atcomsystems.ca/forum Forums Wiring - Telecom Carriers - Trunks & Services Cabling Telephones Systems Wiring Shielded cable fails tests

Im doing an install at a hospital. For this round we ran 3 legs consisting of cat5 shielded feeding the 20 or so 7Ghz RF id tag sensors, that track patient and doctor motions by way of badges. I guess its a quality of service deal where the hospital gets more oney if they can prove theyre behaving well.

Now as weve come into the testing part of it we are getting a lot of errors in the form of return loss on random pairs. There arent reversals or anything but the return loss is recurring.

At first we suspected the hp tester needed calibration or proper settings but then when we got a few passes we were scratchin heads.

Meanwhile the clock is ticking, the guys from the manufacturer of the sensors are lookin up our ladders frowning at the sound of our failing tests on the machine but our terminations look topnotch.

I dont know..what do you think?

Runs too long. are you testing 5 or 5e.

Bad connectors or jacks?

Hard bend in the pull someplace?

Are you testing channel or permanent link? Mave you checked the contacts on your tester/remote/cables? Could be a couple of pins are bent in slightly on the connector on the remote.

I guess the first thing to ask is are you testing your runs with nothing connected to them? I would terminate them on both sides and just check your runs. If they pass, then you know it's the devices hooked up to your runs.

Don't know why you are certifying runs of cable that go to tag sensors. Just run a test for opens, shorts and crossed pairs and connect them up.

-Hal

A few testing rules: 1)Always test basic link. 2)Calibrate before testing 3)Input the manufacturers NVP value, don't use the default. 4) Ensure proper default setup, i.e cable type. 5) Fresh batteries or use AC and calibrate.
Presuming installation is proper are your failures to the local unit or the remote? If testing will not complete then reterm both ends and retest.If anomolies are not detected then you have termination or jack problems. Finally, if all else fails, reset your tester to 10baseT and test! If it passes this test you most likely have a bad section of cable.

Sorry to have disappeared after the help is offered.

To address a few of the things mentioned; the tester itself was fine as far as we could tell, pins were ok, connection was firm, the link cables were A OK. It had been calibrated and set to the factory specs as opposed to default and all aspects of the termination from pins to jacket looked fine...oh yeah and no equipment hooked up.

I guess all the hubub was due to the fact that it was in a hospital emergency room. Theres an awful lot of other sensitive equipment around so in my head the shielded cable makes sense but I dont know that much about it so perhaps it was unecessary.

For now the project is at a standstill. The last day I was there they(the equipment vendors) wanted to test the equipment because they agreed that it shouldve been fine, but something in the hub fried and the test was over.

Now theyre waiting for parts from the other half of the country and we move on to other jobs until theyre ready for some more. Not sure why they have to get a fuse from south carolina but hey what do i know....

but that return loss we were getting on the cat5e shielded cable was still sending errors regardless of whether it works/looks ok or not

Im not even sure what return loss is really so perhaps I should look it up.

Shameless cut and paste (i didnt feel like typing)

In telecommunication, return loss is the ratio, at the junction of a transmission line and a terminating impedance or other discontinuity, of the amplitude of the reflected wave to the amplitude of the incident wave. The return loss value describes the reduction in the amplitude of the reflected energy, as compared to the forward energy. For example, if a device has 15dB of return loss, the reflected energy from that device is always 15dB lower than the energy presented. For all devices that are not perfect transmission lines, or purely resistive loads (perfect black-bodies), the return loss value varies with frequency.

Return loss (as far as telecom goes) only occurs at the 2 wire point a pure 4 wire circuit can't have return loss. On private line voice circuits we always tested return loss. So I'm not sure what your test set is actually testing. Bottom line return loss is the transmit path interferes with the receive path, that's just my over simplified explanation. As Waine said it's reflective power, a more than acceptable level of the signal you are transmitting is coming back to you.

While you are waiting for them, why don't you set up a piece of the same cable, roughly the same length, with the same connectors back at your shop while nobody is bothering you. Try testing that and see what the results are. Try it also with a length of unshielded CAT5e for the heck of it.

-Hal