Cable Fault

I'm going to give this more thought in a bit but something has already struck me. I suggest you get the possibility of moisture out of your thought process. you convinced yourself it was part of your problem and you're yet to have evidence of water. let the answer come to you. don't try to prove out a theory without it sometimes when troubleshooting it's easy to lock onto a theory that sends you in an endless circle, sounds like you have the right tools to tackle this on just isolate all your pathways and get solid readings again, the problem will show itself

I appreciate your advice, but I really need to make a decision asap as this problem is costing alot of money whilst unresolved.

You will never find evidence of water in the cable if the cable is buried. Who knows what's happening under there. There is evidence to suggest moisture ingress, as the resistance is increasing day by day.

Excellent posting, well-written and full of useful information.

It is somewhat unlikely for water ingress to blow all three phase fuses and present all those readings simultaneously, as this would require simultaneous failure of the outer sheath, the inner sheath under the armour and each phase insulation simultaneously. The simultaneous phase-to-earth readings look more like phase-to-armour readings than phase-to-armour-via-rainwater readings. Catastrophic mechanical failure is more likely. Look for places where a digger bucket or something of a similar mechanical nature may have snagged the underground section, or a heavy metal item of some sort attacked the overground sections. A sensitive nose and a pair of binoculars would come in handy.

Review the Permits-to-Work that were open at the time of the fault occurring, as this will give some clues as to what may have happened. Review also the personnel that were registered on site that day, and ask each of them individually if they saw anything or heard anything, as that may give clues as to where the fault is and how it was caused. However, a word of caution. Someone, somewhere, is keeping quiet, with fright of the event and fear of reprisal being the reason for silence. So when the fault is found, concentrate on improving procedures and practices with a view to preventing recurrence; deal with the technicality, not the personality, for the person will already be in a position of never doing it again, and has inadvertently given both self and the organisation a useful learning opportunity. Realise that investment!

Thanks for the response. I agree water ingress is highly unlikely to pierce through 2 layers of PVC insulation and armour. The fault happened at 3am, no work was taking place at this time. I already reviewed what work was taking place in the area over the past week or so, the only thing that was happening was some crane work on some near by vessels. I highly doubt someone has knocked the cable with a crane, as this would've done much more damage than just take out one cable (the cable is installed with lots of other cables on tray).

No digging has taken place in the area either. I agree it's probably phase to armour fault, but could also be phase to earth core fault. Who knows. This one really has me stumped.

One thing is for sure. It can't be seen from here.

you have me going on this one. I hope you find it and report back to us! I'd bet after you send a pulse down the de-energized cable you'll pick up a reading and a spot to dig up. my guess is you'll find a kink or some other "point of compression" in your line that will be the culprit. bestof luck and keep us informed

If possible, check mechanical stresses at each location where the cable is tied down to the cable tray, and obviously where the cable meets and leaves the tray. Long-term mechanical stresses can upset PVC insulation as described here:

http://www.doctorfire.com/PVCwiring.pdf

Mark Bingham
Relativity PL

Kink in the cable perhaps, but it'd take someone fairly strong to kink an armoured cable. I doubt it was installed not adhering to bending radius' 12 years ago and the insulation has finally given way now.

Thanks for the responses, I'll report back when I find something

You did not specify the cable voltage, what type of medium the cable is buried in, nor what the ground water contamination potential is, but there's a dielectric breakdown mechanism called Treeing that is usually completely benign until the failure.

150m isn't a lot of cable but TDR (Time Domain Reflectrometry) is one way to find the fault; however, you should go through the time and expense of replacing the entire length since the cable is breached and there may be other areas of treeing that will cause the next fault if the cable is simply repaired.

Sorry, the voltage is 415V. Cable is buried direct in soil, I'm unsure about ground water contamination, I haven't completed any sampling. Treeing is something I have read about a long time ago but doesn't it only occur toward the end of the cables life? My next best guess was a manufacturing fault.

I did do some reading today on TDR, the problem is getting someone out there to complete the test ASAP.

I'm beginning to conclude replacement of the cable is the way to go.

Treeing can occur anytime, but it's less likely at that voltage. My guess is that the breach was caused during the initial installation of the cable with time and thermal cycling over its operational life providing the catalyst for the breach to grow exponentially until failure ensued.

You could try earth/ground fault testing along the path of the cable. You will need a good length of lead wire for the ground taps along the way, but the cost of troubleshooting will definitely be under the cost of whole replacement.

If you are still getting "good" readings as to your initial assessment i.e., a low resistance rather than infinity, the ground fault tester should give you a nice bell curve and help you in isolating some possible mechanical damage.

What was the rainfall like during the time of the incident? Is there a monitoring point for the water table in the area? Are there any utilities carrying liquid of any kind near the area, and if so, do they have anything in relation to your issue? Any local utilities which may have experience an anomaly may be of some use.

You said there were crane operations at the time nearby. Did the crew there have anything weird happen at the same time as your issue?

Problem solved guys, here's what I found:

Megger reading was incorrectly recorded, it was 50-100meg 2 days later, not 50-100 ohms. I pulled the contactor out of the MCC module and found it had been burnt out on one side. I concluded the cable was fine and the short circuit never occured in the cable, since the other side of the contactor was clean.

Unsure why they read 0 ohms on the day it happened, might've been another incorrect reading.

I have concluded that it was probably loose termination on the cable, causing an arc to form between the terminals and cable strands. The loose termination has probably come as a result of vibration (we have massive inrushes on these contactors as we are starting fans that sometime spin in reverse from high winds when off) or someone has tampered with it.

I replaced the contactor and she's back up and running :)

<...back up and running...>

Well, that's a good thing. So it wasn't a cable fault after all.

The issue points to a need to review the training needs for cable installers and for test equipment operators, and to record the outcomes. The objective is to prevent recurrence.

I thought the moisture theory "wouldn't hold water" thanks for getting back

I did not see in your OP whether the cable is shielded or not.

If it is shielded, you really need to closely examine the corona shield boots and connections for proper installation. (Semi conductor relief spacing/distance, drain connections and configuration, drain conductor routing, etc.)

In addition it is pretty common nowdays for corona shield termination boots to be defective and suffer premature failure due to poor quality of the materials.

This quality issue is very dominant in products from certain countries such as China.

As an aside to the original issue...

In order to eliminate wind-milling of your fan motors (and reduce the in-rush) you could try DC injection.

You would need to perform a cost-benefit analysis for number of fans, power requirements, MTBF/cost of your contactors, etc., but it may end up being beneficial...that's what DC injection units are made for!