0
\$\begingroup\$

I am doing the integration of an aircraft device that uses Ethernet 100BASE-T as its main communication line. The manual says the device comes with its own testing cable which is approximately 2 feet long but it does not feature any twisted pair in it. Instead, the testing cable consist of 51 loose individual wires all crimped to micro D-SUB connector.

Will the ethernet communication work in this kind of setup? On how long of a distance should I expect to have a 100Mb/s ethernet link over loose wires, if at all?

\$\endgroup\$
5
  • \$\begingroup\$ If it uses 100Base-T, then it has an 8P8C modular connector. Where will this D-Sub connector go? What's at the other end of the 51-conductor cable? \$\endgroup\$ Commented Jul 26, 2021 at 13:55
  • \$\begingroup\$ Sorry if this was not clear: The device only has the 51 pin D-Sub connector. The other end of the "cable" is non-terminated wires, the RJ45 connector is to be added on our side. \$\endgroup\$ Commented Jul 26, 2021 at 13:59
  • \$\begingroup\$ If the ambient EMI levels are low, 100BASE-T can work over untwisted wire for a couple dozen meters, in my experience. If the EMI is more severe, as in an industrial environment, it won't work reliably, or at all. YMMV, of course. And for the love of God, don't ask me how I know. \$\endgroup\$ Commented Jul 26, 2021 at 14:03
  • \$\begingroup\$ @Dampmaskin and for the love of god, don't use EMI emitting untwisted Fast Ethernet cabling within the scope of aircraft electronics. The FCC and FAA (or their Canadian equivalents) will beat you with a stick if they find out, and rightfully so. \$\endgroup\$ Commented Jul 26, 2021 at 14:06
  • 2
    \$\begingroup\$ Luckily for all of us, I have never been allowed near aircraft electronics. \$\endgroup\$ Commented Jul 26, 2021 at 14:07

2 Answers 2

Reset to default
1
\$\begingroup\$

Will the ethernet communication work in this kind of setup?

Maybe yes, out of pure luck; 100 Mb/s is really really really old, and 0.7m is not that long, and modern Ethernet hardware has really gotten good. It's impossible to tell how long things might work this way – this cable has the wrong impedance, will pick up interference, interfere with other electronics... Together with how "forgiving" the actual network hardware is, that means you can't make any reasonable statement:

As a cable for testing, this is completely unsuitable, as the point of testing is producing a reliable environment. So, make a better cable, throw this away.

I mean, this is aircraft electronics of all things. The system designers here should know better. Delivering this kind of cable would be a violation of base design requirements in my book, and you should not integrate the device.

\$\endgroup\$
0
\$\begingroup\$

If the "other end" is flying leads, then you could twist them and terminate them into the RJ45.
Put each pair into a hand drill with a chuck and run it slowly until you reach the requisite turns per inch.

\$\endgroup\$
3
  • \$\begingroup\$ And remember that the different pairs have different pitches. \$\endgroup\$ Commented Jul 26, 2021 at 15:01
  • \$\begingroup\$ How does that guarantee correct impedance? \$\endgroup\$ Commented Jul 26, 2021 at 18:05
  • \$\begingroup\$ @Justme Impedance is physical. Match the spec, and you will have the same impedance. This assumes that the wire gauged is the same. Every see a wall ethernet jack on the inside? All the wires are splayed out around the perimeter. Less than perfect matching reduces the distance from the max, it doesn't preclude operation. \$\endgroup\$ Commented Jul 26, 2021 at 19:13

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.