Underground megger test not passing

[roncom0351]

I have a situation where two underground circuits #1 (2 runs of 1300 LF of 1 Ph 250 THWN in 3") #2 (2 runs of 1300 LF of 1 Ph 1/0 THWN in 2") located in the water table (water is present in MH's) does not pass a megger test. The original runs did not pass so the conductors were removed and new conductors installed. (Getting the same results).

I'm being told by the contractors engineer that the cables are not passing because of the water the cable is in.

The first test on the original cable pulls did not pass the megger testing so the contractor applied 120 volts to heat up the cables and try to dry the moisture out. This didn't work.

Both of the second set of conductors are not passing the megger testing.

The contractor is asking us to approve letting them power up the cables without the testing complete.

I have only run into this situation a long time ago when the cables came with paper/insulation wire covering.

The new cables these days don't have the paper covering and should not have moisture in the wire.

Does anybody have suggestions as what to do to get these new cables to pass the megger testing.

 

[petersonra]

What values are coming back from the test? 1300 feet is a long run. You might be getting some anomaly from the testing procedure. What procedure is being used for the test. What value is being used as the pass fail point, and how was it determined?

Are all the conductors failing?

It is possible that when they were pulled the new conductors through the existing water filled conduits that water got into the conductors. I don't think there is any reliable way to get the water out once it gets inside the insulation.

 

[G._S._Ohm]

A TDR test may show a discrete point along the cable that is causing you to fail.

Here's two links to use for benchmark values

http://www.prysmianusa.com/export/s...ngineering_Guide/Megohmeter_Testing_Rev_1.pdf

http://www.nexans.us/eservice/US-en... Voltage Cables in the Field - 2010-06-25.pdf

 

[zog]

What values are coming back from the test? 1300 feet is a long run. You might be getting some anomaly from the testing procedure. What procedure is being used for the test. What value is being used as the pass fail point, and how was it determined?

Are all the conductors failing?

It is possible that when they were pulled the new conductors through the existing water filled conduits that water got into the conductors. I don't think there is any reliable way to get the water out once it gets inside the insulation.

You can do a nitrogen purge

 

[petersonra]

You can do a nitrogen purge

how does that work?

 

[roncom0351]

Megger Values

Megger Values

#1 Circuit: Ph. A to Gnd = 0, Ph. C to Gnd = 0, Ph. A to C = 0

#1 Circuit: Ph. A to Gnd = 0, Ph. C to Gnd =308, Ph. A to C = 232

All values are in Megohms.

 

[roncom0351]

Purge

Purge

Nitrogen purge is not possible................

 

[zog]

how does that work?

Pretty well

http://forums.mikeholt.com/showthread.php?t=99286

 

[G._S._Ohm]

#1 Circuit: Ph. A to Gnd = 0, Ph. C to Gnd =308, Ph. A to C = 232

If A to G is 0 then C to G and A to C should read identically since they are in parallel but this may be a measurement repeatability problem.

 

[K8MHZ]

I don't know if this will help you or not, but a while back we were building a power plant on land with a very high water table. All the trendways had at least a foot of water in them. It was a nasty, wet winter and all the spools were stored outside.

My job was insulation testing. I found that if I did not use a heat gun to dry out the ends of the conductors before testing, over half would fail.

Just something you may want to try. It has to be done on both ends, obviously. IIRC, it only took a minute or two per conductor.

Also, leaving bits of tape on the ends will fail a megger test. Just the sticky stuff from being taped up will do it, too. If we were working with a taped conductor, we always cut the last inch or so off if it failed an insulation test and then retested.

 

[roncom0351]

Thanks K8MHZ

Thanks K8MHZ

Thanks K8MHZ my contractor is going to try this in the morning.

roncom0351

 

[G._S._Ohm]

The 'zero' megohm values you may want to check with a regular DVM.

 

[K8MHZ]

Thanks K8MHZ my contractor is going to try this in the morning.

roncom0351

You are quite welcome.

The most problematic ends were the ones that were made up for pulling. They would get stripped, then taped, then soaped then drug through water. Then they may sit there in that condition for days before I got to check them.

My standard method was always nip a couple inches off the end the 'head' was made up on before I even tried megging them.

Let us know if any of my suggestions worked. I used them with success on literally hundreds of conductors at the power plant job.

 

[brian john]

As noted you can heat the conductors in addition clean all ends dry and tape one end to keep them clean and dry and free from accidental shorts.

 

[Cow]

I've had the same problem when megging, dry the ends of the wires if they aren't and retest. That's one more lesson I learned the hard way when I replaced some motor leads that weren't actually bad, I think they call that experience?:ashamed1:

 

[jeremysterling]

The 'zero' megohm values you may want to check with a regular DVM.

I agree that the "zero" result is not a valid value for resistance. If your zero value indicate you're off-scale high, then I would say your wire is good. If "zero" is actually some ohmic value that is a dead short to ground, repull the conductors.

 

[don_resqcapt19]

A "megger" will often read "0" where a DMM reads infinity. As the others have said, contamination at the ends of the conductors being tested often leads to unacceptably low megger readings.

 

[K8MHZ]

A "megger" will often read "0" where a DMM reads infinity. As the others have said, contamination at the ends of the conductors being tested often leads to unacceptably low megger readings.

I have a megger that the scale is reversed upon setting the meter to the 'meg' setting. It has other resistance settings like an ohm meter, but the scale is backwards.

It's an older analog meter. Really confuses the guys used to a digital meter. Set to ohms, a 'pass' would peg the needle to the right, set to meg, a 'pass' would peg toward the left.

 

[G._S._Ohm]

if I did not use a heat gun to dry out the ends of the conductors before testing, over half would fail.

After the heat gun was used did all pass?

 

[K8MHZ]

After the heat gun was used did all pass?

With heating and/or trimming, they all passed. The pulls were all done properly with the right eqipment and they did not spare manpower when it come to pulling in conductors. All the conduit was way oversized. I almost got stuck on the pulling crew (which would have sucked). I did that for about a week before they found out I could read blue prints and operate test equipment, including the hi-pot, which not many j-men wanted anything to do with. (66,000 volts in the rain. What could possibly go wrong?)

The environment was the culprit. Everything was wet or frozen and our lay down area was outside and unprotected.

The crew that started the project, which I took over shortly, was baffled by the amount of megger failures. No way could those readings be reflective of the insulation integrity, so until they figured it out, there were no re-pulls.

Just when I started doing the testing, someone contacted the cable manufacturer and told us to try trimming and heating. It worked and became standard procedure after a failure.

The goo on duct tape will read. They used duct tape to protect the head during the pulls. So I just started trimming every conductor on the pull side before we even started the testing. If they were obiously wet, out would come the heat at that point instead of waiting for the inevitable.