Structured Cabling
An Overview of Structured Cabling
Over recent years the technical development of copper cabling and data transmission systems has progressed at an incredible rate.
In the late 1970s when Ethernet was under development, one of the few readily available cables deemed suitable for a data
transmission rate of 10Mbit/s was a screened 50 ohm coaxial cable, used extensively for closed circuit television applications.
Similarly, when IBM was developing its Token Ring LAN System, it was thought that only a high quality electrically shielded
cable would be capable of supporting a transmission speed of 4mbit/s/s over the distances required.
Today, manufacturers are demonstrating systems that can transfer 1.2Gbit/s over low cost Unshielded Twisted Pair (UTP)
cables. This has been achieved by the use of sophisticated transmission and coding systems and improvements in the
transmission characteristics of the cables.
The following table indicates the relationship between data transmission speeds and signal frequencies:
Technology |
Data Transmission (Mbps) |
Maximum Signal Frequency (MHz) |
Ethernet |
10 |
10 |
Ethernet |
100BaseTx |
31.25 |
Ethernet |
100BaseT4 |
25 |
Ethernet |
100VG AnyLAN |
30 |
TP-PMD (CDDI) |
100 |
32.35 |
Token Ring |
4/16 |
4/16 |
ATM |
155 |
100 |
Gigabit Ethernet |
1000 |
125 |
Manchester encoding being used for the first three technologies, Non Return to Zero (NRZ) coding
for ATM and Pulse Amplitude Modulation level 5 (PAM 5) for Gigabit Ethernet.
It is interesting to see how critical the quality of the cable installation is nowadays
as end users place vastly increased demands on the infrastructure. A bad installation
of a Category 5 system could very easily cope with a 10 MHz signal. The user would
only notice the poor quality if they suddenly switched to using 70MHz or 125MHz frequencies.
The following schematic illustrates the basic components and sub-systems of a structured cabling system such as Avaya
Communication's Systimax(R) offering:
The cabling system is split into five subsystems:
- Campus Backbone Sub-system - includes both fibre and copper cables linking buildings locally,
external to internal cable splicing (external cables are often armoured and contain a water resist gel to protect
the copper cables) and lightning protection circuitry built into the splice boxes and frames.
- Riser Backbone Sub-system - includes both fibre and copper cables running between frames.
- Equipment Sub-system - includes equipment (or host) cables linking equipment to the frames plus
any baluns or adapters that may be required to run equipment protocols over the UTP. The Equipment
cables are often 4-pair RJ45 ended cables or 25-pair telco (RJ21) ended cables.
- Horizontal Sub-system - includes 4 pair UTP star-wired cables, plus 25 pair cable runs to outlet
blocks and includes the RJ45 (RJ stands for 'Radio Jack' and the RJ45 is an 8-wire plug/socket
similar to the American 4-wire RJ11, but larger) sockets themselves. The RJ45 socket typically has
a lifetime usage of about 1000 insertions/de-insertions.
- Work Area Sub-system - includes baluns or adapters if required, and the 4 pair flyleads linking
between the RJ45 socket and the PC/terminal/telephone etc.
- Administration Sub-system - includes jumper wire or 110/RJ45 plug-ended patch leads, the 110/RJ45 patch frames,
any adapters that are required at the frame and the labelling system used at the frames.
The MDF is the Main Distribution Frame, and is often located wherever the external data and voice services are supplied
into the building. It is common for this main administration point to be in the main computer room and is close
to the main servers and hubs/switches. Traditionally, voice services are administered here and extension numbers
distributed via copper riser cables to the IDFs (Intermediate Distribution Frames).
The IDF is where localised data services are administered and hubs/switches provide users with access to
the servers via fibre riser cables linking back to the MDF or locally placed servers. In addition, copper services
such as telephone extensions can be administered here.
The Copper patching frame commonly uses a number of 110 frames wall-mounted together and separated into a number
of fields, such as outlet, equipment or riser fields. Riser, equipment or outlet cables are terminated onto the back of these
frames using the Insulation Displacement Connection (IDC). The connection is made by use of a punchdown tool that pushes
each wire (24 AWG) into a narrow slot containing metal blades just wide enough to allow the solid copper core of the
wire to pass through, but not the insulation. The insulation is completely displaced and the copper has a clean firm
connection on the frame. Patching occurs on the front of the frame via the use of jumper wire, terminated in the same
fashion, or a patchlead of between 1 and 4 pairs which uses metal blades that push into 110 blocks mounted in rows
on the front of the frame.
The 110 frames come in 100pair, 300pair and 900pair sizes and you would typically require a patch cord management
trough (118D3) for every two vertical frames.
Often, RJ45 patching is requested and this can be incorporated into 19" racks easily linking into the equipment that
is also installed in 19" racks. Using RJ45 patching takes up more room than wall-mounted 110 patching, but it tends to
look more aesthetically pleasing. Due to the nature of RJ45s, there is some compromise on Near End Crosstalk (NEXT)
and the performance of the connection can not match a high quality 110 connection.
The LIU is the Lightguide Interconnection Unit and is effectively the patch frame for the fibre connections.
The fibre comes in a number of flavours:
- Multi-mode 62.5/125 micron. This was used originally in favour of the European 50/125 micron standard due
to difficulties in original SMA connectors attaching to the 50 micron core. Although 62.5/125 micron
has been widely used there are moves afoot to go back to using 50/125 micron due to the increased
distances attained for Gigabit Ethernet over multi-mode fibre without resorting to single-mode.
- Multi-mode 50/125 micron.
- Single-mode 8.3/125 micron.
And each cable could contain separately sheathed cores or bonded multi-core fibres in 'ribbon' format.
The connections occur via STII (spring held bayonet), STII+ (fixed bayonet) or SC (fixed) terminations.
The LIUs contain rows of 'couplers' each containing a terminated fibre on the back ready for a fibre
patch lead to connect into the front.
Provided that the installation has been carried out in strict accordance with the guidelines laid down within the
two courses 'Design and Engineering' and 'Installation and Maintenance' and engineers that have been through these
courses have overseen the design and installation, then Lucent will supply an Applications and parts warranty on the
cabling system that covers a period of 15 years from the point of completion of the installation. Three of the five
subsystems have to be in place before a complete system is deemed suitable.
Design Considerations
1. The horizontal cable runs must never exceed 90m. This includes any up and down routes that the cable may take,
e.g. rise and fall of cable trays, the distance up to the frame termination points etc. To cover yourself in a design
it is a good rule of thumb to allow only 80m runs. Lucent allow 100m horizontal runs but this includes flylead,
patch lead and equipment lead lengths. This requires you to design an IDF in as central an area as you can to minimise
the number of IDFs that are required. Consequently this minimises the number of places where expensive comms
equipment needs to be installed.
2. There are minimum bend radii requirements for the different cables used in SCS and they are as follows:
- 4 pair UTP (Cat 5) - 22mm smooth bends (or folded up to 90 degree angles)
- 200 pair UTP - 8 times the cable diameter or greater (EIA/TIA-TSB40)
- 4-core fibre - 51mm
- 12-core fibre - 77mm
Due to the development and improvements in the performance of UTP cabling, a number of different types of UTP cables are
available today. Each of these has specific advantages in terms of transmission characteristics and cost. To help classify the
types of UTP cables a number of organisations have developed a programme which characterises the cables.
The programme was originally instigated by a company called Anixter in the USA and was embraced by the Underwriter
Laboratories (UL). UL is a well recognised testing centre in the USA, which has produced a number of specifications relating
to cable performance, including fire and electrical safety. Both the Anixter and UL systems grade cables in terms of Levels
with Level 1 being the lowest grade cable with the worst transmission characteristics and Level 5 currently being the highest
grade of cable with the best transmission characteristics.
This programme has been officially sanctioned by the Electronic Industries Associated (EIA) and Telecommunications
Industry Association (TIA) of America which has also received support from a number of Associations and Organisations
outside of the USA, especially in the UK. EIA/TIA568 (issued in 1989) identifies the restraints for installing Premises
Structured Cabling Systems. The International Organisation for Standardisation (ISO) has adopted these standards and
has expanded them to produce an international standard.
Caution should be exercised when comparing the performance of cables classified under the Level and Category programmes.
Although both classification systems specify the same performance criteria, until recently there was fundamental differences
between the test methods used to assure compliance. The performance parameters at specific frequencies are the same in the
two systems but the Category Programme requires that cables are tested at all frequencies up to the maximum specified
frequency, not just at 'spot' frequencies. It is therefore important that users ensure that they are comparing 'like for like'
when comparing performance criteria.
|