REFERENCE T3 R3 T2 R1 T1 R2 T4 R4 WHT GRN T2 R2 T3 R1 T1 R3 T4 R4 WHT

Transcription

1 P O L A R I Z A T I O N G L O S S A R Y & A L P H A N U M E R I C REFERENCE S Y S T E M A P P L I C A T I O N S C A B L I N G S T A N D A R D S T3 R3 T2 R1 T1 R2 T4 R4 WHT GRN GRN WHT ORG BLU WHT BLU ORG WHT BRN BRN INFORMATION AT YOUR FINGERTIPS In this section you will find information on many of the topics that you as an IT professional are likely to encounter. From application standards to cabling practices, from wiring sequences to a glossary of commonly used industry terms, we hope that you are able to quickly find the information you need. As always, we are available toll-free, from 8 a.m. to 5 p.m., Monday through Friday to answer any questions you may have. Our technical support engineers are available to provide you with design assistance and to make recommendations based on the latest standards updates. In the Cabling Practices section, highlights of standard cabling practices as they are recommended by industry trade organizations such as BICSI and ANSI/TIA/EIA are listed for your convenience. Where applicable, we have highlighted a recommended Molex Premise Networks solution T2 R2 T3 R1 T1 R3 T4 R4 WHT ORG ORG WHT BLU WHT GRN WHT BRN GRN BLU BRN In the Applications section, we have listed in a consolidated fashion the most common applications you are likely to encounter and the Molex Premise Networks products that we recommend for that particular application. In many cases there is more than one way to configure an application but we have chosen a brief list for maximum simplicity in your decision-making process. The Polarization and Wiring Sequences section visually depicts, as defined by industry standards, correct wiring sequences to meet EIA 568 A and B. The Glossary lists industry terms, commonly encountered acronyms and defines the nature of the various standards setting bodies. Page 78

2 REFERENCE In this section... Description Page Cabling Standards and Practices System Applications Polarization and Sequence Glossary Alphanumeric Index T3 R3 T2 nc nc R2 T4 R4 WHT GRN GRN WHT ORG BLU WHT BLU ORG WHT BRN BRN T2 R2 T3 nc nc R3 nc nc WHT ORG ORG WHT BLU WHT GRN WHT BRN GRN BLU BRN Page 79

3 Structured Cabling Practices A structured cabling system is an integrated network that handles all information traffic -- voice, data, video, even large, complex building management systems. The greatest expense of any network is the addition, removal or changing of devices especially after the initial installation. Therefore, the decision to implement a structured cabling system can pay for itself over and over again since it eliminates the need to re-cable every few years. Installing a structured cabling system reduces this cost by allowing new sections to be added to the network with minimum effort, expense and downtime. The following pages contain an outline of structured cabling standards with guidelines and examples for implementing a Molex Premise Networks structured cabling system. Level 5 Area II Page 80

4 Telecommunications Entrance Facility (As defined in the EIA/TIA-569 Commercial Building Standard for Telecommunications Pathways and Spaces) A building entrance facility provides the point at which the telecommunications service enters the building, including the entrance through the wall and the entrance room or space. Non-supportive equipment must have a separate entrance facility. The telecommunications entrance facility may contain the backbone pathways that link to other buildings in campus environments. Gross Floor Space Wall Length sq. m sq. ft. mm in , ,000 10, ,000 20,000 1, ,000 40,000 1, ,000 50,000 2, ,000 60,000 2, ,000 80,000 3, , ,000 3, Minimum Equipment and Termination Wall Space Page 81

5 Equipment Room The equipment room is a centralized space housing only that equipment which directly relates to the telecommunications system and its environmental support systems. It should meet the known space requirements of the specific equipment to be installed. However, if this is unknown, assume 0.07 sq. m (0.75 sq. ft.) for every 10 sq. m (100 sq. ft.) of work station space supported with a minimum size of 14 sq. m (150 sq. ft.). In special use buildings (such as hospitals, hotels and laboratories) the size must be based on number of work stations as follows: Area Work Stations sq. m sq. ft. Up to to to Equipment Room Floor Space for Special Use Buildings Telecommunications Closet The telecommunications closet accommodates the telecommunications function and support facilities only with a minimum of one closet per floor. Additional closets should be provided for each area up to 1000 sq. m (10,000 sq. ft.) when: (a) The floor area served exceeds 100 sq. m (10,000 sq. ft.) (b) The horizontal distance exceeds 90 m (300 ft.) Telecommunications Closet Serving Area Closet Size sq. m sq. ft. sq. m sq. ft x x x x 9 Recommended Closet Sizing (Based on one work station per 10 sq. m [100 square ft.]) Page 82

6 CABLING STANDARDS & PRACTICES Frame Layout Frames should be arranged in logical sections, grouping cross-connections of similar classification together. These sections may be as small as two panels and as large as multiple racks. The sections are then combined into racks so that patch cords or cross wires flow between areas thus minimizing the length of cross-connections. A Main Distribution Frame (MDF) has separate voice and data trunks and systems, but common local horizontal service should be laid out as illustrated. Since local service and trunks are generally connected to systems, patch cord length is minimized. In a large data MDF, the trunks should be placed on both sides with system patching located in the center to minimize patch cord length. Ring run panels in adjacent racks should be placed at the same level to form a wiring trough between racks. Ring run panels should be placed above and below a patching section and above a section of blocks. Open bay racks should be installed so that rear access is available for installation and maintenance. Racks should be located no closer than 1m (36") from any wall or equipment behind, in front or to one side. It is acceptable to butt one side of the rack against a wall or other structure. Two racks may be ganged together side-to-side. Racks should be bolted to the floor with anchors in concrete floors or with toggle bolts through raised computer flooring. Rack tops should be securely attached to the wall from behind using a Mod Strut Mounting Kit. Page 83

7 Block Based System Layout DESIGNING A KATT PDS WALL-MOUNTED SYSTEM LAYOUT When wall mounting KATT PDS Blocks, there are two basic configurations of blocks and rings: 600-pair column 1200-pair column A 600-pair column is generally sufficient for IDF and small MDF applications while a 1200-pair column is used for larger MDFs. The 600-pair column (see Figure 1) is approximately 1m (3 ft.) high and consists of two sections of three 100-pair blocks, each with a double set of rings in the center for demarcation and a single ring top and bottom. To design a wall-mounted system, specify the number of columns needed to provide sufficient pairs or channels. This configuration is generally applicable for terminating up to 4200 pairs (7 columns). Figure 1 Molex Premise Networks The 1200-pair column (see Figure 2) is approximately 2m (6 ft.) high and consists of four sections of three 100-pair blocks. Double rings are located on top of each, except for the third block down which has a single run. An additional single run is located below the last panel. To design a wall-mounted system, specify the number of columns needed to provide sufficient pairs or channels. This configuration is generally applicable to terminate up to 24,000 pairs (20 columns). Molex Premise Networks (a) Add additional ring runs at the top and bottom of the section, or (b) Do item "A" and add additional ring runs (2 or 3 total) between each three-block section. Figure 2 Page 84

8 Designing a KATT PDS Rack-Mounted System Layout SMALL SYSTEMS: < 2000 PAIR OR 10 PANELS TOTAL Small systems use the 3U panel since the integral 1U ring accepts 50 Category 3 cables or 40 Category 5 cables. Each panel supports 48 channels of Category 5 (4-pair). It is highly unlikely that 100% of the channels will be activated at Category 5. Since there are two additional ring runs (demarcation plus the bottom row), the 3U panel is appropriate for small Category 5 applications. A single 42U rack using 3U panels supports 13 panels plus three additional 1U ring runs. This is equivalent to 2600 pair (13 panels with 2 blocks of 100 pair each). Therefore, systems smaller than 2000 pairs only require a single rack or cabinet to house the KATT PDS panels and still have space for active equipment or future expansion. For these small systems, divide the rack into sections vertically, placing an additional ring run between the sections to form the double ring demarcation. Use an additional ring run below the last panel, so the minimum system (two sections) has two ring runs in addition to those provided on the panels. (see Figure 3) Figure 3 Molex Premise Networks Molex Premise Networks Page 85

9 MEDIUM SYSTEMS: 2000 TO 6000 PAIR OR UP TO 30 PANELS Medium systems require two to four racks or cabinets to provide mounting for the number of panels required. In this case, define each section as one or two racks and cross-connect horizontally between racks. Molex Premise Networks recommends using 4U panels with integral 2U rings because the rings on the center rack must support the cable coming from the blocks on that panel as well as the cables passing through that rack from and to adjacent ones. Lay out each rack to include two additional 2U ringrun panels one to create a double ring demarcation and one below the last panel in the rack. If the number of panels in each rack is not the same, lay out the racks so that the ring runs are horizontally-aligned using 4U blank panels to fill in unoccupied space. The 2U ring on the 4U panel can accept up to 100 Category 5 or 140 Category 3 cables, so this configuration will support up to 4 racks side-by-side. (see Figure 4) Figure 4 Molex Premise Networks Molex Premise Networks Molex Premise Networks Molex Premise Networks Molex Premise Networks Molex Premise Networks LARGE SYSTEMS: > 6000 PAIR In most cases, installations requiring more than four frames do not use full patch cord or Category 5 cross-connection. Instead, most channels are hard wired with cross-connection cable or jacketed IWC cable. Use patch cords for temporary changes and special applications (called "Patching by Exception"). Define a ratio that limits the percentage of channels that can be cross-connected with jacketed cable (either patch cords or IWC) by taking the number of frames required and dividing by four. For example, if eight frames are required, the percentage of channels that can be cross-connected is limited to 50%. If a higher percentage of cross-connection is required, add additional 2U ring runs between panels to increase the available cable management. Page 86

10 Backbone Cabling System Structure Backbone cabling systems are those cabling infrastructure segments that interconnect telecommunications closets (TC), equipment rooms and building entrance facilities. Backbones can either interconnect several buildings (interbuilding or "campus") or wholly contained within a building (intrabuilding or "premises"). Regardless of the type, the backbone infrastructure contains the following cabling system components: Backbone cables Terminating hardware at each end of cable segments like patch panels, cross connects, connectors and adapters Patch cords or jumper cables used to connect the cabling plant to the active equipment or for backbone-to-backbone cross-connection INTERBUILDING BACKBONES Interbuilding backbones (see Figure 5) interconnect multiple buildings in a campus-style facility. Examples are colleges/universities, industrial parks and military bases. The Main Cross-connect (MC) is located in a centrally-located building on the campus. The MC is commonly referred to as the "campus distribution frame" because it is the central connection point for the entire campus. Intermediate Cross-connects (IC) in each building interconnect the Horizontal Crossconnects (HC) with the MC or Building Distribution Frame (BDF). Very often, ICs are referred to as building distribution frames because they are the central connection point for HCs in that particular building. Figure 5 MC INTRABUILDING BACKBONES Intrabuilding backbones (see Figure 6) interconnect the IC with the main and horizontal cross-connects when all distribution frames are located in the same building. IC ARCHITECTURE Backbone cabling uses a hierarchical star wiring topology per ANSI/EIA/TIA-568A, where each HC in a telecommunications closet is cabled to a MC directly or through an IC. This physical topology offers users maximum flexibility since it supports virtually any logical topology like point to point, ring, bus, tree or a combination of these topologies. Figure 6 HC BACKBONE CABLES (MEDIA SELECTION) ANSI/EIA/TIA-568A recognizes the following cables for individual use or in combinations like UTP for voice transmission and fiber for data transmission: Four-pair 100 Ω unshielded twisted pair (UTP) cable Four-pair 150 Ω shielded twisted pair (STP-A) cable 62.5/125 µm optical fiber cable Single-mode optical fiber cable HC HC HC Today, optical fiber is the preferred media for backbone cabling and is recommended by Molex Premise Networks. Its low loss, extended distance and high bandwidth potential allow this media to support current and high-speed multimedia networks. Figure 7 shows maximum cabling distances for each cabling media with voice transmission over UTP 100Ω UTP (voice) and data transmission over fiber. Figure / 125 Fiber 800m MC 300m IC 500m HC 2000m MC 1500m IC 500m HC 3000m Singlemode Fiber MC 2500m IC 500m HC Page 87

11 When specifying optical fiber cables, make sure they comply with the attenuation and bandwidth specifications of ANSI/EIA/TIA-568A as follows: Table 1: ANSI/EIA/TIA-568A Fiber Specifications 62.5/125 FIBER SINGLE-MODE FIBER INDOOR/OUTDOOR 850 nm 3.75 db/km nm 1.50 db/km 1.0 / nm / 0.50 db/km 850 nm 160 MHz-km 1300 nm 500 MHz-km * *No bandwidth specifications exist for single-mode fibers. Band-width specifications are represented as pulse broadening as a product of source emitter quality and system length. For premise and campus backbones, it is virtually unlimited. CONNECTING HARDWARE The duplex SC, also called the 568SC, is the recommended connector and adapter for new installations per ANSI/EIA/TIA-568A. Its duplex design helps maintain polarity in a fiber optic backbone installation (see Figure 8). Beige colored connectors and adapters designate multimode while blue connectors and adapters designate single-mode connections. When expanding an existing backbone, continue to use the same connector and adapter type used in the cable plant. This is most likely an ST. Insertion loss for all connectors, field installed or on patch cords, should not exceed 0.75 db per mated pair. Any splices in the backbone, fusion or mechanical should be less than 0.30 db insertion loss. Need to update to include small form factor connectors Figure 8 TSB-72 "CENTRALIZED OPTICAL FIBER CABLING SYSTEM" GUIDELINES A "Centralized Optical Fiber Cabling System" is a fiber-based system architecture approved and introduced by EIA/TIA. It features guidelines for implementing multimode fiber to the desktop by centrally locating all active hubs/concentrators at the main cross-connect (MC). In practice, it passively combines horizontal and backbone cabling segments by crossconnecting, splicing or home-running cables from the Workstation Outlets (WO) to the MC. The maximum distance between the Hub/Concentrator at the MC and the WO is increased from 90 meters under 568A to 300 meters according to TSB-72 (see Figure 9). This architecture makes Optical Fiber To the Desk (OFTD ) a more affordable solution by eliminating connection points, providing better workgroup hub port utilization and allowing easier network cable man-age-ment from one centrally located cross-connect point. Figure 9 Wiring Closet Outlet Patch Cord High Fiber Count Cabling PC Centralized Hub Horizontal Cabling (low fiber count) Page 88

12 Level 5 Area II CABLING STANDARDS & PRACTICES HORIZONTAL CABLING SYSTEM STRUCTURE The horizontal section of the cabling infrastructure is the part of the system that extends from the Telecommunications Outlet (TO) to the Horizontal Cross-connect (HC) located in the Telecommunications Closet (TC). It usually runs horizontally along a single floor and is where adds, moves and changes are made. Cross-connections are made at the HC by patching or punchdown blocks in the TC. Because this is the most dynamic section of the cabling system and is in closest proximity to the end user, great care must be taken in its design. Large scale changes to this subsystem are very disruptive since ceiling or floor access in the user environment is typically necessary to make changes. When designing horizontal sections, keep in mind that facilities may be used for 10 to 15 years. The design must include significant extra capacity to provide for future as well as current applications. ARCHITECTURE The horizontal section basically is a star-wired cabling system made up of single-channel cables from the TO to the HC. This wiring style is also called homerun wiring. Cable lengths in the horizontal section are limited to 90 meters. Combined workstation line cords and HC patch cords make up another 10 meters. This is true for all media except the pull-through fiber architecture allowed under TIA TSB (Technical Service Bulletin) 72. Two new architectures are allowed for use with modular furniture the consolidation point and the Multi User Telecommunications Outlet (MUTO). Page 89

13 CABLING STANDARDS & PRACTICES Consolidation Point This architecture uses a termination point located in the ceiling, floor or on a power pole. Home runs are terminated to this point. An additional length of solid conductor cable is then run from the consolidation point (CP) to the wallplate. This allows the section of cable between the TC and the consolidation point to remain in place when furniture is moved. Only the last few meters are replaced when relocating the furniture pod. A consolidation point is not a splice, nor is it a multi-user telecommunications outlet (MUTO). The difference being that a CP requires an additional connection for each horizontal cable run. When correctly installed, the consolidation point will comply with ANSI/TIA/EIA Commercial Building Cabling Standards TSB-75. Cross-connections or active equipment are not permitted at a CP. Additionally, a CP must be installed in a secure, permanent location such as building columns and permanent walls not in modular furniture. Open Office Cabling: Multi User Telecommunication Outlet A Multi-user Telecommunications Outlet (MUTO) assembly is a grouping in one location of several telecommunications outlets. In this architecture, a ganged wallplate is centrally located in the furniture pod and longer than normal line cords are permitted for the final connection at the workstation. No additional wallplate at the workstation is allowed. MUTOs must be fully accessible and located on permanent, non-moveable walls or building columns. They cannot be mounted in ceilings, modular furniture or other obstructed areas. Molex Premise Networks MUTO is a rugged, high density outlet that provides voice and data service to up to twelve work areas as defined in TIA TSB75. Each MUTO is completely configurable for multimedia solutions using USO II modules including twisted pair copper, coax, video, and optical fiber connections. Length of Maximum Length of Horizontal Cables Work-Area Cables 90m 5m 5m 85m 7m 7m 80m 11m 11m 75m 15m 15m Maximum Length of Work Area Cables Molex Premise Networks MUTO provides voice data for up to 12 work areas. Page 90

14 Performance Standards CATEGORY 5E This is currently the most commonly installed cabling system. A system installed to the specifications of Category 5e performs to Category 5e and additional class D requirements of amendment 3 of ISO/IEC and Addendum 5 to ANSI/TIA/EIA-568-A. It will support the 1000 Base T Gigabit Ethernet Protocol. Requirements are specified to an upper frequency limit of 100MHz. CATEGORY 6 This is a standard currently under development. This standard provides significant future-proofing of a network as it is designed for the most demanding, bandwidth hungry applications. A system meeting this specification will support category 6 and Class E requirements currently under development. Category 6 is expected to be specified to an upper frequency limit of 250 MHz. CATEGORY 7 Although still very much still under development, this is a standard with transmission characteristics expected to be specified to 600 MHz. It will meet Category 7 and class F standards. As software applications continue to grow ever more elaborate and bandwidth intensive, the Category 7 standard will continue to be discussed with greater frequency. Attenuation NeXT Return Loss ELFeXT ACR Link Channel Link Channel Link Channel Link Channel Link Channel Category Category 5e Category 6* Category Standards *(Cat 6 Proposal) Obtaining Copies of Standards The Telecommunications Industry Association issues standards pertaining to communications cabling. The dominant standards are TIA 568A and TIA 569A on Commercial Building Cabling, Pathways and Spaces. Molex Premise Networks recommends a copy of each for anyone regularly involved with communications cabling. These standards can be purchased through: Global Engineering Documents 15 Inverness Way East Englewood, CO USA and Canada Internationally Recognized cable types are as follows: Four-pair, 100 Ω unshielded twisted pair Four-pair, 150 Ω shielded twisted pair (STP-A) Two-fiber, 62.5/125 µm, multimode, optical fiber Page 91

15 RECOMMENDATIONS: FUTURE PROOF YOUR NETWORK Use Category 5e or 6 UTP products as the horizontal media. There is a small price premium over Category 5, but it will ensure that whatever protocol your organization requires, the cabling will convey it. Molex Premise Networks structured cabling system recommends the use of ScTP in certain situations with high noise levels or where security is of utmost concern. ScTP is more expensive to purchase and installation labor costs are higher. However, there are also significant issues in grounding and in maintaining ScTP which make it unattractive. There are situations, however, where it does make sense to use ScTP in cabling system sections. For example, when: 10% of the network goes through a laboratory which contains sensitive measuring equipment that may be affected by energy radiated from UTP cable. 10% of the network passes through a high EMI environment like a manufacturing floor or welding shop, where large electric motors are present. If requirements exist for protecting the data from the environment or the environment from the data to a greater degree than afforded by, choose fiber media. Fiber optic cabling is the ultimate choice, but bandwidth is not the reason to choose fiber. PowerCat has more bandwidth capability than will be used during the expected lifetime of a cabling system, and IEEE is currently working on a 1 gigabit per second version of Ethernet. However, there are many good reasons for using fiber optic cabling systems, as listed below: Low attenuation translates into greater distances. This is important where manufacturing operations, schools and other large scale facilities are concerned. EMI is not a concern. This means low bit error rates, even in hostile environments. There is also no radiated energy from the cabling. This means that data is much more secure in fiber systems than in UTP or ScTP systems. The size and weight of fiber systems are lower, making them suitable for applications that are physically constrained. SELECTING TERMINATIONS: OVERALL ISSUES There are many things to consider when purchasing the wallplates, patch panels, cables and cable management products that make up a system. A few points to consider: Is the manufacturer reputable? Will they stand behind their products and services? Is the warranty strong? Beware of application restrictions hidden in the fine print. Are the products certified by UL for performance as well as safety? Are the products manufactured locally or in third-world countries? Are the products designed for durability? The cabling system may last 10 to 15 years. Are the products designed for ease of installation and maintenance? This will affect life cycle costs. Is there adequate labeling? Do the aesthetics reflect a well-designed quality installation? Page 92

16 Punch Down Blocks VS Patch Panels Termination in the TC can be accomplished by using either patch panels or punch down blocks. Each has advantages and disadvantages. Use patch panels for telephone and data in small-to-medium size installations (less than 2000 people serviced) and where secure closets are available. This allows the lowest maintenance cost because reconfiguration is done with RJ45. Larger installations tend to get unwieldy with patch panels because of the lower density and larger diameter patch cords. Finally, a locked closet or cabinet is important to avoid changes by unauthorized personnel. Use blocks in larger installations (more than 2000 people serviced). This will keep the management under control by allowing only trained personnel to make changes, keeping space requirements low and by cross-connecting with field-installed station wire. Cable Management One of the most important elements of a structured cabling system is proper cable management. A poorly maintained termination point, whether punchdown blocks or patch panels, can quickly lead to chaos at the Main or Intermediate Distribution Frame. Molex Premise Networks recommends laying out equipment and patch panels in a logical manner and using an adequate number of horizontal and side (vertical) ring runs. Proper cable management is as important at the rear as it is on the front of equipment racks. Molex Premise Networks rear cable management trays are a unique option that can be added to our HD Patch Panels. They ensure cables have the proper bend radius and strain relief to ensure a properly installed Category 5/5e/6 panel. The frame containing system connections should include all connections for both frame-mounted and remotely-located equipment. System connections include multichannel devices like CPUs, MUXes and servers. The frame containing drop-side connections should be laid out for maximum density and provide enough ring runs to prevent patch cord clutter in front of the patch panels. A drop-side connection is a cable run from the network to a single channel device like a terminal, modem or telephone. Horizontal ring runs are placed between framemounted system equipment and groups of punch-down blocks and patch panels. Side ring runs should be located just below horizontal ring runs on each side of the rack to keep patch cords to the side and away from the face of the patching field. Rear cable managers should be mounted on the back of patch panels. One rear cable manager should be used for each 8-jack group. Page 93

17 Cable Routing Whenever possible, primary cable routing paths should follow the logical structure of the building. Cable should follow hallways. When a wall must be breached, the cable should pass through pre-established and, preferably, sleeved openings. Cabling should enter and exit major run areas at 90 angles while adhering to applicable bend radius specifications. This minimizes potentially harmful field effects on the data signal from powered devices like fluorescent lighting and air handlers in the run area. In addition, cable should run parallel and perpendicular to corridors with a minimum of corridor crossovers. 90 Bend Cable run above a suspended ceiling should be supported by either cable trays or cable hangers. Additionally, cable should run above all metallic framing like floor joists or roof trusses. Typically, all cabling should be supported at spaces of four to five feet. Constant heating and cooling changes cause cable to expand and contract over time and may actually change the electrical characteristics of the conductors. All cable must be tension free at both ends and over the length of the run. Where a cable must bear some stress, use Kellum grips to spread the strain over a longer length of the cable. For a cabling network to effectively support higher data rates, it must be free of bridges, splices and taps from the outlet back to the wiring closet. Joining two cables together creates a reflection point in the communications channel. Depending on severity, this reflection causes degradation in signal quality and diminishes maximum overall cable run distance. Cables requiring service loops or additional length should be coiled from % of their recommended minimum bend radius. The coil is then wire tied and attached to a nearby support. Depending on where a cable is located, labels may be affixed at specified intervals over the entire length of the run. These labels should bear the cable identifications as a descriptive numbering scheme and should be at both ends of the cable run. This greatly increases the effectiveness of future troubleshooting and reduces network maintenance. All powered devices or power sources emit a certain amount of electromagnetic interference (EMI). To reduce or eliminate the field effect of this EMI on data traffic on a given cable channel, keep cable runs a minimum distance from these sources. Running cable through the center of a building minimizes external interference. Page 94

18 Fiber Cable Routing Optical cables are run in much the same manner as copper media with a few notable exceptions and additions. There are specifications for maximum tension and minimum bend radius for each fiber optic cable. Your cable manufacturer or supplier can provide these specs for your particular cable. It is important when routing the cable to make sure that all installation personnel are aware of these specifications. Maximum tension is rarely exceeded during hand installation of cable, but care must be taken if mechanical pulling devices, like winches, are used. Minimum bend radius specs may be easily violated if care is not taken when cable is routed through walls or around corners. Check all corners to be sure that all stored coils have sufficient diameter. Since there is no problem with electromagnetic interference in fiber optic transmission, cable routing near power sources is admissible and quite common. It is only necessary to be aware of high heat sources (steam pipes, etc.) as would be the case with any cable. Grounding Even if there is no immediate need for grounding in a cabling system, it is good practice to design a grounded cable infrastructure to support any devices or cable that may require it in the future. A #6 AWG ground-wire should be run from each Intermediate Distribution Frame (IDF) back to the Main Distribution Frame (MDF). This cable should connect to both frames, making sure the connection is positive and not likely to corrode. The MDF must be connected to earth ground the main building power ground is a good choice. Firestopping The final area of consideration in installation, and possibly the most important in terms of safety, is firestopping. This includes the use of adequate firestopping methods and materials to fill all openings created in firewalls. The majority of deaths and the rapid spread of fire in commercial buildings are often due to inadequate firestopping. Firestop Material Insulated Cables Steel Sleeve (optional) Packing Material Solid Concrete Floor Slab When firestopping through a concrete barrier, fill the opening with fire-resistant packing material to within the manufacturer's prescribed distance from the edge of the opening. Firestop material is then used to fill the remainder of the opening. TIA/EIA provides more information on this subject. Refer to your local fire codes for in-depth information on this important subject. Testing Buying Enhanced Category 5 cabling products does not guarantee an Enhanced Category 5 installation. Molex Premise Networks recommends that each installed channel be tested to TIA TSB67 Field Testing specifications with a commercially available Level 2 tester. Page 95

19 System Documentation Proper documentation is critical for the maintenance and maximum utilization of a structured cabling system. Documentation should be compiled as the system installation progresses and then delivered to the owner as the last element of the completed job. The complete documentation package should include: Marked-up blueprints showing outlet locations, associated numbering, DF locations and major run paths and risers Test results including a hard copy of all link characteristics for each cable run The cross-connect log in hard copy and software-based (if available) A synopsis of the numbering scheme A list of major components and their place in the network Any additional supporting documentation NUMBERING SCHEME Use an intelligent numbering scheme in the cable plant for cable identification. Base the numbering scheme on the cable plant itself and not any technological or physical aspect of the building it supports. The numbering scheme should break down into three areas: Horizontal cabling Backbone cabling System equipment HORIZONTAL CABLING Label each cable, user outlet and patch panel (or punch block) with a designation developed to the following formula at each end of the run: DF # - Group # - Channel # DF # is the terminating distribution frame to which the cable connects ("00" typically designates the MDF). Group # is sequential in each DF and typically represents a patch panel or punch block. Channel # is sequential within the group, as indicated by the logical or physical channel numbering on the components. Page 96

20 BACKBONE CABLING Label each riser cable, user outlet and patch panel (or punch block) with a designation developed to the following formula at each end of the run: ODF # - TDF # - Cable # - Channel # ODF # is the distribution frame from which the cable originates. TDF # is the distribution frame in which the cable terminates. Cable # is sequential between each DF. Channel # is sequential within the cable, as indicated by the logical or physical channel numbering on the components. It is not required where the cable represents a single physical communications channel. Use the following designation formula for each piece of network equipment that is represented directly or through some type of patch panel: Device # - Group # - Port # Device # is sequential for each device in a DF. Group # is sequential in each device and represents a chassis card or other logical group. Port # is sequential within the group, as indicated by the logical or physical channel numbering on the group component. Cross-Connect Log As the final procedure in any network installation, the certified installer should provide a set of cross-connect logs for each DF in the system. A cross connect log may be a simple hard copy or a software based log documenting the cross connections of rack- and wall-mounted termination components (i.e., patch cables or cross wires). It should follow a simple "from-to" format using the numbering scheme to identify interconnecting ports. Page 97