Tsb67 Data Testing
Tsb67 Data Testing
Tsb67 Data Testing
SECTION 16715
L u c e n t T e c h n o lo g ie s
Bell Labs Innovations
Introduction
Characterizing cabling as a separate entity provides a useful data point for the design of future
applications equipment. Field measurements verifying the transmission performance of cable
plant result in a more professional hand-off from the cabling contractor to the applications
provider. This should improve the overall effectiveness of providing premises networking
solutions, resulting in higher customer satisfaction.
These guidelines present a high level summary of the results of our evaluation and experience
using the instruments. The intent of these guidelines is not to duplicate the information in
TIA/EIA TSB67, or the contents of individual field tester manuals, but to provide additional
information to address issues relevant to the testing of SYSTIMAX SCS installations. These
guidelines should be considered generic to all testers and where specific tester issues are involved,
the mechanism that we have evolved is to talk directly with the tester vendors and recommend
they implement changes in their firmware and/or documentation. So it is important to upgrade to
the latest firmware and maintain the current documentation (including readme files) of the testers
to ensure up to date operational procedures.
The document is organized in the natural sequence a user will be involved with a filed tester as
shown:
2.0
Selection
Calibration
Operation
Interpretation
Application Testing
Upgrades
Trouble Shooting
07/22/1999
ATTACHMENT 1
SECTION 16715
Annex A describes the results of a Bell Labs evaluation and other considerations that are
important in choosing field testers.
3.0
Different field test instruments have slightly different calibration procedures. It is important that
these procedures are followed. Proper calibration of the field testers is required to ensure accurate
measurements.
3.1
A short length of tester equipment cord (supplied with the field tester) is used to connect the
main unit to the remote unit and the calibration sequence activated from the main unit. This will
result in the main unit determining attenuation loss inherent in the testers and calibrate this loss
to be 0 dB (See Figure 1). Different manufacturers provide different calibration kits. If adapters
are needed for the measurement of the installed cabling, it is a good idea to include these adapters
in the calibration. This is because adapters are not part of the installed cable plant and should
consequently be calibrated out of the attenuation measurement, resulting in better measurement
accuracy for attenuation.
Figure 1
3.2
07/22/1999
ATTACHMENT 1
SECTION 16715
Selecting the cable type closest to the cable being tested will yield better accuracy, especially for
length and propagation delay measurements. Often, the field tester is pre-loaded with various
cable types, including 1061/2061 cables. The main result of selecting a cable type is the selection
of the nominal velocity of propagation (NVP). Values for NVP are obtained by tester vendors
from cable data sheets, or sometimes verbally from individuals representing Lucent Technologies.
Consequently, these NVP values are often incorrect or out of date. It is critical that the correct
cable type be selected and the NVP determined using the field tester calibration procedure.
3.3
TSB67 defines two test configurations, a basic link and a channel (see Figures 2 and 3). A channel
assumes two connections at each end while a basic link assumes one connection at each end. For
example, if you are testing the cabling installed by a contractor from the wall outlet to the
termination in the telecommunications closet, you need to configure the field tester for a basic
link. If on the other hand, you are diagnosing an end-to end cabling circuit that will be attached to
equipment at both ends, including cross-connections, patch cords, and equipment cords, you
need to configure the field tester to measure a channel. Sometimes, you may have a situation
where there are two connections at one end and a single connection at the other end. For routine
autotest measurements, it is acceptable to measure this configuration as a channel in the interest
of efficiency. However, for diagnostic or single parameter tests, this configuration will need to be
tested as a channel and a basic link. The end that has 2 connections must pass channel NEXT
requirements, while the end with one connection must pass basic link NEXT requirements. This
type of dual configuration testing may be necessary only if you are working in a detailed analysis
or diagnostic mode.
07/22/1999
ATTACHMENT 1
SECTION 16715
Figure 2
07/22/1999
ATTACHMENT 1
SECTION 16715
Figure 3
3.4
Calibration of NVP
TSB67 outlines a simple calibration procedure for NVP using a known length of cable. Since
various testers use different internal procedures to find the pair with the shortest electrical delay
(corresponds to maximum velocity of propagation), calibration is the only sure way of initializing
the field tester with the right NVP for the particular cable type you are testing. This will result in
more accurate length measurements and less variability of length between different pairs within a
4-pair cable. Figure 4 illustrates the NVP calibration setup.
07/22/1999
ATTACHMENT 1
SECTION 16715
Figure 4
Calibration of NVP
To calibrate, reel off 25 meters (82 feet) of cable and attach modular jacks to both ends. Note the
exact jacket foot marks and determine the length of cable that is cut from the reel. Attach this
connectorized cable to the main unit and the remote unit using tester. Select NVP calibration from
the menus and enter the length of the cable. Note that since the length is determined from the foot
marks on the jacket, it is necessary to make sure that the field tester is set to the appropriate
units (metric or U.S. Customary units) before running the calibration command from the main
unit. The main unit will display the NVP determined for the 1061/2061 cable that you are
calibrating. Save this NVP as a default for the appropriate cable type, so that you do not have to
re-calibrate NVP again for the same type of cable. If you move on to another installation, you
may have a different cable, which requires that you calibrate NVP again.
4.0
It is important to develop uniform test practices and provide adequate training. Although each
cabling customer may have different requirements, it is risky letting each job get tested
differently. These units are much more complicated than continuity buzzers, and improper
settings or inadequate battery management could waste hours worth of testing. Suitable
calibration procedures and intervals must be enforced. For example, calibration of remote unit
every day of use is a good practice. Calibration of NVP for each job will ensure that the NVP
represents the value of the cable being installed at a site.
07/22/1999
ATTACHMENT 1
SECTION 16715
4.1
Sanity Checks
Any tester should undergo regular checkups, or some form of sanity check. Users should develop
some expertise to assess whether a tester is providing useful results. One simple method is to
construct a sample link to be used as a reference. Make sure that it is physically protected and
not subject to movement. Expect some variation with temperature, humidity, and time, but this is
a valid check for malfunctioning units. A link such as this is required for Nominal Velocity of
Propagation (NVP) calibration may be used as a reference link and will serve as a simple test to
verify the consistency of a field tester. If this same link is used to check periodically for tester
repeatability, one will get a useful indication that the field tester is functioning properly (see
Figure 4).
4.2
With the release of TIA-568-A, TIA TSB67 and IS 11801, customers are faced with confusing
choices in specifying testing requirements and interpreting test results. Most of the confusion
arises from attempting to test against the performance limits of IS 11801. It should be noted that
several of the performance limits listed in IS 11801 are marked "ffs" ("for future study"). Some
of the reasons behind the "ffs"
entries include lack of agreement from cable vendors,
inconsistent results from field testing trials, and lack of accuracy determination methods for
hand held testing equipment in relation to such parameters. In addition, while specifying
performance limits in IS 11801, the international standards community is still to release
detailed test procedures and test configurations.
The horizontal link from which the IS 11801 performance includes two connections in the floor
Distributor ( Horizontal Cross-connect in TIA-568-A), a 5 meter patch cord, 90 meters of cable
and a connection at the work area. This link specifically excludes the equipment cable and
work area cable. It is not clear whether hand-held instruments, compliant to TIA TSB-67 are
capable of accurately testing this link, as this would require the test set to "exclude" the
equipment and work area cables.
One of the parameters marked "ffs" in IS 11801 is Return Loss (RL). While the method for
accurately measuring RL in the field is yet to be specified, additional difficulties can be
encountered when using test leads or patch cords incorrectly matched to the cabling system, or
degraded by repeated use. At least one of the TSB67 compliant test sets in the market
includes two choices for IS 11801 configurations (with and without RL), and the test set
manufacturer advises the users to use the "IS 11801 without RL" option. The difficulties
experienced by some users when testing RL with incorrectly matched test leads highlights the
advantages of a cabling system designed with matched components.
07/22/1999
ATTACHMENT 1
SECTION 16715
In contrast with IS 11801, the TIA 568-A Standard and TSB-67 include practical test
configurations (Figures 2 & 3) and accuracy specifications for hand-held instruments. While
work is underway to harmonize the differences, and to reach practical agreement on the IS
11801 specifications marked "ffs", the practical choice is to refer to TIA TSB-67 for accurate
field testing.
4.3
Sometimes, an interface adapter is necessary to attach field testers to the channel/link under test.
These adapters are an extra element in the link configuration and may affect the accuracy of a
measurement. In general, unknown adapters should be avoided. To measure a basic link with a
110 interface, an adapter cord with 110 on one end and the interface of the tester on the other end
is preferred. Test instrument vendors should provide this adapter cord (see A.1). For channel
measurements, testers provide an adapter from the tester interface connector to a modular jack.
Again, it is important that only cords and adapters provided by the testers be used to attach field
testers to the link under test. Note that TSB67 requires that field tester cords do not exceed 2
meters.
4.4
An important issue not yet handled by any of the testers is keeping track of the life cycle of test
interfaces. Any jack or plug has a reliable limit on connect/disconnect cycles. Interfaces should be
serialized (marked) and users should keep track of how many tests are performed so their reliable
limit is not exceeded. Additional interfaces should be ordered and available for replacement at
suitable intervals. Field testers currently do not have a mechanism for storing the number of times
an instrument has been used. Hence it is a good practice to maintain approximate (to the nearest
100) cycles of usage so that plug/jack contacts can be inspected and replaced as necessary.
Different testers claim different number of usage cycles for accurate measurements. As a first
step these vendor guidelines may be sufficient. If measurements are not repeatable or if the failure
rate of links increases, it may be time to replace the plug/jack interfaces.
4.5
Backup Procedures
Backup protection is another important issue. A simple strategy could be to print tests every
few hours. This might be suitable for some small jobs, but may be too time consuming for most.
A better strategy is to uplink to a PC every 4 hours and make duplicates on floppy disks. Also,
make sure an extra set of charged batteries are ready.
4.6
ATTACHMENT 1
SECTION 16715
Method 3, or equivalent. SRL is not specified as a field measurement for links and channels either
in TIA/EIA-568-A or in IS 11801 standards. Hence any claims to measuring SRL in the field
should be considered non-standard and questionable.
Return loss is a measure of reflected signal and is used to characterize either individual
components or links. When characterizing links, return loss is an indication of impedance
mismatches of different components within links and is not specified in either TIA/EIA-568-A,
or the associated TIA/EIA TSB67 field testing document. Further TIA/EIA TSB67 clearly states
in the note under section 5.1 that return loss is under study. The IS 11801 document has a
slightly different approach when dealing with items under study. It shows the possible target
values for future revision of the IS 11801 document and qualifies these values as FFS ( For
Future Study). The reason these are marked FFS is because industry has not reached agreement
on the possible values, test methods, and measurement instruments.
In spite of the above background, certain field test manufacturers may implement proprietary
test schemes for measuring return loss to differentiate their instruments in the market. These
proprietary test schemes have not been reviewed and accepted by industry and must therefore be
viewed with caution.
4.7
Different pairs within a 4-pair cable use different twist rates to improve NEXT performance.
This introduces a physical difference in length between different pairs. Additionally, the velocity
of propagation may vary between pairs because of material differences. Such differences are
normal, and within the operating limits of cabling and applications standards. Furthermore, these
differences are not significantly affected by installation of cabling, and should therefore not be a
major source of concern when checking the quality of the installation.
Delay skew is the term commonly used to refer to the difference in propagation delay over
different pairs. The worst case delay skew allowed in application standards is 50ns for a 100
meter channel. Given that the speed of light in open space is 1 ft/ns and assuming a relative
velocity of 0.6 c for copper cabling, the maximum difference in electrical length allowed is 30 feet
(50 ns x 0.6) for a 100 meter channel. Differences of up to 20 feet in the length measurements of
pairs for a typical cable run are not uncommon and fall within the requirements of application
standards.
Calibration of NVP for each different type of cable is critical to reduce errors in length
measurements. Different cables require re-calibration. Note that the NVP calibration results must
be saved to ensure the new NVP values are effective. To test if the calibration was saved, power
down the tester, power up and check if the new NVP values for the cable type being installed are
current in memory.
07/22/1999
ATTACHMENT 1
SECTION 16715
Note: Because of accuracy issues, TIA standards do not endorse measurement of propagation
delay and delay skew using field testers.
4.8
25-Pair Cabling
TIA considers 25 pair cabling to be backbone cabling and requires NEXT to conform to powersum requirements. Power sum computations involve hundreds of pair combinations and several
thousand data points for conformance testing. Power sum NEXT testing is not possible with
currently available field testers. The alternative is to check individual 4 pair units within a 25-pair
cable and verify each unit meets the 4-pair horizontal cabling requirements. For example, to test a
zone cabling installation with a 25 pair cable serving 6 work areas, the 6 individual circuits can be
verified as basic links or channels, depending on the current test configuration (see3.3). Note that
the 25-pair cable is a different cable from a similar 4-pair cable, and its velocity of calibration
must be calibrated and stored in the field tester. Any 4-pair unit from the 25-pair cable may be
used for calibration.
5.0
Once a measurement has been made, the data has to be interpreted properly. Portions of the data
can be printed in a summary report or saved in records that are transferred later into a PC.
Detailed information necessary to display the data as a graph are only available immediately
following a test. This level of detail data is not stored in the filed tester and is overwritten by new
test information. Hence any detailed analysis of measured data has to be done immediately after
collection of data for a single link. This mode of operation will typically be used only for detailed
analysis. Site testing requiring only a PASS/FAIL indication should follow the faster autotest and
save cycle to optimize testing speed.
5.1
Individual parameter failures can be one of two types : FAIL, or *FAIL. The * indicates that the
failure is questionable because the measured value is so close to the specified limit that the field
tester is not sufficiently accurate to resolve the measurement. It is a good idea to check any FAIL
or *FAIL parameters to understand what may be the underlying cause. Note that any *FAIL of a
single parameter will result in an overall FAIL for the link.
A *PASS indicates that the result is better than the specified limit, but within the accuracy of the
instrument. *PASS for any parameter will be treated as a PASS and if all other parameters are
acceptable, the result in an overall PASS for the link under test. The *FAIL and *PASS area of
uncertain measurements is often referred to as the gray zone and is illustrated in Figure 5.
SYSTIMAX SCS installations exceed the requirements and will not fall in the gray zone when
using Level 2 instruments.
07/22/1999
10
ATTACHMENT 1
SECTION 16715
PASS
NEXT (dB)
PASS*
+1.6dB
TIA CAT 5 Channel
-1.6dB
FAIL*
FAIL
0
10
20
30
50
60
70
80
90
100
FREQUENCY (MHz)
5LB0275.25
Figure 5
5.2
40
In the unlikely event that a link fails, it is time to stop and understand what may be the cause of
the failure. Note that a failure is a measurement failure and could be caused by the cabling, the
tester, poor termination practices, or all of these.
1. Ensure that the tester is working properly by checking it against the benchmark /calibration
link mentioned in section 4. If the benchmark link measures within the accuracy limit of a
level 2 tester ( about 2 dB for NEXT and 1 dB for Attenuation) the tester is functioning
properly.
2. Follow the trouble shooting guidelines in section 7.0 for different types of failures.
3. If the trouble has been identified, take corrective action to fix the cabling.
4. Re-test to ensure that the corrective action has worked effectively.
5.3
Printing Reports
Printing reports is usually done after a continuous set of autotest measurements. Depending on
customer preferences, all test results may be printed and signed off by the installer and/or
customer. If the customer does not require documentation of the testing, any test results that
look odd or have gray zone measurements should be marked and printed from the field tester.
It is important to use the printer cable supplied with the tester unit and ensure that a compatible
printer is used.
07/22/1999
11
ATTACHMENT 1
SECTION 16715
5.4
Saving Data in a PC
For long term data analysis and statistical interpretations, it is a good idea to save all the data into
a PC. All testers provide some mechanism for uploading data into a PC, and freeing up memory
in the field tester for new measured data. This is a good practice that can be used to gain valuable
insights into the quality of an installation and as source data for the ongoing administration of the
premises network. Often, two levels of data can be transferred to a PC. The primary level is the
autotest summary report for each link or channel. A secondary level is detailed graphical data for
plotting and/or detailed analysis. The detailed data is available only after immediate measurement
of a link and cannot be stored as a record in the field tester (See Figure 5).
Figure 5
5.5
SYSTIMAX SCS warranties category 5 performance for all category 5 products sold under
SYSTIMAX SCS. If the SYSTIMAX SCS products are installed by certified VARS, the
complete installation is supported by a 15 year warranty. The SYSTIMAX SCS warranty
requires detailed inspection of installed cabling. Field testers do not substitute for checking over07/22/1999
12
ATTACHMENT 1
SECTION 16715
stressed cable, over-scoring into copper, and other improper installation methods. Hence field
testing is neither complete nor necessary for SYSTIMAX SCS certification. It does provide some
enhancements to the physical inspection process and a degree of comfort to the customer. Field
testing should be based primarily on customer requests and performed in addition to the normal
inspection process, not instead of actual inspection.
5.6
SYSTIMAX SCS does not require or support testing of individual applications in the field. Such
testing is performed in the laboratory as needed. Field testing should be limited to generic cabling
tests as specified in TIA/EIA TSB67.
07/22/1999
13
ATTACHMENT 1
SECTION 16715
6.0
Field testers are evolving rapidly and probably have a life cycle as fast as a PC ( 2-3 years).
Hardware updates are generally not possible and usually involve replacement of the tester.
Software upgrades are relatively easy and with the rapid evolution of field testers, are quite
frequent. These software upgrades are usually obtained directly by downloading new executable
files into a PC and transferring these to the field tester. The following sections describe
procedures used to obtain software for different field testers.
There are a number of ways to obtain the latest firmware for a particular tester; A diskette from
the manufacturer, access to downloadable software from an electronic bulletin board service
(BBS), or access to software and other information at world wide web sites on the Internet. Each
of the manufacturers offers a BBS and/or Internet site. They are listed below in alphabetical order
by manufacturer.
6.1
Datacom
LANcat V tester
BBS # 1-206 290 1603
The file to download is "UPDATE.EXE". It is a self-extracting file with the update program and
a "readme" file packed within it. It can be downloaded into an empty directory on a PC or into a
directory already dedicated to Datacom data and programs.
6.2
Fluke
DSP-100 tester
Internet site: http://fluke.com Not currently stocked with software upgrades.
BBS # 1-206 353 5966
The file to download DSPALL23.EXE. This will self-extract and install a new version of the
DSP-link software on the PC under Windows. The DSP-link software application will enable the
user to connect to the DSP-100 and upload file DSP_23.dsp to the main unit and SR_23.DSP to
the smart remote.
6.3
Microtest
Pentascanner Plus
07/22/1999
14
ATTACHMENT 1
SECTION 16715
BBS # 1-602 957 7716
File to download is "PS30.EXE" Self-extracting with help file and upgrade program unpacked as
part of the operation. Requires on-line product registration to obtain password for software
upgrades.
Internet site: http://www.microtest.com
File to download "PS26.EXE". Follow instructions for BBS program
6.4
Scope
Wirescope 100 or Wirescope 155
BBS # 1-508 393 0158
File to download is "Release 1.6 "for Wirescope 155 and "Release 4.40" For the Wirescope 100.
Internet site: http://www.scope.com
Same file and instructions as BBS.
6.5
Wavetek
Lantek Pro XL
Internet site: http://www.wavetek.com
6.6
The executable files listed in sections 6.1 through 6.5 change from time to time. Hence its is a
good idea to check at the time of downloading to ensure that you are obtaining the desired
firmware upgrade. A note of interest is that all the testers manufacturers are able to send out
prior versions of uploadable software if it is needed. As an example, if a new upload prevents a
tester from performing properly, one can revert to an older version of the firmware. Also, there
may be changes to the interface that some users are not comfortable with using. To obtain these
prior firmware versions, it is recommended that the user contact the manufacturer directly at the
number printed in the tester documentation.
07/22/1999
15
ATTACHMENT 1
SECTION 16715
07/22/1999
16
ATTACHMENT 1
SECTION 16715
7.0
Solution
remote
a) fails length
c) fails Attenuation
d) fails NEXT
07/22/1999
17
ATTACHMENT 1
SECTION 16715
ANNEX A
A.1
Bell Labs at Middletown completed an evaluation of available field testers in October 1995 by
comparing tester results with measurements made using a network analyzer for various basic link
and channel test configurations. The testers listed below (in alphabetical order by company
name) are acceptable for field testing of SYSTIMAX SCS. Since it is possible that results may
vary with different firmware revisions, the firmware revision at the time of testing is also shown.
COMPANY
Datacom Technologies
Fluke Corporation
Microtest Inc.
Scope Communications Inc.
Wavetek Corporation
TESTER
LANcat V
DSP-100
PentaScanner Plus
Wirescope-155
LANTEK PRO XL
FIRMWARE
2.001
1.3
3.10.013
3.91
1.10
Availability of adapter cords from the tester interface to 110 as well as modular jack
interfaces needed to test a basic link.
Life cycle operational costs of each instrument, including replacement of connecting hardware
which can be used for a limited number of insertions.
Warranty, support, and upgrade program.
A.2
User Evaluation
The following additional areas for user evaluation discuss some sample issues for the instruments
that can be taken into account as part of the decision making process. The issues covered are not
exhaustive and there may be others that an astute user will consider important.
A.2.1 Packaging
There are two types of carrying cases: briefcase or doctor's bag. A briefcase opens along the edge.
On a table or desk, zip it open and the units and accessories are laid out side by side. A doctor's
bag opens from the top. This seems to provide more protection and could be easier to open from
the floor. In addition to the case style, consider accessory pockets and the overall size and
weight.
A.2.2 Handling
07/22/1999
18
ATTACHMENT 1
SECTION 16715
Most of the units come equipped with straps for carrying or hanging, an important function since
only 2 meters of cord is allowed at each end of a basic link test configuration. Several of the units
are also equipped with stands that might be useful at the work area end of the link.
Some testers have non-slip rubber cases that are easier to grasp and seem less susceptible to
dropping. Bright color schemes could be an advantage under desks and in dim (non-TIA/EIA
compliant) telecommunications closets.
07/22/1999
19
ATTACHMENT 1
SECTION 16715
All units provide a serial cable and software for uploading test data into DOS or Windows based
PCs. Some restrict how much information could be uplinked, but all provide enough information
for normal field use.
07/22/1999
20
ATTACHMENT 1
SECTION 16715
A.2.6 Batteries
All of these units rely on batteries, in most cases using NICAD packs. Some testers charge with
the pack in the tester, while others offer a separate charging unit. Charging time, operating time,
and battery life varies between units. Some units also allow for tests using AC power or while
charging.
Battery STATUS is an important issue to consider, because many hours of testing will depend
on a reliable battery charge. Continuous battery status in the main display window is a useful
feature. Alternately, warning messages with beeps to alert users about a low battery condition
may be sufficient for most users. In any case, suitable battery management strategies should be
outlined by each manufacturer.
07/22/1999
21