0
\$\begingroup\$

Ethernet up to 100Mbps uses only 2 of the 4 twisted wires in an ethernet cable while gigabit is using all of them. Connecting an LED to one of the gigabit-only pairs should therefore make it easy to detect a gigabit transmission.

Well, actually not. Problem is, that PoE may or may not be used, the ethernet data is AC while an "this is gigabit"-led is DC, ethernet has no fixed current and I have no clue what I'm talking about.

I really ain't versed in electronics. A rectifier between ethernet and the LED would solve the AC problem, but this can't be the whole solution. Ethernet has no fixed voltage as described in here. Is there a module that counters fluctuation in a way that the LED gets the power it needs? Even with the low energy from a ethernet cable? If yes, would this work with PoE injection on the unused wirepairs as well?

Background: An intrusion detection system based on Microsofts ATA has to have a physical connection to the network so I've built a network tap with a maximum throughput of 100Mbps. I forward only two of the wirepairs, forcing any network clients to use a maximum of 100Mbps. This may obviously cause a giant bottleneck, but while not preventable, I would at least like to know when this is a problem. A litte status LED would be nice, flickering upon gigabit data connection.

First post and glad about criticism. Thanks

\$\endgroup\$
  • 3
    \$\begingroup\$ Cables have wires, not veins ... \$\endgroup\$ – brhans Jan 22 at 15:14
  • \$\begingroup\$ Do you mean 100 Mbps rather than what you said (100mb)? \$\endgroup\$ – Andy aka Jan 22 at 15:16
  • \$\begingroup\$ jep. mbps. Sorry for the veins \$\endgroup\$ – catchMyException Jan 22 at 15:22
  • \$\begingroup\$ Attaching anything to the data lines will result in signal degradation... What is the point of forcing 100Mbps? \$\endgroup\$ – MadHatter Jan 22 at 15:32
  • 1
    \$\begingroup\$ This question is off-topic because the user has no understanding of RF effects of a non-linear diode capacitance on a balanced RF differential signal, asking for a hopelessly simple solution. \$\endgroup\$ – Tony Stewart Sunnyskyguy EE75 Jan 22 at 16:06
1
\$\begingroup\$

No, that would not work. Even if you ignore completely messing up what is basically an RF link by introducing a nonlinear element, it just has not way to work. You have to monitor actual traffic to do what you want to do.

An ethernet link is basically negotiated at the physical layer when the link is initiated. If the physical layer sees that the two additional pairs cannot be used it would not just try to use them if the link becomes congested.

The additional pairs are not like a overspill gate on a dam. You either have a Gigabit ethernet link or you don't. You either use the extra pairs, or you don't. There is no in-between.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.