Cable Tester - Abnormal Termination?

[sirhc]

I guess it is also possible your cables are getting a lot of noise from something confusing the tests but it is NOT a software bug but I am betting there is some thing wrong somewhere with your cables.

[sirhc]

Now if your cables were ever KINKED real hard or pulled too hard (stretched) it can deform the cable and cause cross talk across the pairs but if that is the case your cable is trashed should be replaced.

Ethernet cables are designed for a specific geometry and if this geometry is changed from kinking or stretching the cable is trashed!

We are very careful when we pull Ethernet cable up a tower as to not Stretch or kink it as to cause cross talk between pairs.

So some Goggling in this issue. Then make a 100' cable and run the test on it then stretch it and kink it and run it again and see what happens!

THERE IS NO BUG IN THIS PART OF THE CODE TRUST ME!

[lligetfa]

...either there's a short or there isn't. But if there's a short PoE isn't going to work.... soo.... which is it?

If a false short is reported on the PoE lines, then I can understand how the load that the end device puts on the pair could produce such a result. Also, with PoE, two pairs are bonded together so I am not surprised that it could report a short.

Some of my tests reported shorts on an unterminated length of wire which IMHO should never happen. In my case it turned out that using a 3 foot jumper between the switch and the keystone I had on the near end of the wire caused a reflection due to impedance mismatch. A longer jumper gave different results. Using no jumper and keystone but rather crimping the end, gave the best results.

If you are getting odd results on a cable that has no patches, nothing in between, and nothing on the end, then I would suspect kinks or other damage.

[mhoppes]

Chris and I just had a phone chat on this, and there could be several things causing this.

I did hold back some information that this is the tower site that took a direct lightening hit :icon_eeeek: over the summer. However, some of the cable runs and radios are new since then, so that's not the issue.

POSSIBLE CAUSES FOR THESE ANOMOLIES:
* We are using EZ-RJ45 ends which have the open copper ends, RF from the test may be escaping the cable end and cross-talking.
* We are using di-electric grease in each Ethernet port which may not be completely RF transparent and causing cross-talk at the end - couple this with the EZ-RJ45 end.
* There may have been a bad batch of cable when we hung some of these radios.

An airFiber on the same tower which is using EZ-RJ45 ends, but is NOT using di-electric grease is coming up OK/OK/OK/OK lending some credibility to the di-electric theory. Additionally, a NanoBridge connected to the switch with no di-electric also comes up OK/OK/Abnormal/Abnormal.... so... di-electric maybe?

I'll be testing this in the office over the next few days when I get a moment.

[lligetfa]

If using the this cable standard: T-568B:
Orange is pair 1
Green is pair 2
Blue is pair 3
Brown is pair 4

Maybe a legend like what Ubiquiti does?

Legend

Selection_006.jpg (97.12 KiB) Viewed 11646 times

[sirhc]

We will probably add some fluff to it but for now we just want to make sure everything works they way it is intended to work.

We also intend to add online/screen help as well menus.