,

-Ben

 

[DanDel]

Benlanz/pdlocator, thank you for coming clean, though I seriously doubt that yours is the testing company with "the largest database on cable testing to date". I refrain(as I always do) from mentioning my company directly here, though with 43 total full-service testing locations across the country (which include PD as well as Hipot and PF cable testing), we certainly can try to dispute that statement.