CAT6 Patch Panel Termination / Certification Problems 2

[disinfo]

Hello,

Can anybody give me some advice on where I may be going wrong with this as its frustrating me somewhat.

I've just installed two CAT6 Right angle patch panels with an 18M CAT6 link x 4 between them, all using the same brand of hardware (Excel)

The cabling was installed very carefully and has suffered no snags / kinks or other damage, although cabling was pulled in rather than being laid it was done very gently.

All four of the links fail due to high next between 36-12

I took plenty of time on the terminations and have maintained the twist right up until the point of termination with a minimal amount taken from the outer jacket.

Please note that I am not a pro cable installer, nor am I doing this work for money, its for a friend on a non-commercial premises.

 

[skip555]

Please note that I am not a pro cable installer, nor am I doing this work for money, its for a friend on a non-commercial premises.

then what are you using for a certifier if your not a pro ?

its a expensive piece of equipment for a hobbyist

my thoughts would be off brand supplies (excel ? )

inaccurate test equipment or test equipment not properly used

sounds like their getting what their paying for

 

[disinfo]

Thanks for your informative reply.

I'm UK based so you may not be familiar with Excel, but its one of the larger cabling suppliers in the UK. Its comes under the parent company of Mayflex, who also supply Brand-Rex and Belden.

As I said, this is a non-commericial job. Just try to help a friend out in his home.

The tester I am using is a newly purchased and calibrated Metrel MI2016, its a lower end tester but still accurate. I also have a Microtest Omniscanner (calibrated).

Does anyone have any advice?

P.S. Thanks skip555 for your welcome to the forum

 

[wires]

Do both the Metrel and the Microtest instruments show "high next between 36-12"?

If both do then it is kind of hard to believe that there is not a problem. The next step would be to make a test cable with some left over cat6 and the same connectors as used on the job. If you get a fail on the test cable then swap the connectors for another brand, my choice would be Panduit. If you still get a failure try different cable.

The failure has to be one of three things:

1. Faulty installation.
2. Inaccurate test (equipment or operation).
3. Bad hardware (cable or jacks).


Good luck!

 

[trvlr1]

I am not familiar with either of the testers you have used, what I would first test, if they have the capability, is start basic and make sure the pin out is correct, 1-1, 2-2 etc..
Also test the patch cables/injectors etc. being used for the tests to make sure they are all calibrated and correct as well as charged up, if all comes up good on this let us know and we can move on to the next possibilities

 

[disinfo]

Hi there wires, thanks for the reply.

Yes, both testers indicate high next on pairs 36,12 (around -1.3db headroom)

We were working against the clock so I only had time to terminate and test once, I will be able to go and have a go at retermination early next week.

Using both testers TDNXT facilities, both indicate the faults at the point of connection, so it has to be something to do with my terminations.

I'll have a go again but if I have no luck I will look into sourcing some panduit connectors to try.

Thanks again

 

[disinfo]

Hi trvlvr1

Thanks for the reply.

The microtest is an old scanner now, and the metrel is not so common, more of a european tester I think.

The pinout is definitely correct (568B standard)

As I say the metrel is brand new and the Omni recently calibrated, as they are both giving similar results to 0.1 of a db I would be doubtful that the tester is inaccurate.

 

[wires]

One observation. The wiring you installed will most likely work just fine. The slight bit of crosstalk on the data pairs probably won't impair data transmission to a significant degree. You could always send a lot of data over the links and check error rates. Many routers have basic statistics on their status or log page. If you have a Linux box on the network then ifconfig will report errors.

I suspect that you want to understand the problem and that is why I suggested trying to replicate in a test environment. Doing this sort of thing on site with installed wiring usually leads to much frustration.

Whenever I transition from one technology to another I spend some time in the shop making a connections and testing them until I get repeatable results.

 

[disinfo]

Thanks wires,

You are dead on with your suggestion.

I think when I go back on Monday I will take the two patch panels away from the cabs and try again with a test length of cable as you suggested.

You're also right about me just really wanting to understand the problem.

I've installed plenty of CAT5e over time and not had one faliure, used to a decent amount of headroom even on a tricky install. I was quite shocked when I saw the fail appear on the tester Yesterday.

Perhaps I have too much faith in Excel, its the Cat5e cable i've used for years. Maybe its time to try a different brand.

As long as i'm not doing something totally stupid on the install, i'm happy to go back and experiment with reterminating.

It would be nice to be able to leave my freind with a "PASS" certificate.

 

[ignatkarobkin2]

Are you using CAT5 or CAT6 cable?

I am not clear of your problem anymore.

http://www.aaa-avad.com

http://www.ivdsonline.com

 

[disinfo]

CAT6, I was simply comparing it to installing cat5 i.e. harder

 

[ignatkarobkin2]

Why is that so essential for you to have that cable certified?
Few things that might get your cable fail:
90 degree angles
tie wraps

http://www.aaa-avad.com

http://www.ivdsonline.com

 

[oldtimerbob]

You mentioned you went from panel to panel.

You may to check the terminations at the panels. You are only allowed 1/2 inch of exposed pairs at either side(tho I would reccomend less if possible)for each cable.

Sometimes,just re-terminating may correct the problem.

Good Luck,

Has been in the cabling business for about twenty years and is now the Sr PM for a cabling company located in the Los Angeles area.

http://www.globusinc.net

Also a General Class Amatuer Radio Operator.

 

[oldtimerbob]

On my last post,I forgot to mention that you may want to check the cable you are using from your tester ends to the panel ports. If you are testing for Cat6,then you want to be sure and use Cat6 patch cords.

Thanks,


Has been in the cabling business for about twenty years and is now the Sr PM for a cabling company located in the Los Angeles area.

http://www.globusinc.net

Also a General Class Amatuer Radio Operator.

 

[KTM72]

Another thing that you might want to ensure is that you are setting up the parameters in your test instrument correctly. I have used Fluke and Agilent test instruments and I am not really up to speed with the units you are using. Fluke and Agilent have options in their menus to allow you to specify the type of cable, connector etc that you are using. Also the firmware of your instrument, is it current and updated? I will assume that the Omni is since you said it was recently calibrated.
Another question, have you played with the testing algorithms in the units or on a PC in any way...that can cause some major problems too.
I am in the US so the materials you are using are not all that familiar to me, I have heard of brand rex, but know nothing about it. Krone/ADT makes some excellent patch panels and jack in my opinion. Very good connectivity and they grip the wires like a gorilla once you have punched them all down.
Another thing you might want to do is check to see how much exposed wiring you have on each end of the run, for example, did you strip the wiring 12" away from the termination and punch it down to save space on the back of the panel? The TIA/EIA standard is 1" that can potentially cause you to have some cross-talk.
Ultimately as someone already mentioned, if it works dont sweat it too much. I have had to (out of necessity) exceed maximum cable length before by a significant amount, it did not hurt the connection at all.
Most of the established standard are set at their levels due to a couple of failures in a test lab, so the entire standard is lowered across the board. In reality most telecom and data applications will work well in excess of established guidelines.

I was miserable, then someone told me "smile and be happy, things could be worse". So I smiled and was happy...and they were right, things were worse!

 

[disinfo]

Thanks for all the excellent replies.

Just to close this post;

I went back yesterday to correct the issue, after several re-terminations I managed to get all of the links certified, not with a great amount of headroom but at least they passed.

This must be to do with the quality of the cable / patch panels as my terminations were as tight as possible.

I'm now eager to try a different brand of cable, i'm looking at Beldens 4800LX System which seems good.

I've also registered on the BICSI ITS course

Thanks again for your replies.

 

[acableconnection]

Are you testing with cat 6 patch cables?
set up for link or channel?