A reverberant chamber CAN be made to do service for precompliance; we did at a former employer. However, one must rely on experience and trickery.
If you see a bump, move the antenna sideways a few feet. A reverberant chamber will have few peaks more than 6 dB above the actual value, but many deep, deep nulls, so you should move the antenna anyway. Test close; 1 meter. We ended up doing the testing with the operator inside the chamber, which was convenient for rotating manual tables, and using an older HP 141-T mainframe analyzer as the digital ones required modification to make them quiet enough. However, an 8590 only took a screen over the CRT and refurbishing the existing gasket by turning it around. We had to draw lines on the table for cable routing. Moving them away from the "standard" placement caused large difference in radiated fields. but this is PRE compliance, not the real thing. By using a standard cable set and routing, we were able to do good "delta" testing with results that proved out on later certification. A few cones or tiles strategically placed can _reduce_ though not eliminate resonances in the area of interest and this is always a help in reproducing readings. Just don't move the cones after you get them set up. Cortland ====================== Original Message Follows ==================== >> Date: 05-May-99 08:52:12 MsgID: 1068-2224 ToID: 72146,373 From: Patrick Lawler >INTERNET:[email protected] Subj: Re: Characterizing a screen room Chrg: $0.00 Imp: Norm Sens: Std Receipt: No Parts: 1 Thanks for the responses, everyone. I was probably misleading when I used the phrase 'characterize a screen room'. It sounds too much like 'calibrate'. I was (and still am) pessimistic that we can any get decent information from a screen room. I was hoping to at least figure out a way to mark frequencies where the room was unreliable. Enough said. In light of comments made (especially the one below that mentions 20dB variations), it sounds like pre-scans to _discover_ problem spots are invalid as well. - If you saw a bump, it may be a reflection/resonance. - If you saw a quiet area of spectrum, it may be a null. It sounds like the process should be: 1) Discover frequencies by going to an OATS and doing a valid scan. 2) Put the same test setup in the screen room, and move it around until you saw the same relative amplitudes (ignore the absolute amplitudes). Mark the location of that exact test setup! 3) Work on the unit to reduce the relative amplitudes. BTW, I am going to try measurements in the parking lot. If a screen room isn't purchased, there should be lots of money available for lounge chairs and sun screen (I live in Southern California). On Wed, 28 Apr 1999 11:14:38 -0500, Patrick Lawler <[email protected]> wrote: >At 04:21 PM 4/27/99 GMT, Robert Bonsen <[email protected]> wrote: >>My company is planning to purchase a screen room for radiated emissions >>precompliance testing. >> >>I'm aware that reflections can cause resonances and drastically influence >>readings. What kind of testing could I do to characterize the room (aside >>from simple experience)? >> >The simple answer would be this: don't even try. You're much better off >using the company parking lot to do pre-compliance radiated emissions >testing. For conducted emissions/immunity, and to a certain extent radiated >immunity, a shielded room is great. But not for RE. > >The reason for this is the reflections/resonances you get from the walls >and the ceiling. You can get higher than 20 dB ripples on your measurements >in an untreated (no absorber materials on walls/ceiling) shielded room. And >these ripples are not very repeatable, they will change considerably with >position (eg, moving your antenna or EUT less than an inch may result in >field variations of much more than 10 dB). Because of these huge >variations, testing cannot help you characterize your room and take these >reflections into account in your emissions measurements. > >If you absolutely need to use a shielded room, try lining it with absorber >materials. Even a few absorbers are better than none at all. Or try using >another type of pre-compliance device like a GTEM or something similar. >Another alternative would be to turn the shielded room into a mode-stir >chamber. By rotating the properly designed mode stirrer, you will even out >the variations which will result in fairly usable, repeatable numbers. The >size of the room determines the usable frequency range. -- Patrick Lawler [email protected] --------- 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). ====================== End of Original Message ===================== --------- 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).
