Sign up ×
Network Engineering Stack Exchange is a question and answer site for network engineers. It's 100% free, no registration required.

I'm in an A+ course and was just working on wiring Ethernet cables. I was wondering, will an Ethernet cable function if the wires are reversed? Say instead of wiring 568B like striped orange, orange, striped green, blue, striped blue, green, striped brown, brown. Would it be possible for them to work when they're wired brown, striped brown, green, striped blue, and so on?

share|improve this question

3 Answers 3

up vote 13 down vote accepted

When you use TIA/EIA-568B on both sides this is a straight through cable. The colors of the inner jackets don't really matter, much the same as it makes no difference to the operation of the network if you use a network cable with a black or yellow outer jacket.

However, the standard is in place for a real reason, and that is that the cabling system should be implemented in a standard way so that anyone who works on it will intuitively understand what they are working on.

For instance, imagine if someone used your reversed wiring in a real environment, left, and you were now troubleshooting a cabling issue at that site. On inspection, you determine the cable is damaged and it need to be re-terminated on one side and did so with the standard TIA/EIA-568B. After doing so, the cable still will not work.

It isn't because your termination is bad, but because the cable is now a rolled cable. You may waste additional time/resources on trying to re-terminate the cable additional times. Unless you use a cable tester, you may not easily figure out that the cable is now a rolled cabled.

Ultimately, this potentially causes more problems/downtime and isn't really justifiable by any possible reason the original person may have had to not follow the standard.

Old answer (thinking the OP was talking about using TIA/EIA-568B on one side and the reverse on the other):

What you are describing is what is often referred to as a "rolled" or "rollover" cable. Pin 1 connects to pin 8 on the other side, pin 2 to pin 7, etc.

This type of cable would not work for network communications, but is used by many vendors for serial communication, such as console port access.

share|improve this answer

Simply addressing the connectivity of Ethernet cable, for short one as long as you use the same pairs at the same position it's OK.

For longer ones, you might have some crosstalk issue if you do it "however you want as long as it's the same pairs at both ends"

TIA/EIA-568A & B are only conventions to help everyone work together with less headache, there are no UTP-Police that will check your setup. Sure, they're supposed to be optimized (alternating signal/ground strands in the connector, centering Pair 1 to be able to use it for telephony if needs be, etc) but it's not like each pairs will refuse a signal if it's not at the right position in the connector. Also, if you follow the standard, it's quite easy to differentiate Straight/Roll-Over/Cross-Over cables.

It's the same thing as the Jacket color (In UTP structured cabling. Fiber jacket color actually means something). Even if the cables are Blue, Red, White or Black, they're still an Ethernet Cable (barring rolled-over or crossed-over ones), meaning you can use whichever colors you like, but it's still a good idea to be able to see their function by using different colors (as long as you stick to it!)

share|improve this answer
UTP Police == Auditors. If your infrastructure wiring is audited / certified, it will fail if it doesn't follow the correct standard. –  Ricky Beam Jun 12 '14 at 18:38
Yes, but fortunately that is part of the Certification process, not the Ethernet standards. It's not a requirement to "simply have the network working" –  Remi Letourneau Jun 12 '14 at 18:44
If seen cables < 2m not working properly. So short is really short. I've also seen people working changing plugs on about 150 cables because they thought they could ignore the standard and ignore the cable tester (smalle Fluke). –  Jens Link Jun 13 '14 at 11:15
if a <2m cable isn't working properly, it rarely because you've selected to use other pairs than the TIA/EIA-568x standard, but mostly because you've untwisted too much or crimped badly Like I said, if you're in your home, that's not a problem. I wouldn't do that in a datacenter though. You're right about test equipment, they're designed to follow the standard so if you don't follow it, it might work while the test equipment will tell you it's failing. –  Remi Letourneau Jun 13 '14 at 17:38

The real answer is simple. Ethernet is a differential line technology.

There are two ways to transmit information on a transmission line : Balanced(Analog)/Differential(Digital) or Unbalanced(Analog)/Single ended(Digital) line.

A differential transmission line is one where the signal is transmitted at two opossite polarities thru two lines. To represent a bit 1 you might send, lets say, +12V in a conductor and -12V on the other (Fast/Giga Ethernet is more complicated than that). Why this is done so ?

Because on a single-ended line, interference induced on the cable cannot be distinguised from a legitimate signal. If you put +12V on a single a conductor and receive, due to attenuation, +2V on the other side, this +2V might not be enough to prevent interference from destroying your signal, and if this does happen you cannot recover (you can, by higher abstraction means, but lets ignore this for a while) your information.

If you send the information on two conductors, by transmitting the oposite signals on each one, a interference that causes a, lets say, +5V spike on the line will interfere with BOTH conductors at the same time on the same way. If you send +12V on a conductor and -12V on the other, you might get on the other side +17V and -7V on the other. The voltages are wrong, but if you compare +17V to -7V (=24V) you get the same value as if you compare +12V to -12V (=24V difference). Think of this difference not as a normal subtraction but as a comparision of the linear distance of both points on a line.

So, a differential signal line will have higher immunity from noise.

But what does this means ?

On the ethernet connector of the network adapter, differential signals are paired on a very specific way. If you are dealing with Fast Ethernet, they are present on the pins 0-1 and 3-6. If you simply put the cables color paired side by side, you will produce not a differential line but two sepparate single-ended transmission lines (Part of the signal will go via Pair 2, part via Pair 3). Pin 3 will go thru a pair and Pin 5 will go thru another pair. They wont match.

This will cause all sort of effects. First they are not twisted togheter, they have a higher probability of receiving different kinds of interference values. They are not paired, so they will provide a totally different charactheristic impedance than what is expected.

As a historical reference, we starter using twisted pair cables to reuse twisted pair cables used for telecom. Twisted pair provided a cheaper way to install ethernet on older buildings that had already telephony equipment. Under such circunstance, you might use the two left-over pairs to transmit two telephone lines. They can - if done correctly - provide one more Fast-Ethernet connection (Yes, you can use a single cable to transmit two ethernet runs). Under a gigabit standard (not all of them, there are various standards for gigabit) all pairs are used so you have no such possibility.


The standard is set that way because ethernet is a differential transmission line technology and you must match the output/input differential pins on the adapter to the pairs on the cable. Thats why orange/light orange conductors jump over (usually) blue/light blue (or something like that) pair at the center.

share|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.