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

Problem with wiring between main panel and sub-panel in unattached building

Problem with wiring between main panel and sub-panel in unattached building


Old 07-02-15, 01:23 PM
Thread Starter
Join Date: Jul 2015
Location: USA
Posts: 3
Problem with wiring between main panel and sub-panel in unattached building


I am located in a rural, unincorporated area of Mississippi. A three-wire connection between a main panel and an out-building is still allowed as long as the two structures do not share another common electrical path between them. Truthfully, there is not code enforcement period. Please do not hijack the thread with commentary about that. It is what it is.

I have two sections of direct-burial aluminum wire. I believe it is 2-2-4. I may have that part wrong, but suffice it to say the wire size is suitable for the application—whatever it is. I had all that checked out and recommended and cross-checked. As I am typing, I just do not remember for sure.

Section 1 of wire goes from the main panel to a 100 amp disconnect switch located in the edge of the yard. Section 2 goes from the disconnect switch to a 100 amp sub-panel in a metal storage building. Total distance of about 150ft. The metal building was built and section 2 of the wiring has been in place for a number of years, but section 2 of the wire was not connected to anything until about 2 months ago. (The wire had been installed in anticipation of the house being located where the disconnect switch now sits. House location was built a few feet over from the originally intended location.)

The disconnect was installed about two months ago as was Section 1 of the wire which runs from the main panel to the disconnect. The disconnect was installed to connect Section 1 to Section 2 rather than a splice. The disconnect, main panel, and sub-panel are all grounded with their own grounding rods.

While the sub-panel is 100 amp, there is only one breaker installed, and it is a 20 amp breaker with fluorescent lights powered from it. When everything was connected 2 months ago, sub-panel showed 250 volts or so across the hot legs coming into the sub-panel. For the past two weeks, I have not been in shed and had not turned on lights. Now, I discover lights are not working. Sometime in those two weeks, things went awry. There has been no digging or disturbance of any kind along the path of either section of wire since everything was connected two months ago. Here is what I have done to diagnose problem.

1. I checked voltage on sub-panel in the storage building. Voltage across the hot legs is about 20-30 volts. I did not write that number down, but suffice it to say it is small. Voltage from hot leg #1 to ground is 27 volts. Voltage from hot leg #2 to ground is 4.5 volts. Same for both when checked against neutral. Voltage was checked by putting probe on the slight amount of exposed wire at end that is going into the lugs on the panel rather than probing on the lugs themselves. I was assuming that might eliminate the possibility of a bad connection of the wire to the lugs by probing against the wire itself.

2. I checked the outgoing voltage on the disconnect which connects to sub-panel in metal building. Voltage across the hot legs was 251 volts. Once again, I put the probes on the wire itself rather than the lugs at the bottom of the disconnect. Somewhere between the disconnect and the sub-panel, there is obviously a problem.

My questions:

1. Does conventional wisdom suggest that I have a break (or almost break) in both the individual hot wires along section 2 of wire? It seems terribly unlikely to me that both wires would have a fault in them considering they recently worked and there has been no opportunity for physical damage to occur. If only one failed, I assume I would still get a 120-130 volts between the good leg and ground at the sub-panel?

2. Can anyone think of a scenario where this is just a bad connection at either the disconnect or the sub-panel? Is it possible for this to be the case considering I am getting low voltage when probing against the wires instead of the lugs into which they are connected? Could the outside of the exposed wire be oxidized too much to get a reading? The aluminum paste stuff that goes on aluminum wire was used, but perhaps not enough was used. Would it be worthwhile to remove and reconnect the wires coming out of the disconnect and the ones going into the subpanel?

As you can imagine, I really do not want to spend another $500-$1000 for a new trench and wiring to run new wires. Any opinions or suggestions would be appreciated.
Sponsored Links
Old 07-02-15, 02:28 PM
Temporarily Suspended
Join Date: Jul 2008
Location: NY
Posts: 10,986
Does conventional wisdom suggest that I have a break (or almost break) in both the individual hot wires along section 2 of wire?
Leakage might be a better word than a break. What else is crawling underground besides your wires? Could some critters have gotten hungry?
Old 07-02-15, 09:25 PM
PJmax's Avatar
Group Moderator
Join Date: Oct 2012
Location: NJ - USA
Posts: 43,459
Welcome to the forums.

It sure sounds like your wires have gone bad underground. If the subpanel has no loads connected then you should be reading normal voltage even with corroded fittings at the supply panel.

I would disconnect the hot wires at both ends. Short them together at one end and check for continuity at the other end.

If the direct burial cable wasn't set in sand then it may have come into contact with a rock. A rock will cut the insulation and when power is turned on.... the voltage leaking to ground eats the cable.
Old 07-03-15, 07:54 AM
Thread Starter
Join Date: Jul 2015
Location: USA
Posts: 3

Thanks for the welcome. Clearly this is not the answer I was hoping for, but what you say makes sense. Soil here is sand-clay mix. Not at all rocky. I do not think a rock is the problem, however, read on.

Remember from my full description, Section 2 of the cable (where the problem is) has been buried for 5+ years but only first connected 2 months ago. If there was a small nick or fault in that cable--perhaps from the initial installation 5+ years ago, when power was connected, I could have had a small, unnoticeable leakage. With power connected, the voltage leaking through that "small" fault would cause further decay of the insulation and/or wire leading to a larger fault that would eventually lead to a complete failure? Unfortunately, that scenario does make sense. I am going to do as you suggested later today and short the cables together at one end and test the resistance.

Thanks again for the reply.

Old 07-03-15, 04:21 PM
PJmax's Avatar
Group Moderator
Join Date: Oct 2012
Location: NJ - USA
Posts: 43,459
I remembered your first post but it would be pretty hard to get a random nick in BOTH wires.
That's why I mentioned a rock.
Old 07-03-15, 06:01 PM
Thread Starter
Join Date: Jul 2015
Location: USA
Posts: 3
This afternoon, I disconnected the hot wires from the sub-panel and shorted them together. I went to the other end and measured the resistance by probing the ends of the wires. I got about 1540 ohms (1.540 on the 20K ohm setting). How do I interpret this result?

I then shorted the neutral wire and the two hot wires at the sub-panel and went back to the disconnect. On that end, I measured between each hot and the neutral. For each one, the reading was about 2000-2200 ohms on each one. I did not write the numbers down, but they were not equal. They were close to one another though. In retrospect, I should have probably shorted the neutral and hot one at a time in the sub-panel rather than shorting all three together. I will re-do that tomorrow and check them one at a time.

As for the rock thing, we do not have rocky soil whatsoever in this area. Perhaps there could be a rock buried along the path somewhere, but if so, it was probably placed there. It would be highly unusual in this area if it was naturally occurring. As unlikely (and unfortunate) as it may be, it would appear that I do have something wrong with both wires. At this point, finding where the fault is will be very difficult.

I read somewhere that a potential way of finding a complete break in a wire is to do the following. If anybody is familiar with this technique and can comment on it, please do. I have heard that you can wrap a wire around the insulated part of a spark plug on a small gasoline engine and then connect the other end of that wire to the end of the buried cable. You can then crank the engine and use a portable AM radio and pick up the noise signal as you walk down the path followed by the cable. When you reach a point where the noise signal stops on the radio, you have found your break. Anybody ever heard of doing this? If so, can you provide more detailed information on whether to use bare or insulated wire to wrap around the spark plug and more detail on exactly how to attach it to the spark plug? Assuming this method is valid, am I likely to find the fault in my wire since it is not a complete break (otherwise my resistance would read infinity)?

Any help appreciated. Thanks PJMax for your help already. Do you have any more advice you could offer? Thanks!

Old 07-06-15, 06:28 AM
Forum Topic Moderator
Join Date: Feb 2005
Location: Near Lansing, Michigan
Posts: 10,292
I got about 1540 ohms (1.540 on the 20K ohm setting). How do I interpret this result?
The cable is very bad, essentially completely severed. A healthy cable should have about 0.25 ohms per 1000 feet length. Any chance it was kinked or crimped, backed over with a car, etc, prior to or during installation?

I have heard of the radio method and a few other similar ones. They might work. The real tool that does that process is called a pulser fault locator, and they work very well, but are pretty expensive. If you can find a local electrical contractor with one or an underground utilities marking contractor, they can probably locate the break to within a couple feet. You can then dig it up and evaluate repair or replacement.

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