CHAPTER 6. Wiring the Network—Cables, Connectors, Concentrators, and Other Network Components

SOME OF THE MAIN TOPICS IN THIS CHAPTER ARE

---

Structured Wiring 74

Important Definitions 77

Physical Cable Types 81

Terminations and Connections 92

Telecommunications Rooms 99

Bridging the gap between the stated standards and the actual implementation of bringing a network to a user’s desktop workstation is not the simple task you might expect. Connecting tens, hundreds, or even thousands of computers can become an exercise in futility if proper planning is not done. Planning and installing your cable plant carefully is vital for ease of future upgrading and expanding your network.

For more information about the planning necessary for a successful network implementation, see Chapter 2, “Overview of Network Topologies,” and Chapter 3, “Network Design Strategies.”

This chapter covers quite a few technical details that relate to the network cables and other components used in your network. Although the definitions and other material you’ll find in this chapter might seem overwhelming at first, this chapter is a good reference when you encounter some of these terms later in the book.

Structured Wiring

In the 1980s, the Telecommunications Industry Association (TIA) and the Electronics Industries Association (EIA) formed a task force to establish a set of standards for installing network wiring in buildings. The first draft was completed in 1991 and became known as EIA/TIA-568 (referred to in this chapter as 568). A more recent standard is named ANSI/TIA/EIA-569-A (referred to in this chapter as 569-A). These standards documents encompass structured wiring, cables, network topology, connectors and hardware, electrical performance specifications, physical termination, and support mechanisms.

The 568 and 569-A standards describe the physical layout and specifications for the physical plant as it relates to the various topological standards. The physical plant, in this context, comprises everything having to do with what leads up to your desktop, from routers, cables, patch panels, and so on.

These are the basic topics covered in the standards:

The work area—The termination point of the network at a user’s work space.

The backbone cabling system structure—Connections between multiple telecommunication rooms, equipment rooms, and entrance facilities.

The horizontal cabling system structure—Connection from the telecommunications outlet in the work area, terminating in the telecommunications room.

The telecommunications closet—The central wiring point for a floor. The telecommunications closet can contain both network devices and concentrators (such as switches), as well as telephone equipment.

Other specifications—Such issues as intrabuilding connections and such factors as electromechanical interference.

In the following sections you will learn about these topics, as well as others. For a more complete explanation of the standards, it is suggested that you obtain the standards and read them. It is beyond the scope of this book to describe the standards in detail. Instead, those that apply to networks are discussed. In addition, several terms are defined for those who are not well versed in the terminology used by these standards. Many of these terms are also used throughout other chapters in this book. Between this chapter and the glossary, you should be able to locate the definition of almost any word used by network administrators, and those who put together LANs, MANs, and WANs.

The Work Area

The work area includes the telecommunications outlet (that is, the faceplate into which you plug your computer’s network cable at your desk), which serves as the work area interface to the entire network cabling system. Work area equipment includes cables used to connect to the telecommunications outlet. The following are the work area cabling specifications:

Equipment cords are assumed to have the same performance as patch cords (in the telecommunications closet) in the same typing category, for example, Category 5, 5e, and 6 network cables.

When used, adapters are assumed to be compatible with the transmission capabilities of the equipment to which they connect.

Horizontal cable links are specified with the assumption that a maximum cable length of 5 meters (16 feet) is used for equipment cords in the work area. This can depend on the actual length of cabling used to connect the work area back to the telecommunicatons closet. The important factor to remember is that there is a maximum distance that all cables can add up to, depending on your topology. Thus, if the cable from the telecommunications closet is less than the specified length allowed by the standards, you can use a longer cable from the termination point at the work area.

The Backbone Cabling System Structure

The backbone cabling system of the standard provides interconnections between telecommunication rooms, equipment rooms, and entrance facilities (see Figure 6.1).

Figure 6.1. The backbone of the network includes the cables that connect different areas of the network.

This cabling system includes backbone cables, intermediate and main cross connects, mechanical terminations, and patch cords or jumpers used for backbone-to-backbone cross connections. The backbone also extends between buildings in a campus environment.

There are some points specified for the backbone of the cabling system:

Equipment connections to the backbone cabling should be made with cable lengths of 30 meters or less.

The backbone of cabling should be configured as a star topology.

The backbone is limited to no more than two hierarchical levels of cross connects—main and intermediate. No more than one cross connect can exist between a main and a horizontal cross connect, and no more than three cross connects can exist between any two horizontal cross connects.

A total coax backbone distance of 90 meters is specified for high-bandwidth capability over copper. This distance is for uninterrupted backbone runs.

The distance between terminations in the entrance facility and main cross connect should be documented and made available to a service provider.

Recognized media can be used individually or in a combination, as required by the installation.

Multipair cable is allowed, as long as it satisfies the requirement of a minimum of cross-talk.

The proximity of cabling to sources of electromagnetic interference should be taken into account.

Cross connects for different cable types must be located in the same facility.

Note that in these specifications, bridge taps and splitters are not allowed.

The Horizontal Cabling System Structure

The horizontal cabling system (shown in Figure 6.2) extends from the telecommunications outlet in the work area and terminates in a horizontal cross connect in the telecommunications room. It includes the telecommunications outlet.

Figure 6.2. The horizontal cabling extends from the telecommunications closet to the user’s work area.

The distance covered by the horizontal cabling is limited by the network topology chosen for your network. For example, in most Ethernet networks, this distance is 90 meters. Token-Ring has various specifications, depending on the cables used. For more information about Token-Ring, see the chapter “Token-Ring Networks,” located on the upgradingandrepairingpcs.com Web site.

The Telecommunications Closet

Telecommunications rooms generally are considered to be floor-serving facilities for horizontal cable distribution. They also are used for intermediate and main cross connects. The telecommunications room is where you place patch panels, as well as hubs or switches that are used to connect individual workstations or servers to the network backbone.

Important Definitions

In discussing network cables and troubleshooting wiring problems, there are several important terms and concepts to understand. This section contains definitions of some of the terms used earlier in the chapter. This section should be considered as an expanded glossary of terms associated with the Physical layer of the OSI model (discussed in Appendix A, “Overview of the OSI Seven-Layer Networking Reference Model”). In other chapters you will find references to these terms. Although it is not required reading for the casual reader, the following can serve as an invaluable reference when making purchasing decisions, as well as during the design phase of a new network or when upgrading an existing one.

Attenuation to Cross-Talk Ratio (ACR)

ACR is a critical factor in determining the capability of an unshielded twisted-pair cable or shielded twisted-pair cable. Attenuation to cross-talk ratio (ACR) is the value of the attenuation less the crosstalk value, both expressed in decibels (db) at a particular frequency. This is a quality factor for cabling. Before you can understand this ratio, you need to understand what the term attenuation means.

Attenuation

Attenuation is the decrease in magnitude of the signal as it travels through any transmitting medium, such as wire or glass. Attenuation is measured as a logarithm of the ratio between the input and the output power or between the input and the output voltage of the system. It’s expressed in db. All good things must come to an end, and this is the case with electricity as well as light. As the signal travels down the copper wire (or the fiber-optic cable), some of the signal is lost. This is why it is necessary in a network topology to impose specific limits on the lengths of cable you can use. After you get past certain limits imposed by a particular topology, the signal becomes so degraded that the data transmitted cannot be reliably recovered at the destination.

Figure 6.3 shows that attenuation occurs as the signal travels down the wire. The amplitude of the electrical signal decreases the farther it travels from the transmitting side of the communications channel.

Figure 6.3. An electrical signal degrades as it travels through a copper cable (attenuation).

Bandwidth

Bandwidth is the range of frequencies required for proper transmission of a signal. This is expressed in hertz (Hz) as a difference of frequencies. For example, the bandwidth used on a copper wire for voice communications (via the PSTN) is 4MHz. Because copper cables are capable of carrying frequencies well above this 4MHz limit, DSL transfers are possible via use of frequencies above the 4MHz used by voice communications.

Characteristic Impedance

Characteristic impedance is the value of impedance (a combination of resistance and reactance) of a transmission line measured over a specific frequency range. Impedance is expressed in units of Z, because it is a calculation based on both resistance and reactance of the network media. Whereas resistance is the capability of a medium to resist the transmission of electrons, reactance is another thing altogether. Reactance, for alternating current (AC), is the medium’s tendency to store and then release the current as it flows through the medium.

Cross-Talk

Cables are made up of two or more copper wires that are bundled together with an outer cover so that it’s easier to route them through the conduits that form the path your physical network takes. The coupling of signals from one pair of wires in a cable to another pair of wires in the same cable actually can cause the signals to interfere with each other. The electrical signal in a copper wire not only travels down that particular wire, but also radiates out perpendicularly and can interfere with other copper wires in the same cable or bundle. This is called cross-talk. This coupling also can occur between wires of different cables that are close to one another. In Figure 6.4, you can see that some of the signal has radiated from one wire and produced noise on another.

Figure 6.4. A portion of an electrical signal radiates from an adjacent wire, producing interference (cross-talk).

Dialectic

To keep the individual copper wires separated from each other within a cable, an insulating material called dialectic material is used to help prevent interference between two conductors. It can be a simple plastic nonconducting material, or a more complex formulation used in some high-capacity wire bundles.

Electromagnetic Field

As electrons move through a medium, two fields are associated with this movement: electric fields and magnetic fields. These fields exist at varying distances from the conductors (the wires) as they are brought closer together.

Electromagnetic Interference

Electromagnetic interference (EMI) refers to the interference that electromagnetic signals produce by frequent changes of electrons moving through certain media. Network wiring and equipment can be very susceptible to EMI, and they also emit EMI.

Far-End Cross-Talk (FEXT)

Far-end cross-talk occurs between two twisted pairs of the cable at the far end (destination) of the cable from the measuring source. The transmitting end of a cable pair produces the stronger electrical signal (because the signal attenuates, or becomes weaker, as it passes through the copper wire or fiber-optic cable), so FEXT can be a more difficult problem to tackle. However, you should be cognizant that connectors are properly created at the far end of a connection to prevent interference between copper wires at that point. The signal might be weaker at the end point of a connection, yet it still exists. This is why the specifications allow only a very small amount of exposed copper wire when connecting a cable to an actual connector (such as an RJ-45 jack).

Frequency

Frequency is a measurement of the number of times a periodic action occurs in a measure of time. In terms of alternating current, this is the number of cycles per second and is usually expressed in hertz.

Full-Duplex and Half-Duplex Communications

Full-duplex communications means that communications between two network nodes can occur in both directions simultaneously. Obviously, this is a communications method in which both transmitted and received signals are not simultaneously present. They alternate in time on the transmission medium. Another method for creating a full-duplex connection is to use separate wires for transmission and reception. Using this method, both ends of a communications link can send or receive data simultaneously. Using half-duplex communications, only one side of the communications line can transmit at any point in time.

Impedance

Impedance is the total resistance and reactance offered by a circuit component. The units are expressed in ohms. The common symbol for impedance is the Greek letter zeta, or Z. This is a complex numerical value, mathematically expressed as either a complex number or the polar coordinate number.

Impedance Match

Impedance match is a condition in which the impedance of a device or wiring system is matched to another wiring system or device.

Leakage

Leakage is the undesirable passage of current through an insulator or over the surface of a conductor. This can occur in older cable bundles in which the insulating material has become degraded over time and signals from one wire in the bundle interfere with signals in other wires in the same bundle. This could happen, for instance, if a small animal were to attempt to chew through a cable. Feel sorry for the rat, but feel sorrier for yourself when you have to replace the cable!

Near-End Cross-Talk (NEXT)

Near-end cross-talk is cross-talk that occurs between two twisted pairs measured at the same location, and it usually occurs between wires in a twisted-pair cable. One of the conditions that can introduce this interference is a crushed cable, so care must be used when pulling network cabling and attaching connectors.

Nominal Velocity of Propagation

Nominal velocity of propagation is the speed at which a signal travels through a medium expressed as a decimal fraction of the speed of light in a vacuum.

Power SUM

Power SUM is a measured parameter that includes the sum of the contributions of power from all pairs of a cable system, excluding the pair under test. This is done with all the other pairs of the cable having signals present.

Radio Frequency

Radio frequencies are frequencies in the electromagnetic spectrum that are used for radio communications. These generally occur above 300KHz.

Radio Frequency Interference (RFI)

Radio frequency interference is electromagnetic interference at radio frequencies.

Shield

A shield is a metallic foil or wire screen-mesh that encircles a cable or wires in a cable to prevent electromagnetic or radio frequency fields from entering or leaving a cable. This prevents interference with other cables in the same cable bundle or cables that are in close proximity to the shielded cable. You’ve probably heard the term STP used in connection with network cables. This is an abbreviation for shielded twisted-pair cabling, typically found in Token-Ring networks. Although shielding cables were considered to be important in earlier network implementations, unshielded twisted-pair cables (in which the twisting of wires helps to reduce interference) are the norm today for cables connecting the work area to the telecommunications closet.

Time Domain Reflectometry (TDR)

Time domain reflectometry is a method of measuring cable length or faults by timing the period between a test pulse and its reflection from an impedance discontinuity on the cable. A TDR measuring instrument can enable you to determine the approximate location of a problem on a cable. It can also be used to determine any defects in a spool of cable before you deploy it in your network. You can learn more about TDR in Chapter 49, “Network Testing and Analysis Tools.”

Physical Cable Types

Much attention is given to the specifications used for Ethernet and Token-Ring networks. Most of these specifications deal with the physical makeup of the cabling involved in connecting the individual components of the network. Understanding the cable types, the number of wires in a cable (both shielded and unshielded varieties), and several other electrical factors is critical to successfully upgrading or maintaining your network system. The type of cable you use depends on the type of LAN you are creating. The connectors, terminations, and distances that can be covered by particular cable types will be a factor in determining any cable length restrictions and overall quality of LAN you can create. For example, an Ethernet network card and a Token-Ring network adapter card use different connectors and cables.

Twisted-Pair Cabling

The most basic wire type used for LAN wiring is twisted-pair wiring. This wire type also is referred to as unshielded twisted pair, or UTP. This wire is a derivative of the more common cable that was used in telephone installations in most commercial facilities for years. This type is versatile, is easy to install, and has favorable performance characteristics. It comes in various colors, wire gauges, insulation materials, twisting methods, and outer jacket materials.

The basic cable assembly for UTP cable can contain a large number of conductors (or copper wires). Most conductors are grouped into pairs that are twisted around each other. Telephone cables are available in 2, 4, 6, 25, 100, and even larger groupings of conductors. Most of the cable that is used for LAN wiring comes as cable consisting of four pairs of wires.

The four-pair cable has become a standard and is referenced in the EIA/TIA-568B cabling standards. This is the cable around which most of the cable standards and performance tests are based. Several of the LAN topology configurations use only two of the four-wire pairs; however, some use all four pairs. Another common cable type that is found in twisted-pair installations is a 25-pair jumper cable. This cable type is primarily used between patch panels and connector type punchdown blocks.

The main difference between typical telephone wiring and LAN wiring is the grading of the assembly of the twisted pairs within the cable. The primary factor that differentiates one cable type from another is the number of twists per foot that each individual pair of conductors has within the cable. The twisting of the individual pairs to the cable is significant. The twisting of the two wires has a twofold effect electrically on the cable assembly. First, it causes the interline capacitance to be reduced. This is a good thing because the reduction of capacitance reduces any signal shorting between the conductors at high frequencies. Second, twisting the wire couples the electromagnetic fields equally, thus helping to cancel out any interfering signals. This operation is referred to as a balanced transmission. One effect of achieving a balanced transmission is that the high frequencies of a LAN signal do not interfere with the other use of the wires in a cable assembly. Some radiation of the signal does occur, but because the transmitted signal is kept to a low amplitude, random emissions remain within acceptable limits.

Typically, wire sizes for UTP cable range between 18 AWG and 32 AWG. AWG, which stands for American wire gauge, is the standard for sizing wires in the U.S. Wire size is based primarily on the current-carrying capacity of the wire as set by the National Electrical Code. As the wire gauge increases, the physical diameter of the wire decreases. So a number 10 AWG wire is physically smaller than a number 8 AWG wire. Number 10 AWG wire is approximately 0.1 inch in diameter and usually can carry approximately 30 amps of current.

So for telephone wiring or LAN wiring, the number 18 AWG wire is much larger than the 32 AWG wire. Common sizes for LAN wiring are typically 22 to 24 AWG. This wire is typically solid, not stranded, for ease in termination on insulation displacement connectors.

Categories of Twisted-Pair Cables

As mentioned previously, the twisting of the wire pairs of conductors that make up cables is important—so important that the cable used for LAN wiring is graded into categories. Category 1 was used for POTS (or plain old telephone service). Category 2 was used in early networking wiring schemes, such as ARCnet, and for connecting terminals to multiuser computer networks. Category 3 uses four twists per foot and is still graded for operating as a LAN wiring system. Category 3 is rated for speeds up to 16MHz and is still used as a cable in some Token-Ring networks. Category 4 is rated up to 20MHz. Category 5 is rated for up to 100MHz operation. Category 5 was, until recently, the de facto standard for LAN wiring; however, it has now been replaced by two new categories: Category 5e and Category 6.

Category 5E differs from Category 5 in that it adheres to standards 568B.1 and B.2 and additional Class D requirements of the ISO/IEC 11801. These standards require a tunable frequency limit of 100MHz and are a superset of Category 5 and Class D.

During the first quarter of 2001, Category 6 cabling was certified for use as a standard. Transmission characteristics are specified up to 250MHz. Also called Class E according to ISO/IEC, this cable probably will represent the last generation of unshielded twisted-pair cabling that is used in LAN wiring. This cable is different from the standard UTP cable because it contains filler material to separate the twisted pairs from each other, and thereby reduces cross-talk between wire pairs. One of the biggest problems with using higher frequencies through the pairs of the cables is that adjacent conductor capacitance is reduced and cross-talk increases. Separating these conductors reduces this capacitance and cross-talk. This also is a consideration when installing cables because if cables are bundled too tightly there can be a resulting chance of interference of data signals between individual cables. Hence, modern standard practice dictates that when cables are installed, they are to be installed with either loose cable ties or Velcro straps.

The latest category, Category 7 UTP cabling, offers a different approach to twisted-pair cabling architecture. The cable is assembled with an overall shield and individually shielded pairs. The most significant improvement with this type of cable will be in the higher performance bandwidth achieved with ratings up to 600MHz. It is also designed to be backward-compatible with lower-performance categories and classes, although it appears it will have a new interface for the jack and plug. It is interesting to note that TIA is not actively developing a standard for Category 7. This organization probably will try to assimilate a standard with class F standards put forth by the ISO.

Performance Comparison

The choice of cabling insulation material is important. Requirements set forth in the National Electrical Code (NEC) specifically and stringently place requirements on the type of cable insulation allowed in certain portions of buildings. There’s an increasing use of large amounts of cable for LAN wiring, and these cables are usually installed above drop ceilings and below computer-room raised floors (also known as the plenum space). Unfortunately, these areas are most often used to handle cooling and environmental air. Conventional wire installations installed in these locations were found to be flammable at the very least, and at their worst, would produce toxic gases from the materials that surround cable bundles that would be carried with the cooling or environmental air, thus placing people in the other parts of the building at risk. Additionally, fire can actually be spread through the plenum areas.

Manufacturers soon developed cable installations that were less flammable and could be used in plenum-rated areas. The National Electrical Code differentiates cable types by voltage, power classifications, and insulation types. It should be noted that there is a definite difference between plenum- and riser-rated cable. It would seem that riser-rated cable would be classified higher than plenum-rated cable, because riser-rated cable is intended for use in vertical shafts that run between floors. The shafts are not normally used to handle environmental or cooling air except in ductwork. Thus, the cable installed in risers does not have to have insulation rated as stringently as that for cabling installed in plenum-rated areas.

The special requirements for cabling insulation and power ratings are covered in detail in the NFPA National Electrical Code Sections 770 and 800, for those who want to pursue these details further.

Color Coding and Marking

Each pair of wires in a twisted-pair cable assembly is color coded so that each wire can be identified at each end of the cable assembly and terminated properly. This color code is shown in Figure 6.5.

Figure 6.5. Wire pairs in a cable are color coded.

As you can see, each pair of the cable is color coded in a complementary fashion. Pair-one wires are color coded white-blue and blue-white. The blue-white wire has as its base color blue insulation with a white stripe molded at intervals along its length. The stripe is sometimes called a tracer. The white-blue wire is color coded in reverse, with a white wire that has a blue tracer. The color code is unique for each pair and is repetitive.

The color coding is important in LAN wiring because the system signals are polarity sensitive. If pairs on the cable are reversed, the signals are reversed, causing a failure in the receiving equipment. The terms tip and ring, used to designate the polarity of each pair of wires, stem from the days of the old telephone patch panels. The equipment used consisted of quarter-inch phone plugs, which fit into corresponding jacks on a patch board or switchboard. The switchboard plug consisted of two parts. The tip of the plug was wired through the sleeve or ring of the plug. The plug used on audio equipment and musical instruments is the same plug. The primary color was wired to the ring and the secondary color was wired to the tip.

As mentioned before, there is also a use for 25-pair jumper cable. This color code, broken down by pair, is shown in Table 6.1.

Table 6.1. Color Coding for a 25-Pair Jumper Cable As Specified by the ICEA

This cable must be rated for the category for which it’s to be used. Physically, it mostly is used between patch panels and punchdown blocks, or between patch panels to patch panel installations. Cable sizes above 25 pairs are usually in groups of 25-pair cables. Each of these groups of cables is marked within the larger bundle with a wrapped colored leader that, by design, is color coded with the same color code that is used on the twisted-pair cabling scheme. Thus, on a 50-pair cable, which would have two 25-pair cables, the 25-pair bundle has an outer spiral wrap of a blue plastic streamer, and the second group of 25 has a group wrapped with an orange streamer. This color code can be repeated ad infinitum for a very large group of 25-pair cables.

Coaxial Cables

Coaxial cables are the original LAN cable. This cable was first used in Ethernet networks, IBM PC net broadband networks, and ARCnet networks, besides being used for video and cable television applications. It still is in use in many older locations, even though newer installations have converted to twisted pair. Coaxial cable has been around long enough that it has a mature construction technology and is relatively inexpensive. The primary advantages of coaxial cable are its self-shielding properties, its low attenuation at high frequencies, and its moderate installation expense. For example, if you have a cable modem in your house, a coaxial cable is used for both the video signals and the cable modem frequencies.

Coaxial cables consist of the conductor centrally positioned in a cable surrounded by an insulating medium, which then is enclosed by a shield (see Figure 6.6). The shield can consist of a foil wrapping within an integral drain wire or a wire braid. The coax that is used for thick Ethernet might have a double-shield layer.

Figure 6.6. Coaxial cable consists of a shielded copper wire.

Placing the center conductor in an insulating medium surrounded by a shielding material theoretically traps all the electromagnetic fields inside the cable assembly. Because this shield has to be grounded, the mode of propagation of the signals in the cable is analogous to that of a mechanical pipeline. The grounded shield helps prevent interfering signals outside the cable from impinging on the center conductor. Conversely, the grounded shield also prevents signals from leaking out of the cable structure. Grounding is very important in this cabling system. A cable installation without proper grounding is susceptible to outside EMI and RFI interference.

Types of Coaxial Cabling

Two types of coax cabling are used in wiring local area networks. One type is called thicknet and the other is referred to as thinnet. Thicknet was used in the original Ethernet coax trunk distribution cable now known as 10BASE-5. The cable has a large center diameter conductor of number 12 AWG and has an overall diameter of approximately 0.4 inch. This cable typically was run close to a workstation either in the ceiling or in the walls. A connection is made to the cable by literally tapping to the wire by punching a hole through it (commonly known as a vampire tap), and then the connection is directly made by running a cable from the tap to the networked device, such as a terminal or a PC. The term thicknet was introduced because this coaxial cable is almost a half-inch in diameter, and when newer, smaller diameter coaxial cable was introduced it looked very large. A newer standard cable, which is approximately a quarter-inch diameter and is much more flexible, was named thinnet (10BASE-2). This type of cable uses BNC connectors with T-adapters. It is less expensive, but the distances for a thinnet Ethernet segment are limited compared to those for thicknet. ARCnet is another LAN topology that uses coaxial cable. In this topology, the workstations are connected directly to the coax in a star arrangement. Each leg terminates in an active or a passive hub.

ARCnet is one of the oldest networking technologies still in use today, although you’re more likely to find it used in point-of-sale mechanisms, linking electronic cash registers, for example, than in other types of networks.

Typically, the sizes for coaxial cables are designated by an RGB number or a manufacturer’s numbering system. A summary of cable types follows:

Cable RG 59/U-1 (105 482 624) is a 75-ohm coaxial cable with a 22 AWG (7?30) center conductor, a foamed polyethylene dielectric, a bare copper braid (mm. 95% coverage) outer conductor, and a PVC jacket. (Similar to RG 59/U type.) UL style 1354.

Cable RG 59/U-lA (105 521 561) is a 75-ohm coaxial cable with a 22 AWU (7?30) center conductor, a foamed polyethylene dielectric, a bare copper braid (mm. 95% coverage) outer conductor, and a PVC jacket. (Similar to RU 59/U type.) UL style 1354, UL Listed Type CL2.

Cable RG 59/U-2 (105 482 632) is a 75-ohm coaxial cable with a 22 AWG copper covered steel center conductor, a polyethylene dielectric, a bare copper braid (mm. 80% coverage) outer conductor, and a PVC jacket. (Similar to RU 59/U type commercial.) UL style 1354.

Cable RG 59/U-2A (105 521 579) is a 75-ohm coaxial cable with a 22 AWG copper covered steel center conductor, a polyethylene dielectric, a bare copper braid (mm. 80% coverage) outer conductor, and a PVC jacket. (Similar to RU 59/U type commercial.) UL style 1354. UL Listed Type CL2 per 1987 NEC.

Cable RG 59/U-5 (105 482 665) is a 75-ohm plenum coaxial cable with a 22 AWG copper covered steel center conductor, an FEP dielectric, a bare copper braid (mm. 95% coverage) outer conductor, and an FEP jacket. (Similar to RU-59/U type.) UL Listed Type CL2P per 1987 NEC.

Cable RG 62 A/U-1 (105 482 723) is a 93-ohm coaxial cable with a 22 AWG copper covered steel center conductor, an air dielectric polyethylene dielectric, a bare copper braid (mm. 95% coverage) outer conductor, and a PVC jacket. (Similar to RU 62 A/U type.)

Cable RG 62 A/U-lA (105 521 660) is a 93-ohm coaxial cable with a 22 AWU copper covered steel center conductor, an air dielectric polyethylene dielectric, a bare copper braid (mm. 95% coverage) outer conductor, and a PVC jacket. (Similar to RU 62 A/U type.) UL Listed Type CL2 per 1987 NEC.

Cable Ethernet (105 482 798) is a 50-ohm coaxial cable with a 0.0855 AWU solid tinned copper center conductor, a foamed polyethylene dielectric, a foil shield bonded to dielectric, and a PVC jacket. (Similar to Ethernet Type.) UL style 1478 DEC approved.

Cable Ethernet-1A (105 538 037) is a 50-ohm coaxial cable with a 0.0855 AWU solid tinned copper center conductor, a foamed polyethylene dielectric, a foil shield bonded to dielectric, a tinned copper braid (mm. 93% coverage), a foil shield, a tinned copper braid (mm. 90% coverage), and a yellow PVC jacket. (Similar to Ethernet Type.) UL style 1478. UL Listed Type CL2 per 1987 NEC.

Cable Ethernet-2 (105 482 806) is a 50-ohm plenum coaxial cable with a 0.0855 AWU solid tinned copper center conductor, a foamed FEP dielectric, a foil shield, a tinned copper braid (mm. 93% coverage), a foil shield, a tinned copper braid (mm 90% coverage), and an FEP jacket. (Similar to Ethernet.) UL Listed Type CL2P per 1987 NEC. DEC approved. Xerox specifications/IEEE 803.

Characteristic Impedance

As you can see from the preceding list, the characteristic impedance is the first electrical consideration mentioned. This is because this cable was first used for the needs of RF signal propagation, and the coax impedance was specified so that proper load matching could be made at the head end of the RF equipment. Standard impedances for coaxial cable are 50 ohms, 75 ohms, and 92 ohms. The diameter of the center conductor, the dialectic material, and the mechanical properties of the shield contribute and also help define the coaxial cable’s characteristic impedance. This impedance value is the value of the impedance at the maximum frequency for which the cable is designed.

For example, if you were using coaxial cable for video service, you would expect to see 75-ohm impedance exhibited at the maximum operating frequency of 900MHz. For cable, there’s always a trade-off between frequency headroom and attenuation per cable length. Typically for most coax cable, the attenuation is less than 1.5 decibels per hundred feet at 10MHz. At 100MHz, the attenuation is up around 5 decibels per hundred feet; consequently, as you increase your cable run, your attenuation goes up. When installing coax, there’s always a tradeoff in signal strength versus cable length versus frequency bandwidth.

A 10Mbps Ethernet segment can be up to 500 meters or approximately 1,640 feet in length using thicknet cable. For thinnet cable, the length can be up to 185 meters or 607 feet. Attenuation and frequency-based signal distortion limit segment lengths in Ethernet systems, whereas network lengths are limited by timing constraints that will be seen as bit-rate errors. Of course, at 100MB per second the maximum length is again reduced.

The network topology and distances that can be covered in a typical Ethernet network are discussed more fully in Chapter 13, “Ethernet: The Universal Standard.”

The two most common types of connectors are the BNC and the TNC. These are both named after their designers. The BNC connector has been around since World War II. It is a bayonet type and can be installed as a crimp type, a three-piece type, or a screw-on connector.

BNC connectors are similar to those used by cable companies to connect coax cabling to your set-top box. There is a single wire in the middle of the connector that carries the signal. To attach this connector to a cable, a crimping tool is used. A small portion of the cable is peeled back and inserted into the rear of the connector. The crimping tool then applies pressure to hold the cable to the connector. A three-piece type looks like a T-shaped connector, so you can connect cables to each side of the connector and use the third to screw into the receptacle on your computer. This last type was crucial in allowing a connection to a computer using coax cabling, yet letting the signal flow through the connector if the third part of the connector was removed from the computer.

The TNC connector usually is configured as a screw-on type and has been specifically developed for ease of installation with video-type cable. It is not used much in computer networking.

These are the advantages of coaxial cables:

Low susceptibility to EMI and RFI pickup

High-frequency bandwidth

Longer segment lengths than with twisted-pair cables

Can be matched with fiber-optic and twisted-pair cables

Lower signal distortion

Less cross-talk between cables

Better information security than with twisted-pair cable

These are the disadvantages of coaxial cable:

More difficult to install than twisted-pair cable

Heavier than twisted-pair or fiber-optic cables

Usually must be daisy-chained or home-run to workstations

Does not have the adaptability of twisted-pair cable

Is more expensive and takes more time to install

Fiber-Optic Cables

Fiber-optic technology is significantly different from copper and uses light transmitted through hair-thin fibers. Fiber-optic cable offers higher bandwidth and lower signal losses. It also allows higher data rates over longer distances.

These are the advantages of using fiber-optic cables:

Information carrying capacity—Fiber-optic bandwidth capacities are well in excess of what’s required by today’s network applications. The 62.5/125 Micrometer fiber recommended for building use has as its minimum bandwidth a capacity of over 160MHz per kilometer. The bandwidth at over 100 meters is well over 1.5Gbps. If the wavelength is different, the actual bandwidth can rise to 5Gbps. With the advent of Gigabit Ethernet, and 10Gigabit Ethernet, signaling techniques have greatly increased the bandwidth available on fiber-optic cables.

Low signal loss—Optical fibers offer low signal loss. This low signal loss permits longer transmission distances. In comparison with copper, the longest recommended copper horizontal link is 100 meters; when using fiber it is 2,000 meters or more. Again, this distance can increase as newer signaling techniques are used. In addition, long-distance runs of fiber-optic cables, joined with repeaters and other similar devices, can increase the distance achieved by fiber-optic cables dramatically.

The biggest drawback in using copper cable is that signal loss increases with signal frequency. Attenuation, or signal loss, is higher at 100MHz than at 10MHz. Consequently, high data rates increase power loss and decrease practical transmission distances. Loss does not significantly change with signal frequency in fiber-optic systems. Attenuation does change with frequency of the light transmitted through the fiber, but the data rate does not. So if you have both a 10MHz and a 100MHz signal traveling through the fiber, they are attenuated alike.

Electromagnetic Immunity

The basic transmission medium in a fiber-optic cable consists of either plastic or glass material. Both of these are considered to be insulators or dielectrics, so these materials are immune to electromagnetic interference. The transmitted signals consist mainly of modulated light signals that are tunneled through the fiber medium and do not escape. No signals emanate outside the cable, so it does not cause cross-talk, which is the main limitation in twisted-pair cable. It can be run in electrically noisy environments such as high-density computer-room installations, factory floors, and other electrically dense environments without concern because the cables are immune to outside noise sources. Yet, as you will learn in other chapters, fiber-optic cables do suffer some signal interference within the cable itself. For example, single-mode fiber, in which only one signal is transmitted, can cover a longer distance than multi-mode fiber, which injects multiple modes of light into a larger fiber-optic cable. In such a case, multi-mode fiber can suffer from degradation of the signal as different wavelengths of light interfere with other wavelengths.

Size and Weight

Fiber-optic cable weighs considerably less than copper cable. It is typically 22% to 50% lighter than comparable four-pair Category 5 cable. Less weight makes fiber-optic cable easier to install depending on its durability, which has improved over time. Typical weights for 1,000 feet are as listed here:

Two-fiber cable: 11 lbs.

12-fiber cable: 33 lbs.

4-pair Category 5 UTP: 25 lbs.

25-pair backbone UTP: 93 lbs.

10BASE-2 coax: 24 lbs.

Fiber-optic cable is smaller than copper cable. Typically, it’s about 15% less in volume than Category 5 twisted-pair cable.

Safety

As stated before, the glass and plastic that compose the transmission medium of fiber-optic cable are dielectrics, or insulators, and thus do not conduct electricity. Fiber-optic cable therefore does not present a spark hazard and can be used in explosive environments. It also does not attract lightning. Fiber-optic cable has jacket ratings that are comparable to the copper cable jackets and has the same flammability ratings that meet code requirements in buildings.

Security

It wasn’t until just recently that the capability to physically tap fiber-optic cables was developed. This requires extremely expensive equipment and a skilled operator. Typically, because fiber-optic cables do not emanate electromagnetic radiation, they are fairly secure against tapping. When compared to other methods of transmission, fiber-optic cable is the most secure medium for carrying sensitive data.

Fiber Construction and Operation

Fiber optics is a technology in which signals are converted from electrical into light signals. The signals then are sent or transmitted through a thin glass or plastic fiber and converted back to electrical signals. The fiber-optic cable consists of three concentric layers differing in optical parities. As shown in Figure 6.7, a fiber-optic cable consists of the following:

The core—The inner, light-carrying portion of the cable.

The cladding—The middle layer, which confines the light in the core.

Buffer—The outer layer, which serves as a shock absorber to protect the core and the cladding from damage.

Outer jacket—The covering, which protects the cable.

Figure 6.7. Components of a fiber-optic cable.

How Light Travels Through a Fiber-Optic Cable

Light transmission is not random. It is channeled into modes, which are possible paths for light rays to travel. There can be as few as one mode (single-mode fiber) or as many as several thousand modes in the design of the fiber (multi-mode fiber).

Although the number of modes is significant, it actually relates to determining the fiber’s bandwidth. More modes means lower bandwidth. The cause of this is dispersion. As a pulse of light travels through the fiber, it spreads out over distance. Although there are several reasons for such dispersion, the two principal concerns are modal dispersion and material dispersion. Different path lengths followed by light rays as they bounce down the fiber cause modal dispersion. Material dispersion is caused by different light wavelengths traveling at different speeds. To limit material dispersion, you limit the wavelengths of light transmitted. In other words, don’t use multi-mode fiber if you need to transmit a lot of data from one place to another. Use single-mode fiber for long distances. Use several cables of multi-mode fiber when you need to increase the bandwidth, without compromising on the actual bandwidth that can be achieved. Multi-mode fiber does allow for more than one data channel to travel through the same fiber-optic cable, but it should be limited as distance increases.

Fiber-optic cable can be modified in several ways to achieve different signal transmission characteristics. Modifications can be made to affect bandwidth and attenuation, and to facilitate coupling the light into and out of the fiber.

The stepped index multi-mode fiber has a large core with uniform optical properties. This fiber supports thousands of modes of operation and offers the highest dispersion and, hence, the lowest bandwidth (see Figure 6.8).

Figure 6.8. Light can reflect off the internal cladding as it travels through the fiber-optic cable.

The graded index multi-mode fiber has different optical properties in the core. This type reduces dispersion and increases bandwidth. The graded index makes light following longer paths travel slightly faster than light following shorter paths. The net result is that the light does not spread out as much. Nearly all multi-mode fiber that is used in networking and data communications has a graded index score.

The single-mode fiber has the highest bandwidth and the lowest loss of performance. The core of single-mode fiber is smaller than that of multi-mode fiber. The bandwidth that this fiber exhibits is much greater than the capacities of today’s electronics. This fiber can support speeds in excess of many gigabytes per second.

The most common fiber for networking is the 62/125-micron fiber (multi-mode). The two numbers designate the core diameter and the cladding diameter, respectively. In this case the core diameter is 62.5 microns and the cladding diameter is 125 microns. Other common sizes are 50/125-micron and 100/140-micron cable.

To summarize:

Graded index multi-mode fiber is the preferred fiber for horizontal cable and most backbone applications.

Single-mode fiber, by virtue of its immense bandwidth and long transmission capabilities, is the best choice for covering longer distances.

Attenuation in Fiber-Optic Cables

Similar to the degradation of an electrical signal in copper wires, attenuation in fiber-optic cables is a loss of power. During transmission, light pulses lose some of their energy, which shows up as a loss in signal strength. Attenuation is specified for fiber in decibels per kilometer. Attenuation ranges from under one decibel per kilometer for single-mode fibers and up to 2,000 decibels per kilometer for large-core plastic fibers.

Attenuation varies with the wavelength of light. There are three prime low-loss windows of wavelengths that are used today:

850 nanometers

1,300 nanometers

1,550 nanometers

The 850-nanometer wavelength is the most widely used because it was developed first, and optical devices such as LEDs (light-emitting diodes) operating at 850 nanometers are inexpensive and plentiful. The 1,300-nanometer wavelength offers low loss with a slight increase in cost for LEDs. The 1,550-nanometer wavelength is mainly used in long-distance telecommunications applications.

Terminations and Connections

For copper cabling, the main criterion is to provide an intimate, gas-tight joint between the connector contacts and the cable conductor. In reality, this seldom exists and there are two approaches to terminations in premise cabling: crimping the center conductor and using an insulation displacement contact.

When doing either type of connection, it is important to use the correct size and type of wire and also the correct tool. For example, if a contact is rated for a 24-gauge solid conductor, using a stranded or smaller wire, such as a 28-gauge wire, would result in a connection that could become loose or could fail.

Crimping

When a conductor is crimped, the contact is crushed around the center conductor. This cold-welds the contact to the center conductor. Many crimping tools are available today for almost any type of connection and cable type. You must use the proper tool for a successful crimp. Crimping tools are designed to provide the correct pressure by closing the dies a fixed amount. Using the wrong tool or die can result in either an under-crimp or an over-crimp. Under-crimping results in either a high resistance or a loose connection. Over-crimping can crush the wire or the connector so badly that it will be damaged and fail.

Insulation Displacement Contact

Insulation displacement contact uses a slotted beam. The wire is driven between the slotted beams. The beams are under spring tension and pierce the wire insulation and provide contact to the conductor inside. The contact can be either a flat form bar or a slotted barrel. These terminations are the most common in premise cabling applications.

Modular Jacks and Plugs

Modular jacks and plugs have been around a long time and are familiar to everyone as the connectors that plug into telephone handsets, bases, and wall outlets. The connectors and jacks that are used in premise wiring are different. Residential wiring and equipment use four-position plugs and jacks. The ones used for premise wiring are eight-position and terminate all four pairs of the cable. In typical nomenclature, the plug is the male end and the jack is the female end.

Modular jacks and plugs often are referred to as RJ connectors. The RJ comes from the term registered jack, and is specified in the USOC specification. USOC stands for Universal Service Order Code. This is a Bell Telephone specification that was developed for specific wiring connections, patterns, and applications within the telephone system.

An RJ-11 is a six-position connector and an RJ-45 is commonly referred to as an eight-position connector. Each of these basic jack styles can be wired for different RJ configurations. For example, the six-position jack can be wired as an RJ-11 C, which is a one-pair jack. It can also be wired as RJ-14 C, which is a two-pair, or an RJ-25 C, which is a three-point configuration. An eight-position jack can be wired for configurations such as RJ-61 C, four-pair, and RJ-48 C. The key eight-position jack can be wired for RG-45 RAS, RJ-46 S, and RJ-47 S. The fourth modular jack style is a modified version of the six-position jack, commonly called an MMJ. It was designed by Digital Equipment Corporation, along with the modified modular plug, to eliminate the possibility of connecting DEC data equipment to voice lines and vice versa. See Figure 6.9 for an example of these types of jacks.

Figure 6.9. Several types of modular jacks can be used for network cabling.

Modular Plug Pair Configurations

It is important that the pairing of wires in the modular plug match the pairs in the modular jack as well as the horizontal and backbone wiring. If they do not, the data being transmitted might be paired with incompatible signals. Modular cords wired to the T 568A color scheme on both ends are compatible with the 568B systems and vice versa. See Figure 6.10 for a breakdown of jack types and how they are wired.

Figure 6.10. A color scheme is used to match up wires at each end of a cable when joining the cable to a modular jack.

Common Outlet Configurations

Several outlet configurations were shown in Figure 6.9; however, it should be noted that the T 568A and T 568B have been adopted by the 568B.1 and 11801 standards. They are nearly identical except that pairs two and one are reversed. T 568A is the preferred scheme because it is compatible with one-or two-pair USOC schemes. Either configuration can be used for an ISDN service or high-speed data applications. Transmission categories 3, 5, 5E, and 6 are applicable only to this type of pair of grouping.

As shown in Figure 6.11, USOC wiring is available for one-, two-, three-, or four-pair systems. Pair one occupies the center conductors, pair two occupies the next two contacts out, and so forth. One advantage to this scheme is that a six-position plug configured with one, two, or three pairs can be inserted into an eight-position jack and still maintain pair continuity. The disadvantage is the poor transmission performance associated with this type of pair sequence. None of these pair schemes is cabling-standard compliant.

Figure 6.11. USOC wiring is available for one, two-, three-, or four-pair systems.

Various other standard schemes appear in Figure 6.12.

Figure 6.12. Different network cables require different wire connections to standard jacks.

There are a few guidelines you should follow when using the modular jacks and plugs:

For each category application, you must use plugs and jacks for that category.

You must be sure that you’re using the correct plug or jack for your conductor type.

You must follow termination procedures carefully. With the higher category cables in particular, proper installation procedures are essential to meet performance specifications.

ANSI TIA/EIA-568B standards specify that all pairs be terminated at the outlet.

The length of exposed wire (untwisted) shall not exceed 13mm for Category 5 or higher cables.

The length of exposed wire (untwisted) for Category 3 shall be within 75mm from the point of termination.

Patch Panels

Patch panels provide a means of rearranging circuits so that adding, subtracting, and changing workstations is made easier. Patch panels are where the circuits are connected and reconnected. Several patch panels use a feed-through connector set into which a cable can be plugged on both sides. Some configurations can have the horizontal cables going to the work areas plugged into one side of the panel.

Typically, feed-through patch panels are not suited for high-speed operation. Category 5 and higher panels feature IDC contacts on the back and modular jacks on the front. Modular jacks are usually 110-style or barrel style. This configuration offers a better electrical performance to reduce NEXT. Fiber-optic patch panels often offer a transition between different connectors. Transition among S C, FDDI, and S T are common.

There are also other ways of connecting and terminating cabling. Two of these use IDC connections. The first is the Type 66 cross connect block. This type of block has 50 rows of IDC contacts to accommodate the 50 conductors of 25-pair cable. Each row contains four contacts. Type 66 blocks represent an older style designed originally for voice circuits. Some of the newer designs meet Category 5 requirements. You should check to be sure that the block is rated for the category you’re installing, because older block designs have high cross-talk, which makes them unsuitable for high-data-rate designs.

A 110 cross connect consists of three parts: mounting legs, wiring block, and connecting block. The legs provide cable routing management and also hold the wire block. The wire block is composed of small plastic blocks that position the cable with index strips. Conductors are placed in the slot of the index strip. The strip usually has 50 slots to accommodate a 25-pair cable. It is marked every five pairs to help visually simplify the installation and reduce errors. This is also color-coded using the standard blue/orange/green/brown/slate color code. Wires used are punched into place with the 110-installation tool. This, however, does not terminate the conductors; it simply positions them. The device that does the termination is the connecting block. The IDC connecting block has contacts at both ends. One set of contacts terminates the contacts of the wiring block, and the other set on the outside is used for performing the cross connect.

This wiring system can accommodate as many as 300 pairs. Each horizontal strip can handle 25 pairs. A 100-pair cross connect requires four index strips. A 200-pair cable requires eight index strips, and so forth.

The system can be used as a prewired assembly for specific applications. One variation uses a 25-pair connector. In this situation, the block is prewired to the connector to allow a 25-pair cable from a hub or PBX to simply plug into the cross-connect.

There are pros and cons to using cross-connect blocks. They offer higher densities and require less space than patch panels, and also are less expensive. On the other hand, they are the least friendly for making moves, additions, and changes to the configuration. Skill is involved in removing and rearranging cables. When using patch panels, almost anyone can rearrange the system. In both situations security, ease of attachment, expense, and physical space are all considerations.

Terminating Fiber

What used to be a challenging task in the past, and is still an important task today, is terminating fiber-optic cable. There is a big difference between terminating electrical wiring and terminating a glass fiber that is only 62.5 microns in diameter. For one, electrical connections require a low resistance connection; the fiber requires a tight tolerance alignment. Misalignment in fiber connections will cause energy to be lost as light crosses a junction of the connector.

There are three functions of the termination process:

To prepare a smooth, flat, or rounded surface capable of accepting as much transmitted light as possible.

To provide a precise alignment of the clad fiber within the connector or splice to allow maximum coupling effectiveness.

To provide a secure physical attachment of the connector or spliced unit to the buffer cable.

Several varieties of common connectors are used in fiber optics. The following list does not include all connectors but does include those most commonly used for communication applications:

ST

SC

Biconic

SMA

Mini BNC

Data Link

Dual Fixed-Shroud (FDDI)

You can see examples of some of these in Figure 6.13.

Figure 6.13. Several kinds of connectors are used with fiber-optic cables.

Various techniques are used for installing fiber-optic connectors, but five tasks are common to any termination process:

1. The outer jacket, strength members, buffer tube, and coating must be removed.

2. The fiber must be threaded through the connector housing.

3. Fiber must be secured inside the connector.

4. The connector must be securely attached to the outside of the fiber.

5. The end of the clad fiber extending through the tip of the connector must be cut in preparation to accept the light signal.

The correct installation of fiber-optic connectors requires training, specific equipment, and consumable materials. Thankfully, manufacturers are constantly reducing the amount of time and training required to install the fiber-optic equipment they produce.

Epoxy terminations have long been used to ensure that the fiber is properly held in the connection ferrule. It does have its drawbacks, however. It is an extra step in the process, it’s potentially messy, and it requires curing the epoxy. Curing time can be shortened by utilizing an oven. This requires having another piece of equipment. There is also a possibility of spilling the epoxy on carpets and furniture in the finished building.

The epoxyless connectors require only a crimping tool and eliminate the need for epoxy. The simplest connectors use an internal insert that is forced snugly around the fiber during crimping. Inserts clamp and position the fiber while the crimp secures the cable strength members.

Another variation uses a short piece of fiber, which is factory-assembled and polished and inserted into the end of the connector ferrule. The inserted cable fiber butts up against this internal fiber, which is enclosed by an index-graded gel, and then the cable is crimped in place.

There also is a connector that uses a hot melt adhesive that is preloaded into the connector. This eliminates external mixing and loose components. Next, the connector is placed into an oven for a minute or so to soften the adhesive. The prepared fiber then is inserted in the assembly and is left to dry. Finally, the prepared connector is lightly polished.

The main benefit of epoxyless connectors is that they take less time and hence increase productivity. Remember that installing cables is perhaps the most expensive part of your network. Although servers and desktops computers are not cheap, the cost of labor and cabling to connect those computers to an enterprise network can be quite expensive.

Fiber-Optic Splicing

The preceding section discussed connecting fiber-optic cables to terminators. This section focuses on connecting segments of the fiber-optic cable itself. The splicing techniques are used only for extending the distance of the cable segment by adding another segment to it, or for repairing a cut or a damaged fiber cable. Now, splicing techniques and products have become user-friendly and often are considered a fast, low-loss alternative to traditional connector terminations.

There are two main types of fiber-optic splicing: fusion and mechanical.

Fusion Splicing

Fusion splicing is a process in which two sections of fiber are heated and, in effect, welded together. Fuse splices typically are good connections with attenuation losses as low as 0.1 db. Mechanical strength exhibited by this splice is often as strong as the original fiber.

The steps of a fusion splice include the following:

1. The ends of both fiber sections are prepared.

2. Both fibers are inserted into the splicing unit and precisely aligned.

3. Heat is applied at the interface of both fiber surfaces, and they are fused.

4. This place is tested for light loss.

Mechanical Splicing

Mechanical splicing is the process in which two sections of fiber are aligned and either glued or crimped in place within a permanent hood or shell. Mechanical splices are less labor-intensive than fuse splices. In some cases the splices can be installed in less than one minute per unit. Until recently, mechanical splices were relatively high-loss connections and were used for very limited applications. In the past several years, manufacturing processes have developed very low-loss mechanical splice units.

The low-loss connection attributes of today’s mechanical splice units created a new technique for terminating fiber. After cleaning and polishing each installed fiber, installers have the option of purchasing fiber jumpers or tails with connectors on one end and quickly splicing these tails to the installed fiber.

Fiber-Optic Patch Panels

Fiber-optic panels are termination units, which are designed to provide a secure, organized chamber for housing connectors and splice units. The typical termination unit consists of the following components:

Enclosed chamber—This can be a mountable wall or equipment rack.

Coupler panels—These hold the connector couplers.

The connector couplers.

Splice tray—Organizes and secures splice modules.

General Considerations for Fiber-Optic Cabling

For the insulation of optical fiber connecting hardware, the following recommendation should apply. Connectors should be protected from physical damage and moisture. Optical fiber cable connecting hardware should incorporate high-density termination to conserve space, provide for ease of optical fiber cable, and patch cord management on installation. Optical fiber cable connecting hardware should be designed to provide flexibility for mounting on walls, racks, or other types and distribution frames, and standard and mounting hardware.

You should insist that a minimum of 1 meter of two-fiber cable be accessible for termination purposes. Testing is recommended to ensure correct polarity and acceptable link performance. Clause 2 of 568B.1 provides recommended optical fiber link performance testing criteria.

Connections

Telecommunication outlet and connector boxes should be securely mounted at planned locations. The telecommunications outlet box or connector box should provide cable management means to assure a minimum bend radius of 25mm and should have slack storage capability.

The fiber types should be identified:

Multi-mode connectors or visible portions of it and adapters are to be identified with the color beige.

Single-mode connectors or visible portions of it and adapters are to be identified with the color blue.

The two positions in a duplex connector are referred to as position A and position B.

Small Form Factor Connectors (SFF)

Figure 6.14 shows a popular connector on the market today.

Figure 6.14. Small form factor (SFF) fiber-optic connectors are available today.

Some advantages of SFF connectors include compact size, modular compatibility with the eight-position modular copper interface, and adaptability to high-density eight-network electronics. Qualified SFF duplex and multi-fiber mode connector designs can be used in the main cross connect, intermediate cross connect, horizontal cross connect, and consolidation points and work areas. A TIA fiber-optic connected inter-mateability standard shall describe each SFF design. This design should satisfy the requirements specified in Annex A of the 568 B-B.3 standard.

Centralized optical fiber cabling provides users with flexibility in designing optical fiber cabling systems for centralized electronics typically in single tenant buildings.

Telecommunications Rooms

There have been some major changes from the EIA/TIA 568A to the EIA/TIA 568B and 569 standards transition. One of the major changes is that the telecommunications closet has evolved into the telecommunications room. These rooms are generally considered to be floor-serving facilities for horizontal cable distribution, and they also can be used for intermediate and main cross connects.

The telecommunications room is now defined for design and equipment according to ANSI/TIA/EIA 569A. Some of the specifications include specifications for wire management, relieving stress from tight bends, cable ties, staples, and so on. Horizontal cable terminations cannot be used to administer cabling system changes. Jumpers, patch cords, or equipment cords are required for reconfiguring cabling connections. There’s also a further restriction that application-specific electrical components cannot be installed as part of the horizontal cabling.

Open Office Cabling

Additional specifications for horizontal cabling in areas with movable furniture and partitions have been included in TIA/EIA 568B.1. Horizontal cabling methodologies are specified for open office environments by a means of multiuser telecommunications outlet assemblies (MUTO). It is preferable to use MUTOs only when the entire length of the work area cord is accessible to facilitate tracing and to prevent erroneous disconnection. Up to 22 meters or 71 feet of work area cable are allowed.

Consolidation Points

Consolidation points or transition point connectors are interfaces between the patch panels and MUTOs.

General Horizontal Cabling Subsystem Specifications

Note that the ISO/IEC 11801 allows 120 and unshielded twisted-pair horizontal cabling. Grounding must conform to applicable building codes as well as ANSI/EIA/TIA 607. You must have a minimum of two telecommunications outlets in the work area. The first outlet must have a 100-ohm twisted-pair category 5E, and the next outlet must also have one outlet with twisted-pair Category 5E. Two-fiber multi-mode optical fiber must also be installed.

Additional outlets can be provided. These outlets are in addition to and cannot be replaced by the minimum requirements of the standard. Bridged taps and splices are not allowed for copper-based horizontal cabling. Additional specific components cannot be installed as part of the horizontal cabling. When needed, they must be placed external to the telecommunications outlet or horizontal cross connects.

Finally, the proximity of horizontal cabling to sources of electromagnetic interference should be taken into account.

Documenting and the Administration of the Installation

When in place, the wiring plant is not a static, immobile structure. It is a dynamic and evolving part of your network. If the plant is not maintained with an up-to-date documented record, then any additions, changes, moves, and upgrading are out of the question.

The TIA/EIA-606 standard provides a guide to documenting a cabling system to make its administration efficient and effective. This provides the administrator with several benefits:

Allows better asset management

Increases network reliability and up time

Speeds and simplifies troubleshooting

Facilitates movement, additions, and changes

Allows for disaster recovery plans

Allows for capacity planning, upgrading, and acceptance of new emerging applications

Allows the generation of management reports

Documentation should include equipment and component labels, electronic or real records, drawings, work orders, and reports. Realistically, every piece of the physical plant should be labeled. This includes cables, termination hardware, cross connects, patch panels, closets, and anything else that will assist you in developing a meaningful overall view of your system.

Good labeling requires a unique coding scheme that makes sense to you. The label can include a location scheme, a component scheme, or a combination of both. You must realize that a label cannot have all the required information for a specific component, so it should contain enough information to uniquely define the component and point to a specific record. These pointers, sometimes referred to as linkages, will point to other records.

Records

The TIA/EIA 606 standard divides a record into four types of information:

Required information—The essential information about the component.

Required linkages—Links to other records.

Optional information—Additional information that makes the record understandable and comprehensive.

Optional linkages—Links to additional records that might be helpful to include. For example, if you are connecting a new PC in a new work area, you might want to include a linkage from the PC record to the record for the new workstation.

Drawings

Drawings are an essential part of your physical plant. They are necessary to locate components within a building. Drawings, especially “as built” drawings, will show the locations of conduits, pull boxes, and other components hidden from view behind walls and ceilings and under floors. These enable you, your installers, and network administrators to define and control space requirements, estimate cable densities, and keep track of equipment. In other words, you need to document your network. In Chapter 3, “Network Design Strategies,” you can learn about several applications and techniques that can be used to assist you in keeping track of your cable installations, as well as other components in your network.

Work Orders

Work orders should record all equipment and cabling moves, adds, and changes. This should form a history of the cabling system’s life or evolution. The records of the pertinent equipment involved should be updated every time the work order is performed.

Reports

A report is a group of records organized in a specific manner. This can be in the form of a database that can be selected to show a selected part of a record. For example, a report might show the number of hard drives on a specific server, or it can show the cables running to the device. You can include the hard drive identifiers or, in the case of the cables, the cable numbers.

Several types of cable management software are available that can be used to maintain records and generate reports. There are also specific programs aimed at managing a cabling system. These programs are compatible with structured cabling systems and conform to the TIA/EIA standards.