Gigabit Ethernet’s most popular standards are 1000Base-T twisted copper cable and 1000Base-X (Fiber Optic). Let's take a closer look at them and also analyze 10GBase - 10Gigabit Ethernet which is the standard for datacenters (server connection, network switches - backbone etc) supporting transfer speeds of up to 1250MB/sec.

1000Base-T

The 1000Base-T is the IEEE 802.3ab standard for Gigabit Ethernet over copper wiring. Similar to 100BaseT, 1000Base-T can run up to a maximum of 100 meters or 330 feet and is supported on CAT5e, CAT6 and CAT7 cabling. Users can also visit our Unshielded Twisted Pair (UTP) - CAT 1 to CAT5, 5e, CAT6 & CAT7 article which contains diagrams and valuable information on each category cable, their maximum length, supported speeds and more.

With 1000Base-T, all 4 pairs of the copper cable are used, however only two pairs are used for the negotiation of the link. If the cable only has 2 pairs available, the negotiation between the two devices might succeed, however the link will not come up.

Crossover cables are not required for 1000Base-T links as the two endpoints will automatically negotiate and bring up the link even when connected via a straight-thru UTP cable.

1000Base-X

The 1000Base-X is the industry standard for Gigabit Ethernet over fiber optic. There are four Gigabit Ethernet over fiber optic standards of which the most popular are the 1000Base-SX and 1000Base-LX standards.

The 1000Base-SX standard is for operation over multi-mode fiber using 770 to 860 nanometer light wavelength. Multi-mode Gigabit Ethernet fiber can run up to 200 meters when using 62.5/125 fiber or 550 meters when using 50/125 fiber. Multi-mode Gigabit Ethernet is usually used for backbone connections between switches located in datacenter, short-distant buildings or different levels within a building.

The 1000Base-LX standard is for operation over single-mode fiber using 1,270 to 1,355 nanometer light wavelength. Single-mode Gigabit Ethernet fiber can run up to 5Km depending on the fiber type and is commonly used for backbone connections between buildings and long-distant endpoint connections.

10GBase

The 10Gigabit Ethernet standard has become popular in the past couple of years, however the IEEE standard (802.3ae) has been developed since 2002. 10Gigabit Ethernet provides speeds 10 times faster than Gigabit Ethernet and is supported over copper and fiber optic. 10Gigabit Ethernet is rarely used for endpoint connectivity (workstations, laptops) and found mainly in datacenters connecting servers, switches and storage devices. In some cases, it is also used for backbone links between buildings and switches.

10GBase-T

10GBase-T is the Ethernet standard covering copper wiring, standardized in 2006 via the IEEE802.3an standard and is supported only by CAT6 and CAT7 twisted pair cables. When using CAT6 cables, 10Gbps is supported only for up to 55 meters, whereas CAT7 cabling can provide 10Gbps support for up to 100 meters.

10 Gigabit Ethernet over fiber is defined by standards 10GBASE-SR, 10GBASE-LR, 10GBASE-ER, 10GBASE-LX4/LRM, plus a few more. Following is a brief overview of these standards:

10GBASE-SR: Know as Short Reach (SR), used for short distant connections via multi-mode fiber optic using the 850 nanometer wave length. Maximum distance is 400 meters, depending on the type of fiber optic cable.

10GBASE-LR: Known as Long Reach (LR), used for long distant connections up to 10Km via single-mode fiber optic cable using the 1310 nanometer wave length.

10GBASE-ER: Known as Extended Reach (ER), used for extra-long distant connections up to an impressive 40Km via single-mode fiber optic cable using the 1550 nanometer wave length.

10GBASE-LX4/LRM: The LX4 standard was replaced by the LRM (Long Reach Multi-mode). LX4 operates at distances up to 300 meters for multi-mode fiber or 10Kms for single-mode fiber and uses the 1310 nanometer wave length. The LRM standard uses multi-mode fiber only and can cover distances of up to 220 meters, again using the 1310 nanometer wavelength.

This completes our article covering 1000Base T/LX/SX Gigabit & 10GBase (10Gigabit) Ethernet standards. We mentioned the IEEE standards that define these technologies and also took a look at the different UTP Category cables supporting 1Gigabit and 10Gigabit Ethernet speeds.

Morse code is type of binary system which uses dots and dashes in different sequences to represent letters and numbers. Modern data networks use 1s and 0s to achieve the same result. The big difference is that while the telegraph operators of the mid 19th Century could perhaps transmit 4 or 5 dots and dashes per second, computers now communicate at speeds of up to 1 Giga bit, or to put it another way, 1,000,000,000 separate 1s and 0s every second.

Figure 1. Morse Code device used to transmit '0' and '1'

Although the telegraph and the teletypewriter were the forerunners of data communications, it has only been in the last 35 years that things have really started to speed up. This was borne out of the necessity for computers to communicate at ever ncreasing speeds and has driven the development of faster and faster networking equipment, higher and higher specification cables and connecting hardware.

Development Of New Network Technology

Ethernet was developed in the mid 1970's by the Xerox Corporation at its Palo Alto Research Centre (PARC) in California and in 1979 DEC and Intel joined forces with Xerox to standardize the Ethernet system for everyone to use. The first specification by the three companies, called the 'Ethernet Blue Book', was released in 1980, it was also known as the 'DIX standard' after their initials.

It was a 10 Mega bits per second system (10Mbps, = 10 million 1s and 0s per second) and used a large coaxial backbone cable running throughout the building, with smaller coax cables tapped off at 2.5m intervals to connect to the workstations. The large coax, which was usually yellow, became known as 'Thick Ethernet' or 10Base5 - the '10' refers to the speed (10Mbps), the 'Base' because it is a base band system (base band uses all of its bandwidth for each transmission, as opposed to broad band which splits the bandwidth into separate channels to use concurrently) and the '5' is short for the system's maximum cable length, in this case 500m.

The Institute of Electrical and Electronic Engineers (IEEE) released the official Ethernet standard in 1983 called the IEEE 802.3 after the name of the working group responsible for its development and, in 1985, version 2 (IEEE 802.3a) was released. This second version is commonly known as 'Thin Ethernet' or 10Base2; in this case the maximum length is 185m even though the '2' suggest that it should be 200m.

Other popular Ethernet standards developed and widely used include:

• IEEE 802.3
• Ethernet II
• 802.3 SNAP
• Novell Proprietary 802.3 Raw

Since 1983, various standard have been introduced because of the increased bandwidth requirements, so far we are up to the 40Gigabit rate!

Today, all computers are equipped with a network card and using straight-thru or cross-over network cables, we are able to quickly transfer data at speeds much greater than a serial or parallel cable. However , there are still times where we require a transfer via the serial or parallel port, and that's what this page is about.

Transferring data between computers via direct cable connections can be performed using the following methods:

• Serial Cross-over cable
• Parallel cable – also known as ‘LapLink’ cables
• USB Transfer or Data Link cable
• UTP Cross-over or Straight-thru cable

Different speed rates are achieved depending on the transfer method selected, however there are some requirements for each.

Direct transfer via Serial cross-over cables require the existence of one free serial port on both source and destination computer. Furthermore, the maximum attainable speed for data transfer via a serial port is only 14 kb/sec (0.014Mbps), making it unpractical and extremely slow for large file transfers. Apart from the very low speed rates, most laptops, workstations and servers do not offer a serial port anymore as the port has become obsolete since the introduction of USB ports.

Transferring data via a parallel port is more practical as opposed to a serial port thanks to the greater speed rates of up to 1.1 Mb/sec (8.8Mbps), however just like serial ports, parallel ports are almost non-existent as they too have been replaced by USB interfaces.

Data transfers via USB ports are a more popular solution, however a special USB Transfer (or USB Data Link) cable is required. Users cannot use a standard USB cable used to connect printers or other USB devices. The special USB Transfer cable contains electronic circuits allowing two USB ports to connect and transfer files.

Speed transfers with USB vary depending on the USB port version on the machines. USB v1.1 specification provides a maximum of 12Mbps while the USB v2.0 specification further increases this figure to 480Mbps, almost 5 times the speed of a 100Mbps network. USB 3.0 and USB 3.1 support speeds close to 10 times that of USB v2.0 however data transfer cables beyond the v2.0 specification do not exist or are extremely difficult to find.

UTP Cross-over & Straight-thru cables is the final method covered and happens to also be the most popular. Cross-over cables are required for older network cards that do not support Auto-MDIX while newer and Gigabit network cards support Auto-MDIX and require simple Straight-thru cables. Transfer speeds supported are 10Mbps, 100Mbps and 1Gbps (1000Mbps), surpassing USB v2.0 transfer speeds by more than two times.

We will also be learning how to create the cables required to meet our goals and comparing the speed of the two (Serial and Parallel)

In the next pages, we’ll analyze Serial - Parallel ports and connection cables, and then move to USB ports & Data Transfer Cables.

The Serial Direct Connection is the one which utilizes the COM ports of your computers. Every computer has at least two COM ports, COM1 and COM2. The "COM" stands for "Communications". Its pinouts are a lot simpler when compared to the parallel port, but the speed is also a lot slower.

To give you an idea of how fast (or slow) a serial port is, at its best you will get around 12 to 14 KB per second. That's pretty slow when you're used to a network connection, but let me show you how serial data is transferred so you can also understand why it's a lot slower:

Figure 1. Transfer of data via serial port

The above picture gives you an idea on how serial data is transferred. Each colored block that is numbered is sent from PC 1 to PC 2. PC 2 will receive the data in the same order it was sent, in other words it will receive data block 1 first and then 2, all the way to block 7. This is a pretty good representation of data flow in a serial cable. Serial ports transmit data sequentially over one pair of wires (the rest of the wires are used to control the transfer).

Another way you can think of it is like a one lane road where the road is wide enough to only fit one car at a time (one data block at a time in our example above) so you would imagine that the road cannot process several cars at one time.

The Serial Port

Most new computers have two COM ports with 9 pins each; these are DB-9 male connectors. Older computers would have one DB-9 male connector and one DB-25 male connector. The 25 pin male connector is pretty much the same as the 9 pin, it's just bigger.

Let's have a look at a serial port to see what we are talking about:

Figure 2. Physical Serial interface - DB-9 (usually COM1) and DB-25 (usually COM2)

Different pinouts are used for the DB-9 and DB-25 connectors and we will have a look at them in a moment. Let's just have another quick look at the COM ports of a new computer:

Figure 3. Serial ports (COM1 & COM2)

Notice the COM ports, they are both DB-9 connectors, there is no more DB-25! The connector above the two blue COM ports is an LPT or Parallel port.

The serial port of a computer is able to run at different speeds, thus allowing us to connect different devices which communicate at different speeds with the computer. The following table shows the speeds at which most computers' serial ports are able to run and how many KB/sec they translate to:

Figure 4. Serial port speeds (bits/second) and data transfer speeds

Now we will have a look at the pinouts of both DB-9 and DB-25 connectors:

Figure 5. Serial COM port pinouts for both DB-9 & DB-25 serial interfaces

Serial Data Transfer Cable

All that's left now is the pinouts required to allow us to use the serial cable for direct connection. There is a special term for this type of a cable, its call a "null modem" cable, which basically means you need to have TX and RX crossed over. Because you can have different configurations, e.g. DB-9 to DB-9, DB-9 to DB-25, and DB-25 to DB-25, I have created different tables to show you the pinouts for each one:

1) DB-9 to DB-9. You use this configuration when you need a cable with a DB-9 connector on each end:

Figure 6. DB-9 to DB-9 Null Modem (serial) Cable

2) DB-9 to DB-25. You use this configuration when you need a cable with one DB-9 and one DB-25 connector on either end:

Figure 7. DB-9 to DB-25 Null Modem (serial) Cable

3) DB-25 to DB-25. You use this configuration when you need a cable with a DB-25 connector on each end:

Figure 8. DB-25 to DB-25 Null Modem (serial) Cable

Well, that pretty much covers everything about serial direct connection via a null modem cable.

If you're using third party software to connect your computers, you probably won't stumble into big problems, but if you're using Windows software be sure you have unique names for each of your computers because Windows will treat the direct connection as a "network" connection. This means you will be able to see the other computer via Network Neighborhood.

 Next - LPT Ports - Parallel Direct Cable Connection and pinouts or Back to Network Cabling Section

Users interested in transferring files using parallel direct cables can visit the following Microsoft support page which explains How to Install and Configure the Direct Cable Connection Feature (https://support.microsoft.com/en-us/kb/298446).

As we’ll see further below, there are three different type of LPT ports, SPP, EPP and ECP, each supporting different speeds and features, but all use the same direct cable connection.

A standard LPT port will provide speeds of 40Kb/s to 60Kb/s while the faster ECP ports will deliver up to 1.1 Mb/sec or 8.8 Mbps.

To better understand why parallel links are much faster than serial links, we’ll need to analyze the way data is transferred. This is clearly shown and explained in the diagram below:

Figure 1. Transfer of Data via Parallel (LPT) Port

This diagram shows data transfer via parallel ports and we can see multiple data blocks being simultaneously transferred from one host to another, increasing significantly the overall throughput. Serial ports are capable of transferring one data block per time, therefore unable to match speeds of parallel ports.

What does the parallel port (LPT) look like?

The picture below shows a parallel port, also known as LPT port, on a computer:

Figure 2. Physical LPT port on a computer motherboard

With older computers, you will always find the LPT port right above the two COM ports and it's usually color coded purple. When looking at an LPT port, it is impossible to identify whether the port is a SPP, EPP or ECP type because all three types visually look the same. and all LPT ports use DB-25 female connectors.

Examining the Different Type of LPT Ports

Before we begin analyzing the pin outs of the LPT port, let's have a look at the different type of LPT ports available. The diagram below shows all three LPT ports along with their supported speed and description:

Figure 3. Different LPT Ports: SPP, EPP & ECP and their specifications

4-bit LTP Ports

The port can do 8 bit byte output and 4 bit nibble input. These ports are often called "unidirectional" and are most commonly found on desktop bus cards (also called IO expansion cards, serial/parallel cards, or even 2S+P cards) and older laptops. This is still the most common type of port, especially on desktop systems. 4 bit ports are capable of effective transfer rates of about 40-60 Kbps in typical devices but can be pushed upwards of 140 Kbps with certain design tricks.

8-bit LTP Ports

These ports can do both 8 bit input and output and are sometimes called "bidirectional ports" but that term is often misused by vendors to refer to 4 bit ports as well. Most newer laptops have 8 bit capability although it may need to be enabled with the laptop's vendor-specific CMOS setup function. This is discussed below. A relatively smaller percentage of LPT bus cards have 8 bit capability that sometimes must be enabled with a hardware jumper on the board itself. True 8 bit ports are preferable to 4 bit ports because they are considerably faster when used with external devices that take advantage of the 8 bit capability. Eight bit ports are capable of speeds ranging from 80-300 Kbps, again depending on the speed of the attached device, the quality of the driver software and the port's electrical characteristics.

EPP LTP Ports

Can do both 8 bit input and output at ISA bus speeds. These ports are as fast as 8 bit bus cards and can achieve transfer rates upwards of 600 Kbps. These ports are usually used by non-printer peripheral devices such as external CDROMs, tape drives, hard drives, network adaptors and more.

ECP LTP Ports

Can do both 8 bit input and output at bus speeds. The specification for this port type was jointly developed by Microsoft and Hewlett-Packard. ECP ports are distinguished by having DMA capability, on-board FIFOs at least 16 bytes deep, some hardware data compression capability and are generally featured more than other ports. These ports are as fast as 8 bit bus cards and can achieve transfer rates upwards of 1 Mbps and faster on PCs whose buses will support it. The design is capable of faster transfer rates in the future.

Let's now have a quick look at the pin outs of an LPT port:

Figure 4. LPT Port Pinouts & their functions

The LPT Direct Connect Cable

As explained, there are different LPT ports, but all use the same parallel cable for direct transfer between two hosts.. Depending on your computer bios’s LPT settings you will be able to achieve different speed transfers as outlined previously.

The diagram below clearly shows the pin out configuration of the LPT Direct Connect Cable :

Figure 5.  DB-25 to DB-25 LPT Direct Connect Cable Pinouts

We should note that one wire should be attached to the metal body of the male pins on both sides; this is also shown as the "metal body" on the diagram.

 Next - Serial Direct Cable Connection, DB9, DB25, COM Ports and Pinouts or Back to Network Cabling Section

USB DCC has been around for well over a decade but has failed to become a popular method of transferring data because of the rapid growth of networks and their significant speeds. In most cases, users with Gigabit Ethernet ports on their workstations simply require a standard UTP straight-thru cable to begin transferring data at double the rate of a USB DCC.

Let’s now take a closer look at USB interfaces and how they work.

About USB

USB stands for Universal Serial Bus and it’s the standard interface for all computer peripherals (printers, faxes, CDROMs, mice, joysticks etc) and mobile devices today.

The Universal Serial Bus gives you a single, standardized, easy-to-use way to connect multiple devices to a computer. The USB port is also capable of providing power to the connected devices, but it should be noted that each USB port has a limit of delivering a maximum amount of power, depending on the USB version. !

The initial specification of USB v1.1 was designed to deliver a maximum of 150mA (0.75 watts) at 5Volts. USB 2.0 specification increased the maximum power to an impressive 500mA (2,5 watts). USB 3.0 specification next came along and further increased the maximum power to 1,5Amps (7,5 watts) at 5Volts, allowing more power demanding peripherals to connect and be powered by USB.

Finally, the latest USB specification v3.1, jumped from 7,5 watts to a whopping 100 watts, but at the same time providing support for 5Volts, 12Volts and 20Volts.

Figure 1. USB Ports on a PC motherboard

As mentioned previously, there are 4 different versions of USB specifications: v1.1, v2, v3.0 and v3.1. Most USB ports today on computers and laptops support USB v3.0 specification while peripheral devices with USB v3.0 interfaces are already available in the market.

USB v3.1 and v3.0 are fully backwards compatible with USB v2.0 and v1.1 peripherals, ensuring full compatibility with any USB device.

The table below compares the different USB versions currently available, along with their most important technical specifications:

Figure 2. USB Specifications, different versions, Names, Data rates and power capabilities

Keep in mind that when you're using a USB DCC cable, you won't get such great speeds, but somewhere around the 500 Kbps. This also depends on the type of CPU, O/S, the quality of the cable and electronic components and protocols running on your system.

When transferring data between two computers, the effective throughput (speed) achieved will depend on a variety of factors such as USB port version on both ends, USB Direct Cable version, CPU and HDD speed, how busy both systems are more. A USBv2.0 cable connected to two USBv2 ports is expected to achieve a 480Mbps transfer rate, however if the HDD on either side is unable to keep a sustained transfer speed, its likely the throughput will dramatically decrease.

Another important detail is the operating system used. Today, all Windows and Linux operating systems fully support USB ports, however older operating systems are not completely compatible. The table below shows which operating systems provide full support for USB ports, regardless of their specification version:

Figure 3. USB support by Microsoft Windows operating system

The Standard USB Cable

The USB standard uses A and B connectors to avoid confusion. "A" connectors head "upstream" toward the computer, while "B" connectors head "downstream" and connect to individual devices. This might seem confusing to some, but it was designed to avoid confusion between consumers because it would be more complicated for most people to try and figure out which end goes where.

This is what the USB cable and connectors look like:

Figure 4. USB 'A' and 'B' type connectors

As mentioned earlier, the USB port can power certain devices and also transfer data at the same time. For this to happen, the USB port must have at least four cables or two pairs,of which one pair is used to power the USB device (e.g hdd) and the second pair is used for data transfer between the device and computer.

The diagram below shows a standard USB cable with its internal 4 wires and their function. The shielding located at the far left is used to protect the cable from electromagnetic interference:

Figure 5. USB Cable - Wires inside the USB cable

The USB Direct Connection Cable (DCC)

As previously mentioned, transferring data between two computers using the USB ports, requires the purchase of a USB Direct Connection Cable. (DCC). This cable is also known as a USB Transfer or Data Link cable. The DCC cable is not a simple cross-over cable, but contains electronic circuits that make it possible to use it to transfer data between computers.

Figure 6. USB Transfer or Data Link cable

Searching for USB DCC cables will reveal a number of manufactures however almost all offer only USB 2.0 DCC cables, which means that the maximum transfer speed expected would be 480Mbps.

In addition, DCC cables are plug-and-play, not requiring additional drivers and are powered directly from the USB port. In most cases, when the USB cable is plugged into the computer, it will see it as an external drive, containing a special executable application that needs to be run on each computer to begin transferring files between them. This completes the discussion on USB Direct Cable Connection & port specifications. More information on Data transfer methods, Serial & Parallel ports, Ethernet (Fast/Gigabit/10Gigabit) and Fiber optic can be found in our Network Cabling Section.

I've successfully used Laplink cable to link two PCs for FILE TRANSFER only (not playing Games), with WIN95 and Direct Cable Connection program using the NetBeui protocol on each computer. You can quickly check to see if the protocol is installed by doubleclicking on the "Network Section" in Control Panel of your Windows operating system.

In addition to the above, you must have installed "Client for Microsoft Networks", "File and Printer Sharing for Microsoft Networks" and optionally the TCP/IP protocol, which will require some configuration. Providing a simply IP Address and subnetmask will be enough for our purposes, the rest of the fields can be ignored. If you would like to allow users to access your files and printer, then ensure both the options in "File and Print Sharing" are selected.

Once you have completed the above steps, you should have the following listed in the "Network Selection" window::

• Client for Microsoft Networks
• TCP/IP
• Netbeui
• File and Printer Sharing for Microsoft Networks

Once your changes are complete, Windows might prompt you to reboot the system, so make sure all work is saved before answering "yes"!

You should also share the Disks on both computers by right-clicking on the selected disks installed in your system and select the "Sharing" option that will appear in the menu. You can access them via your "My Computer" icon on your desktop.

After you complete these actions, you will see a blue hand "holding" your shared drives, indicating that the drive is shared with the rest of the network!

Back to Network Cabling Section

In the 1950s, more research and development into the transmission of visible images through optical fibers led to some success in the medical world where it was being used in remote illumination and viewing instruments. In 1966 Charles Kao and George Hockham proposed the transmission of information over glass fiber and realized that to make it a practical proposition, much lower losses in the cables were essential.

This was the driving force behind the developments to improve the optical losses in fiber manufacturing and today optical losses are significantly lower than the original target set by Charles Kao and George Hockham.

The Advantage of Using Fiber Optics

Because of the low loss, high bandwidth properties of fiber cables they can be used over greater distances than copper cables. In data networks this can be as much as 2 km without the use of repeaters. Their light weight and small size also make them ideal for applications where running copper cables would be impractical and, by using multiplexers, one fiber could replace hundreds of copper cables. This is pretty impressive for a tiny glass filament, but the real benefit in the data industry is its immunity to Electro Magnetic Interference (EMI), and the fact that glass is not an electrical conductor

Figure 1. A Fiber Optic Cable

Because fiber is non-conductive it can be used where electrical isolation is needed, for instance, between buildings where copper cables would require cross bonding to eliminate differences in earth potentials. Fibers also pose no threat in dangerous environments such as chemical plants where a spark could trigger an explosion. Last but not least is the security aspect; it is very, very difficult to tap into a fiber cable to read data signals.

Fiber Construction

There are many different types of fiber cable, but for the purposes of this explanation we will deal with one of the most common types -- 62.5/125 micron loose tube. The numbers represent the diameters of the fiber core and cladding, these are measured in microns which are millionths of a meter.

Figure 2. Construction of Fiber Optic Cable - Components

Loose tube fiber cable can be indoor or outdoor, or both. Outdoor cables usually have the tube filled with gel to act as a moisture barrier to the ingress of water. The number of cores in one cable can be anywhere from 4 to 144.

Over the years a variety of core sizes have been produced but these days there are three main sizes that are used in data communications, these are 50/125, 62.5/125 and 8.3/125. The 50/125 and 62.5/125 micron multi-mode cables are the most widely used in data networks, although recently the 62.5 has become the more popular choice. This is rather unfortunate because the 50/125 has been found to be the better option for Gigabit Ethernet applications.

Figure 3. ST Duplex Patch Lead

The 8.3/125 micron is a single-mode fiber cable which until now hasn't been widely used in data networking due to the high cost of single mode hardware. Things are beginning to change because the length limits for Gigabit Ethernet over 62.5/125 fiber has been reduced to around 220m and now using 8.3/125 may be the only choice for some campus size networks.

What's the Difference Between Single-Mode and Multi-Mode?

With copper cables larger size means less resistance and therefore more current, but with fiber the opposite is true. To explain this we first need to understand how the light propagates within the fiber core.

Light Propagation

Light travels along a fiber cable by a process called 'Total Internal Reflection' (TIR); this is made possible by using two types of glass which have different refractive indexes. The inner core has a high refractive index and the outer cladding has a low index. This is the same principle as the reflection you see when you look into a pond. The water in the pond has a higher refractive index than the air and if you look at it from a shallow angle you will see a reflection of the surrounding area, however, if you look straight down at the water you can see the bottom of the pond

Figure 4. Light transmitted through a Fiber Optic cable

At some specific angle between these two views points the light stops reflecting off the surface of the water and passes through the air/water interface allowing you to see the bottom of the pond. In multi-mode fibers, as the name suggests, there are multiple modes of propagation for the rays of light. These range from low order modes, which take the most direct route straight down the middle, to high order modes, which take the longest route as they bounce from one side to the other all the way down the fiber.

Figure 5. Light bouncing while it travels inside a Fiber Optic cable

This has the effect of scattering the signal because the rays from one pulse of light arrive at the far end at different times; this is known as Intermodal Dispersion (sometimes referred to as Differential Mode Delay, DMD). To ease the problem, graded index fibers were developed. Unlike the examples above which have a definite barrier between core and cladding, these have a high refractive index at the centre which gradually reduces to a low refractive index at the circumference. This slows down the lower order modes allowing the rays to arrive at the far end closer together, thereby reducing intermodal dispersion and improving the shape of the signal.

So What About the Single-Mode Fiber?

Well, what's the best way to get rid of Intermodal Dispersion? Easy, only allow one mode of propagation. So a smaller core size means higher bandwidth and greater distances. Simple as that!

 Back to Network Cabling Section

The Meaning of 100Base-TX/T4/FX

We are going to break the "100Base-T" into three parts so we can make it easier to understand:

100

The number 100 represents the frequency in MHz (Mega HertZ) for which this cable is made. In this case it is 100 MHz. The greater the MHz, the greater speeds the cable can handle. If you try to use this type of cable for greater frequencies (and, therefore, speeds) it will either not work or become extremely unreliable. The 100 MHz speed translates to 100Mbit per second, which in theory means 12 Mbps. In practice though, you wouldn't get more than 4 Mbps.

Base

The word "Base" refers to Baseband. Baseband is the type of communication used by Ethernet and it means that when a computer is transmitting, it uses all the available bandwith, whereas Broadband (cable modems) shares the bandwidth available. This is the reason cable modem users notice a slowdown in speed when they are connected on a busy node, or when their neighbor is downloading all the time at maximum speed! Of course with Ethernet you will notice a slowdown in speed but it will be smaller in comparison to broadband.

TX/T4/FX

The "T" refers to "Twisted Pair" physical medium that carries the signal. This shows the structure of the cable and tells us it contains pairs which are twisted. For example, UTP has twisted pairs and this is the cable used in such cases. The 100Base-T is used sometimes to refer to the 100Base-TX cable specification. For more information, see the "UTP -Unshielded Twisted Pair" page where you can find information on pinouts for the cables. All 100MB rated cables, except the 100Base-FX, use CAT5 cable.

100Base-TX

The TX (sometimes referred as "T" only) means it's a CAT5 UTP straight through cable using two of the four available pairs and supports speeds up to 100 Mb. Maximum length is 100 meters and minimum length between nodes is 2.5 meters.

100Base-T4

The T4 means it's a CAT5 UTP straight through cable using all four available pairs and supports speeds up to 100 Mb. Maximum length is 100 meters and minimum length between nodes is 2.5 meters.

100Base-FX

The FX means it's a two strand fiber cable and supports speeds up to 100 Mbs. Maximum length is usually up to two kms.

To summarize, keep the following in mind:

• 100Base-TX/T4 works for 100 Mb networks only and uses unshielded twisted pair cable with RJ-45 connectors at each end
• All CAT5 UTP cables have four pairs of cables (eight wires).
• 100Base-TX (sometimes called 100Base-T) uses two of the four available pairs within the UTP cable, whereas the 100Base-T4 uses all four pairs.
• 100Base-FX also works for speeds up to 100 Mb but uses fiber optic cable instead of UTP.

Next - 1000Base-T 1Gigabit & 10GBase Ethernet or Back to Network Cabling Section

Deciphering 10 BaseT/2/5/F/35

To make it simpler to distinguish cables they are categorized: that's how we got CAT1, 2, 3, etc. Each category is specific for speed and type of network. But since one type of cable can support various speeds, depending on its quality and wiring, the cables are named using the "BaseT" to show exactly what type of networks the specific cable is made to handle.

We are going to break the "10 Base T" into three parts so we can make it easier to understand:

10

The number 10 represents the frequency in MHz (Megahertz) for which this cable is made. In this case it is 10 MHz. The greater the MHz, the greater speeds the cable can handle. If you try to use this type of cable for greater frequencies (and, therefore, speeds) then it either will not work or become extremely unreliable. The 10 MHz speed translates to 10Mbit per second, which in theory means 1.2 Mbps. In practice though, you wouldn't get more than 800 kilobits per second (Kbps). 

Base

The word "Base" refers to Baseband. Baseband is the type of communication used by Ethernet and it means that when a computer is transmitting, it uses all the available bandwidth, whereas Broadband (cable modems) shares the bandwidth available. This is the reason cable modem users notice a slowdown in speed when they are connected on a busy node, or when their neighbor is downloading all the time at maximum speed! Of course with Ethernet you will notice a slowdown in speed but it will be smaller in comparison to broadband.

T/2/5/F/35

The "T" refers to "Twisted Pair" physical medium that carries the signal. This shows the structure of the cable and tells us it contains pairs which are twisted. For example, UTP has twisted pairs and this is the cable used in such cases. For more information, see the "UTP -Unshielded Twisted Pair" page where you can find information on pinouts for the cables.

10Base-T

A few years ago, the 10 BaseT cables used CAT3 cables, which are used for speeds up to 10Mbit, but today you will find mostly CAT5 cables, which are good for speeds up to 100 MHz or 100 Mbps, these cables are also used for 10 Mbps networks. Only two pairs of the UTP cable are used with the 10Base-T specification and the maximum length is 100 meters. Minimum length between nodes is 2.5 meters.

10Base-2

This specification uses coaxial cable which is usually black, sometimes also called "Thinwire coax," "Thin Ethernet," or "RJ-58" cable. Maximum length is 185 meters while the minimum length between nodes is 0.5 meters. 10Base-2 uses BNC connectors which, depending on the configuration, require special terminators. The 10Base-2 specification is analyzed here in great detail (also contains pictures) if you wish to read more about it.

10Base-5

This specification uses what's called "Thickwire" coaxial cable, which is usually yellow. The maximum length is 500 meters while the minimum length between nodes is 2.5 meters. Also, special connectors are used to interface to the network card, these are called AUI (Attachment Unit Interface) connectors and are similar to the DB-15 pin connectors most soundcards use for their joystick/MIDI port.

Most networks use UTP cable and RJ-45 connectors or Coaxial cable with BNC "T" connectors, for this reason special devices made their way to the market that allow you to connect an AUI network card to these different cable networks.

The picture below shows you a few of these devices:

Figure 1. AUI Connector to RJ-45 connector & AUI Connector to BNC 'T' Coaxial Connector

10Base-F

This specification uses fiber optic cable. Fiber optic cable is considered to be more secure than UTP or any other type of cabling because it is nearly impossible to tap into. It is also resistant to electro magnetic interference and attenuation. Even though the 10Base-F specification is for speeds up to 10 Mbits per second, depending on the type of fiber and equipment you use, you can get speeds of up to 2Gigabits per second!

10Base-35

The 10Base-35 specification uses broadband coaxial cable. It is able to carry multiple baseband channels for a maximum length of 3,600 meters or 3.6 Kms.

Summary

To summarize, keep the following in mind:

• 10Base-T works for 10 Mbps networks only and uses unshielded twisted pair cable with RJ-45 connectors at each end and maximum length of 100 meters. They also only use two pairs of cables.
• 10Base-2 works for 10 Mbps networks only and uses coaxial cable. Maximum length is 185 meters and BNC "T" connectors are used to connect to the computers; there are special terminators at each of the coaxial cable.
• 10Base-5 works for 10 Mbps networks only and uses thick coaxial cable. Maximum length is 500 meters and special "AUI" connectors (DB-15) are used to interface with the network card.
• 10Base-F works for 10 Mbps networks only and uses fiber optic cable.

 Next - 100Base-(T) TX/T4/FX - Ethernet or Back to Network Cabling Section

Why do we need an cross-over cable?

When sending or receiving data between two devices (I.E. computers) one will be sending while the other receives. All this is done via the network cable and if you look at a network cable you will notice that it contains multiple cables. Some of these cables are used to send data, while others are used to receive data and this is exactly what we take into account when creating a crossover cable. We basically connect the TX (transmit) of one end to the RX (receive) of the other!

The diagram below shows this in the simplest way possible:

CAT-5/Cat-5e/Cat6 Cross-Over Cable

There is only one way to make a CAT5e crossover cable and it's pretty simple. Those who read the "Wiring UTP" article know a crossover cable is a 568A on one end and a 568B on the other. If you haven't read the wiring section, don't worry because we’ll provide enough information to help understand about the concept.

As mentioned previously, the purpose of a crossover cable to connect the transmitting side (TX) from one end, to the Receiving side (RX) at the other end, and vice versa.

Let's now have a look at the pinouts of a typical crossover CAT5e cable:

As you can see, only four pins are needed for a crossover cable. When you purchase a crossover cable, you might find that all eight pins are used, these cables aren't any different from the above, as the rest of the pins are not used. Whether your crossover cable has 4 or 8 pins connected, it won’t make any difference in performance.

Here are the pinouts for a crossover cable with all eight pins connected:

It is important to note that Gigabit Ethernet does not require crossover cables. Straight-thru cables are used and the two ends are automatically crossed over by the network cards (Auto-MDIX).

Where else can I use a cross-over?

Crossover cables are not just used to connect computers, but a variety of other devices. Prime example are switches and hubs. If you have two hubs and you need to connect them, you would usually use the special uplink port which, when activated through a little switch (in most cases), makes that particular port not cross the tx and rx, but leave them as if they were straight through.

What happens though if you haven't got any uplink ports or they are already used? The Crossover cable will allow you to connect them and solve your problem. The diagram below shows a few examples to make it simpler:

As you can see in the above diagram, thanks to the uplink port, there is no need for a crossover cable.

Let's now have a look at how to cope when we don't have an uplink to spare, in which case we must make a crossover cable to connect the two hubs:

Almost all switches on the market today support the Auto-MDIX feature, allowing the connection between two switches with the use of a straight-thru network cable.

All the above should explain a cross-over cable, where we use it and why we need it. I thought it would be a good idea to include, as a last picture, the pinouts of a straight thru and a cross-over cable so you can compare them side by side:

 This completes our discussion on UTP CAT5, CAT5e, CAT6 cross-over cables. We' explained why cross-over cables are necessary, when we use them, their pinouts and much more. Next article deals with 10Base-T/2/5/F/35 - Ethernet standard, analyzing the 10BaseT technology and specification.

Next - 10Base-T/2/5/F/35 - Ethernet or Back to Network Cabling Section

We will be mainly focussing on the wiring of CAT5e & 6 cables as they are the most popluar cables around! We'll also cover wiring classic CAT1 phone cables. It is very important to understand UTP cabling standards and how to correctly terminate them.

Cabling is the foundation for a solid network, and implementing it correctly the first time will help avoid hours of frustration and troubleshooting. On the other hand, if you are dealing with a poorly cabled network, this knowledge will help you to find the problem and fix it more efficiently.

Wiring the UTP cables

We are now going to look at how UTP cables are wired. There are two popular wiring schemes that most people use today: the T-568A and T-568B. These differ only in which color-coded pairs are connected -- pairs 2 and 3 are reversed. Both work equally well, as long as you don't mix them. If you always use only one version, you're okay, but if you mix A and B in a cable run, you will get crossed pairs.

UTP cables are terminated with standard connectors, jacks and punchdowns. The jack/plug is often referred to as a "RJ-45," but that is really a telephone company designation for the "modular eight-pin connector" terminated with the USOC pinout used for telephones. The male connector on the end of a patch cord is called a "plug" and the receptacle on the wall outlet is a "jack."

Figure 1. A RG-45 Jack and RJ-45 Plug / Connector

As already mentioned, UTP has four twisted pairs of wires. The illustration shows the pairs and the color codes they have. As you can see, the four pairs are labeled:

Figure 2. Colour codes & Pairs of UTP CAT 5, CAT 5e, CAT6, CAT7 Cable

Pairs 2 and 3 are used for normal 10/100 Mbps networks, while pairs 1 and 4 are reserved. In Gigabit Ethernet, all four pairs are used.

The picture below shows the end of a CAT5e cable with an RJ-45 connector, commonly used to connect computers to a switch. It also shows a stripped CAT5e cable and identifies the four twisted pairs:

         
Figure 3. UTP CAT5e cable and stripped twisted pairs

UTP cables are now available in a variety of colours, making it possible to have different coloured cables for different applications.

Figure 4. Different coloured UTP cables

T-568A & T-568B 4-pair Wiring

Ethernet is generally carried in eight conductor cables with eight-pin modular plugs and jacks. The connector standard is called "RJ-45" and is just like a standard RJ-11 modular telephone connector, except it is a bit wider to carry more pins.

Note: Keep in mind that the wiring schemes we are going to talk about are all for straight-through cables only.

The eight-conductor data cable contains four pairs of wires. Each pair consists of a solid colored wire and a white wire with a stripe of the same color. The pairs are twisted together. To maintain reliability on Ethernet, you should not untwist them any more than necessary (about 1 cm). The pairs designated for 10 and 100 Mbps Ethernet are orange and green. The other two pairs, brown and blue are used when Gigabit Ethernet is supported or can be used for a second 10/100 Ethernet line or for phone connections. It should be noted that running a second Ethernet or phone line over an existing UTP cable, is not recommended as UTP wasn’t designed for such applications.

There are two wiring standards for UTP cables, called "T568A" (also called "EIA") and "T568B" (also called "AT&T" and "258A"). The only difference between the two standards is the wiring of two out of four pairs, which are swapped, as shown below.

T-568A is supposed to be the standard for new installations, while T-568B is an acceptable alternative. However, most off-the-shelf data equipment and cables seem to be wired to the T568B specification. T568B is also the AT&T standard. It’s perfectly right to use either wiring standard, however special consideration should be taken so that the same standard is used throughout the whole cabling infrastructure.   For existing installations, its best to first check which of the two standards is being used and continue with that standard.

Pin Number Designations for T568B

Note that the odd pin numbers are always the white with stripe color (1,3,5,7). The wires connect to RJ-45 8-pin connectors as shown below:

Figure 5. RJ-45 Connector and Jack with EIA/TIA 568B pinouts

The following table provides the color codes, pair numbers and their function, for the T568B wiring specification:

Figure 6. T568B Pinout configuration, colour code, pairs and their functionality

The wall jack may be wired in a different sequence because the wires are often crossed inside the jack. The jack should either come with a wiring diagram or at least designate pin numbers.

Note that the blue pair is on the center pins; this pair translates to the red/green pair for ordinary telephone lines which are also in the centre pair of an RJ-11 (green=white/blue, red=blue).

Pin Number Designations for T568A

The T568A specification reverses the orange and green connections so that pairs 1 and 2 are on the center four pins, which makes it more compatible with the telephone company voice connections. (Note that in the RJ-11 plug at the top, pairs 1 and 2 are on the center four pins.) The illustrations show the order of colors in T568A:

Figure 7. RJ-45 Connector and Jack with EIA/TIA 568A pinouts 

The following table provides the color codes, pair numbers and their function, for the T568A wiring specification:

Figure 8. T568A Pinout configuration, colour code, pairs and their functionality

The diagram below shows the 568A and 568B in comparison:

Figure 9. Comparing T568A and T568B wiring schemes and pinouts. Front side of the RJ-45 connectors are shown

Where are they used?

The most common application for a straight-through cable is a connection between a PC and a hub or switch. In this case, the PC is connected directly to the hub or switch, which will automatically cross over the cable internally, using special circuits. In the case of a CAT1 cable, which is usually found in telephone lines, only two wires are used. These do not require any special cross over since the phones connect directly to the phone socket.

Figure 10. Straight-thru UTP CAT5/CAT5e cable pinouts

The picture above shows a standard CAT5e straight-through cable used to connect a PC to a hub or switch. You might expect the TX+ of one side to connect to the TX+ of the other side, but this is not the case. When you connect a PC to a hub, the hub will automatically x-over the cable by using its internal circuits. The result is that pin 1 from the PC (which is TX+) connects to pin 1 of the hub (which connects to RX+).This happens for the rest of the pinouts as well.

If the hub didn't cross-over the pinouts using its internal circuits (this happens when you use the uplink port on the hub), then pin 1 from the PC (which is TX+) would connect to pin 1 of the hub (which would be TX+ in this case). So, no matter what we do with the hub port (uplink or normal), the signals assigned to the eight pins on the PC side of things will always remain the same. The hub's pinouts, however, will change depending whether the port is set to normal or uplink.

This article covered UTP Straigh-thru cable pinouts and wiring schemes for both T568A and T568B standards. We examined the UTP cable pairs along with their colour codes along with the RJ-45 connectors and jack.

Next - CAT5 UTP Cross-Over Cable or Back to Network Cabling Section

Characteristics of UTP

The characteristics of UTP are very good and make it easy to work with, install, expand and troubleshoot and we are going to look at the different wiring schemes available for UTP, how to create a straight through UTP cable, rules for safe operation and a lot of other cool stuff !

So let's have a quick look at each of the UTP categories available today along with their specifications:

Figure 1. The Different UTP Categories and their specifications

Category 1/2/3/4/5/6/7 – a specification for the type of copper wire (most telephone and network wire is copper) and jacks. The number (1, 3, 5, etc) refers to the revision of the specification and in practical terms refers to the number of twists inside the wire (or the quality of connection in a jack).

CAT1 is typically used for telephone wire. This type of wire is not capable of supporting computer network traffic and is not twisted. CAT1is also used by telco companies providing ISDN and PSTN services. In such cases the wiring between the customer's site and the telco’s network is performed using CAT 1 type cable.

CAT2, CAT3, CAT4, CAT5/5e,CAT6 & CAT 7 are network wire specifications. This type of wire can support computer network and telephone traffic. CAT2 is used mostly for token ring networks, supporting speeds up to 4 Mbps. For higher network speeds (100 Mbps or higher) CAT5e must be used, but for the almost extinct 10 Mbps speed requirements, CAT3 will suffice.

CAT3, CAT4 and CAT5 cables are actually 4 pairs of twisted copper wires and CAT5 has more twists per inch than CAT3 therefore can run at higher speeds and greater lengths. The "twist" effect of each pair in the cables ensures any interference presented/picked up on one cable is cancelled out by the cable's partner which twists around the initial cable. CAT3 and CAT4 are both used for Token Ring networks -- where CAT 3 can provide support of a maximum 10Mbps, while CAT4 pushed the limit up to 16Mbps. Both categories have a limit of 100 meters.

The more popular CAT5 wire was later on replaced by the CAT5e specification which provides improved crosstalk specification, allowing it to support speeds of up to 1Gbps. CAT5e is the most widely used cabling specification world-wide and unlike the category cables that follow, is very forgiving when the cable termination and deployment guidelines are not met.

CAT6 wire was originally designed to support gigabit Ethernet, although there are standards that will allow gigabit transmission over CAT5e wire.. It is similar to CAT5e wire, but contains a physical separator between the four pairs to further reduce electromagnetic interference. CAT6 is able to support speeds of 1Gbps for lengths of up to 100 meters, and 10Gbps is also supported for lengths of up to 55 meters.

Today, most new cabling installations use CAT6 as a standard, however it is important to note that all cabling components (jacks, patch panels, patch cords etc) must be CAT6 certified and extra caution must be given to the proper termination of the cable ends.

In 2009, CAT6A was introduced as a higher specification cable, offering better immunization to crosstalk and electromagnetic interference.

Organizations performing installations using CAT6 cabling should request a thorough test report using a certified cable analyzer, to ensure the installation has been performed according to CAT6 guidelines & standards.

CAT7 is a newer copper cable specification designed to support speeds of 10Gbps at lengths of up to 100 meters. To achieve this, the cable features four individually shielded pairs plus an additional cable shield to protect the signals from crosstalk and electromagnetic interference (EMI).

Due to the extremely high data rates, all components used throughout the installation of a CAT7 cabling infrastructure must be CAT7 certified. This includes patch panels, patch cords, jacks and RJ-45 connectors. Failing to use CAT7 certified components will result in the overall performance degradation and failure of any CAT7 certification tests (e.g using a Cable Analyzer) since CAT7 performance standards are most likely not to be met. Today, CAT7 is usually used in DataCenters for backbone connections between servers, network switches and storage devices.

The next pages show you how UTP cable is wired and the different wiring schemes. It's well worth visiting and reading about.

Reader interested can also visit our Network Cabling section to select amongst a number of articles covering UTP, X-Over cables, Fiber Optic cables and much more.

Back to Network Cabling Section