Dan, If the (#4/0) loop under the floor is tied to the main ground bar in each electrical room currently what good would it do to run a separate loop external to the building and tie it in to the same ground bar in each electrical room?
Thanks for your help on this.
Ed
There's a historical electrical experiment which demonstrates something interesting.
http://en.wikipedia.org/wiki/Faraday_cage
"In 1836, Michael Faraday observed that the charge on a charged conductor resided only on its exterior and had no influence on anything enclosed within it. To demonstrate this fact, he built a room coated with metal foil and allowed high-voltage discharges from an electrostatic generator to strike the outside of the room. He used an
electroscope to show that there was no electric charge present on the inside of the room's walls.
Although this cage effect has been attributed to Michael Faraday, it was
Benjamin Franklin in 1755
who observed the effect by lowering an uncharged cork ball suspended on a silk thread through an opening in an electrically charged metal can. In his words, "the cork was not attracted to the inside of the can as it would have been to the outside, and though it touched the bottom, yet when drawn out it was not found to be electrified (charged) by that touch, as it would have been by touching the outside. The fact is singular." Franklin had discovered the behavior of what we now refer to as a Faraday cage or shield (based on one of Faraday's famous ice pail experiments which duplicated Franklin's cork and can). "
I think of the inside and outside grounding conductor loops like this above. I would not say the inside loop would be free from charge and current flow present on the outside loop. I would expect some noise or signal would travel everywhere. However, the bulk of the circulating ground loop current will take a preferred, lower impedance path, especially a dedicated path provided directly for it.
Two different grounding paths for two different problems. At the services and UPS rooms, the grounding paths have circulating currents because of the nature of the equipment. Connection on that grounding path can be expected to see or take a share of the circulating current, which is done as necessary but not as a basis for communication signal referencing.
In the raised floor area, the problem is to provide an earth ground that also serves as an equipotential 0 volt reference for communication signalling, to prevent stray circulating currents from also flowing on the communication wiring, and to meet 250.30 (A) 3 or 4, grounding separately derived power sources in the IT space.
The experiment above is showing charge present on the outside of the conductor transfers preferentially to the outside and not to the inside. That's just how I would visualize the implementation, there's going to be significant noise and I would provide a path for that. The raised floor grounding loop, I could only hope the noise prefers to travel on the outside path and not significantly through the IT space. Note I say hope and not something more deterministic. Trying to gain an edge by choosing the path for the noise and choosing the point where the IT equipment is referenced to the earth.
Newer IT equipment has much better immunity from stray current flow on the grounding paths. The older stuff has a high rate of failure to operate from noisy grounding.
