Microtest, Inc.

FIELD TESTING OF HIGH PERFORMANCE

PREMISE CABLING

AN UP-TO-DATE REVIEW OF
PHYSICAL LAYER
MEASUREMENTS,
CABLING STANDARDS,
TROUBLESHOOTING PRACTICES
AND CERTIFICATION
TECHNIQUES

Copyright © 1994,1995,1996, 1998. Microtest, Inc. All rights reserved.

1
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

While every effort has been made to ensure all information in

this handbook is correct, Microtest assumes no liability or
obligation to the user regarding suitability for purpose or accu-
racy. This document is not intended as a replacement for reading
and understanding the standards referred to herein. Technology
and standards change rapidly, and we encourage you to check
with vendors and standards bodies for current information.
Copyright © 1994, 1995, 1996, 1998 Microtest Inc. All rights
reserved. No part of this document may be reproduced in any
form without the express prior written consent of Microtest Inc.
Printed in the United States of America
Written by Mark Johnston, Director of Technology Development

2
 Microtest, Inc.

Contents
Preface .......................................................................................... 5
Introduction .................................................................................. 6
How to Use This Handbook ....................................................... 7
High Performance Cabling ..................................................................... 7
Physical Layer Tests ................................................................................ 7
Cabling Standards ................................................................................... 7
LAN Application and Standard Performance Summaries ......................... 7
Troubleshooting and CertiFication Tools ............................................... 7
Link Troubleshooting Guide .................................................................. 8
Appendix 1: Cable Installation Tips ........................................................ 8
Appendix 2: A Guide to Modular Connector Wiring .............................. 8
Appendix 3: Standards Organizations ...................................................... 8
High Performance Cabling ........................................................ 9
High Performance Cabling Alternatives .................................................. 9
Shielding Issues .................................................................................... 12
Fiber Optic Cabling .............................................................................. 13
Physical Layer Tests ................................................................ 16
Copper Cabling Measurements ............................................................. 16
Length ................................................................................................. 16
Delay .................................................................................................... 18
Delay Skew ........................................................................................... 18
Near End Crosstalk (NEXT) ................................................................ 19
Power Sum NEXT ................................................................................ 21
FEXT and ELFEXT .............................................................................. 21
Attenuation .......................................................................................... 22
Attenuation to Crosstalk Ratio (ACR) .................................................. 24
Bandwidth ........................................................................................... 25
Wire Map .............................................................................................. 26
Impedance ........................................................................................... 28
Structural Return Loss/Return Loss ..................................................... 28
DC Loop Resistance ............................................................................. 29
Noise .................................................................................................... 29
Fiber Optic Measurements ................................................................... 30
Cabling Standards .................................................................... 33
Current Published Standards ................................................................ 34
Standard Description Scope .................................................................. 34
Link Definitions ................................................................................... 35
Cabling Standards in Development ...................................................... 37
TIA 568A Addendum 2: Additional Transmission
Performance Specifications for UTP ..................................................... 37

3
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

TIA 568A Addendum 3: Additional Transmission

Performance Specifications for Enhanced Category 5 Cabling ............. 38
TIA PN#3727: Enhanced Twisted Pair Cabling Systems ...................... 38
Amendment to EN 50173:1998 .......................................................... 39
EN 50173:2000 .................................................................................. 39
EN 50xxx:Cabling Test Methods ......................................................... 39
LAN Application and Standard Performance Summaries .. 41
Introduction ........................................................................................ 41
Overview of Categories and Classes ..................................................... 42
Table 6: Categories and Classes ............................................................. 42
TIA ....................................................................................................... 42
Basic Link Performance Summary ........................................................ 42
TIA/ISO/IEC/CENELEC: Channel Performance Summary ............ 44
ISO/IEC/CENELEC: Permanent Link Performance Summary ......... 46
Technology Application Matrix ............................................................. 48
Fiber LAN Application Performance Requirements .............................. 49
Troubleshooting & Certification Tools ................................... 51
The Right Tool for the Right Job ......................................................... 51
Network Analyzers ............................................................................... 52
Handheld Certification Tools ............................................................... 53
Cable Testers ........................................................................................ 58
Continuity Testers ................................................................................ 58
Link Troubleshooting Guide .................................................... 59
Solving Wiremap Problems ................................................................... 59
Solving Length/Opens/Shorts Problems ............................................. 59
Solving Resistance Problems ................................................................. 60
Solving Attenuation Problems .............................................................. 61
Solving NEXT Problems ...................................................................... 61
Solving ACR Problems ......................................................................... 63
Solving Noise Problems ........................................................................ 63
For More Information ................................................................ 64
Glossary ..................................................................................... 65
Appendix 1: Cable Installation Tips ..................................... 75
Appendix 2: A Guide to Modular Connector Wiring ........... 79
Appendix 3: Standards Organizations ................................. 83
Index ........................................................................................... 86

4
 Microtest, Inc.

Preface
Physical layer cabling is the basic building block for most local area
networks. Cabling challenges arise as these networks reach
100Mbps and beyond, especially for unshielded twisted pair (UTP)
networks. Standards are rapidly evolving, and the quality of today’s
cabling and connecting hardware far exceeds the best products of
only a few years ago.
While the rate of change in cabling technology has increased, the
level of education has not kept pace. A great deal of confusion exists
regarding cable measurements for high speed cabling, and how to
interpret evolving standards.
Installing and maintaining a reliable cable plant is essential to the
well being of today’s mission-critical LANs. Category 5E UTP is
today’s preferred choice for LAN cabling. Standards for Gigabit
Ethernet are almost complete, providing 1000 Mbps over UTP.
Category 6 will provide significant performance advantages over
previous types of UTP, and Category 7 takes copper further still.
While Category 5 UTP has become the preferred choice for LAN
cabling, enhanced Category 5, Category 6, Category 7, and optical
fiber have made significant strides in the last 12 months. There are
many new alternatives, and it can be difficult to separate hype from
useful information.
This handbook is intended for those who work with high speed
cabling, both twisted pair and optical fiber. It provides an up-to-date
review of cabling alternatives, physical layer measurements, cabling
standards, troubleshooting practices, and certification techniques.

5
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

Introduction
How do you know if Category 5 is good enough for your appli-
cation? Is a Class A link better than a Class F link? What is the
ISO/IEC/JTC1/SC25/WG3 US TAG up to and should I care?
Are changes in international standards likely to affect me? What
do I really need to know about cabling standards? What simple
things can I do to minimize network downtime?
A great deal of confusion surrounds network cabling issues. Yet
cabling accounts for 50-70% of all LAN failures, and the result-
ing network downtime costs organizations thousands of dollars
per year in lost productivity, idle resources, and lost revenues. If
you are a network administrator, it is essential to understand
how to get the best performance from your cabling investment,
and find faults quickly when they occur.
If you are a cable installer, you need to be up to date on the
latest cable installation and conformance techniques to ensure
your work meets changing international standards.
Fortunately, success with high performance premise cabling is
not difficult once its capabilities and characteristics are under-
stood. This handbook is intended to help you navigate through
the clutter of standards, cable classification systems, and con-
formance requirements, so that you can increase your productiv-
ity and deliver better value to your organization.

6
 Microtest, Inc.

How to Use This Handbook

Review the following chapter summaries for areas of individual
interest. If you are relatively new to the subject, simply read the
chapters in the order they are presented. If you are familiar with
the subject, you may be interested in Cabling Standards as
well as LAN Application and Standard Performance
Summaries. A guide to each chapter follows.

High Performance Cabling

Describes and compares high performance cabling alternatives.

Physical Layer Tests

Describes all the key measurements used in cable testing and
certification. Read this chapter if you don’t know what all the
terms mean or which ones are most important.

Cabling Standards
Compares and contrasts domestic and international cabling
standard bodies. Provides a summary of existing cabling stan-
dards as well as a detailed look ahead to what is in development
and coming soon.

LAN Application and Standard Performance

Summaries
Defines categories 3-7 and Class C-F. Outlines Basic Link,
Permanent Link, and Channel performance for all categories and
classes. Maps LAN applications into their corresponding re-
quirements. Summarizes fiber optic LAN application require-
ments at the physical layer.

Troubleshooting and CertiFication Tools

Describes the pros and cons of using all common types of cable
troubleshooting and conformance tools.

7
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

Link Troubleshooting Guide

Includes tables designed to help you quickly pinpoint the spe-
cific reason for any cable faults you may encounter.

Appendix 1: Cable Installation Tips

Provides recommendations for improving the quality of your
cable installations, which should lead to greater customer
satisfaction and reduced network downtime.

Appendix 2: A Guide to Modular Connector Wiring

Summarizes common connector types, wiring practices, and
color coding for connectors, patch cords, and horizontal wiring.

Appendix 3: Standards Organizations

Where to go for more information and copies of the standards
referred to in this handbook.

8
 Microtest, Inc.

High Performance Cabling

High Performance Cabling Alternatives

High performance premise cabling is available in both twisted pair
and optical fiber alternatives. This section will describe and com-
pare the many cabling alternatives. Table 1 provides an overview of
the name, construction, advantages, expected performance, and
relative cost of today’s most popular cable types.

Name Cross Section Construction Expected Performance Cost/

Illustration node1
Cable consists of 4 pairs Mixed and matched
of 24 AWG (0.50mm) cables and connecting
copper with thermoplastic hardware from various
Basic polyolefin or fluorinated manufacturers that have
Category ethylene propylene (FEP) a reasonable chance of 1
5 UTP jacket. Outside sheath meeting TIA Cat 5
consists of polyvinyl- Channel and ISO Class
chlorides (PVC), a fire D requirements. No
retardant polyolefin or manufacturers warranty
fluoropolymers. is involved.
Cable consists of 4 pairs Category 5 components
of 24 AWG (0.50mm) from one supplier or from
copper with thermoplastic multiple suppliers where
polyolefin or fluorinated components have been
Enhanced ethylene propylene (FEP) deliberately matched for
Cat 5 jacket. Outside sheath improved impedance and 1.2
UTP consists of polyvinyl- balance. Offers ACR
(Cat 5E) chlorides (PVC), a fire performance in excess of
retardant polyolefin or Cat 5 Channel and Class
fluoropolymers. Higher D as well as a 10 year or
care taken in design and greater warranty.
manufacturing.
Cable consists of 4 pairs Category 6 components
of 0.50 – 0.53mm copper from one supplier that
with thermoplastic are extremely well
polyolefin or fluorinated matched. Channel zero
ethylene propylene (FEP) ACR point (effective
Category jacket. Outside sheath bandwidth) is guaran- 1.5
6 UTP consists of polyvinyl- teed to 200 MHz or
chlorides (PVC), a fire beyond. Best available
retardant polyolefin or UTP. Performance
fluoropolymers. Ex- specifications for
tremely high care taken in Category 6 UTP to 250

9
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

Name Cross Section Construction Expected Performance Cost/

Illustration node1
design and manufacturing. MHz are under
Advanced connector development.
designs.
Cable consists of 4 pairs of Category 5 components from
24 AWG (0.50mm) copper one supplier or from multiple
with thermoplastic polyolefin suppliers where components
or fluorinated ethylene have been deliberately
propylene (FEP) jacket plus designed to minimize EMI
Foil a drain wire. Pairs are susceptibility and maximize
Twisted surrounded by a common EMI immunity. Various 1.3
Pair metallic foil shield. Outside grades may offer increased
sheath consists of polyvinyl- ACR performance.
chlorides (PVC), a fire
retardant polyolefin or
fluoropolymers.
Cable consists of 4 pairs Category 5 components
of 24 AWG (0.50mm) from one supplier or from
copper with thermoplastic multiple suppliers where
polyolefin or fluorinated components have been
Shielded ethylene propylene (FEP) deliberately designed to
Foil jacket. Pairs are minimize EMI susceptibil-
surrounded by a common ity and maximize EMI 1.4
Twisted immunity. Offers superior
metallic foil shield, followed
Pair by a braided metallic EMI protection to FTP.
shield. Outside sheath
consists of polyvinyl-
chlorides (PVC), a fire
retardant polyolefin or
fluoropolymers.
In the Type 1 version, Positive ACR to 300
consists of 2 pairs of 22 MHz. However, rarely
AWG copper with used since it is only a 2
thermoplastic polyolefin or pair implementation.
fluorinated ethylene STP has fallen out of
STP-A propylene (FEP) jacket. favor in most LAN
(IBM Pairs are individually installations. It is included N/A
Type 1) surrounded by a metallic in the table only because
foil shield, followed by a its acronym is often
braided metallic shield. confused with the more
Outside sheath consists of recent 100 ohm cabling
polyvinyl-chlorides alternatives.
(PVC), a fire retardant
polyolefin or
fluoropolymers.

10
 Microtest, Inc.

Name Cross Section Construction Expected Performance Cost/

Illustration node1
Also called PiMF (for Pairs Category 7 cabling
in Metal Foil), SSTP provides positive ACR to
consists of 4 pairs of 22-23 600 – 1200 MHz.
AWG copper with a Shielding on the
thermoplastic polyolefin or individual pairs gives it
Category fluorinated ethylene phenomenal ACR. A
7 propylene (FEP) jacket. recommendation on a
Pairs are individually Category 7 connector is
Shielded- surrounded by a helical or expected in 1998, 3.1
Screened longitudinal metallic foil though it will not be
Twisted shield, followed by a compatible with the
Pair braided metallic shield. popular modular 8 RJ45.
Outside sheath consists of Detailed performance
polyvinyl-chlorides (PVC), specifications for
a fire retardant polyolefin orCategory 7 are under
fluoropolymers. development.
Two-Fiber zipcord Multimode fiber has much
multimode cable. There higher performance than 2.4
are many types of any copper alternative in

(does not include optical electronics)

multimode cable. This this table. It is capable of
representative type has a supporting data rates up
62.5 micron (or 50 to 2 Gbps at distances up
Multimode micron) core surrounded to 2 km. It is completely
Fiber by 125 micron cladding. immune to EMI and
The cladding is covered NEXT, and has very low
by a thermoplastic buffer, attenuation. The added
which is in turn covered cost of optical electronics
by a aramid (usually has traditionally been the
Kevlar) strength member, main reason for fiber’s
followed by a PVC jacket. low penetration to the
desktop.
Single mode fiber optic Single mode fiber has
cabling has a similar much lower loss, higher 3.5
construction to multimode bandwidth, and similar
(does not include optical electronics)

fiber, except that the core cost to multimode fiber. It

is 8-9 microns in diameter has not achieved much
Single and constructed of market penetration in
l extremely low-loss glass premise wiring applica-
Mode designed to pass tions because the
Fiber monochromatic light. necessary laser
transceivers add
significant cost to the
overall LAN implementa-
tion. Labor costs and
testing also requires a
higher skill level.
Table 1: High Performance Cabling Alternatives

11
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

1
Cost comparisons in the table are based on estimates available as of
early 1998, will vary by location and supplier, and may change
significantly in a short period of time. Please check with local
suppliers for current information. The cost of a basic Cat 5 installa-
tion is used as a unity reference point. The cost is calculated based
on the total cost of the cable installation divided by the number of
outlets. The average length of a link is 50 meters. The multimode
fiber example used four ST connections per segment. All labor to
design, install, test, and manage the installation was included.

There are other factors to consider as well. Is it likely you will

need bandwidth over 100 MHz? Even GigaBit Ethernet, at 1000
Mbps, is being designed to run on 100 MHz category 5E chan-
nels. Category 6 is much better than Category 5E, providing
superior ACR and return loss. Fiber optic cabling beats copper
alternatives in all respects except for cost. It is good practice to
buy and install the best quality cabling you can afford.
Cabling technology is changing, and there are three questions
that you should ask your cabling supplier:
1. What will it cost?
2. What will it do for me?
3. Why do I need it?
You cabling is a long-term, mission critical investment. Be sure
you are satisfied with the answers to these questions before
making a cabling decision.

Shielding Issues
Generally speaking the US, UK, and Canada are pro-UTP and
against ‘unnecessary’ shielding. Experts from these countries
maintain that well-balanced cabling, properly installed, will meet
all RF/EMI regulations. Experts from Europe counter that such
good balance is never seen in the field, and shielding is not only
prudent but also necessary. Airport radar runs at about 600 kW.

12
 Microtest, Inc.

This produces a field of 4.2 volts per meter one kilometer away
from the source. A GSM phone (the common cell phone used in
Europe) produces a field of 4.3 volts per meter 2 meters away
from the phone. Some European experts maintain such emis-
sions will affect data transmission in unshielded cabling. There
is also a heightened concern about the deleterious health effects
of RF/EMI in many European countries, which also tends to
support a market for shielded cabling alternatives to UTP.
Performance of FTP can be comparable to UTP, however, for
conductors of the same diameter FTP will exhibit slightly more
attenuation due to the parallel ground plane. Some FTP manu-
facturers compensate for this effect by slightly increasing the
copper gauge. FTP also costs slightly more per meter and costs
about 10-15% more to install due to the extra time to manage
shield connections.
Both shielded and unshielded cables can provide high perfor-
mance. Which type you choose is a matter of personal prefer-
ence, cost, and performance requirements.

Fiber Optic Cabling

An optical fiber consists of a thin glass or plastic core sur-
rounded by cladding. Fiber optic cabling comes in two main
types: single mode and multimode. Multimode fiber optic cable
has a wider core, so that a beam of light has room to follow
multiple paths through the core. It is called multimode because
there are multiple light paths (modes) in the fiber. Multimode
fiber is best suited for LAN applications because it is easier to
terminate, uses wider aperture, less expensive LED transmitters,
has plenty of bandwidth, and is more than adequate for short
distance (< 5 km) LAN applications.
In single-mode fiber-optic cable, the core is so narrow that the
light can take only a single path (i.e. a single mode) through it.

13
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

Single mode fiber has very little signal attenuation, which is why
it is used extensively in telephony applications, where extremely
long distances and extremely high bandwidths are needed. This
type of cable is the most difficult to install, because it requires
the greatest precision. It also requires laser transmitters, which
are much more expensive than LED transmitters.
Fiber most often used in LANs is often referred to “62.5/125
micron”. What does this mean? Figure 1 below illustrates a
cross-section of a duplex fiber optic cable. The core is the glass
medium through which the light travels. It is 8 to 9 microns
(mm, or millionths of a meter) in diameter for single mode fiber,
and either 50 mm or 62.5 mm for multimode. The cladding is a
protective layer with a lower index of refraction than the core.
The lower index means that light that hits the core walls will be
redirected back to continue on its path.
Thermoplastic
buffer
Core : Aramid
SM: 8-9 µm strength
MM: 50 or member
62.5 µm
PVC jacket
Cladding: 125 µm
Figure 1: Construction of Typical
Two-Fiber Multimode Fiber Cable

Multimode fiber cabling is designed to have minimal attenuation

at three windows centered around the following wavelengths:
850 nanometers (1 nm = 1 billionth of a meter), 1300 nm, and
1550 nm. Fiber LAN transceivers designed to attach to multi-
mode cabling are optimized to either 850 or 1300 nm. For
example, 10BASE-F uses an 850 nm wavelength, while FDDI
(Fiber Distributed Data Interface) uses a 1300 nm wavelength.
Because the installer generally can’t predict with certainty what
LAN application may be attached, it is important to test in-

14
 Microtest, Inc.

stalled fiber at both 850 nm and 1300 nm.

A
T
T
E
N
U
A
T
I
O
N
850 1300 1500
Wavelength (nm)

Figure 2: Attenuation vs Wavelength for a Typical

Fiber Optic Cable

The above graph shows that there is much less loss at 1300 nm
than 850, so why don’t most LAN networks use 1300 nm
transceivers? It is because 850 nm transceivers are considerably
less expensive.

15
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

Physical Layer Tests

Copper Cabling Measurements

Length
Why measure length? For most UTP connections, provided the
signal propagation delay does not exceed system requirements,
the actual length is much less important than the attenuation
and NEXT of the link. Length is the best measurement for
finding opens, shorts, and breaks in the cable. Excessive cable
length is more common to coaxial than to UTP LAN’s, but the
principles of length measurement are the same.
Length is most commonly measured using a technique called
Time Domain Reflectometry (TDR). TDR can be thought of as
‘cable radar’. Typically, the cable tester generates a pulse down
the wire. When the pulse encounters a change in impedance,
such as an open, short, or poor connection, all or part of the
pulse energy is reflected back to the tester. The tester measures
the delay between when the pulse was sent and when the
reflection was received. Knowing the speed of electron travel in
the cable (Nominal Velocity of Propagation or NVP), the tester
can then calculate the distance to the fault. NVP is normally
expressed as a percentage of the speed of light (c), such as
0.82c.
The size of the reflected pulse will be proportional to the magni-
tude of the change in impedance. Thus, a large change in imped-
ance such as an open or short will cause a large reflection, and a
minor impedance change such as a poor connection will create a
smaller reflection.
Your measurement of length will only be as accurate as the NVP
setting you use, which varies from about 0.6c to 0.9c for most
cable types. Thus, if your tester’s NVP is set for 0.6c and your
cable’s NVP is actually closer to 0.9c, your result could be more
than 30% inaccurate. To avoid these gross errors in the field, it

16
 Microtest, Inc.

is a good idea to look for a tester with a wide variety of prepro-

grammed cable types.
End or
Scanner Cable fault
Transmitted pulse
OPEN
Reflected pulse

Transmitted pulse
SHORT
Reflected pulse

Transmitted pulse
TERMINATOR
No reflected pulse

Figure 3: TDR
It is very difficult to achieve an accuracy better than 2% using
TDR, because the NVP also varies by up to 4 - 6% between
different pairs in the same cable. Plus, the twisted copper length
is always greater than the linear sheath length. Finally, for long
cables, the incident rectangular pulse is often distorted into a
rough sawtooth shape by the cable itself. Depending upon
whether you sample voltage at the ‘top’ or the ‘bottom’ of this
irregularly-shaped returned pulse, your time delay can vary by
several nanoseconds, which will then correspond to a length
difference of several feet in your result.
Many cable testers have a ‘dead zone’ of about 6 meters (18
feet) where they cannot see any faults. This is because the
incident pulse they use is about 20 nanosecond (ns) long, and
given that signals travel about 3 ns/meter, this means the pulse
hasn’t completely left the tester until it’s about 6 meters long.
Such testers are unable to look for a reflection until the trans-
mitted pulse has completely left. Given that many cable faults
are found in patch cables, wallplates, and other things less than
6 meters from the test point, it’s a good idea to try to find a
TDR without this limitation.
If you are looking for twisted pair faults with a TDR, be sure to
get one that shows you the length of all pairs. Just because one

17
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

pair is 50 meters long doesn’t mean another one isn’t open 3

meters away at the punch block.
A TDR is a powerful tool for finding faults in coax or twisted
pair wiring, and should be the first line of defense for cable
installers or LAN managers against cabling problems.
An equally valid approach for measuring length is to look at
impedance variations in the frequency domain using return loss.
In the mid-90s, some UTP cabling was introduced that con-
tained three pairs designed with a similar delay and a fourth pair
with a widely dissimilar delay. This produced misleading results
when tested in the field, since the length algorithm produced a
result suggesting one pair is significantly shorter or longer than
the other three.

Delay
Delay, or propagation delay, is a measure of the time required
for a signal to propagate from one end of the circuit to the other.
The delay of a pair is directly proportional to its NVP (discussed
under Length). Delay is usually measured in nanoseconds (ns).
Typical delay for category 5 UTP is around 5.7 ns per meter.
International standards dictate that the worst case delay not
exceed 1 microsecond for a 100 meter link.
Delay is the principle reason why there is a length limitation in
LAN cabling. In some networking applications, such as
Gigabit Ethernet, the variation in delay between different pairs of
the same cable (see Delay Skew) can be an important parameter.

Delay Skew
Propagation delay skew (skew) is the difference between the

18
 Microtest, Inc.

propagation delay on the fastest and slowest pairs in a UTP

cable. Due to a shortage of Teflon, which is often used as an
insulation material in category 5 cables, some cable manufac-
tures have substituted other materials on one or two of the four
pairs. Such hybrid cables, while still fully complying with all
category 5 parameters, may exhibit high skew.
Skew is important because several high speed networking
technologies, notably Gigabit Ethernet uses all four pairs in the
UTP cable. If the delay on one or two pairs is significantly
different, then signals which are all sent at the same time from
one end may arrive at significantly different times at the re-
ceiver. While all receivers are designed to allow for some slight
variations in delay, a large skew will make it impossible to
recombine the original signal.
Worst case skew for a 100 meter link should be kept under 50
nanoseconds, though 35 nSec or less is preferable.

Near End Crosstalk (NEXT)

When current flows in a wire, an electromagnetic field is created
which can interfere with signals on adjacent wires. As frequency
increases, this effect becomes stronger. Each pair is twisted because
this allows opposing fields in the wire pair to cancel each other.
The tighter the twist, the more effective the cancellation, and the
higher the data rate supported by the cable. Maintaining this twist
ratio is the single most important factor in any successful UTP installation.
If wires are not tightly twisted, the result is near end crosstalk
(NEXT). If you have ever had a telephone call where you could
hear another conversation faintly in the background, you have
experienced crosstalk. In fact, the name crosstalk derives from
telephony applications where ‘talk’ came ‘across’. In LANs, NEXT
occurs when a strong signal on one pair of wires is picked up by an
adjacent pair of wires. NEXT is the portion of the transmitted

19
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

signal that is electromagnetically coupled back into the received

signal. The receiver may not be able to distinguish between the real
received signal (which having been sent from the far end, is now at
its weakest strength) and the crosstalk noise (see Figure 4).
Weak received signal Strong TX signal

Workstation
Hub

Strong transmitted signal Weak RX signal

Figure 4: Near End Crosstalk

If we allow the electromagnetic radiation created by crosstalk to

get too high we create two problems: not only do we impair
performance in the cabling, but we also create emissions unac-
ceptable to the FCC. This creates a frustrating dilemma. If you
make the transmit levels too strong you won’t be able to meet
FCC requirements, but if you make the signals too weak it will
be difficult to design receivers that avoid errors due to crosstalk
noise.
Note that low NEXT is a high number (e.g. 45 dB). This means
that there is a big difference between the magnitude of a trans-
mitted signal and its associated crosstalk. High NEXT, which is
undesirable, is a low number (e.g. 20 dB) meaning the crosstalk
is more significant relative to the strength of the transmitted
signal.
To meet category 5 standards, all cable connections are allowed
a maximum of 13 mm (0.5 in) of untwisted conductors at the
point of termination.
Common causes of excess NEXT include:
Ø use of untwisted patch cables
Ø poor termination practices
Ø use of old 66 punch down blocks
Ø use of non-datagrade connectors

20
 Microtest, Inc.

Ø use of voice grade cable

Ø use of female to female modular 8 couplers
It is extremely important to measure NEXT from both ends of the
cable or link under test. NEXT is specific to each end of the link,
and it is normal to get significantly different results at each end.
All standards which include NEXT measurements require this.
Figure 5 illustrates the typical shape of a NEXT curve for a
category 5 link.
-25

-35

NEXT -45
(dB)

-55

-65
0 20 40 60 80 100 MHz
Frequency

Figure 5: Near End Crosstalk of a Category 5 Link

Power Sum NEXT

Power sum NEXT (PSNEXT) is actually a calculation, not a
measurement. PSNEXT is derived from an algebraic summation
of the individual NEXT effects on each pair by the other three
pairs. PSNEXT and ELFEXT (discussed below) are important
measurements for qualifying cabling intended to support 4 pair
transmission schemes such as Gigabit Ethernet. There are four
PSNEXT results at each end of the link per link tested.

21
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

FEXT and ELFEXT

Far End Crosstalk (FEXT) is similar to NEXT, except the signal
is sent from the local end and crosstalk is measured at the far
end. FEXT must be measured from both ends.

Figure 6: FEXT

FEXT by itself is not a useful measurement. Why? FEXT is

highly influenced by the length of the cable, since the signal
strength inducing the crosstalk is affected by how much it has
been attenuated from its source. So two links with similar
construction with different lengths will have significantly differ-
ent FEXT. For this reason, Equal Level FEXT or ELFEXT is
measured instead. ELFEXT simply subtracts attenuation from
the result, so that the result is normalized for attenuation
(length) effects. Another way to understand ELFEXT is to
think of far-end ACR, which amounts to the same thing. Fi-
nally, just to make things interesting, we also have power-sum
ELFEXT, or PSELFEXT.
To make accurate ELFEXT measurements requires a field tester
with a dynamic range at least 20 dB lower than the expected
signal level. The Microtest OMNIScanner has a noise floor <
100 dB, which means it can easily make accurate Category 6
ELFEXT measurements.

Attenuation
All electromagnetic signals lose strength as they propagate away

22
 Microtest, Inc.

from their source, and LAN signals over cabling are no excep-
tion. This loss of signal level with distance is called attenuation
loss. Attenuation is measured in decibels (dB). Since it is a loss,
it is usually expressed as a negative value. Thus, -10 dB is a
weaker signal than -8 dB. Decibels are logarithmic, such that if
any two signals are 6 dB different in strength, one is twice the
voltage of the other. Thus, a -10 dB signal has twice the voltage
of a -16 dB signal, and four times the voltage of a -22 dB signal.

Strong signal Signal

Test Set Weak signal
Injector

Figure 7: Cable Attenuation

The two main contributions to attenuation are skin effect and

dielectric loss. When current flows in a wire at high frequencies,
its density is not uniform across the conductor but is concen-
trated at the surface or skin of the conductor. This effectively
reduces the area of the conductor carrying current, and results in
losses which are approximately proportional to the square root
of frequency. This explains why attenuation increases with
frequency, and why solid core cable generally outperforms
stranded cable.
Temperature also affects attenuation in some cables. The
dielectric materials which form the conductor insulation and
cable jacket absorb some of the transmitted signal as it propa-
gates along the wire. This is especially true in cables containing
PVC such as many category 3 cables. PVC contains a chlorine
atom which is electrically active and forms dipoles in the insu-
lating materials. These dipoles oscillate in response to the
electromagnetic fields surrounding the wires, and the more they
vibrate, the more energy is lost from the signal. Temperature
increases exacerbate the problem, because they make it easier
for the dipoles to vibrate in the insulation. This results in in-
creasing loss with temperature.

23
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

That is why standards bodies tend to specify attenuation re-

quirements adjusted for 20°C (68°F). If your cables are likely to
be operating in much higher temperatures, check with your cable
supplier to choose a brand which is designed to be more tolerant
of a warmer environment.
Figure 7 indicates the typical shape of an attenuation curve for a
category 5 link. Note that in contrast to NEXT, the attenuation
characteristics of a cable are typically quite smooth and well
behaved.
8

(db) 4

0
1 10 100 Mhz
Frequency

Figure 7: Attenuation of a Category 5 Link

To measure attenuation accurately, it is essential that you make
a one-way measurement, rather than a loopback measurement
and divide the total loss by two. See the Troubleshooting
and Certification Tools section for more information.

Attenuation to Crosstalk Ratio (ACR)

Due to the effects of attenuation, signals are at their weakest at
the receiver end of the link. But this is also where NEXT is the
strongest. Signals that survive attenuation must not get lost due

24
 Microtest, Inc.

to the effects of NEXT.

In Unshielded Twisted Pair (UTP) networks, attenuation deter-
mines the strength of the received signal, while the noise is
primarily the NEXT from the station’s own transmitter. Attenu-
ation-to-Crosstalk Ratio (ACR) is the difference between the
crosstalk loss and the attenuation for the pair in the link being
tested. ACR is a figure of merit for the cable. In essence, it is a
measure of how much ‘overhead’ you have between the size of
the signal at the receiver, and the size of the crosstalk noise.
Thus, the greater the ACR, the better. ACR is specified in ISO
and IEEE standards, but not the current draft of EIA 568A,
though it can be explicitly derived from NEXT and attenuation
requirements.
Note that because NEXT is different at each end, ACR results
will also be different at each end. The worst case ACR results
must be used.
Using PSNEXT and attenuation, power sum ACR (PSACR) can
also be calculated.

Bandwidth
Bandwidth is a measure of the usable frequency range of the link
under test. For most networks, this is the highest frequency at
which the signal can easily be distinguished from background
noise. Bandwidth can be determined by finding the lowest
frequency at which the worst case pair-to-pair NEXT comes
within 3 dB of the corresponding attenuation. See figure 8
below.

25
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

0
Attenuation (dB)

-10

-20
3 dB ACR
-30

-40

-50

-60
NEXT (dB)

-70

-80
0 10 20 30 40 50 60 70 80 90 100
Frequency (MHz)

Figure 8: Bandwidth

Wire Map
Wiremap is used to identify installation wiring errors. For each
of the 8 conductors in the cable, wire map indicates:
Ø proper pin termination at each end
Ø continuity to the remote end
Ø shorts between any two or more conductors
Ø crossed pairs
Ø reversed pairs
Ø split pairs
Ø any other miswires
Correct connectivity is illustrated in Figure 9:

26
 Microtest, Inc.

Correct Connectivity
1 1
2 2
3 3
6 6
5 5
4 4
7 7
8 8

Figure 9: Correct connectivity

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). A
crossed pair occurs when the two conductors in a wire pair are
connected to the position for a different pair at the remote
connector. Split pairs occur when pin to pin continuity is main-
tained but physical pairs are separated. Refer to Figure 10 for an
illustration of a reversed pair, crossed pairs, and split pairs.

Reversed Pair Crossed Pairs Split Pairs

1 1 1 1 1 1
2 2 2 2 2 2
3 3 3 3 3 3
6 6 6 6 6 6
5 5 5 5 5 5
4 4 4 4 4 4
7 7 7 7 7 7
8 8 8 8 8 8

Figure 10: Reversed Pairs, Crossed Pairs and Split Pairs

The one exception to watch for when doing a wire map is

Copper Data Distributed Interface (CDDI). CDDI’s normal
wiring is intentionally nonstandard and will ‘fail’ a wire map test.
See page 42 for more information.
Note: In this handbook, we will refer to pairs by their associated
pin numbers (example, pair 12). This is because several different
accepted standards exist for pair nomenclature. See Appendix

27
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

2: A Guide to Modular Connector Wiring for more

information.

Impedance
Impedance is a measure, in ohms, of the opposition to the flow
of current in a cable. It includes the effects of resistance, induc-
tance, and capacitance. UTP is commonly rated at either 100Ω
or 120Ω, and this value must be constant over the operating
bandwidth of the cable.
Impedance can be measured in the time domain or the frequency
domain. There is generally a good correlation between these
measurements above 10 MHz. Impedance of the cable itself is
usually uniform and is not affected by the installation. Imped-
ance mismatches caused by poor installation techniques are best
found with TDR techniques (see the Length discussion start-
ing on page 12).

Structural Return Loss/Return Loss

Structural Return Loss (SRL) is a measure of the uniformity of
the cable’s impedance. Cables are not perfectly uniform in
construction, and these variances cause slight impedance
changes in the cable. Each impedance change contributes to loss
in signal strength. Structural Return Loss is a function of the
cable’s design and manufacture, and unlike NEXT, is less
affected by installation practices. SRL is expressed in dB, and
the higher the number, the better the cable.
The Return Loss of a link is a measure of the overall uniformity
of the link’s impedance, relative to a nominal value (usually 100
ohms in the case of UTP). Just because the cable in the hori-
zontal span has good SRL does not necessarily mean the link
will have good return loss. Return loss in the link will have

28
 Microtest, Inc.

deleterious effects on the measurement accuracy of any field

tester that uses the modular 8 connector, which is one reason
such devices can achieve no better than TSB-67 Level I accu-
racy.
Manufacturers of very high quality UTP have taken special care
to ensure impedance is very uniform throughout the link, and
also that all components are very well matched. So while return
loss wasn’t a big issue when Category 5 cabling first appeared, it
is an important differentiator for Cat 5E and Cat 6.

DC Loop Resistance
DC loop resistance is a measure of the linear DC loss of a
conductor. This is usually a function of the cable design and
rarely varies. This measurement is usually done to ensure there
are no gross misconnections which add significant resistance to
the link. DC loss for category 5 cable should not exceed 9.4Ω
per 100 m.

Noise
Noise is undesirable electrical activity in a conductor. Noise
sources in LAN environments typically include AC transmission
lines, fluorescent light fixtures, industrial equipment, and eleva-
tors. Noise problems also occur when other wire pairs in the
same sheath are used for non-LAN applications.
Testing for noise involves turning off all LAN devices and
measuring for background electrical activity in the cable. Often,
noise problems can be corrected by moving the cable away from
the source or removing the source, if possible. Different noise
sources radiate at different frequencies, so if your tester is
capable of telling you the frequency of the noise source, that can
assist you in pinpointing the cause.

29
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

The two common sources of noise are Electromagnetic Interfer-

ence (EMI) and Radio Frequency Interference (RFI). EMI is
characterized by lower frequencies and higher amplitudes, and
comes from sources such as motors, fluorescent light fixtures,
and AC power lines. RFI is characterized by higher frequencies
and lower amplitudes, and comes from sources such as cellular
phones, radio, TV, and high frequency switching power supplies.
Fortunately, UTP LANs are surprisingly resistant to noise (other
than NEXT). For this reason, neither IEC 11801 nor EIA/TIA
568A has seen the need to provide noise measurement require-
ments in the standards. While cable standards do not require
noise measurements, some high speed applications standards do.
For example, 100VG-AnyLAN has an impulse noise require-
ment. Please refer to the Network-Specific Cabling
Requirements section for more information.

Fiber Optic Measurements

In testing fiber optic LANs, there are two main classes of test
instruments: Optical Loss Test Sets and OTDRs. Optical Loss
Test Sets include optical sources and power meters. They are
used to measure the end-to-end loss through a fiber link. They
are also used to measure the strength of transmitted power, the
power of the received signal, or the link loss. Optical Loss Test
Sets are relatively inexpensive and by far the most commonly
used tool of the fiber optic installer.
The Optical Time Domain Reflectometer or OTDR, is a much
more sophisticated instrument which is capable of locating
breaks in the fiber, identifying poor connections, and measuring
fiber length. It is commonly used in the single mode fiber world,
where the fibers are typically much longer and include many
more connections in the link. OTDRs can also be used to
measure end-to-end loss, although less accurately than a power

30
 Microtest, Inc.

meter. They are rarely used in the multimode fiber LAN world
except to measure length, because short fiber LANs have few
connections (usually 4 or less) and are easy to troubleshoot.
OTDR’s are also difficult to operate and cost 5 – 10 times as
much as a power meter – light source combination.
An alternative to the OTDR is the Fault Finder. These are
essentially smaller versions of OTDR’s with reduced functional-
ity, speed, accuracy, and price. The primary purpose of a Fault
Finder is to find the exact location of a fault on CATV fiber or
large, campus LAN fiber cabling.
A new classification in fiber optic test tools is the Certification
Tool. A fiber certifier is a tool that automates the test process
and compares results to predefined standards. For example, the
Microtest CertiFiber has an Autotest function that performs the
following tests: measures bidirectional power loss at both 850
and 1300 nm on two fibers, measures the length and propaga-
tion delay of the link, then compares all these results to industry
standards such as TIA 568A, IS11801, FDDI, 10BaseF, Fiber
Channel, and others. The operator then gets an overall pass/fail
for the link.
Typical fiber measurements include attenuation and length. An
important requirement when testing fiber cabling is to know the
optical link budget, or OLB. The OLB is maximum allowable
loss for the circuit or application. It alone allows you to know
whether or not your application will run or whether the installa-
tion has been properly installed.
As an example, consider an 88 meter multimode fiber link to be
tested at 850 nm. What is the Optical Link Budget (or in other
words, how do I know if it’s any good)?

31
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

88 meters

Splice
Mated Mated
connection connection

Figure 11: Typical Fiber link

TIA 568A allows: 0.75 dB per mated connection, + 0.3 dB per

splice, + 3.75 dB per km @ 850 nm
In this example, we have:
• 4 Connections @ 0.75 = 3 dB
• 1 splice @ 0.3 dB = 0.3 dB
• 0.088 km x 3.75 dB/km = 0.33 dB
Optical Link Budget = 3 + 0.3 + 0.33 = 3.63 dB @ 850 nm
Most installers don’t do this for each link and every wavelength.
The problem: they don’t know where ‘PASS’ is! This is a real
problem. An installer in the field who tests two fibers, and finds
both were 6.2 dB, would probably assume both were OK. If
one has only two mated connections, no splices, and is short,
while the other has 4 mated connections and is long, “6.2 dB”
represents a PASS for the first and FAIL for the second. This is
why it is important to have the ability for your field tester to
automatically calculate the OLB for you when testing fiber
cabling.

32
 Microtest, Inc.

Cabling Standards
Several different standards organizations influence data cabling.
The most important of these are:

Organization Specific Subcommittee Key Area of Influence

EIA/TIA TR41.8.1 North America
CENELEC TC 215 WG1 Europe
ISO/IEC ISO/IEC JTC1 SC25 WG3 Worldwide
IEEE 802.3x Worldwide
ATM Forum PHY Worldwide
Table 2: Cable Standards

In addition, there are industry associations of cabling profession-

als that have growing amounts of influence in this industry. The
largest of these is BICSI (Building Industry Consulting Service
International) based in Tampa, Florida.
As this handbook is being written, there is much activity hap-
pening in the cabling standards area. Let’s first examine pub-
lished cabling standards, then look at what’s ahead.

33
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

Current Published Standards

Standard Description Scope
TIA TSB-67 Specifies field test requirements North
for installed UTP America
TIA 568A Commercial Building Tele- North
communications Standard America
ISO/IEC IS11801 Generic Cabling for Worldwide
Customer Premises
CENELEC EN 50173 Customer Premises Cabling Europe
for Information Technology
(very similar to IS11801)

Table 3: Published Standards

These standards define and specify performance of cabling links

up to 100 MHz. TIA link performance standards are specified
as Categories. International specifications use the Category
designation for components and cables, but use the term Class
for link performance. The link naming protocol and bandwidth
of current and proposed cabling is shown in tables 4 and 5.

Standard TIA ISO/IEC CENELEC

Connectors Category Category Category
Cables Category Category Category
Links Category Class Class

Table 4: Link Naming Protocol

34
 Microtest, Inc.

Standard TIA ISO/IEC CENELEC

16 MHz Links Category 3 Class C Class C
100 MHz Links Category 5 Class D Class D
200 MHz Links Category 6 Class E Class E
600 MHz Links Category 7 Class F Class F

Table 5: Cabling Bandwidth, Categories, and Classes

Link Definitions
In all standards, there are two principal link definitions. First is
the channel, which includes every cabling element necessary to
pass data from a device at one end of the link to a device at the
other end. The permanent or basic link, described in detail
below, is essentially the subset of the channel that excludes the
user patch cords at either end.
Figures 12a and 12b show the models for the channel, the
permanent link and the basic link as described in current stan-
dards.
The channel comprises no more than 100 m of horizontal
cabling and four connections.
The permanent link is the part of the cabling link from the
telecommunication outlet via the horizontal cable to the first
connection of the cross-connect in the floor distributor. The
permanent link may optionally incorporate a transition point. A
complete cross-connect is not part of the permanent link.

35
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

Channel specification

Permanent link1)

Horizontal cabling
(balanced pair)
EQP C C C C C TE

equipment patch TP TO work area

FD cable3) cord2),3) cable3)
Cabling link

NOTE 1: 90 meters mechanical length maximum

NOTE 2: 5 meters mechanical length, electrically
equivalent to 7.5 m of horizontal cabling.
NOTE 2: Combined 10 m mechanical, 15 m electrical.

Figure 12a: Model of the channel, cabling link

and permanent link

Permanent link

Horizontal cabling
(balanced pair)
FT C C C C FT

test TP TO test
FD cord cord
Basic link

Figure 12b: The “basic link” test configuration

for a permanent link

In order to measure certain electrical performance characteristics

for the permanent link it is necessary to attach test cords to the
permanent link. Is desirable though not necessarily practical to
attempt to eliminate the effects of the test cord on the perma-
nent link. At higher frequencies and low signal levels the
interaction between test cords is unique for each link tested.
For this reason a “basic link” configuration is defined which
reflects the modified acceptance criteria as shown in Figure 12b.

36
 Microtest, Inc.

Cabling Standards in Development

A number of cabling standards are currently in development.
These are being driven by two issues. First, the Gigabit Ether-
net specification requires the specification of return loss and
ELFEXT, which are not specified completely in any published
cabling standard. Second, there is a desire to specify cabling
performance to higher levels, namely Enhanced Category 5 (Cat
5E), Category 6, and Category 7.
The following is a summary of the description and current status
of active cabling standards work. Note: while correct for May
1998, this information is changing rapidly. Please check with
your preferred cabling supplier for up to date information if
required.

TIA 568A Addendum 2

Additional Transmission Performance Specifications
for UTP
This standard is being written because it was found that some
new LAN technologies, notably Gigabit Ethernet, require
performance parameters to be specified that were not specified
in TIA 568A. The four additional parameters will be:
• Return Loss
• ELFEXT (Equal Level Far End Crosstalk)
• Propagation Delay
• Delay Skew
It is expected that most existing legacy Category 5 channels will
meet the new published requirements, but it is not assured. This
document has gone to ballot and is expected to be approved in
the late summer of 1998.

37
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

TIA 568A Addendum 3

Additional Transmission Performance Specifications
for Enhanced Category 5 Cabling
This standard recognizes advances in cabling technology,
defines additional measurements, and provides much stricter
performance limits to 100 MHz. The following additional
measurements are to be specified:
• Power Sum NEXT
• Return Loss
• Worst Pair-to-Pair ELFEXT
• Power Sum ELFEXT
• Propagation Delay
• Delay Skew
This document needs further development and will likely be
approved in the second half of 1998. It is important to note that
international committees do not recognize CAT5E.

TIA PN#3727
Enhanced Twisted Pair Cabling Systems
This study group was originally tasked with looking at higher
bandwidth performance specifications for UTP (up to 300
MHz). The members of this group were until recently fully
engaged on the prior two work items, and have essentially been
overtaken by international events (see IS11801:2nd Edition
information below). This work has recently become reactivated,
and it is expected to closely parallel international cabling stan-
dards work, so that a harmonized result is obtained.
The focus of this group will be on Category 6 specifications.
Due to a request from the IEEE, while Category 6 will provide
a positive ACR at 200 MHz, it is to be specified to 250 MHz.

38
 Microtest, Inc.

This is because advanced network interface cards employ

NEXT cancellation technology that allows them to use band-
width past the zero-ACR point. This group has a Category 6
draft published, and will work quickly to reach agreement on
Category 6 link specifications.

Amendment to EN 50173:1998
This work is being done in the CENELEC TC215 committee.
It adds permanent link and channel definitions, and for each
type of link gives performance specifications in Classes A-D
(where applicable) for the following measurements:
• Length • NEXT
· • Nominal Impedance • ACR
• Return Loss • Propagation Delay
• Attenuation • Delay Skew
Additional measurements (balance, ELFEXT, and others) are
under consideration. This document is expected to be ratified in
late 1998.

EN 50173:2000
This is a complete redraft of EN 50173. Only a skeletal outline
exists at present. Significant work will begin after 50173:1998 is
substantially completed.

EN 50xxx: Cabling Test Methods

This provides a European field-testing standard specifying the
methods required for the testing of installed balanced copper
and optical fiber cabling. It will include a full treatment of all
EN 50173 measurements and performance requirements.

39
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

40
 Microtest, Inc.

LAN Application and Standard Performance

Summaries

Introduction
This section provides a summary of performance specifications
for:
• TIA Basic Links
• TIA Channels
• ISO/IEC/CENELEC Channels
• ISO/IEC/CENELEC Permanent Links
• Copper LAN Application vs. Category/Class matrix
• Fiber LAN Application Performance Requirements
Special note: While every effort has been made to ensure the
accuracy of this information, standards and LAN application
requirements are changing at a rapid rate. We strongly urge you
to check with your networking vendor, cable supplier, or test
equipment supplier for up to date information.

41
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

Overview of Categories and Classes

Category 3 Category 5 Category 5E Category 6 Category 7

Class C Class D Class E Class F
Bandwidth 16 MHz 100MHz 100MHz 200MHz 600MHz
Cable type UTP UTP/FTP UTP/FTP UTP/FTP SSTP
Connector Mod 8 Mod 8 Mod 8 Enhanced To be
type Mod 8 specified
Standard ü ü Q3 98 Q1 99 Q1 99
status
Link Cost 0.7 1 1.2 1.5 2.2
(Cat 5 =1)

Table 6: Categories and Classes

TIA
Basic Link Performance Summary

Horizontal cabling
(balanced pair)
FT C C C C FT

test TP TO test
cord cord
Basic link

Figure 14: Basic Link

Length < 94 meters including test equipment cords

Wiremap correct pin-to-pin connectivity (no opens, shorts,
crossed, reversed, or split pairs)
See Table 7 for Basic Performance data across Categories 3-6.

42
 Microtest, Inc.

Category 3 Category 5 Category 5E* Category 6

Propagation < 548 nS < 548 nS < 548 nS
Delay
Delay Skew < 45 nS < 45 nS <45 nS
Attenuation (dB)
@ 1 MHz 3.2 2.1 2.1 2.1
@ 10 MHz 10.0 6.3 6.3 6.2
@ 100 MHz 21.6 21.6 20.7
@ 200 MHz 30.4
NEXT (dB)
@ 1 MHz 40.1 60.0 64 73.5
@ 10 MHz 24.3 45.5 49 57.8
@ 100 MHz 29.3 32.3 41.9
@ 200 MHz 36.9
PSNEXT (dB)
@ 1 MHz 60 71.2
@ 10 MHz 45.5 55.5
@ 100 MHz 29.3 39.3
@ 200 MHz 34.3
ELFEXT (dB)
@ 1 MHz 57 61 65.2
@ 10 MHz 37 41 45.2
@ 100 MHz 17 21 25.2
@ 200 MHz 19.2
PSELFEXT (dB)
@ 1 MHz 54.4 58 62.2
@ 10 MHz 34.4 38 42.2
@ 100 MHz 14.4 18 22.2
@ 200 MHz 16.2
Return Loss (dB)
1 – 20 MHz N/A 15 17 19
20 – 100 MHz 15-10log(f/20) 17-7log(f/20) 19-10log(f/20)
100 – 200 MHz 19-10log(f/20)

* TIA 568A Addendum 3, March 1998

TABLE 7: Basic Link Performance

43
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

TIA/ISO/IEC/CENELEC
Channel Performance Summary

Channel specification

Horizontal cabling
(balanced pair)
EQP C C C C C TE

equipment patch TP TO work area

cable cord cable

Figure 15: Channel

Length < 100 meters including test equipment cords

Wiremap correct pin-to-pin connectivity (no opens, shorts,
crossed, reversed, or split pairs)
See Table 8 for Channel Performance data across CAT3/ClassC
- CAT6/Class E.

44
 Microtest, Inc.

Cat3/Class C Cat5/Class D Category 5E Cat6/Class E1

Propagation < 548 nS < 548 nS < 548 nS
Delay
Delay Skew < 45 nS < 45 nS <45 nS
Attenuation (dB)
@ 1 MHz 4.2 2.5 2.1 2.2
@ 10 MHz 11.5 7.0 6.3 6.4
@ 100 MHz 24.0 21.6 21.6
@ 200 MHz 31.8
NEXT (dB)
@ 1 MHz 39.1 60.3 63 72.7
@ 10 MHz 22.7 44.0 47 56.6
@ 100 MHz 27.1 30 39.9
@ 200 MHz 34.8
PSNEXT (dB)
@ 1 MHz 60 71.2
@ 10 MHz 44 54.0
@ 100 MHz 27 37.1
@ 200 MHz 31.9
ELFEXT (dB)
@ 1 MHz 57 59 63.2
@ 10 MHz 37 39 43.2
@ 100 MHz 17 19 23.2
@ 200 MHz 17.2
PSELFEXT (dB)
@ 1 MHz 54.4 56 60.2
@ 10 MHz 34.4 36 40.2
@ 100 MHz 14.4 16 20.2
@ 200 MHz 14.2
Return Loss (dB)
1 – 20 MHz 15 15 17 19
20 – 100 MHz 15-10log(f/20) 17-7log(f/20) 19-10log(f/20)
100 – 200 MHz 19-10log(f/20)

1
based on March20, 1998 draft working documents

Table 8: Channel Performance

45
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

ISO/IEC/CENELEC
Permanent Link Performance Summary

Permanent link

Horizontal cabling
(balanced pair)
FT C C C C FT

test TP TO test
cord cord

Figure 16: Permanent Link

Length < 90 meters excluding test equipment cords

Wiremap correct pin-to-pin connectivity (no opens, shorts,
crossed, reversed, or split pairs)
See Table 9 for Permanent Link Performance data across
ClassC-F.

46
 Microtest, Inc.

Class C1 Class D1 Class E2 Class F2

Propagation < 504 nS < 504 nS < 504 nS
Delay
Delay Skew < 43 nS < 43 nS <43 nS
Attenuation (dB)
@ 1 MHz 3.1 2.1 1.7 2.0
@ 10 MHz 9.2 6.1 5.5 5.6
@ 100 MHz 20.6 18.9 17.5
@ 200 MHz 27.9 45.8@600 MHz
NEXT (dB)
@ 1 MHz 40.1 61.2 73.4 74
@ 10 MHz 24.3 45.5 57.8 74
@ 100 MHz 29.3 41.8 65.1
@ 200 MHz 36.9 54@600 MHz
PSNEXT (dB)
@ 1 MHz 71.2
@ 10 MHz 55.5
@ 100 MHz 39.3
@ 200 MHz 34.3
ACR
@ 1 MHz 37.1 58.9 71.7
@ 10 MHz 14.8 39.1 52.3
@ 100 MHz 8.5 23
@ 200 MHz 9.1
ELFEXT (dB)
@ 1 MHz 65.2
@ 10 MHz 45.2
@ 100 MHz 25.2
@ 200 MHz 19.2
Return Loss (dB)
1 – 10 MHz 15 15 19 19
10 – 20 MHz 15 19 19
20 – 100 MHz 15-7log(f/20) 19-7log(f/20) 19-7log(f/20)
100 – 200 MHz 19-7log(f/20) 19-7log(f/20)
1
based on ISO/IEC TR 14763-4 11-25-97
2
based on March 20, 1998 draft working documents

Table 9: Permanent Link Performance

47
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

Technology Application Matrix

All LAN applications are designed to be supported on UTP
using the modular 8 (RJ45) connector. The following table
summarizes the pin assignments and cabling class necessary to
support each application.

Application Pins 1 & 2 Pins 3 & 6 Pins 4 & 5 Pins 7 & 8

10Base-T Cat 3/Class C Cat 3/Class C
Token Ring-4 Cat 3/Class C Cat 3/Class C
Token Ring-16 Cat 3/Class C Cat 3/Class C
TPPMD Cat 5/Class D Cat 5/Class D
100Base-T2 Cat 3/Class C Cat 3/Class C
100Base-TX Cat 5/Class D Cat 5/Class D
100Base-T4 Cat 3/Class C Cat 3/Class C Cat 3/Class C Cat 3/Class C
VGAnyLAN Cat 3/Class C Cat 3/Class C Cat 3/Class C Cat 3/Class C
ATM-25 Cat 3/Class C Cat 3/Class C
ATM-155 Cat 3 Cat 3/Class C Cat 3/Class C
ATM- 155 Cat 5 Cat 5/Class D Cat 5/Class D
1000Base-T Cat 5E Cat 5E

Table 10: Technology Application Matrix

48
 Microtest, Inc.

Fiber LAN Application Performance Requirements

The following matrix provides a summary of all popular LAN
applications that run on multimode fiber cabling, with their
corresponding performance requirements. OLB is the optical
link budget, or maximum allowable loss.
Standard 850 nm 1300 nm Max. Max. Reference
OLB OLB Delay Length
TIA 568A Formula 1 Formula 2 N/A 2000 m TIA 568A
IS 11801 Formula 3 Formula 4 N/A 1500m IS 11801
ATM155 -- 7.0 dB -- 2000m ATM UNI V3.0
ATM155SWL 9.0dB -- -- -- ATM UNI V3.0
ATM622 -- 6.0 dB -- 500 m Af-phy-0046.000
FDDI -- 11.0 dB -- -- ANSI X3T9.5
10BASE-FL 12.5dB -- 10,000ns 2000m IEEE 802.3J
(1) 10BASE-F 15.3.2.2
10BASE-FB 12.5dB -- 10,000ns 2000m IEEE 802.3J
(2) 10BASE-F 15.3.2.2
100BASE-F -- Formula 4 -- 2000m IEEE802.12- 1995
(3) sec.18-10
1000BASE-SX 7 dB -- -- 550 m Draft IEEE
50 micron 802.3z draft 2
1000BASE-LX - 7.5 dB -- 550 m Draft IEEE
50 micron 802.3z draft 2
1000BASE-SX 7 dB -- -- 220 m Draft IEEE
62.5 micron 802.3z draft 2
1000BASE-LX -- 7.5 dB -- 550 m Draft IEEE
62.5 micron 802.3z draft 2

Token Ring 13.0 dB -- -- -- IEEE 802.5

Fiber Channel -- 6.0 -- 175m ANSI X3T11/
(5) 755D/Rev 4.3

Table 11: Fiber Performance Requirements

49
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

Notes:
1. 10BASE-FL is a system for interconnecting repeaters or DTEs with up to 2 km per
segment and where backwards compatibility with FOIRL is desirable.
2. 10BASE-FB is a system for interconnecting repeaters with up to 2 km per segment
and for backbone applications where a large number of repeaters must be cascaded.
3. The 100BASE-F standard explicitly states fiber optic cabling must meet the
specifications of IS11801:1995 at 1300 nm for distances up to 2000m.
4. Fiber Channel numbers assume a 100 Mbps implementation with four mated
connections.

Standard Optical Link Budget Formula

TIA 568A (850 nm) OLB (dB) = (# mated connections)(0.75) +
Formula 1 (# splices)(0.3) + (length)(3.75 dB/1000m)
TIA 568A (1300 nm) OLB (dB) = (# mated connections)(0.75) +
Formula 2 (# splices)(0.3) + (length)(1.5 dB/1000m)
IS 11801 (850 nm) OLB (dB) = (# mated connections)(0.75) +
Formula 3 (# splices)(0.3) + (length)(3.5 dB/1000m)
IS 11801 (1300 nm) OLB (dB) = (# mated connections)(0.75) +
Formula 4 (# splices)(0.3) + (length)(1.0 dB/1000m)

Table 12: OLB Formulas

50
 Microtest, Inc.

Troubleshooting & Certification Tools

The Right Tool for the Right Job

Cable troubleshooting tools fall into four broad classes with
increasing price and capability: continuity testers, cable testers,
handheld certification tools, and network analyzers.
Many problems are caused in troubleshooting and conformance
applications due to the improper use of test equipment. It’s not
a good idea to try to troubleshoot CAT 5 UTP with a toner, nor
is it intelligent to try to compare the absolute dynamic range of
two different CAT 6 cables using a handheld cable tester.
So the first question to answer is, what is your objective?

If you need to: The right tool for the job is a: Example:
Make specialized Toner, DMM, or basic Progressive
measurements on continuity tester Electronics 200 EP
data cabling in a lab
environment
Diagnose faults on Cable tester Microtest
a UTP, STP or MICROSCANNER
coaxial cable
Certify that an in- Handheld certification Microtest
stalled UTP segment OMNIScanner
meets specific net-
work or link standards
Compare the Network analyzer HP 4195A
performance of raw
Category 6 cabling

Table 13: Choosing the Appropriate Cable Test Tool

51
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

Network Analyzers
The standard by which all cables are ultimately measured is the
network analyzer, such as the HP4195A. This device contains a
highly accurate radio frequency source and a narrowband track-
ing receiver. By varying the network analyzer output frequency
over a wide range onto the transmit wire pair and using the
tracking receiver to measure the signal on the receive wire pair,
NEXT and attenuation measurements can be made. The
narrowband receiver approach allows the network analyzer to
focus on receiving only the signal that it is using to make the
measurements. That way, any noise present during the measure-
ment will be rejected.
The bandwidth of this receiver can be adjusted to allow the
speed and resolution of the measurement to be varied - the
narrower the bandwidth (higher resolution) the longer the
measurement takes. The converse is also true - the wider the
bandwidth (lower resolution) the faster the measurement goes.
The user has to make the trade-off between speed and resolution.
One of the difficulties in making the measurement is that the
unbalanced 50 ohm outputs and inputs of the network analyzer
must be transformed with balun type transformers to match the
100 ohm input of the cable. These baluns must be of sufficient
bandwidth and balance so as not to affect the measurement.
Also, since the range of signals to be measured can vary by as
much as 1000 to 1 (60 dB), the network analyzer must be
capable of measuring at least 10 dB better than the expected 60
dB range in order for the errors induced by the analyzer to be
minimized; generally more than 10 dB is desirable.
The network analyzer is a large (60+ pounds), expensive
($20,000+), 120 volt AC-powered instrument that contains
multiple RF generators, multiple RF receivers, a CRT display for
review of results and disk storage for retrieval of setups and
data. In addition to the network analyzer, additional test equip-

52
 Microtest, Inc.

ment and balun transformers may have to be attached to make

the measurement connections to the cable to be tested.
Operating the network analyzer is not for the faint of heart.
Since the network analyzers in use today are general purpose lab
instruments, they contain many more analytical features and
have higher accuracy’s than are necessary for the measurement
of typical network cable parameters. This sophistication tends to
make the units cumbersome to set up and operate in the field.
Also, due to the generality of the devices, a price is paid both in
dollars and in the presentation of the data – network analyzers
do not judge the data they display.
Network analyzers are intended for use in laboratory environ-
ments by highly trained technical personnel who are capable of
creating their own custom interface equipment. They are typi-
cally used to evaluate the relative performance of unterminated
datagrade cabling.

Handheld Certification Tools

Unlike the network analyzer, certification tools are intended for use
by unsophisticated users, on a budget, in a field environment. This
has major design implications which include size, weight, reliability,
power requirements, automation of testing, speed, ability to judge
results, ease of connectivity, and ease of use.
These products are typically used to verify that installed links
meet TIA Category 5 or 6, or ISO Class D or E requirements.
They are also ideal for performing physical layer troubleshooting
on LANs.

A leading device in this product category is the Microtest

OMNIScanner®. This device resembles an oversize mobile phone
and will all the same functions that a network analyzer would
need to measure and certify cable links for category 6 installa-

53
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

tions. The OMNIScanner® represents a significant advance over

the second generation Category 5 cable testers that most install-
ers owned in 1997. For perspective on the differences achieved,
see table 15.

Typical Laboratory Typical Second Third Generation

Network Analyzers Generation OmniScanner®
Field Tester
Bandwidth 500 MHz 100 – 155 MHz 300+ MHz
Dynamic range > 110 dB 65 dB >100 dB
Highest performance Category 7 Category 5 Category 7
cabling accurately tested
Measurement Domain Frequency Frequency Frequency
Architecture Vector based Scalar based Vector based
(magnitude and phase) (magnitude only) (magnitude and phase)
Measurement channels 1 (+ 2 external baluns 4 4
required for UTP tests)
Time for TSB-67 Autotest Hours ~20 seconds ~ 8 seconds
Powersum No No Yes
measurements
Accurate Return Loss Yes No Yes
measurements
ELFEXT measurements No No Yes
Project-based No No Yes
Certification
Cabling supported UTP, FTP, Cat 3-5; coax, Cat 3-7; coax, multi-
SSTP, coax multimode fiber and single mode fiber
Component-level testing Yes No Yes
Expertise required Electrical Engineer Novice installer Novice installer
Data interpretation None Yes Yes - advanced
Size Benchtop Handheld Handheld
Price $30,000+ $4000 $5500

Table 14: Certification Tools

54
 Microtest, Inc.

Figure 16: The Microtest OMNIScanner

The OMNIScanner® makes all necessary measurements for

certification and troubleshooting all types of copper and fiber
cabling, including length, Wiremap, attenuation, NEXT, PSN-
EXT, ELFEXT, PSELFEXT, return loss, resistance, ACR, and
PSACR. It also includes advanced diagnostics for troubleshoot-
ing the cause and location of any cabling challenges.
Like a network analyzer, the OMNIScanner® is a narrowband,
frequency-domain, DSP-based full vector design. However,
unlike the network analyzer, it is a four-channel design, and
includes the high frequency baluns necessary for UTP tests.
The purpose of the narrow-band receiver is to filter out and
minimize any noise or interference that may be on the receive
pair while the NEXT measurement is being made. Noise that

55
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

typically is impressed on the cable wired inside the walls and

ceilings of a building is made up of radio interference from
strong radio stations, switching noise and its harmonics from
fluorescent light fixtures or dimmers, and radiated or conducted
noise from wiring that may be adjacent to the network cable.
This interference, if left unfiltered, would cause the receiver to
“see” more signal than was actually being coupled from the
transmit wire pair. The result of receiving this noise would be an
erroneous determination of the near end crosstalk.
Another important benefit of the narrow-band receiver is that
NEXT measurements are not affected by the wave shape
(harmonic distortion) of the transmit frequency. The tunable
narrow-band receiver acts much like a very narrow bandpass
filter that only responds to the fundamental frequency compo-
nent of the transmitted signal. Near end crosstalk produced from
the higher order harmonics of the transmitter signal are rejected
by the narrow-band filtering in the receiver. Consequently, a true
NEXT versus fundamental frequency measurement is made.
As an illustration of the bandwidth and accuracy of the
OMNIScanner®, figure 17 shows the results of a high perfor-
mance cable tested to 300 MHz by both the HP 4195A network
analyzer and the Microtest OMNIScanner®.

56
 Microtest, Inc.

-30

-40
NEXT (dB)

-50

-60

-70
Network Analyzer NEXT
OmniScanner NEXT

-80
0 50 100 150 200 250 300
Frequency (MHz)

Figure 17: NEXT Comparison of Network Analyzer and OMNIScanner

The level of agreement shown is as good as would be obtained

with any two network analyzers. This shows that the
OMNIScanner® is capable of making highly accurate measure-
ments, on all types of cabling, to frequencies well beyond the
proposed Category 6 maximum limit.

57
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

Cable Testers
This class of tool can verify end-to-end connectivity, perform a
wire map, measure length via TDR, find split pairs and other
miswires, and has the intelligence to interpret most results.
Prices are a bargain, at under $500. Network downtime often
exceeds $500 per minute, so these tools are an invaluable
investment for IT managers. They are extremely easy to oper-
ate, lightweight, and practically indestructible. A good example
is the Microtest MICROSCANNER. These products are gener-
ally available from VARs and large networking products dis-
tributors.

Continuity Testers
The simplest cable troubleshooting tools are toners and continu-
ity checkers. These are generally less than $200 and verify end
to end connectivity, continuity, and can check for crossed pairs.
Common manufacturers are Paladin, Progressive Electronics,
and the Siemon Company. These are useful tools, but are usually
insufficient to diagnose problems on LANs. They are available
from most popular catalog supply companies.

58
 Microtest, Inc.

Link Troubleshooting Guide

In this guide, tables are provided which suggest possible causes
for ‘failed’ links. When you refer to the measurement that failed
in your test, you will see several possible causes for the failure.
For each cause, one or more solutions to correct the cause are
provided.

Solving Wiremap Problems

Possible Wire Map Fault Solution

Crossed pins: 1236 Unless it is an unusual application requiring crossed wires,

< 2136 rewire the connection. Check Appendix 2 for correct color
coding of horizontal cable and patch cord wires at termination
points.

Expecting: 1236 2-pair Token Ring cable is being used for a 10BASE-T
application.
Actual: 3645
Change the cable or rewire connectors.

Expecting: 3645 2-pair 10BASE-T cable used for Token Ring application.

Actual: 1236 Change the cable or rewire connectors.

Wires missing ISO and EIA standards require all 8 wires.

10BASE-T application requires:1236
Token Ring application requires: 3645
ArcNet application requires: 45
Use Length to measure lengths of individual wires to pinpoint
the location of the break.

Solving Length/Opens/Shorts Problems

Excessive length is not normally a problem found in properly
designed structural wiring systems. Use TDR to find the location
of open, shorted, or intermittent connections.

59
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

Possible cause Solution

Reported cable length is Punch blocks or other intermediate
shorter than the actual connections may cause a reflection
cable length before the end of the cable. If other test
(Intermediate reflection) parameters are within limits, it may not
cause a problem.

Length exceeds limits Check for excess cable coiled in walls

or ceilings. May not be a problem if
attenuation and NEXT readings are
within spec.

Length measured Verify NVP in Scanner is set correctly

inaccurately for the cable under test. Also check for
excess cabling in walls and ceilings.

Can’t trace cable to find Use a toner to trace path from source to
location of short or open point of break.

Solving Resistance Problems

Resistance is measured in a loop through each pair in the cable.
Given that the value for resistance per foot is usually extremely
consistent, excess resistance results are almost always due to
either too much cable or additional contact resistance at a
termination point.

Possible cause Solution

Excess length See ’Solving Length Problems’

Poor connection Poor connections at punch blocks or RJ-45

connectors can cause excessive resistance
and thus signal loss. Inspect all termination
points for good contact on connections.
Inspect all contact surfaces of connectors
and reseat all connections.

60
 Microtest, Inc.

Solving Attenuation Problems

Attenuation is the decreasing of a transmitted signal as it travels
along a cable. This loss increases with cable length, frequency,
and temperature. Above a certain amount of loss, the cabling
may not send network data reliably.

Possible cause Solution

Cable grade unsuitable for R eplace cable with appropriate

data rate, e.g. using Cat 3 grade.
for 100 Base-T
applications

Excessive length Is length within manufacturer’s

guidelines for your network type?
Move user to a closer hub
R eplace with higher grade cable.

Non-twisted or poor quality R eplace with category 5 patch

patch cable (e.g. silver cables.
satin)

Poor punch-down block Check and reconnect if necessary.

connections 66 blocks will typically exhibit much
more loss than 110 or Krone
blocks.

Poor connections or other Verify that conductors are seated

terminations properly. Check for tight wire twists
at all termination points.

Solving NEXT Problems

Excessive NEXT is usually caused by poor (loosely twisted)
terminations. Using poor quality patch cables is another com-
mon cause for NEXT.

61
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

Possible
Solution
NEXT cause

Use of E limination of couplers will reduce NE X T. T he fewer the

couplers total number of connections in the link, the lower will be
your total NE X T.

Grade of cable Verify that the grade of cable is suited to the application.

Multiple Use of additional pairs to carry data for other applications

applications can affect NE X T. E nsure wiring is used for hub-
workstation communication only. Other traffic should be
re-routed to different cables.

Substandard T he overall quality of the link will be determined by the

components quality of its weakest component. F or category 5 links,
all components: cables, patch cords, terminations,
connectors, patch panels, and punch blocks must be
rated category 5.

Patch cables W here patch cables are necessary, use high grade
cables with the minimum possible length. Use of non-
twisted pair patch cables, i.e. silver satin, will adversely
affect NE X T.

Pair Twists E xamine cable for loose or untwisted cabling, particularly

up to blocks, connectors, and/or wall plates. Category 5
connections require a maximum of 13 mm (0.5 inch)
untwisted wires at any termination.

Split pairs E nsure logical pairs are twisted together. US OC wiring

causes split pairs. S ee Appendix 2 for a guide to
connector wiring and color codes.

Unbalanced S ome R J-45 designs are capacitively-coupled only, and

connectors can cause common mode voltages to create much more
NE X T than is expected. S uch problems are usually
identifiable because they occur only on short links ( < 25
meters). T he only solution in this case is to use a
different connector. E ven a perfect termination job will
not fix this problem.

E arly category 6 links are carefully constructed from

Category 6 matched components. T his means the test cord on the
NEXT failure field tester should match the link manufacturer's
recommendation for the exact type of cable and
connector. F ailures of category 6 links are often traced
to using a different supplier's test cord and plug that
does not exactly match the category 6 link under test.

62
 Microtest, Inc.

Solving ACR Problems

Given that ACR is derived directly from attenuation and NEXT
measurements, you should review possible causes for failures in
those areas if you have a poor ACR result. See the previous
sections, Solving Attenuation Problems and Solving
NEXT Problems, for more details.

Possible noise source Solution

AC lines, light fixtures Do not lay data cables alongside power

cables. Move data cables away from power
sources and light fixtures, especially florescent
light fixtures.

Machinery and other If possible, move data cables away from the
impulse noise source of noise. If this is not possible and the
generating equipment noise is excessive (for example some factory
environments), you may need to replace UTP
with ScTP, STP, or fiber.

Radio Frequency Move the data cable away from the source of
Interference (RFI) noise. If this is not possible, you may need to
sources replace UTP with ScTP, STP, or fiber.

Solving Noise Problems

This refers to external noise rather than crosstalk noise. Caution
should be used when evaluating whether or not noise is an issue
since there are no standards for what constitutes unacceptable
levels of ambient noise.

63
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

For More Information

Got a tough question? Puzzled about some unique aspect of
cabling? For those with Internet access, the Usernet Group
comp.dcom.cabling is a great place to exchange ideas, ask ques-
tions, and pick up information. Like most discussion groups,
there is no guarantee on the quality or bias of information
presented, but it is useful resource. Another, more advanced
forum for cabling issues with an international perspective can be
found at http://www.teldor.com/ttcf.html. This is a closed
user group to people who are seriously interested in cabling
issues, but any cabling professional can get access. Once on the
forum, you may receive regular emails on subjects such as fiber
modal bandwidth, discussions on how to make very high
frequency measurements on UTP, shielding/emissions issues,
and other advanced topics.
White papers on cabling subjects, troubleshooting tips, links to
many cabling oriented sites, and cabling standards information
can be found in the Industry and Technology section of the
Microtest web site http://www.microtest.com.
You may also contact your closest Microtest office for addi-
tional information:
Microtest, Inc., 4747 North 22nd Street, Phoenix, AZ 85016
1-800-526-9675, Phone: (602) 952-6400, FAX: (602) 952-6401
Microtest Europe Ltd., 2A Kingfisher House, Northwood Park,
Gatwick Road, Crawley, West Sussex, RH102XN, United
Kingdom
Phone: +44 (1293) 894 000, FAX: +44 (1293) 894 008
Microtest GmbH, Am Soeldnermoos 17, D-85399 Munich-
Hallbergmoos, Germany
Phone: +49 (89) 607 6861-0, FAX: +49 (89) 607 6861-1

64
 Microtest, Inc.

Glossary

66-Type Punch-Down Block

A block consisting of four columns of 50 pins each, onto which
the 50 wires of the 25-pair group are placed. Each wire is placed
in a pin, and then punched into place, stripping the insulation in
the process. The pins in columns 1 and 2 are shorted together,
and the pins in columns 3 and 4 are shorted together. This
creates an input and output side of the block. To connect the
two halves, bridging or shorting clips are used between pins 2
and 3 of any particular row. The 66-Type Punch-Down Block
provides easy access to each wire and is used to terminate most
twisted-pair cable.
Most installed 66-Type Blocks do not support Category 5.

802.3
The Institute of Electrical and Electronic Engineers (IEEE)
physical layer standard for Ethernet coaxial cable. It specifies
the carrier sense multiple access with collision detection
(CSMA/CD) access method on bus topologies.

802.5
An IEEE standard for token ring networks that provides for a
ring topology configuration (a closed set of active taps con-
nected by point-to-point links). Access to the ring is granted
when a token is received and passed in a logical (and physical)
ring sequence between the workstations.

10BASE-T
A twisted-pair Ethernet cable standard that conforms to IEEE
802.3 specifications and can carry data at 10 Mbps.

65
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

Alternating Current (AC)

An electric current whose direction is periodically reversed. This
is the common form of electricity generated in homes and
offices. All electronic digital systems use direct current (DC) in
their circuits and have built-in power supplies to convert the AC
into DC.

ANSI
American National Standards Institute.

ARCNET
An acronym for Attached Resource Computer Network, a
token-passing bus architecture network developed by Datapoint
Corporation. ARCNET, the first local area network technology,
has a flexible architecture that allows both star and bus topology
networks, or a combination that can be described as a distrib-
uted star with branches. ARCNET can use single twisted-pair,
coax and duplex fiber optic cabling.

Attenuation
The decreasing of a transmitted signal as it travels along a cable.
The longer the cable the more loss there will be. Above a certain
amount of loss, the cabling may not transmit network data
reliably.

Attenuation to Crosstalk Ratio (ACR)

An indication of how much larger the received signal of a pair is
compared to the noise (NEXT) on the same pair.

Bandwidth
Maximum information carrying capacity of a channel.

Basic Link
A basic link is a model defined by TSB-67. A basic link is

66
 Microtest, Inc.

similar to what an installer might work with – including the wall

plate, horizontal wiring and first cross connection. It is intended
to be used by system designers and users of data telecommuni-
cations systems to verify the performance of permanently
installed cabling. A basic link consists of up to 90 meters of
horizontal cabling, a connection at each end, up to 2 meters of
test equipment cord from the main unit of the field tester to the
local connection, and up to 2 meters of test equipment cord
from the remote connection to the remote unit of the field
tester.

CENELEC
Comité Européen de Normalisation Electrotechnique

Certification
An automated test process that verifies that test results meet
established standards requirements.

Channel
All elements of the communications link connecting a DTE
(typically a PC or server) to a hub in the wiring closet. Includes
fiber cabling, patch cords, connections, and splices if any.
A channel is a model defined by TSB-67. It is similar to what a
user works with to transmit information between a personal
computer and its hub, or concentrator. A channel includes up to
90 meters of horizontal cable, a work area equipment cord, a
telecommunications outlet/connector, an optional transition
connection close to the work area, and two cross-connect
connections in the telecommunications closet. According to the
TSB-67, the total length of equipment cords, patch cords and
jumpers shall not exceed 10 meters. It is important to note that
the connections to the equipment at each end of the channel are
not included in the channel definition.

67
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

Cladding
A protective layer, made of glass or plastic, which surrounds the
core. The cladding has a lower index of refraction than the core,
which helps keep light in the core.

Core
The center of a fiber optic cable, usually made of extremely clear
glass or plastic, through which the light travels.

Electronic Industries Association (EIA)

An organization of manufacturers of electronic parts and sys-
tems. The EIA sets such standards as the EIA-232 interface
specification.

Electromagnetic Interference
An unwanted signal that enters the transmission line from
fluorescent lights, fan motors, etc. See also RF Interference.

ELFEXT
Equal Level Far End Crosstalk. This is far end crosstalk ad-
justed for attenuation. It can be thought of as far end ACR.

FDDI
Fiber Data Distributed Interface

Hertz (Hz)
The frequency of electrical vibrations (cycles) per second. One
Hz equals one cycle per second.

IDC-Type Punch-Down Block

A twisted-pair wiring panel in which each wire is placed in a pin,
and then punched into place, stripping the insulation in the
process.

68
 Microtest, Inc.

IEC
International Electrotechnical Commission

Institute of Electrical and Electronic Engineers (IEEE)

A membership organization, including engineers, scientists and
students in electronics and allied fields, that is involved with
setting standards for the computer and communications field.

Impedance
The resistance to the flow of alternating current in a circuit.

ISO
International Standards Organization

Local Area Network (LAN)

A communications network serving multiple users within a
confined geographical area (as in the same building or group of
adjacent buildings). It usually refers to the interconnection of
personal computers. Shared data is stored in a high performance
PC called a fileserver which serves as a remote disk drive to all
network users. Users may also share printers, modems and other
peripheral devices.

Loopback
The shorting together of wires in a connector so that a signal
sent along the wire returns to its place of origin.

Megahertz (MHz)
One million cycles per second. See Hertz (Hz).

Micron
1 millionth of a meter

69
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

Multimedia network
A network that allows access to Integrated Voice and Data
(IVD) over standard unshielded telephone twisted-pair wiring.
Multimedia LANs use existing wiring to carry both voice and
medium-to-high speed data in typical office environments.

Multimode
Fiber optic cabling with a wider core (typically 50 or 62.5
microns) that allows light to travel in multiple paths, such that it
is reflected back from the cladding back into the core as it
travels down the core.

Near End Crosstalk (NEXT)

The interference measured on a wire adjacent to the wire on
which the signal is being sent. Near end crosstalk measurements
show the amount of signal leaking when measured close to the
signal generation point. If the crosstalk coupling across at the
near end is great enough, it can interfere with signals coming
from a remote point which are diminishing as they reach the
same spot.

Noise
Any extraneous signal that invades the transmission of electrical
pulses or frequencies along a cable. Noise is measured as im-
pulse or Root-Mean-Square (RMS).

Nominal Velocity of Propagation (NVP)

The speed of data transmission along a cable relative to the
speed of light in a vacuum.

OLB
Optical Link Budget

70
 Microtest, Inc.

OTDR
Optical Time Domain Reflectometer. A sophisticated fiber test
tool that locates fiber breaks, and can determine loss per con-
nection, splice, or segment.

Patch cable (Test Cable)

A short, flexible cable terminated at both ends that is used to
interconnect LAN equipment.

Pin
One of the male leads on a multiple line plug, such as an RJ-45
connector or an EIA-232 connector.

Plenum cable
Cable certified to be fire resistant and to produce a minimum of
smoke. It can be installed in the space between the false ceiling
and the floor or ceiling above, called the plenum.

Punch-Down Block
A central termination point for twisted-pair cable. Each wire is
placed in a pin, then punched into place, stripping the insulation
in the process. See also 66-Type Punch-Down Block and IDC-
Type Punch-Down Block.

Resistance
A property of a conductor which resists or opposes the flow of
current in an electronic circuit.

Return Loss
This is a measure of the overall uniformity of a link’s impedance
relative to 100 ohms.

RF Interference
An unwanted signal that enters the transmission line from radio

71
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

and television transmitters. With this type of interference, the

cable acts as an antenna. See also Electromagnetic Interference
and Noise.

Screened Twisted-Pair (ScTP) cable

Four pair UTP, with a single foil or braided screen surrounding
all four pairs in order to minimize EMI radiation or susceptibil-
ity. Screened twisted pair is sometimes called Foil Twisted Pair
(FTP).

Screened/Shielded Twisted Pair (SSTP)

Four pair cabling, with each pair having its own individual
shield, in addition to an overall shield surrounding all four pairs.
SSTP offers similar performance to Type 1 STP except with 4
pairs (rather than 2) and in a 100 ohm impedance (rather than
150).

Shield
An insulating material used with coax or twisted-pair cable to
reduce electrical interference.

Shielded Twisted Pair (STP)

Two pair cabling, with both pairs having its own individual
shield, in addition to an overall shield surrounding both pairs.
STP is most familiar to LAN users as the heavy black IBM Type
1 cabling found in token ring networks.

Single-mode
Fiber optic cabling with a narrow core (8-9 microns) that only
allows light to travel in one path (i.e. one mode).

TIA
Telecommunications Industry Association

72
 Microtest, Inc.

Time Domain Reflectometry (TDR)

A diagnostic technology in which a pulse of known amplitude
and duration is sent along a cable. When an open or short is
detected, a pulse reflects back to the device generating the
signal. The device then measures how long the round trip took
and converts this to distance using the cable’s NVP.

Token Ring Network

A physical ring network topology. On a token ring network, data
moves around the entire ring, from one workstation to the next,
eventually receiving the transmitted information back after it
completes one round trip around the ring. The token, a specific
bit sequence that circulates around the nodes, gives permission
to transmit.

Twisted-Pair cable
Cable in which two wires are wrapped (or twisted) around each
other between two and 12 times per foot of length. Most
twisted-pair cabling contains either four or 25 pairs of wires.

Unshielded Twisted-Pair (UTP) cable

A common type of twisted-pair cable that does not have a shield
wrapped around the wires.

Wire closet
An area in a building in which punch-down blocks are centralized; a
central termination area for network or telephone cabling.

73
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

74
 Microtest, Inc.

Appendix 1:
Cable Installation Tips
The following is a compilation of UTP installation tips. If you
follow most of these recommendations you will greatly reduce
future downtime.
Ø Buy the best UTP cable you can afford
Ø Buy termination equipment (modular 8 jacks, punch
blocks, etc.) that meets or exceeds the quality of your
cable
Ø Label cable ends
Ø Modular 8 plugs should match the intended cable type
(flat versus round, and solid-core versus stranded)
Ø Always maintain the wire twist for all pairs up to the
point of termination (within 13 mm [0.5 in] for category
5 installations)
Ø Avoid sharing voice and data applications in the same
cable
Ø Always terminate voice and data cables into different
punch blocks or racks
Ø Strive to minimize the number of connections between
the Network Interface Card (NIC) and network equip-
ment. This will help reduce NEXT, attenuation, and
intermittent connection problems.
Ø Use a minimum bend radius of four times the cable
diameter for 4-pair cable (10 times for 25-pair cable)
Ø Tie and dress cables neatly. Do not cinch category 5
cables too tightly or you can damage them.
Ø Never use bridged taps (multiple appearances of the
same cable pair at several distribution points)

75
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

Ø Allow at least one square foot of closet floor space for

every 250 square feet of usable floor space
Ø Allow for sufficient cooling and airflow in the wire closet
Ø Put a lock on the wire closet and use it, but make sure
access is available to those who need it.
Ø On multi-floor installations, locate the wiring closet
where there is common vertical access, preferably in a
location <100 meters from the furthest user connection.
The location is often near the center of the floor, which
helps ensure cable distances don’t exceed the maximum.
AT&T research has shown that 99 percent of the time
user stations are located 93 meters or less from a wiring
closet.
Ø Install sufficient power circuits and outlets for future
unanticipated equipment
Ø Be sure you have permission to work in the wiring
closet. In some buildings, the owners of the closet are
not the tenants
Ø Do not use old 66 blocks (which exhibit high frequency
attenuation and poor NEXT). Use them only for voice
applications. Some new 66 blocks do support category 5;
check with your vendor first.
Ø Do not use screw type terminations. Use clip or barrel
IDC (insulation displacement contact) connections such
as those used in IDC blocks
Ø Pay very careful attention to termination practices.
Maintain a tight wire twist up to the point of termination
at punch blocks, wire plates, and connectors
Ø Check building and fire codes and follow them
Ø Avoid routing cables near EMI noise sources, such as

76
 Microtest, Inc.

motors and AC power lines

Ø Keep patch cables as short as possible, and choose
category 5 patch cable when possible.
Ø Watch out for patch cables terminated such that the
wrong pairs are twisted (i.e., 12/34/56/78)
Ø Beware of silver-satin or flat (untwisted) patch cables,
especially from the wall plate to the NIC. Many vendors
recommend you buy factory terminated patch cables to
ensure connection terminations meet category 5 require-
ments.
Ø Insist that the installation be tested with equipment that
meets or exceeds the requirements for the installation,
and get a conformance report for each cable. A toner or
DMM is insufficient. Such tools tell you nothing about
the cable’s ability to reliably transfer high speed data.
Ø Ensure that any measurements of NEXT are done from
both ends. It is possible to pass NEXT requirements at
one end and fail at the other.

77
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

78
 Microtest, Inc.

Appendix 2:
A Guide to Modular Connector Wiring
This appendix provides an overview of modular connector
wiring, and includes a comparison of wiring schemes, color
codes for connectors and cables, and pin assignments. It also
includes an example of a structured wiring connection from the
hub in the wire closet to the workstation in the users office.

Modular 8 Connectors
Modular 8 connections are comprised of a female jack and a
male plug. All modular 8 jacks are not the same. Careful inspec-
tion will reveal there are two main types: those designed to work
with solid core UTP and those designed to work with stranded
UTP (see Figure A2.1). Some universal designs support both
types.

Stranded Solid

Figure A2.1: Solid core versus stranded modular-8 jacks

In the solid core design, the blade in the pin connection is split
into two tines, which firmly grasp the solid wire. In the stranded
wire design, the blade is designed to pierce the strand. If you
attach the wrong type of connector to your cabling, you are
likely to get an intermittent connection.

79
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

Modular 8 Jack Pin/Pair Assignments

Most networks will use T568A or T568B assignments. USOC
wiring, though suitable for voice applications, should be avoided
in data applications, as it causes split pairs (and thus excessive
NEXT).

Designation Pin Pair Color

T568A 1 T3 White-Green (W-G)
2 R3 Green (G)
3 T2 White-Orange (W-O)
4 R1 Blue (BL)
5 T1 White-Blue (W-BL)
6 R2 Orange (O)
7 T4 White-Brown (W-BR)
8 R4 Brown (BR)

T568B 1 T2 White-Orange (W-O)

2 R2 Orange (O)
3 T3 White-Green (W-G)
4 R1 Blue (BL)
5 T1 White-Blue (W-BL)
6 R3 Green (G)
7 T4 White-Brown (W-BR)
8 R4 Brown (BR)

USOC 1 T4 White-Brown (W-BR)

2 T3 White-Green (W-G)
3 T2 White-Orange (W-O)
4 R1 Blue (BL)
5 T1 White-Blue (W-BL)
6 R2 Orange (O)
7 R3 Green (G)
8 R4 Brown (BR)

80
 Microtest, Inc.

Figure A2.2: 8-Position Pin/Color Assignments for T568A,

T568B, and USOC Jack Configurations

Beware of mixups where one end the cable is terminated using

T568A, and the other end using T568B.

Network Cabling Standards

10BASE-T, token ring, and TP-PMD each use unique pin
assignments for transmitting and receiving data. The following
table summarizes the requirements for each network type.

ATM User ATM Network

Pin 10BASE-T Token Ring TP-PMD
Device Equipment
1 TD+ T3 T+ R+

2 TD- R3 T- R-

3 RD+ T2

4 R1

5 T1

6 RD- R2

7 T4 R+ T+

8 R4 R- T-

81
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

Color codes for 100 W UTP horizontal solid cable and

stranded patch cords
Normally, horizontal wiring and patch cords share the same
color code scheme. However, an approved option exists, as
shown in the following table. The EIA T568A configuration is
assumed.

Pair Horizontal Option 1 Option2 Pin

Pair 1 White-Blue (W-BL) White-Blue (W-BL) Green (G) 5
Blue (BL) Blue (BL) Red (R) 4

Pair 2 White-Orange (W-O) White-Orange (W-O) Black (BK) 3

Orange (O) Orange (O) Yellow (Y) 6

Pair 3 White-Green (W-G) White Green (W-G) Blue (BL) 1

Green (G) Green (G) Orange (O) 2

Pair 4 White-Brown (W-BR) White-Brown (W-BR) Brown (BR) 7

Brown (BR) Brown (BR) Slate (S) 8

Beware of 25-pair UTP, as most of it is not category 5. Further-

more, at the present time, no field test equipment exists that can
verify that it is category 5.

82
 Microtest, Inc.

Appendix 3:
Standards Organizations

American National Standards Institute

1430 Broadway
NY, NY 10018
Phone (212) 642-4900
www.ansi.org/
Canadian Standards Association
178 Rexdale Blvd.
Toronto Ontario
Canada M9W 1R3
Phone (416) 747-4044
www.csa.ca/
Comité Européen de Normalisation Electrotechnique
Rue de Stassart, 35
B - 1050 BRUSSELS
Phone + 32 2 519 68 71
Fax.: + 32 2 519 69 19
http://server.cenelec.be
EIA/TIA Standards Office
2001 Pennsylvania Ave. N. W.
Washington, DC 20006.
Phone (202) 457-4966
www.eia.org/
www.tiaonline.org/
Federal Communications Commission
U.S. Government Printing Office
Superintendent of Documents
Mail Stop SSOP
Washington, DC 20402-9328
www.fcc.gov

83
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

Institute of Electrical and Electronics Engineers

Attn: Standards Department
445 Hoes Lane
PO Box 1331
Piscataway, NJ 08855-1331
Phone (908) 981-0060
stdsbbs.ieee.org/
International Standards Organization/
International Electrotechnical Commission (ISO/IEC)
www.iso.ch
ANSI is the U.S. member for ISO and IEC, and is the
source in the United States of their publications
(see ANSI).

International Electrotechnical Commission (IEC)

PO Box 131
1211 Geneva 20
Switzerland
www.iec.ch
International Telecommunications Union (ITU)
Place des Nations
1211 Geneva 20
Switzerland
www.itu.ch

Many of the standards referred to in this handbook may be

obtained from:
Global Engineering
1990 M Street N. W.
Washington, DC 20036
Phone (800) 854-7179
or (202) 429-2860
global.ihs.com

84
 Microtest, Inc.

Index
- Symbols-
10BASE-T network cabling standards 81
- A -
ACR problems 24–32
Attenuation, causes of 23
problems 61
Attenuation to Crosstalk Ratio (ACR) 24
- B -
Bandwidth 25
- C -
Cable radar 16
Cladding 68
Color codes 82
Copper Data Distributed Interface (CDDI)and wire map 27
Core 68
Crossed pairs 27
- D -
DC loop resistance 29
Dead zone 17
Delay 18
Delay Skew 18
- E -
Electromagnetic Interference (EMI) 30
ELFEXT 22
- F -
FEXT 22
Fiber Optic 30
Figure 1: Construction of Typical Two-Fiber Multim 14
Figure 2: Attenuation vs Wavelength for a Typical 15
Figure 3: TDR 17
Figure 4: Near End Crosstalk 20
Figure 5: Near End Crosstalk of a Category 5 Link 21
Figure 6: FEXT 22
Figure 7: Cable Attenuation 23
Figure 8: Attenuation of a Category 5 Link 24
Figure 9: Correct connectivity 26
Figure 10: Correct connectivity 27
Figure 11: Reversed Pairs, Crossed Pairs and Split 27
Figure 12: Typical Fiber link 32

85
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

Figure 13a: Model of the channel, cabling link an 36

Figure 13b: The “basic link” test configuration f 36
Figure 14: Basic Link 42
Figure 15: Channel 44
Figure 16: Permanent Link 46
Figure 17: The Microtest OMNIScanner 55
Figure 18: NEXT Comparison of Network Analyzer and 57
Figure A2.1: Solid core versus stranded modular-8 79
Figure A2.2: 8-Position Pin/Color Assignments 81
- I -
Impedance 28
- L -
Length 16
Length/opens/shorts problems 59
Link Troubleshooting Guide 59
- M -
MICROSCANNER 58
Modular 8 connectors 79
jack pin/pair assignments 80
- N -
Narrow-band receiver 55
Near End Crosstalk (NEXT) 19
causes of excess 20
problems 61
Network analyzers 51
Network cabling standards 81
Noise, causes of 30
problems 63
Nominal Velocity of Propagation (NVP) 16
- O -
OTDR 30
- P -
Physical layer tests 16
Power Sum NEXT 21
Problem solving
attenuation 61
length/opens/shorts 59
NEXT 61
noise 63
resistance 60
wire map 59
Propagation delay 18
PSELFEXT 22

86
 Microtest, Inc.

PSNEXT 21
- R -
Radio Frequency Interference (RFI) 30
Resistance problems 60
Return Loss 18, 28
Reversed pair 27
- S -
Split pairs 27
SRL 28
Standards 33
Standards organizations 83
Structural Return Loss 28
- T -
Table 1: High Performance Cabling Alternatives 11
Table 3: Cable Standards 33
Table 4: Published Standards 34
Table 5: Link Naming Protocol 34
Table 6: Cabling Bandwidth 35
Table 7: Categories and Classes 42
Table 8: Basic Link Performance 43
Table 9: Channel Performance 45
Table 11: Technology Application 48
Table 12: Fiber Performance Requirements 49
Table 13: OLB Formulas 50
Table 14: Choosing the Appropriate Cable Test Tool 51
Table 15: Certification Tools 54
Time Domain Reflectometry (TDR) 16
Token Ring network cabling standards 81
Tools
troubleshooting & certification51
TP-PMD
network cabling standards 81
Troubleshooting
ACR problems 63
attenuation problems 61
length/opens/shorts problems 59
NEXT problems 61
noise problems 63
resistance problems 60
wire map problems 59
- W -
Wire map 26
problems 59

87
FIELD TESTING OF HIGH PERFORMANCE PREMISE CABLING

Microtest, Inc., 4747 North 22nd Street, Phoenix, Arizona 85016-4708

1-800-526-9675, Phone: (602) 952-6400, FAX: (602) 952-6401

Microtest Europe Ltd., 2A Kingfisher House, Northwood Park, Gatwick

Road, Crawley, West Sussex, RH102XN, United Kingdom
Phone: +44 (1293) 894 000, FAX: +44 (1293) 894 008

Microtest GmbH, Am Soeldnermoos 17,

D-85399 Munich-Hallbergmoos, Germany
Phone: +49 (89) 607 6861-0, FAX: +49 (89) 607 6861-1

http://www.microtest.com/

2938-4505-05 Rev. 01 5/98

88