A Tutorial on Centralized Optical Fiber Cabling Networks Part IV

Mary Spring Corning, Perspection Inc.

Informations

• Publié par: Necher
• Nombre de lectures: 33
• Langue: English

Extrait

Part IV
A Tutorial on Centralized Optical Fiber
Cabling Networks
by Douglas E. Harshbarger and George H. Sellard
he centralized optical fiber A continuity test can be per- In a more sophisticated test, an
network design consolidates formed on multimode fibers with a optical time domain reflectometer
T all network electronics in a flashlight, a walkie-talkie set and (OTDR) measures and documents
single main cross-connect (MC) and two people. No tone generator or discretely all points of loss along a
provides direct connections amplifiers are necessary. Just select cabling link.
between the MC and workstations. the color-coded fiber to be tested, Attenuation testing after installa-
FIBER
When compared with convention- darken the area,and shine the flash- tion ensures the network cabling
OPTICS
al, decentralized premises data net- light at the end of the fiber. Confir- meets established attenuation spec-
works, this design minimizes mation from a partner, in a ifications and link loss power bud-
significantly the number of darkened area at the other end of gets. Documenting the test results
telecommunications closets (TCs), the fiber, indicates link continuity. provides an invaluable as-built
connectors and splices needed. Attenuation testing is nearly as sim- record of the system, useful for any
Consequently, the centralized opti- ple. (Note: To protect your eye- subsequent troubleshooting or
cal fiber design greatly simplifies sight, do not look directly into the accountability.
network testing. end of a fiber that may be part of a
With optical fiber cabling sys- system using lasers.) Standards
tems, the properties most fre- Acceptance values and test pro-
Testing and Documenting
quently tested are continuity and cedures for optical fiber cabling
attenuation. Simply put,continuity Attenuation systems are specified in Annex H,
shows that a link will pass light The primary parameter field-test- “Optical Fiber Link Performance
from one end to the other, and ed in optical fiber cables is attenua- Testing,” of ANSI/TIA/EIA 568A, the
attenuation measures how much tion. This is the optical power loss, Commercial Building Telecommuni-
light passes from one end of a link expressed in decibels (dB), cations Cabling Standard. Test
to the other. Usually, only attenua- between cable termination end equipment is accurate and easy to
tion testing is performed because points. Attenuation is the sum of use.
it also proves continuity. However, loss caused by the physical proper-
Measuring End-to-End
a continuity check can be per- ties of fiber, splices, connectors,
formed very quickly and easily couplers and switches. End-to-end
Attenuation
prior to attenuation testing to veri- attenuation testing involves using a End-to-end attenuation is per-
fy that fibers are terminated in the light source and a power meter to formed using the insertion loss
right positions and labeled cor- measure optical power loss method, comparing the power of
rectly. between cable termination points. light injected into the fiber at the
PAGE 58 CBM DECEMBER 1998some sources of loss, such as a very tight
Figure 1 Zeroing out test equipment includes taking a reference measurement with one jumper, and then compar-
cable bend. In addition, the OTDR mea-
ing that measurement to another taken with an adapter and a second jumper.
sures and documents cable footage,
which is a helpful piece of data.
1: Reference Setup
Light Direction
By sending pulses of laser light through
a fiber and measuring the level and time
Test Jumper No. 1
delay of the pulses as they return, the
OTDR pinpoints where and how loss
Connector
Connector
occurs. The resulting signal trace depicts
the optical power as it is attenuated over
-6.10 dBm
distance by the fiber, connectors and
splices. (See Figure 3.) The signal trace
dBm
Power Received = P (dBm)
ref
documents the integrity of the system as
OS
OM
built and provides a baseline for future
troubleshooting. This record could prove
2: Jumper Check
Connector Connector
to be a critical cost-saver, as digging up
interbuilding cable to correct mistakes is
Test Jumper
Test Jumper
expensive.
No. 1
No. 2
Connector Interconnection
Connector
Sleeve How the Centralized Design
-6.70 dBm
Simplifies Testing
The centralized design “just makes
Power Received = P
dBm
check everything easier and faster,” said Charles
OS
OM B. Orr of Sellard Communications, who
has installed and tested a number of cen-
tralized optical fiber networks. With
Figure 2 End-to-end attenuation testing: Light from an optical source (left) is sent through a fiber to an optical
meter (right) where a power reading is recorded. This reading is compared with the reference measurement to
Attenuation testing
obtain the end-to-end attenuation of the fiber link.
begins with cleaning
Domain Reflectometer
Connector Connector Connector Connector
connectors and
Test Jumper
Test Jumper
System
adapters and zeroing Connector No. 1
No. 2
Connector
Patch
Patch
Panel
Panel
out the equipment to
-9.30 dBm
dBm
establish a reference
OS
OM
measurement.
nected to the power meter. (See Figure direct connections between the single
near end (P reference) and light exiting at
the far end (P test). A stabilized light 1.) This check power level must be with- main cross-connect and workstations,
source, handheld power meter, test cords in 0.5 dB of the reference. setup for testing is done only one time.
and an adapter are required. Testing Next, an optical source/test jumper is “The biggest advantage is not having to
should be performed in accordance with connected to one end of the fiber to be move anything around.” Orr said. “You
TIA/EIA standard procedures: TIA/EIA tested, and a power meter/test jumper is only set up once. With a conventional net-
526-14A, Method B for multimode fiber; connected to the other end. The resulting work design,you have to test between the
power reading is subtracted from the ref- computer room and the horizontal cross-
TIA/EIA 526-7, Method A.1 for single-
erence measurement to obtain the end-to- connect (HC). Then you have to move to
mode fiber.
end attenuation (Figure 2). the HC to test the cable between the HC
Attenuation testing begins with clean-
and the workstations. You have to move
ing connectors and adapters and zeroing
Testing with an Optical Time
out the equipment to establish a refer- people, equipment and documentation.
ence measurement. This is accomplished “Also, you’re supposed to zero out the
by transmitting a reference power level Analysis with an optical time domain meters when you turn them off and on
over a test jumper cable to the power reflectometer (OTDR) verifies acceptable again,” Orr continued. “When you test a
meter, and then delivering a check power losses through the cable and connector or centralized design, you don’t have to turn
level over the test jumper connected to splice joints. The OTDR also will detect the meters off until you’re done. That
any unexpected and potentially trouble- saves time and trouble.”
an adapter and a second test jumper con-
DECEMBER 1998 CBM PAGE 59tralized networks is completed only
Figure 3 In this sample OTDR signal trace, slopes indicate gradual attenuation over a section of fiber, while vertical
rarely, due to generally short distances in
drops represent point losses at connectors, splices and faults. Spikes indicate reflective events such as connectors.
horizontal cabling.
The final spike marks the end of the fiber.
“If there’s trouble and you think there’s
Test Fiber
a broken fiber, you would do the OTDR
Mechanical
OTDR Patch Panel
Box
test. But that doesn’t happen very often.
Patch Panel Fusion Splice
Splice
(Fiber End)
You’ve got to get really aggressive with
those cables to break anything,” he said.
B D
A C E
“Actually, copper cable is more sensitive
to bend radius than fiber is.”
When OTDR testing is necessary, a cen-
tralized design simplifies this procedure
too, because of the single setup. “And
you’re not moving a $30,000 piece of
equipment around,” Orr said.
Testing optical fiber involves working
with tried-and-true methods, which are
uncomplicated and reliable. The central-
ized network design makes that proce-
CBM
dure even simpler.
Douglas E. Harshbarger, RCDD, is
market development engineering man-
ager, premises systems, for Corning Inc.
0 300 600 1200
900
George Sellard is president of Sellard
Distance (feet or meters)
Communications. Special thanks to
Completing all testing from a single source remain stationary during testing Patrick Scanlon of Rochester Institute of
location not only saves time and effort, because any jostling can change the refer- Technology for his assistance.
but it also improves the accuracy of mea- ences,” Orr said. “So staying in one place All figures are from Siecor Universal
surements, according to Orr. helps.” Transport System Design Guide, Release
“It’s recommended that your light According to Orr, OTDR testing of cen- III, Siecor Corp., 1995.
PAGE 60 CBM DECEMBER 1998
Relative Loss (dB)