Structured Cabling
Structured Cabling
Structured Cabling
STRUCTURED CABLING
Foundation for the Future
Table of Contents
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3
Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .17
Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .18
Introduction
As today’s communication networks become more complex—as more users share periph-
erals, as more mission-critical tasks are accomplished over networks and as the need for
faster access to information increases—a good foundation for these networks becomes
increasingly important. The first step toward the adaptability, flexibility and longevity
required of today’s networks begins with structured cabling—the foundation of any infor-
mation system.
It is vital that communications cabling be able to support a variety of applications and last
for the life of a network. If that cabling is part of a well-designed structured cabling sys-
tem, it can allow for easy administration of moves, adds and changes and smooth migration
to new network topologies. On the other hand,“worry-about-it-when-you-need-to” systems
will make moves, adds and changes a hassle and make new network topologies too diffi-
cult to implement. Network problems occur more often, and are more difficult and time-
consuming to troubleshoot.When communication systems fail, employees and assets sit
idle, causing a loss of revenues and profits. Even worse, the perceptions of customers and
suppliers can be adversely affected.
The purpose of this white paper is to present the advantages of using a standards-based
structured cabling system for a business enterprise. The paper will cover a brief historical
perspective of structured cabling, a review of the current standards, media types and per-
formance criteria, system design and installation recommendations. Particular attention
will be given to the ANSI/TIA/EIA-568-A standard and the horizontal cabling subsystem in
that standard.
By defining every aspect of their system, manufacturers “locked” customers into a propri-
etary system. One manufacturer’s system would not work with another, or run over any
other type of cabling. If a customer decided to change systems, not only would new
electronics and software need to be purchased, but new cabling would need to be
installed as well.
Moves, adds or changes were also difficult with a proprietary system. Each time a new
machine was added to the network, new cable had to be installed and inserted into the
ring or attached to the bus. Furthermore, the whole system might have had to come down
to add the new user.
Compounding the problem of proprietary systems was the telephone cabling system.
As part of its 1984 divestiture agreement, AT&T was no longer responsible for the
cabling inside a customer’s building. From then on the service provider maintained
the system only up to the demarcation point where the telephone service entered the
building. Beyond that point, maintenance and upgrade of the telephone system was the
customer’s responsibility.
As a result, network managers had (and many still do have) two distinct systems demand-
ing their full and undivided attention.The desire for one system that could run any appli-
cation without the accompanying hassles and headaches of past systems was magnified
exponentially. Enter structured cabling.
A flexible cabling plant is made possible with standards adopted by both the cable plant
manufacturers and by the electronic manufacturers. If an end-user also follows these same
standards, then any application, cable, connector or electronic device built for compliance
will work in the same system.
Each of these standards works in conjunction with 568-A. Additional standards like the
National Electric Code (NEC) or local provisions and laws also must be reviewed when
designing or installing any telecommunications system. This paper concentrates on 568-A
and describes some of the basic elements of a generic cabling system, cable types and
some of their advantages and disadvantages, and installation requirements and practices.
1. The entrance facility is the point where outside plant cables and associated hardware
are brought into the building. The entrance facilities may be used for public network
services, private network customer services or both. The demarcation point between
carrier and customer, and overvoltage protection devices are located here.
5. The horizontal cabling consists of the physical media used to connect each outlet to a
closet.Various types of cable can be used for horizontal distribution. Each type has its
own performance limitations, size, cost and ease-of-use. (More on this later.)
5
4
6
3
1. Entrance Facility 4. Telecom Closet
2. Equipment Room 5. Horizontal Cabling
3. Backbone Cabling 6. Work Area
2
1
Star-wired Topology
The 568-A standard specifies that a structured cabling system use a star topology (refer to
Figure 1). Each work-area telecommunications outlet must be connected to a cross-con-
nect in a telecommunications closet. All cables from a floor or area in a building therefore
run back to one central point for administration. Each telecommunications closet must be
star-wired back to the equipment room for the building. In a campus environment each
building is star-wired back to one main administration area.
Using a star-wired topology eliminates many of the pitfalls encountered with proprietary
systems. First, a physical star will still work as though connected to a ring or bus topology
so that in the event of a station failure, the networking electronics can bypass that particu-
lar station.This keeps station or lobe problems local, avoiding a global system crash. Star-
wiring allows for changes in network applications—such as going from a ring or chain-
based application to a bus-oriented one—without pulling new cable, thus saving time,
effort and money.
With good planning, changing a telephone outlet to a workstation location is easily accom-
plished by changing patch cords in the closet and equipment at the outlet. No additional
cable needs to be pulled or terminated.
Media Types
One of the first choices faced when planning or developing a structured cabling system is
the type of media to be used. 568-A recognizes three different media:
• Unshielded Twisted Pair (UTP) — 4-pair, 24-gauge, 100 Ohm copper cable
• Shielded Twisted Pair (STP) — 2-pair, 22-gauge, 150 Ohm copper cable
• Single-mode and multimode optical fiber cables
Coaxial cables were recognized by the original 568 standard mainly because their installed
base was used for Ethernet (10BASE2 and 10BASE5) applications. In the 568-A document,
coax is grandfathered, but not recognized. In other words, if a system has already been
installed using coax, it may be maintained, added to or changed. However, it should not be
used for new system installations.
Today, Category 5 is the most popular standards-recognized media for high-speed data
applications due to ease of installation, lower installation costs and lower space consump-
tion. Compared to STP, UTP cables are smaller, more flexible and less expensive.The elec-
tronic components used with UTP are also the least expensive of the three recognized
media. And since these make up a large portion of the overall network investment, this
expense is factored heavily in the decision to use UTP.
A word of caution: As with any chain, a structured cabling system is only as strong as its
weakest link. Therefore, to obtain Category 5 performance from a link, the entire link
must be composed of components that comply to Category 5 standards. Using Category 5
distribution cable, connecting hardware and outlets along with Category 3 patch cords
will result in only Category 3 performance.
The purchase of Category 5 cables and components, however, does not guarantee
Category 5 performance from an installed system.The installation of Category 5 structured
cabling systems has a tremendous effect on the final performance level. In fact, without
proper installation, the high performance of a Category 5 system can be reduced to that of
a simple phone system.
The high performance of STP systems is a result of shielding. In an STP cable, each twisted
pair is foil-wrapped and a metallic braid is placed just underneath the cable sheath. These
components reduce emissions from the cable and protect the pairs from outside interfer-
ence when the shielding is properly grounded.
Decibels
Attenuation and NEXT are both reported in decibels (dB) as negative numbers. Since the
minus (-) sign is assumed for these two parameters, a statement of “... 40 dB of NEXT...”
actually means -40 dB. Attenuation and NEXT dBs are relative measures of changes in
voltage. Further, an increase of 10 dB means a 10-fold increase in the measured parameter.
Figure 3 shows the logarithmic progression of decibels. Note that the minus signs have
been assumed.
3 dB = 2X
10 dB = 10X
20 dB = 100X
30 dB = 1000X
40 dB = 10,000X
50 dB = 100,000X
60 dB = 1,000,000 X
Near-End Crosstalk
Crosstalk occurs along a circuit when a portion of the energy from one signal is jumping
or crossing from one pair to another. In data communications, the main concern is with
crosstalk that occurs at the “near end” or close to the transmitters. UTP and STP cables are
designed for bidirectional conversations.That is, at each end of the cable length, one pair
is used for transmitting and another for receiving.The transmit pair at one end becomes
the receive pair at the other.
The close proximity of the transmit and receive pairs increases the probability of exces-
sive NEXT. Since transmissions attenuate so rapidly, a relatively powerful signal must be
transmitted for something recognizable to arrive at the other end. Simultaneously, relative-
ly weak signals are being received on an adjacent pair. Even small imbalances in the trans-
mitted signal can cause NEXT. The first 50–60 feet of a cable run are especially vulnerable
to NEXT because the greatest disparity exists there between the strengths of the transmit
and receive signals. Further down the run, the transmitted signal has attenuated to the
point where little, if any, of the effects of NEXT are present. However, connection points
are also susceptible to crosstalk. Quality products and proper installation will ensure that
NEXT does not overpower received signals.
To measure NEXT, a known signal level is generated onto a transmit pair. Any portion of
that signal “heard” on the receive pair is crosstalk. Less of this signal on the receive pair
and bigger numbers are ideal. Therefore, NEXT of 40 dB is better than 30 dB performance.
Both environmental and installation-related factors can cause increases in noise and NEXT.
Some of these factors include untwisting of pairs, cable jacket removal, bend radius, fluo-
rescent lights and motors.
Attenuation-to-Crosstalk Margin
The effect of attenuation on data transmissions requires the reduction of all noise forms in
cabling, including crosstalk. Too much noise will prevent the receiver from distinguishing
the transmitted signals from unwanted noise. The result can be garbled, incoherent data,
retransmissions and slow network response time.
0 dB
Attenuation
Near-End Crosstalk
80 dB
0 MHz 100 MHz
High-quality products and proper installation techniques will attain the greatest possible
margin between NEXT and attenuation—optimizing the performance and reliability of a
network. For example, if products are used that meet, but do not exceed, the minimum
Category 5 requirements, the installed system may not perform at Category 5 levels. Each
time a cable is pulled, bent, untwisted and terminated, a small amount of attenuation
and/or NEXT is added to the link. Necessary installation tasks can cause the attenuation-to-
crosstalk ratio of marginal Category 5 products to fall below the 3 dB minimum. Choosing
products that exceed the standards—ones that provide “head room”—prevents this situa-
tion. The greater the head room of a product, the more use and abuse that product will
withstand before it no longer meets Category 5 requirements.
Figure 5 compares the three data grade categories of 4-pair UTP cables. The results com-
pare the performance of these categories at 16 MHz, the peak frequency for Category 3. A
look back to Figure 3 shows that the 21 dB differential between Category 3 and Category
5 represents a 100-fold increase in NEXT performance.
Fiber optic cables consist of a glass core and cladding surrounded by a protective coating.
The core and cladding are part of the same glass rod but have different optical properties.
Light pulses are injected into the core. As light pulses travel down the cable, the cladding,
acting as a mirror, reflects the pulse back to the center of the core. A plastic protective
coating, called a buffer, surrounds the core and cladding. Figure 6 shows the construction
of a fiber optic cable.
Coating
Cladding
Core
Single-mode and multimode fibers also have different optical light sources. Lasers transmit
optical signals into a single-mode fiber because they emit a focused, high-powered light
beam. The lasers’ wavelengths in single-mode applications are 1,310 and 1,550 nanome-
ters (nm). Since the core size of multimode fibers is much larger than that of single-mode,
lower-powered LEDs (light emitting diodes) in the 850 and 1,300 nm range are typically
used as their optical source.
Fiber optic cable construction differs depending on the purpose and application of the
cable. A tight buffer construction is used for indoor applications.This buffer brings the
total fiber size to 900 um, providing extra protection to allow for direct connectorization
of a fiber. Outdoor cables use a loose tube design in which 250 um coated fibers float in a
moisture-resistant gel. The gel combats the harmful effects of temperature, humidity and
mechanical loads. In addition to the buffer tubes, fiber optic cables may include Kevlar
yarn, armor sheaths, steel messengers or other components.
System Design
Since the 568-A standard recognizes the need for both data and voice communications, it
requires a minimum of two ports per workstation. The first port must be supported by a
4-pair, 100 Ohm, UTP cable, Category 3 or higher. The second port must be supported by
one of the following media: a UTP cable (Category 5 recommended), STP or a multimode
62.5 um optical fiber cable (two strand minimum).
For horizontal cable runs from a telecommunications closet to a work-area outlet, the max-
imum transmission distance allowed for any media type is 295 feet (90 meters). An addi-
tional 33 feet (10 meters) provides for patch cords both in the closet and the work area.
Figure 7 shows the TIA horizontal link model with corresponding distance limitations for
cables and cordage.
Installation Practices
It is vital to use proper installation practices to ensure that the performance of the entire
system is not diminished simply through improper installation. Using a Category 5 system
as an example, any one of the following practices may not, by itself, reduce a system
below Category 5 compliance. Since many of these items such as pulling cables and
untwisting pairs are unavoidable, care must be taken not to allow their combined effect to
lower the performance of the system.
One of the first installation tasks is pulling cables from the telecommunications closet to
each outlet location. The maximum pulling tension for Category 5, according to 568-A, is
25 lbf. Higher tension on the cable may stretch the twists, or in effect, untwist the pairs.
Extreme tension will also increase attenuation.Visual inspection of an installed link cannot
identify that tension limitations have been exceeded. Using proper pull techniques, pull
cords and cable lubricants will prevent this type of damage.
The next step in the installation process is to prepare the cable for termination. This
involves stripping away some of the jacketing material and untwisting the conductors.
The jacketing material should not be removed any more than necessary to complete
a termination.
As stated in 568-A, the pairs in a Category 5 cable should never be untwisted more than
1/2 inch from the point of termination. Any further untwisting of the pairs will increase
crosstalk and susceptibility to EMI/RFI.
After all of the conductors in a cable have been terminated, the cable must be dressed or
managed. Cable management keeps termination fields neat and orderly, and provides strain
relief for the cable. Strain relief transfers the weight of the cable from the cable itself to
some sort of supporting hardware.Without strain relief, the weight of the cable will cause
it to sag and eventually pull away from its termination.
Cable ties are the most popular method of attaching cables to support hardware. These
thin plastic strips are looped around a bundle of cables and cinched tight enough to sup-
port the cables. Cable ties also make large bundles of cables more manageable and give
the installation a neat appearance. But over-cinching of cable ties has the same effect as
tight bends. The conductors of the outermost cables in the bundle tend to become flat-
tened. A proper cinch of a cable tie will allow the cables to slide easily back and forth
within its loop. Staples are not recommended for Category 5 installations.
Proper planning includes examining all applications, network technologies and telecom-
munication outlet locations that are currently used or might be used in the future.
Accounting for all foreseeable scenarios allows the physical infrastructure to be installed
once, yet serve business requirements not yet on the horizon. Moves, adds and changes no
longer require pulling new cables—except when your physical space expands.
Choosing the proper media for a new cable installation depends on the applications and
services that the network is expected to provide. Category 3 UTP cables are sufficient if a
network is designed strictly for telephony or low- to medium-speed data applications like
Ethernet. Electrically noisy areas such as X-ray labs, HVAC equipment rooms or near motors
may lend themselves more to a shielded or optical fiber cable. Complete fiber solutions are
ideal for enterprises that want to cable once regardless of the application being used today
or tomorrow—or for groups or users that demand large quantities of information.
Budgetary concerns should impact the decisions made to this point. However, keep in
mind that standards-based systems are designed to last a minimum of 10 years from time
of installation. Further, many of today’s products are warranted for longer periods such as
15 or 25 years. Therefore, cutting corners on the foundation of the network will have ram-
ifications for many years to come.
A properly planned and installed system will allow companies to spend their time, atten-
tion and scarce capital resources in other areas for years. The ultimate goal is to run any-
thing, anywhere, at any time. The other option is to face each network problem and issue
as it arises. The choice is yours.
TSB-67, Transmission Performance Specifications for Field Testing of UTP Cabling Systems