"daestrom" wrote in message news:Koykb.33644$ snipped-for-privacy@twister.nyroc.rr.com...
No, the neutral MUST be bonded to the gounding electrode conductor. In a remote building such as the one you describe in 1), you treat grounding and neutral issues just like you would in a separately derived system.
What you are describing is a mutliple-fault condition. If this were to occur, you would indeed be in peril; the gradient from the g-rod to any point out further than several inches would be high enough to be a hazard. But then again your senario could just as likely (or unlikely) occur in a regular service and be just as dangerous. The only saving grace would be if you were lucky enough to have a nice copper water-service with a low impedance path out to a steel main with the utility grounded to it. I've seen dropped neutrals that were never even noticed because the water service was carrying the service imbalance (neutral) current. In one case (a bank), the only clue was copious green stains in the restroom sinks and toilets. All I had to do to prove it to the super (he didn't believe me) was place my clamp-on ammeter around the incoming water service. I've also seen a drop-neutral knock out a CATV distribution amplifier out at the street in an underground subdivision. The service imbalance current was going out on the 500Coax shield. While I'm in the story-telling mode, let me tell you how years ago, before I had my own business, four people - including my boss, electrical inspector, untility engineer and a journeyman - made mistakes that almost killed a cuctomer: The company I worked for had wired a residence that was built over a crawl space with a rat slab. The main panel, which also served as the service disconnect, was installed in a utilty closet located near the middle of the house. This was an underground 200-amp service with the meter socket mounted on an H-frame near the begining of a 100' driveway. The service entrance conduit was run along the rat-slab about 20' in the crawl and then up into the bottom of the panel in the closet above This conduit was really just a sleave because after the service entrance left the building it was direct-burial. This was unacceptable to both the AHJ and LILCO engineer because the conduit was run too far into the structure without an OCD protecting it (we have a 5' rule in our area; the code simply says the disconnect means must be as close as possible). Well the right way to have installed such a service would be to have installed a 200-amp combination meter-main out at the driveway and run four conductors into the house (L1, L2, N, ECG) or encase the conduit in the crawl with at least 2" of concrete. Now when this violation was picked up the whole house had be trimmed-out with all appliances installed. My old boss, the local inspector and the LILCO engineer decided the concrete approach was impractacle as well as adding a third conductor for the ECG (remember the service was direct burial outside of the foundation). Instead, they arrived at what they felt was a decent compromise - not a good word when dealing with the safety of electrical installations. It consisted of pulling out the meter socket and replacing it with a meter-main, leaving the 3-conducors as-is, driving a separate set of G-rods out at the meter location, and finally converting the panel from a main to a sub-panel by removing the main bonding jumper (a green screw that goes through the neutral bar and into the box frame) and, finally, separating the ECGs (branch circuit grounds) from the nuetrals. Sounds bad enough does it? It gets worse. Here's were the journeyman that performed the panel conversion came in: The house had a Jenn-Air electric indoor down-draft grill which was fed by a 6/2 SEU service cable. When he moved all the ECGs off the neutral bar and onto a newly-installed ground bar, he saw the bare stranded wire neutral of the 6/2 SEU and mistook it for an ECG. He diligently moved it to the now-floating ground bar! This why the NEC only permits the usage of service cable for appliances only when it originates from a "main" or service panel - not a sub-panel. The house passed inspection and no one living in it has any idea of the peril. One day I show up to work and was sent out to check two problems: the fan on the Jenn-Aire was not working and everytime someone steps out of the shower, they get belted. Funny thing though: they lived in that house for a month and never got belted getting out of that shower. How could that be? Simple. The problem never showed up until someone tried to turn on the Jenn-Aire fan, found it didn't work and walked away leaving the switch in the "on" position!
Never a bad idea!
Paul