The Ethernet Wiring Conundrum
Ethernet wired networking is an industry standard. The Institute of Electrical and Electronics Engineers has painstakingly defined specifications and standards for all aspects of Ethernet networking.
Need an Ethernet cable or adding equipment to your home network? Just buy any Ethernet cables and connectors and your should be done, right?
Unfortunately, it’s not that easy. As typical with standards, Ethernet has succumbed to that joke in Geek circles – “Standards are great. That’s why we have more than one”.
IEEE 802.3 is the official spec for Ethernet, but there are many related standards and specifications.
Let’s Get Physical (Connectors)
An Ethernet cable is actually a set of eight copper wires wrapped together in a single outer insulated cover.
Each of the eight inner wires is insulated themselves. The insulated cover on each wire is a different color to make it easy to tell the wires apart.
At the ends of an Ethernet cable is a modular connector that makes it easy to install or remove a cable without a mechanical tool.
Commonly referred as an RJ45 jack, it is made of clear or translucent plastic and does require a tool to install; a process know as wire termination.
Plugging in an Ethernet cable just takes a gentle push and it locks into the mating socket with solid audible and haptic feedback.
Removing the cable is just as easy, a light press of a small plastic tab releases the connector for removal.
When it comes to the actual wiring, there are two different ways to arrange the wiring inside an Ethernet cable: These two methods are called T568A and T568B.
What’s The Difference Between T568A and T568B?
If you Google that, you’ll get over 1 million results. It is a common question and there is a lot of confusing information on this topic.
And there’s the rub. Even professionals that install Ethernet networks for a living, often get confused between T568A and T568B.
It’s become sort of a slang greeting when teams meet up at a client or customer job site.
Instead of exchanging the usual “Good Morning” pleasantries, the first words between myself and the wiring crew is the question “568A or B?”
I’m going to give you my summary and interpretation. This is not exhaustive, but should help cut through all the confusion and guide you in understanding which to choose or sorting out what may already be installed in your home or office.
“Chaos is merely order waiting to be deciphered.”
José Saramago (Literature Nobel Prize Winner)
Terminating an Ethernet wire requires lining up the eight 8 individual internal wires in a row and inserting them into the connector in a pre-determined order.
There’s the rub – With 8 wires, there are potentially 40,320 different ways those same wires can be arranged, so how to choose?
The Order Doesn’t Matter?
If you think about it for a while, this is actually a reasonable solution. If you wire both ends the same way, the order may not matter.
As long as the signals in wires 1 to 8 are connected to the same signals in wires 1 to 8 of the next device or cable, the internal order of the wires in the connector might not matter.
This is actually the case with the Ethernet cables you buy in retail store or online from an eCommerce website.
These “patch cables” are used to connect devices over short distances and are not intended for use inside the walls of your home or office.
As long as both ends of the cable are wired the same way (and they always are), you can simply buy them without ever asking “568A or B” or even caring.
Just plug in that Lutron Caseta bridge to your home router, or plug your desktop PC Ethernet port into your Ethernet switch nearby and you are set to go.
Technical aside: In the early days of Ethernet one did have to worry about straight through or cross-over Ethernet cables. The former were used to connect devices to a switch and the latter used for direct device to device connections.
Fortunately, all Ethernet equipment now has hardware based self-configuring connections that magically configure either way as needed.
(I still remember some discussions with engineers and product designers on whether a particular port on a product we were building should be “MDI” or “MDI-X”, but fortunately the introduction of “Auto MDI-X” chipsets put that argument to rest.)
Two Flies In The Ointment
Unfortunately, it is not that simple as ignoring the wire ordering inside Ethernet cables in modern Ethernet wiring for house-wide or system-wide installations.
First, with electronic signals, transmitting them down tiny thin wires of copper is actually an astounding feat of electromagnetic physics.
Real world issues of interference (other signals or background radiation) can corrupt the signals along with other esoteric problems such as induced noise, near-end crosstalk, far-end crosstalk, and other effects that are just as inscrutable to most of us.
An Ethernet cable is actually a type of unshielded twisted pair (UTP) multiconductor cable.
Kind of like a tug of war – the cable is unshielded (bad for protection from interference) so pairs of wires are loosely twisted together (good) to provide some protection.
If you really want to get deep into it, there is a whole discussion of how many twists per foot, and which pairs of wires in the eight-wire cable should be twisted together which is completely complementary to the topic today of what order to use for the individual wires themselves.
Second, that bugaboo of modern products and systems we all face sooner or later – backward compatibility, or the lack thereof with older wiring – specifically telephone wires.
Spoiler Alert – Those Two Conflicting Requirements Yields Two Competing Solutions
T568A and T568B are two different standardized ways of ordering the individual wires inside an Ethernet cable.
T568A is designed for backward compatibility with older telephone wires; T568B is designed for better signal isolation and noise protection for newer networking systems and products.
So the choice was supposed to be simple – if you need backward compatibility, use T568A, if you want maximum future expansion to higher speeds and newer equipment, use T568B.
Landline? What’s That?
Rapidly becoming extinct, the traditional telephones used in residences use two copper wires.
In 1973, Bell Telephone introduced the RJ11 modular jack to make it easier to install and change telephone wiring. Instead of looping individual wire around a screw terminal, the now ubiquitous module jack made it easy to install telephones in homes and business.
The RJ11 jack allows for telephone wire with up to 4 individual wires. Since a telephone line needs only two wires, this means a single RJ11 wire and jack can handle up to two separate physical phone lines.
The order of the wires was clever – If you only have one telephone line, then only the two inner pins in the jack are used and the telephone cord only has two copper wires inside.
If you have two lines, then the outer two pins are also used (for a total of 4) to send two phone calls down a single telephone cord to be split out at the other end.
Brilliant or Foolish?
As more lines were needed for larger homes and business, the RJ11 modular jack was redesigned to become the bigger RJ45 that we know well today.
Still for telephone use, the clever mechanical engineers designed the RJ45 as a mechanical superset of the RJ11.
You can take an RJ11 jack and plug it into a larger RJ45 physical connector. The inner 2 wires or 4 wires connect to the appropriate pins working perfectly for a single line or two-line telephone systems.
This composite photo tries to show this – From left to right: Two identical RJ45 Ethernet jacks side-by-side; A classic 2-wire silver-satin modular telephone cord with an RJ11 jack; A modern Ethernet cable with 8 wires and RJ45 jack.
If you look very carefully, you can see how the RJ11 telephone cord fits inside the RJ45 without touching the outermost extra pins.
What’s In Your Wall?
In the last thirty years or so, most newly constructed homes included “structured wiring” pre-installed.
Homes have been pre-wired with RJ45 jacks and wiring that can be used for either traditional landline telephones or data networks.
By installing RJ45 jacks instead of RJ11 and using data grade cables, the homeowner could configure the wiring for telephone, data, or a mixture of both.
To preserve this dual use of data or telephone, the majority of mass market home builders choose to use the T568A standard.
Only custom built homes where smart home automation is planned from the beginning typically use T568B to maximize the options for faster speeds and better signal isolation.
For most home networking installations, it really doesn’t matter. Either T568A or T568B will work fine – and will work with high speed Ethernet networks as fast as 1 Gbps or even 10Gbps.
The most important thing is to keep everything the same for all your infrastructure – patch panels, in-wall wires, switches, and hubs. When expanding or repairing the Ethernet wiring system in your home or business, use the same wiring scheme, T568A or T568B that is already in place.
To keep things straight and avoid a trip online to Google everything you remember there are two different Ethernet wiring schemes but not sure which one is which, I use a simple mnemonic:
“B” is for business, so that means “A” must be residential.
Choose T568B (business standard) for maximum speeds and noise isolation, T568A (residential standard) for backward compatibility with home telephone wiring, whether you have a landline or not.