Agilent
Physical Layer Test System
(PLTS Version 3.0)
Technical Overview
Vector Network Analyzer-based and
Time Domain Reflectometer-based Systems
Includes:
“What’s New in PLTS Software Version 3.0”
Why is Physical Layer Testing
Required?
The next generation computer and communication systems now
being developed will handle data rates of multiple gigabits/second. Many systems will incorporate processors and SERDES chip
sets that exceed GigaHertz clock frequencies. New and troubling
input/output issues are emerging as switches, routers, server
blades, and storage area networking equipment moving toward
10 Gbps data rates. Digital design engineers choosing chip-to-chip
and backplane technologies for these systems are finding signal
integrity challenges that have not been encountered before.
In order to maintain signal integrity throughout the complete
channel, engineers are moving away from single-ended circuits
and now use differential circuits. The differential circuit provides
good Common Mode Rejection Ratio (CMRR) and helps shield
adjacent PCB traces from crosstalk. Properly designed differential
transmission lines will minimize the undesirable effect of mode
conversion and enhance the maximum data rate throughput
possible. Unfortunately, differential signaling technology is not
always an intuitive science.
Traditional parallel bus topologies are running out of bandwidth.
As parallel busses become wider, the complexity and cost to route
on PC boards increase dramatically. The growing skew between
data and clock lines has become increasingly difficult to resolve
within parallel busses. The solution is fast serial channels. The
newer serial bus structure is quickly replacing the parallel bus
structure for high-speed digital systems. Engineers have been
turning to a multitude of gigabit serial interconnect protocols with
embedded clocking to achieve the goal of simple routing and more
bandwidth per pin. However, these serial interconnects bring their
own set of problems.
Differential transmission lines coupled with the microwave
effects of high-speed data have created the need for new design
and validation tools for the digital design engineer. Understanding
the fundamental properties of signal propagation through measurement and post-measurement analysis is mandatory for today’s
leading edge telecommunication and computer systems. The
traditional Time Domain Reflectometer (TDR) is still a very useful
tool, but many times the Vector Network Analyzer (VNA) is needed
for the complete characterization of physical layer components.
There is a strong need for a test and measurement system that
will allow simple characterization of complex microwave behavior
seen in high speed digital interconnects. In fact, many digital
standards groups have now recognized the importance of specifying
frequency domain physical layer measurements as a compliance
requirement. Both Serial ATA and PCI Express have adopted the
SDD21 parameter (input differential insertion loss) as a required
measurement to ensure channel compliance. This parameter is an
indication of the frequency response that the differential signal
sees as it propagates through the high-speed serial channel.
An example of a proposed SDD21 compliance mask is shown in
Figure 1 for the Channel Electrical Interface (CEI) working group
for the Optical Internetworking Forum (OIF).
In order to maintain the same total bandwidth as the older parallel
bus, the new serial bus needs to increase its data rate. As the
data rate increases through serial interconnects, the rise time of
the data transition from a zero logic level to a one logic level
becomes shorter. This shorter rise time creates larger reflections
at impedance discontinuities and degrade the eye diagram at the
end of the channel. As a result, physical layer components such as
printed circuit board traces, connectors, cables, and IC packages
can no longer be ignored. In fact, in many cases, the silicon is so
fast that the physical layer device has become the bottleneck.
SDD21 (dB)
0
ISI Loss > 4 dB
-11.4
Sample Compliance Interconnect
-15
0 312.5 MHz
1.5625 GHz
3.125 GHz
Figure 1. Today's digital standards are now using frequency domain
measurements for compliance testing, such as this input differential
insertion loss (SDD21) mask for XAUI.
2
A Single Test System Can Provide
the Total View
As the combination of both time-domain and frequency domain
analysis becomes more important, the need for multiple test
systems becomes difficult to manage. A single test system that
can fully characterize differential high-speed digital devices, while
leaving domain and format of the analysis up to the designer, is a
very powerful tool. Agilent’s Physical Layer Test System (PLTS) is
designed specifically for this purpose.
PLTS has been designed specifically for signal integrity analysis.
PLTS software guides the user through hardware setup and
calibration, and controls the data acquisition. It automatically
applies patented transformation algorithms to present the data
in both frequency and time domains, in both forward and reverse
transmission and reflection terms, and in all possible modes
of operation (single-ended, differential, and mode-conversion).
A powerful virtual bit pattern generator feature allows a userdefined binary sequence to be applied to the measured data to
convolve eye pattern diagrams. Next, highly accurate RLCG 1
models can be extracted and used to enhance the accuracy of
your models and simulations.
PLTS Provides Design Confidence
Through Complete Characterization
Physical-layer structures have increasingly become the bottleneck
in high-speed digital system performance. At low data rates,
these interconnects are electrically short. The driver and receiver
are typically the biggest contributors to signal integrity. But as
clock speeds, bus speeds, and link speeds all push past the gigabit-per-second mark, physical layer characterization becomes
more critical.
Another challenge for today's digital designers is the trend to
differential topologies. Fully understanding device performance
requires analysis in all possible modes of operation.
Single ended
Port 1
For translating device performance into standards compliance,
eye diagrams add an important statistical analysis. And for leveraging this complete characterization into improved simulations,
measurement-based s-parameter or RLCG model extraction
completes the picture.
Mode
Differential
Diff-to-comm
Comm-to-diff
Common
Time domain
TDR
TDT
TDD11 TDD21
TDD22 TDD12
TCD11 TCD21
TCD22 TCD12
TDC11 TDC21
TDC22 TDC12
TCC11
TCC21
TCC22
TCC12
Frequency domain
Reflection Transmission
SDD11
SDD21
SDD22
SDD12
SCD11
SCD21
SCD22
SCD12
SDC11
SDC21
SDC22
SDC12
SCC11
SCC21
SCC22
SCC12
Single-ended
T11
T22
T33
T44
T21 T31 T41
T12 T32 T42
T13 T23 T43
T14 T24 T34
S11
S22
S33
S44
S21 S31 S41
S12 S32 S42
S13 S23 S43
S14 S24 S34
Figure 3. Complete characterization includes forward and reverse
transmission and reflection, in all possible modes of operation, in both
frequency and time domains.
+
Differential
Port 1
Differential
Port 2
–
Single ended
Port 3
Frequency-domain analysis, again in all possible modes of operation, is also necessary for fully characterizing these physical-layer
structures. The s-parameter model describes the analog behavior
exhibited by these digital structures. This behavior includes
reflections from discontinuities, frequency dependent losses,
crosstalk, and EMI performance.
Single ended
Port 2
Device under test
+
Time-domain analysis is typically used for characterization of
these physical-layer structures, but often, the designer concentrates only on the intended modes of operation. For a complete
time-domain view, step and impulse responses in reflection and
transmission (TDR and TDT) must be seen. The analysis must
include the unintended modes of operation as well.
–
Single ended
Port 4
Figure 2. A differential structure operates in many modes.
Single-ended analysis can reveal sources of asymmetry on this
differential transmission line.
1. An RLCG equivalent circuit model, also known as Telegrapher’s
Parameters, describes the electrical behavior of a passive transmission
line. The model is a distributed network consisting of series resistance
and inductance (R and L) and parallel capacitance and conductance
(C and G).
3
PLTS Enables Mode-Conversion
Analysis for Early Insight into
EMI Problems
Mode Conversion
The benefits of differential signaling include lower voltage swings,
immunity from power supply noise, a reduced dependency on RF
ground, and improved EMI performance (reduced generation and
susceptibility). The extent to which a device can take advantage
of these benefits is directly related to device symmetry.
A practical application of how mode conversion helps identify
problems in physical layer devices is shown in Figure 5 . This shows a
XAUI backplane with two daughter cards that typically transmit data
at 3.125 Gbps. The design objective for this high-speed differential
channel is to minimize the crosstalk between adjacent differential
PCB traces throughout the length of the channel. The channel
consists of the linear passive combination of the backplane and
two daughter cards. Any mode conversion from differential mode
to common mode will generate EMI and create crosstalk that will
be incident upon other channels and will degrade performance.
However, locating the exact structure within the channel that
creates the most mode conversion is not simple.
Symmetric devices only respond to, and only generate, differential
signals. These ideal devices do not respond to or generate common-mode signals, and they reject radiated external signals (i.e.,
power supply noise, harmonics of digital clocks or data,
and EMI from other RF circuitry).
Asymmetric devices however, do not exhibit these benefits. When
stimulated differentially, an asymmetric device will produce a common-mode response in addition to the intended differential
response, and cause EMI radiation. Conversely, with a commonmode stimulus, an asymmetric device will produce an unintended
differential response. This mode conversion is a source of EMI
susceptibility.
Mode-conversion analysis is an important tool for understanding
and improving device symmetry, and provides the designer with
early insight to identify and resolve EMI problems at the design
stage.
Passive differential structure
Differential
stimulus
Differential
response
Common-mode response
(unintended mode conversion)
Figure 4. Asymmetric devices cause mode-conversions, which are
indicators of EMI generation and susceptibility.
4
Figure 5. By aligning the impedance profile with the mode conversion
profile, PLTS allows the pinpointing of crosstalk-generating structures
within physical layer devices.
Looking at Figure 5, the differential to common mode conversion
time domain reflection parameter (TCD11) is time aligned with the
differential impedance profile of the channel (TDD11) below it. A
marker is placed on the largest magnitude peak of TCD11. This is
where the physical structure within the channel is creating the
most mode conversion and thus the source of the most crosstalk.
We can align the TDD11 to the TCD11 in time and therefore colocate the problematic structure on TDD11. To relate this structure to the channel, we use the differential impedance profile as a
reference. From previous analysis, we know that the two capacitive discontinuities on TDD11 are the daughter card via field and
motherboard via field, respectively. Since the marker falls upon
the second discontinuity on TDD11, it is deduced that the motherboard via field is the biggest culprit to causing crosstalk in adjacent channels. The motherboard via field was subsequently rerouted and the crosstalk generation was reduced considerably.
This shows how identifying the mode conversion in a channel can
be intuitive with proper analysis.
Dynamic range
Four-Port TRL (Thru-Reflect-Line)
High-dynamic range is important for a number of reasons.
Certainly, measurements of very low levels of crosstalk are
possible, but this is only one parameter where dynamic range is
important.
TRL calibration is primarily (but not exclusively) used in non-coaxial
environments, such as in-fixture or microprobe measurements. It
determines the same error model as the SOLT calibration, although
it uses different calibration standards - a thru, a reflection standard
(a short), and one or more offset length transmission lines.
More important is the ability to overcome masking effects of
multiple discontinuities, which in systems with lower dynamic
range would attenuate the stimulus such that deep structures
would become invisible.
And most importantly for differential devices, high-dynamic range
allows for identification of very low levels of mode-conversion,
which are the direct result of device asymmetry. This allows early
resolution of potential EMI issues.
Accuracy Through Error-Correction
VNA-based systems
Four-Port SOLT (Short-Open-Load-Thru)
The most common technique for both coaxial and non-coaxial
environments, SOLT calibration, is a vector error correction
process that characterizes systematic error by measuring known
calibration standards. This data is used to calculate a 72-term
error model, which is then used to remove the effects of
systematic errors from subsequent measurements.
Figure 7. The vector-correction process characterizes systematic error
to provide superior accuracy in network analyzer-based systems.
Utilizing three impedance standards (the short, open, and load)
and a thru standard, the error model compensates for directivity,
source match, load match, reflection and transmission tracking
(frequency response), and crosstalk in both forward and reverse
directions.
Four-Port LRM (Line-Reflect-Match)
TRL calibration typically provides more accuracy than SOLT and is
only available for PNA-based systems (it is not supported for
872x-based systems).
LRM is a variation of TRL, and determines the same error model
as the SOLT and TRL calibrations, although it uses different (and
fewer) calibration standards – a transmission line, a reflection
standard (a short), and a load.
In broadband non-coaxial environments, specifically microprobing,
LRM can offer an accuracy improvement over TRL due to bandwidth limitations of the TRL line standards.
LRM calibration is only available for PNA network analyzer-based
systems (it is not supported for Agilent 872x-network analyzer
based systems).
Four-Port Electronic Calibration (ECal)
For electronic calibration to 9 GHz, the Agilent N4430B ECal
Module (30 kHz to 9 GHz) is supported by all of the PLTS VNAbased systems. With one set of connections, this solid-state
tuner simulates all of the impedance states required for full fourport error correction with accuracy that is generally better than
SOLT, but somewhat less than TRL.
Figure 6. The PNA network analyzer-based calibration interface simplifies
a complex process, allowing full error-correction in minutes, not hours.
5
TDR-based systems
Module calibration
Module calibration, also called vertical calibration, calibrates the
gains, offsets, and timing for each channel.
At the start of the TDR calibration process, PLTS will report that a
module calibration is valid, recommended, or required.
This calibration is available for all supported TDR-based systems.
De-skew
Asymmetry in a differential system – between the two step
generators, between the two receivers, etc… – can potentially
lead to imbalances or errors in resulting measurements.
When differential measurements are selected in the calibration
process, PLTS automatically performs this operation to remove
length variance from cables or test fixtures.
Figure 8. Like the PNA network analyzer-based calibration interface,
the TDR-based calibration interface guides the user through the
calibration(s) reducing operator error and saving time.
This calibration is available for all supported TDR-based systems.
Reference plane calibration (RPC)
RPC adjusts the calibration reference plane to the end of the
test cables and automatically performs the de-skew operation
(see above).
This calibration is available for all supported TDR-based systems.
Normalization
Normalization is an error correction process that characterizes
systematic error by measuring known calibration standards (a
short, a load, and a thru) to calculate an error model, which is
then used to remove the effects of systematic errors from
subsequent measurements.
The error model compensates for directivity, source match, and
reflection tracking (frequency response). This is the most
accurate calibration type for TDR.
Note: Normalization is available for the Agilent 86100 family of
oscilloscopes only.
6
Remove Unwanted Effects from
the Measurement
Error correction
Over the years, many different approaches have been developed
for removing the effects of the test fixture from the measurement
(shown in Figure 9). The level of difficulty for each error correction
technique is linearly related to the accuracy of each method. Time
domain gating is perhaps the simplest and most straightforward
method, but it is also the least accurate. Likewise, de-embeding is
the most complicated method, but it is the most accurate. It is
important to have a test system that will allow flexibility of choosing the method of error correction desired for each application.
Error correction techniques fall into two fundamental categories:
direct measurement (pre-measurement processing) and deembedding (post-measurement processing). Direct measurement
requires specialized calibration standards that are connected to
the end of a coaxial test cable and measured. The accuracy of the
device measurement relies on the quality of these physical standards. De-embedding uses a model of the test fixture and mathematically removes the fixture characteristics from the overall
measurement. This fixture de-embedding procedure can produce
very accurate results.
Easier
Port Extension
Ease of Use
º
Time Domain Gating
TDR Reference Plane Calibration
TDR Normalization
SOLT
TRL
LRM
De-embedding
More Accurate
Accuracy
Figure 9. PLTS has advanced error correction techniques to allow flexibility for many applications.
Port Extension (also known as Phase Rotation) mathematically
extends the calibration reference plane to the DUT.
This technique is easy to use, but assumes the fixture – the
unwanted structure – looks like a perfect transmission line:
a flat magnitude response, a linear phase response, and constant
impedance. If the fixture is very well designed, this technique
can provide good results.
Because gating essentially considers the magnitude of
the unwanted discontinuity, and Port Extensions consider
phase (electrical length), using the two tools together
may provide optimum results.
Figure 10. In this rather extreme example of time-domain gating, the top
plots show the measured differential step impedance and return loss.
The lower left plot shows a gate added to remove the large discontinuity
in the center of the trace. On the lower right, the measured and the
recalculated return losses are displayed. In this case, the gate improved
the return loss by more than 10 dB within the frequency band of interest.
De-embedding uses an accurate linear model of the fixture, or
measured s-parameter data of the fixture. This fixture data can
then be removed mathematically from the DUT measurement
data in post-processing.
Time-domain gating is similar to port extension, in that it is also
very easy and fast. The user simply defines two points in time or
distance, and the software mathematically replaces the actual
measured data in that section with data representing an "ideal"
transmission line. The return loss is then recalculated to show
the effects of the change in the frequency domain.
One practical application of time-domain gating is as a confidence
check before replacing a suspect connector.
Figure 10 illustrates how this technique might be used.
Figure 11. The effects of test fixtures can be removed from the device in
post-processing through de-embedding.
7
Calibration at the DUT reference plane has the advantage that
the precise characteristics of the fixture don't need to be known
beforehand, as they are measured and corrected for during the
calibration process.
An example of this technique is microprobing using a calibration
substrate, where the calibration reference plane is established at
the probe tips, rather than at the end of the coaxial test cables.
Advanced calibration techniques (TRL/LRM) – originally developed
for wafer probing applications – provide additional options.
Figure 12. A microprobing application, where the calibration is performed
using an impedance standard substrate, establishes the calibration
reference plane at the probe tips.
Microprobing can offer the user the ability to forego the test
fixture and launch the stimulus directly at the device input.
The response can be measured directly at the device output.
Additionally, when calibration substrates are available,
calibration can be performed directly at the probe tips. This
achieves co-location of the calibration reference plane with
the device measurement reference plane.
PLTS has the flexibility to accommodate many microprobe
configurations. By adding the calibration substrate coefficients
as a calibration kit, the process becomes as straightforward
as a coaxial calibration.
Figure 13. Cascade Microtech’s Summit probe station.
PLTS Support for Microprobing
Applications
Agilent works closely with leading microprobe and probe station
suppliers to provide the best complete system solutions possible.
One of the most significant measurement challenges is connectivity. Test equipment provides a controlled coaxial environment, but
what if the DUT – the backplane, the interface connector,
the IC package – is non-coaxial?
Test fixtures can provide the required connectivity, but at a
cost. The quality of the test fixture – its connectors, impedance
discontinuities, parasitics, and dielectric losses – all contribute
to less than ideal performance of the fixture. Subsequently,
the accuracy of the device measurement is degraded.
Several techniques are available to remove these fixture effects
(see Remove Unwanted Effects from the Measurement on page
6), but the accuracy of these techniques is greatly impacted by
the quality of the fixture itself, or the availability of an accurate
s-parameter model of the fixture (used for de-embedding).
8
Figure 14. GigaTest Lab’s GTL-4060 probe station.
Simplify the Measurement Process
Device characterization with PLTS software is straightforward.
The user interface has been designed to make setup, calibration,
and measurement intuitive and error-free. A wizard guides the
user through all of the required steps. The last prompt is to connect
the device-under-test and initiate the measurement. Setup and
calibration differs slightly between TDR-based and VNA-based
systems. However, in both cases the PLTS software provides an
intuitive wizard to assist in the step-by-step process.
Step 1
System setup
TDR
setup
process
Step 1.
Select calibration
and measurement
parameters
Step 3
Device
measurement
TDR
calibration
Setup and
calibration
complete
VNA
setup
process
Easy Setup
TDR
setup
Step 2
Calibration
VNA
setup
VNA
calibration
Figure 15. PLTS has a three-step system set up to make measurements
intuitive and error-free.
In the first step, PLTS software automatically polls the GPIB, and
prompts the user to accept or modify the default parameters
based on hardware capabilities.
TDR Setup
Figure 16. PLTS software completely controls the hardware setup
via GPIB.
Figure 17. The user may calibrate only for the specific parameters of
interest. Estimated calibration times are calculated based on selections.
VNA Setup
Parameter selection not required in the VNA setup
Figure 18. Default parameters are calculated from system capabilities.
User modifications are interactive.
9
Step 2. Calibration
Step 3. Device measurement
After the system setup, the calibration method is selected. Again,
the wizard simplifies the process and provides the greatest flexibility to the user.
For both TDR and VNA systems, when the calibration is completed,
PLTS prompts the user to connect the device-under-test, and
select an initial analysis format.
The calibration interface shows the required calibration standards
as icons, which are initially represented in red. As the user connects the standards, and mouse-clicks the corresponding icon,
the system makes the measurement, and the icon color changes
to green (indicating completion). When all of the icons are green,
the calibration is complete.
Then, the system makes all of the measurements required for the
complete characterization.
With one setup, calibration and measurement, up to sixty-four
time-domain and frequency-domain device parameters are available.
TDR calibration
Setup and calibration complete
Figure 19. The TDR system calibration wizard provides status and controls
module and reference plane calibrations, de-skew, and normalization.
VNA calibration
Figure 21. All domains and formats are available immediately after the
measurement is completed.
Figure 20. The VNA system calibration wizard simplifies four-port SOLT,
TRL, and LRM error correction.
10
Data Analysis with PLTS
File and view management
with the data browser
Format, scaling,
and marker
control with
easy access
toolbars
Plot and trace
management with
context-sensitive
parameter buttons
Time domain analysis:
• 32 parameters
• 7 formats
• time or distance
Eye diagram analysis:
• 16 parameters
• 8 formats
Frequency domain analysis:
• 32 parameters
• 8 formats
RLCG model extraction
• 32 parameters
11
Data Analysis
All supported analysis types and formats1 are available immediately after the measurement is completed, and at any time there
after. PLTS flexibility allows the user to begin where they are most
familiar.
Frequency-domain analysis
Time-domain analysis
Initially, sixteen parameters are displayed in thumbnail view as
shown below. These thumbnails represent four modes of device
operation: differential, common-mode, and the two mode-conversion types (common-mode stimulus with differential response and
differential stimulus with common-mode response). A double
mouse-click on any of these thumbnails will expand the selected
parameter to full screen for closer analysis.
The mixed-mode time domain is a common starting point. Initially,
sixteen parameters are displayed in thumbnail view as shown
below. These thumbnails represent four modes of device operation: differential, common-mode, and the two mode-conversion
types (common-mode stimulus with differential response and
differential stimulus with common-mode response). A double
mouse-click on any of these thumbnails will expand the selected
parameter to full screen for closer analysis.
The mixed-mode frequency domain is another common starting point.
Not shown here are the additional sixteen single-ended
frequency-domain parameters.
Not shown here are the additional sixteen single-ended
time-domain parameters.
Figure 23. The mixed-mode frequency-domain matrix.
Figure 22. The mixed-mode time-domain matrix.
1. See Parameters and Formats on page 15 for more detail.
12
Measurement-based eye-pattern diagrams
RLCG model extraction
Using the built-in digital pattern generator, the user is able to
define virtual bit pattern (as wide as 232-1 bits). PLTS then convolves the selected bit pattern with the device impulse response
to create an extremely accurate measurement-based eye pattern
diagram.
RLCG (resistance, inductance, capacitance, and conductance)
models describe the electrical behavior of passive transmission
lines in an equivalent circuit model.
This eliminates the need for a hardware pulse/pattern generator,
and its flexibility allows for a great deal of "What if…" analysis.
From the measured s-parameters of a device, PLTS calculates the
R, L, C, G, complex propagation constant, and complex characteristic impedance.
This provides a highly accurate, measurement-based coupled
transmission line model for export into modeling and simulation
software such as Agilent ADS, Synopsis HSPICE, and others.
Figure 24. The digital pattern generator.
After the eye pattern is generated, marker functions can be used
to make typical measurements like jitter, eye opening, rise and fall
times, and more.
Figure 26. RLCG model extraction (W-Element shown).
Figure 25. Eye pattern diagram.
13
What’s New in PLTS Version 3.0?
New hardware support
• Agilent N5230A Option 240/245 (4-Port PNA-L microwave
network analyzer)
• Agilent E8361A (67 GHz 2-port PNA microwave network analyzer)
Note: N4421B 50 GHz test set is compatible with E8361A
up to 50 GHz. N4421BH67 is available for 67 GHz 4-port
system.
• Agilent electronic calibration modules N4430A, N4430B,
N4431B, N4691B, N4692A, N4693A, N4694A and 85093C
tools such as Cadence Allegro PCB SI
630.
Figure 27. PLTS version 3.0 enables support of the new Agilent N5230A
Option 245 for a one-box system and eliminates the test set for 20 GHz
applications.
New software support
• Improved file export capability
– Output 1-port, 2-port and 3-port touchstone files
(*.1sp, *.s2p and *.s3p)
– Modify port configuration
– Modify frequency range
• File export to Agilent Advanced Design System and Cadence
Allegro PCB SI 630 Simulation Suite
• TDR peeling algorithm for corrected impedance profile
Figure 29. PLTS version 3.0 enables data export to powerful simulation
14
Figure 28. PLTS version 3.0 enables
support of electronic calibration modules
for faster and simpler calibrations.
Parameters and Formats
Time domain: Sixteen single-ended and sixteen mixed-mode time
domain parameters are available. These parameters correspond to
input and output TDR, and forward and reverse TDT in singleended, differential, common-mode, and mixed-modes.
Frequency domain: Sixteen single-ended and sixteen mixed-mode
frequency domain parameters are available. These parameters
correspond to forward and reverse transmission and reflection in
single-ended, differential, common-mode, and mixed-modes.
Vertical amplitude formats for impulse or step function are impedance, volts, real part of complex parameter, and log magnitude.
The horizontal scale may be displayed in time (nS) or distance (cm).
Formats are log or linear magnitude, phase, group delay, Smith,
Polar, real part of complex parameter, imaginary part of complex
parameter.
Stimulus
T11
T12
T13
T14
TDR
Port 1 to 1
TDT
Port 2 to 1
TDT
Port 3 to 1
TDT
Port 4 to 1
T21
T22
T23
T24
TDT
Port 1 to 2
TDR
Port 2 to 2
TDT
Port 3 to 2
TDT
Port 4 to 2
T31
T32
T33
T34
TDT
Port 1 to 3
TDT
Port 2 to 3
TDR
Port 3 to 3
TDT
Port 4 to 3
T41
T42
T43
T44
TDT
Port 1 to 4
TDT
Port 2 to 4
TDT
Port 1 to 4
TDR
Port 4 to 4
Response
Response
Stimulus
S11
REFL
Port 1 to 1
S12
TRANS
Port 2 to 1
S13
TRANS
Port 3 to 1
S14
TRANS
Port 4 to 1
S21
TRANS
Port 1 to 2
S22
REFL
Port 2 to 2
S23
TRANS
Port 3 to 2
S24
TRANS
Port 4 to 2
S31
TRANS
Port 1 to 3
S32
TRANS
Port 2 to 3
S33
REFL
Port 3 to 3
S34
TRANS
Port 4 to 3
S41
TRANS
Port 1 to 4
S42
TRANS
Port 2 to 4
S43
TRANS
Port 1 to 4
S44
REFL
Port 4 to 4
Figure 32. Single-ended frequency domain formats.
Figure 30. Single-ended time domain formats.
Stimulus
Common
TDT
Port 2 to 1
DIFF to DIFF
TDD21
TDD22
TDT
Port 1 to 2
DIFF to DIFF
TDR
Port 2 to 2
DIFF to DIFF
Common
TCD11
TCD12
TDR
TDT
Port 1 to 1
Port 2 to 1
DIFF to COM DIFF to COM
TCD21
TDR
TDT
Port 1 to 1
Port 2 to 1
COM to DIFF COM to DIFF
TDC21
TDC22
TDT
TDR
Port 1 to 2
Port 2 to 2
COM to DIFF COM to DIFF
TCC11
TCC12
TDR
TDT
Port 1 to 1
Port 2 to 1
COM to COM COM to COM
TCD22
TDT
TDR
Port 1 to 2
Port 2 to 2
DIFF to COM DIFF to COM
TDC12
Differential
TDR
Port 1 to 1
DIFF to DIFF
TDC11
TCC21
TCC22
TDT
TDR
Port 1 to 2
Port 2 to 2
COM to COM COM to COM
Figure 31. Mixed-mode time domain formats.
Response
TDD12
Common
Differential
Common
Response
Differential
Differential
TDD11
Stimulus
SDD11
SDD12
SDC11
REFL
Port 1 to 1
DIFF to DIFF
TRANS
Port 2 to 1
DIFF to DIFF
REFL
TRANS
Port 1 to 1
Port 2 to 1
COM to DIFF COM to DIFF
SDD21
TRANS
Port 1 to 2
DIFF to DIFF
SDD22
REFL
Port 2 to 2
DIFF to DIFF
SDC21
SDC22
TRANS
TRANS
Port 1 to 2
Port 2 to 2
COM to DIFF COM to DIFF
SCD11
SCD12
SCC11
SDC12
SCC12
REFL
TRANS
Port 1 to 1
Port 2 to 1
DIFF to COM DIFF to COM
REFL
TRANS
Port 1 to 1
Port 2 to 1
COM to COM COM to COM
SCD21
SCD22
TRANS
REFL
Port 1 to 2
Port 2 to 2
DIFF to COM DIFF to COM
SCC21
SCC22
TRANS
REFL
Port 1 to 2
Port 2 to 2
COM to COM COM to COM
Figure 33. Mixed-mode frequency domain formats.
15
Eye diagram: Twelve single-ended and four mixed-mode
parameters are available (reflection terms and mode-conversion
terms are not applicable for eye-diagrams). These parameters
correspond to forward and reverse transmission in single-ended,
differential, common-mode, and mixed-modes.
Stimulus
T12
TDT
Port 2 to 1
_
T13
TDT
Port 3 to 1
T14
TDT
Port 4 to 1
T23
T24
TDT
Port 3 to 2
TDT
Port 4 to 2
Response
T21
TDT
Port 1 to 2
_
T31
T32
TDT
Port 1 to 3
TDT
Port 2 to 3
_
T41
T42
T43
TDT
Port 1 to 4
TDT
Port 2 to 4
TDT
Port 1 to 4
T34
TDT
Port 4 to 3
_
Figure 34. Single-ended eye diagram formats.
RLCG model extraction: Thirty-two transmission line parameters
are available. These correspond to resistive loss, inductance,
capacitance, dielectric loss, characteristic impedance (real and
imaginary), attenuation constant, and propagation constant, in
differential, common-mode, W-element, and self and mutual
terms. Current capability allows RLCG extraction for coupled
transmission lines only.
RD
Resistance
LD
Inductance
CD
Capacitance
GD
Conductance
ZORD
Char. Imp.Real
ZOID
Char. Imp.Imag.
AD
Atten.
constant
BD
Propagation
constant
Figure 36. Differential RLCG formats.
RC
LC
CC
GC
Resistance
Inductance
Capacitance
Conductance
ZORC
Char. Imp.Real
ZOIC
Char. Imp.Imag.
AC
Atten.
constant
BC
Propagation
constant
Figure 37. Common-mode RLCG formats.
Stimulus
Common
Differential
Common
Response
Differential
TDD21
TDD12
TDT
Port 1 to 2
DIFF to DIFF
TDT
Port 2 to 1
DIFF to DIFF
–
–
–
TCC21
–
TCC12
TDT
TDT
Port 1 to 2
Port 2 to 1
COM to COM COM to COM
Figure 35. Mixed-mode eye diagram formats.
R11
L11
C11
G11
Self
resistance
Self
inductance
Self
capacitance
Self
conductance
R12
Mutual
resistance
L12
Mutual
inductance
C12
Mutual
capacitance
G12
Mutual
conductance
Figure 38. W-element RLCG formats.
RS
Self
resistance
LS
Self
inductance
CS
Self
capacitance
GS
Self
conductance
RM
Mutual
resistance
LM
Mutual
inductance
CM
Mutual
capacitance
GM
Mutual
conductance
Figure 39. Self/mutual RLCG formats.
16
PLTS System Performance
Definitions
To specify the performance of a Physical Layer Test System
(PLTS), this data sheet lists the dynamic range, measurement
uncertainty, and measurement port characteristics for each system configuration. Two types of performance numbers are offered:
specifications and characteristics.
Specifications describe the instrument’s warranted performance
over the temperature range of 23 °C ± 3 °C.
Characteristics are typical but non-warranted performance
parameters. These are further denoted as "typical" or "nominal."
• Typical (typ.): Expected performance of an average unit, not
including guardbands.
• Nominal (nom.): A general, descriptive term that does not
imply a level of performance.
Measurement port characteristics indicate the RF performance
of network analyzer and test set port leakages, mismatches, and
frequency response. The specification for the test set’s crosstalk
does not include noise.
Dynamic range (Signal-to-Noise Ratio) is further defined as
Pref – Pmin, where Pref is the nominal or reference power out of
a source test port, and Pmin is the minimum power into a receiver
test port that can be measured above the peaks of the system’s
noise floor (10 dB above the average noise floor). System dynamic
range is the amount of attenuation that can be measured from a
0 dB reference.
Calibration is the process of measuring standards that have fully
defined models (and are thus called "known" standards) in order
to quantify a network analyzer’s systematic errors based on an
error model. Calibration must be performed within the operating
temperature specified for the calibration kit. For all calibration kits
the operating temperature is 23 °C ±3 °C. For a calibration to
remain fully verifiable, the temperature of the network analyzer
must remain within ±1 °C around the initial measurement calibration temperature.
Error correction is the process of mathematically removing from
the measurement those systematic errors determined by measurement calibration.
Measurement uncertainty curves show the worst-case uncertainty in reflection and transmission measurements using full
error correction with the specified calibration kit. This includes
residual systematic errors, as well as system dynamic accuracy,
connector repeatability, noise and detector errors. Cable stability
and system drift are not included. Furthermore, the graphs for
reflection measurement uncertainty apply to a one-port device.
The graphs for transmission measurement uncertainty assume a
well-matched device (S11= S22 = 0). In the phase uncertainty
curves, the phase detector accuracy is better than 0.02 degrees,
useful for measurements where only phase changes.
17
System Performance Summary
N1953B Physical Layer Test System
36 ps (10 MHz to 20 GHz)
The following characteristics are applicable for a system in the
following configuration:
Network analyzer
Test set
Test cables
Calibration kit
Calibration technique
Characteristic measurement uncertainties
Agilent E8362B, Option 014/UNL
Agilent N4419B
Agilent N4419B, Option B20
Agilent 85052D, 3.5 mm
Four-port SOLT
Dynamic range (signal-to-noise ratio)
Transmission measurements at 10 Hz IF bandwidth, with full
four-port error correction, and -5 dBm output power.
Frequency
10 to 45 MHz
45 to 500 MHz
500 MHz to 10 GHz
10 to 20 GHz
Dynamic range
60 dB typical
70 dB typical
100 dB
85 dB
Figure 40. Worst case 3.5 mm transmission magnitude and
phase uncertainty.
Measurement port characteristics
Residual uncertainties for corrected data. These apply for 25 °C
with less than 1 °C variation from calibration.
Directivity
Source match
Load match
Refl. tracking
Trans. tracking
45 MHz
to 2 GHz
56 dB
42 dB
56 dB
± .0025 dB
± .020 dB
Test set typical performance
Frequency range
Transition time
(10 to 90%, TR=.72/BW)
Impedance
Maximum operating level
Damage level
Test port connectors
RF connectors
Weight
18
2 to
10 GHz
42 dB
36 dB
42 dB
± .009 dB
± .032 dB
10 to
20 GHz
40 dB
33 dB
40 dB
± .013 dB
± .050 dB
Figure 41. Worst case 3.5 mm reflection magnitude and phase uncertainty.
10 MHz to 20 GHz
36 ps
50 ohms (nom)
+20 dBm
+30 dBm (typ)
3.5 mm (m)
SMA (f)
9 kg
System Performance Summary
N1955B Physical Layer Test System
18 ps (10 MHz to 40 GHz)
The following characteristics are applicable for a system in the
following configuration:
Network analyzer:
Test set:
Test cables:
Calibration kit:
Calibration technique:
Characteristic measurement uncertainties
Agilent E8363B, Option 014/UNL
Agilent N4420B
Agilent N4420B-B20
Agilent 85056A, 2.4 mm
Four-port SOLT
Dynamic range
Transmission measurements at 10 Hz IF bandwidth, with four-port
error correction and -17 dBm output power.
Frequency
10 to 45 MHz
45 to 500 MHz
500 MHz to 20 GHz
20 to 40 GHz
Dynamic range
65 dB typical
55 dB
70 dB
55 dB
Figure 42. Worst case 3.5 mm transmission magnitude and
phase uncertainty.
Measurement port characteristics
Residual uncertainties for corrected data. These apply for 25 °C
with less than 1 °C variation from calibration.
45 to
500 MHz
Directivity
43 dB
Source match
38 dB
Load match
43 dB
Refl. tracking ± .001 dB
Trans. tracking ± .015 dB
Test set typical performance
Frequency range
Transition time
(10 to 90%, TR=.72/BW)
Impedance
Maximum operating level
Damage level
Test port connectors
RF connectors
Weight
500 MHz
to 10 GHz
39.5 dB
34 dB
39.5 dB
± .002 dB
± .020 dB
10 to
20 GHz
39 dB
34 dB
39 dB
± .008 dB
± .040 dB
20 to
40 GHz
33 dB
27 dB
33 dB
± .026 dB
±.20 dB
Figure 43. Worst case 3.5 mm reflection magnitude and phase uncertainty.
10 MHz to 40 GHz
18 ps
50 ohms (nom)
+20 dBm
+30 dBm (typ)
2.4 mm (m)
2.4 mm (f)
9 kg
19
System Performance Summary
N1957B Physical Layer Test System
14 ps (10 MHz to 50 GHz)
The following characteristics are applicable for a system in the
following configuration:
Network analyzer
Test set:
Test cables:
Calibration kit:
Calibration technique:
Characteristic measurement uncertainties
Agilent E8364B, Option014/UNL
Agilent N4421B
Agilent N4421B-B20
Agilent 85056A, 2.4 mm
Four-port SOLT
Dynamic Range
Transmission measurements at 10 Hz IF bandwidth, with four-port
error correction and -17 dBm output power.
Frequency
10 to 45 MHz
45 to 500 MHz
500 MHz to 20 GHz
20 to 50 GHz
Dynamic range
55 dB typical
55 dB
70 dB
55 dB
Figure 44. Worst case 2.4 mm transmission magnitude and
phase uncertainty.
Measurement port characteristics
Residual uncertainties for corrected data. These apply for 25 °C
with less than 1 °C variation from calibration.
45 to
500 MHz
Directivity
43 dB
Source match
38 dB
Load match
43 dB
Refl. tracking ± .001 dB
Trans. tracking ± .015 dB
Test set typical performance
Frequency range
Transition time
(10 to 90%, TR=.72/BW)
Impedance
Maximum operating level
Damage level
Test port connectors
RF connectors
Weight
20
500 MHz
to 10 GHz
39.5 dB
34 dB
39.5 dB
± .002 dB
± .020 dB
10 to
20 GHz
39 dB
34 dB
39 dB
± .008 dB
± .040 dB
20 to
40 GHz
33 dB
27 dB
33 dB
± .026 dB
± .20 dB
Figure 45. Worst case 2.4 mm reflection magnitude and phase uncertainty.
10 MHz to 50 GHz
14 ps
50 ohms (nom)
+20 dBm
+30 dBm (typ)
2.4 mm (m)
2.4 mm (f)
9 kg
PLTS Ordering Guide
Physical Layer Test Systems
Quick configuration guide
Network analyzer based systems
See the PLTS ordering guide on page 23 for more detail.
System
N1957B
N1955B
N1953B
Description
E8364B
PNA, 4-receiver
N4421B
S-parameter test set
N1930A
Physical Layer Test System SW
E8363B
PNA, 4-receiver
N4420B
S-parameter test set
N1930A
Physical Layer Test System SW
E8362B
PNA, 4-receiver
N4419B
S-parameter test set
N1930A
Physical Layer Test System SW
Frequency range
Transition time
(10-90%, TR = 0.72/BW)
10 MHz to 50 GHz
≥ 14.4 ps
10 MHz to 40 GHz
≥ 18 ps
10 MHz to 20 GHz
≥ 36 ps
Receiver bandwidth (3 dB)
Transition time
(10-90%, TR = 0.72/BW)
Time domain reflectometer (TDR) based systems
See the PLTS ordering guide on page 26 for more detail.
System
Description
• Agilent 86100 family Infiniium DCA wide-bandwidth oscilloscope1
• Agilent 54754A TDR module(s) (1 or 2 modules supported)
40 ps (nominal)
12.4 or 18 GHz
≥ 10 ps (normalized)2
20 GHz
28 ps (typical)
• Agilent N1930A PLTS SW
• Tektronix CSA 8000 communications signal analyzer3, 4 or
Tektronix TDS 8000 digital sampling oscilloscope3, 4
• Tektronix 80E04 dual channel, 20 GHz TDR sampling module(s)
(1 or 2 modules supported)
• Agilent N1930A PLTS software
1. PLTS compatibility requires 86100 firmware revision 03.06 or higher.
2. Adjustable from larger of 10 ps or 0.08 x time/division.
3. Tektronix bandwidth and transition time specifications from Tektronix
data sheet 85W-13499-8, dated 10/03.
4. PLTS compatibility requires firmware revision 1.3.3 or higher.
21
PLTS Ordering Guide
PNA-based systems
A complete PNA-based Physical Layer
Test System includes the following:
• Four-port network analyzer system
• Physical Layer Test System software,
running on a user-supplied PC
(see requirements below)
• Software License (node-locked or
server-based)1
• Sample DUT board
(balanced transmission line)
N1957B
Physical Layer Test System
N1955B
Physical Layer Test System
10 MHz to 50 GHz (14 ps)
10 MHz to 40 GHz (18 ps)
E8364B
E8364B-014
N4421B
N1930A
N1930A-010
N1930A-020
E8363B
E8363B-014
N4420B
N1930A
N1930A-010
N1930A-020
PNA network analyzer, 4-rcvr
Configurable test set
S-parameter test set
Physical Layer Test System SW
Node-locked license
Server-based/floating license
Recommended accessories:
N4421A-B20
Test port cables, 3 ft.,
2.4 mm (m-f), quantity 4
N4421B-1CP
Rack mount kit, test set
85056A
Calibration kit 2.4 mm
• User documentation
Precision test cables and rack mount kits
with handles are available as options.
Calibration kits and ECal modules can be
ordered separately, or as accessories in a
bundled system. Start-up and productivity
assistance is also available.
PNA network analyzer, 4-rcvr
Configurable test set
S-parameter test set
Physical Layer Test System SW
Node-locked license
Server-based/floating license
Recommended accessories:
N4421A-B20
Test port cables, 3 ft.,
2.4 mm (m-f), quantity 4
N4420B-1CP
Rack mount kit, test set
85056A
Calibration kit 2.4 mm
N1953B
Physical Layer Test System
10 MHz to 20 GHz (36 ps)
User-supplied PC requirements:
• 1 GHz Pentium® III or better
E8362B
E8362B-014
N4419B
N1930A
N1930A-010
N1930A-020
• ≥ 512 MB of main memory
(≥ 1 GB recommended)
• Windows® 2000 or XP
(XP is recommended)
PNA network analyzer, 4-rcvr
Configurable test set
S-parameter test set
Physical Layer Test System SW
Node-locked license
Server-based/floating license
Recommended accessories:
N4419A-B20
Test port cables, 3 ft.,
3.5 mm (m-f), quantity 4
N4419B-1CP
Rack mount kit, test set
85052D
Calibration kit 3.5 mm
• A CD-ROM drive
• A supported GPIB card:
- Agilent 82340, 82341, 82350 GPIB
Interface
- Agilent 82357A USB/GPIB Interface
- any National Instruments GPIB card
Upgrades from two-port PNA to four-port PLTS
Supported network analyzers
PNA
model
number
E8364A
E8364B
E8363A
E8363B
E8362A
E8362B
Network analyzer options
Required
Compatible
N4421B
014
010, 022, UNL
1. See software licensing sidebar on page 28.
22
Not tested
or specified
Required
test set
016, 080, 081, 083
H08, H11
N4420B
N4419B
System
frequency
range
45 MHz to 50 GHz
10 MHz to 50 GHz
45 MHz to 40 GHz
10 MHz to 40 GHz
45 MHz to 20 GHz
10 MHz to 20 GHz
Required
application
software
N1930A
Ordering Guide
TDR-based systems
A complete TDR-based Physical Layer Test System includes the
following:
User-supplied PC requirements:
• 1 GHz Pentium III or better
• An oscilloscope mainframe with (1) or (2) TDR plug-in modules
• ≥ 512 MB of main memory (≥ 1 GB recommended)
• Physical Layer Test System software, running on a user-supplied
PC (see requirements at right)
• Windows 2000 or XP (XP is recommended)
• Software License (node-locked or
server-based)1
• User documentation
Calibration kits and test port cables are available separately.
Start-up and productivity assistance is also available.
• A CD-ROM drive
• A supported GPIB card:
- Agilent 82340, 82341, 82350 GPIB Interface
- Agilent 82357A USB/GPIB Interface
- any National Instruments GPIB card
TDR-Based Physical Layer Test Systems
TDR module
Receiver bandwidth
Transition time
Requred application
(1 or 2 modules supported)
(3 dB)
(10 to 90%, TR=0.35/BW)
software
Mainframe
Agilent 86100 family Infiniium DCA Agilent 54754A
wide-bandwidth oscilloscope
Differential
(86100A/B/C)2
TDR/TDT plug in module
Tektronix CSA 8000
communications signal analyzer3
Tektronix 80E04 dual
channel, 20 GHz TDR
sampling module
Tektronix TDS 8000
digital sampling oscilloscope3
Tektronix 80E04 dual
channel, 20 GHz TDR
sampling module
12.4 or 18 GHz
40 ps (nominal)
≥ 10 ps (normalized)
Agilent N1930A
20 GHz
28 ps (typical)
Agilent N1930A
Recommended accessories
N4419B-B20
Test port cables, 3 ft., 3.5 mm (m-f), quantity 4
8120-4948
Cable assembly, 3 ft., SMA (m-m)
N1020A
Time domain reflectometry probe, 6 GHz, variable pitch
Includes N1020A-K05 calibration substrate
1250-2604
Adapter, radial bend, SMA (m-f)
1250-2151
Termination, SMA (f), 50 ohm load
SMA female load required for normalization
1250-2152
Termination, SMA (f), short
SMA female short required for normalization
1250-2153
Termination, SMA (m), short
SMA male short required for normalization
1810-0118
Termination, SMA (m), 50 ohm load
SMA male load required for normalization
5061-5311
Adapter, 3.5mm (f-f)
Thru adapter required for TDT normalization
1. See software licensing sidebar on page 28.
2. PLTS compatibility requires 86100 firmware revision 03.06 or higher.
3. PLTS compatibility requires Tektronix CSA/TDS 8000 firmware
revision 1.3.3 or higher.
23
www.agilent.com
PLTS application software uses FLEXlm licensing. Two licensing options
are available:
Option 010 Node-locked license (default). Node-locked licenses are locked to a single
PC through its host ID (e.g., the MAC address) with or without connected system hardware. Node-locked licenses do not require any networking or license server processes.
Option 020 Floating (server-based) license. Floating licenses allow users to share a
single license, or multiple licenses, over a network. The application software may be
installed on an unlimited number of PC's with or without connected system hardware.
The number of available licenses determines the number of concurrent users. Floating
licenses require a license server and a TCP/IP (or IPX/SPX) connection between
clients and server(s).
Web Resources
Visit our Web sites for additional product and
literature information.
Physical Layer Test Systems
www.agilent.com/find/plts
Network Analyzers
www.agilent.com/find/na
Electronic Calibration (ECal)
www.agilent.com/find/ecal
Agilent Technologies’ Test and Measurement Support,
Services, and Assistance
Agilent Technologies aims to maximize the value you
receive, while minimizing your risk and problems. We
strive to ensure that you get the test and measurement
capabilities you paid for and obtain the support you need.
Our extensive support resources and services can help
you choose the right Agilent products for your applications
and apply them successfully. Every instrument and system
we sell has a global warranty. Two concepts underlie
Agilent’s overall support policy: “Our Promise” and “Your
Advantage.”
Our Promise
Our Promise means your Agilent test and measurement
equipment will meet its advertised performance and
functionality. When you are choosing new equipment, we
will help you with product information, including realistic
performance specifications and practical recommendations
from experienced test engineers. When you receive your
new Agilent equipment, we can help verify that it works
properly and help with initial product operation.
Your Advantage
Your Advantage means that Agilent offers a wide range of
additional expert test and measurement services, which
you can purchase according to your unique technical and
business needs. Solve problems efficiently and gain a
competitive edge by contracting with us for calibration,
extra-cost upgrades, out-of-warranty repairs, and onsite
education and training, as well as design, system integration, project management, and other professional engineering services. Experienced Agilent engineers and technicians worldwide can help you maximize your productivity, optimize the return on investment of your Agilent
instruments and systems, and obtain dependable measurement accuracy for the life of those products.
For more information on Agilent Technologies’ products,
applications or services, please contact your local
Agilent office.
Agilent Email Updates
www.agilent.com/find/emailupdates
Get the latest information on the products and applications you
select.
Agilent Direct
www.agilent.com/find/agilentdirect
Quickly choose and use your test equipment
solutions with confidence.
Agilent Open
www.agilent.com/find/open
Agilent Open simplifies the process of connecting
and programming test systems to help engineers
design, validate and manufacture electronic products. Agilent offers open connectivity for a broad
range of system-ready instruments, open industry
software, PC-standard I/O and global support, which
are combined to more easily integrate test system
development.
Microsoft® is a U.S. registered trademark of Microsoft Corp.
Windows® and MS Windows® are U.S. registered trademarks of Microsoft Corp.
Pentium is a US registered trademark of Intel Corporation
Phone or Fax
United States:
(tel) 800 829 4444
(fax) 800 829 4433
Canada:
(tel) 877 894 4414
(fax) 800 746 4866
China:
(tel) 800 810 0189
(fax) 800 820 2816
Europe:
(tel) 31 20 547 2111
Japan:
(tel) (81) 426 56 7832
(fax) (81) 426 56 7840
Korea:
(tel) (080) 769 0800
(fax) (080)769 0900
Latin America:
(tel) (305) 269 7500
Taiwan:
(tel) 0800 047 866
(fax) 0800 286 331
Other Asia Pacific
Countries:
(tel) (65) 6375 8100
(fax) (65) 6755 0042
Email: tm_ap@agilent.com
Contacts revised: 05/27/05
The complete list is available at:
www.agilent.com/find/contactus
Product specifications and descriptions in this document
subject to change without notice.
© Agilent Technologies, Inc. 2004, 2005, 2006
Printed in USA, January 4, 2006
5989-0271EN