1 members (Scott35),
406
guests, and
30
robots. |
Key:
Admin,
Global Mod,
Mod
|
|
|
Joined: Nov 2007
Posts: 764
Member
|
Okay, but just to clarify things, most generators have the yoke of their 3-wire receptacles in direct contact with a metal panel that is also connected to the generator frame. There usually is no way to bond the receptacle grounding terminal to the neutral conductor of the generator receptacle without also bonding the frame and vice versa. There may be some that have this ability, but it’s not a very common feature from what I’ve seen. I don’t see why putting yourself between the line and "neutral" wouldn’t result in current flow or why giving it a second path to an independent "ground" couldn’t still allow enough flow to cause trouble for someone. My generator is pretty typical. Each side of the 20A/120V duplex is connected to one leg of the generator, so 240V between hots. With a 500W load [halogen work light] connected to one half of the duplex receptacle, I can read 56V AC from the hot of the other half of the receptacle with no connected load to an independent building ground using my Fluke 337 meter. It seems that the hazard could possibly increase when multiple loads are connected to the receptacles at the same time.
|
|
|
|
Joined: Jul 2004
Posts: 9,931 Likes: 34
Member
|
If neither winding of the generator is connected to anything but the current carrying conductors in the receptacle there is no path to the frame or to ground. The output is totally isolated from both. If you touch ther center tap of the winding that we think of as neutral and one of the other generator windings you have a fault but none of these windings will create a fault to the EGC pin, ground or to the generator frame
Greg Fretwell
|
|
|
|
Joined: Jul 2004
Posts: 787
Member
|
How much extension cord would be required to capacitively couple 5 mA of current thru a short between a phase and earth from a unbonded small generator? How much crappy abused cheap extension cable is required to leak 5 mA of current thru a defective power tool thru a sweaty person to a return path via the crappy abused extension cord? I guess it is all dependent on how much nannying we are willing to tolerate. In most cases, there is probably not a problem bonding or floating the neutral with respect to earth when using "small" generators. When it comes to connecting to an existing structure that is supposed to be NEC compliant, requiring the earth neutral bond will be necessary, if for nothing else, because of the size and variety of loads in the structure. Plus the potential for interaction with other services that may be subject to voltage transients such as cable TV, phone service, isolated water or gas/propane supplies, etc. Larry C
|
|
|
|
Joined: Nov 2007
Posts: 764
Member
|
I think there may be some confusion about what I’m saying. The last two paragraphs highlighted in the regulation below seem to describe what I’m driving at. This link has some pretty decent graphics too. Generator Regs Grounding Requirements for Portable and Vehicle-mounted Generators Under the following conditions, OSHA directs (29 CFR 1926.404(f)(3)(i)) that the frame of a portable generator need not be grounded (connected to earth) and that the frame may serve as the ground (in place of the earth): • The generator supplies only equipment mounted on the generator and/or cord and plug-connected equipment through receptacles mounted on the generator, § 1926.404(f)(3)(i)(A), and • The noncurrent-carrying metal parts of equipment (such as the fuel tank, the internal combustion engine, and the generator’s housing) are bonded to the generator frame, and the equipment grounding conductor terminals (of the power receptacles that are a part of [mounted on] the generator) are bonded to the generator frame, § 1926.404(f)(3)(i)(B). Thus, rather than connect to a grounding electrode system, such as a driven ground rod, the generator’s frame replaces the grounding electrode.
|
|
|
|
Joined: Jul 2004
Posts: 9,931 Likes: 34
Member
|
KJay, you are just saying the non current carrying metal is all bonded together, not that it is bonded to any current carrying conductor. As far as I am concerned, connecting this to the current carrying conductor, without actually grounding it, is creating an additional hazard.
Larry, I don't think anyone is saying you can connect a generator to a structure without grounding it.
Greg Fretwell
|
|
|
|
Joined: Jul 2004
Posts: 787
Member
|
If neither winding of the generator is connected to anything but the current carrying conductors in the receptacle there is no path to the frame or to ground. The output is totally isolated from both. I disagree with the statement that the output is totally isolated from both the frame or the ground. There will be SOME coupling. Is it enough to cause a shock or fatality? That depends. It depends on the state of the insulation of whatever is connected to the generator and the condition of the generator itself. It depends on the applied voltage to the insulation. It comes down to Who or What are we trying to protect? Larry
|
|
|
|
Joined: Jul 2004
Posts: 9,931 Likes: 34
Member
|
You say there is some coupling and then you say you will fix that by intentionally bonding it. Isn't that 100% coupled? At that point you have to make the arbitrary decision of which current carrying conductor you want to bond. Until you bond one, they are all the same and none of them has a particular affinity to the frame of the generator.
As I said before, small Hondas are not bonded and I have not heard of any problems with them.
Greg Fretwell
|
|
|
|
Joined: Jul 2004
Posts: 787
Member
|
I am stating that there is "some" coupling. How much is dependent on the individual generator and the connected load. By bonding the "neutral" to the frame/earth of the generator, the variability of this coupling is no longer a variable. With a known configuration, certain protective devices can act in a predictable manner. (GFCI) Are protective devices necessary? That is a different question. Is bonding of any of the output leads required or even desirable? It all depends on the design of the system. Will not bonding any output lead cause a problem? In most cases of using small portable generators, I think not. Can it cause a problem? Under the right circumstances, absolutely! Do I think the bond is required? Right now, as long as the generator is a truly independent system, I do not believe it is necessary. As soon as it gets connected to an external system, such as an fuel supply, battery charger, remote monitoring system, block heater, an external structure like a building, hard wired phone system, or anything else that makes it not independent, then a bond MAY be required. Larry
|
|
|
|
Joined: Jul 2004
Posts: 9,931 Likes: 34
Member
|
I agree with all of that and when the generator is connected to a structure it must be bonded. I suppose I just prefer 2 pole transfer equipment and leaving that bonding to the service main bonding jumper. When you start switching neutrals you open yourself up to all sorts of potential problems.
Greg Fretwell
|
|
|
|
Joined: Nov 2007
Posts: 764
Member
|
From what I'm reading, it sounds like the bigger shock hazard may not be from electrical so much, as from the possible fine OSHA may issue if someone brings a portable generator on a jobsite without the grounding terminal of the receptacles being bonded to the neutral or not having GFCI protection for them.
|
|
|
Posts: 61
Joined: August 2007
|
|
|
|