located short in u/g cable

Thread Starter

tlukwin

Joined May 30, 2012
3
Hello all, hoping to get some input on a cable problem. Have a 2C SPC8 cable running u/g for ~6750 ft. It has recently shorted to itself, not to ground. (Meggared) Resistance readings of the cable open on both ends are 4.4 Ω from the east end and 4.64 Ω from the west. Doing the math and assuming a dead short, the short should be 48.7 % the length of the cable from the east end. Took a TDR and shot the cable from each end, using a VOP of .71 (just guessing), got readings of 4468' from the east end and 4681 from the west, (means the VOP was wrong), but makes the math jive with the resistance tests. Digging up at the surmised spot(4' down), the resistance was ∞ to the west and 0.8Ω to the east. Another TDR shot shows the fault ~600 ' to the east.
I tried using a cable locator/faultfinder to try to locate the short, but since it is not to ground, the faultfinder doesn't see any problem. I have loacted it np, it runs right along the railway, so I know exactly where it runs. I tried hooking the + of the transmitter to one wire, and grounding the other, and grounding both the wires at the far end and hooking the locator up in the CD mode, hoping to see the current direction change, but to no avail. Running out of ideas... Anyone????:confused:
 

strantor

Joined Oct 3, 2010
6,782
Do you have access to, or funds to buy, a percentage meter? I work in a Wire & cable making plant, and that's what our guys use to find the conductor>conductor shorts in miles-long cables. It basically does that same thing you did with your megger, but much more accurately. I think it usually gets them within a few feet of the short, then they use a "thumper" which, as I understand it, is basically just a big bank of capacitors which they discharge across the short, and you can hear the "pop" where the short is. However, I think to use the percentage meter you need access to both ends of the cable at the same time.
 

ErnieM

Joined Apr 24, 2011
8,377
An alternate way to get a better resistance reading is to use a power supply in the current limit mode. This depends on voltmeters being more accurate for low voltages then the ohmmeter function is for low resistances.

You drive the shorted line with say .5 amps and read the voltage on the lines. If you can use the same current both ends (or even better measure the current) you should get a more precise answer as to how much cable is between you and the short.

One trick we used to use building hybrids when there was a "weak" short buried down some layers was to charge a large cap (say 100 uF) to 50 volts, then use that as a power source to zap the short. You get a brief burst of large current this way, and if the short is the weakest link you may burn it out without having to dig anything up.

You also run the risk of making an open anywhere along the line so do use this method as a last resort.
 

Thread Starter

tlukwin

Joined May 30, 2012
3
The line was used to carry 600 V, so I think that would have opened up the short. BTW I shot it with a TDR to 600', dug at that location, and unfortunately, found a conduit, so it seems its in the conduit, so much for that idea. Time to dig in a new cable !!!!
 

JMW

Joined Nov 21, 2011
137
Before you dig, calibrate your TDR. You now have access to a known physical length of cable, adjust the velocity factor to get that length and then remeasure to your short. If it is in conduit there is a possibility the conduit has been crushed, you may be able to locate this by inspecting the route. Look for new track work, not unusual for a backhoe to excavate your line.
 

Mongrel_Shark

Joined May 30, 2012
22
I worked in Concrete for over ten years. One day I gat called out to cut a hole in a 5 acre carpark. so an electrician could fix a short (in this case was bad join). I was laughing on the way to the job, saying we'd be there all day cutting up half the carpark...

The electrician had a very sensitive induction sensor, with wheels. about the size of a lawnmower... He already had path of cable marked on concrete from plans, before we arrived. All he did was push the sensor down the line, it had a light and a buzzer on when it detected ac. about halfway down the line the light went out... A few quick circles to makes sure the cable didn't wander off with regard to the plans, and a few sensitivity adjustmants, and he marked a squre on the concrete about 400mm (1'4") across, for me to cut out. I was convinced the sensor couldn't be that accurate andtried to get him to make a bigger hole (We got paid by lineal meter of cut :D) But he was pretty sure. So I cut the square out and helped him dig. convinced I would be doing more cutting when he failed to find the short. Turns out it was smack bang dead center of the square he marked.....


So. Long story short. Can you use induction sensor? I think the electrician I worked with hired or borrowed the one he had. it was made just for finding cable problems underground, either shorted or broken!

I have a pen does the same thing. but not through a meter of dirt, concrete and reinforcing steel....
 

Mongrel_Shark

Joined May 30, 2012
22
I just saw the condiuit issue. There are ways around that! no need to dig up the whole trench!

You could attach new length to on end of old wire and pull through old conduit.

Or cut conduit where short is and fix cable, then sleeve the conduit where you cut it. This is easiest if you put the sleeve on ones side while wire is cut...


There are more ways around the problem. but I'm hungry and dinner is calling...
 

Thread Starter

tlukwin

Joined May 30, 2012
3
Yes, plan B will be to go downstream from the fault, dig it up there, and tie into another 600V feed to power the location. I like the idea of the induction sensor, since I wasn't having much luck with the cable locator, and recalibrating the TDR to a known length of identical cable pointed me to dig where the conduit was. There has been no excavation activity around there for 5 years, and the road crossing was closed off 2 yrs ago, so the only thing I can forsee, it that the cable was spliced initially, 15+ years ago(red black on one end, black white on the other) and somehow the splices have bared and shorted to each other. As a last ditch effort, was going to hook 120 V to one conductor, and let the AC run down the line, through the short and back, and power a device to have current run down the line, through the short, but not any further. Will take the locator and see if it can sense the AC that far down the line, but not hopeful. If that still points to the conduit location, see plan B
 
Top