Go Back  DoItYourself.com Community Forums > Electrical, AC & DC. Electronic Equipment and Computers > Electrical - AC & DC
Reload this Page >

Outdoor Underground Circuit: Ground Fault? Neutrals seem crossed. Help!

Outdoor Underground Circuit: Ground Fault? Neutrals seem crossed. Help!


Old 08-30-14, 10:36 AM
Thread Starter
Join Date: Aug 2014
Location: USA
Posts: 6
Outdoor Underground Circuit: Ground Fault? Neutrals seem crossed. Help!

I have a weird problem that suddenly popped up with my 15 amp underground circuit (running for for 16 years) that feeds couple of lampposts, a few outdoor outlets and a small shed.
A month or so ago, all the exterior lights and outlets on this run went out. The GFCIs on individual receptacles went out. Then the inside breaker wouldn't turn on.

This past weekend I set out to hunt down the problem. Here's the run: Outlet #1 (containing the feed coming from the panel) feeding Outlet #2, which continues on to feed Junction box #3 which has two feed... one to a separate lamppost... the other to the small shed and then onto a couple more receptacles and the final roadside lamppost.

Here's what I found. When I disconnected the feed coming FROM Outlet #1 to the rest of the circuit, I can get the breaker (and the GFCI in this box) to turn on. Next, I connected Outlet #2 again, leaving off the feed to the rest of the run. Ok... so #1 and 2 work fine.

Now, when I add the feed back to Outlet #2, the GFCI in Outlet #1 pops.

So, I disconnected (inside Junction box #3) all the leads. There are three pairs going into this junction box: One is the feed from Outlet #2, one pair goes to the 1st lamppost, and the other cable goes toward the rest of the run (shed, outlets, 2nd lamppost). I then turned the circuit back on to make certain which pair was the feed from Outlet #2. I found the hot lead using a no-contact tester and marked it. I then used a 2-wire probe voltage tester to find the neutral from this feed cable (it's impossible to tell inside the box which is which). To my surprise and confusion, no matter which neutral OR black wire I touched (of the remaining 5) they all lit up the tester as 110 volts... completing the circuit! Mind you, one pair goes to the disconnected lampost and the other paid goes to the rest of the run... disconnected. Shouldn't only a single wire complete the circuit as neutral??

Huh? This really got me. What do I have going on here? The cables are in an underground trench (about shovel depth), some of which shares the trench with my black plastic pipe feeding my garden spigots. They are in sand at the bottom of the trench (I did all this myself between 8-15 years ago... the 8 being that 1st lamppost run). Am I right in thinking that all the cables seem to be compromised somewhere? Why would all the 5 remaining cables in the junction box function as the returning neutral in the main feed?

If I have to dig the cables to this part of the run it would be about 70 feet worth. I've already changed out the GFCIs in the run. Cleaned out any ant's living in a couple of the boxes. A couple of outlets showed rust so I cleaned and replaced the outlets. Nothing has helped and I'm left with this strange 5-wire bunch of neutrals in that junction box.

Is this a short or cross of the neutrals (and the blacks too)? Is this a ground fault underground? Is the cable shielding compromised?

My next test will be bypassing the longest feed run from Outlet #2 to Junction #3 above ground with new temporary cable... just to see if the GFCI and breaker will stay on. I figure if this works, the it would definitely be something wrong with that single pair feed. But the multi-neutral thing really throws me. Am I dealing with all the cables crossing somewhere? (Tight, rusty staple or something else?)

I'd really appreciate your expert advice on this. I've been doing my own electric work for over 40 years but this kind of troubleshooting is new to me.

Sponsored Links
Old 08-30-14, 10:58 AM
PJmax's Avatar
Group Moderator
Join Date: Oct 2012
Location: Northern NJ - USA
Posts: 46,091
Welcome to the forums.

C'mon Jerry..... 40 years of electrical This should be an easy repair.

You're mentioning multiple GFI's. One GFI should not be feeding another GFI. You can have several on a circuit but each GFI should be wired to line only and treated as a standalone device. It's also ok to have one GFI feeding multiple receptacles.

To test the wiring.... disconnect it at both ends and use a VOM set to ohms. A high resistance setting for analog or auto on a digital. You should measure no continuity between any wire and earth ground. If you measure any continuity to earth.... the cable is bad.

What is your wiring method.... UF, PVC ?
Old 08-30-14, 11:37 AM
pcboss's Avatar
Forum Topic Moderator
Join Date: Jul 2002
Location: Maryland
Posts: 14,587
You really need something besides a non-contact detector to troubleshoot this. An analog tester like a Vol-con or a multimeter is much better for troubleshooting.
Old 08-30-14, 01:20 PM
Thread Starter
Join Date: Aug 2014
Location: USA
Posts: 6
Well, the repair part will be a breeze once I find out what is causing this. The digging part is what's making me want to cover all bases and troubleshoot the problem as closely as possible before having to dig. I hate not having my exterior lights on at night.

As for the GFCIs, I did have them all separate (one for each outlet/device and junctions carrying to line downstream). Since the run is around 200' total with varied receptacles and lights along the way, it made it easier to simply press a local GFCI rather than have to walk back up to the house to a single GFCI or into the cellar to the panel after a heavy rain or snow. (My property is terraced).

When I started troubleshooting this problem, I swapped out the GFCI in Receptacle #1 at the beginning of the circuit and tied in the rest of the run to this GFCI. I figured--keep it simple until I find the glitch. After I trouble shoot and repair, I might go back to the way I had it originally.

Ok, so you are saying to get my volt meter (I have an analog), set to ohms on the higher setting (my options are X 1K or X10K) and test each lead while they are disconnected at each end of the run? (This is starting to make sense). So, if the needle moves at all, that would mean the cable sheathing is compromised (and possibly touching other cables too) and grounding the neutrals (and black leads) to earth?

By Jove, I think I've got it!

The runs are UF. Decent quality. I remember installing in sand at the bottom of the trench, but you never know... with frost/heave up here in Pennsylvania along with the crazy winters we've had lately (tons of snowmelt last year) maybe a sharp rock popped up, or a tight staple on a post rusted too far and cut into the shielding.

I suppose I've been lucky over my 40 years of learning and doing my own electric (I wouldn't install a service panel from the meter though). Even with outdoor runs, I've near had one fail.

BTW... I appreciate your advice on this forum. You pros are very generous with your knowledge... unlike some other pro oriented forums that shun DIYers. I also want to tell other DIYers to learn as much as possible, research the Codes, and by all means, ask a pro or hire someone if it's over your head (and understanding).

As for me, I'm a commercial advertising photographer and always share my expertise when asked. Electrify my back and I'll photograph yours. (wait... that sounded weird!) I wired my Manhattan loft studio myself over the years... darkrooms, dressing rooms, 600 amps worth of studio power for lighting, etc. I've also renovated a couple of vintage houses like my current 1868 brick colonial. I've got 90% of this house rewired and updated and still moving forward to a 2nd bathroom/laundry and new kitchen next year.

Let me know if it seems I am understanding your suggestion. My son is starting school next Tuesday... after that I'll try your idea and start digging.

Old 08-30-14, 02:44 PM
PJmax's Avatar
Group Moderator
Join Date: Oct 2012
Location: Northern NJ - USA
Posts: 46,091
Now, when I add the feed back to Outlet #2, the GFCI in Outlet #1 pops.
That tells me that you are using the load side from GFI 1 to power receptacle 2. That's wrong.
Only an appliance plugged into GFI 1 should trip it. Based on what you said.... you should not be using any load side(output) terminals on the GFI's

set to ohms on the higher setting (my options are X 1K or X10K) and test each lead while they are disconnected at each end of the run?
Use the highest resistance scale you have- x 10K or even x 100K if you have it.

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off

Thread Tools
Search this Thread
Display Modes