S 5.3 Selection of cable types appropriate in terms of communications technology
Initiation responsibility: Head of IT
Implementation responsibility: Planner, Building Services Manager
In terms of communications technology, the selection of the cable is determined by the required data transfer rate (which is often referred to as the bandwidth, although this is not entirely correct) and the distances between the transmission units. In addition, the structural conditions, i.e. the cable trays and the ambient conditions present during installation and operation, must be considered. These conditions must also be taken into account when selecting the cables, because they also affect the cable design. In the following, the advantages and disadvantages in terms of IT security aspects are described.
The transmission systems used today use electrical or optical interfaces for cable-bound communication. Accordingly, the cables must provide metal conductors as the transmission medium for electrical transmission or synthetic or glass optical fibres (FO) for optical transmission.
In the following, copper and fibre optic cables are examined in more detail:
Twisted-pair cable
Copper cables for use in IT are designed symmetrically. In this cable design, two wires each are twisted together to form a pair, and four of these pairs are placed together to form a cable (twisted-pair cable or TP). The possible bandwidth and the immunity to interference of the cables differ depending on the diameter of the wires, their insulating material, including the dyes, the type of twisting, and the type of shielding used for the twisted pairs. For uniform labelling of the cable types, the 2nd edition of ISO/IEC 11801 "Information technology - Generic cabling for customer premises" suggests the following uniform type designation scheme to uniquely identify the construction elements, which are read from left to right:
Figure: Type designation system for copper cables
For example:
- unshielded U/UTP,
- unshielded with an overall shield for all wire pairs (F/UTP or SF/UTP),
- shielded, but the pairs of wires are shielded individually (U/FTP) - this used to be referred to as Pair in Metal foil (PiMf) - and
- like the design above, but with an additional overall shield (F/FTP, S/FTP and SF/FTP).
The standards specify categories and classes for the limit values of the transmission properties of cables and connection components. The categories describe the requirements and limit values for each individual element in the cabling infrastructure, and the classes describe these for the overall system installed.
The transmission properties for the individual components are currently divided into the categories 1 to 7. The following applies in this case: the higher the category number, the higher the possible transmission bandwidth.
High transmission quality can only be maintained reliably when a harmonious blend of cables and connection components (plugs and sockets) is selected and properly installed. The devices do not "detect" the length of the cables installed and instead react to electrical signals. For this reason, the specifications of the electrical limits of the cables are the most important parameters. According to ISO/IEC 11801, the maximum length for copper cables is 90m (100m including patch and connection cables). This maximum length can be exceeded, though, when the required electrical transmission parameters are maintained.
The TP cable is the standard according to the cabling standards when cabling the general access area on a floor. This type of cable offers the following advantages:
- TP cables, and in particular their fabrication, are relatively inexpensive compared to FOCs when less bandwidth is needed.
- TP cables are relatively easy to install and fabricate.
- TP cables can also be considered a type of universal cabling since other services (e.g. telephony) can use these cables without additional complicated technology.
- The performance of the installations is easy to measure.
- TP cables allow devices built according to the requirements in the "Power over Ethernet" (PoE) specification to be supplied with power.
They have the following disadvantages, though:
- Due to the alternating currents flowing when data is transmitted and minor asymmetries which are always present in the twistings of the wires, electromagnetic fields are generated which can be detected in the environment (risk of eavesdropping) and disrupt systems. In turn, electromagnetic fields in the environment can also disrupt the transmission of data in the cable as well.
These effects are minimised by using shields in the cable design (see also U/UTP through SF/FTP). The specifications for the minimum distance between different types of cables, lines, and systems as well as for the grounding of shields must be maintained. - The effects mentioned also have an effect inside the cable. Unshielded installation cables (U/UTP) provide the least protection against crosstalk between separate pairs of wires. In this case, only twisting the individual wires has an effect.
Fibre optic cable (FOC)
When signals are transmitted in fibre optic cables, light in the visible to far infrared range is used for transmission. Diodes or lasers are used to generate the light signals. These elements convert the electrical signal to light modes with different polarisations and/or different concentrations.
The fibre optic cable, also referred to as the fibre, consists of a core used for transmission and a cladding material around the core. The materials differ in terms of their so-called indices of refraction.
The cabling standards define the categories OM-1, OM-2, and OM-3 for multimode fibre optic cables. These specifications are for fibre optic cables with a gradient refractive index and a nominal core/cladding diameter of 50/125 or 62.5/125 micrometers. Single-mode fibre optic cables are in category OS-1. The nominal core/cladding diameter of single-mode fibre optic cables is 9/125 micrometers.
While several light modes of a signal are coupled in a multimode fibre, only one light mode is coupled in single-mode fibres due to the small diameter of the core. This means the possible bandwidths and maximum lengths that can be reached without additional amplification are different for the two types of fibre. Different types of fibre cannot be combined when connecting systems in some cases.
Fibre optic cables are used in the following areas, amongst other things:
- when spanning large distances in wide area networks (WANs),
- in metropolitan area networks (MANs),
- in company networks (local area networks or LANs) for connections between buildings and on floors,
- in areas subject to high electromagnetic interference, and
- in storage area networks (SANs) in computer centres to connect systems at the highest possible data rates.
The connectors selected for the optical fibre infrastructure are also a deciding factor for the quality of the connections.
Using fibre optic cables offers the following advantages:
- Fibre optic cables permit higher bandwidths with longer maximum lengths when compared to copper cables.
- Fibre optic cables are insensitive to electromagnetic fields.
- there are no crosstalk effects like with electrical conductors.
- Fibre optic cables provide a potential-free connection between the end points of the cable.
- Fibre optic cables can only be tapped using complex technology.
- Cables with a large number of fibres can be built more compactly than comparable copper cables and weigh much less at the same time.
- The fire load of fibre optic cabling is lower in comparison to the fire load of copper cabling. The reasons for this include the comparably less material required, the combination of materials in the cable design, and the possible high number of fibres without massively increasing the size of the finished cable.
Fibre optic cables have the following disadvantages, though:
- The cost of installation for FOC is higher than for copper cables, primarily due to the splicing work required.
- The coupling components used to operate the FOC, especially for single-mode FOC, are more expensive than those for copper cables.
- Most common workstation computers provide better support for LAN connections using TP cables than FOCs. Workstation clients nowadays are usually connected to the LAN using copper cable.
The following table provides an overview of the length restrictions for cables for some of the most common protocols:
| Network access protocol | Cable type | Max. length | |
|---|---|---|---|
| Ethernet | 10Base-T | TP | 100m |
| 10Base-FL mono-mode | Multimode FO | 2,000m | |
| 10Base-FL single-mode | Single-mode FO | 25,000m | |
| Fast Ethernet | 100Base-TX | TP Cat 5 | 100m |
| 100Base-FX | Multimode FO | 400m | |
| Gigabit Ethernet | 1000Base-T | TP Cat 5e | 100m |
| 1000Base-SX | Multimode FO | 550m | |
| 1000Base-LX | Multimode FO | 550m | |
| 1000Base-LX | Single-mode FO | 10,000m | |
| 10 Gigabit Ethernet | 10GBase-T | TP Cat 6a | 100m |
| 10GBase-LX4 | Multimode FO | 300m | |
| 10GBase-LW4 | Single-mode FO | 10,000m | |
| 10GBase-SR | Multimode FO | 300m | |
| 10GBase-LR | Single-mode FO | 10,000m | |
| 10GBase-ER | Single-mode FO | 40,000m | |
| 10GBase-LW | Single-mode FO | 10,000m |
Table: Length restrictions for common types of cable
Note that all lengths stated above are maximum lengths. The maximum length often consists of the length of the installation cable itself plus the connecting cables (patch cables). For 1000Base-T, for example, the length of the installation cable should not exceed 90 meters so that there is enough length left for the patch cables.
Summary
In WAN and MAN networks, fibre optic cabling with single-mode fibres are standard. It is absolutely recommended to use theses fibres in LAN cabling nowadays between buildings and for floor distributors located farther away due to the length restrictions of 10 Gigabit Ethernet.
The use of fibre optic cables up to the workstations, and therefore the elimination of copper cabling on the floor, can only be evaluated when assessing the situation as a whole.
The following items argue in favour of using fibre optic cables:
- the lower fire load situation,
- the higher level of security available with fibre optic cables,
- EMC neutrality,
- possible savings in cable channel and tray construction,
- saved space due to the lower number of distribution rooms required and therefore lower expenses for electrical cabling for the distribution rooms, and
- simplified UPS and grounding concepts.
The following items argue against using fibre optic cables on the other hand:
- the higher cost of interface cards for the terminal devices and for the network components,
- the continuing need for the most part to connect telephone equipment using copper cables, and
- possible restrictions on the ability to implement power-over-Ethernet for IP telephony or even to connect access points in the WLAN.
For new installations and when modernising, it is therefore recommended to develop and evaluate the technical, safety, and economic requirements together with a qualified planner (see also S 5.2 Selection of an appropriate network topology).
Review questions:
- Does the selection of cable types take into consideration both communications- and security-related and constructional requirements?