Mistyfying Electrical Issue

[egaito]

I have found what I believe is a tie between the ground bus and the bonding system, but I'll have to crawl around a bit more to confirm that."

That is the correct configuration. The ac ground (green wire, NOT THE AC NEUTRAL), the dc ground and the bonding system are all connected.

In the panel with the rotary switch (gen/shore/off), I can see a green labeled "generator neutral" which I assume ties neutral to ground at the genset, when the genset is in use, which I've read is the correct way to do things.

A thunderstorm approached and offered me another clue. Since I had my hands in the panel, I physically disconnected the cables from the pedestal in case the dock took a hit (steel frame, tin roof) When I went back to the panel, my neutral/ground continuity was gone.

I'm not sure if this is correct....I guess if it's tied together on shore, my boat becomes part of that circuit, but should I see continuity, even with the breakers off on the pedestal?

 

[MikeP]

I have to admit I'm not really understanding exactly what the problem is. If its that the breakers are popping at 28 amps, that seems more or less OK - maybe weak breakers as someone pointed out.

Remember that the boat wiring and shore wiring are different. On shore, the neutral and ground systems are tied together, on the boat they are not. The genny has the neutral and ground connected together.

So again, what you are seeing sound normal.

I apologize if I'm not understanding what you are describing..

The Calder book describes all this stuff quite well and the oem Hatteras wiring diagrams depict the specifics connections as they were when the boat left the factory. Of course, who knows what they may be now...

 

[rsmith]

When I rebuilt my house after Hurricane Frances I was required to separate the grounds in my house panel from the neutral with an isolated buss bar. I was also required to increase the size of the conductor that goes to the ground rod. At the time I wondered how it would effect the boat as I have a 100 a branch circuit running to the dock. 9 years now and I haven't noticed any change in the zink life.
On a side note reading MikeP's block heater thread got me wondering about having an ac device inserted into the engine. I've had a # of water heater elements that shorted out and blew the breaker. Every one ,the casing on the heater element was ruptured and it appeared that the internals of the element were shorting to ground thru the water. If the engine block is not tied to the ground and only the bonding system could the bonding system become "hot" if the element fails internally???

 

[egaito]

I have to admit I'm not really understanding exactly what the problem is. If its that the breakers are popping at 28 amps, that seems more or less OK - maybe weak breakers as someone pointed out.

Remember that the boat wiring and shore wiring are different. On shore, the neutral and ground systems are tied together, on the boat they are not. The genny has the neutral and ground connected together.

So again, what you are seeing sound normal.

I apologize if I'm not understanding what you are describing..

The Calder book describes all this stuff quite well and the oem Hatteras wiring diagrams depict the specifics connections as they were when the boat left the factory. Of course, who knows what they may be now...

I may not have explained the breaker situation well. Each cord is on it's own 30 amp breaker. Pre-issue, if one side or the other is overloaded, that individual breaker would do it's job and let go. That only happened on this boat occasionally during the winter months when I'm using a space heater to supplement the onboard reverse cycle units. Now, if the sum of the boat's draw across both cables comes close to 30 (i.e. stove, microwave and hot water heater on the galley panel and AC on the salon panel), they both let go simultaneously. Its as if the two legs are joined in some way that they didn't used to be.

From what I've read, clamping the power cord should read 0, or at worst below 30 milliamps. I'm reading amps.

I wonder if my rotary selector could be going bad......

 

[Pascal]

"In the panel with the rotary switch (gen/shore/off), I can see a green labeled "generator neutral" which I assume ties neutral to ground at the genset, when the genset is in use, which I've read is the correct way to do things."

Doesn't sound original, the n to g bond is at the Genset not outside in the panel. Where does it go and is it possibly always on even when the rotary is on shore position?

Rotaries can go bad but I doubt that a the problem. I don t see how the hots of both legs could be connected without tripping breakers or a massive short, unless maybe the hots are on the same phase, but then if you d connect elsewhere on opposing phase you d have some serious smoke coming out. If you measure voltage at the pedestal across both hots, what voltage do you get? 220?

 

[LUV'N LIFE]

Thanks Jim. I've only heard that go off once since I've had the boat. never figured out what caused it, but it was years ago.

This is behind the original Hatteras panel. The other, newer panel has a light on the breaker.

I traced the wire that showed continuity from neutral to ground and found it was an outlet box in the ER, into which my ice maker is plugged in. Unplug the ice maker, continuity goes away. Simple enough, but upon attaching everything back onto the bus, once again I have continuity, so apparently there is more than one issue going on here.

A question about the ice maker: When I unplug the ice maker, and test across the three prongs of the plug, what should I see? I get continuity between hot and neutral, but not between the round prong and either of the other two. This doesn't make sense to me considering when it is plugged into the boat's system, I do get continuity between ground and neutral.

Ed, the round prong or safety ground will show continuity to the ice make case.

 

[madhatter1]

OK, I am gonna jump in on this one. Background is controls and power metering but I won't bill myself as an "expurt" on this as I am learning as I go. I had never heard of reading the entire cable with an amprobe as this does not make sense (based on what I know or think I know). Then the other thread pops up about electrical drownings and that artice says how to test yout boat by "clamping" the shore cord. I read that and still don't think it is right and here is why. When we meter power we are only concerned with the amps going through the phases (Hots). No amps are taken at the neutral as they don't matter (other than a reference for power factor). I will explain this best I can and will use ficticious ballpark numbers because it is not cut and dry. If you have 20 amps on the hot you will not have 20 amps on the neutral. The load "consumed" the power and it is now heat in some form. There is some amperage on the neutral (a residual amount) that is opposite the hot phase. So if the neutral had 5 amps it will show as 20-5=15 amps on the cord. My amprobe is broken but tomorrow I will get a new one and take some readings on my boat and house and supply some factual detailed readings on this. I will turn stuff on, meter the cord, meter the panel feeds, and the neutral. Also will meter ground in several locations.

Please feel free to call me out on anything that is incorrect. I have years of field experience but not one piece of paper (other than a contractor license) that says I am qualified to do what we do. Hopefully an electrical engineer posts right after I do to send us all in the right direction. Tuesday is a good day to do this as all the weekend rum is out of my system by then.

 

[Boatnut]

Ed,

Just a couple of comments for what they may be worth.

First don't you have a strange and often difficult problem to resolve.

Turning off bratkers will not help much, as ground does not go through a breaker, I doubt they wired your boat to pass neutral through a breaker. Typically only hot goes through the breaker.

There is often a difference of voltage between neutral and ground, the amout of voltage will vary depending on the length and size of the wires. Many times a safety wire is small that neutral.

When you disconnect a device, even if it checks out OK, leave it disconnected and continue trouble shooting. The ground leads are in effect in parrel, meaning that is they is high resistance in more that on lead, the resistance decreases. The lower the resistance the higher the current will be.

On the refridgerator plug, reading the plug on the appliance should never show continuity if it does there is a problem. You will read continuity between the 2 when the unit is plugged in, this is because neutral and ground are connected at the source.

Always remember nuetral carries current, ground never carries current. If you put you ampmeter on a ground wire and it shows current there is a problem.

I am speaking of 120v circuits, not 220v, do not get confused when you hear someone say the installations say connect neutral and ground together. They are most likely talking about a 220v system. Tying neutral to ground IMHO, is a bad pratice.

When on generator power, the same basic rules apply, the only difference is the genset is the source, treat it the same as you would dock power. More than one person has been severely shock or killed on genterator power.

One last thing, keep an eye out for missing grounds,, and especially missing neutral, when a neutral is open current will seek a path to ground any way it can. Be aware of what is connected where, like a reversed sockets someone with no experience could have wired incorrectly, we see this one a lot.

I hope this helps, I am working off shore presently, I will be traveling through Miami on my return, if you have not found it when I pass through, I'd be glad to layover and help you out.

GOOD LUCK

 

[egaito]

Ed,



I hope this helps, I am working off shore presently, I will be traveling through Miami on my return, if you have not found it when I pass through, I'd be glad to layover and help you out.

GOOD LUCK

More food for thought...thanks. Unfortunately, Misty is in Atlanta. Sanctuary, in Miami, has isolation tranformers, so hopefully, I'll never have to track anything like this down on her!

 

[Boatnut]

Atlanta? I had no idea there was an ocean in Atlanta, or is it hotlanta? lol

 

[egaito]

Atlanta? I had no idea there was an ocean in Atlanta, or is it hotlanta? lol

Lake Lanier isn't quite an ocean, but it's rougher than the ocean on weekends ;-)

 

[egaito]

Update:

Problem still not identified, but I did some further experimenting. I used one of the prototypes of the device I'm helping a guy bring to market that detects elc without shutting the boat down, and allows the hot and neutral to be clamped without the ground. Of course, it alarmed immediately when the breaker was turned back on, but that was expected. On the other cord, I installed a patch cable with the insulation cut back so I could put the clamp meter on any combination of the three conductors I wished.

Putting the clamp meter on all three conductors yield equal readings, in line with what the total draw of both legs is. Clamping the ground alone, yields about 40ma. Clamping the hot and neutral, without the ground, reads a bit lower than clamping all three. It's tough to say whether they all balance since readings are fluctuating a little bit (I'd need three clamp meters and really fast math to figure out if it's exact)

As it turns out, one of these test devices indicated a problem on another boat on the lake a few days ago. On a 50 amp cord, 18 amps was being read with the clamp meter.

To make a long story short, in their case it appears one of the mounting screws to a dryer outlet had been driven through the neutral and ground, joining them (it was a houseboat....not the first time something like this has been found). It had been this way for years, and fortunately no one was hurt by the leakage. It appears the ground wire did it's job.

I guess the next task is to physically start disconnecting every circuit from the panel until I find something

A photo of the test device we're building is attached. In most cases, when this device alarms, one can simply start cycling breakers to find the offending circuit.

 

[jim rosenthal]

Actually, on my 1971 Hatteras 36, each AC circuit is wired with two breakers, I think of equal amp rating; one breaks the hot, the other the neutral. I don't think they do this anymore, but they did then. I don't know what year Ed's Hatteras is, or if it has the double-breaker system.

 

[Boatsb]

What are you trying to read with that meter on the main cable?

Im pretty sure its properly used on a single conductor at a time.

 

[egaito]

What are you trying to read with that meter on the main cable?

Im pretty sure its properly used on a single conductor at a time.

Generally speaking yes, if you are looking for the load on that wire. If you clamp the whole cord, it shows the difference, which should be 0. Anything else indicates a leak somewhere.

 

[egaito]

Actually, on my 1971 Hatteras 36, each AC circuit is wired with two breakers, I think of equal amp rating; one breaks the hot, the other the neutral. I don't think they do this anymore, but they did then. I don't know what year Ed's Hatteras is, or if it has the double-breaker system.

The boat is a '68, but was completely refit in the mid 90's. One panel is factory, and the other was new at the time of the refit. Each has a single main breaker.

 

[MikeP]

"The boat is a '68, but was completely refit in the mid 90's."

Well, if there's a problem it's a pretty safe bet, IMO, that it dates from the refit.

 

[Boatsb]

mix 120 and 240 with the the cord and you could see a variance. Also the fact that you have 3 conductors positions differtdistances from the probe will make it less accurate.

I would check each conductor individually and do the math.

 

[Boatnut]

Generally speaking yes, if you are looking for the load on that wire. If you clamp the whole cord, it shows the difference, which should be 0. Anything else indicates a leak somewhere.

In theory you a correct, in reality it rarely if ever happens. What you want to see is a current level so low it will not be a dqanger. We run ground fault tests every time we install an invasive machine, the leakage current we look for is less than 10 nano amps. No doubt you will find the problem and resolve it Ed, you might pull out a few haairs along the way, but perserverance pays off.

I take no credit nor offer no credit for mis spelled wurds, this tiperiter cant spull tu gud.