CAT5e cables 'woven' - engineering opinion needed

[No1Bombay]

THE SETTING: The rack room; plenum floor.
CIRCUMSTANCES: Installation by in-house IT Staff and Building Maintenance Staff of 125+ drops of CAT5e in preparation for testing and certification by a Hubbell certified installer.
At some point the once neatly laid cable bundle became quite tangled and 'woven' with approximately 25% of the total number of cables wrapped around the remaining 75% and crossing the core bundle and themselves at various angles, a given cable sometimes crossing several times in direct contact with itself.
THE PROBLEM: Seeing the 'tumbleweed' the cable plant had become at the rack room end, I feared there existed potential for the performance to compromised, especially in view of moving everything from 10Mbps/half duplex to 100Mbps/full in a new, robust MS SQL environment. Since cables are certified in a static state, would there be a possibility of inductance compromise once everything was fired up? Cable tests that I've read of late seem to suggest this is possible (or probable in view of some studies if I'm interpreting the test results correctly) unless the cables are laid parallel as best as is practical.
Comments or referral to an engineer (with verifiable credentials) for an opion (paid or unpaid) would be helpful.
Much Thanks!

 

[MasterRacker]

I've read exactly the opposite. According to what I've read, "combing" the cables so the run nice and parallel, like you see in magazines and ads, actually can increase crosstalk between cables since they are running side by side. Having them jumbles in the raceways minimizes this. This means we've each read conflicting information.

My own bundles are randomized and we're running a full 100 Mb switched arrangement with no problems. My cabling has not been spectrum tested and fully "certified" however. I do know that in day to dy operation our network has no problems.


Jeff
I haven't lost my mind - I know it's backed up on tape somewhere ....

 

[MattWray]

It will not harm anything. I am not an engineer, but I have worked with cabling and used Wirescope 350 to certify cables. My boss has worked with cabling for over 20 years. It is virtually impossible to not have any twisting or weaving of the cable as it is pulled throughout the site. The purpose of making the cables straight down the wall or tray is to make it look good and professional. It doesn't affect transmission. Matt Wray
CCNA, MCP
mwray77518@yahoo.com

 

[Guest_imported]

I agree with the two answers above. Crossing each other at or near a rightangle will give minimum pickup of the opposing cable. This holds good for all applications, include power, audio, and RF, etc. Parallel cabling looks tidier or more organised than a rats nest of wires, and of course if you need to withdraw one or two cables it's usually a darned sight easier if they were neatly bundled. Each pair of wires within the cable forms a balanced pair, and since we are only talking about minute currents within these wires, there is virtually no radiation worth bothering about.


ROGER - GØAOZ.

 

[Guest_imported]

bundle that cable in parallel or rats nest style, it's all good! worry about fluoresant(sp?) lighting, electric motors, power converters, bad terminations, low spec office jacks, and cheap patch cords. then you're good to go. we've done many-many closets bundling hundreds of lines all over each other neatly, and in a mess, and it all goes 100tx without issue.