Don: One thing that hasn't been mentioned so far is that you should place your room ground stud as close as possible to the powerline filters. If your room is a modular design (and not double electrically isolated), then the room is actually an assembly of conductive panels with clamp plates bridging the gap between the panels. You should be sure to put the ground stud on the same panel that carries the powerline filters.
First, for safety, you don't want the capacitive leakage current from the filters flowing through the mechanical room joints as it heads to the ground rod. Second, the filters are removing RF noise currents from the incoming power lines. These currents are shunted to the case of the filter and flow onto the conductive surface of the room, following the same path to the ground rod as the power frequency leakage current. If you allow these RF currents to flow across the impedance discontinuities of the clamp plates, then there is the possibility of radiation at those discontinuities. The result is equivalent to degrading the shielding effectiveness of your room. You can take some comfort in seeing what my facility looks like. I have a large and a small modular room. The small room is permanently affixed to the large room with a 24" x 24" duct (or tunnel) facing into a penetration port on the side of the large chamber. So, I have two rooms whose outer skin is electrically common, and provides RF shielding from the outside world. Plus, the rooms are RF isolated from each other. The large room has 2 filter lines for DC, 4 filter lines for 60Hz 3 phase, and 4 filter lines for 400Hz 3 phase. Per the NEC, each three phase AC is a 5 wire Wye configuration, with a safety ground running back to the distribution panel. The powerline filters are at about eye level on the outer wall of the room. Immediately below the filters, I drilled two separate holes in the concrete floor for two ground rods. Each ground rod has a short 2" wide braid connecting to the ground stud on the room wall. Now, just to complicate things, my small room also has 60Hz and 400Hz 3 phase power (similar 5 wire Wye feeds). But for these, I was able to negotiate a compromise with my facilities engineer. The 3 phase lines and the neutral line for both the 60Hz and 400Hz are wired from the distribution panel as you would expect. But, the safety ground for both systems loops down the parent room wall (in conduit) to a disconnect box, and then back up to the ceiling to rejoin the other wires and proceed to the small room ground stud. The disconnect box houses a 3 pole double throw switch. In one position, two of the switch blades separately bridge the 60Hz and 400Hz safety grounds, providing a second, fully compliant 5 wire feed for both 60Hz and 400Hz feeds. In the other position, the switch breaks the safety grounds to the small room filters (and also turns on an embarrassingly bright red warning light visible to all). The rationale is that the rooms are one continuous structure, and two safety grounds are always connected at the large room filters. Plus, there are also two ground rods attached to the room. I am allowed to disconnect the second safety ground set only during my physical presence in the lab (the lab is a locked restricted area) while performing a test. (And yes, as I expected, the facilities electricians were watching me to make sure I didn't cheat.) The funny thing is that I don't see the small room ground system causing any problems, even over the extended military spec frequency range. So I usually just leave that fancy switch in the closed position, thoroughly violating single-point grounding. Finally, as Lou mentioned, you have to be careful of surprise ground paths on the instrumentation you connect to the room. The coax back to an RF power amplifier, and then the coax back to a synthesizer, and the coax back to your oscilloscope might carry a current through to the oscilloscope safety ground. (I'll salute anybody who has never smoked a scope probe ground lead!) This same effect can introduce ambient signals into your emission data if RF currents flow along your signal coax from the room to the spectrum analyzer. I always use an isolation transformer with my spectrum analyzer, and I use a very heavy braided strap to connect the chassis of the spectrum analyzer to a ground point at the room penetration port. Regards, Ed > -----Original Message----- > From: [email protected] [SMTP:[email protected]] > Sent: Wednesday, November 03, 1999 6:25 AM > To: [email protected] > Subject: FW: Chamber Grounding > > > > > Our chamber is grounded/isolated per the instructions of the vendor. We > > have one copper clad ground rod installed through a hole drilled in the > > slab adjacent to the corner of the semi-anechoic chamber. Other grounds > > are isolated from the chamber (conduits, air pipes, water pipes, service > > entrance safety ground, etc.). The ground comes from the ground rod, > not > > the service entrance. > > > > 1) Is "single-point-ground" as described above for Tempest? Is the > > degree of isolation useful for typical commercial work? The chamber > spec > > is 100 dB isolation. For our immediate work, 60 dB of isolation is > > adequate. Is there any correlation between chamber isolation and > > effectiveness of the ferrites for the uniform field required for > immunity > > testing? Is there some other observable sensitivity such as degraded RF > > measurements that would result from not observing the isolation? What > > would the manifestations be? > > > > 2) Using the isolation as described above, has anyone experience ground > > loop problems between the service entrance power and the local chamber > > power distribution due to the "single point ground" concept defined > above? > > > > > > > > Don Umbdenstock > > Sensormatic > > > :-):-):-):-):-):-):-):-):-):-):-):-):-):-):-):-):-):-):-):-):-):-):-):-):-) Ed Price [email protected] Electromagnetic Compatibility Lab Cubic Defense Systems San Diego, CA. USA 619-505-2780 (Voice) 619-505-1502 (Fax) Military & Avionics EMC Services Is Our Specialty Shake-Bake-Shock - Metrology - Reliability Analysis :-):-):-):-):-):-):-):-):-):-):-):-):-):-):-):-):-):-):-):-):-):-):-):-):-) --------- This message is coming from the emc-pstc discussion list. To cancel your subscription, send mail to [email protected] with the single line: "unsubscribe emc-pstc" (without the quotes). For help, send mail to [email protected], [email protected], [email protected], or [email protected] (the list administrators).
