Stuff, I don't get your examples of things gone wrong, it sounds like your saying the hot is shorted and you pick up the tool and it the current now flows through you to ground, cause you don't have a ground line, But the second that starts the GFCI would trip.
Why is a ground wire required from a generator?
- Thread starter Billy_Bob
- Start date
Users who are viewing this thread
Stuff
Well-Known Member
What I am saying is that in that example the GFCI would not trip. You would get shocked.
Sponsor
Paid Advertisement
Jadnashua
Retired Defense Industry Engineer xxx
Even without a ground, a GFCI will work just fine. For current to run through you, there has to be a voltage difference, and line voltage is high enough so that just standing , your skin and contact with the ground can do it if there's a problem. That's why you can legally install a 3-prong receptacle on a 2-wire circuit (you have to put a label on it) and it will still function fine. If some current ends up going through you, it will not be keeping the hot/return paths equal, which is why it trips. If you were well insulated from some other path, yes, you could touch a line and not be hurt, but if there were any voltage differences, that could induce current flow, and then problems occur.
In the scenario where when the ground wire is connected and a GFCI trips, but does not when the ground is installed, that implies that there is a connection between the ground and neutral wires, and some of the current is not returning via the neutral, thus unbalancing the GFCI, which trips it. It also implies that if you touched some grounded point on one of those devices it is powering, you could become part of the circuit, and current could travel through you as well, which also would trip the GFCI. The fact it hasn't tripped (yet) when you remove the ground implies that you just haven't touched the defective device or if you did, you were well insulated, and there wasn't enough current to trigger the GFCI to trip. I'd bet if you put a quality ammeter on the ground wire, you'd measure current, which should never occur unless there's a fault. Remember, it only takes 4-5 ma 0.004-0.005A to trip one of them. There should never be current on the ground/safety wire until there is a fault. The GFCI is designed to detect that problem with grounding, but it's indirect...earth, even if not perfect, can still act as a ground, and current can travel through it. Throw in potential problems with bonding to gas, sewer, water pipes, and touching one of them can be a source of problems.
In the scenario where when the ground wire is connected and a GFCI trips, but does not when the ground is installed, that implies that there is a connection between the ground and neutral wires, and some of the current is not returning via the neutral, thus unbalancing the GFCI, which trips it. It also implies that if you touched some grounded point on one of those devices it is powering, you could become part of the circuit, and current could travel through you as well, which also would trip the GFCI. The fact it hasn't tripped (yet) when you remove the ground implies that you just haven't touched the defective device or if you did, you were well insulated, and there wasn't enough current to trigger the GFCI to trip. I'd bet if you put a quality ammeter on the ground wire, you'd measure current, which should never occur unless there's a fault. Remember, it only takes 4-5 ma 0.004-0.005A to trip one of them. There should never be current on the ground/safety wire until there is a fault. The GFCI is designed to detect that problem with grounding, but it's indirect...earth, even if not perfect, can still act as a ground, and current can travel through it. Throw in potential problems with bonding to gas, sewer, water pipes, and touching one of them can be a source of problems.
Last edited:
http://cdn.powerequipment.honda.com/pe/pdf/manuals/00X31Z037140.pdf says
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.
Somebody may have jumpered yours, or there may be an accidental connection.
Or Honda has changed the design since that maual was printed in 2007, Honda has about 7 different manuals for this model depending on serial number here is the one for mine: http://cdn.powerequipment.honda.com/pe/pdf/manuals/00X31Z037230.pdf
Where on page 30 it says
System Ground
This generator has a system ground that connects generator frame components to ground terminals in the AC output receptacles. The system ground is connected to the AC neutral wire.
Stuff
Well-Known Member
A GFCI without a ground works in your house because ground and neutral are connected at the service disconnect. That provides a "return" path for electricity when you touch hot while touching "ground" which then causes the imbalance that the GFCI needs to trip. With an ungrounded (improperly installed) generator the return current comes back on the neutral and the GFCI never sees an imbalance so does not trip.
Jadnashua
Retired Defense Industry Engineer xxx
All a GFCI does is measure the power going out on the hot lead and compares it to the return. IF they aren't equal within the design limits, it trips...plain and simple. HOw that power gets diverted is irrelevant. A GFCI will work just fine, even if the ground and neutral at the panel were never connected, or if there's no ground in the system anywhere at all. The 'ground' fault is describing a path for power that is not supposed to happen. How it happens is irrelevant, but from a safety viewpoint, when it occurs, it is a safety issue, as there should not be power traveling to ground...it should be following the power lead path - out one, back the other, not be diverted somewhere else which might be through you.
We are in agreement that without the ground path wired up, it is harder for the GFCI to trip, as seen in the OP's situation. But, the fact that it does when the ground is attached implies there is a system fault somewhere that is allowing power to be diverted to the ground lead when it shouldn't be. Note, if you were to touch or find that errant point, even with the ground disconnected, the GFCI would still trip, as then, there would be an alternate path for electricity. It could go from the chassis of one device to the other if they had the ground and neutral bonded together inside of them.
There is a system problem with one or more of the devices on that circuit that is tripping the GFCI.
We are in agreement that without the ground path wired up, it is harder for the GFCI to trip, as seen in the OP's situation. But, the fact that it does when the ground is attached implies there is a system fault somewhere that is allowing power to be diverted to the ground lead when it shouldn't be. Note, if you were to touch or find that errant point, even with the ground disconnected, the GFCI would still trip, as then, there would be an alternate path for electricity. It could go from the chassis of one device to the other if they had the ground and neutral bonded together inside of them.
There is a system problem with one or more of the devices on that circuit that is tripping the GFCI.
But the Generator has the ground and neutral connected just like the house, so how is this different?
Yep. Pretty clear... there is the sentence you quoted, and there is the paragraph above.
It appears they don't change model numbers when they make major changes.
I guess I am the one who introduced what now has me confused, as people keep referring to my situation or the condition in my post. But I described 4 different situations in my post, so I never know which one is being discussed. Let me try to clear up the mess I think I have made.
Knows in all situations; Generator has ground bonded to neutral. house has ground bonded to neutral. 3 appliances (Furnace, normally hard wired to house, Freezer chest normally running on GFCI, Ref/Freezer unit not normally on GFCI), Furnace has outlet at it for powering condensate pump.
Using generator as is to back feed house through a 2 way transfer switch will cause GFCI to blow, as a loop of neutral and ground is formed which splits current causing a current difference between hot and neutral. Recommended practice is to either remove bonded neutral and Ground at generator label as such and not use on any construction sites as it is no longer OSHA approved. Or put in a 3 way transfer switch to break the house ground.
The 4 conditions I went through
1) Turned off main, flipped every circuit breaker off. Diconnected ground wire from outlet back to panel at furnace plugged in generator to plug, GFCI popped, (ground wire from outlet went into furnace and terminated at screw attached to frame, no other ground wires internally attached at that screw), removed that ground wire from outlet into furnace, and everything worked fine. -- Not sure why generator GFCI was popping. Furnace does power a DC thermostat???
2) Plugged in 2 extension cords which ran to Freezer and Ref/Freezer. Everything ran fine.
3) Ungrounded furnace plugged into generator, went to add Ref/Freezer and Freezer extension cords Ground plug went in first and the minute it did GFCI popped. -- I read somewhere that plugging in ungrounded and grounded devices into a GFCI outlet can cause it to pop, there was no explanation as to why this would happen. So why does the Ungrounded furnace which works fine alone, and the Grounded Ref/Freezer and Freezer which work fine alone on extension cord, blow the GFCI when plugged in together?
4) I then got two 3 prong to 3 prong addapters, such that the freezer and ref/freezer unit extension cords no longer had grounds and plugged them and the furnace into the generator and everything worked fine. -- Since the devices are plugged into GFCI outlet it would seem if there was a short in one of the appiances makeing the skin hot, the minute it was touched current would flow through the person to the ground they were on and cause the GFCI to blow. The only risk is if you somehow connected the hot and neutral with your body such that the current was flowing through you and returning to the generator, but in this case having a ground wire there would not save you anyhow. Can someone explain the situation where being on a GFCI without ground puts you at risk of shock?
At this point, I think the risk in 4 is minimal, and I really don't understand why what is going on in 1. nor do I understand why 3 causes it to pop. If I could understand what is happening in 1 and get the furnace online grounded I feel pretty good it would solve 3 and eliminate the need to run situation 4. But I also don't know why running ungrounded and grounded devices at the same time on GFCI causes it to pop.
Jadnashua
Retired Defense Industry Engineer xxx
Your furnace is (probably) running on a 240vac circuit. It may or may not have a neutral, but it will have a ground (or should). A resistance of 24K or less, will produce a current of 5ma or more with a 120vac supply. It doesn't take much in a device to get that much current to flow. Power does not have to flow back to the generator directly, but it must flow somewhere. Throw in the fact that the devices in question are sitting on the ground, even running from one device to another before it can get back to the generator will unbalance the flow. The power on one of the 120vac devices, might be flowing back on the other leg of the 240vac furnace. If you were to change which leg the devices are connected to, you might find a combination that kept them in balance and not trip the GFCI. For a GFCI to trip, there must be in the order of 4-5ma of current that is not going where it is supposed to. In the real world, finding that exact path can be tricky. Some devices internally bond neutral to ground, so it could be there without you making that connection externally. The older the device, the more likely there could be some carbon dust, or crud that could allow some current to leak to places you wouldn't expect it to. A little corrosion can divert power. The generally higher humidity levels in a basement can become problematic, especially when you throw in a concrete slab (one reason why GFCIs are required there, including outdoors).
IF you were using a shared neutral circuit, the power on the neutral might not cancel out - in fact, it would be rare that both legs were perfectly matched, and that can create an imbalance on hot-neutral that a GFCI is seeing, depending exactly on how things are wired.
The fact that the GFCI trips as soon as the ground prong is inserted implies ground and neutral are connected in the devices being powered, and the two 120vac devices are flowing some current between them via that alternate path. Are the two things next to each other and touching? Just sitting on the basement slab may be enough of a path.
IF you were using a shared neutral circuit, the power on the neutral might not cancel out - in fact, it would be rare that both legs were perfectly matched, and that can create an imbalance on hot-neutral that a GFCI is seeing, depending exactly on how things are wired.
The fact that the GFCI trips as soon as the ground prong is inserted implies ground and neutral are connected in the devices being powered, and the two 120vac devices are flowing some current between them via that alternate path. Are the two things next to each other and touching? Just sitting on the basement slab may be enough of a path.
No the furnace is not 240 its 120, on a 120 breaker, I know the difference, there is one white, one black and one copper wire going to it.
The ref/ Freezer is upstairs the furnace is in the basement, the Ref / Freezer runs on the GFCI by itself perfectly fine, which says the ground and neutral are not bonded in the device. Its only when plugged in at the same time as the ungrounded 120v furnace does it pop the GFCI breaker.
