PXA-1000 User Manual
PXA-1000
Distributed Polarization
Crosstalk Analyzer – PolaX™
User Guide
Version: 2.1
Date: March 9, 2015
PXA-1000 User Guide
General Photonics Corporation is located in Chino California.
For more information visit the company's website at:
www.generalphotonics.com
or call 909-590-5473
GP-UM-PXA-1000-V21
Page 2 of 68
PXA-1000 User Guide
SAFETY CONSIDERATIONS
The following safety precautions must be observed during operation of this product.
Failure to comply with these precautions or with specific warnings elsewhere in this manual
violates safety standards of design, manufacture, and intended use of the product. General
Photonics assumes no liability for customers’ failure to comply with these requirements.
Before operation, the user should inspect the product and review the manual carefully.
Properly ground the chassis and work space using the chassis ground terminal.
Use only in a safe work environment in terms of temperature, humidity, electrical power and risk
of fire or shock. The product is designed for indoor use. Avoid exposure to liquids or water
condensation. Provide adequate ventilation for cooling.
Operate the product on a stable surface. Avoid excess vibration.
Standard laser safety procedures should be followed during operation.
Never look into the light source fiber connector when the light source is
turned on. THE OUTPUT LIGHT FROM A HIGH POWER LASER IS HARMFUL
TO HUMAN EYES. Follow industry standard procedures when operating a
high power laser source. Since the light from the PXA-1000 is invisible, it
is safer to turn it off before changing connections and when the light
source is not in use.
OPERATION CONSIDERATIONS
•
To ensure measurement accuracy, allow 10 minutes warm-up time before taking
measurements.
•
When powering the instrument off, wait at least 20-30 seconds before powering
it back on to avoid damage to electrical components.
•
The PolaX software program should be closed before powering off the PXA-1000.
GP-UM-PXA-1000-V21
Page 3 of 68
PXA-1000 User Guide
GP-UM-PXA-1000-V21
Page 4 of 68
PXA-1000 User Guide
Section 1.0
Overview ............................................................... 7
Section 2.0
Features ................................................................ 9
2.1 Front Panel and Optical Inputs ................................................ 9
Fiber Connectors ............................................................................9
Ferrule Cleaning Procedure ..........................................................10
2.2 Rear Panel: Electrical and Remote Control Interfaces ........... 11
Section 3.0
Operation Instructions ...................................... 13
3.1 Unpacking ............................................................................. 13
3.2 Setup ..................................................................................... 13
3.3 Software Interface Quick Reference ...................................... 15
3.4 Measurements ....................................................................... 21
Distributed Polarization Crosstalk Measurement (PM fiber) .........21
Polarization Extinction Ratio (PER) Measurement (PM fiber) .......27
Birefringence Measurement of PM Fiber .......................................31
Coherence Length Measurement of a Light Source .......................34
Polarization Extinction Ratio (PER) Measurement (System) ........38
3.5 Advanced Data Analysis ........................................................ 41
Graph Operations .........................................................................41
Data Interpretation ......................................................................47
Saving Data..................................................................................50
3.6 PER and Position Reference Calibrations (Optional Feature) . 51
X Position Reference ....................................................................51
PER Reference..............................................................................53
3.7 Troubleshooting .................................................................... 54
Section 4.0
Specifications ..................................................... 55
Optical..........................................................................................55
Electrical/Communication ............................................................55
Physical and Environmental .........................................................56
GP-UM-PXA-1000-V21
Page 5 of 68
PXA-1000 User Guide
Appendices ................................................................................. 57
Appendix 1.0 Comparison of PXA-1000 to Traditional White Light
Interferometer............................................................................ 57
Appendix 2.0 Spatial Resolution of PXA-1000 ............................. 59
Appendix 3.0 Polarization Crosstalk in PM Fiber ......................... 61
Classification of Polarization Crosstalk by Cause..........................61
Classification of Polarization Crosstalk by Measurement Results .61
Capabilities and limitations of the PXA-1000 ...............................66
GP-UM-PXA-1000-V21
Page 6 of 68
PXA-1000 User Guide
Section 1.0
Overview
The Distributed Polarization Crosstalk (X-Talk) Analyzer (PXA-1000) is a white light
interferometer designed to obtain space-resolved stress information by analyzing stress-induced
polarization cross-coupling along a length of polarization maintaining (PM) fiber. Its unique optical
design eliminates the strong zero-order interference and reduces the multi-coupling interference
common in traditional white light interferometers; as a result, the PXA-1000 has higher
measurement sensitivity, higher dynamic range, and higher spatial measurement accuracy than
traditional white light interferometers. The PXA-1000 enables the use of the PM fiber itself as the
sensing medium, eliminating the need to place multiple fiber gratings along the fiber. It can
therefore obtain higher spatial resolution of the stress distribution than grating-based systems.
Because no discrete sensing elements are required, the system is easy to install and calibrate,
making it ideal for monitoring space-resolved structural changes along bridges, tunnels, dams, oil
pipes, or buildings. It can also be used as an intrusion detection system, because any mechanical
disturbances to the PM fiber will cause polarization coupling. Another important application is PM
fiber quality inspection. The PXA-1000 easily identifies defective sections of PM fiber, enabling the
manufacturers or users to remove them. Furthermore, the PXA-1000 is ideal for quality inspection
and screening of PM fiber coils, since it can pinpoint the locations of imperfections or areas of local
stress on the fiber coil induced during the fiber winding process. The software displays the location
and polarization coupling ratio of each stress point as a function of distance. It also generates a
table listing the locations and polarization coupling strengths of all crosstalk peaks above a userdefined threshold. Other applications of the instrument includes measuring the extremely high
polarization extinction ratio of a polarizing waveguide, obtaining the autocorrelation function of a
light source, measuring the birefringence of a PM fiber and the lengths of PM and SM fibers, and
matching the optical path lengths of an interferometer.
More detailed information on measurement principles is provided in section 3.
Figure 1 PXA-1000 Distributed Polarization Crosstalk Analyzer
GP-UM-PXA-1000-V21
Page 7 of 68
PXA-1000 User Guide
GP-UM-PXA-1000-V21
Page 8 of 68
PXA-1000 User Guide
Section 2.0
Features
2.1 Front Panel and Optical Inputs
The front panel of the PXA-1000 is shown in Figure 2.
Figure 2 PXA-1000 front panel
Front panel features:
Power:
Power on/off switch
Light:
Light source safety key
Light Out:
Adapter (narrow-key PM FC/PC standard) for SLD output (to FUT)
Light In:
Adapter (narrow-key PM FC/PC standard) for interferometer input (from
FUT)
The recommended (default) connector type is a narrow-key FC/PC PM connector, although
other connector types are available by customer request. The connector keys are aligned to the
slow axis of the PM fiber.
Fiber Connectors
The front panel adapters are universal connector interfaces (UCI), which feature a maletype adapter top piece that can be removed for direct access to the ferrule end for routine
cleaning and maintenance without removing the entire adapter from the panel. This feature helps
avoid high insertion loss, high return loss and measurement instability caused by dirty or
contaminated connectors
External fiber connectors should be cleaned using industry standard cleaning methods
before connection to the PXA-1000. If this procedure is followed before each connection, the
instrument’s internal connector ferrules should not need regular cleaning. However, high insertion
loss or measurement instability that does not improve after cleaning the external connectors may
indicate that the instrument’s internal connector ferrules require cleaning.
GP-UM-PXA-1000-V21
Page 9 of 68
PXA-1000 User Guide
Ferrule Cleaning Procedure
Make sure light source is off before cleaning connectors.
Each connector ferrule is contained in a universal connector interface consisting of a front
piece that connects to the external fiber connector, and a base piece that is mounted on the front
panel of the instrument, as shown in Figure 3. To clean a connector ferrule, first, make sure no
external connector is connected to the universal connector interface. Then, using a Phillips
screwdriver, remove the two small screws connecting the front and back parts of the adapter, and
carefully pull the front flange straight out. (Note: never remove the adapter base from the front
panel). The ferrule end should now be exposed. Clean the ferrule using standard cleaning
procedures (compressed air or a fresh lint-free tissue and alcohol or other connector-cleaning
solvent), taking care to avoid scratching the ferrule surface. Finally, replace the front flange
(position it so that the key notch faces up, and the small alignment pin lines up with the hole in
the base piece, before pushing it in) and the screws.
Hole for
alignment pin
Remove screws
Front flange
Adapter basedo not remove
Ferrule end
Figure 3 Diagram of universal connector interface
GP-UM-PXA-1000-V21
Page 10 of 68
PXA-1000 User Guide
2.2 Rear Panel: Electrical and Remote Control Interfaces
The rear panel of the PXA-1000 is shown in Figure 4.
Figure 4 Rear panel
Rear Panel Features:
Cooling fan air intakes
USB interface port
Line: External AC input connector
: Chassis ground
The PXA-1000 uses a USB interface to communicate with the control computer. The
control program and USB driver are pre-installed on the control computer.
Fuse location:
Figure 5 shows the location of the fuse compartment
under the power cord plug. There are two fuses in the
compartment- the one in use and a spare. The fuse
further inside the compartment is active. The one
closer to the compartment opening is the spare.
Replace the fuse with one with the exact rating of the
original.
Figure 5 Fuse compartment
GP-UM-PXA-1000-V21
Page 11 of 68
PXA-1000 User Guide
GP-UM-PXA-1000-V21
Page 12 of 68
PXA-1000 User Guide
Section 3.0
Operation Instructions
3.1 Unpacking
Inspect PXA-1000 for any physical damage due to shipping and transportation. Contact
carrier if any damage is found. Check the packing list to see if any parts or accessories are
missing.
Packing List
Item #
Description
1
PXA-1000
2
Power cord
3
USB cable
4
User guide
6
Control computer (laptop) with control program and drivers pre-installed
7
Power supply for computer
8
PM patchcord
9
Manual polarization controller
3.2 Setup
1.
Connect instrument power cord and plug it into wall receptacle. Make sure the ground pin
of the power cord is connected to earth ground.
2.
Power on control computer.
3.
Connect instrument to computer with USB cable. Wait for “device detected” message to
appear.
4.
Connect input and output fibers (see next several sections for details on measurement
setups). Make sure SLD source is turned off while cleaning connectors and making
connections.
5.
Power on PXA-1000.
6.
Turn on the safety key for the internal light source, if applicable.
Note: The safety key enables control of the internal SLD light source. It does not by itself
GP-UM-PXA-1000-V21
Page 13 of 68
PXA-1000 User Guide
turn on the light source. Once the safety key is in the “on” position, the PolaX
measurement software is able to turn on the light source.
7.
Run the program “PXA-1000” from the desktop or Start menu shortcuts. The program files
are in the folder C:\DPXA\. The user interface screen shown in Figure 6 will appear.
8.
The PXA-1000 will run through an initialization sequence, which takes about 1 minute.
During this time, a “System is initializing” message will be displayed in the system
message box on the bottom left of the screen, and the progress bar immediately above
the message box will show the status of the process. Once initialization is complete, the
system is ready for measurement and the message box contents will change to “System is
ready to test”.
Setup and Analysis
Trace Display and Fitting Options
Data Display Window
System message box
Progress bar
Function keys
Figure 6 Main program interface
GP-UM-PXA-1000-V21
Page 14 of 68
PXA-1000 User Guide
3.3 Software Interface Quick Reference
This is a quick reference guide for the software interface. Individual features and functions
are described in more detail in the following sections.
Function Keys (bottom line of screen, or corresponding keys on keyboard)
F1
(About)
Displays the PolaX™ software version.
Features and interfaces described here correspond to PolaX v.2.2.
F2
(Run)
Executes the selected measurement.
F3
(Load)
F5
(Default)
F6
(Save)
Stores measured data to file.
F8
(Initialize)
Delay line initialization; resets position to zero.
(Exit)
Exits PXA-1000 control software PolaX.
The PXA-1000 will run through an exit sequence before exiting the
program. Do not disconnect or power down the instrument during this
process.
F9
Loads saved data from file.
This function does not require connection to the PXA-1000.
Resets system parameters to default values:
Delay Start: 0mm
Delay End: 50mm
Attenuation: 5 dB
Gain: Low
Δn: 5 × 10E−4
Plot Options (top left side of screen)
Curve Titles
Curve Icons
(gray = hidden)
GP-UM-PXA-1000-V21
Page 15 of 68
PXA-1000 User Guide
The plot options allow the user to customize curve appearance, as well as to export data
from the selected plot.
To customize the appearance of a plot, click on the icon for that plot to bring up a
customization pull-down menu. “Plot Visible” hides or displays the plot. The pull-down menu also
gives options to copy that curve’s data to the clipboard or export it to an Excel file.
Trace Display and Fitting Options (left side of screen)
Smooth
Gaussian
Cursor 1
Cursor 2
Marks
Options to apply or remove smoothing from one or more curves.
Curve smoothing method can be selected from the “Curve Analysis” tab
on the right side of the screen. Once smoothing is applied, original data
and curve are replaced by smoothed version.
Fits selected curve using Gaussian function for coherence length
measurement or general peak characterization. The fitting curve is
displayed in the graph window for comparison to the raw trace, and the
fitting parameters are displayed in the “Curve Analysis” tab on the right of
the screen.
Displays/removes cursor 1. When the cursor is present, its coordinates on
the active curve (selected at top right of screen) are displayed next to it.
Displays/removes cursor 2. When the cursor is present, its coordinates on
the active curve (selected at top right of screen) are displayed next to it.
Options to insert or remove markers 1 and 2. The selected marker is
inserted at the active cursor’s position. Marker coordinates are displayed
in the “Curve Analysis” tab on the right of the screen.
Note: Cursor must be on screen to insert marker.
GP-UM-PXA-1000-V21
Page 16 of 68
PXA-1000 User Guide
Curves
Curve options include:
Copy one trace to another
Clear one or more traces
Display
Display or hide one or more curves.
Sets x position of active cursor as x-reference (zero point) for active
trace.
X-Ref
Sets y position of active cursor as y reference (value set by the system)
for active trace.
Y-Ref
Measured optical power at interferometer input
(DUT output).
Power Monitor
Measurement Setup Options (tab on right side of screen)
Curve Selection
Select curve A, B, or C for Setup, Curve
Analysis, Fiber information, or Events.
Measurement
Xtalk Distribution
Birefringence
Coherence length
Raw Interference
PER
Sets system to measure polarization
crosstalk distribution.
Sets system to measure birefringence of
PM fiber.
Sets system to measure the coherence
length of an input light source.
Sets system to measure and display raw
interference signal data vs delay.
Measures the system’s polarization
extinction ratio (quick check for system
setup).
Delay Scan Range
Set up scan range for variable delay line
Start Position
0mm by default
Stop Position
Set end position depending on length of
fiber or waveguide to be measured.
SLD
Attenuation
GP-UM-PXA-1000-V21
Set attenuation of internal VOA
Page 17 of 68
PXA-1000 User Guide
Software switch turns light source on or
off. Running a measurement will
automatically turn on the light source.
Sets the amplifier gain to low or high.
When input optical power is low, setting
the amplifier gain to high will increase
SNR
and
system
measurement
sensitivity. The low gain setting is the
default; it works well for most
measurements.
Sets the birefringence of PM fiber under
test. This value is used to calculate the
positions of coupling points on the PM
fiber. To change the value, click on the
number and type in the desired value,
then press ENTER.
System calibration functions
(may not be available to all users)
Calibrates the system using a 0 dB
calibration standard.
SLD switch
Gain
Δn of PM fiber
Calibration
PER Calibration
Set Mark1 as Zero
Position
Sets Marker 1 position as x reference (0
position) for the system.
Curve Analysis (tab on right side of screen)
The top block summarizes the x
coordinates
and
corresponding
y
coordinates on the selected curve of
cursors 1 and 2. Y-coordinates on other
curves can be displayed by selecting
another curve at the top of the screen.
The next block summarizes the x
coordinates
and
corresponding
y
coordinates of markers 1 and 2, as well
as the x and y distances between them.
Markers are placed on specific curves.
Displays the fit results with Gaussian fit
Cursor positions
Marker positions
model
Gaussian
Results
Fit
⎛ ( x − x0 ) 2
a ⋅ exp⎜⎜ −
2σ 2
⎝
⎞
⎟,
⎟
⎠
where a is
the amplitude, σ is the standard
deviation (STDV) and x0 is the center.
Bandwidth is the 3dB bandwidth of the
Gaussian fit curve. The fit residual is
defined by
1
N
N −1
∑(y
i =0
fit ,i
− y raw,i ) 2
.
Smoothing Method
Third-order
polynomial
Moving Average
GP-UM-PXA-1000-V21
Obtains a smooth curve using the thirdorder polynomial intercept method.
Obtains a smooth curve using the
moving average method:
y[i]=(y[i]+y[i+1]+…+y[i+n])/n, where
n is the average number.
Page 18 of 68
PXA-1000 User Guide
Moving Average
number
Sets the average number for moving
average method.
Fiber (tab on right side of screen)
Fiber measurement data
Fiber PER Analysis
PER including
internal fiber
Estimated PER of
FUT
Set Fiber START
Set Fiber END
Calculation Method
Calculate PER of FUT
Fiber
Birefringence
Length of Fiber
Delay between Input
and Output (mm)
Measured
Birefringence
Beat Length
Recalculate
GP-UM-PXA-1000-V21
Measured PER of the entire system,
including internal fiber in the PXA-1000.
Calculated PER of the fiber under test
(FUT) over the designated range.
Sets start point for PER calculation and
events table.
Sets end point for PER calculation and
events table.
Select calculation method for PER
measurement.
Peak: Include all peaks ≥threshold value
within designated range in the
calculation.
Area: Integrate over area under the
curve within the designated range
to calculate PER.
Recalculate PER after changing settings
(calculation method, fiber range, or peak
threshold)
Data summary for birefringence
measurement
Length of fiber under test (input by
user)
Measured delay for fiber under test
Measured birefringence of fiber under
test
Calculated beat length of fiber under
test.
Recalculate birefringence and beat
length after changing measurement
settings (e.g. fiber length).
Page 19 of 68
PXA-1000 User Guide
Events (tab on right side of screen)
The Events tab provides a table summarizing the position and amplitude of all measured
crosstalk peaks between the designated start and end points (specified on the “Fiber” tab) that
exceed the user-defined threshold and width settings.
Sort by:
Table
Xtalk (Average)
Xtalk (Maximum)
Save Events
Peak Detection
Parameters
Threshold
Width
Crosstalk peaks can be sorted by
position or amplitude
Lists position and amplitude of all peaks
between Marker 1 and Marker 2 that
meet the threshold criteria.
Lists average crosstalk value, calculated
using peaks listed in the table.
Lists the maximum crosstalk, from
peaks listed in the table.
Saves data in peak table to a file.
Saved data includes index, position, and
amplitude of listed peaks.
Define threshold parameters for table
listings. To change a parameter, select
it, type the new value, and press ENTER.
The table will update automatically.
“Threshold” rejects peaks that are too
small. The peak detection function
ignores any peak found whose fitted
amplitude is less than the specified
threshold.
“Width”
specifies
the
number
of
consecutive data points to use in the
quadratic least squares fit. The width is
coerced to a value greater than or equal
to 3. The value should be no more than
about 1/2 of the half-width of the peaks
and can be much smaller (but > 2) for
noise-free data.
The peak analysis region can be modified by changing the values of “Fiber START” and
“Fiber END” on the “Fiber” tab.
GP-UM-PXA-1000-V21
Page 20 of 68
PXA-1000 User Guide
3.4 Measurements
The PXA-1000 can perform several types of measurements related to polarization
crosstalk. Measurement principles, setup descriptions, and results analysis for each type of
measurement are described in this section.
Distributed Polarization Crosstalk Measurement (PM fiber)
Polarization maintaining (PM) fiber is widely used in fiber optic sensor systems. The
performance of such a sensor system is directly limited by the quality of the PM fiber coil,
especially by the magnitude of cross coupling between the two principal polarization modes. The
PM fiber itself may have imperfections such that the local intrinsic stresses may cause polarization
cross coupling. External stress on the PM fiber during the fiber winding process may also cause
polarization cross coupling. In order to optimize PM fiber coil quality, one would first select PM
fiber with low intrinsic polarization cross coupling and then wind the fiber so as to minimize
externally induced cross coupling. In practice, one would also like to identify the exact stress
points which cause polarization coupling during the winding process so as to remove the induced
stress as it occurs. The distributed polarization crosstalk analyzer (PXA-1000) can reveal the
location and magnitude of polarization coupling induced by both intrinsic and external stresses.
PXA-1000 System Schematic
A
B
C
Δz
z=Δz/Δn
Δz = Δnz
PM fiber coil
SLD
PM fiber
Digital
circuit
Polarizer, 45 deg alignment
Coupler
Mirrors
Variable delay line
Amplifier PD
Delay line control
Figure 7 PXA-1000 system schematic
GP-UM-PXA-1000-V21
Page 21 of 68
PXA-1000 User Guide
As shown in Figure 7, the PXA-1000 system consists of three main parts: light source,
interferometer and light detection/processing circuit. The light input to the DUT is linearly
polarized and aligned to the slow-axis of the PM fiber, with a 30nm bandwidth and 1310nm or
1550nm center wavelength. Its power can be adjusted by a variable optical attenuator (VOA).
After passing through the DUT, the light aligned to the fast- and slow axes are mixed together
through an analyzer, generating interference peaks in a fiber based interferometer as the delay
line is adjusted.
Distributed Polarization Crosstalk Measurement Principle
slow axis
fast axis
force
A
Fixed
mirror
B
C
z
45° oriented
polarizer
BS
Moving
mirror
(ns-nf)z
PD
Figure 8 Polarization crosstalk measurement principle
As shown in Figure 8, when light enters the PM fiber at position A, it has only one
polarization component, aligned to the slow axis (red). Stress at position B induces polarization
coupling and produces a polarization component aligned to the fast axis (blue). Because the two
polarization components travel at different group velocities, at the output of the fiber (position C),
the two components will experience a delay difference
Δz = n s z − n f z = Δnz
(1)
where ns and nf are the refractive indices of the slow and fast axes, respectively, their difference
Δn is the birefringence, and z is the distance between the coupling point B and the output point C.
If a polarizer oriented at 45 degrees from the slow axis is placed at the fiber output, half of the
power in each of the polarization components will pass through the polarizer and emerge with the
same polarization state (linear, aligned to the polarizer axis). Consequently, they will interfere in
the Michelson interferometer shown in Figure 8 to produce interference peaks as the delay is
adjusted. The distance between the two adjacent interference peaks is Δnz; therefore, from Eq.
(1), the location of the coupling point is
z = Δz Δn .
The coupling point can therefore be located
using the interference graph. The coupling ratio can also be calculated from the strength of the
interference peaks.
GP-UM-PXA-1000-V21
Page 22 of 68
PXA-1000 User Guide
Setup for Distributed Polarization Crosstalk Measurement
The recommended distributed polarization crosstalk measurement setup is shown in Figure
9. All fibers used in the measurement should be PM fibers with connector keys aligned to their
slow axes. The external connectors should be carefully cleaned before connecting to the PXA1000.
FUT
Light output (FUT in)
Light input (FUT out)
Figure 9 Distributed polarization crosstalk measurement setup
Distributed Polarization Crosstalk Measurement Procedure (for PM fiber)
1.
Connect fiber under test (FUT) as shown in Figure 9.
GP-UM-PXA-1000-V21
Page 23 of 68
PXA-1000 User Guide
2.
On the measurement information section on the right side of the screen, select the curve (A,
B, or C) on which to display measured data.
3.
Select “Xtalk Distribution” from the Measurement menu on the Setup tab.
4.
Set the delay scan range. The start position is fixed at 0mm. Set the “Stop Position” to the
desired value (range is 0 to 700 mm for the 1.3 km PXA-1000 or 0 to 1500 mm for the 2.6 km
PXA-1000). Generally, the “Stop Position” should be longer than the minimum delay calculated
by
Minimum Delay= fiber length* Δn*1000 (mm)
(2)
The “Stop Position” can be changed either by using the up/down arrow buttons next to the
value or by selecting the value and typing in a new number.
5.
Check the attenuation setting.
a.
Make sure the light safety key is in the “on“ position.
b.
If the SLD on/off button is red, as in the example above, click the button
to turn on the SLD.
c.
Click the “Power Monitor” button on the lower left side of the screen.
Check that the optical power is at an appropriate level. Power ~1mW or so
GP-UM-PXA-1000-V21
Page 24 of 68
PXA-1000 User Guide
usually works well, although higher power may be necessary for sensitive
measurements. If the power is too high or too low, change the attenuator
setting (range 0 to 30 dB) and check the power again.
d.
If the measurement is run with the power level too high, the detector may
saturate and the resulting data may not be accurate. In this case, the
PXA-1000’s message box will show “Detector is saturated. Please increase
attenuation of VOA”.
e.
The optical power level affects the position and width of the noise floor of
the measurement. If the noise floor is too high or too broad, the optical
power may be too low.
6.
Set the amplifier gain. “Low gain” is recommended for most applications. “High gain” can be
used for fiber systems with very high loss.
7.
Set the birefringence of the PM fiber under test. If the birefringence is not known, measure it
by following the procedure described in the next section. The birefringence setting can be
changed by selecting the value and typing the new value in the box. The fiber birefringence
setting determines the x positions of data points if the scale is set to “fiber length”.
8.
Click the “Run” button at the bottom of the screen or press the F2 function key on the
computer keyboard to start the measurement. During measurement, the message “Measuring
xtalk” will be displayed in the message box at the bottom left of the screen, and the progress
bar will indicate the status of the measurement.
Progress bar
Message box
9.
When measurement is finished, the x-talk vs. position trace will be displayed (see below). The
x position of the PXA light input connector (which connects to the DUT output) is zero on the
plot (Marker 1). To the left of that is a peak corresponding to an internal input reference point.
Marker 2 is placed at the PXA-1000 light output connector (DUT in). To the right of that is a
peak corresponding to an internal output reference point.
Section 3.5 provides more details on analysis of the measured results.
GP-UM-PXA-1000-V21
Page 25 of 68
PXA-1000 User Guide
Input connector
Input reference
Output connector
Output reference
Noise floor
Figure 10 Crosstalk measurement of a 3m PM fiber.
Default plot shows crosstalk (dB) vs. fiber length (m).
GP-UM-PXA-1000-V21
Page 26 of 68
PXA-1000 User Guide
Polarization Extinction Ratio (PER) Measurement (PM fiber)
PER Determination from Distributed Polarization Crosstalk Measurement
Polarization extinction ratio (PER) is the ratio between the power in the principal
polarization component of a light beam and the power in the orthogonal polarization component,
expressed in dB. It is a measure of the linearity and degree of polarization of a polarized light
source, or of the polarization preserving or suppressing properties of a fiber or optical component.
It is one of the principal parameters used to evaluate the quality of a PM fiber. When evaluating a
PM fiber or PM fiber system, the principal polarization component is usually the one aligned to the
slow axis of the PM fiber. In this case, the PER can be expressed as:
(3)
PER = 10 log(Ps Pf )
where Ps is the total power in the slow axis, and Pf is the total power in the fast axis.
Most PER meters and measurement systems can only measure the total PER of the fiber or
system under test. Because of this, it is impossible to isolate the effects of particular connectors or
splice points. In addition, the measured value is heavily dependent on the alignment of the input
light at its launch point into the FUT, which limits both the accuracy and repeatability of the
measurement.
As described in the previous section, the PXA-1000 measures crosstalk at all points along
the fiber under test. The PER of a particular section of the FUT can be calculated by integrating the
effects of all crosstalk events between the designated points on the fiber.
When a distributed polarization crosstalk measurement is done, the PXA-1000 provides
measurement results for system PER and PER of the fiber under test on the Fiber tab (Figure 11).
“PER including Internal Fiber” is the PER of the entire system, including the effects of
internal fibers and reference points in the PXA-1000.
“Estimated PER of FUT” is the PER of the fiber under test (FUT), calculated using the
crosstalk contributions from all crosstalk events that meet the threshold conditions (as listed in
the events table) between the designated start point and the designated end point for the
calculation.
When
the
distributed
crosstalk
measurement
is
first
completed,
the
algorithm
automatically identifies and excludes the contributions of misalignments at the FUT input and
output connectors. The start point for the PER measurement is denoted by Marker 1, which is
automatically placed at the PXA-1000 input connector (FUT output), and the end point for the PER
measurement is set at Marker 2, which is automatically placed at the PXA-1000 output connector
(FUT input).
GP-UM-PXA-1000-V21
Page 27 of 68
PXA-1000 User Guide
System PER, includes the effect of
internal fiber in the PXA-1000
PER of DUT, calculated using the
measured crosstalk from the
designated start point
to the designated end point
Figure 11 Results of polarization crosstalk measurement of 388m fiber coil,
with closeup of PER results section on Fiber tab
GP-UM-PXA-1000-V21
Page 28 of 68
PXA-1000 User Guide
Fiber Range for PER Measurement
The user can choose to calculate the PER between any two points on the FUT (for
example, to exclude the effects of connectors, splice points, etc.) by changing the start and end
points used for the PER calculation. Figure 12 shows some examples.
Figure 12 Polarization crosstalk curves of a 13 m jumper with two FC/APC connectors a) and a 250
m PM fiber coil spliced with two FC/APC connectors. PER measurement with a commercial PER
meter always includes the contributions of the input connector and two splices, while the PXA1000 has the ability to identify and eliminate the polarization crosstalk contributions of all
connectors and splices in the measurement system. Note that fiber length in the horizontal axis is
obtained by dividing the fiber delay line distance ∆Z by the average birefringence obtained using
the procedure described in Section 3.2 of the referenced paper. 1
Start or end points can be set to an active cursor position (place cursor, drag to desired
position, and click “Set Fiber START” or “Set Fiber END”) or can be set to particular values by
selecting the values in the start or end point boxes and typing in new values.
Note: either cursor 1 or cursor 2 can be used to reset the fiber start or end position. If
both cursors are on the screen, the active cursor is the one most recently moved.
PER Calculation Method
There are two methods available for PER calculation. “Peak” uses all crosstalk peaks within
the designated range that meet the user-defined peak threshold criteria (defined on the Events
tab on the right side of the screen). This method is generally more suited to a FUT that includes
multiple discrete crosstalk peaks (for example, from multiple connector interfaces, splice points, or
localized stress points on the fiber).
“Area” uses the integrated area under the curve to calculate the PER. This method is
generally better for a single length of fiber or a system with more continuous or quasi-continuous
coupling rather than discrete crosstalk events. If this method is selected when the distributed
crosstalk measurement is made, the initial PER measurement is made from the input connector to
the output connector and will include half of the area under the peaks for the input and output
1
Zhihong Li, X. Steve Yao, Xiaojun Chen, Hongxin Chen, Zhuo Meng, Tiegen Liu, “Complete characterization of
polarization-maintaining fibers using a distributed polarization crosstalk analyzer” Preprint paper.
GP-UM-PXA-1000-V21
Page 29 of 68
PXA-1000 User Guide
connectors. The user can choose to exclude these peaks by redefining the start and end points for
fiber PER measurement.
Cursor 1
Cursor 2
PER of DUT, recalculated with new
start and end points
Start point: Cursor 1 position
End point: Cursor 2 position
Peak threshold setup (from Events tab)
used for PER calculation
Click “Calculate PER of FUT” button to
recalculate PER after changing setup
parameters (fiber range, peak threshold, etc.)
Figure 13 PER measurement of same 388m fiber coil as in Figure 11, with different fiber range
GP-UM-PXA-1000-V21
Page 30 of 68
PXA-1000 User Guide
Birefringence Measurement of PM Fiber
Birefringence Measurement Principle
slow axis
fast axis
A
B
ΔnL
L
PXA-1000
light output
PXA-1000
input
Figure 14 Birefringence measurement principle
As shown in Figure 14, when light enters the PM fiber under test from the PXA-1000 light
source output position A, two polarization components aligned to the slow axis and fast axis,
respectively, of the fiber are launched because of the misalignment between the PM fiber output of
the PXA-1000 light source and the fiber under test. Due to the birefringence of the PM fiber, the
two polarization components travel at different group velocities and experience a delay difference
at the end (position B) of the fiber under test:
Delay slow− fast = n s L − n f L = ΔnL
(4)
where L is the length of the fiber under test.
The delay
slow-fast
generated by the PM fiber under test can be measured by adjusting the
delay line in the PXA-1000’s interferometer. The birefringence of the PM fiber can therefore be
calculated by
Δn = n s − n f =
Delay slow− fast
(5)
L
It should be noted that the delay measured by the PXA-1000 is related to the group
velocity of light in the fiber, not the phase velocity. Therefore, the measured birefringence is group
birefringence, not phase birefringence, and the beat length calculated by λ/Δn is consequently
only an approximate value.
Setup for Birefringence Measurement
The recommended setup for measuring the birefringence of PM fiber is shown in Figure 15.
The slow axis of the PM fiber under test should be aligned to the connector key.
GP-UM-PXA-1000-V21
Page 31 of 68
PXA-1000 User Guide
PM fiber
under test
Light output (DUT in)
Light input (DUT out)
Figure 15 Setup for birefringence measurement of PM fiber
Birefringence Measurement Procedure (for PM fiber)
1.
Precisely measure the length of the PM fiber under test, then connect it to the PXA-1000 as
shown in Figure 15.
2.
On the measurement information section on the right side of the screen, select the curve (A,
B, or C) on which to display measured data.
3.
Select “Birefringence” from the Measurement menu on the Setup tab.
4.
Set the delay scan range. The start position is fixed at 0mm. Set the “Stop Position” to the
desired value. For short fibers (up to several tens of meters), the default value of 50mm
should be sufficient. For longer fibers, the delay range should be increased.
5.
Check the attenuation setting.
GP-UM-PXA-1000-V21
Page 32 of 68
PXA-1000 User Guide
a.
Make sure the light safety key is in the “on” position.
b.
If the SLD on/off button is red, click the button to turn on the SLD.
c.
Click the “Power Monitor” button on the lower left side of the screen.
Check that the optical power is at an appropriate level. Power ~1mW or so
usually works well. If the power is too high or too low, change the
attenuator setting (range 0 to 30 dB) and check the power again.
6.
Set the amplifier gain. “Low gain” is recommended for most applications. “High gain” can be
used for fiber systems with very high loss.
7.
Click the “Run” button at the bottom of the screen or press the F2 function key on the
computer keyboard to start the measurement. During measurement, the message “Measuring
birefringence” will be displayed in the message box at the bottom left of the screen.
8.
When the measurement is finished, the message “Birefringence measurement is done” is
displayed, and the dialog box shown below appears. Enter the length of the PM fiber and click
“OK”.
9.
After the fiber length is entered, the birefringence will be calculated. Go to the Fiber tab on the
right side of the screen to view the data. The “Fiber Birefringence” box shows the fiber length,
measured delay between input and output connectors, measured birefringence, and estimated
beat length. The user can change the fiber length used for birefringence calculation by
selecting the value in the “fiber length” box and typing in a new value, then clicking the
“Recalculate” button.
On the plot, Marker 1 and Marker 2 indicate the PXA-1000 input and output connector
positions, respectively. The default x-axis for the plot is delay, so the distance between the
markers is the measured delay D = (ns-nf)L, where L is the fiber length.
GP-UM-PXA-1000-V21
Page 33 of 68
PXA-1000 User Guide
Input connector
Output connector
Delay = (ns−nf)L
Figure 16 Birefringence measurement of a 3m PM fiber.
Default plot shows crosstalk (dB) vs. delay (mm)
Coherence Length Measurement of a Light Source
Measurement Principle
The interference pattern generated by the PXA-1000 is shown in Figure 17. It has the form
of an amplitude modulation signal. The modulation frequency is a constant determined by the
center wavelength of the light source and the sweep speed of the delay line. The amplitude of the
envelope represents the x-talk magnitude, and its full width at half maximum (FWHM) is related to
the linewidth of the light source.
FWHM =
λ2
2 ln 2 λ20
≈ 0.44 0
π Δλ
Δλ
(6)
where λ0 is the center wavelength of the input light source and Δλ is the FWHM of the
power spectrum of the light source. The coherence length of the light source can be calculated by
l c = FWHM *
GP-UM-PXA-1000-V21
π
2 ln 2
≈ 2.266 * FWHM
(7)
Page 34 of 68
PXA-1000 User Guide
FWHM
z
z
Envelope of interference signal
after demodulation circuit
Interference signal of coupling
point when delay line is moving
Figure 17 Interference signal and its envelope
Setup for Coherence Length Measurement
The recommended setup for coherence length measurement of a light source is shown in
Figure 18. If the degree of polarization (DOP) of the light source under test is very low, the
interference signal can be too small to be detected by the PXA-1000. It is therefore recommended
that a polarizer be placed between the light source and the PXA-1000 input. A PM pigtailed
polarizer is preferred because SM pigtail fiber may result in coupling between different interference
peaks and reduce the measurement accuracy.
PM fiber
Light source
under test
Polarizer
Figure 18 Setup for coherence length measurement
GP-UM-PXA-1000-V21
Page 35 of 68
PXA-1000 User Guide
Coherence Length Measurement Procedure
1.
Connect light source under test to a PM pigtailed polarizer, then connect the polarizer’s PM
output to the PXA-1000’s input connector.
2.
On the measurement information section on the right side of the screen, select the curve (A,
B, or C) on which to display measured data.
3.
Select “Coherence Length” from the Measurement menu on the Setup tab.
4.
The coherence length measurement uses the default delay range settings.
5.
Check the light source power level.
a.
Click the “Power Monitor” button on the lower left side of the screen.
Check that the optical power is at an appropriate level. Power ~1mW or so
usually works well. If the power is too high or too low, change the power
level of the light source (if possible) and check the power again.
6.
The coherence length measurement uses the default gain setting.
7.
Click the “Run” button at the bottom of the screen or press the F2 function key on the
computer keyboard to start the measurement. During measurement, the message “Coherence
GP-UM-PXA-1000-V21
Page 36 of 68
PXA-1000 User Guide
length is being measured” will be displayed in the message box at the bottom left of the
screen, and the status of the measurement will be indicated by the progress bar.
8.
When the measurement is finished, the message “Coherence length measurement is done” is
displayed in the message box at the bottom left of the screen.
b) Set scale to linear
c) Set y-scale limit
d) Fit the peak
a) Select zoom tool and zoom
in on one peak
Figure 19 Initial data display after coherence length measurement
Plot shows crosstalk (dB) vs. delay (mm).
9.
The plot will show the crosstalk (dB) vs. delay (mm).
a.
Select and zoom in on one peak from the measured curve (see section 3.5
for details on zooming the graph). For best results, choose a peak with a
good shape that does not overlap with other peaks.
b.
Set the y-scale to linear using the y-scale pull-down menu at the top left
of the plot.
c.
If necessary, change the y-scale limits by selecting the value of the upper
y-scale limit and typing in a new value, then pressing ENTER.
d.
From the “Gaussian” pull-down menu on the left of the screen, select the
Gaussian fitting option for the active curve (curve A in the example
above). The calculated coherence length will be displayed in the message
box at the bottom of the screen. The corresponding Gaussian fitting curve
is displayed as a red line on the plot and the fitting parameters are
displayed in the Curve Analysis tab on the right of the screen.
GP-UM-PXA-1000-V21
Page 37 of 68
PXA-1000 User Guide
Measured data (white)
Fitted curve (red)
Estimated coherence length
Gaussian fitting parameters
Figure 20 Coherence length measurement results after curve fitting
Polarization Extinction Ratio (PER) Measurement (System)
In addition to the PER measurement from the crosstalk data, the PXA-1000 has a quick
system PER measurement function that can be used to check a measurement setup before doing
crosstalk characterization of a device or fiber under test.
In general, this function gives the same PER value as the “PER including internal fiber”
calculated from the distributed polarization n crosstalk measurement.
Measurement Principle
For the quick PER measurement, the PXA-1000 uses an optical switch to block or transmit
slow axis-aligned light. Assuming that the measured light powers are Ptotal and Pfast, respectively,
when the slow axis is unblocked or blocked, the PER of the system including the PM fiber under
test can be calculated by
PER = −10 log(
GP-UM-PXA-1000-V21
Pfast
Ptotal − Pfast
)
(8)
Page 38 of 68
PXA-1000 User Guide
It should be noted that the measured PER includes the x-talk generated by the
misalignment at both the input and output connectors, so the measured PER is not equal to the
PER of the PM fiber under test.
Setup for PER Measurement
PM fiber
under test
Light output (DUT in)
Light input (DUT out)
Figure 21 Setup for PER measurement
PER Measurement Procedure
1.
Connect DUT as shown in Figure 21.
2.
Select “PER” from the Measurement menu on the Setup tab.
3.
Click the “Run” button at the bottom of the screen or press the F2 function key on the
computer keyboard to start the measurement. During measurement, the message “PER is
being measured” will be displayed in the message box at the bottom left of the screen.
4.
When the measurement is finished, the measured PER will be displayed in the message box at
the bottom left of the screen.
GP-UM-PXA-1000-V21
Page 39 of 68
PXA-1000 User Guide
GP-UM-PXA-1000-V21
Page 40 of 68
PXA-1000 User Guide
3.5 Advanced Data Analysis
Graph Operations
Using the graph operation functions shown in Figure 22, the user can zoom in and out in
the data plot, use cursors to read the coordinates of measured points, place two markers on the
curves and measure the coordinate differences between the two markers.
Markers
Cursor and marker coordinates
Trace Display and Fitting Options
Cursors
Graph palette
Figure 22 Graph operations interface
GP-UM-PXA-1000-V21
Page 41 of 68
PXA-1000 User Guide
Graph Palette
The graph palette can be used to move cursors and to
zoom and pan the graph display. Click the
corresponding button in the graph palette to enable
cursor movement, display zooming, or display panning.
Each button has a status indicator in its upper left
corner which turns green when that option is enabled.
•
Cursor
movement tool
Panning
tool
•
•
Cursor Movement Tool (graph only)—Allows
cursor or marker to be dragged on the display.
Zoom—Zooms in and out of the display.
Panning Tool—Picks up the plot and moves it
around on the display.
Zoom
The Zoom tool (middle button on the graph palette)
allows the user to zoom in or out on the graph. When
the Zoom tool is clicked, a pop-up menu of zoom
options appears. This menu is shown below.
Zoom by selection rectangle. Only the area in the
selected rectangle is displayed.
Zoom by rectangle; with zooming restricted to x data,
(the y scale remains unchanged).
Zoom by rectangle; with zooming restricted to y data,
(the x scale remains unchanged).
Undo last zoom. Resets the graph to its previous
setting.
Zoom in about a point. If you hold down the mouse on a
specific point, the graph continuously zooms in until you
release the mouse button.
Zoom out about a point. If you hold down the mouse on
a specific point, the graph continuously zooms out until
you release the mouse button.
GP-UM-PXA-1000-V21
Page 42 of 68
PXA-1000 User Guide
Graph Scale Options
Editable text
Figure 23 Graph scale options
The y axis parameter is always crosstalk or interference amplitude. The scale of the plot
can be log scale (dB) or linear. The default is log scale, because smaller peaks may not be visible
in linear scale; however, linear scale is useful for viewing some data and for curve fitting.
The x axis parameter can be the position along the fiber (fiber length) or the delay. The
default parameter depends on the measurement being made. If the x parameter is fiber length, it
will be expressed in meters. If the x-parameter is delay, it can be displayed in terms of length
(mm) or time (ps).
Both the x and y scale limits can be edited by selecting the text and typing in new values.
This can be a more precise way of rescaling the plot than using the zoom options.
GP-UM-PXA-1000-V21
Page 43 of 68
PXA-1000 User Guide
Cursor Options
Curve A is selected
Cursor pull-down menus
Curve Analysis
Figure 24 Cursor control
The cursor pull-down menus on the left side of the screen allow the user to add or delete
cursors. Selecting “Create Cursor x” from a cursor menu causes the corresponding cursor to
appear on the plot. With the cursor-movement tool
selected, the cursor can be moved to
different positions on a measured curve using the mouse. The coordinates (x, y) of the current
cursor position will be displayed next to the cursor. The (x, y) coordinate display can also be
moved with the mouse. The cursor coordinates are also shown in the table on the Curve Analysis
tab on the right side of the screen.
GP-UM-PXA-1000-V21
Page 44 of 68
PXA-1000 User Guide
If there is data for more than one curve on-screen, the user can select which curve’s
coordinates will be displayed for a particular cursor position. In the example shown in Figure 24,
curve A is selected, so the cursor coordinates displayed on the plot and in the curve analysis table
are the coordinates on curve A that correspond to the current positions of cursor 1 and cursor 2. If
the user selects curve B, the coordinates displayed will be those that correspond to the cursor
positions on curve B. This is useful for comparing peak heights or widths on different curves.
For example, in the plot above, curves A, B, and C are the measured crosstalk in the same
test fiber with different amounts of pressure applied at a particular point. The effects can be
compared by placing a cursor at the peak in the center of the plot and switching the curve
selection between A, B, and C.
Cursors can also be used to update the fiber range to be used for PER calculation and peak
detection. After a crosstalk measurement is made, the PER will be calculated for the FUT between
Marker 1 and Marker 2 (PXA-1000 input and output connectors, respectively). This is also the
range used for peak detection for the Events table. The user can select a more limited range to be
used for the PER measurement and peak detection by dragging a cursor to the desired position on
the plot, then clicking on “Set Fiber START” or “Set Fiber END” on the Fiber tab on the right side of
the screen. Then, click “Calculate PER of FUT” to recalculate the PER over the new range. The
Events table will also update to cover the new range.
Marker Options
GP-UM-PXA-1000-V21
Page 45 of 68
PXA-1000 User Guide
The marker pull-down menu on the left side of the screen allows the user to add or delete
markers. Selecting “Create Mark x” from the menu adds (or moves, if the marker is already onscreen) the specified marker at the current active cursor position. If both cursor 1 and cursor 2
are on-screen, the active cursor position is the position on the selected curve of the cursor that
was most recently moved. The coordinates of the markers are displayed in the table at the top
right of the screen. The coordinate differences between the two markers are automatically
calculated and displayed in the bottom line of the table on the Curve Analysis tab on the right side
of the screen.
Unlike the cursors, the marker coordinates do not change with curve selection. In fact, the
two markers can be placed on different curves. This is also useful for calculating differences in
position or amplitude of crosstalk events on different curves.
Curve Smoothing
In most cases, curve smoothing is not necessary, but occasionally, it can be useful to help
enhance hard-to-resolve crosstalk features.
The bottom section of the Setup tab allows the user to select the smoothing method to be
used, and for the moving average method, to specify the moving average number. Smoothing can
be applied to or removed from one or more curves from the pull-down menu at the left of the
screen.
Smoothing pull-down menu
Smoothing method selection
on Setup tab
The third-order polynomial method is useful to resolve detailed features of peaks. It fits
the peaks with no changes to peak widths.
GP-UM-PXA-1000-V21
Page 46 of 68
PXA-1000 User Guide
The moving average method can change the width and amplitude of crosstalk peaks, so it
is not recommended for detailed analysis of individual peaks. However, it is useful for separating
crosstalk features from noise when the noise is too high.
Gaussian Fit
The Gaussian Fit function was described in the Coherence Length Measurement section.
Besides its use for light source coherence length measurement, it is also useful for characterizing
individual crosstalk peaks. It can be used to determine the center and width of a particular peak.
Data Interpretation
Figure 25 shows a typical x-talk measurement curve. Marker 1 and Marker 2 show the xtalk peaks generated by the input and output connectors, respectively, of the PXA-1000. The
curve between the two markers represents the x-talk distribution of the PM fiber under test. There
are also some small peaks outside the region bounded by the two markers. These are caused by
x-talk reference points built into the system for x-talk calibration and from the misalignment
between the polarizer chip and the pigtailed PM fiber.
Estimated PER of
FUT/DUT without
connectors
PER of entire
system including PM
fiber in PXA-1000
Input connector
Output connector
x-talk reference
Output reference
(generated by pigtailed
polarizer just after SLD
source)
Noise floor
Figure 25 Measurement example 1: 3m PM patchcord
GP-UM-PXA-1000-V21
Page 47 of 68
PXA-1000 User Guide
Figure 26 Measurement example 2: PM fiber coil
Figure 26 shows the results of a distributed polarization crosstalk measurement of a fiber
coil. Marker 1 and Marker 2 show the positions of the input and output connectors of the coil.
Between them are several smaller crosstalk peaks at semi-regular intervals, possibly due to
winding defects.
The table in the Events tab on the right of the screen summarizes the crosstalk peaks
within the analysis area defined by Markers 1 and 2 that meet the threshold conditions. The “Sort
by” pull-down menu at the top of the table allows the user to sort peaks by position or amplitude.
In this example, the table lists the amplitudes and locations of all crosstalk peaks >−50 dB.
Clicking on the index number of any peak in the table causes a cross-shaped marker to appear at
that peak (the marker is at peak 6 in this example).
The average and maximum crosstalk, calculated from the peaks listed in the table, are
shown below the table.
The threshold level can be changed by typing in the text box. After changing the
threshold, press ENTER or click outside the text box to update the table to include peaks above
the new threshold.
The peak analysis region can be redefined by moving Markers 1 and 2 on the curve to be
analyzed. The table will update automatically.
The information in the table (index, position, and amplitude of crosstalk peaks) can be
saved to a file by clicking on the “Save Events” button. The user will be prompted for a filename
and location to which to save the data. The saved data file will have file extension “.pks” to avoid
GP-UM-PXA-1000-V21
Page 48 of 68
PXA-1000 User Guide
confusion with other types of data files, but it can be opened using applications such as Notepad
or Excel.
Pressure applied at this point
on the fiber
Figure 27 Measurement example 3: PM patchcord with 900µm loose tube, with different amounts of
pressure applied over a 1cm section of fiber.
Curve A (blue): no pressure applied.
Curve B (green): 1kg applied
Curve C (red): 500g applied
Figure 27 illustrates how the PXA-1000 can detect changes in crosstalk in a fiber due to
environmental factors. It shows 3 consecutive crosstalk measurements of a PM patchcord with
900µm jacket, with different weights placed over a particular 1cm section of the fiber. Curve A
(blue) is the baseline measurement with no weight applied. Curve C (red) shows the same fiber
with a 500g weight placed on it. The weight causes crosstalk of about −55 dB at the point where it
was applied, but the rest of the curve is relatively unchanged. Curve B (green) shows the same
fiber with a 1kg weight placed on it at the same point. The crosstalk at that point increases to
almost −35 dB.
Note that many factors can affect the amount of crosstalk caused by pressure on a fiber,
including the axis along which the pressure is applied (for example, pressure applied along the
slow or fast axis of a fiber will have a different effect than the same pressure applied along an axis
45° between the fiber axes), how much the fiber is insulated from the pressure (jacketing, etc.),
and whether the pressure is applied to a discrete point or distributed over a longer length of fiber.
GP-UM-PXA-1000-V21
Page 49 of 68
PXA-1000 User Guide
Saving Data
Data from any completed measurement can be saved to a file using the “Save” button
(F6). The user will be prompted to specify which curve is to be saved:
Select the desired curve and click “OK”.
The user will be prompted for a filename and location to which to save the data. The saved
data file will have file extension “.pxa” to avoid confusion with other types of data files, but it can
be opened using applications such as Notepad or Excel.
Saved curve data (with file extension “.pxa”) can be loaded for display, further analysis, or
comparison with other curves by using the “Load” button (F3). The software can be used for
display and analysis of saved data without the PXA-1000 connected to the computer.
GP-UM-PXA-1000-V21
Page 50 of 68
PXA-1000 User Guide
3.6 PER and Position Reference Calibrations (Optional Feature)
To ensure maximum accuracy and repeatability, General Photonics recommends periodic
factory calibration of the PXA-1000. However, an option is available for users to quickly reestablish the PER measurement and input connector position references.
Note: These user calibrations write to the system file, so they must be done carefully.
X Position Reference
X-talk and position references in the PXA-1000
In order to calibrate the x-talk in the PXA-1000, a standard coupling point is built into the
optical path just after the input connector. Its position and coupling ratio (x-talk) have been
factory calibrated and stored in the PXA-1000 system files. After every x-talk measurement, the
PolaX control software automatically locates the reference point position and adjusts the y-position
of the measured trace such that the x-talk at the reference point matches the calibrated value.
Then, the software automatically locates the input connector and sets its x-axis position to zero.
Therefore, the x-position of a coupling point indicates its distance from the input connector of the
PXA-1000.
Re-establishing the X=0 position reference
Under normal circumstances, the PXA-1000 places Marker 1 at the input connector
position and references the x-axis of the plot such that the input connector position corresponds to
X = 0.
If the x-reference becomes corrupted such that the PXA-1000 consistently misidentifies
the position of the input connector, the user can re-establish the x-reference using the “Set Mark1
as Zero Position” function in the “Calibration” block at the bottom of the Setup tab on the right
side of the screen. The procedure is as follows:
1.
Connect a good PM patchcord, for which the input connector peak can be easily identified, as
the DUT.
2.
Follow the standard procedure for a distributed polarization crosstalk measurement.
GP-UM-PXA-1000-V21
Page 51 of 68
PXA-1000 User Guide
3.
Identify the peak corresponding to the input connector and place Marker 1 on that peak.
4.
Click “Set Mark1 as Zero Position” at the bottom of the Setup tab.
5.
You will be prompted for a password.
6.
Enter the password and click “OK”. The Marker 1 position will be set as the system xreference.
GP-UM-PXA-1000-V21
Page 52 of 68
PXA-1000 User Guide
PER Reference
General Photonics can provide an optional 0-dB PER calibration artifact for users to verify
and, if necessary, re-establish the PXA-1000’s PER reference.
The procedures are as follows:
PER Verification
1.
Connect the 0-dB PER artifact as the DUT. Make sure connectors are clean, connections are
good, and fibers are stationary.
2.
Run the PER measurement. If the measured system PER is ≤0.2 dB, the PXA-1000 does not
require PER calibration.
Re-establish the PER reference
1.
Click “PER Calibration” at the bottom of the Setup tab.
2.
You will be prompted for a password.
3.
Enter the password and click “OK”. You will be prompted to connect the 0-dB PER artifact.
4.
Connect the 0-dB PER artifact as the DUT. Make sure connectors are clean, connections are
good, and fibers are stationary. Click “OK”.
5.
The PXA-1000 will perform a measurement and recalibrate its PER reference. When it is
finished, the progress bar and message box at the lower left of the screen will indicate that
the process is finished.
GP-UM-PXA-1000-V21
Page 53 of 68
PXA-1000 User Guide
3.7 Troubleshooting
The following table lists some common issues and probable causes.
Symptom
Program shows “PXA-1000 not
connected”
Plot shows inaccurate fiber length or
peaks not at expected positions.
Noise floor of plot is too high.
Program shows “Detector saturated”
message in the message box.
Program shows estimated PER of
FUT as “Inf”
Markers incorrectly placed after
measurement of a long coil or a
waveguide.
GP-UM-PXA-1000-V21
Probable Cause
Check the USB connection. Reconnect the cable and
wait for the “device detected” message before running
the control program. If necessary, restart the PXA-1000
and the control computer before reconnecting.
1. Check Δn of fiber. If the value used is not the
actual Δn of the fiber, the displayed fiber length
will be incorrect.
2. Check that the DUT connector keys are aligned
to the slow axis of the FUT/DUT.
If the DUT output power is too low, the noise floor of the
plot may increase. Generally, it should be 10 dB below
the lowest crosstalk peaks to be measured. Use the
“Power Monitor” function to check the DUT output
power, and, if necessary, adjust the attenuation.
DUT output power is too high. Use the “Power Monitor”
function to check the DUT output power, and increase
the attenuation as needed.
There are not enough detected peaks to calculate a PER.
Either change the PER calculation method to “Area” or
change the threshold level used for peak detection for
“Peak” method PER calculation (for example, from −50
to −60 dB).
Check that the delay range for the MDL is set correctly.
If the measurement curve does not include a peak at
the right corresponding to the output connector, the
MDL scan range may be too short to measure the entire
DUT, and the PXA-1000 software will not be able to
place Marker 2 correctly. Check the required scan length
for the DUT and adjust the MDL stop position
accordingly.
Page 54 of 68
PXA-1000 User Guide
Section 4.0
Specifications
Optical
Operating wavelength
1310 or 1550nm
Polarization X-talk measurement
sensitivity
Polarization X-talk measurement
noise floor1
Polarization X-talk resolution
<−75 dB (for DUT output power >5dBm )
−80dB typical
−95 dB
Polarization X-talk repeatability2
±0.5 dB
Polarization X-talk accuracy
3
0.25 dB
±0.5 dB
Measurement or sensing range
(Assuming PM fiber Δn of 5 x10-4)
Measurement speed4
1.3 km or 2.6 km standard
3.1 km optional
8 s/ 100 m (with fiber Δn = 5 x10-4)
Spatial resolution5
PER measurement range
6 cm (assuming no fiber dispersion,
birefringence Δn=5x10-4)
> 30 dB
Spatial accuracy6
±20 cm (with fiber Δn =5x10-4 )
Waveguide polarization dependent
attenuation
LiNbO3 waveguide spatial
resolution
SLD Power
Up to 75dB (for DUT power output >5dBm)
SLD bandwidth
> 30 nm
SLD PER
>20 dB
0.75 mm
>10 dBm
Electrical/Communication
Power Supply
100-240VAC, 50-60 Hz
Communication Interfaces
USB 2.0
Display
Laptop control computer (supplied)
Software
PolaX polarization X-talk measurement program:
Identifies polarization X-talk magnitude/location
Zoom-in function
X-talk related calculations
GP-UM-PXA-1000-V21
Page 55 of 68
PXA-1000 User Guide
Physical and Environmental
Dimensions
Fiber Type
2U ¾ 19” rack mount size
14” (L) × 14” (W) × 3.5” (H)
PM fiber
Fiber Input/Output Connectors
FC/PC narrow key standard
Weight
13.5 lb
Operation temperature
10 to 50°C
Storage Temperature
−20 to 60°C
Notes:
Specifications listed in table apply for standard 1550 or 1310nm operation at 23±5°C.
1. Defined as the system noise displayed on the polarization X-talk curve if the input power is
disconnected during a measurement for which the DUT output power is >5dBm.
2. Defined as the standard deviation of twenty successive measurements of the amplitude of
an X-talk peak between −15 and −40 dB.
3. At 23 ± 5°C. Guaranteed by design and calibration in manufacturing process.
4. Average speed for full-length scan.
5. Defined as the minimum resolvable distance between two polarization X-talk points of
equal amplitude, based on Sparrow Criterion (two peaks of equal height overlap at 3dB
point, resulting in a flat-top curve) and measured when the peaks are between −15 and
−40 dB.
6. Defined as the standard deviation of twenty successive measurements of the distance of
an X-talk peak of height between −15 and −40 dB from the X-talk peak induced by the
input connector (zero position).
GP-UM-PXA-1000-V21
Page 56 of 68
PXA-1000 User Guide
Appendices
Appendix 1.0 Comparison of PXA-1000 to Traditional White Light
Interferometer
It is well known that only the two eigenpolarization modes HE11s and HE11f can propagate in
polarization maintaining fiber. HE11s is polarized along the slow axis and HE11f along the fast axis
of the fiber. After traveling through a piece of PM fiber, an input wave packet is split into a series
of small wave packets separated in time because of the birefringence and mode coupling of the PM
fiber. This section describes a simple model to simulate the wave packets aligned to the slow and
fast axes of the fiber. Assume that the light input to the PM fiber has no fast axis component and
that there are three coupling points x1, x2 and x3 along the fiber (see Figure 28). It should be
emphasized that light is coupled not only from the slow axis to the fast axis, but also from the fast
axis to the slow axis at each coupling point; therefore, the resulting wave packet series will
include wave packets caused by multiple couplings.
Input wave
packet polarized
along slow axis X0
X1
Xout
X3
X2
PM fiber
S13 S23
S0
S12
slow axis
Optical path length
f1
fast axis
f2
f3
Optical path length
Figure 28 Wave packet sequence generated by coupling between slow and fast axis
Figure 28 shows the significant output wave packets at the end of the fiber (xout). S0 is the
principal wave packet, aligned to the slow axis. f1, f2, and f3 are wave packets aligned to the fast
axis, generated by first-order coupling. S12, S23, and S13 are wave packets aligned to the slow
axis, generated by second-order coupling.
GP-UM-PXA-1000-V21
Page 57 of 68
PXA-1000 User Guide
S0
f1
f2
f3
S13 S23
S12
Mixed wave packet
sequence in fixed arm
S0
f1
f2
f3
S13 S23
S12
Mixed wave packet sequence
in moving arm
moving direction
Figure 29 Wave packet sequence in traditional interferometer after passing a 45°-oriented analyzer
After passing through a 45° oriented analyzer, the wave packets aligned to the slow and
fast axes will be mixed together (see Figure 29). If this mixed light is input to an interferometer, a
series of interference peaks can be observed as the delay in one arm of the interferometer is
changed. Table 1 lists all possible interference peaks for the example shown in Figure 28. Only the
peaks generated by the interference between S0↔f1, S0↔f2, and S0↔f3 represent the coupling
points x1, x2 and x3. The other peaks listed in Table 1 are ghost peaks that can cause errors in the
identification of coupling points. They can also be superimposed on the real peaks, reducing the xtalk measurement accuracy.
In order to minimize the number and magnitude of ghost peaks, the PXA-1000 uses a
patented technique to prevent the zero order, second order and most higher order interference
signals from being generated as the delay line scans (see Table 1). Consequently, the PXA-1000 is
able to achieve higher position measurement accuracy, higher dynamic range and higher
sensitivity than traditional white-light interferometers.
Table 1 Interference peaks for PM fiber model shown in Figure 28
Zero-order interference
Interference from first
order coupling
Interference from
second order coupling
Interference from higher
order coupling
Traditional White-light Interferometer
PXA-1000
S0↔S0, S12↔S12, S23↔S23, S13↔S13,
f1↔f1, f2↔f2, f3↔f3
S0↔f1, S0↔f2, S0↔f3
None
S0↔S12, S0↔ S23, S0↔S13
f1↔f2, f1↔f3, f2↔f3
S12↔f3, S12↔ S23, S12↔S13,
S12↔f2, S12↔f1
f3↔ S23, f3↔ S13
S23↔S13 S23↔f2, S23↔f1
S13↔f2, S13↔f1
None
GP-UM-PXA-1000-V21
S0↔f1, S0↔f2, S0↔f3
most higher order
couplings are eliminated
Page 58 of 68
PXA-1000 User Guide
Appendix 2.0 Spatial Resolution of PXA-1000
According to interference theory, the full width at half maximum (FWHM) of an
interference single envelope can be calculated by
l FWHM
λ20
2 ln 2 λ20
=
≈ 0.44
π Δλ
Δλ
(9)
where λ0 and Δλ are the center wavelength and spectral width, respectively, of the light source
used for measurement. Thus, the spatial resolution of a PM fiber measurement can be obtained by
Lresolution =
lFWHM
Δn
(9)
where Δn is the birefringence of the PM fiber.
For example, when λ0=1310nm, Δλ=30nm, and Δn=5x10-4, then the spatial resolution of the PXA1000 will be about 5cm.
GP-UM-PXA-1000-V21
Page 59 of 68
PXA-1000 User Guide
GP-UM-PXA-1000-V21
Page 60 of 68
PXA-1000 User Guide
Appendix 3.0 Polarization Crosstalk in PM Fiber
Classification of Polarization Crosstalk by Cause
Polarization crosstalk in a PM fiber arises from three principal causes. 1) Fiber axis
misalignment at fiber connection interfaces, such as connectors or fusion splices, typically causes
extremely localized, large-amplitude crosstalk. The amplitude depends on the misalignment angle.
Examples are shown on the left side of Figure 30. Figure 30a shows crosstalk sources along a
fiber, and Figure 30b shows the resulting crosstalk measurement plot. 2) PM fiber imperfections,
such as local birefringence variations, internal shape variations, or internal stress, cause
polarization coupling that is generally small in amplitude and occurs gradually over a certain
length of the PM fiber (see center section of Figure 30). 3) External mechanical stresses on
sections of the fiber, such as fiber bending, fiber crossing, fiber squeezing, or pressure on the
fiber, can cause complicated composite crosstalk effects that can include polarization couplings
that occur at sharp points in space, as well as some that occur gradually along a length of fiber,
with varied amplitudes that depend on the stress orientations with respect to the slow axis and on
the stress intensities, as shown in the right section of Figure 30.
Classification of Polarization Crosstalk by Measurement Results
In general, the PXA-1000 distributed polarization crosstalk analyzer can accurately
measure the strength of polarization crosstalk occurring at different locations along a fiber with a
spatial resolution of a few centimeters. Although the causes of the crosstalk cannot always be
identified from measurement results, educated guesses can be made based on the shape and
strength of the measured crosstalk at each location. It is also feasible to classify the crosstalk
based on the shapes of the measured curves, as discussed below.
GP-UM-PXA-1000-V21
Page 61 of 68
PXA-1000 User Guide
Point Splice or
stress connector
Multiple-point
stresses
Line
stress
Internal fiber
variations
a
Discrete coupling
Continuous
coupling
Quasi-continuous
coupling
X-talk
b
Z
Figure 30 Illustration of different types of polarization crosstalk. a) Different sources of
polarization crosstalk. b) The resulting crosstalk peak profiles. Left: discrete polarization x-talk
peaks induced by a point stress or a splice. Each such peak is a Gaussian curve with a shape
determined by the coherence function of the light source. The spatial resolution is also determined
by the width of the coherence function. Center: continuous polarization x-talk induced by a line
stress and by internal fiber imperfections, respectively. Right: quasi-continuous x-talk induced by
multiple densely packed stress points spaced on the order of or less than the resolution of the
instrument.
X-talk caused by discrete polarization coupling points
This category includes polarization coupling induced by a sharp stress, a splice point or
multiple stress/splice points separated by distances much larger than the resolution of the
measurement instrument, as shown on the left side of Figure 30a. These types of discrete coupling
result in sharp, distinct peaks in the x-talk measurement trace, with the width of the peak
determined by the spatial resolution of the instrument, as shown in Figure 30b. For this type of
coupling, the peak x-talk value for each coupling point conveys useful information. The x-talk
values listed in the table at the right of Figure 31b result from such discrete coupling points.
GP-UM-PXA-1000-V21
Page 62 of 68
PXA-1000 User Guide
a)
A
B
Section A
Discrete couplings
b)
39 40
42
41 Threshold = −60 dB 43
44
Quasi-continuous couplings
Figure 31 a) X-talk measurement of a 340 meter long fiber coil. Two significant sections (labeled A
and B) are marked. b) Closeup view of section A, showing both discrete and quasi-continuous
coupling peaks. The table on the right of the screen lists the magnitudes of discrete x-talk peaks
larger than −60 dB. Note that the shape and width of discrete x-talk peaks are determined by the
coherence function of the light source.
X-talk caused by continuous polarization coupling
This category includes polarization coupling that accumulates gradually over a section of
fiber, induced by a line stress or by fiber internal imperfections, where the length of the affected
section of fiber is comparable to or larger than the resolution of the measurement instrument, as
shown in the center section of Figure 30a. The crosstalk measurement result of such continuous
coupling is a broad dome with a width and shape determined mainly by the length of the section
GP-UM-PXA-1000-V21
Page 63 of 68
PXA-1000 User Guide
of fiber under stress, as shown in Figure 30b. In general, crosstalk caused by a section of
imperfect fiber is very small in amplitude - on the order of −60 dB or lower. Because of this curve
structure (low amplitude, wide peak), the peak x-talk value for crosstalk resulting from continuous
polarization coupling is not meaningful. However, the cumulative coupling occurring in a section of
fiber can be obtained by defining the starting and ending positions of the continuous-coupling
section of fiber using the PXA-1000 software’s cursors, as shown in Figure 32. In this example, the
cumulative value is −61.32 dB.
Section B
a)
Continuous
Discrete
b)
Z1
Z2
Figure 32 Closeup view of section B of the x-talk measurement plot shown in Figure 31a, in which
two continuous or quasi-continuous x-talk peaks are identified. b) Cumulative x-talk value of a
continuous/quasi-continuous coupling is obtained by setting the locations of cursors Z1 and Z2 and
calculating the integrated PER of the corresponding fiber section. The resulting value, −61.32 dB, is
shown at the bottom right of the screen.
GP-UM-PXA-1000-V21
Page 64 of 68
PXA-1000 User Guide
X-talk caused by quasi-continuous polarization coupling
This category includes polarization coupling induced by multiple stress points spaced on
the order of or less than the resolution of the measurement instrument, as shown in Figure 30a.
This type of polarization coupling appears in polarization crosstalk measurements as a broad
composite peak with height variations, with a width and shape determined by the number of
stress points, their relative positions, and their relative strengths, as shown in Figure 30b. Quasicontinuous coupling cannot reliably be distinguished from continuous coupling. As in the case of
continuous coupling, it is not meaningful to give a peak x-talk value for quasi-continuous coupling.
However, the cumulative coupling occurring in a section of fiber can be obtained by defining the
starting and ending positions of the continuous-coupling section of fiber using the PXA-1000
software’s cursors, as shown in Figure 33b.
GP-UM-PXA-1000-V21
Page 65 of 68
PXA-1000 User Guide
a)
b)
Quasi-continuous
Figure 33 a) X-talk measurement of a low quality PM fiber coil of length 309 meters. b) Closeup
view of the boxed section in a), showing more detailed structure of quasi-continuous couplings.
The cumulative x-talk of the region between the two cursors is −26.46 dB.
Capabilities and limitations of the PXA-1000
The PXA-1000 can take crosstalk measurements at spatial intervals of about 4-6 mm,
much finer than the specified x-talk resolution of the instrument (on the order of 5 cm). The exact
spacing between two adjacent data points is dependent on the birefringence of the fiber; it is
defined as the ratio of the delay resolution of the variable delay line used in the PXA-1000 to the
GP-UM-PXA-1000-V21
Page 66 of 68
PXA-1000 User Guide
fiber birefringence. However, depending on the type of the polarization coupling, a given x-talk
reading may not represent the true x-talk value at that point in space, as will be discussed below.
For X-talk induced by discrete polarization coupling points, the PXA-1000 is able to display
the corresponding discrete x-talk peaks, to provide an accurate x-talk value for each x-talk peak,
and to list in a table all peak values above a defined threshold, as shown in Figure 31b. Note that
each peak has a Gaussian shape corresponding to the coherence function of the light source used;
however, only the peak value is meaningful and represents the x-talk value at the point in space
at which the x-talk occurs. The other points on the Gaussian curve are due to the light source’s
coherence function and do not represent meaningful x-talk values for the corresponding points, as
shown in Figure 34. Note that the instrument’s x-talk accuracy specification is based on
measurement of the peak values of such discrete x-talk points.
For discrete coupling, only
the peak value is meaningful
For continuous or quasi-continuous
coupling, only the integrated x-talk
value over a distance is meaningful
X-talk
meaningless
z
Z
Z
Figure 34 Left: Illustration of a crosstalk peak caused by discrete coupling, with peak and off-peak
values marked. The off-peak values are artifacts caused by the coherence function of the light
source; they have no relation to real x-talk points on the fiber. Right: Illustration of a x-talk dome
induced by continuous or quasi-continuous polarization coupling. A point on the dome does not
correspond to a x-talk point on the fiber. In this case, only the integrated cross coupling between
the points Z1 and Z2 is meaningful. Points Z1 and Z2 can be defined in the software interface.
For X-talk induced by the continuous or quasi-continuous coupling shown in Figure 30, the
x-talk value of any single point on the broad x-talk composite peak is not meaningful. The PXA1000 is unable to give an accurate x-talk value for such a point, although the x-talk data file
includes data points every 4-6 mm. For a x-talk composite peak caused by continuous or closely
packed quasi-continuous coupling points, only the cumulative cross-talk value is meaningful, as
shown in Figure 32b and Figure 33b. The PXA-1000’s data display and analysis software, PolaX,
has a function that calculates the cumulative x-talk from point Z1 to point Z2, where Z1 and Z2 are
defined by the locations of the cursors, as shown in Figure 32b and Figure 33b. In general, the
distance between Z1 and Z2 should be much larger than the spatial resolution of the instrument in
order to obtain an accurate result. In addition, the two points should also be chosen at valleys on
the x-talk curve and the “Area” calculation method should be selected, as shown at the bottom
right of Figure 32b and Figure 33b.
GP-UM-PXA-1000-V21
Page 67 of 68
PXA-1000 User Guide
Note that the primary purpose of the PXA-1000 is to obtain accurate x-talk measurements
of discrete x-talk peaks; the accuracy of a cumulative x-talk calculation is not guaranteed.
However, the instrument records measurement data every 4-6 mm, and this data is available to
users to optimize the calculation for specific cases where higher accuracy is required. Data is
available to the user in two forms: the raw interferometer signal data as a function of the relative
delay between the two arms of the interferometer (“Interferometer Only” data) and the x-talk
data as displayed on the screen. The x-talk data is the raw interferometer data with the horizontal
and vertical axes shifted according to internal position and x-talk references.
GP-UM-PXA-1000-V21
Page 68 of 68
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement