Underground fault

Status
Not open for further replies.
Z

Zmax25

Member
Location
Colorado
Occupation
Electrician
Question: What is the worst you've ever seen a piece of direct buried URD faulted?

I had a job last week where a guy lost power to his house. Upon taking some reading with my meter it was pretty obvious the cable was faulted underground. The service from the utility was in the ally at the back of the property with the meter and main panel on the detached garage at the back. After doing some locating I learned the feeder for the house went underneath the neighbors garage, into their back yard for about 50' and then back under the fence and into this guys house panel. I located the initial fault, it was right below the fence that the feeder crossed under and a 4x4 wood post with concrete was placed directly on top of it. The fence had been replaced this past summer. I started digging to find the other end of the cable that was still good and was unsuccessful. Did more locating and realized the other "good end" of the cable was some 40' away from where the fault initially occurred. I've never seen cable burn like that underground. I kept thinking how this could have happened and the only thing I can come up with is not being properly grounded. Since the meter and main panel were in the back on the garage then an EGC should have been ran underground with the feeders to the house panel, however it was only the 2 hots and one neutral(sweetbriar URD). If an EGC would have been installed there would have been a path for the fault current and thus tripped the OCD(200A), I think...

More questions:
How long do you think it would take to burn up 40' of underground cable? did this happen all the sudden or was it a slow process the last 6 months since the post was placed on top of the cable? I feel it was a slow process, however I also think there would have been other signs indicating the feeder to the house was failing. The homeowner reported never having any problems prior.

Could this situation create hazardous voltage gradients?

New user alert: I'd like to upload pictures with this but it appears I can only post pictures from online not directly from my computer...or maybe I just don't know how to use this forum yet
 
Welcome to the forum.

It sounds to me like a case of insulation damage to a direct bury cable and aluminum conductor rot set in.

If the underground wire was installed before the adoption of the 2005 code it would have been permissible to not run an EGC with the feed to the house and rebond the neutral there as long as there were no other metallic paths between the two buildings.
 
How warm was the ground (dirt) you were digging out?
cable burning in the ground heats the dirt up and many times will NOT trip an OCPD. The breakers have to have minimum fault to trip. It could be there was something like 40 amps continuous burning the cable up over time.
hard to believe the guy didn’t notice it before. I could see it if it were a garage or something that didn’t use 240.

If it was a feeder how high did the bill rise?
 
Welcome to the forum.
Zmax25 said:
I kept thinking how this could have happened and the only thing I can come up with is not being properly grounded. Since the meter and main panel were in the back on the garage then an EGC should have been ran underground with the feeders to the house panel, however it was only the 2 hots and one neutral(sweetbriar URD). If an EGC would have been installed there would have been a path for the fault current and thus tripped the OCD(200A), I think...
Click to expand...
Not correct. The neutral and an EGC would have been connected to the same point at the supply end.

The OCPD did not open because the resistance of the fault kept the current too low to trip it.

Zmax25 said:
Could this situation create hazardous voltage gradients?
Click to expand...
If the neutral had burned open, then its current could indeed have created some voltage differences, yes.

Zmax25 said:
New user alert: I'd like to upload pictures with this but it appears I can only post pictures from online not directly from my computer...or maybe I just don't know how to use this forum yet
Click to expand...
Click on the little button that looks like the sun over a mountain, then click in the box, then select your file.
 
How would running an EGC prevent this? The 4 wire URD I have seen (I believe sold as “mobile home feeder” AL 4/0-4/0-2/0-2) has an insulated grounding conductor, as needed for underground use, being aluminum. Unless it somehow managed to burn the jacket off the grounding conductor I don’t see any difference…

In my mind, isolating grounded and grounding conductors for separate structures only serves as to keep the neutral current off the grounding path, thus lowering impedance against a potential fault. It would not protect against a ground fault “leaking” caused by insulation failure or damage, unless it was somehow located close enough to the ground rods to flow back onto the neutral and trip the OCPD. Pretty unlikely though, as the resistance of the earth is usually pretty high

…if I’m missing a basic concept someone set me straight. I’m merely an appretice
 
Hv&Lv said:
How warm was the ground (dirt) you were digging out?
cable burning in the ground heats the dirt up and many times will NOT trip an OCPD. The breakers have to have minimum fault to trip. It could be there was something like 40 amps continuous burning the cable up over time.
hard to believe the guy didn’t notice it before. I could see it if it were a garage or something that didn’t use 240.

If it was a feeder how high did the bill rise?
Click to expand...
By the time I was digging it up the power had been out for at least 2 days so no way to really feel the warmth of the dirt. I haven’t spoke with the customer about his electric bill yet so I’m not sure on that. However he had only recently bought the home so if the cable had been burning for some time then it may have seemed like just a normal usage bill to him without having any history to go off of.
 
LarryFine said:
Welcome to the forum.

Not correct. The neutral and an EGC would have been connected to the same point at the supply end.

The OCPD did not open because the resistance of the fault kept the current too low to trip it.


If the neutral had burned open, then its current could indeed have created some voltage differences, yes.


Click on the little button that looks like the sun over a mountain, then click in the box, then select your file
Click to expand...

Thanks Larry!
I guess I was just thinking if an EGC would have been there then it would have served as the low impedance path but you're right they are at the same point on the supply end so the neutral should have served as that path as well.

The neutral was certainly burned open too. I dug up 6' of charred cable trying to find the other good end. without success i went back to the locator and determined it was about 40' burned underground.

I still cant seem to get the pictures to work. clicking on the sun/mountain icon brings up a box with the 2 chain links on it for me(not sure why) and then there is a box to type an http address into.
 
MattG0311 said:
How would running an EGC prevent this? The 4 wire URD I have seen (I believe sold as “mobile home feeder” AL 4/0-4/0-2/0-2) has an insulated grounding conductor, as needed for underground use, being aluminum. Unless it somehow managed to burn the jacket off the grounding conductor I don’t see any difference…

In my mind, isolating grounded and grounding conductors for separate structures only serves as to keep the neutral current off the grounding path, thus lowering impedance against a potential fault. It would not protect against a ground fault “leaking” caused by insulation failure or damage, unless it was somehow located close enough to the ground rods to flow back onto the neutral and trip the OCPD. Pretty unlikely though, as the resistance of the earth is usually pretty high

…if I’m missing a basic concept someone set me straight. I’m merely an appretice
Click to expand...
Well it depends on the installation I guess. yes, a 4 wire URD(underground residential distribution) would have all 4 conductors insulated and would therefor need to burn thru the insulation over the ground wire before the current would travel back on the EGC and trip the breaker. However I have seen on multiple occasions people run a 3 or 4 wire URD and then run a separate bare ground in the trench with it. In that case once the Ungrounded conductor/s burned open they would have had that path back to the source instead of just traveling thru the high resistance path of the earth.
 
Zmax25 said:
Well it depends on the installation I guess. yes, a 4 wire URD(underground residential distribution) would have all 4 conductors insulated and would therefor need to burn thru the insulation over the ground wire before the current would travel back on the EGC and trip the breaker. However I have seen on multiple occasions people run a 3 or 4 wire URD and then run a separate bare ground in the trench with it. In that case once the Ungrounded conductor/s burned open they would have had that path back to the source instead of just traveling thru the high resistance path of the earth.
Click to expand...
That sounds a lot more useful than just an insulated low impedance grounding path…but your cable is still toast, hopefully not 40ft worth lol.
 
MattG0311 said:
That sounds a lot more useful than just an insulated low impedance grounding path…but your cable is still toast, hopefully not 40ft worth lol.
Click to expand...
I'm pretty certain it was 40' worth, but I sure wasn't going to keep digging for 40' to try and see, but I trust that my locator was correct. The section of faulted cable was actually in the neighbors back yard and a good portion of it was under a slab of concrete too. We ended up abandoning the underground and putting in an overhead feeder.

And yes, that would certainly be more useful.
 
Zmax25 said:
I still cant seem to get the pictures to work. clicking on the sun/mountain icon brings up a box with the 2 chain links on it for me(not sure why) and then there is a box to type an http address into.
Click to expand...
The chain is for entering a URL; you don't want that. Just click in the box where it says (Or click here).
 
LarryFine said:
The chain is for entering a URL; you don't want that. Just click in the box where it says (Or click here).
Click to expand...
Today I clicked on it and it worked! heres a couple pics
burned close up.jpg
the entire 6-7' that i dug up looked like this. I can only imagine thats what the entire 40' section of it looked like.
burned close up2.jpg

all 3 conductors totally gone. I just dont understand how this guys power worked for so long with cable like this. I dont think that this is something that happened in a single instant moment
faulted cable section.jpg
This is the section of cable I determined was bad. Keep in mind this feeder is is for the house on the left of the fence.
 
Metrotech.
Nice setup. We have several of those..

I’m surprised it worked AT ALL like that.
The only time I ever saw wire that disintegrated and surrounding dirt that bubbled was a lightning strike.
It was instantaneous..
That really had power flowing?
I really don’t see how.
 
Hv&Lv said:
Metrotech.
Nice setup. We have several of those..

I’m surprised it worked AT ALL like that.
The only time I ever saw wire that disintegrated and surrounding dirt that bubbled was a lightning strike.
It was instantaneous..
That really had power flowing?
I really don’t see how.
Click to expand...
I have no way of knowing if power was flowing on it, im simply assuming it had to be. There certainly wasnt any lightning in colorado in the last month. But its really hard to tell. Im quite certain the fault started at this fence postPost placed on cable.jpg
fence was replaced sometime late last summer or early fall.
the burned section of cable was from here back towards the main panel/meter, not toward the house panel. The owner of the house had only recently purchased it and reports not having any power issues since he lived there. Then he was on vacation and came back to no power. so no way of telling how long it was out. At least two days but maybe up to 5 or 6.
 
Status
Not open for further replies.
Top