That all makes sense. Thanks for the analysis! Every time I do (or=20 say, in this case) something really wrong, I always get more information = than when I only go a little wrong. :-) -Adam Joe Geller wrote: >NO, this is wrong. (bonding neutral to GND at a subpanel in an out buil= ding) 1) If the neutral is riding "several dozens of volts" above ground,= the wire gauge is too small for the load. 2) The very point of the gro= und is to provide a safe working surface such as the cabinet. If you hav= e an exposed shock hazard it is from improper grounding. 3) MOST importa= nt - If the wire gauge was so small that the neutral is riding so high a= bove ground, then when a large fraction of the normal 'return' current (w= hat should have been AC neutral) is flowing now through ground (in the im= proper installation), the ground potential at the cabinet can rise too! = 4) If the circuits get really corrupted, where what should have been curr= ent in the neutral line, is now flowing into the gnd (earth) spike at the= garage ... it can be flowing back to the house circuit through the earth= between the buildings causing voltage potentials between your earth poin= ts, not good. > >The point is, NEVER use the ground circuit for normal return currents in= place of the neutral circuit, which is what the improper installation in= effect does. Ground circuits ideally have low impedance to the system = ground, but little to no normal return current flow (what should be neut= ral current in single phase power circuit). > >On Wed, 25 Aug 2004 12:44:42 -0400, M. Adam Davis wrote: > =20 > >> I would bond them together. There will be a ground loop, but if >> you don't bond them then neutral will be several or dozens of volts >> away from ground - not good. >> >> The problem gets worse with an intermittent or poor ground. I used >> an extension cord to power a computer across a basement once. The >> ground was bad somewhere along the line, and the computer case >> ended up floating and giving a nasty shock to anyone who touched it. >> >> If you bond the ground wire, the ground loop stays in wiring and >> not in equipment (and people) if you don't bond it then you have a >> potential problem since the ground loop may terminate in one of >> more pieces of electrical equipment (or they may push ground and >> neutral further apart). >> >> Treat the remote panel like a regular circuit breaker box - ground >> rod, bonded neutral. >> >> I am not a licensed electrician, though, so follow my advice at >> your own peril. >> >> -Adam >> >> Matt Redmond wrote: >> >> =20 >> >>> Please no flaming about inspectors, etc... that we got off-topic >>> on last time (the 'what cable gauge?' discussion)! I have >>> another question about wiring my detached garage - this one is >>> about grounding. >>> >>> The way I read things, I have two acceptable choices when it >>> comes to grounding in the detached garage. They are: >>> >>> (a) Run H-H-N-G from main panel to garage. Do NOT bond N-G in >>> the remote panel. >>> >>> (b) Run H-H-N from main panel to garage. DO bond N-G in the >>> remote panel. >>> >>> In either case the garage gets its own earth ground. >>> >>> Any opinions on which is the 'betterer' approach. Any advantages >>> to one over the other? I'm already going to be running 3- >>> conductor cable + ground - any reason not to use (a)? >>> >>> Thanks! >>> >>> -matt redmond >>> >>> -- >>> http://www.piclist.com hint: The list server can filter out >>> subtopics (like ads or off topics) for you. See >>> http://www.piclist.com/#topics >>> >>> >>> =20 >>> >> -- >> http://www.piclist.com hint: The list server can filter out >> subtopics (like ads or off topics) for you. See >> http://www.piclist.com/#topics >> =20 >> > > >-- >http://www.piclist.com hint: The list server can filter out subtopics >(like ads or off topics) for you. See http://www.piclist.com/#topics > > > > =20 > -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics