All,
We have been using a Comtest G310 0.7m x 0.7m x 0.7m Stripline cell for a
few years now. For those unfamiliar with a stripline cell, it is a cubic
chamber with a type-N coax connection on each side. The coax center
conductor is routed through the bottom of the chamber under a false floor.
The coax shield is connected to the outside of the chamber which is metal.
The chamber is split in half with hinges and latches so that it opens/closes
in clam shell fashion. The seam is gasketed and the entire outer shell of
the chamber is metal so that it forms a faraday cage when closed. The
inside of the chamber is lined with ferrite loaded foam to prevent
reflections. I believe that there is a crude description of a stripline
chamber in EN 61000-4-3 (IEC 1000-4-3).
We purchased this system with a signal generator, an RF amplifier and
software which allows us to perform radiated immunity testing in the
chamber. The chamber is specified to be able to perform fully compliant
radiated immunity testing on DUT's that fall within a 23cm x 23cm x23cm
imaginary cube centered in the chamber. We have used the chamber for this
purpose with good results.
However, I have also used the chamber for radiated emissions measurements.
This is where my question comes up. The chamber is not "calibrated" for
emissions. However, over the last few years, every time I have taken a DUT
to a 10meter OATS (at an outside test lab), I put the same DUT in the
chamber and connect the chamber's coax line to our spectrum analyzer. I
connect a 50 Ohm load to the coax connector on the other side of the
chamber. I then take spectrum analyzer readings from the DUT. I repeat
this process for all orientations of the DUT that I can set up. (i.e. face
up, face down , on its side ...) I use a software program to collect and
graph these readings. I have saved these spectrum analyzer readings as an
unofficial archive in my files. I have used these archives as a baseline for
many comparisons. I have used the archives to evaluate design changes on
previously tested DUT's. (I have even included the graphs as updates to our
Compliance Folders for DUT's that have been tested this way.) I have also
used these archives to estimate how completely new DUT's would perform at a
10m OATS. To me it seems almost too easy. There are no ambient emissions
in the chamber to cloud the measurements. I just get a little uptight when
things seem too easy. I start wondering if I'm missing something.
After having said the "easy" statement above, I should say that it is a
little bit of a pain monitoring the DUT through the window, and sometimes I
have trouble trying to loop the DUT's cabling into the "compliant" area of
the chamber. I am also limited by size and cabling interfaces. I can only
bring power and a couple of signal cables in through the filtered connectors
in the chamber wall.
Is there anybody in the group that would have an idea how accurate this type
of testing is? I'm not looking for somebody to calculate the +/-dB
uncertainty. I'm asking if there are basic physical properties (i.e. near
field vs. far field, inaccurate measurement of cable emissions, fringing
effects ...) that would make these types of comparisons risky? Is there
anybody who has tried to correlate these types of measurements to an OATS?
Are there ways to improve this test method (i.e. certain ways to route the
DUT's cables) that would make this type of measurement more accurate.
P.S. I do have an antenna to go with my spectrum analyzer, but our factory
is in the middle of a city. And I'm cramped for room. I have tried some 1m
prescan testing with my antenna with some success. The ambients are very
hard to work through.
Well there's some food for thought. Anybody care to bite?
Chris Maxwell, Design Engineer
GN Nettest Optical Division
6 Rhoads Drive, Building 4
Utica, NY 13502
PH: 315-797-4449
FAX: 315-797-8024
EMAIL: chr...@gnlp.com
-------------------------------------------
This message is from the IEEE EMC Society Product Safety
Technical Committee emc-pstc discussion list.
To cancel your subscription, send mail to:
majord...@ieee.org
with the single line:
unsubscribe emc-pstc
For help, send mail to the list administrators:
Jim Bacher: jim_bac...@mail.monarch.com
Michael Garretson: pstc_ad...@garretson.org
For policy questions, send mail to:
Richard Nute: ri...@ieee.org
