Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

Re: RJ-45 common mode radiated emissions testing

Title: RJ-45 common mode radiated emissions testing
Rick, thanks for providing the test points.  I have more questions than comments (probably the general response to someone that has just done a pot load of work).  I notice that the tests indicated that your transmitted voltage levels of about 50 - 70 db above a microvolt caused the radiated emission to fail class B by about 30 to 50 dB, depending on the data point.  We all know that there is very loose coupling between power on the lines and radiated power, since there is som much that is dependent on geometry.  However, you have created a nice stake in the ground.
Based on the measurements you have made, would you care to make any very rough estimates of the power levels that might cause radiation problems, and how those same power levels would impact noise.  I would dearly love to be able to bound the problem of how much noise might be transferred to the signal lines from powering the signal pairs, and was wondering if you have any insights that might tie some of your measurements back to this problem?
Thanks again for the good work Rick!
Best regards,
Robert D. Love
President, LAN Connect Consultants
7105 Leveret Circle
Raleigh, NC 27615
Phone: 919 848-6773
Fax: 720 222-0900
email: rdlove@xxxxxxxx
----- Original Message -----
Sent: Monday, August 28, 2000 11:51 AM
Subject: RJ-45 common mode radiated emissions testing

I have posted the test setup and results of a radiated emissions test that I did recently in a 10 meter anechoic chamber.

The test was to try to determine what common mode signal voltage would produce a radiated field that would be at the
compliance limit, class B in this case.
I hope that this data will help us to specify a noise requirement for the DTE power from the PSE, as well as the PD.

It is important to understand that the radiated emissions requirements only cover from 30MHz and higher.
The spectral region from 500 kHz to 30 MHz, is obviously of importance to 10/100/1000, but I don't cover it here.

The pdf file is at:

comments, questions?

- Rick