Perhaps someone can help us ascertain what caused this near disaster to occur:

Check out the attached pictures.

This is a Norcold Model 6162 gas/electric refrigerator installed on our houseboat. This particular model has an "AUTO" setting which we usually kept it on that switches back and forth between gas and electric. Our houseboat has no shore power but has a Honda EU3000 generator. This generator has a 20A breaker built in. The generator has a cable wired into the breaker box of the houseboat. These are 15A breakers and they seemed to be working fine. I had that fridge wire plugged into the same outlet with the plug for our 700W microwave. Because they were both 90 degree plugs, I had to use that outlet expander. The outlet expander has an indicator light on the side to show it is operating properly. I tested it at home and it appears to be working fine. It will be going in the trash. It is probable that the microwave and fridge were running at the same time. But at most the microwave was on for 3 or 4 minutes and it is only 700 W. The fridge heating coil is supposed to only pull 2.5 amps. The house 15 amp breaker never tripped.

I don't know when this damage occurred. I opened the access panel last weekend to bleed the gas line to light the fridge for the first time this year and discovered the damaged wire and charred wood. It must have occurred sometime last season as evidenced by green corrosion formed on the copper wire. We never smelled anything hot or burning and it's a very small cabin inside.

If you note in the pictures, it was the middle conductor-- the green chassis ground wire that fried. That wire was never part of the block that plugged into the back of the fridge. It was screwed to the chassis of the fridge. I checked the cord with a continuity tester--- both the + & - conductors show no break from end to end of the cord. I checked the house outlets with my meter and they all seem wired correctly.

Besides that ground burning, the minus side of the 12V DC wires that run the gas igniter for the propane burner for the fridge melted too like the AC was looking for ground through it. The + side of the 12V wiring has a 5A fuse right there before it plugs into the fridge and the fuse was not tripped. The + side showed no signs of heating. But the negative wire jacket was melted from the fridge to where it was wirenutted to conductors coming into the compartment. The wire nut even melted.

The refrigerator operates fine in gas mode. I ran it for 24 hours last weekend checking it continuously and there was no heating of the 12V wires. Of course I had that AC line disconnected.

Note in the pictures that the + & - conductors appear to not have overheated.
It is that 3rd ground wire that shows overheating.

Any idea what could have caused this?

Thanks,

Mark [email protected]

Is the outlet or breaker eqipped with a GFI? I would think on a house boat they would be, but you don't mention it. If there is indeed a GFI, have you tested it recently?

Looks to me like there may have been a short from the hot AC lead in the fridge to the chassis, so that the ground was the return path to the breaker box. If you had a GFI ths would have caused it to trip.

Another thought - could the ground current have been come from some other external source? Hw about lightning? Or another appliance, extension cord, or something that could have come in contact with the chassis of the fridge? Any electrical sources on the house boat that are nearby, such as 12 volt DC? You can get a lot of current out of a 12 volt starter battery.

What looks like what happened is that you had a ground loop or ground fault.

One way that this could happen is if the 12 VDC ground is connected to the AC ground in more than one place. e.g. The generator and the AC panel.

It may also be related to how your breaker panel is set up.

The generator usually has neutral connected to ground and your AC fuse panel may also have the same connection. This should not be. The generator may have so

The 12 V ground is probably made at the breaker panel.

I can't download the manual from the slow link I am using now, but I'll take a look at it later.

I doubt the outlet expander had anything to do with it.

Chasing down ground loops can be difficult.

It may also have been a lightning strike.

I'll try to look at the manual of the EC 3000 tonight.

That manual states:

GROUND SYSTEM
Connections for standby power to a building electrical system must be
Made by a qualified electrician. The connection must isolate the
Generator power from utility power, and must comply with all
Applicable laws and electrical codes. A transfer switch, which isolates
Generator power from utility power, is available through authorized
Honda generator dealers.
Honda portable generators have a system ground that connects
Generator frame components to the ground terminals in the AC output
Receptacles. The system ground is not connected to the AC neutral
Wire. If the generator is tested by a receptacle tester, it will not show
The same ground circuit condition as for a home receptacle.

We, then therefor have to look at the electrical panel of the boat.

If that's the only panel and there is no dock connector, then the ground terminal of the generator would need to be connected to the ground terminal of the boat's panel.

In the boat's panel, neutral would be connected to the ground bar.

Thinks get messier, if there is a dock connector. In that case, you need a transfer switch that also switches the neutral.

Can you take a pic of the inside of the boat's panel?

I think I have the answer to what caused that wiring to burn.

One key incident was not originally considered --- During the Hurricane Ike wind storm last fall, the deck roof got ripped off the boat. There were two AC lines and a DC line for the Navigation lights that were bundled together and ran up into that roof. After the roof was ripped off there was a spaghetti of wires hanging out of the top of the boat.

Here is what I believe happened:

When the roof ripped off the DC + wire crossed with an AC ground wire in that wiring mess. It could have been actual wire to wire or there was a lot of metal trim and railing parts that they both could have touched.

Current flowed from the + side of the battery through 10 gauge primary wire to 14 gauge wire that went to the nav lights. There is no fuse in that run. There is a switch for those lights but it is on the negative side.

The current then jumped from that DC + to the 3rd wire on one of the 14 gauge AC lines. It then came down to the breaker box buss bar and out through the houseboat wiring.
It was able to find ground when it came in through the 3rd prong on the refrigerator plug, to the chassis of the fridge, to the 12V DC ground wire for the igniter also grounded to the chassis, and back to the battery on the 14 gauge wire.

The fridge AC wire and initial run of 12V wire for the fridge are both 16 gauge. So those were the thinnest conductors and most likely to burn.

It was a fluke, but the current by-past all of the safety precautions--- the 12V fuses and switch for those nav lights are on the ground side, it by-past the 15A breakers in the AC box, and it by-passed the 5A fuse on the fridge 12V line.

We were not on the boat during the storm so that explains why we never smelled any burning.

The day after the storm, I put the boat into storage for the winter. I never opened that inspection hatch for the fridge.

CaptMark

So, the "ground fault" fits as an explanation.

So, you now know the importance of fusing. Even cars have "fusible links" in them. In one of my previos vehicles, it was an actual piece of wire in the "fuse box" and it would burn in a controlled manner.

Fuses, circuit breakers should always be used, In a boat, vibration is always there and it's possible for a light to fail shorted.

I remember driving at night when I was a teenager and the headlamps would start blinking. The headlamps had a resetable thermal fuse. A later vehicle had fuses for each beam.

There are circuit breaker that have switches built into them that can be used as an indicator.

Your explanation is valid and it explains why the NEC won't allow low and high voltage in the same conduit. A 6" separation for parallel cables not in conduit is suggested for low voltage wiring like Ethernet and telephone.