TBITS 6.9 - Profile for the Telecommunications Wiring System in Government Owned and Leased Buildings - Technical Specifications

1. INTRODUCTION

1.1 Objective and Scope

This Canadian Government Open Systems Interconnect (OSI) Profile specification establishes the minimum requirements for the telecommunications wiring system used to provision voice, data, and video applications in government-owned and leased buildings (hereinafter called 'federal buildings').

The use of standardized design and installation methodologies, together with standard products, will reduce government costs throughout the life-cycle of the telecommunications wiring infrastructure, and facilitate management, administration, maintenance, training, and future expansion. This specification describes the selection of specific design, installation, and management criteria, allowing for uniform specification documents. This will ensure multi-vendor supply for both initial installations, and expansion as required.

The CSA standards referenced by this profile define a telecommunications cabling infrastructure serving a building (i.e., not separate ones for each individual organization or Department located in a building). This infrastructure is designed to permit flexible re-arrangements of people and equipment within the building. In a building housing multiple organizations, this infrastructure is also constructed to facilitate changes in space allocated to the various organizations within it.

The infrastructure consists of telecommunications spaces (such as telecommunications closets and equipment rooms) and pathways (such as conduit) along with the cable and associated hardware within these spaces and pathways. The requirements for spaces and pathways are defined in CAN/CSA-T530, while the requirements for cable and associated hardware are defined in CAN/CSA-T529.

It cannot be emphasized enough that properly-designed telecommunications spaces and pathways are essential to provide the capability for a building to accommodate new demands imposed by changing technologies. These facilities, when properly implemented, permit new cabling and telecommunications equipment to be installed at lower cost and with less disruption to the occupants.

Telecommunications spaces are defined by T529 and T530 as being shared by the organizations requiring access to them. Consequently, all organizations in the building require access to the main telecommunications equipment room. This room contains, among other things, the main cross-connects for all telecommunications services (voice, data, image, etc.) serving all occupants. Similarly, the various organizations served by a telecommunications closet require access to it. While the standards specify that telecommunications equipment (such as LAN hubs) are typically housed in these shared telecommunications spaces, computer equipment (such as LAN servers) need not be contained within these shared spaces. In fact, this specification underscores the need to have such devices housed in a separate room, particularly in multi-tenant buildings.

Occupants' security requirements, particularly in multi- organization buildings, should be determined by referring to information technology security standards published by the RCMP and by completing a Threat and Risk Assessment.

1.2 Summary

Over the last few years the growth in the power of business software applications, coupled with the increasing demand for communication between individuals and businesses, has resulted in a dramatic increase in the importance of an organization's structured cabling infrastructure. The need to store, index, and retrieve increasingly large amounts of information, engage in workgroup computing, utilize electronic mail, store and retrieve voice and communicate rapidly across great distances continue to grow. Video conferencing and software applications that need to pass large amounts of data rapidly around the corporate network are coming on stream rapidly. In order to satisfy these needs, telecommunications and data communications have become a highly complex and extremely important element of our lives.

The continuous increase in the technical complexity of structured cabling and the never-ending push for cost reduction have resulted in a number of notable changes in the way that internal cabling systems are designed, installed, managed and controlled. It is now generally accepted that organizations must have a management strategy in place if they are to take advantage of the best technology available. This includes both the standardization of their cabling infrastructure and the adoption of agreed upon planning and management methodologies. Many of the organizations that failed to recognize this in the early days have learned the hard way. Those who were early adopters have reaped the benefits and have been able to further augment their management methods.

This Canadian Open Systems Application Criteria (COSAC) embraces contemporary technical standards as well as operational experience in the federal government and the private sector, thus providing for a reduction in costs and improved design, installation and management over the life-cycle of the system.

1.3 Definitions

The majority of these definitions are from the Canadian Standards Association standard CAN/CSA T529-95, "Telecommunications Cabling Systems in Commercial Buildings," and as updated by TSB75, and this Federal Government information technology standard.

adapter: a device that enables any or all of the following:

a) different sizes or types of plugs to mate with one another or to fit into a telecommunications outlet/connector;

b) the rearrangement of leads;

c) large cables with numerous wires fanning out to smaller groups of wires;

d) interconnection between cables.

administration: The method for labelling, identification, documentation and usage needed to implement moves, additions, and changes of the telecommunications infrastructure.

backbone: a facility (e.g. pathway, cable, or conductors) between telecommunications closets, or floor distribution terminals, the entrance facilities, and the equipment rooms within or between buildings.

bonding: a low impedance path obtained by permanently joining all non-current-carrying metal parts to assure electrical continuity and having the capacity to conduct safely any current likely to be imposed on it.

bridged tap: the multiple appearances of the same cable pair at several distribution points.

cable: an assembly of one or more conductors or optical fibres with an enveloping sheath, constructed so as to permit use of the conductors singly or in groups.

cable sheath: a covering over the conductor assembly that may include one or more metallic members, strength members, or jackets.

cable tray: a type of raceway

cabling: a combination of all cables, wire, cords, and connecting hardware.

campus: the building and grounds of a complex; e.g., a university, college, industrial park, government establishment, or military establishment.

channel: the end-to-end transmission path between two points at which application-specific equipment is connected.

commercial building: a building, or part thereof, that is intended for office use.

conduit: a raceway of circular cross-section of the type permitted under the electrical code and this Profile. Includes EMT (electrical-metallic tubing) conduit.

connecting hardware: a device providing mechanical cable terminations.

consolidation point: a location for interconnection between horizontal cables that extend from building pathways, and horizontal cables that extend into work area pathways.

cord, telecommunications: a cable using stranded conductors for flexibility, as in distribution cords or line cords.

cross-connect: a facility enabling the termination of cable elements and their interconnection, and/or cross-connection, primarily by means of a patch cord or jumper.

cross-connection: a connection scheme between cabling runs, subsystems, and equipment using patch cords or jumpers that attach to connecting hardware on each end.

customer premises: building(s) with grounds and belongings under the control of the customer.

demarcation point: a point where the operational control, or ownership changes.

device (as related to a workstation): an item such as a telephone, computer, graphic or video terminal.

device (as related to protection): a protector, a protector mount, a protector unit, or a protector module.

distribution frame: a structure with terminations for connecting the permanent cabling of a facility in such a manner that inter-connection or cross-connections may readily be made.

duct:

a) a single enclosed raceway for wires or cables. See also conduit, raceway;

b) a single enclosed raceway for wires or cables usually buried in soil or concrete;

c) an enclosure in which air is moved. Generally part of the HVAC system of a building.

entrance facility, telecommunications: an entrance to a building for both public and private network service cables (including antennae) including the entrance point at the building wall and continuing to the entrance room or space.

entrance point, telecommunications: the point of emergence of telecommunications conductors through an exterior wall, a concrete floor slab, or from a rigid metal conduit or intermediate metal conduit.

entrance room or space, telecommunications: a space in which the joining of inter- or intra-building telecommunications backbone facilities takes place. An entrance room may also serve as an equipment room.

equipment cable (cord): a cable or cable assembly used to connect telecommunications equipment to horizontal or backbone cabling.

equipment room, telecommunications: a centralized space for telecommunications equipment that serves the occupants of the building. An equipment room is considered distinct from a telecommunications closet because of the nature or complexity of the equipment.

furniture cluster: a contiguous group of work areas, typically including space division, work surfaces, storage, and seating.

ground: a connection to earth obtained by a grounding electrode.

horizontal cabling: the cabling between, and including, the telecommunications outlet/connector and the horizontal cross-connect.

horizontal cross-connect: a cross-connect of horizontal cabling to other cabling, e.g., horizontal, backbone, or equipment.

hybrid cable: an assembly of two or more cables (of the same, or different types or categories) covered by one overall sheath.

infrastructure, telecommunications: a collection of those telecommunications components, excluding equipment, that together provide the basic support for distribution of all information within a building or campus.

interconnection: a connection scheme that provides for the direct connection of a cable to another cable or to an equipment cable without a patch cord or jumper.

intermediate cross-connect: a cross-connect between first level and second level backbone cabling.

jumper: an assembly of twisted wires without connectors, used to join telecommunications circuits/links at the cross-connect.

keying: the mechanical feature of a connector system that guarantees correct orientation of a connection, or prevents the connection to a jack, or to an optical fibre adapter of the same type intended for another purpose.

ink: a transmission path between two points, not including terminal equipment, work area cables, and equipment cables.

main cross-connect: a cross-connect for first level backbone cables, entrance cables, and equipment cables.

media, telecommunications: wire, cable, or conductors use for telecommunications.

modular jack: a telecommunications female connector. A modular jack may be keyed or unkeyed, and may have six or eight contact positions, but not all positions need to be equipped with jack contacts.

modular plug: a telecommunications male connector for wire or cords. A modular plug may be keyed or unkeyed, and may have six or eight contact positions, but not all the positions need be equipped with contacts.

multimode optical fibre: an optical fibre that will allow many bound modes to propagate. The fibre may be graded-index or step-index fibre. See, also, optical fibre cable.

multi-user telecommunications outlet assembly: a grouping in one location of several telecommunications outlets/connectors.

open office: a floor space division provided by furniture, movable partitions, or other means, instead of building walls.

optical fibre cable: an assembly of one or more optical fibres.

optical fibre duplex adapter: a mechanical media termination device designed to align and join two duplex connectors.

optical fibre duplex connection: a mated assembly of two duplex connectors and a duplex adapter.

optical fibre duplex connector: a mechanical media termination device designed to transfer optical power between two pairs of optical fibres.

outlet box, telecommunications: a metallic or non-metallic box mounted within a wall, floor, or ceiling, used to hold telecommunications outlet/connectors, or transition devices.

outlet/connector, telecommunications: a connecting device in the work area, on which the horizontal cable terminates.

patch cord: a length of cable with connectors on one or both ends used to join telecommunications circuits/links at the cross-connect.

patch panel: a cross-connect system of mateable connectors that facilitates administration.

pathway: a facility for the placement of telecommunications cable.

pull strength: see pull tension.

pull tension: the pulling force that can be applied to a cable without affecting specified characteristics of the cable.

raceway: any channel designed for holding wires, cables, or busbars, and, unless otherwise qualified in the rules of the CE Code, the term includes conduit (rigid and flexible, metallic and non-metallic), electrical metallic and non-metallic tubing, underfloor raceways, cellular floors, surface raceways, wireways, cable trays, busways, and auxiliary gutters.

shield (screen): a metallic layer placed around a conductor or group of conductors. NB - the shield may be the metallic sheath of the cable or the metallic layer inside a non-metallic sheath.

single-mode optical fibre: an optical fibre that will allow only one mode to propagate; such fibre is typically a step-index fibre.

small buildings: typically having no more than two telecommunications closets or less than 100 work stations

space, telecommunications: an area used to house the installation and termination of telecommunications equipment and cable, e.g., telecommunications closets, work areas, and access holes/handholes.

splice: a joining of conductors, meant to be permanent, generally from different sheaths.

splice closure: a device used to protect a cable or wire splice.

star topology: a topology in which each telecommunications outlet/connector is directly cabled to the distribution device.

telecommunications: any transmission, emission, or reception of signs, signals, writings, images, and sounds, that is information of any nature by cable, radio, optical, or other electromagnetic systems.

telecommunications closet: an enclosed space for housing telecommunications equipment, cable terminations, and cross-connect cabling. The closet is the recognized location of the cross-connect between the backbone and horizontal facilities.

telecommunications grounding busbar: a common point of connection for the telecommunications system and bonding to ground; located in the telecommunications closet or equipment room.

terminal:

a) a point at which information may enter or leave a communications network; or

b) the input-output associated equipment; or

c) a device by means of which wires may be connected to each other.

topology: the physical or logical arrangement of a telecommunications system.

transfer impedance: the ratio of induced voltage of the conductors enclosed by the shield to the shield of the cable, connector, or cable assembly.

transition point: a location in the horizontal cabling where flat under-carpet cable connects to round cable.

work area (work station): a building space where the occupants interact with a workstation device(s).

work area cable (cord): a cable assembly connecting the telecommunications outlet/connector with the terminal equipment.

1.4 Acronyms and abbreviations

2. TELECOMMUNICATIONS CABLING SYSTEM STRUCTURE

The figure illustrates a model for the various functional elements that comprise a modern telecommunications wiring system.

3. SPECIFICATIONS

3.1 General

These specifications are mandatory for all new federal buildings, and retrofits of existing ones.

Additionally, managers are also encouraged to apply appropriate elements of this Profile to existing buildings, even when no retrofit has been planned. By doing so, significant reductions in operational costs can be realized.

3.1.1 This Profile generally adopts the Canadian Standards Association standard CAN/CSA-T529-95 (EIA/TIA-568A), "Telecommunic-ations Cabling Systems in Commercial Buildings." With the exception of the limitations and enhancements to that standard, described in the remainder of this section, T529-95 shall be utilized in the design and execution of cabling systems in government buildings as described above.

While this Profile primarily addresses building cabling, there are a number of other CSA standards that must be used in conjunction with CAN/CSA-T529-95 in the design and deployment of a structured telecommunications cabling system in order to achieve the most cost-effective operation. The most important are:

· CAN/CSA-T527-94 (EIA/TIA-607): "Grounding and Bonding for Telecommunications in Commercial Buildings";

· CAN/CSA-T528-93 (EIA/TIA-606): "Design Guidelines for Administration of Telecommunications Infrastructure in Commercial Buildings";

· CAN/CSA-T530-M90 (EIA/TIA-569): "Building Facilities, Design Guidelines for Telecommunications".

3.1.2 This Profile generally adopts these CSA standards.

Note: in addition to the limitations and enhancements introduced by this Canadian Open Systems Applications Criteria Profile, certain specification and design information of importance to federal government installations has been reproduced, unchanged, from the above standards throughout this COSAC. In such cases, original references are given.

Other related standards are listed in section 4, References.

3.2 Component Certification

3.2.1 This Profile requires that all hardware components procured as part of a structured cabling system shall be UL or ETL certified.

3.3 Horizontal Cabling

T529 specifies three transmission media types which may be used individually, or in combination, in the horizontal cabling infrastructure:

· four-pair 100 W unshielded twisted-pair (UTP) cable;

· two-pair 150 W shielded twisted-pair (STP-A) cable;

· two-fibre, 62.5/125 m m optical fibre cable.

3.3.1 This Profile endorses only the unshielded twisted-pair (UTP) and 62.5/125 m m optical fibre cable types meeting the specifications contained in T529.

Specifications for cross-connect jumper, patch cord, and work area cables/cords media are found in 3.6.

NB: 50 W coaxial cable, typically used in traditional 10Base2 and 10Base5 Ethernet networks, is a recognized (but grandfathered) media type in T529. New installations are specifically recommended against, and this cable type is expected to be removed from the next revision to the standard.

T529 recognizes various configurations in the horizontal cabling portion of the telecommunications wiring system.

3.3.2 Each work area shall be fitted with a minimum of two telecommunications outlets, each consisting of an eight-position modular jack (ISO8877, commonly known as RJ45) meeting the requirements of T529, Figure 10-1 (designation T568A), and with each terminating a run of four-pair Category 5 UTP cable no longer than 90 metres.

When considering quantities, it should be recognized that frequent changes in the use of floor space require that the floor be provisioned with the same density of cables, irrespective of immediate intended use of individual portions of the floor. Space used as a conference room today could well be used for individual workers tomorrow. Accordingly, sufficient cables should be provided to permit at least one work area per 10 square metres of useable floor space. This density may be increased (for example, providing cables to serve one work area for every 8 square metres) if justified by operational requirements. A lower density could be provided if justified by immediate and anticipated long-term requirements for the space.

T-529 generally requires that each telecommunications outlet be connected to the telecommunications closet by an unbroken run of 4-pair UTP. However, Annex G of T-529 recognizes that 25-pair cabling may be run from the telecommunications closet to a remote distribution terminal. Four-pair UTP is used from this point to the telecommunications equipment installed at the work area. This alternative approach was subsequently refined in TSB75. Consequently, either of the two approaches is considered valid.

T-529 states that 25-pair cables for horizontal distribution should be undertaken only as a special case considering system engineering guidelines. T-529 further requires that the location of this remote distribution terminal be readily accessible and visibly marked. Effectively, TSB75 requires that the distribution points be installed in physically easily accessible spaces, and prohibited in ceiling spaces.

A multi-user telecommunications outlet (refer to figure 3.3A) connector has the individual work-area jacks installed at the end of the horizontal cable from the telecommunications closet. Work area cables/cords (4 pair) are used to extend the service to the telecommunications equipment installed at the work area. There is no additional jack.

Figure 3.3A

A consolidation point (refer to figure 3.3B) contains no jacks. The horizontal cable is directly connected (i.e., not cross-connected) to the solid-core 4-pair cables serving the individual work areas using IDC-type hardware. A standard "RJ45" jack is installed at the work area.

Figure 3.3B

Because the work area cables/cords are made of stranded wire having poorer transmission characteristics than solid conductor wire, the maximum overall distance from the telecommunications closet to the work area must be reduced. Table 3.3 shows the maximum combined lengths of horizontal, work area cables, patch cords, and equipment cable allowable in such installations.

Table 3.3: Maximum Length of Horizontal and Related cables

3.3.3 Where the need is anticipated, a minimum of two strands of 62.5/125 m m optical fibre cable, terminated in SC connectors, should be installed in addition to the two or more runs of four-pair UTP cable.

3.4 Horizontal Conduit and Zones

3.4.1 A horizontal pathway, either EMT conduit or cable tray, must be provided. Zoned EMT conduit is the preferred horizontal pathway infrastructure, and should take precedence over all other types of pathway design.

As noted in Section 1 of this Profile, a properly-designed horizontal pathway system is essential to the long-term operation of a cabling system. A horizontal pathway must be provided except in leased accommodation, where the Crown has only a relatively short-term interest in the space. In such situations, however, a zoned wiring design shall still be used.

The design of the horizontal pathway must obviously take into account other related design aspects of the building (e.g. available physical space, false ceilings,) and while design must be to some extent on a per-building basis, it shall be governed in general by CAN/CSA-T530: "Building Facilities, Design Guidelines for Telecommunications.

Figure 3.4, below, illustrates the area served by a telecommunications closet divided into zones. Each zone is serviced individually, and directly, from the telecommunications closet.

Cables are run from the telecommunications closet through conduit to the zone being served. From the end of the conduit, they are run through ceiling space and down to the work area through utility poles, and where applicable, through raceways in screens. Wall-mounted outlets are typically served through conduit stubbed up into the ceiling space.

Whatever zone distribution system is used, all cables in the zone shall be of the same length. This length shall be sufficient to reach the most distant location within the zone. Slack, and unused cables should be coiled and left at the end of the zone conduit, or in the zone if another distribution system is used.

The Canadian Electrical Code requires metallic raceways to be bonded and grounded. The National Building Code requires adequate firestopping to be applied where the raceway penetrates the walls of the telecommunications closet.

Flexible metal conduit is not recommended by Clause 4.4.1.1 of T530, and is not covered by that standard. Nevertheless, there may be special applications requiring this type of conduit.

PVC is permitted by the National Building Code (NBC) but subject to stringent requirements, especially if installed in plenum spaces. It is important to note that provincial or municipal requirements may be more stringent. This Profile does not recommend the use of PVC unless it is installed in the slab.

Figure 3.4

Certain other specifications, because of their importance, are reproduced hereunder in whole, or with slight modifications, from T530.

3.4.2 No section of conduit shall be longer than 30 metres (100 feet) or contain more than two 90° bends between pull points or pull boxes. (T530, clause 4.4.2.1).

3.4.3 The inside radius of a bend in conduit shall be at least six (6) times the internal diameter. When the conduit size is greater than 50 mm (2 in), the inside radius shall be at least ten (10) times the internal diameter of the conduit. For optical fibre cable, the inside radius of a bend shall always be at least ten (10) times the internal diameter of the conduit (T530, clause 4.4.2.2).

3.4.4 EMT Conduit shall be reamed to eliminate sharp edges and terminated with an insulated bushing (T530, clause 4.4.2.5).

3.4.5 Pull boxes shall not be placed in a fixed false ceiling space unless it is above a suitably marked and hinged, panel (modification of T530, clause 4.4.2.6.2)

3.4.6 Pull boxes shall be placed in an EMT conduit run (T530, clause 4.4.2.6.3) where:

(a)

(i) the length is more than 30 metres (100 feet) ⁽¹⁾

(ii) there are more than two 90° bends; or,

(iii) if there is a reverse bend in the run.

3.5 Backbone Cabling

T529 specifies four transmission media types which may be used individually, or in combination, in the backbone cabling infrastructure:

· 100 W unshielded twisted-pair (UTP) cable;

· 150 W shielded twisted-pair (STP-A) cable;

· 62.5/125 m m optical fibre cable;

· single-mode optical fibre cable.

3.5.1 This Profile endorses only the unshielded twisted-pair (UTP) and 62.5/125 m m optical fibre cable types meeting the specifications contained in T529, and requires that they be installed as part of the backbone network infrastructure.

3.5.2 The quantity of UTP cable⁽²⁾ requirements. As a guide, three (3) pairs of category 3 UTP should be provided between the main cross-connect and each telecommunications closet for each work area planned to be served by the closet. For example, if one work area is planned per 10 square metres of floor space and the closet serves 500 square metres, 150 pairs (50 work areas x 3 pairs per work area) should be provided.

3.5.3 UTP cable sufficient to meet the estimated speed high speed data requirements should also be provided when required. As a guide, one run of 4-pair category 5 UTP should be provided between the main cross-connect and each telecommunications closet for every six (6) work areas planned to be served by the closet. Alternatively, appropriately-sized category 5 multi-pair UTP cable may be used. Refer, also to 3.7, note 2, for other related considerations.

3.5.4 A minimum of 12 strands (six pairs) of 62.5/125 m m optical fibre cable shall be provisioned between the main cross-connect and each telecommunications closet for high speed backbone requirements except in small buildings. All strands shall be terminated and tested.

3.5.5 Single-mode optical fibre cable is not covered by this specification, but may be added if required. If such cable is required, it shall meet the specifications described in section 12.3 of T529.

NB: 50 W coaxial cable, typically used in traditional 10Base2 and 10Base5 Ethernet networks, is a recognized (but grandfathered) media type in T529. New installations are specifically recommended against, and this cable type is expected to be removed from the next revision to the standard.

3.6 Connecting Hardware for UTP Cable

This section covers the minimum specifications for cross-connection hardware for the 100 W UTP cabling system found in the main cross-connect; intermediate cross-connect(s); horizontal cross-connects; horizontal cabling transition points; and telecommunications outlets/ connectors. Typical cross-connect facilities consist of cross-connect jumper cables or patch cords, and terminal blocks or patch panels.

Terminal blocks consist of hardware designed to use punched-down jumper wires (IDC method) to make the required cross-connection.

Patch panels are designed to use compatible plug-ended patch cords to make the required cross-connection. Typically, these patch panels use ISO 8877 ("RJ45") jacks.

An advantage of patch panels is that the cross-connections may be made by relatively untrained ("non-technical") personnel. In such cases, however, caution must be exercised to ensure that changes are made only by those authorized to make them (i.e. the staff responsible for the cable plant, and not "just any user"), and that change records are faithfully entered into the cable management system. As patch cords are available in fixed lengths, physical management of the patching system (e.g. rings, cable trays, etc.) is required to ensure a neat and manageable installation.

An advantage of terminal blocks is that only the desired number of pairs need to be cross-connected. For example in voice applications, because only one pair usually is required to operate a telephone set. The technician only needs to cross-connect one pair from the backbone cable to the horizontal cable run. By comparison, in a typical RJ45 patch panel, all four pairs are cross-connected. As a result, three pairs in the backbone are 'wasted'. However, if the terminal block approach is taken, personnel having more skills are required to make changes, especially if category 5 performance is to be achieved and maintained.

Generally, manufacturers of terminal block technology also manufacture plug-ended patch cords that may be plugged into their terminal blocks. These patch cords are typically available in different pair sizes. Terminal blocks, especially used in combination with compatible patch cords, typically offer better transmission performance than that achieved using the RJ45 patch panel approach, while obtaining virtually the same ease of use as RJ45 patch panels.

3.6.1 This Profile recommends that identical cross-connection hardware be used for both (nominal) voice and data fields such that patch cords are interchangeable.

Terminal blocks and patch panels may be mounted on a wall, on racks, or in cabinets. Wall mounting is typically less expensive than the other approaches, and additionally, saves on floor space that might be at a premium. The floor space is then available for other purposes, such as housing active equipment

T-529 permits data equipment to be either interconnected or cross-connected to the cabling system. Under both scenarios, the cabling is terminated on connecting hardware.

Equipment is considered to be cross-connected when the equipment ports are also terminated on dedicated connecting hardware. Cross-connections are made using either jumper wires or patch cords connected between the two fields of connecting hardware. Equipment is considered to be interconnected when the equipment ports are not terminated on dedicated hardware. Under this scenario, patch cords make the connection directly between the equipment port and the cable connecting hardware. These two scenarios are illustrated in T-529, and reproduced below as figure 3.6.

Figure 3.6

3.6.2 This profile standardizes on the cross-connection approach. Accordingly, a dedicated connecting field shall be provided for equipment ports.

As T-529 notes, interconnection is only possible when each individual equipment port has its own dedicated jack (typically ISO 8877, commonly known as "RJ45"). Equipment having other types of connectors (such as a 50-pin connector) has multiple ports appearing on each connector. Accordingly, these ports must be terminated on connecting hardware in order to provide access to each individual port. By providing a dedicated connecting field for data equipment, flexibility is provided to use equipment equipped with either type of connector.

3.6.3 Hardware used to terminate 100? UTP cables shall be of the insulation displacement contact (IDC) type.

Work area cabling is critical to a well-managed distribution system. While a separate specification of the work area cabling is stated generally as being outside the scope of T529, certain minimum requirements must be met, otherwise there may be detrimental effects on transmission performance.

3.6.4 Cables and connectors used in the work area shall meet or exceed the patch cord requirements specified in sections 10 and 12 of T529.

3.7 Telecommunications Closets and Equipment Rooms

T529 and T530 describe a Telecommunications Closet as an enclosed space for housing telecommunications equipment, cable terminations, and cross-connect cabling. The closet is the recognized location of the cross-connect between the backbone and horizontal facilities. The equipment room is defined as a centralized space for telecommunications equipment that serves the occupants of the building. Equipment rooms are normally considered distinct from telecommunications closets because of the nature or complexity of the equipment housed therein. However, any or all of the functions of a telecommunications closet may alternatively be provided by an equipment room; this is often likely to be the case in a small building.

Telecommunications spaces are also defined by T529 and T530 as being shared by the organizations requiring access to them. Consequently, all organizations in the building require access to the main telecommunications equipment room. This room contains, among other things, the main cross-connects for all telecommunications services (voice, data, image, etc.) serving all occupants. Similarly, the various organizations served by a telecommunications closet require access to it. While the standards specify that telecommunications equipment (such as LAN hubs) are typically housed in these shared telecommunications spaces, computer equipment (such as LAN servers) need not be contained within these shared spaces. In fact, this specification underscores the need to have such devices to be housed in a separate room, particularly in multi-tenant buildings.

3.7.1 This Profile requires that the Telecommunications Closet and Equipment Room space in all new buildings, and existing buildings when they are retrofitted, be dedicated to the telecommunications function and associated support requirements. Their design shall be governed by CAN/CSA-T530: "Building Facilities, Design Guidelines for Telecommunications".

3.7.2 Design shall also be governed by CAN/CSA-T527-94, "Grounding and Bonding for Telecommunications in Commercial Buildings".

T530 highlights the requirement for important environmental support systems. Typical examples of such support requirements are security and environmental controls, including heating, ventilation, and air-conditioning (HVAC).

3.8 Cabling Directly Between Telecommunications Closets

T529 notes that if requirements for certain bus or ring configurations are anticipated, direct connections between telecommunications closets (vertical or horizontal) are allowed. Such cabling is in addition to the basic star topology described in Section 2.

3.8.1 Cabling direct between telecommunications closets shall meet the same specification requirements for media, connecting hardware, etc., as are found elsewhere in this Profile.

3.9 Intra- and Inter-Building Distances

Maximum distances of individual components (e.g., telecommunications closet to work area, patch cords), as well as overall (e.g., between a main cross-connect, intermediate cross-connect and telecommunications closet) are application-dependent. The maximum distances specified in figure 3.9 are based on voice and data transmission for UTP and data transmission for optical fibre.

Figure 3.9

NOTES:

[1] The distance between the entrance point and the main cross-connect shall be included in the total distance calculations when regulatory policies within the jurisdiction, which relate to the location of the demarcation point, deem it appropriate. The demarcation point between the service providers and the customer premises cabling may be part of the entrance facilities. The location of this point for regulated carriers is determined by the regulatory agency (normally the CRTC). The carrier should be contacted to determine the location policies in effect. The length and type of media (including gauge size for copper) shall be recorded and should be made available to the service provider upon request.

[2] While T529 allows a distance of up to 800, 500, and 300 metres for spans B, C, and D, respectively, for voice and low-speed data applications, the use of category 3, 4, or 5 multipair UTP backbone cabling for applications whose spectral bandwidth is in the range of 5 Mhz to 16 Mhz, 10 Mhz to 20 Mhz, and 20 Mhz to 100 Mhz respectively, must be limited to a total of 90 metres. The 90 metre distance assumes a total of 10 metres is needed for equipment cables (patch cords, jumpers).

3.9.1 While this standard recognizes that the capabilities of single-mode optical fibre allow for backbone link distances of up to 60 km, such extended networks are generally considered to extend outside the scope of this standard which is concerned with building and campus environments.

3.9.2 Specific applications may exist, or become available in the future, that do not operate properly over the distances specified. For example, to support local exchange carrier and other provider services, it may be necessary to insert repeaters or regenerators (which are outside the scope of the standard) along the backbone cabling.

3.9.3 When the HC to IC distance is less than maximum, the IC to MC distance for optical fibre can be increased accordingly but the total distance from the HC to the MC shall not exceed the maximum of 2000 m for 62.5 µm optical fibre cable or 3000 m for single-mode optical fibre cable.

3.9.4 When the HC to IC distance is less than maximum, the IC to MC distance for UTP cabling can be increased accordingly, but the total distance from the HC to the MC shall not exceed the maximum of 800 m (but still 90 m in the case of category 5 high-speed backbones).

To minimize cabling distances, it is often advantageous to locate the main cross-connect near the centre of a campus. Installations that exceed these distance limits may be divided into areas, each of which can be supported by backbone cabling within the scope of the standard. Interconnections between the individual areas, which are outside the scope of the standard may be accomplished by employing equipment and technologies normally used for wide-area networking applications.

3.10 Pathway/Space Separation from Electro-magnetic and Radio Frequency Energy Sources

Horizontal and backbone cabling must be routed to avoid sources of electro-magnetic interference (EMI) and radio frequency interference (RFI) such as electrical power wiring, motor-generators, induction heaters, arc welders, elevator equipment, fluorescent lighting ballasts, transformers, and other similar sources. T530 specifies pathway/space separation from electrical facilities that generate high levels of EMI and/or RFI. Accordingly:

3.10.1 Cable pathways and telecommunications closet and equipment room spaces shall be designed such that they are separated from nearby EMI and RFI sources per the requirements of CAN/CSA-T530 (ref. clauses 4.8.5, 7.2.1.1, 8.1.3, and 8.2.1.5).

3.11 UTP Installation, Testing and Certification

T529 warns that UTP cabling systems may not be classified as category 3, 4, and 5 compliant unless all the cabling system components are installed to satisfy the requirements of section 10.6 of the standard. Channel performance depends on cable characteristics (including cross-connect jumpers and patch cords), the total number of connections, and the care with which they are installed and maintained. T529 notes that field testing of UTP cabling runs at frequencies up to 100 Mhz poses numerous technical difficulties and that test methods and apparatus for transmission testing are under study.

3.11.1 To ensure reliable operation over the usable life of the cabling system, it shall meet all relevant requirements of T529, and installation shall be carried out in accordance with Section 10.6 - "UTP Installation Practices".

3.11.2 All cables shall be tested for wire map and length compliance. A random sample of 5 - 10% of category 5 cables, chosen by the client Department, not the installation company, shall also be tested with a Level II tester for conformance with channel performance (attenuation, NeXT) as described in TSB67. The sample selection should include the longest and shortest runs, with the proviso that the shortest run selected is greater than 15 metres in length (ref. TSB67, section 7.8). NeXT shall be tested from both ends (ref. TSB67, section 7.4). Where work area equipment cables/cords are not provided by the installation company (typically the case), the installation company may use one representative sample for all tests.

3.11.3 Departmental RFPs shall require that the complete cabling system proposed by Bidders is covered by a certification program provided jointly by the Manufacturer of the equipment and the bidding company. Such certification shall provide an assurance that the system will support applications for which it is designed (e.g. 100BaseT, 155 Mbps ATM), and additional applications introduced in the future (e.g. 622 Mbps ATM) by recognized standards fora that use the CAN/CSA-T529 or TSB67 (or subsequent relevant TSBs) component and link/channel specifications during the lifetime of the certified system.

3.11.4 Departmental RFPs shall require that Bidders be certified system vendors of the Manufacturer's components being bid, and use only technicians fully trained and qualified on installation and testing of the components to be installed.

Annex B, "Installation Testing and Certification," contains more detailed information on these issues.

3.12 Optical Fibre Installation, Testing, and Certification

3.12.1 To ensure reliable operation over the usable life of the cabling system, design, installation, and testing shall meet the requirements described in Section 12 and Annex H of T529.

For several years the defacto standard for optical fibre connectors was the ST type (formally known as BFOC/2.5). However the industry is in the process of migrating to the SC type (formally known as 568SC), which is specified in the current T529 standard. The standard notes that networks with an installed base of BFOC/2.5 optical fibre connectors and adapters may remain with the same connector types for both existing and future additions to their network(s). The standard also recommends that networks be migrated to the newer, standardized SC connector and adapter for new additions to the installed base, or to retrofit the existing network.

3.12.2 It is mandatory that for new network installations, departments and agencies specify the type 568SC connectors and adapters specified in T529, Annex F.

3.13 Cabling System Administration

Administration of the telecommunications infrastructure includes documentation (e.g., labels, records, drawings, reports, work orders) of cables, termination hardware, patching and cross-connect facilities, conduit, other cable pathways, telecommunications closets, and other telecommunications spaces. The collection, and timely updating, of infrastructure information is critical to the success of the administrative process. Annex C of this Profile provides a more detailed examination of the critical success factors involved in cabling system administration approaches.

3.13.1 It is strongly recommended that any cable management system product being acquired generally conform to the mandatory, optional, and advisory specifications contained in CAN/CSA-T528-93: "Design Guidelines for Administration of Telecommunications Infrastructure in Commercial Buildings".

3.13.2 For all but small buildings, this Profile requires that the cable management system be computer-based.

3.13.3 It is a mandatory requirement that RFPs for structured cabling systems require that data entry be carried out by the installation company as the cabling system is installed, and be handed over, ready for operational usage, at the time of system acceptance.

3.13.4 Departments shall ensure that procedures are developed and made operational that guarantee that cable records are kept current, reflecting the impact of on-going moves, adds, and changes.

3.14 Local and National Codes

Telecommunications wiring is regulated by the applicable building code. Departments are warned that there is no one building code applicable homogeneously across Canada. Next, while office buildings are generally constructed of non-combustible material, buildings under a certain size or a particular occupancy are permitted to be constructed of combustible materials such as wood.

As part of the implementation of the Canada-United States Free-Trade agreement, the electrical codes in Canada and the US have been harmonized with regard to fire resistance requirements and marking of communications cables. The current CSA standard, CAN/CSA-214, incorporates this alignment. A cable substitution hierarchy has been introduced into the Canadian Electrical Code, while the US National Electrical Code has a new cable marking to indicate cables that have passed the Canadian FT4 vertical tray fire test.

Approved cable markings, and their equivalent fire test ratings resulting from the 1994 Canadian Electrical Code and the 1993 US National Electrical Code are:

Table 3.14

The flame tests in table 3.14 have been listed in descending order of severity. The following cable substitutions may be used as shown below:

a) Cables marked MPP, CMP, MPR, CMR, MPG, CMG, MP, CM, CMX, CMH, FT6, and FT4 may be substituted for FT1;

b) Cables marked MPP, CMP, MPR, CMR, MPG, CMG, and FT6 may be substituted for FT4;

c) Cables marked MPP and CMP can be substituted for FT6.

It is federal government policy that buildings which are owned by the federal government comply with the NBCC and NFCC, and therefore need not comply with potentially more stringent Provincial or municipal codes and bylaws. The NBCC requires that FT4 rated cable be installed in plenum spaces. By contrast, two provinces - Ontario and British Columbia - have more stringent requirements, generally requiring the use of FT6-rated cable. But, there are exemptions within provincial jurisdictions: for example the City of Vancouver stipulates the use of FT4. The telecommunications wiring project in a new or retrofitted building is not only an expensive proposition, but one fraught with human safety and liability issues. If the necessary research has not been carried out, and approvals have not been obtained, project managers can find themselves in disastrous liability situations. Therefore, this Profile requires that:

3.14.1 Proposed building plans: site, architectural, structural, mechanical, and electrical (this includes cabling systems) must be submitted to the appropriate Human Resources Development Canada/Labour Canada (HDRC) regional office for review and acceptance prior to the commencement of the project.

3.14.2 Local municipal building code requirements, and the extent of their applicability. if any, must be determined, and subsequent permits obtained (by the constructor/contractor), prior to the project.

4 REFERENCES

4.1 Standards of Intrinsic Relevance to this Profile

CAN/CSA-T527-94, "Grounding and Bonding for Telecommunications in Commercial Buildings" (Canadian Standards Association)

CSA home page: http://www.csa.com 

CAN/CSA-T528-93, "Design Guidelines for Administration of Telecommunications Infrastructure in Commercial Buildings"

CAN/CSA T529-95, "Telecommunications Cabling Systems in Commercial Buildings"

CAN/CSA-T530-M90⁽³⁾, "Building Facilities, Design Guidelines for Telecommunications"

TSB67, "Transmission Performance Specifications for Field Testing of Unshielded Twisted Pair Cabling Systems" (Electronic Industries Association)

EIA/TIA documents including TSBs available from hhtp://global.ihs.com/

TSB75, "Additional Horizontal Cabling Practices for Open Offices"

"Generic Base-Building Telecommunications Cabling Infrastructure Normative References & Generic Technical Criteria" (Public Works and Government Services Canada, March 1992).

4.2 Canadian Electrical Code, Part II Standards

CAN/CSA-C22.2 No. 182.4, "Plugs, Receptacles and Connectors for Communications Systems"

CAN/CSA-C22.2 No. 214, "Communications Cables"

CAN/CSA-C22.2 No. 225, "Telecommunications Equipment"

CAN/CSA-C22.2 No. 226, "Protectors in Telecommunications Networks"

CAN/CSA-C22.2 No. 233, "Cords and Cord Sets for Communications Systems"

4.3 ISDN Standards

CAN/CSA-T531, "Acoustic-to-Digital and Digital-to-Acoustic Transmission Requirements for ISDN Terminals"

CAN/CSA-T538, "Carrier-to-Customer Installation - DS1 Metallic Interface"

CAN/CSA-T540, "Integrated Services Digital Network (ISDN) Primary Rate - Customer Installation Metallic Interfaces (Layer 1) Specification"

CAN/CSA-T542, "Integrated Services Digital Network (ISDN) - Data-Link Layer Signaling Specification for Application at the User-Network Interface"

CAN/CSA-T543, "Integrated Services Digital Network (ISDN) - Minimal Set of Bearer Services for the Primary Rate Interface"

CAN/CSA-T544, ""Integrated Services Digital Network (ISDN) - Minimal Set of Bearer Services for the Basic Rate Interface"

4.4 Treasury Board Information Technology Standards (TBITS)

TBITS-6.1, Canadian Open Systems Application Criteria

TBITS-6.2, COSAC LAN Profile

4.5 ANSI/IEEE Standards

IEEE P1143/D7, "Guide on Shielding Practice for Low Voltage Cables" (Draft)

ISO/IEC 8802-3: 1996 [ANSI/IEEE Std 802.3, 1996 Edition], Information technology--Local and metropolitan area networks--Part 3: Carrier sense multiple access with collision detection (CSMA/CD) access method and physical layer specifications.

IEEE Std. 802.3u-1995, Supplement to ISO/IEC 8802-3:1993, Local and Metropolitan Area Networks: Media Access Control (MAC) Parameters, Physical Layer, Medium Attachment Units and Repeater for 100 Mbps Operation, Type 100BaseT

ISO/IEC 8802-5: [ANSI/IEEE Std 802.5-1995], Information Technology--Local and metropolitan area networks--Part 5: Token ring access method and physical layer specifications.

ISO/IEC 10038: 1993 [ANSI/IEEE Std 802.1D, 1993 Edition], Information technology--Telecommunications and information exchange between systems--Local area networks--Media access control (MAC) bridges.

IEEE Std. 802.12-1995, Demand Priority Access Method Physical Layer and Repeater Specifications for 100 Mbps Operation

4.6 ATM Forum Standards

af-phy-0015.000: "ATM Physical Medium Dependent Interface Specification for 155 Mbps over Twisted pair Cable"

af-phy-0040.000: "Physical Interface Specification for 25.6 Mbps over Twisted Pair"

af-phy-0046-000: "622.06 Mbps Physical Layer"

af-phy-0047-000: "155.52 Mbps Physical Layer Specification for Category 3 UTP"

http://www.atmforum.com/atmforum/approved-specs.html for all of the above and related ATM Forum standards

4.7 Bibliography

"ATM to the Desktop - The Skeptics are only Half Right" - Ron Jeffries, Telecommunications magazine, September 1995

http://www.telecoms-mag.com/tcs.html

BICSI (Building Industry Consulting Service International)

http://www.bicsi.org

"Cabling Issues for Enterprise ATM Networks" - Sandia National Laboratories, Business Communications Review, April 1995

http//www.bcr.com/

"Computer and Communications Standards" lexicon

http://www.cmpcmm.com/cc/standards.html

"Cost Effective Fibre Optics in the Subscriber Loop" - G.Bradley, Sasktel, 1995

Ethernet Switching, ATM over UTP, and other white papers - Anixter Inc. (1996)

http://www.anixter.com/homepage.html

Fast Ethernet Consortium (University of New Hampshire).

http://www.iol.unh.edu/consortiums/fe/fast_ethernet_consortium.html

FDDI - Finally, There's Money to be Made" - Martin A. Palka, Business Communications Review, 1994

http://www.bcr.com/

"Fibre-to-the-Terminal: Video on Demand through Networked MultiMedia" - Sasktel; 8th annual National Fiber Optic Engineers Conference (Washington) proceedings Volume 1, April 1992

Gigabit Ethernet Alliance

http://www.gigabit-ethernet.org/

"How ATM Lost the Race to the Desktop" - Dr. John McQuillan, Business Communications Review, March 1996

http://www.bcr.com/

"How to Effectively Manage your Structured Cabling Infrastructure" white paper - Anixter Inc. (1996)

http://www.anixter.com/scsmgmt.html

IEEE (Institute of Electrical and Electronics Engineers) standards

http://stdsbs.ieee.org

"Implementing the Canada-United States Free-Trade Agreement: Harmonizing Fire Requirements for Cables" - Stanley Kaufman, AT&T Network Cable Systems, Norcross, Georgia, 30071.

"Open Office Cabling System - Application Guidelines" - Jerry Solomon, Amp. Inc.

http://www.amp.com/

"Parallel Paths Emerge for Fast Ethernet and ATM" - Al Chiang, Telecommunications magazine, March 1996

http://www.telecoms-mag.com/tcs.html

"Residential Broadband: Once More Around the Block" - Dr. John McQuillan, Business Communications Review, December 1995

http://www.bcr.com/

"Structured Cabling Systems for Intelligent Buildings" (LucentTechnologies)

http://www.lucent.com/systimax/papers/scs_iba.html

"The Politics of Cable Management" - Walter A. Chiquoine, Business Communications Review, February 1995

http://www.bcr.com/

"25 Mbps ATM - Can it Fly?" - Ron Jeffries, Business Communications Review, August 1994

http://www.bcr.com/

"Understanding the Electrical Impulses of UTP Cabling Sheds Light on Cable Performance" -tutorial no. 93, Lee Chae, LAN Magazine.

http://www.lanmag.com

"Why Fiber? Why Now?" - Amp. Inc., December 1995

http://www.amp.com/

"Wire and Cable, notice no.40" - Canadian Standards Association, June 24 1994

"Wobbly D.C. to Megablinking Light in 3.3 Billion Seconds" - G. Bradley, Research & Development, Sasktel, 1995

ANNEX A: HIGH BANDWIDTH APPLICATIONS

A.1 Introduction

The rapid advancements in technologies and applications for the enterprise network are straining the physical wiring plant of many organizations, particularly those still using category 3, or lower quality, cabling).

A number of competing, high-speed technologies have recently been, or are in the process of being introduced to meet to the bandwidth demands of today and the future.

As an alternative to copper, optical fibre has long been touted as the ultimate telecommunications highway with its virtually unlimited capacity, immunity to EMI and RFI, enhanced security, and long transmission distances without repeaters. Beginning with inter-office trunking by the telcos in 1980, long-haul trunking (from 1981 onwards), metropolitan area networks in and around cities (starting in 1984), to cable television supertrunks (from 1988), optical fibre deployment has been rapid and pervasive. From its telecommunications carrier and broadcasting origins it has also found its way into the local area network.

A.2 Ethernet

100BASE-T is one of several solutions that has been proposed for providing 100 Mbps bandwidth over copper UTP (and optical fibre in one variant) to the large installed base of 10 Mbps Ethernet users.

The IEEE 802.3u 100BaseT standard was approved in June 1995. "Fast Ethernet" and "100 Mbps CSMA/CD" are other names that describe the specification, which retains the same CSMA/CD protocol of the earlier 10 Mbps Ethernet. 100BaseT encompasses four approaches, three for unshielded twisted pair copper, and one for optical fibre:

· 100Base-TX provides 100 Mbps operation over 2 pairs of Category 5 cable;

· 100Base-T4 calls for 4-pair operation over categories 3, 4, and 5 UTP;

· 100Base-T2 is currently being studied by a task force (IEEE 802.3y). The intention is to support 100 Mbps full duplex operations over two-pair category 3, 4, and 5 UTP;

· 100Base-FX defines operation over standard two-strand multi-mode 62.5/125 >m m optical fibre. It will support runs up to 2000 metres (full duplex) and 412 metres (half duplex).

Another approach, 100VG-AnyLAN (the "VG" stands for voice-grade), is being studied by the IEEE802.12 committee. It is designed to support voice and video running at 100 Mbps over category 3 UTP.

Additionally, two further standards for high-speed Ethernet networking over UTP have been proposed to the IEEE: "Gigabit Ethernet" and "1G-AnyLAN." Two specifications, now under study by the 802.3z and the 100VG-AnyLAN (802.12) working groups, call for a nominal 1 gigabit/second, initially over single- and multi-mode optical fibre, and subsequently over category 5 UTP. Ethernet-oriented engineers argue that gigabit Ethernet is already cheaper to build than 622 Mbps ATM and may be cheaper than 155 Mbps ATM by 1997. Finalization of a standard is estimated to be sometime in 1998.

A.3 Token Ring

Ethernet and FDDI networks have been rejuvenated through a performance boost from new technologies. Token Ring is no different: while the basic 4 and 16 Mbps speeds of Token Ring have not increased (unlike Ethernet), switching increases the speed and efficiency of networks by providing dedicated transmission paths between users and resources. Aggregate bandwidth and productivity increase as a result of multiple simultaneous connections through the switching fabric. Switching also offers a simpler and more efficient means of connecting multiple rings than the traditional routing or source route bridging.

A.4 FDDI

The Fibre Distributed Data Interface (FDDI) was the first 100 Mbps network and the first one designed from the ground up to use optical fibre. Its span is significantly greater than that of typical horizontal cabling infrastructures: up 1000 nodes over a 100 kilometre span; the distance between attached stations can be as much as 2 km for multimode fibre and 40 km for single-mode fibre. Its use has been primarily as a high-speed backbone for connecting Ethernet and Token Ring LANs, and to support high-end compute-intensive workstations.

While the FDDI market has grown slowly for the past several years, primarily for use in backbones, or to support high end workstations, some analysts and vendors believe that it is on the threshold of greater growth. These projections are based on the expectation of integrated, interoperable, standards-based chipsets that enable FDDI over unshielded twisted pair, as distinct from the optical fibre which FDDI was originally based on.

The ANSI X3T9.5 FDDI (PMD) specification allows 100 Mbps FDDI to be used over category 5 UTP.

A.5 ATM

The utilization of signaling and encoding techniques from the Synchronous Optical Network SONET standard, used in high speed metropolitan area optical fibre networks, have found new uses in ATM networking using unshielded twisted pair cable plant within buildings. The ATM Forum has completed a standard specification (ATM25), a 25.6 Mbps ATM interface specifically designed to function over category 3 unshielded twisted pair, and products have been shipping since 1995.

Work is being carried out on a more complex and demanding specification which will support 51.84 Mbps over category 3 UTP.

Using category 5 UTP, ATM is already supported at 155 Mbps. The ATM Forum has started work on a new scheme designed to support 155 Mpbs over category 3 UTP; estimates are that specifications will be complete and products will become available in 1997.

Finally, the ATM Forum also has study groups examining several other standards-based specifications, including one for 622 Mbps over category 5 UTP.

A.6 Fiber Channel

An existing standard, "Fibre Channel", which has been mainly used to date in the closely-coupled computers and high-speed storage connect world (e.g., between computers and computers, ands between computers and RAID storage devices), is designed to run over optical fibre and copper at speeds up to 1.062 Gbps. Fibre Channel has now been specified as the physical layer for both Gigabit Ethernet (802.3z) and 100VGAnyLAN (802.12)

A.7 Conclusion

Since the structured cabling approach and 10BaseT first became commonly available in the early nineties, here has been endless discussion and debate about the relative costs and attributes of copper and optical fibre. Recently, the question has been whether the availability of ATM over unshielded twisted pair copper should influence the choice of copper or fibre to the desktop. Pricing issues have been much discussed, and comprehensive comparisons of copper versus fibre have been advanced. These comparisons include the cost per metre of the media, costs for termination costs, labour, network interface cards, jumper cables, and testing. Some real-world cases have shown that in certain circumstances, optical fibre to the desktop has the edge over unshielded twisted pair copper, particularly where an integrated ATM networking approach is being taken.

From the foregoing, it should be clear that, faced with preparing a cabling design for a new building or a renovation project, departments should carry out a full functional and technical requirements analysis before developing an RFP. The major thing to remember is to get the sizing of pathways and spaces right - unfortunately, undersized telecommunications rooms and conduit systems have been all too common, resulting in expensive incremental add-ons later. 

ANNEX B: INSTALLATION TESTING & CERTIFICATION

This annex is an edited version of a Microtest Inc. white paper, and published by Anixter Inc., both companies to whom grateful appreciation is extended.

B.1 TESTING & CERTIFICATION

B.1.1 Introduction

It is often overlooked that certification of the various categories (3, 4, and 5) of cable has been of the various components in isolation from one another. Until 1994, there was little or no effort to understand the effects of cable, connectors, punch-down blocks, wall plates, patch cords, or installation practices together.

T529 includes an informative Annex E which describes the worst-case performance characteristics of a UTP channel in the horizontal portion of the system. It defines the attenuation and NeXT (Near End Cross Talk) requirements under defined conditions. In the absence of a normative specification while the standard was near completion, Annex E was used as the best information against which to test an installed link. However, it is important to realize that there is no pass/fail standard. From 1993 through the end of 1994, most field test equipment (FTE) verified category 5 performance against the Annex E definitions. In some cases, Category 5 links were failed by FTE when in fact the link had been carefully installed and "known" Category 5 components were used. Because of concerns about discrepancies between results obtained with FTE and with laboratory network analyzers, a task group under TIA was established to study these issues.

After two years of hindsight, research and study, it became clear that while some early units did have limited accuracy, other factors made a bigger contribution to errors in measurements and differences in comparison tests. Some of these factors include:

· unbalanced components, especially unbalanced modular 8 (RJ45) connections. These are tested with simple 100W resistors, yet perform quite differently, and with much higher crosstalk, when attached to a real cable. This is the root cause of the "short-link" problem;

· poorly conducted comparison tests. For example, the FTE would attach to the circuit with an extra patch cord while the network analyzer used a different connection. The patch cord changes the NeXT of the circuit, so it would be impossible to obtain the same result in the FTE and the network analyzer;

· the lack of a standard for balun performance in areas that really matter, such as output-signal balance or common-mode impedance. Variations in these unspecified parameters wreak havoc when attempting correlation studies;

· cabling tests that had little or no control over the stability of the pair and twist orientation between consecutive tests. Of course, as the twists and pair spacing varied, so did the NeXT measurements;

· the use of logarithmic-spaced sweeps on NeXT tests. Some tests used linear-spaced sweeps, while others used logarithmic-spaced sweeps. The log sweeps had gaps at the high end over 2.5 Mhz wide between consecutive measurements. It was easy to miss a NeXT peak;

· tests on FTE performed at levels well below their intended dynamic range. Poor results down at the noise floor were then erroneously extrapolated to the normal operating range where the real accuracy was actually much better;

· finally, there was no agreed upon method to specify NeXT accuracy. There is no standard unit with which to measure NeXT, unlike weight (kilogram), electrical potential (volt) or length (metre).

Clearly, all these unresolved issues made a certain amount of disagreement between instruments inevitable. Fortunately, the task force studied all of these issues and many others in detail and resolved, documented, and standardized ways to handle them. The result was Telecommunications System Bulletin TSB 67.

TSB67 was ratified in September 1995. It is of great interest to cable installers, test-equipment manufacturers, telecommunications administrators and LAN administrators because it provides, for the first time, detailed requirements on how to test and certify installed unshielded twisted-pair (UTP) cabling.

Briefly, TSB67 includes a link model, a description of which tests must be performed to certify the link (wiremap, length, attenuation, and NeXT), and specifications for how each test is to be performed. In addition, TSB67 contains detailed procedures for verifying the accuracy of field test equipment against both a theoretical model and a laboratory network analyzer. Finally, TSB67 specifies performance criteria for FTE.

The primary field test parameters are:

• Wire Map;
• Length;
• Attenuation; and,
• Near-end Crosstalk (NeXT) loss.

The wire map test is intended to verify pair-to-pin termination at each end and check for installation connectivity errors. For each of the eight conductors in the cable, the wire map test indicates:

• continuity to the remote end;
• shorts between any two or more connectors;
• crossed pairs;
• reversed pairs;
• split pairs;
• any other miswiring.

Correct connectivity of telecommunications outlets/connectors is defined in T529, section 10.4.5 and is illustrated thus:

Figure B.1.1A

A reversed pair occurs when the polarity of one wire pair is reversed at one end of the link (also called a Tip/Ring reversal). Refer to figure B.1.1B(a) for an illustration of a reversed pair.

A transposed pair occurs when the two conductors in a wire pair are connected to the position for a different pair at the remote location. Refer to figure B.1.1B(b) for an illustration of transposed pairs. NB: Pair transpositions are sometimes called crossed pairs.

Split pairs occur when pin-to-pin continuity is maintained but physical pairs are separated. Refer to figure B.1.1B(c) for an illustration of split pairs.

Figure B.1.1B

The physical length of the basic link/channel is defined as the sum of the physical lengths of the cables between the two end points. Physical length of the basic link/channel is determined by measuring the length(s) of the cable(s), determined from the length markings on the cable(s), when present, or estimated from the electrical length measurement. The maximum physical length of the basic link shall be 94 metres, including test equipment cords, while the maximum physical length of the channel shall be 100 metres, including equipment cords and patch cords.

Attenuation is a measurement of signal loss in the basic link or channel, and is dependent on length. Tests are carried out to determine the worst-case attenuation of all pairs within a link relative to the maximum attenuation allowed.

Near-end crosstalk (NEXT) loss is a measure of signal coupling from one pair to another with a UTP cabling link, and is length independent. It is important that all pair combinations be measured.

B.1.2 The Channel and Basic Link Models

Before the issue of accuracy and how it is addressed in TSB67 is studied, it is necessary to understand TSB67's two link definitions: the channel link and the basic link. Figure 1 illustrates the TSB67 channel definition. Cables A and E are user patch cords, almost always terminated in RJ45 connectors. Note that these are not and cannot be test equipment cords; they must be the user's actual patch cords.

Figure B.1.2A: TSB67 Channel Definition

As Figure B.1.2A shows, the mated connection at the ends of these cords is not included in the channel definition. It is considered a part of the field tester. This connection is typically an 8-position RJ45. This means any measurements taken on the channel must be made through the mated connection and do not include the connection's characteristics. The mated RJ45 connection has significant NeXT, which becomes a source of error in NeXT measurements that will significantly differentiate the accuracy of channel and basic link measurements.

The channel was defined because it is important to know the performance of the sum of all the components between the hub and the workstation so that the quality of communications from end to end can be predicted. This information is essential to circuit designers and important to end users. However, cable installers are typically not responsible for installing patch cords, as workstations are not usually present when cabling is installed and tested.

Figure B1.2B: TSB67 Basic Link Definition

In Figure B.1.2B, the basic link represents a minimal link and has only one connection at each end. The channel has two. In addition, the basic link can only be 90 metres in length, while the channel can extend to 100 metres. For these reasons, both attenuation and NeXT will be higher on the channel than on the basic link.

B.1.3 Accuracy Levels

Recognizing that the basic link and the channel link represent two different models, the authors of TSB67 chose to define two distinct accuracy levels: Level II (high accuracy) and Level I (lower accuracy).

The reason for the two levels is that when a channel is being tested, the tester is almost always forced to measure through (but not include) the NeXT effects of a RJ45 interface directly on the FTE. The unpredictable crosstalk in this connection sets a limit on the achievable accuracy of the measurement. In contrast, when testing a basic link, field test equipment manufacturers can choose to use a very low crosstalk interface directly on the FTE. This reality is reflected in the TSB67 description of two accuracy levels for field test equipment. Level I reflects the performance boundaries imposed by the reality of having to test through an RJ45 connection. Level II sets a much higher accuracy requirement, attainable only if a different, low-crosstalk interface is used. The uncertainty caused by the higher crosstalk RJ45 interface can be avoided, and thus a much higher level of accuracy can be achieved.

An earlier TSB (TSB-40A, subsequently incorporated into T529) specified the worst-case NeXT performance of any RJ45 connection to be 40 dB at 100 Mhz. So, while some connections might achieve 42 or 43 dB, 40 dB performance is all that can be guaranteed. This unpredictable, high level of inherent crosstalk limits any tester's ability to make measurements at a Level II accuracy level when testing a channel through an RJ45 interface to the FTE.

B.1.4 Accuracy Measurements

The TIA task force that established TSB67 determined six key performance parameters that affected the accuracy of field testers (see Table 1.4). The largest error term for field testers is Residual NeXT. This consists of the sum of the tester's internal NeXT plus the NeXT of the interface to the tested link. Remember, this mated connection is not included in the link definition.

When testing a channel, this residual NeXT will include the NeXT of a mated RJ45 connection. Even if the residual NeXT on the field tester's internal circuits is zero, its overall residual NeXT will be limited to 40 dB by the mated RJ45 connection specified by TSB-40A. Thus, the Level II accuracy performance requirement of 55 dB cannot be met when testing a channel. This is, in fact, the reason Level I and Level II were created. When testing a basic link, the field tester can make use of an interface with much lower inherent crosstalk, thus making the Level II residual NeXT requirement of 55 dB achievable.

TSB67 specifies that for an instrument to meet Level I or II accuracy, it must meet all six of the requisite performance parameters. The crosstalk and balance characteristics of an RJ45 connector immediately limit any tester using it to no better than Level I accuracy. It is important to note that even Level II accuracy tools are reduced to Level I accuracy when forced to test through an RJ45 interface because of the uncertainty created by the RJ45 connector. This uncertainty has an unpredictable magnitude and phase, so it cannot be compensated for, or subtracted by, hardware or software.

Table B.1.4: TSB67 Accuracy Performance Parameters

TSB67 also requires that field testers agree with Annex B of TSB67, which states that agreement with network analyzers must be demonstrated. The reason for this is that different field testers may employ different methods to make measurements. Some of these methods, such as time-domain measurements, may have additional error sources unaccounted for in the theoretical error model.

The Level I performance limitations of an RJ45 connection hold true even when time-domain measurement techniques are used to attempt to 'time-gate' away the high crosstalk. The outgoing pulses used to make time-domain measurements have a duration of several nanoseconds, which equates to several feet. This means a "NeXT dead-zone" is created where the tester cannot read the crosstalk on the first few feet. As Figure 1 shows, this measurement technique does not comply with TSB67 because the test must begin directly behind the first RJ45 connections, not two or three feet down the cable.

Many products will claim to meet Level II accuracy, but the fine print often shows such products barely meet the minimum requirements, especially with residual NeXT. Proof of compliance with Annex B (network analyzer agreement) of TSB67 is conspicuously absent in most cases. Departments are advised to have their cabling RFPs require bidders to state full compliance with TSB67 specifications.

B.1.5 Length Accuracy Issues

Annex D of TSB67 provides information on how to increase the accuracy of length measurements, or, at the least, how to minimize the inaccuracies of such measurements. Since most FTE measures length using Time Domain Reflectometry (TDR), the accuracy of these products depends upon the Nominal Velocity of Propagation (NVP) setting of the cable being tested. NVP varies up to 5 percent from cable to cable and even from pair to pair. TDR is an excellent method to measure length but requires the cable's precise NVP. 

ANNEX C: MANAGEMENT & ADMINISTRATION

C.1 INTRODUCTION

Modern buildings require an effective telecommunications infrastructure to support the wide variety of services that rely on the electronic transport of information. In contrast with older cabling networks that carried voice and data separately, the modern structured system carries voice and data, and increasingly, any, or all of, video, alarm, environmental control, security and audio. Historically, administration of the cabling system, if carried out at all, took second place to computer system management and network management.

Administration of the modern telecommunications infrastructure includes documentation (labels, records, drawings, reports, and work orders) of cables, termination hardware, patching and cross-connect facilities, conduits and other cable pathways, telecommunications closets and other associated spaces.

C.2 BENEFITS OF COMPREHENSIVE ADMINISTRATION

The implementation of a sound cable management strategy produces the following savings:

• Reduced time required to perform moves, adds and changes
• Reduced downtime when rectifying faults.
• Increased life of cabling system
• Dramatic reduction in life cycle cost.

C.3 WHEN TO START

The traditional, and possibly the easiest time, to implement new management and administration practices is when a department moves to new or retrofitted premises: a "clean start" provides the opportunity for accurately documenting the cabling and associated network equipment. Because day-to-day operations are not static, an organization continues to conduct moves, adds, and changes yet it must have a system to capture information on changes otherwise 'cabling anarchy' ensues. It is imperative that departments mandate that documentation be delivered on the day that the cabling system is handed over, and that it be accurate. Further, it is important to obtain documentation in both hard and soft forms to avoid re-entering the data into the department's system. This applies to drawings, wiring lists, cross-connect tables, narratives, etc.

Additionally, it is important that departments have the internal systems in place to ensure that they are able to maintain documentation accurately as moves, adds and changes take place. This is not an easy job and should not be underestimated. There are considerable advantages for departments in multi-tenant locations (i.e. two or more departments situated in a single building) to get together and pool resources. This can cover a variety of functions: pooling staff, pooling orders for MACs to the local telco in Centrex environments, etc.

Many organizations, even if they are not relocating to a new or renovated building, are coming to recognize that they have to get their cabling back under control. To do this they must audit, clean up, and document while the network remains live and the structured cabling staff continue with their regular duties.

Two fundamental methods can be adopted to get a cabling system under control. The "blitz" method involves the use of a team normally working after hours auditing sections of the building or site and combining the information in a database to gain a complete picture. A less aggressive way is to bite off small chunks over a longer period. By auditing and documenting the horizontal cabling, it is possible to identify which circuits are live and hence which of the patch cords and jumper cables are redundant (for the dead circuits). Redundant cables can then be removed. Done floor by floor, this rapidly gets the cabling system back under control, and particularly in Centrex environments can result in impressive cost savings.

C.4 CABLE MANAGEMENT PRODUCT TYPE OVERVIEW

Cable management systems can be broken down into four distinct groups:

• traditional paper systems;
• computer-based ones using word processing packages, spreadsheets, or DBMS packages;
• specialized software applications;
• electronic cross-connect systems.

C.4.1 Paper-based Systems

Paper systems are prone to human error and have no built-in methods of checking to ensure that the entered information is consistent and logical. As a result, accuracy deteriorates over time. Additionally, the paper itself often gets mislaid, or physically deteriorates. Importantly, there is no cross-referencing inherent in a paper-based system.

C.4.2 Office Software Applications

Spreadsheet, word processing, and DBMS packages have been used commonly for several years for cable system documentation. Their use has generally been to provide an enhanced replacement for a traditional paper system. However, while their use solves some of the problems associated with paper systems, they do not generally provide validity checks and are therefore still prone to human error.

C.4.3 Specialist Software Applications

The first purpose-built applications for cable management became available in the late eighties. The market and the products that serve this market have evolved rapidly since then.

CAD applications use a building drawing as the basis for the documentation. Items on the drawing have database records attached to them; a parallel database is used to record the circuits that result from established connections. The primary user interface is through the CAD system.

Database applications record all of the basic information within a database and cross-reference as necessary. Some also have the capability to display location information on an imported CAD floor plan. This gives the flexibility for the same application to be used with or without floor plans.

C.4.4 Electronic Cross-connect Systems

Electronic cross-connect systems, traditionally known as matrix switches, have been around for 10-15 years in the mainframe data centre environment. Now, a new generation of electronic cross-connect systems designed to handle a plethora of data and voice communications have entered the market place.

What exactly is an electronic cross-connect? It is a cross-connect device that resides in the telecommunications closet, providing any-to-any physical level switching (of copper, but not optical fibre connections). It basically automates all the physical tasks that are carried out in the telecommunications closet. The device is managed from a desktop PC using point-and-click operations. Typical attributes and benefits are:

· MACs are conducted electronically, thus eliminating the problems associated with moving patch cords and jumper cables;

· MACs can be carried out at remote locations - other parts of the building, other buildings on a campus, or even in another city, thus eliminating travel time and costs;

· MACs are automatically documented in the system's database in real time, thus eliminating incorrect or out-of-date information on the cabling infrastructure;

· MACs can be scheduled to be carried out in batch mode - say, overnight, or at a weekend, thus reducing or eliminating 'unavailable' time to users and overtime costs of technicians;

· Since the requirement to make constant visits by technicians to telecommunications closets is eliminated, a much higher degree of security can be guaranteed, and costs reduced. Additionally, access to networks behind the device can be denied by software during off-hours;

· Trouble-shooting using such diagnostic tools as protocol analyzers can be carried out on each and every cable remotely; again, visits are eliminated;

· Virtual LAN capability is automatically provided to low-end, low-cost hubs that do not have that capability built-in: a user can be moved on a port on hub A to any other port on hub B in seconds by dragging and dropping.

These systems also generally have the ability to graphically display floor plans, and user names, together with their workstation identifiers and telephone numbers.

C.4.5 CAD or DBMS?

Departments, when writing an RFP or RFQ, must make a choice between a database-oriented system and a CAD-based system. A number of factors will determine which is most suitable. The obvious starting point is to determine if all of the drawings are available. If they are, then the decision will probably be based upon whether or not the cable management must be integrated with any space planning tool that is already in use, or is planned, by the facilities management team. If it must, the solution is probably to take a CAD-based system. If it does not, then the advantages of database-oriented applications are obvious. Database systems are simpler, faster, more flexible and can be integrated more easily into other non-CAD-based systems. They will also use less powerful hardware to achieve the same performance.

C.4.6 Single User Versus Multi-user Systems

Many systems claim to be networkable but this does not necessarily mean they are truly multi-user. An organization must decide whether or not it needs a multi-user system and whether or not access to the information will be required over a WAN. If it is, then the speed of data retrieval over low bandwidth links must be considered, and if unacceptable, the cost implications of higher-speed links. The method of licensing is also important because it may affect the price. If the system is to be multi-user, it is preferable to use a system with concurrent user licensing rather than application copy licensing.

C.4.7 Ease of Use

Ease of use is very subjective, varies widely, and should be assessed by the procuring Department.

C.4.8 Data Import and Export

Many systems have inflexible or even non-existent modes of data entry. For example, all data must be hand entered since the package has no ability to import it from a spreadsheet, word processor, DBMS, or existing CAD database. This will significantly increase the cost. Of equal importance is the ability to export data. If the chosen system falls behind the competition and a decision is made at some time in the future to change systems, it must be possible to transfer from one system to another. 

C.5 CONCLUSION

The management and administration of a structured cabling infrastructure requires modern management methods and use of the appropriate hardware/software tools. This is possible through the advent of mature applications software designed specifically for the task. Direct and indirect savings provided by competent cable management systems have been shown repeatedly to justify their cost. Organizations that have recognized this fact benefit through reduced operational costs and improved service to their users. Downtime is reduced, which increases productivity and saves time and money. This philosophy is now widely accepted across both the private, and public sectors. 

ANNEX D: ZONE AND CONDUIT DESIGN

It is customary to align zones with the structural grid system of the building. Obviously, the size of the structural grid varies from building to building, so the standard approach to designing cabling implementations will have to be adjusted somewhat on a case-by-case basis.

Figure D.1 below illustrates typical (usually, but not always, square) zone sizes A through D. In cases where structural grids do not align with those shown as being typical, for example in the case of rectangular grids, the area less than or equal to the required zone area should be used. An example is that in the case of a structural grid of 9.4m (20 ft) by 11.8 m (25 ft), the provisioning should be equivalent to that of Zone C. 

Figure D.1

Zone C represents a good compromise between minimizing the number of conduit runs and minimizing the number of cables that need to be installed within the conduit. Note that as zone size increases the length of cable doubled back to the telecommunications closet increases. This must be accounted for when determining the maximum cable length. By extension, zones larger than Zone D are not recommended. Therefore, it is recommended that designers implement zoned conduit based on the Zone C parameters wherever possible.

It is recommended that the parameters in Table D1 below be used as a guide in sizing conduit runs from the telecommunications closet to work areas. At the very least conduit should be sized such that there is room for AT LEAST two runs of 4-pair UTP and one optional duplex run of optical fibre cable: 

Table D.1: Conduit Sizing

Footnotes

(1) Where a conduit is sized for cable fill of less than 30% fill, the distance between boxes may be increased to 60 metres (200 ft.)

(2) "Low-speed data" is arbitrarily defined to be that used by older data communications methodologies, typically having speeds an order of magnitude less than 10 Mbps Ethernet.

(3) A 1996 version is in balloting as this TBITS goes to press.