2

Cabling contractors put wires according to plan however I think the distance is longer than 100 meters as I cannot see devices connected on another end. When the cat5e cable is longer than 100 meters, will I still see lights on the port that is connected to ? IS there any way of test 'signal' strength on the end of the cable ?

Jestem_z_Kozanowa
  • 143
  • 4
  • 2
    I can tell you, from experience, cables over 100 m _can_ work. They may not work _well_ though. Any decent cable tester should be able to verify the cable, measure length, test attenuation, cross-talk, etc. – jscott Apr 30 '15 at 12:58
  • 4
    every contractor I worked with handed over test protocols of a network cable tester when they were done. You should require yours to do the same. – faker Apr 30 '15 at 13:02

3 Answers3

2

You can use (good) Cat5e cables with length > 100 meters. They may not provide with full quality and speed, but may work.

You can test the cable and measure the length with some (pro) cable testers like Fluke Networks CableIQ. Some switches can do this, too. The feature is called TDR (Time-Domain-Reflectometry).

Lights should be on as soon as the cable makes the right contact between to active devices. Lights on do not implicitly mean there is a data connection.

Regards

Sebastian
  • 241
  • 3
  • 7
  • 2
    Is the issue only about signal strength? Originally there was a requirement that packets had to be long enough to reach from one end of the cable to the other. Does that still apply? Or was that only for half-duplex links? – kasperd Apr 30 '15 at 13:09
2

Lights usually mean you have an electrical connection, they do not necessarily mean that the signal is getting through perfectly. You can have lights but still see such high packet-loss that communication is impossible. Imagine a phone call where you can reach the person you call, but their voice is scrambled with static.

The 100 meter length is not a hard limit - It is a recommendation based on expected signal loss over long runs. I have seen 120 meter cables work ok (with some packet loss), and I have also seen 80m cables fail completely. It depends on a lot of factors.

The issue is signal degradation leading to packet loss, not latency. Signal degredation can have many causes, such as EMI (electro-magnetic interference), wire resistance, cable quality, and shielding just to name a few. If you need a longer run, use high-quality shielded cable, and high-quality ends. Make sure that your cable runs do not go near any obvious sources of EMI, such as fluorescent lights, transformers, electric motors, or high-voltage wiring. Don't run them in the same conduit as power cables. Avoid crimps in the middle of the wire.

As other posters have suggested, you need a decent cable tester to verify the quality of the connection. If you cannot get it to work, you can either shorten the cable, or insert a switch or signal booster in the middle of the run.

Superbob
  • 121
  • 2
1

You would probably need a professional meter that would measure all sorts parameters, as the strength of the signal is just one of them, and by all means not the most important. Noise and cross talk are examples of parameters I would personally look at. You can rent a professional Fluke meter for a reasonable price.

Konrad Gajewski
  • 1,518
  • 3
  • 15
  • 29