|
Joined: Apr 2005
Posts: 1,138
Member
|
Member
Joined: Apr 2005
Posts: 1,138 |
I was speaking with a telecom guy today. He was saying that when making an ethernet cable it really doesn't matter what colors go where as long as 1 and 2 are pair and 3 and 6 are a pair. I said he was wrong. But I need docuements to prove why. Does anyone have a link to why 1 and 2 are white orange and why 3 and 6 are white/green, etc...?
|
|
|
Visit Atcom to get started with your new business VoIP phone system ASAP
Turn up is quick, painless, and can often be done same day.
Let us show you how to do VoIP right, resulting in crystal clear call quality and easy-to-use features that make everyone happy!
Proudly serving Canada from coast to coast.
|
|
|
Joined: Jan 2006
Posts: 2,266 Likes: 1
Member
|
Member
Joined: Jan 2006
Posts: 2,266 Likes: 1 |
It doesn't matter as far as working as long as it is the same on both ends. May confuse the next guy who happens to look at the pinouts
|
|
|
|
Joined: Dec 2002
Posts: 9,429 Likes: 3
Member
|
Member
Joined: Dec 2002
Posts: 9,429 Likes: 3 |
Not true. It is done that way because it would be split pairs any other way. Liquid, you should PM Ed about this 
Jeff Moss Moss Communications Computer Repair-Networking-Cabling MBSWWYPBX, JGAE
|
|
|
|
Joined: Jan 2007
Posts: 153
Member
|
Member
Joined: Jan 2007
Posts: 153 |
Orange and green (1-2,3-6) using 568B are the only two pairs that data uses unless your using 1000 base-T. They still should be terminated 568.
Technician I IBEW Local #58 Detroit, MI
|
|
|
|
Joined: Feb 2006
Posts: 826
Member
|
Member
Joined: Feb 2006
Posts: 826 |
Quite some time ago someone told me that you could terminate an ethernet cable any way you liked, as long as both ends looked the same. Didn't follow the standards, but yes, it would work. Until they came out with 100Base-T, when the 3-6 split became more important.
Later I argued with the same colleage about splitting data cables and putting two jacks on each cable. His opinion was that it was a more efficient use of the cable, and afterall, it seemed to work. Sure, it would work, until they came out with 1000Base-T.
Some of the standards may seem arbitrary, but they are developed with an eye towards emerging technologies and future proofing. That Cat5 that was installed in 1995 using best practices and following standards that were overkill for 10Base-T, could be utilized at 100Base-T and then 1000Base-T with no upgrades required.
Rather than putting the burden of proof on yourself, I would turn it around. Can he prove that his sloppy terminations will hold up from 10Base-T all the way to 10Gig? What about 802.3af POE, Cisco's original pre-standard POE, or the upcomming 802.3at standard? Ask him to think a few years down the road. New POE standards are expected to provide power not only to APs and VoIP telephones, but more power hungry devices such as laptops. Would he trust his cabling practices to support these technologies? Is he willing to put equipment at risk in the near future just because he can play with the standards now and it doesn't "appear" to make a difference?
I'm sorry I can't help with supplying a link. I don't think you'll be able to provide adequate proof unless you want to pay for the standards documents.
|
|
|
|
Joined: Sep 2006
Posts: 87
Member
|
Member
Joined: Sep 2006
Posts: 87 |
Technically, he is correct. You "can" wire it any way you like just so long as it is wired "straight through." The electrons transmitting over the conductors do not know, and frankly don't care, what color the dielectric insulation is that surrounds the conductor. Straight through is straight through whether or not it is 568A, 568B, or any other pattern that you devise.
Now, having said that, it is always a wise idea of getting into the habit of doing things "right." And, there is an established "right way" to terminate patch cords. That is with an established standard such as 568A & 568B. It doesn't take any more time to do, and it is very easy to learn.
So, do it right by following the standards.
Hans Broesicke, RCDD
|
|
|
|
Joined: Nov 2002
Posts: 632
Member
|
Member
Joined: Nov 2002
Posts: 632 |
It is my understanding that they must be pairs due the number of twist per pair,Which makes sense to me because of the differences in lengths of the pairs ,as documented when you certify a cable.True the cable doesn't know what color it is ,but I would assume that the signal reach the point of destination better on two equal length conductors.
|
|
|
|
Joined: Jul 2001
Posts: 3,347 Likes: 10
Moderator-Avaya-Lucent, Antique Tele
|
Moderator-Avaya-Lucent, Antique Tele
Joined: Jul 2001
Posts: 3,347 Likes: 10 |
You could wire it with any pair replacing any other pair, BUT, you still need the pairing to be 1&2, 3&6, 4&5, and 7&8.
The most common mistake is pairing 1&2, 3&4, 5&6, 7&8. This gives you a split pair on 3&6, and will cause the connection to fail. The "link" light will come on, because DC-electrical-continuity-wise, both ends are the same. But when the data starts moving through the cable, it will fail because of the cross talk.
|
|
|
|
Joined: Nov 2002
Posts: 632
Member
|
Member
Joined: Nov 2002
Posts: 632 |
|
|
|
|
Joined: Feb 2007
Posts: 2,473
Member
|
Member
Joined: Feb 2007
Posts: 2,473 |
Steve, HJB, Tommy and Gene they are right. It doesn't matter what color you are using unless it is straght through.
|
|
|
Forums84
Topics94,515
Posts639,949
Members49,847
|
Most Online5,661 May 23rd, 2018
|
|
|
|