Gert: What do you have at each end of the UTP cables besides transformers? There are a lot of emitteres bsides 50-60Hz in a hospital to cause such errors on data cables. Transients from elevators stopping and starting induce terrific transients in UTP cables.
Can you actually monitor the data at the terminal end by brodging a scope across it? Do you use the scope in differential mode in order not to disturb the balanced line? Perhaps the transformer coupling requires a grounded center tap or a connection to signal common. Check your cable routing to ensure proximity to high current carrying conductors is as far as possible. Those are just some possibilities. Good luck Ralph Cameron ----- Original Message ----- From: Gert Gremmen To: Emc-Pstc@Ieee. Org Sent: Friday, May 05, 2000 3:14 PM Subject: FW: common mode voltages in utp 100 Mbs network For some reason the message below did not make it to the list. And honest, there are no 4-letter words in it ! ;<) So heres a retry: Regards, Gert Gremmen, (Ing) Ce-test, qualified testing ================================== Web presence http://www.cetest.nl CE-shop http://www.cetest.nl/ce_shop.htm /-/ Compliance testing is our core business /-/ ================================== -----Original Message----- From: Gert Gremmen [mailto:[email protected]] Sent: Thursday, May 04, 2000 2:38 PM To: Emc-Pstc@Ieee. Org Subject: common mode voltages in utp 100 Mbs network Hello group, Does anyone have experiences with common mode voltages 50-60 Hz of high impedance that disturb the network traffic on a UTP Cat 5 10/100 Mbs network. ?? Both 10 and 100 Mbs traffic were affected. A customer complained about varying network performance on several places in a large hospital. As soon as I connected an oscilloscope to ground on one of the network conductors the problem disappeared (for the appr work station) The measure CM voltage was less then 2 Volts at 1MOhm to ground. Unloaded I suspect the CM-voltage to be much larger. Well this makes sense if the receivers of the network card are getting to much CM-voltage and are functioning only during the zero crossing of the CM-voltage. However, the modern utp Cat 5 network cards are all transformer coupled, so they should balance out the CM voltage. Could there be insufficient symmetry in the network cards line transformer to transfer part of the CM voltage into DM voltage ? And why network twisted pairs do not contain grounding resistors of about 1M ? I am not an network specialist, but helped out my customer using a handful of 1M resistors. I think that is highly unusual as a solution , least to say. Any clue will be highly appreciated as usual. Regards, Gert Gremmen, (Ing) Ce-test, qualified testing ================================== Web presence http://www.cetest.nl CE-shop http://www.cetest.nl/ce_shop.htm /-/ Compliance testing is our core business /-/ ==================================
<<Notebook.jpg>>
