Yoopersup
I started out in electrical work as a laborer with a power production crew in the Air Force. This issue was the one that always caused the most discussion. The practice that developed out of necessity was to not bond the neutral at the generator because it did not transfer at the switch. If the neutral does not transfer at the other transfer switch then the generator is not in fact a separately derived system because it's neutral is bonded to ground and to the power companies neutral at that other transfer switch. If it is not an SDS it should not have been bonded at the generator.
Leaving the code alone for the moment what you don't want is any neutral current using the EGCs as a normal current path even under a single failure scenario. What you have now is a neutral from the generator running to the other emergency panel and an equipment grounding conductor or EGC running to the fire pump transfer switch. Bonding the neutral of a non SDS generator to the frame was the error we had to get corrected on almost every generator set prior to acceptance. The project managers never buy anything that is not specified or required at inspection. As a result they never buy the neutral isolation kit from the generator manufacturer and they never order the forth pole in the transfer switch. Since the generators are shipped with bonded neutrals the installers leave them that way because they don't have the neutral isolation kit on hand and it is a rare heavy industry crew that will have someone that will spot the error.
I have installed both SDS systems and common neutral systems from Uganda to french frigate shoals and from the Argentine pampas to Alaska. As long as I kept clear in the documentation and in my head whether or not the system in question was a separately derived system I always got acceptance and passed inspection. That many engineers and electrical inspectors can’t all be wrong. All you need to do is isolate the neutral of the generator from the frame and leave that common neutrals only bonding point at the other transfer switch alone. The only concern that leaves me with is I cannot trace a low impedance fault current path from the case of the fire pump motor back to the neutral of the generator in order to clear a fault under generator operation of the fire pump. What may be true is no such fault clearing path is wanted because fire pumps should run to destruction rather than fail to pump the fire protection water that is the whole reason for their existence. I just get concerned that this mite leave a hazard to personnel. What mitigates against that is the fact that the fire protection piping is very likely to be a large metallic system that includes underground piping that will have been used as a grounding electrode for the buildings utility supplied service the neutral of which is common to the generators neutral. Even though that mite well work as a fault clearing path I’d much rather have the fault clearing pathway be electrician built and maintained.
--
Tom Horne
"This alternating current stuf is just a fad. It is much too dangerous for general use." Thomas Alva Edison
[This message has been edited by tdhorne (edited 10-04-2006).]
