Old Cables 3

[hanksmith]

Currently working on a job that has cable installed that I plan on using, the problem with the cables is that they have been installed for over 40 years now.

The cables are in use so I know they are working, I am just adding additional load to them. If the cables fail they will shut down a good part of the plant.

Are there any de-rating factors to use to adjust for the age of the cable?

Thanks

 

[Zogzog]

What type of cable are you talking about? Either way, after 40 years they should at least be considering replacement.

 

[TurbineGen]

What voltage are the cables? 480V and below, I would say it should be ok to use them if they pass a Megger insulation test. 2300V and above, I'd seriously consider replacement. Between thos voltages is a little grey and I'd consider things such as temperature, moisture and other environmental factors.

------------------------------------------------------------------------
If it is broken, fix it. If it isn't broken, I'll soon fix that.

 

[hanksmith]

Cable are all 5kV

All 1 conductor cable.
some are armoured.
some are un-armourded, un-sheilded.

The un-armoured cables are in concrete duct banks, the armoured are in dedicated cable tray

 

[Zogzog]

Well 40 years is about right for these cables, depending on the type of insulation. Your best bet is to do a condition assesment test, Tan Delta is a shutdown is possible, or PD testing if a shutdwon is difficult. That way you can asses the condition of each cable and "rate" them according to condition so you can replace the worst ones according to your budget.

 

[oldfieldguy]

Your cables are OLD. Not necessarily bad, but still, they're old. I've dealt with 40+ year old cable systems. Like Zog says, get them tested and for your peace of mind, be prepared to replace those that don't look good on the tests.

We did this and our client's unscheduled electrical outages dropped to a small fraction of what it was previous to the test/replace effort.

I know (from experience) that standing in a meeting and telling management that you want to replace cables that are still working is sometimes a hard sell, but you're talking about the future reliability. It's far easier to schedule replacement at your convenience that run around screaming in the dark when they fail at some critical juncture in the future.

Document all conversations for your records. That way, if they turn you down on replacement, you'll have an "I told you so" file for the untimely failures.

Good luck!

old field guy

 

[hanksmith]

Is the tan delta test the same as a "tip up" or "tied up" power factor test??

I think a shut down is possible, from the research I have done the PD test is not great for cables unless they are about 90% failed, it's great for testing the terminations though.

I have now been told that a tied up power factor test is best for testing cables but I have never heard of this test before, I did a google and it came up with lots of hits, if anybody has some decent literature on the topic it would be greatly appreciated.

Thanks

 

[BJC]

A shut down is possible wheather you plan it or not. Cables that old can have brittle insulation, move it just a little and chunks of insulation can fall off or cracks can develope.
As a contingency I would have replacement cable located and available.
Your raceway may be bad as well. If you have embedded raceway it may be cracked and/or collasped. The most common problem is raceways get silted up over the years the cable can cemented in place. It can be impossible to pull out.
Line up your ducks so you can install new cables in a hurry. Tha may include new raceways.

I have done cable that were not that old and we put in new cables and swapped connections to make the transitions. If shutdown time is valuable new cables and racways can be very cheap.

 

[Zogzog]

"Is the tan delta test the same as a "tip up" or "tied up" power factor test??"

Similar, but slightly different equipment.

"I think a shut down is possible, from the research I have done the PD test is not great for cables unless they are about 90% failed, it's great for testing the terminations though."

PD testing has come a long way in just the last year or so, there are different methods and equipment (And databases) out there, some are much better than others, the only problem areas is when there are 2 different cable types spliced together. These guys have a great system and Don is probally one of the leading PD gurus in the industry.

http://www.nooutageelectricaltesting.com/

"I have now been told that a tied up power factor test is best for testing cables but I have never heard of this test before, I did a google and it came up with lots of hits, if anybody has some decent literature on the topic it would be greatly appreciated."

Cable testing methods has been discussed here many times before, and there were some really smart cable guys in the discussion, search for those threads. Dont listen to what you have "heard", do the research, read the standards.

ANSI/NETA
IEEE 400
IEEE 576
ICEA

All 4 of these are recognized MV cable testing standards, none of them discuss Pf, they all are very similar.

One other good reference is the EPRI cable testing procedures that were updated last year, could be the most recent in depth study.

 

[petamm]

My opinion is that 40 old years cables are quite old but not neccessary close to dead.It depends on history of each one - rate of faults, loading, overstressing... and the real condition (make the diagnostics). You did not mention type of insulation which is important for tests recommended too. I would recommend these tests:
1. VLF 3Uo 60 min. withstand test, 2 .tan delta or other integral method test (like RVM,IRC...), 3. PD off-line test. Generally each of the method has some + and -, you always must combine them. For exmpl. if the cable insulation contains a lot of water - there are no air/gas voids - no or less PD´s - PD test seems to be OK but cable can faild under VLF test.Although I am a big PD testing fan, in this case I guess it will not tell you a lot (40 years old 5 kV cable construction - a lot of PD´s along the whole cable lenght possible or even PD location impossible because of the multidischarge activity and multi signal reflection, a high water/humidity cable insulation content possible in case of PILC, WT in case of XLPE/PE insulation). You should think about replacement in the near future as well. Testing can prolong the time and sort the cables.

 

[Ceast]

I'd recommend both, Partial Discharge and Low Freq tan delta. Grab a PD sample and analyze the leakage. Tan delta is a much better test the PD test is a good preliminary test. It will show obvious breakdown issues. You should be able to get both tests done for less (depending on count) than 5k. Always choose a NETA testing firm.

 

[ingamg]

I think you have to change that cable. You can expect faults.

The time in use its enought to take the decision.

 

[benlanz]

Hello and greetings to all again.

I haven’t posted to this site in a while but, with all the confusion I hear in the above discussion, it appears that we need to open up this topic again. We have had this discussion several times on this forum but perhaps many of you don’t remember our previous discussions. I recommend that you do a search on this site for ‘cable testing’ if you would like to review our past discussions.

I will assume the subject cable systems are built with extruded insulation. Are they?

Yes, a 40 year old cable is ‘old’ but, if the financial impact of an outage is low and the cable system’s reliability assured with a thorough diagnostic test, there are very few good reasons to replace it.

1) Moderate loading is not likely to be an issue
The constant loading has very little to do with the insulation reliability of the shielded power cables. If we are talking about the reliability of mechanical connection you might have a point. Extruded cable systems rated 5kV and higher almost never fail by conduction or thermal issues. Extruded cable systems almost always fail through an erosion process associated with partial discharge (PD).

2) PD and TD are not effective with unshielded cables –in fact most electrical tests are not effective on unshielded cable!
hanksmith says that some of the cables are unshielded. Unless a cable has a shield to define a ground plane, you are only testing the points of the cable that do. Otherwise you are testing the air around the insulation, which is not important!

3)On-line PD finds <3% of defects
Why are some of you recommending an on-line PD test for the shielded cables when the circuit can be de-energized? If reliability is critical, the only test that can assure reliability is to repeat the manufacturer’s off-line PD test.

4) TD cannot assure reliability
TD detects aging mechanisms in cable insulation. The problem is that you can not tell where the “losses” are in the cable system. Most TD tests in field are not performed with guard circuits at the terminations. If you do not use guard circuits, your measurement is useless. Most field TD equipment sold in North America do not have guard circuit capabilities! Losses measured over the surface of an old termination or at joints can be 2 to 3 times higher than losses in cable insulation. Even if a TD test is not grossly affected by a termination or joint, what are you going to do with a measurement that tells you that your old cable is, well…. old! This sounds like an expensive calendar to me. If you are trying to compare the relative losses on 1000s of utility cable to prove to management that your old cables are old the TD approach may make some sense. If you are working with critical cable systems, and TD test is a very poor choice to assure reliability.

I know one company that replaced 30+ out of 40+ terminations because of “bad” TD readings. I asked how they knew which terminations to replace. They didn’t know, so they just replaced the terminations at both ends of each phase with a ‘bad’ TD reading! I asked they used a guard circuit. They asked me, “What’s that”. My friends, unless you use guard circuits, your Power Factor, TD, Dielectric Spectroscopy, Relaxation Voltage,….. (all general condition assessment tests) results will have serious errors. Sadly, the company who did the TD testing, is also in the cable repair business. The more ‘bad’ TD readings they measured, the more repair business they received! My advice… Before you decide what test to use and what to do with the results, consult an expert and the IEEE standards.

5) For the record, PD tests do not measure leakage. Leakage has very little to do with extruded cable system reliably.

6) A “Tip up” test refers to the shape of the voltage vs. loss curve. To the best of my knowledge there is no such thing as a “tied up” curve. If there are losses in system under test, the voltage/loss curve will tip up in a non linear fashion as the voltage increases.

7) Careful with VLF destructive withstand – you may do more damage than good
If you can afford an outage, and you simply want to blow out some weak points on a shielded power cable a very low frequency (VLF) AC destructive withstand is a reasonable choice. Remember a destructive HIPOT cannot assure reliability.

8) I agree with Zogzog that you need to read the standards. However, the guys associated with the link provided need to read the cable testing standards too. While they provide many great electrical services for other electrical components, to the best of my knowledge they only provide, on-line PD testing for cable systems. If the manufacturers only did on-line PD testing on all cable components sold over the last 40 years, all of our cable systems would be in big trouble!


Cheers,

-Ben

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

 

[dodgergrl]

Ugh! What kind of test can one perform on Type MV-90 unshielded cable for overall cable condition assesment? My project is a 30 plus year old underground installation in a mobile home park and the owner would like the cable tested. Please do not refer me to IEEE publications as I can not afford to purchase them at this time. We do have a NETA certified testing company on board but, for various, (some would say "silly"), reasons, I do not want to ask them until I know at least a bit more about our options.

Many thanks in advance!

 

[MaXXXXPower]

I am working in an installation company, recently we used an old cable in a new innstallation. Before installing the cable we measured with an insulation tester that it was OK. Now, when the ship is on commissioning there is an earth fault in this cable due to its age i guess.. Next time we are using new cables!

 

[vtpower]

Another option for older cables is something we have done recently which is Cable Cure Injection. The problem is, you need to take the cable out of service to inject them, but you don't need to replace. It will essentially rejuvenate the insulation, filling in any treeing that my have occurred, and most vendors come with a 20 Year guarantee, of course they won't reimburse you for lost plant time if it does go down. It's another option to not replacing aged cables.
VTPOWER

 

[benlanz]

VTpower is right that injection can help with water trees in HMWPE and XLPE cable insulation which do not have strand-filled or solid conductors. However, injection does nothing for defective splices, terminations, cable insulation voids or electrical trees. The industry best practice is to perform a complete off-line 50/60 Hz PD test to repeat the manufacturer's QC test in the field per IEEE recommendations. This process is the only effective way of determining the conditions of the cable system and which cable systems are reasonable candidates for injection. If you inject the wrong cables you will most likely be disappointed with the results. Remember, water trees are not likely to fail cable directly. Electrical trees need to be created first but, they can take years before they go to failure.


Dodgergrl,

I assume that by MV 90 you are talking about 5kV class cables systems insulated with EPR insulation without an outer metallic shield. You have a serious challenge on your hands. Unfortunately there are no highly effective, predictive diagnostic tests on the market for unshielded cable (without an outer metallic shield). Without a shield, the voltage stress distribution is not even during voltage application. Thus any voltage applied is only going substantially to stress the insulation which is in contact with a ground plane. Without a ground plane, you are stressing the air around the cable not the cable insulation!! The best you can do is to apply low levels of DC voltage (less than operating voltage) for less than 5 minutes to determine it there is any substantial tip-up in a micro-ammeter during the voltage ramp. Make sure you ground the cables for a few minutes after applying the DC voltage.

There are some tests that are in the R&D phase that may help in the future but, for now you are stuck with the DC tip-up test for unshielded underground cables. This is one of the many reasons that the industry is moving away from unshielded cable in medium voltage applications.


Cheers,



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