Sorry, i should have indicated there are TWO ethernet connectors. The loopback used at the test house is between the two different ports using a standard CAT5e cable. The connector with integral magnetics is supposed to include common mode rejection. Though a quick test with a network analyzer didnt show anything sectacular. CM Attenuation at 100MHz was around 40dB but at 500MHz it was more like 10dB. The board has no ground plane splits anywhere. It has no supply plane splits that active signal traces run over. The highest near field probed levels at 375/500/625MHz were detected directly above the PHY, not on the rx/tx lines coming from it (though it is definately evident there at higher levels than anywhere else on the board.) My reasoning as to why the signals are common mode is that i can also detect them with an inductive probe easily off the ethernet cable itself, but not off the power or serial cables from the board. Interestingly another product (using a different PHY) done some time ago showed some emissions at 250MHz (the power plane had an impedance hole near this frequency, which probably didn't help, but it passed) and a lot of spurs at 100MHz (which were very marginal pass with UTP cable. And that looks similar with this implementation. I have yet to deal with that). But the other product had everything else passing with considerable margin. Ken, Id certainly be interested in looking at the "image plane" idea. If i understand it right it involves having a separate "quiet" ground for all the IO connectors. That would be rather tricky in this case as one of them is an antenna connector that carries several watts at UHF. Splitting its ground is likely to create a number of design challenges, a significant one being trying to convince my peers to implement it! Regards, Bryan Ackerly On 06/04/2013, at 7:19 PM, <verma.nitya@xxxxxxxxx> wrote: > Hi Bryan, > > You mentioned that the test was done as loop back on the same Ethernet port > which means there is no cable attached to the port but only the loopback > connector. > Could it be possible that source of these emissions might be from the > Ethernet lines but they are coming out of the product via another medium? > > Is it that any other small cable wire or mesh is nearby which is picking up > from and radiating. May be if it is there you might want to isolate them? > > This is just a thought. > > Thanks and Regards > Nitya > > -----Original Message----- > From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx] On > Behalf Of Bryan Ackerly > Sent: Saturday, April 06, 2013 8:39 AM > To: keithK EPD; si-list@xxxxxxxxxxxxx > Subject: [SI-LIST] Re: Ethernet emissions. > ------------------------------------------------------------------ To unsubscribe from si-list: si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field or to administer your membership from a web page, go to: //www.freelists.org/webpage/si-list For help: si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field List forum is accessible at: http://tech.groups.yahoo.com/group/si-list List archives are viewable at: //www.freelists.org/archives/si-list Old (prior to June 6, 2001) list archives are viewable at: http://www.qsl.net/wb6tpu