in any form or by any means (including
electronic storage and retrieval or translation into a foreign language) without prior
agreement and written consent from Agilent Technologies, Inc. as governed by
United States and international copyright
laws.
Trademark
Acknowledgements
Microsoft® is a U.S. registered
trademark of Microsoft Corporation.
Windows
U.S. registered trademarks of
Microsoft Corporation.
Adobe Reader
trademark of Adobe System
Incorporated.
Java™ is a U.S. trademark of Sun
Microsystems, Inc.
MATLAB® is a U.S. registered
trademark of Math Works, Inc.
Norton Ghost™ is a U.S. trademark
of Symantec Corporation.
®
and MS Windows® are
®
is a U.S. registered
Manual Part Number
N9010-90025
Supersedes: August 2009
Print Date
February 2010
Printed in USA
Agilent Technologies, Inc.
1400 Fountaingrove Parkway
Santa Rosa, CA 95403
Warranty
The material contained in this document is provided “as is,” and is
subject to being changed, without
notice, in future editions. Further,
to the maximum extent permitted
by applicable law, Agilent disclaims
all warranties, either express or
implied, with regard to this manual
and any information contained
herein, including but not limited to
the implied warranties of merchantability and fitness for a particular purpose. Agilent shall not
be liable for errors or for incidental
or consequential damages in connection with the furnishing, use, or
performance of this document or of
any information contained herein.
Should Agilent and the user have a
separate written agreement with
warranty terms covering the material in this document that conflict
with these terms, the warranty
terms in the separate agreement
shall control.
Technology Licenses
The hardware and/or software described
in this document are furnished under a
license and may be used or copied only in
accordance with the terms of such license.
Restricted Rights Legend
If software is for use in the performance
of a U.S. Government prime contract or
subcontract, Software is delivered and
licensed as “Commercial computer software” as defined in DF AR 252.227-7014
(June 1995), or as a “commercial item” as
defined in F AR 2.101(a) or as “Restricted
computer software” as defined in FAR
52.227-19 (June 1987) or any equivalent
agency regulation or contract clause. Use,
duplication or disclosure of Software is
subject to Agilent Technologies’ standard
commercial license terms, and non-DOD
Departments and Agencies of the U.S.
Government will receive no greater than
Restricted Rights as defined in FAR
52.227-19(c)(1-2) (June 1987). U.S. Government users will receive no greater than
Limited Rights as defined in FAR 52.22714 (June 1987) or DFAR 252.227-7015
(b)(2) (November 1995), as applicable in
any technical data.
Safety Notices
CAUTION
A CAUTION notice denotes a
hazard. It calls attention to an
operating procedure, practice, or
the like that, if not correctly performed or adhered to, could result
in damage to the product or loss of
important data. Do not proceed
beyond a CAUTION notice until
the indicated conditions are fully
understood and met.
WARNING
A WARNING notice denotes a
hazard. It calls attention to an
operating procedure, practice,
or the like that, if not correctly
performed or adhered to, could
result in personal injury or
death. Do not proceed beyond a
WARNING notice until the indicated conditions are fully
understood and met.
2
Notice
Warranty
This Agilent technologies instrument product is warranted against defects in material and workmanship for
a period of one year from the date of shipment. During the warranty period, Agilent Technologies will, at
its option, either repair or replace products that prove to be defective.
For warranty service or repair, this product must be returned to a service facility designated by Agilent
T echnologies. Buyer shall prepay shipping charges to Agilent Technologies and Agilent Technologies shall
pay shipping charges to return the product to Buyer. However , Bu yer shall pay all shipping char ges, duties,
and taxes for products returned to Agilent Technologies from another country.
Where to Find the Latest Information
Documentation is updated periodically. For the latest information about this analyzer, including firmware
upgrades, application information, and product information, see the following URLs:
http://www.agilent.com/find/exa
To receive the latest updates by email, subscribe to Agilent Email Updates:
http://www.agilent.com/find/emailupdates
Information on preventing analyzer damage can be found at:
This chapter contains the specifications for the core signal analyzer. The specifications and
characteristics for the measurement applications and options are covered in the chapters that follow.
13
Agilent EXA Signal Analyzer
Definitions and Requirements
Definitions and Requirements
This book contains signal analyzer specifications and supplemental information. The distinctio n amon g
specifications, typical performance, and nominal values are described as follows.
Definitions
•Specifications describe the performance of parameters covered by the product warranty (temperature
= 5 to 50°C, unless otherwise noted).
•95th percentile values indicate the breadth of the population (≈2σ) of performance tolerances
expected to be met in 95% of the cases with a 95% confidence, for any ambient temperature in the
range of 20 to 30°C. In addition to the statistical observations of a sample of instruments, these values
include the effects of the uncertainties of external calibration references. These values are not
warranted. These values are updated occasionally if a significant change in the statistically observed
behavior of production instruments is observed.
•Typical describes additional product performance information that is not covered by the product
warranty. It is performance beyond specification that 80% of the units exhibit with a 95% confidence
level over the temperature range 20 to 30°C. Typical performance does not include measurement
uncertainty.
•Nominal values indicate expected performance, or describe product performance that is useful in the
application of the product, but is not covered by the product warranty.
Conditions Required to Meet Specifications
The following conditions must be met for the analyzer to meet its specifications.
•The analyzer is within its calibration cycle. See the General section of this chapter.
•Under auto couple control, except that Auto Sweep Time Rules = Accy.
•For signal frequencies < 10 MHz, DC coupling applied.
•Any analyzer that has been stored at a temperature range inside the allowed storage range but outside
the allowed operating range must be stored at an ambient temperature within the allowed operating
range for at least two hours before being turned on.
•The analyzer has been turned on at least 30 minutes with Auto Align set to Normal, or if Auto Align
is set to Off or Partial, alignments must have been run recently enough to prevent an Alert message. If
the Alert condition is changed from “Time and Temperature” to one of the disabled duration choices,
the analyzer may fail to meet specifications without informing the user.
Certification
Agilent Technologies certifies that this product met its published specifications at the time of shipment
from the factory. Agilent Technologies further certifies that its calibration measurements are traceable to
the United States National Institute of Standards and Technology, to the extent allowed by the Institute’s
calibration facility, and to the calibration facilities of other International Standards Organization
members.
14Chapter 1
Agilent EXA Signal Analyzer
Frequency and Time
Frequency and Time
DescriptionSpecificationsSupplemental Information
Frequency Range
a. In the band overlap regions, for example, 3.5 to 3.6 GHz, the analyzer may use either band for
measurements, in this example Band 0 or Band 1. The analyzer gives preference to the band
with the better overall specifications (which is the lower numbered band for all frequencies
below 26 GHz), but will choose the other band if doing so is necessary to achieve a sweep
having minimum band crossings. For example, with CF = 3.58 GHz, with a span of 40 MHz
or less, the analyzer uses Band 0, because the stop frequency is 3.6 GHz or less, allowing a
span without band crossings in the preferred band. If the span is between 40 and 160 MHz,
the analyzer uses Band 1, because the start frequency is above 3.5 GHz, allowing the sweep to
be done without a band crossing in Band 1, though the stop frequency is above 3.6 GHz, preventing a Band 0 sweep without band crossing. With a span greater than 160 MHz, a band
crossing will be required: the analyzer sweeps up to 3.6 GHz in Band 0; then executes a band
crossing and continues the sweep in Band 1.
Specifications are given separately for each band in the band overlap regions. One of these
specifications is for the preferred band, and one for the alternate band. Continuing with the
example from the previous paragraph (3.58 GHz), the preferred band is ba nd 0 (indicated as
frequencies under 3.6 GHz) and the alternate band is band 1 (3.5 to 8.4 GHz). The specifications for the preferred band are warranted. The specifications for the alternate band are not
warranted in the band overlap region, but performance is nominally the same as those warranted specifications in the rest of the band. Again, in this example, consider a signal at 3.58
GHz. If the sweep has been configured so that the signal at 3.58 GHz is measured in Band 1,
the analysis behavior is nominally as stated in the Band 1 specification line (3.5 – 8.4 GHz)
but is not warranted. If warranted performance is necessary for this signal, the sweep should
be reconfigured so that analysis occurs in Band 0. Another way to express this situation in this
example Band 0/Band 1 crossing is this: The specifications given in the “Specifications” column which are described as “3.5 to 8.4 GHz” represent nominal performance from 3.5 to 3.6
GHz, and warranted performance from 3.6 to 8.4 GHz.
b. N is the LO multiplication factor. For negative mixing modes (as indicated by the “−” in the
“Harmonic Mixing Mode” column), the desired 1st LO harmonic is higher than the tuned frequency by the 1st IF (5.1225 GHz for band 0, 322.5 MHz for all other bands).
DescriptionSpecificationsSupplemental Information
Standard Frequency Reference
Accuracy±[(time since last adjustment × aging
rate) + temperature stability +
a
]
b
Temperature Stability
20 to 30 °C
5 to 50 °C
Aging Rate
calibration accuracy
±2 × 10
±2 × 10
±1 × 10
16Chapter 1
−6
−6
−6
/year
Agilent EXA Signal Analyzer
Frequency and Time
DescriptionSpecificationsSupplemental Information
Achievable Initial Calibration
±1.4 × 10
−6
Accuracy
Settability
Residual FM
Center Frequency = 1 GHz
±2 × 10
−8
≤10 Hz × N p-p in 20 ms
nominal
10 Hz RBW, 10 Hz VBW
a. Calibration accuracy depends on how accurately the frequency standard was adjusted to 10
MHz. If the adjustment procedure is followed, the calibration accuracy is given by the specifi-
cation “Achievable Initial Calibration Accuracy.”
b. For periods of one year or more.
c. N is the LO multiplication factor.
c
,
Chapter 1 17
Agilent EXA Signal Analyzer
Frequency and Time
DescriptionSpecificationsSupplemental Information
Precision Frequency Refere nce
(Option PFR)
Accuracy±[(time since last adjustment
× aging rate) + temperature
stability + calibration
accuracy
a]b
Temperature Stability
−8
−8
±5 × 10
−10
20 to 30 °C
5 to 50 °C
Aging Rate
±1.5 × 10
±5 × 10
Total Aging
1 Year
2 Years
Settability
Warm-up and Retrace
300 s after turn on
900 s after turn on
±1 × 10
±1.5 × 10
±2 × 10
c
−7
−9
−7
±1 × 10
−7
(nominal)
±1 × 10
−8
of final frequency
of final frequency
(nominal)
Achievable Initial Calibration Accuracy
d
±4 × 10
−8
Standby power to reference oscillatorNot supplied
Residual FM
Center Frequency = 1 GHz
≤ 0.25 Hz x N p-p in 20 ms
(nominal)
/day (nominal)
e
10 Hz RBW, 10 Hz VBW
a. Calibration accuracy depends on how accurately the frequency standard was adjusted to 10
MHz. If the adjustment procedure is followed, the calibration accuracy is given by the specifi-
cation “Achievable Initial Calibration Accuracy.”
b. The specification applies af ter the analyzer has been powered on for four hours.
c. Standby mode does not apply power to the oscillator. Therefore warm-up applies every time
the power is turned on. The warm-up reference is one hour after turning the power on. Retrac-
ing also occurs every time the power is applied. The effect of retracing is included within the
“Achievable Initial Calibration Accuracy” term of the Accuracy equation.
18Chapter 1
Agilent EXA Signal Analyzer
Frequency and Time
d. The achievable calibration accuracy at the beginning of the calibration cycle includes these
effects:
1) Te mperature difference between the calibration environment and the use environment
2) Orientation relative to the gravitation field changing between the calibration environment
and the use environment
3) Retrace effects in both the calibration environment and the use environment due to turning
the instrument power off.
4) Settability
e. N is the LO multiplication factor.
Chapter 1 19
Agilent EXA Signal Analyzer
Frequency and Time
DescriptionSpecificationsSupplemental
Information
Frequency Readout Accuracy±(marker freq. × freq. ref. accy. + 0.25%
a
+ 2 Hz + 0.5 ×
b
)
Example for EMC
× span + 5%× RBW
horizontal resolution
d
Single detector only
±0.0032% (nominal)
a. The warranted performance is only the sum of all errors under autocoupled conditions. Under
non-autocoupled conditions, the frequency readout accuracy will nominally meet the specifica-
tion equation, except for conditions in which the RBW term dominates, as explained in exam-
ples below. The nominal RBW contribution to frequency readout accuracy is 2% of RBW for
RBWs from 1 Hz to 390 kHz, 4% of RBW from 430 kHz through 3 MHz (the widest autocou-
pled RBW), and 30% of RBW for the (manually selected) 4, 5, 6 and 8 MHz RBWs.
First example: a 120 MHz span, with autocoupled RBW. The autocoupled ratio of span to
RBW is 106:1, so the RBW selected is 1.1 MHz. The 5% × RBW term contributes only 55 kHz
to the total frequency readout accuracy, compared to 300 kHz for the 0.25% × span term, for a
total of 355 kHz. In this example, if an instrument had an unusually high RBW centering error
of 7% of RBW (77 kHz) and a span error of 0.20% of span (240 kHz), the total actual error
(317 kHz) would still meet the computed specification (355 kHz).
Second example: a 20 MHz span, with a 4 MHz RBW. The specification equation does not
apply because the Span: RBW ratio is not autocoupled. If the equation did apply, it would
allow 50 kHz of error (0.25%) due to the span and 200 kHz error (5%) due to the RBW. For
this non-autocoupled RBW, the RBW error is nominally 30%, or 1200 kHz.
b. Horizontal resolution is due to the marker reading out one of the trace points. The points are
spaced by span/(Npts - 1), where Npts is the number of sweep points. For example, with the
factory preset value of 1001 sweep points, the horizontal resolution is span/1000. However,
there is an exception: When both the detector mode is “normal” and the span > 0.25 × (Npts -
1) × RBW, peaks can occur only in even-numbered points, so the effective horizontal resolu-
tion becomes doubled, or span/500 for the factory preset case. When the RBW is autocoupled
and there are 1001 sweep points, that exception occurs only for spans > 750 MHz
c. Specifications apply to traces in two cases: when all active traces use the same detector, and to
any trace that uses the peak detector. When multiple simultaneous detectors are in use, addi-
tional errors of 0.5, 1.0 or 1.5 display points will occur in some detectors, depending on the
combination of detectors in use. In one example, with positive peak, negative peak and average
detection, there is an additional error only in the average detection trace, which shifts the
apparent signal position left by 0.5 display points.
c
20Chapter 1
Agilent EXA Signal Analyzer
Frequency and Time
d. In most cases, the frequency readout accuracy of the analyzer can be exceptionally good. As an
example, Agilent has characterized the accuracy of a span commonly used for Electro-Magnetic Compatibility (EMC) testing using a source frequency locked to the analyzer. Ideally, this
sweep would include EMC bands C and D and thus sweep from 30 to 1000 MHz. Ideally, the
analysis bandwidth would be 120 kHz at −6 dB, and the spacing of the points would be half of
this (60 kHz). With a start frequency of 30 MHz and a stop frequency of 1000.2 MHz and a
total of 16168 points, the spacing of points is ideal. The detector used was the Peak detector.
The accuracy of frequency readout of all the points tested in this span was with ±0.0032% of
the span. A perfect analyzer with this many points would have an accuracy of ±0.0031% of
span. Thus, even with this large number of display points, the errors in excess of the bucket
quantization limitation were negligible.
Range
Span ≥ 10 Hz, swept1 μs to 500 ms
Span = 0 Hz or FFT−150 ms to +500 ms
Resolution0.1 μs
a. Delayed trigger is available with line, video, RF burst and external triggers.
DescriptionSpecificationsSupplemental Information
TriggersAdditional information on some of the triggers and
gate sources
Video
Independent of Display Scaling and Reference
Level
Minimum settable level−170 dBmUseful range limited by noise
Maximum usable level
Highest allowed mixer level
a
+ 2 dB (nominal)
Detector and Sweep Type
relationships
Sweep Type = Swept
Detector = Normal, Peak,
Sample or Negative Peak
Triggers on the signal before detection, which is
similar to the displayed signal
Detector = AverageTriggers on the signal before detection, but with a
single-pole filter added to give similar smoothing to
that of the average detector
Chapter 1 23
Agilent EXA Signal Analyzer
Frequency and Time
DescriptionSpecificationsSupplemental Information
Sweep Type = FFTTriggers on the signal envelope in a bandwidth
wider than the FFT width
RF Burst
Level Range−50 to −10 dBm plus attenuation (nominal)
Bandwidth (−10 dB)
Most cases 16 MHz (nominal)
Sweep Type = FFT;
FFT Width = 25 MHz;
Span ≥ 8 MHz
Frequency LimitationsIf the start or center frequency is too close to zero,
External Triggers
30 MHz (nominal)
LO feedthrough can degrade or prevent triggering.
How close is too close depends on the bandwidth.
See “Inputs/Outputs” on page 71
a. The highest allowed mixer level depends on the attenuation and IF Gain. It is nominally
–10 dBm + input attenuation for Preamp Off and IF Gain = Low.
24Chapter 1
Agilent EXA Signal Analyzer
Frequency and Time
DescriptionSpecificationsSupplemental Information
Gated Sweep
Gate Methods Gated LO
Gated Video
Gated FFT
Span RangeAny span
Gate Delay Range0 to 100.0 s
Gate Delay Settability4 digits, ≥ 100 ns
Gate Delay Jitter33.3 ns p-p (nominal)
Gate Length Range
Except Method = FFT
Gated Frequency and
Amplitude Errors
Gate SourcesExternal 1
100.0 ns to 5.0 s
External 2
Line
RF Burst
Periodic
Nominally no additional error for gated
measurements when the Gate Delay is greater
than the MIN FAST setting
Pos or neg edge triggered
Chapter 1 25
Agilent EXA Signal Analyzer
Frequency and Time
Nominal Measurement Time vs. Span [Plot]
DescriptionSpecificationsSupplemental Information
Number of Frequency Display Trace
Points (buckets)
Factory preset1001
Range1 to 40,001Zero and non-zero spans
26Chapter 1
Agilent EXA Signal Analyzer
Frequency and Time
DescriptionSpecificationsSupplemental
Information
Resolution Bandwidth (RBW)
Range (−3.01 dB bandwidth)1 Hz to 8 MHz
Bandwidths above 3 MHz are 4, 5, 6,
and 8 MHz.
Bandwidths 1 Hz to 3 MHz are spaced
at 10% spacing using the E24 series
(24 per decade): 1.0, 1.1, 1.2, 1.3, 1.5,
a. The noise marker, band power marker , channel power and ACP all comp ute their results using
the power bandwidth of the RBW used for the measurement. Power bandwidth accuracy is
the power uncertainty in the results of these measurements due only to bandwidth-related
errors. (The analyzer knows this power bandwidth for each RBW with greater accuracy than
the RBW width itself, and can therefore achieve lower errors.) The warranted specifications
shown apply to the Gaussian RBW filters used in swept and zero span analysis. There are four
different kinds of filters used in the spectrum analyzer: Swept Gaussian, Swept Flattop, FFT
Gaussian and FFT Flattop. While the warranted performance only applies to the swept Gaussian filters, because only they are kept under statistical process control, the other filters nominally have the same performance.
Chapter 1 27
Agilent EXA Signal Analyzer
Frequency and Time
b. Resolution Bandwidth Accuracy can be observed at slower sweep times than auto-coupled
conditions. Normal sweep rates cause the shape of the RBW filter displayed on the analyzer
screen to widen by nominally 6%. This widening declines to 0.6% nominal when the Swp
Time Rules key is set to Accuracy instead of Normal. The true bandwidth, which determines
the response to impulsive signals and noise-like signals, is not affected by the sweep rate.
DescriptionSpecificationSupplemental information
Analysis Bandwidth
Standard10 MHz
With Option B2525 MHz
a
a. Analysis bandwidth is the instantaneous bandwidth available about a center frequency over
which the input signal can be digitized for further analysis or processing in the time, frequency, or modulation domain.
DescriptionSpecificationsSupplemental Information
Video Bandwidth (VBW)
RangeSame as Resolution Bandwidth
range plus wide-open VBW
(labeled 50 MHz)
Accuracy±6% (nominal)
in swept mode and zero span
a. For FFT processing, the selected VBW is used to determine a number of averages for FFT
results. That number is chosen to give roughly equivalent display smoothing to VBW filtering
in a swept measurement. For example, if VBW=0.1 × RBW, four FFTs are averaged to generate one result.
a
28Chapter 1
Agilent EXA Signal Analyzer
Amplitude Accuracy and Range
Amplitude Accuracy and Range
DescriptionSpecificationsSupplemental
Information
Measurement RangeDisplayed Average Noise Level to +23 dBm
Preamp OnDisplayed Average Noise Level to +23 dBmOption P03
Input Attenuation Range0 to 60 dB, in 10 dB stepsStandard
Input Attenuation Range0 to 60 dB, in 2 dB stepsWith Option FSA
DescriptionSpecificationsSupplemental Information
Maximum Safe Input LevelApplies with or without preamp
(Option P03)
Average Total Power+30 dBm (1 W)
Peak Pulse Power
<10 μs pulse width,
<1% duty cycle
input attenuation ≥ 30 dB
DC volts
DC Coupled±0.2 Vdc
AC Coupled±70 Vdc
DescriptionSpecificationsSupplemental Information
Display Range
Log ScaleTen divisions displayed;
+50 dBm (100 W)
0.1 to 1.0 dB/division in 0.1 dB steps, and
1 to 20 dB/division in 1 dB steps
Linear ScaleTe n divisions
Chapter 1 29
Agilent EXA Signal Analyzer
Amplitude Accuracy and Range
DescriptionSpecificationsSupplemental Information
Marker Readout
Log units resolution
Average Off, on-screen0.01 dB
Average On or remote0.001 dB
Linear units resolution≤1% of signal level (nominal)
a
a. Reference level and off-screen performance: The reference level (RL) behavior differs from
some earlier analyzers in a way that makes this analyzer more flexible. In other analyzers, the
RL controlled how the measurement was performed as well as how it was displayed. Because
the logarithmic amplifier in these analyzers had both range and resolution limitations, this
behavior was necessary for optimum measurement accuracy. The logarithmic amplifier in this
signal analyzer, however, is implemented digitally such that the range and resolution greatly
exceed other instrument limitations. Because of this, the signal analyzer can make measurements largely independent of the setting of the RL without compromising accuracy. Because
the RL becomes a display function, not a measurement function, a marker can read out results
that are off-screen, either above or below , without any change in accuracy. The only exception
to the independence of RL and the way in which the measurement is performed is in the input
attenuation setting: When the input attenuation is set to auto, the rules for the determination of
the input attenuation include dependence on the reference level. Because the input attenuation
setting controls the tradeoff between large signal behaviors (third-order intermodulation and
compression) and small signal effects (noise), the measurement results can change with RL
changes when the input attenuation is set to auto.
30Chapter 1
Agilent EXA Signal Analyzer
Amplitude Accuracy and Range
Frequency Response
DescriptionSpecificationsSupplemental Information
Frequency ResponseRefer to the footnote for
Band Overlaps on page 15.
Maximum error relative to
reference condition (50 MHz)
Mechanical attenuator only
Swept operation
b
a
Attenuation 10 dB20 to 30 °C5 to 50 °C
th
Percentile (≈2σ)
95
9 kHz to 10 MHz±0.8 dB±1.0 dB±0.40 dB
10 MHz to 3.6 GHz±0.6 dB±0.65 dB±0.21 dB
3.5 to 7 GHz
7 to 13.6 GHz
c d
c d
13.5 to 22.0 GHz
22.0 to 26.5 GHz
c d
c d
±2.0 dB±3.0 dB
±2.5 dB±3.2 dB
±3.0 dB±3.7 dB
±3.2 dB±4.2 dB
a. See the Electronic Attenuator (Option EA3) chapter for Frequency Response using the elec-
tronic attenuator.
b. For Sweep Type = FFT, add the RF flatness errors of this table to the IF Frequency Response
errors. An additional error source, the error in switching between swept and FFT sweep types,
is nominally ±0.01 dB and is included within the “Absolute Amplitude Error” specifications.
c. Specifications for frequencies > 3.5 GHz apply for sweep rates ≤100 MHz/ms.
d. Preselector centering applied.
Chapter 1 31
Agilent EXA Signal Analyzer
Amplitude Accuracy and Range
DescriptionSpecificationsSupplemental Information
IF Frequency Response
Demodulation and FFT
response relative to the
center frequency
Freq (GHz)
≤ 3.6≤ 100.40 dB0.12 dB0.100.03 dB
3.6 to 26.5≤ 100.25 dB≤ 3.610 to ≤ 250.45 dB0.12 dB0.050.04 dB
3.6 to 26.510 to ≤ 250.80 dB
a
FFT Width
(MHz)
b
Max Errorc
(Exceptions
th
95
Percentile
Midwidth
d
)
Error
Slope
(dB/MHz)
e
Rms
(nominal)
a. The IF frequency response includes effects due to RF circuits such as input filters, that are a
function of RF frequency, in addition to the IF pass-band effects.
b. This column applies to the instantaneous analysis bandwidth in use. The range available
depends on the hardware options and the Mode. The Spectrum analyzer Mode does not allow
all bandwidths. The I/Q Analyzer is an example of a mode that does allow all bandwidths.
c. The maximum error at an offset (f) from the center of the FFT width is given by the expres-
± [Midwidth Error + (f ×Slope)], but never exceeds ±Max Error. Usually, the span is no
sion
larger than the FFT width in which case the center of the FFT width is the center frequency of
the analyzer . When the analyzer span is w ider than the FFT width, the span i s made up of multiple concatenated FFT results, and thus has multiple centers of FFT widths so the f in the
equation is the offset from the nearest center. These specifications include the ef fect of RF frequency response as well as IF frequency response at the worst case center frequency. Performance is nominally three times better at most center frequencies.
d. The specification does not apply for frequencies greater than 3.6 MHz from the center in FFT
widths of 7.2 to 8 MHz.
e. The “RMS” nominal performance is the standard deviatio n of th e response relative to the cen-
ter frequency, integrated across a 10 or 25 MHz span. This performance measure was
observed at a center frequency in each harmonic mixing band, which is representative of all
center frequencies; it is not the worst case frequency.
32Chapter 1
Agilent EXA Signal Analyzer
Amplitude Accuracy and Range
DescriptionSpecificationsSupplemental Information
Input Attenuation Switching UncertaintyRefer to the footnote for
Band Overlaps on page 15.
Relative to 10 dB (reference setting)
Frequency Range
50 MHz (reference frequency)±0.20 dB±0.08 dB (typical)
Attenuation > 2 dB, preamp off
9 kHz to 3.6 GHz±0.3 dB (nominal)
3.5 to 7.0 GHz±0.5 dB (nominal)
7.0 to 13.6 GHz±0.7 dB (nominal)
13.5 to 26.5 GHz±0.7 dB (nominal)
Chapter 1 33
Agilent EXA Signal Analyzer
Amplitude Accuracy and Range
DescriptionSpecificationsSupplemental Information
Absolute Amplitude Accuracy
At 50 MHz
a
20 to 30°C
5 to 50°C
At all frequencies
a
20 to 30°C
5 to 50°C
th
95
Percentile Absolute
Amplitude Accuracy
±0.40 dB
±0.43 dB
±(0.4 dB + frequency response)
±(0.43 dB + frequency response)
b
±0.15 dB (95th percentile)
±0.27 dB
Wide range of signal levels,
RBWs, RLs, etc.
0.01 to 3.6 GHz, Atten = 10 dB
Amplitude Reference Accuracy±0.05 dB (nominal)
Preamp On
Option P03
c
±(0.39 dB + frequency
response) (nominal)
a. Absolute amplitude accuracy is the total of all amplitude measurement errors, and applies over
the following subset of settings and conditions: 1 Hz ≤ RBW ≤1 MHz; Input signal −10 to
−50 dBm; Input attenuation 10 dB; span <5 MHz (nominal additional error for span ≥ 5 MHz
is 0.02 dB); all settings auto-coupled except Swp Time Rules = Accuracy; combinations of
low signal level and wide RBW use VBW ≤30 kHz to reduce noise.
This absolute amplitude accuracy specification includes the sum of the following individual
specifications under the conditions listed above: Scale Fidelity, Reference Level Accuracy,
Display Scale Switching Uncertainty, Resolution Bandwidth Switching Uncertainty, 50 MHz
Amplitude Reference Accuracy , and the accuracy with which the instrument aligns its internal
gains to the 50 MHz Amplitude Reference.
34Chapter 1
Agilent EXA Signal Analyzer
Amplitude Accuracy and Range
b. Absolute Ampl itude Accuracy for a wide range of signal and measurement settings, covers the
95th percentile proportion with 95% confidence. Here are the details of what is covered and
how the computation is made:
The wide range of conditions of RBW, signal level, VBW, reference level and display scale
are discussed in footnote a. There are 44 quasi-random combinations used, tested at a 50 MHz
signal frequency. We compute the 95th p ercentile prop ortion with 95 % confidence for this set
observed over a statistically significant number of instruments. Also, the frequency response
relative to the 50 MHz response is characterized by varying the signal across a large number
of quasi-random verification frequencies that are chosen to not correspond with the frequency
response adjustment frequencies. We again compute the 95th percentile proportion with 95%
confidence for this set observed over a statistically significant number of instruments. We also
compute the 95th percentile accuracy of tracing the calibration of the 50 MHz absolute amplitude accuracy to a national standards organization. We also compute the 95th percentile accuracy of tracing the calibration of the relative frequency response to a national standards
organization. We take the root-sum-square of these four independent Gaussian parameters. To
that rss we add the environmental effects of temperature variations across the 20 to 30 °C
range. These computations and measurements are made with the mechanical attenuator only
in circuit, set to the reference state of 10 dB.
A similar process is used for computing the result when using the electronic attenuator under a
wide range of settings: all even settings from 4 through 24 dB inclusive, with the mechanical
attenuator set to 10 dB. Then the worse of the two computed 95th percentile results (they were
very close) is shown.
c. Same settings as footnote a, except that the signal level at the preamp input is −40 to
−80 dBm. T otal power at preamp (dBm) = total power at input (dBm) minus input attenuation
(dB). This specification applies for signal frequencies above 100 kHz.
Chapter 1 35
Agilent EXA Signal Analyzer
Amplitude Accuracy and Range
DescriptionSpecificationsSupplemental Information
RF Input VSWR
at tuned frequency, DC Coupled
Nominal
a
10 dB attenuation, 50 MHz1.07:1
Input Attenuation
Frequency0 dB≥10 dB
10 MHz to 3.6 GHz< 2.2:1See nominal VSWR plots
3.6 to 26.5 GHzSee nominal VSWR plots
Internal 50 MHz calibrator is OnOpen input
Alignments runningOpen input
a. The nominal SWR stated is the worst case RF frequency in three representative instruments.
36Chapter 1
Nominal VSWR [Plot]
Agilent EXA Signal Analyzer
Amplitude Accuracy and Range
Chapter 1 37
Agilent EXA Signal Analyzer
Amplitude Accuracy and Range
DescriptionSpecificationsSupplemental
Information
Resolution Bandwidth Switching Uncertainty
relative to reference BW of 30 kHz
1.0 Hz to 3 MHz RBW±0.10 dB
Manually selected wide RBWs:
4, 5, 6, 8 MHz
DescriptionSpecificationsSupplemental
Reference Level
Range
Log Units −170 to +23 dBm, in 0.01 dB steps
Linear Units 707 pV to 3.16 V, with 0.01 dB resolution (0.11%)
Accuracy
a
b
0 dB
±1.0 dB
Information
a. Reference level and off-screen performance: The reference level (RL) behavior differs from
some earlier analyzers in a way that makes this analyzer more flexible. In other analyzers, the
RL controlled how the measurement was performed as well as how it was displayed. Because
the logarithmic amplifier in these analyzers had both range and resolution limitations, this
behavior was necessary for optimum measurement accuracy. The logarithmic amplifier in this
signal analyzer, however, is implemented digitally such that the range and resolution greatly
exceed other instrument limitations. Because of this, the analyzer can make measurements
largely independent of the setting of the RL without compromising accuracy. Because the RL
becomes a display function, not a measurement function, a marker can read out results that
are off-screen, either above or below, without any change in accuracy. The only exception to
the independence of RL and the way in which the measurement is performed is in the input
attenuation setting: When the input attenuation is set to auto, the rules for the determination of
the input attenuation include dependence on the reference level. Because the input attenuation
setting controls the tradeoff between large signal behaviors (third-order intermodulation and
compression) and small signal effects (noise), the measurement results can change with RL
changes when the input attenuation is set to auto.
b. Because reference level affects only the display, not the measurement, it causes no additional
error in measurement results from trace data or markers.
38Chapter 1
Agilent EXA Signal Analyzer
Amplitude Accuracy and Range
DescriptionSpecificationsSupplemental Information
Display Scale Switching Uncertainty
Switching between Linear and Log
Log Scale Switching
0 dB
0 dB
a
a
a. Because Log/Lin and Log Scale Switching affect only the display, not the measurement, they
cause no additional error in measurement results from trace data or markers.
DescriptionSpecificationsSupplemental Information
Display Scale Fidelity
Log-Linear Fidelity (relative to the
reference condition of −25 dBm input
through the 10 dB attenuation, or
−35 dBm at the input mixer)
Input mixer level
−80 dBm ≤ ML ≤−10 dBm±0.15 dB
ML < −80 dBm±0.25 dB
abc
d
Linearity
Relative Fidelity
e
Applies for mixer leveld range from −10
to −80 dBm, mechanical attenuator only,
preamp off, and dither on.
Sum of the following terms:
high level term
Up to ±0.045 dB
instability termUp to ±0.018 dB
slope term
prefilter term
From equation
Up to ±0.005 dB
g
f
h
Chapter 1 39
Agilent EXA Signal Analyzer
3
σ
320dB()110
SN⁄3dB+()20dB⁄()–
+〈〉log=
Amplitude Accuracy and Range
a. Supplemental information: The amplitude detection linearity specification applies at all levels
below −10 dBm at the input mixer; however, noise will reduce the accur acy of low level measurements. The amplitude error due to noise is determined by the signal-to-noise ratio, S/N. If
the S/N is large (20 dB or better), the amplitude error due to noise can be estimated from the
equation below, given for the 3-sigma (three standard deviations) level.
The errors due to S/N ratio can be further reduced by averaging results. For large S/N (20 dB
or better), the 3-sigma level can be reduced proportional to the square root of the number of
averages taken.
b. The scale fidelity is warranted with ADC dither set to On. Dither increases the noise level by
nominally only 0.24 dB for the most sensitive case (preamp Off, best DANL frequencies).
With dither Off, scale fidelity for low level si gnals, around −60 dBm or lower, will nominally
degrade by 0.2 dB.
c. Reference level and off-screen performance: The reference level (RL) behavior differs from
some earlier analyzers in a way that makes this analyzer more flexible. In other analyzers, the
RL controlled how the measurement was performed as well as how it was displayed. Because
the logarithmic amplifier in these analyzers had both range and resolution limitations, this
behavior was necessary for optimum measurement accuracy. The logarithmic amplifier in this
signal analyzer, however, is implemented digitally such that the range and resolution greatly
exceed other instrument limitations. Because of this, the analyzer can make measurements
largely independent of the setting of the RL without compromising accuracy. Because the RL
becomes a display function, not a measurement function, a marker can read out results that are
off-screen, either above or below, without any change in accuracy. The only exception to the
independence of RL and the way in which the measurement is performed is in the input attenuator setting: When the input attenuator is set to auto, the rules for the determination of the
input attenuation include dependence on the reference level. Because the input attenuation
setting controls the tradeoff between large signal behaviors (third-order intermodulation and
compression) and small signal effects (noise), the measurement results can change with RL
changes when the input attenuation is set to auto.
d. Mixer level = Input Level − Input Attenuator
e. The relative fidelity is the error in the measured difference between two signal levels. It is so
small in many cases that it cannot be verified without being dominated by measurement
uncertainty of the verification. Because of this verification difficulty, this specification gives
nominal performance, based on numbers that are as conservatively determined as those used
in warranted specifications. We will consider one example of the use of the error equation to
compute the nominal performance.
Example: the accuracy of the relative level of a sideband around −60 dBm, with a carrier at −5
dBm, using attenuator = 10 dB, RBW = 3 kHz, evaluated with swept analysis. The high level
term is evaluated with P1 = −15 dBm and P2 = −70 dBm at the mixer. This gives a maximum
error within ±0.025 dB. The instability term is ±0.018 dB. The slope term evaluates to ±0.050
dB. The prefilter term applies and evaluates to the limit of ±0.005 dB. The sum of all these
terms is ±0.098 dB.
f. Errors at high mixer levels will nominally be well within the range of ±0.045 dB × {exp[(P1 −
x
Pref)/(8.69 dB)] − exp[(P2 − Pref)/(8.69 dB)]} (exp is the natural exponent function, e
). In
this expression, P1 and P2 are the powers of the two signals, in decibel units, whose relative
power is being measured. Pref is −10 dBm (−10 dBm is the highest power for which linearity
is specified). All these levels are referred to the mixer level.
40Chapter 1
Agilent EXA Signal Analyzer
Amplitude Accuracy and Range
g. Slope error will nominally be well within the range of ±0.0009 × (P1 − P2). P1 and P2 are
defined in footnote f.
h. A small additional error is possible. In FFT sweeps, this error is possible for spans under 4.01
kHz. For non-FFT measurements, it is possible for RBWs of 3.9 kHz or less. The error is well
within the range of ±0.0021 × (P1 - P2) subject to a maximum of ±0.005 dB. (The maximum
dominates for all but very small differences.)P1 and P2 are defined in footnote f.
DescriptionSpecificationsSupplemental Information
Available DetectorsNormal, Peak, Sample,
Negative Peak, A verage
A verage detector works on RMS,
Voltage and Logarithmic scales
Chapter 1 41
Agilent EXA Signal Analyzer
Dynamic Range
Dynamic Range
Gain Compression
DescriptionSpecificationsSupplemental Information
1 dB Gain Compression Point
(Two-tone)
20 MHz to 26.5 GHz+9 dBm (nominal)
Clipping (ADC Over Range)
Any signal offset−10 dBm
IF Gain set to Low
Signal offset >5 times IF prefilter bandwidth
a. Large signals, even at frequencies not shown on the screen, can cause the analyzer to incor-
rectly measure on-screen signals because of two-tone gain compression. This specification
tells how large an interfering signal must be in order to cause a 1 dB change in an on-screen
signal.
b. Specified at 1 kHz RBW with 100 kHz tone spacing. The compression point will nominally
equal the specification for tone spacing greater than 5 times the prefilter bandwidth. At
smaller spacings, ADC clipping may occur at a level lower than the 1 dB compression point.
42Chapter 1
Agilent EXA Signal Analyzer
Dynamic Range
c. Reference level and off-screen performance: The reference level (RL) behavior differs from
some earlier analyzers in a way that makes this analyzer more flexible. In other analyzers, the
RL controlled how the measurement was performed as well as how it was displayed. Because
the logarithmic amplifier in these analyzers had both range and resolution limitations, this
behavior was necessary for optimum measurement accuracy. The logarithmic amplifier in this
signal analyzer, however, is implemented digitally such that the range and resolution greatly
exceed other instrument limitations. Because of this, the analyzer can make measurements
largely independent of the setting of the RL without compromising accuracy. Because the RL
becomes a display function, not a measurement function, a marker can read out results that are
off-screen, either above or below, without any change in accuracy. The only exception to the
independence of RL and the way in which the measurement is performed is in the input attenuation setting: When the input attenuation is set to auto, the rules for the determination of the
input attenuation include dependence on the reference level. Because the input attenuation
setting controls the tradeoff between large signal behaviors (third-order intermodulation, compression, and display scale fidelity) and small signal effects (noise), the measurement results
can change with RL changes when the input attenuation is set to auto.
d. The ADC clipping level declines at low frequencies (below 50 MHz) when the LO feed
through (the signal that appears at 0 Hz) is within 5 times the prefilter bandwidth (see table)
and must be handled by the ADC. For example, with a 300 kHz RBW and prefilter bandwidth
at 966 kHz, the clipping level reduces for signal frequencies below 4.83 MHz. For signal frequencies below 2.5 times the prefilter bandwidth, there will be additional reduction due to the
presence of the image signal (the signal that appears at the negative of the input signal frequency) at the ADC.
Chapter 1 43
Agilent EXA Signal Analyzer
Dynamic Range
Displayed Average Noise Level
DescriptionSpecificationsSupplemental Information
Displayed Average Noise Level
(DANL)
Option 503, 507, 513, 526
1 to 10 MHz
10 MHz to 2.1 GHz−148 dBm−146 dBm−150 dBm
2.1 GHz to 3.6 GHz−147 dBm−145 dBm−148 dBm
Option 507,513, 526
3.6 GHz to 7 GHz−147 dBm−145 dBm−149 dBm
Option 513, 526
7.0 GHz to 13.6 GHz−143 dBm−141 dBm−147 dBm
Option 526
13.5 GHz to 17.1 GHz−137 dBm−134 dBm−142 dBm
a
b
Input terminated Sample or
Average detector
Averaging type = Log
0 dB input attenuation
IF Gain = High
1 Hz Resolution Bandwidth
20 to 30°C5 to 50°CTypical
−147 dBm−145 dBm−149 dBm
Refer to the footnote for
Band Overlaps on page 15.
17.0 GHz to 20.0 GHz−137 dBm−134 dBm−142 dBm
20.0 GHz to 26.5 GHz−134 dBm−130 dBm−140 dBm
Additional DANL, IF Gain=Low
c
−160.5 dBm (nominal)
a. DANL for zero span and swept is normalized in two ways and for two reasons. DANL is mea-
sured in a 1 kHz RBW and normalized to the narrowest available RBW, because the noise figure does not depend on RBW and 1 kHz measurements are faster. The second normalization
is that DANL is measured with 10 dB input attenuation and normalized to the 0 dB input
attenuation case, because that makes DANL and third order intermodulation test conditions
congruent, allowing accurate dynamic range estimation for the analyzer.
b. DANL below 10 MHz is dominated by phase noise around the LO feedthrough signal. Speci-
fications apply with the best setting of the Phase Noise Optimization control, which is to
choose the “Best Phase Noise at offset < 20 kHz” for frequencies below 25 kHz, and “Best
Phase Noise at offset > 30 kHz” for frequencies above 25 kHz. The difference in sensitivity
with Phase Noise Optimization changes is about 10 dB at 10 and 100 kHz, d eclining to under
1 dB for signals below 400 Hz, above 800 kHz, and near 25 kHz.
44Chapter 1
Agilent EXA Signal Analyzer
Dynamic Range
c. Setting the IF Gain to Low is often desirable in order to allow higher power into the mixer
without overload, better compression and better third-order intermodulation. When the Swept
IF Gain is set to Low, either by auto coupling or manual coupling, there is noise added above
that specified in this table for the IF Gain = High case. That excess noise appears as an additional noise at the input mixer . This level has sub-decibel dependen ce on center frequency. To
find the total displayed average noise at the mixer for Swept IF Gain = Low, sum the powers
of the DANL for IF Gain = High with this additional DANL. To do that summation, compute
DANLtotal = 10 × log (10^(DANLhigh/10) + 10^(AdditionalDANL / 10)). In FFT sweeps,
the same behavior occurs, except that FFT IF Gain can be set to autorange, where it varies
with the input signal level, in addition to forced High and Low settings.
Chapter 1 45
Agilent EXA Signal Analyzer
Dynamic Range
Spurious Responses
DescriptionSpecificationsSupplemental
Information
Spurious Responses
Mixer Level
a
Response
Preamp Off
b
Refer to the footnote for
Band Overlaps on
page 15.
Residual Responses
c
200 kHz to 8.4 GHz (swept)
Zero span or FFT or other frequencies
N/A
−100 dBm
−100 dBm (nominal)
Image Responses
Tuned Freq. (f)Excitation
Freq.
10 MHz to 26.5 GHzf+45 MHz−10 dBm−75 dBc−99 dBc (typical)
10 MHz to 3.6 GHzf+10245 MHz−10 dBm−80 dBc−103 dBc (typical)
10 MHz to 3.6 GHzf+645 MHz−10 dBm−80 dBc−107 dBc (typical)
3.5 GHz to 13.6 GHzf+645 MHz−10 dBm−75 dBc−87 dBc (typical
13.5 GHz to 17.1 GHzf+645 MHz−10 dBm−71 dBc−85 dBc (typical)
17.0 GHz to 22 GHzf+645 MHz−10 dBm−68 dBc−82 dBc (typical)
22 GHz to 26.5 GHzf+645 MHz−10 dBm−66 dBc−78 dBc (typical)
LO Related Spurious Responses
f > 600 MHz from carrier
10 MHz to 3.6 GHz
−10 dBm
−60 dBc
−90 dBc (typical)
Other Spurious Responses
First RF Order
d
f ≥ 10 MHz from carrier
−10 dBm−68 dBcIncludes other LO
spurious, IF feedthrough,
LO harmonic mixing
responses
Higher RF Order
e
f ≥ 10 MHz from carrier
−40 dBm−80 dBcIncludes higher order
mixer responses
Sidebands, offset from CW signal
≤ 200 Hz
200 Hz to 3 kHz
−60 dBc
−68 dBc
f
(nominal)
f
(nominal)
3 kHz to 30 kHz−68 dBc (nominal)
30 kHz to 10 MHz−80 dBc (nominal)
46Chapter 1
Agilent EXA Signal Analyzer
Dynamic Range
a. Mixer Level = Input Level − Input Attenuation.
b. The spurious response specifications only apply with the preamp turned off. When the preamp
is turned on, performance is nominally the same as long as the mixer level is interpreted to be:
Mixer Level = Input Level − Input Attenuation − Preamp Gain
c. Input terminated, 0 dB input attenuation.
d. With first RF order spurious products, the indicated frequency will change at the same rate as
the input, with higher order, the indicated frequency will change at a rate faster than the input.
e. RBW=100 Hz. With higher RF order spurious responses, the observed frequency will change
at a rate faster than the input frequency.
f. Nominally −40 dBc under large magnetic (0.38 Gauss rms) or vibrational (0.21 g rms) envi-
ronmental stimuli.
Chapter 1 47
Agilent EXA Signal Analyzer
Dynamic Range
Second Harmonic Distortion
DescriptionSpecificationsSupplemental
Information
Second Harmonic Distortion
Mixer Level
a
SHIb (nominal)
Source Frequency
10 MHz to 1.8 GHz−15 dBm+45 dBm
1.75 to 7 GHz−15 dBm+ 65 dBm
7 GHz to 11 GHz−15 dBm+55 dBm
11 to 13.25 GHz−15 dBm+50 dBm
a. Mixer level = Input Level − Input Attenuation
b. SHI = second harmonic intercept. The SHI is given by the mixer power in dBm minus the sec-
ond harmonic distortion level relative to the mixer tone in dBc.
Third Order Intermodulation Distortion
DescriptionSpecificationsSupplemental Information
Third Order
Intermodulation Distortion
Tone separation > 5 times IF
Prefilter Bandwidth
Verification conditions
a
b
Refer to the footnote for
Band Overlaps on page 15.
20 to 30°C
Intercept
c
Extrapolated
Distortion
d
Intercept (typical)
100 to 400 MHz+10 dBm−80 dBc+14 dBm
400 MHz to 1.7 GHz+11 dBm−82 dBc+15 dBm
1.7 to 3.6 GHz+13 dBm−86 dBc+17 dBm
3.6 to 5.1 GHz +11 dBm−82 dBc+17 dBm
5.1 to 7 GHz +13 dBm−86 dBc+17 dBm
7 to 13.6 GHz +11 dBm−82 dBc+15 dBm
13.6 to 26.5 GHz +9 dBm−78 dBc+14 dBm
5 to 50°C
10 to 100 MHz
100 to 400 MHz+9 dBm−78 dBc
48Chapter 1
Agilent EXA Signal Analyzer
Dynamic Range
DescriptionSpecificationsSupplemental Information
400 MHz to 1.7 GHz+10 dBm−80 dBc
1.7 to 3.6 GHz+12 dBm−84 dBc
3.6 to 5.1 GHz+10 dBm−80 dBc
5.1 to 7 GHz+12 dBm−86 dBc
7 to 13.6 GHz+10 dBm−80 dBc
13.6 to 26.5 GHz+7 dBm−74 dBc
a. See the IF Prefilter Bandwidth table in the Gain Compression specifications on page 42.
When the tone separation condition is met, the effect on TOI of the setting of IF Gain is negli-
gible. TOI is verified with IF Gain set to its best case condition, which is IF Gain = Low.
b. TOI is verified with two tones, each at −18 dBm at the mixer, spaced by 100 kHz.
c. TOI = third order intercept. The TOI is given by the mixer tone level (in dBm) minus (distor-
tion/2) where distortion is the relative level of the distortion tones in dBc.
d. The distortion shown is computed from the warranted intercept specifications, based on two
tones at −30 dBm each, instead of being measured directly. The choice of −30 dBm is based
on historic and continuing industry practice.
Chapter 1 49
Agilent EXA Signal Analyzer
Dynamic Range
Nominal Dynamic Range at 1 GHz [Plot]
50Chapter 1
Nominal Dynamic Range Bands 1-4 [Plot]
Agilent EXA Signal Analyzer
Dynamic Range
Chapter 1 51
Agilent EXA Signal Analyzer
Dynamic Range
Nominal Dynamic Range vs. Offset Frequency vs. RBW [Plot]
52Chapter 1
Agilent EXA Signal Analyzer
Dynamic Range
Phase Noise
DescriptionSpecificationsSupplemental Information
Phase Noise
Noise Sidebands
b
a
20 to 30°C5 to 50°C
Center Frequency = 1 GHz
Best-case Optimization
Internal Reference
a. The nominal performance of the phase noise at frequencies above the frequency at which the
specifications apply (1 GHz) depends on the band and the offset. For low offset frequencies,
offsets well under 100 Hz, the phase noise increases by 20 × log(f). For mid-offset frequen-
cies, such as [10 kHz, band 0 phase noise increases as 20 × log[(f + 5.1225)/6.1225]. For
mid-offset frequencies in other bands, phase noise changes as 20 × log[(f + 0.3225)/6.1225],
except if in this expression should never be lower than 5.8. For wide offset frequencies, [off-
sets well above 100 kHz], phase noise increases as 20 × log(N). N is the LO Multiple as
shown on page 15; f is in GHz units in all these relationships; all increases are in units of
decibels.
b. Noise sidebands for lower offset frequencies, for example, 10 kHz, as apply with the phase
noise optimization (
Pn Noise Opt) set to Best Close-in φ Noise. Noise sidebands for higher
offset frequencies, for example, 1 MHz, as shown apply with the phase noise optimization set
Best Wid-offset φ Noise.
to
c. Specifications are given with the internal precision frequency reference. The phase noise at
offsets below 100 Hz is impacted or dominated by noise from the reference. Thus, perfor-
mance with external references will not follow the curves and specifications. The internal 10
MHz reference phase noise is about –120 dBc/Hz at 10 Hz offset; external references with
poorer phase noise than this will cause poorer performance than shown.
Chapter 1 53
Agilent EXA Signal Analyzer
Dynamic Range
Nominal Phase Noise of Different LO Optimizations
54Chapter 1
Nominal Phase Noise at Different Center Frequencies
Agilent EXA Signal Analyzer
Dynamic Range
Chapter 1 55
Agilent EXA Signal Analyzer
Power Suite Measurements
Power Suite Measurements
DescriptionSpecificationsSupplemental Information
Channel Power
Amplitude Accuracy
Case: Radio Std = 3GPP W-CDMA, or IS-95
Absolute Power Accuracy
20 to 30 °C
Attenuation = 10 dB
±0.94 dB
Absolute Amplitude Accuracy
Power Bandwidth Accuracy
±0.27 dB (95th percentile)
a. See “Absolute Amplitude Accuracy” on page 34.
b. See “Frequency and Time” on page 15.
c. Expressed in dB.
DescriptionSpecificationsSupplemental Information
Occupied Bandwidth
Frequency Accuracy±(Span/1000) (nominal)
a
+
bc
56Chapter 1
Agilent EXA Signal Analyzer
Power Suite Measurements
DescriptionSpecificationsSupplemental Information
Adjacent Channel Power (ACP)
Case: Radio Std = None
Accuracy of ACP Ratio (dBc)
Accuracy of ACP Absolute Power
(dBm or dBm/Hz)
Accuracy of Carrier Power (dBm), or
Carrier Power PSD (dBm/Hz)
Passbandwidth
e
Case: Radio Std = 3GPP W-CDMA
Display Scale Fidelity
Absolute Amplitude Accuracy
Power Bandwidth Accuracy
Absolute Amplitude Accuracy
Power Bandwidth Accuracy
−3 dB
(ACPR; ACLR)
a
b
+
cd
b
+
cd
f
Minimum power at RF Input−36 dBm (nominal)
ACPR Accuracy
g
RadioOffset Freq
RRC weighted, 3.84 MHz noise bandwidth,
method = IBW or Fast
h
MS (UE)5 MHz±0.22 dB At ACPR range of −30 to −36 dBc with
optimum mixer level
i
MS (UE)10 MHz±0.34 dBAt ACPR range of −40 to −46 dBc with
j
k
BTS5 MHz
±1.07 dB
optimum mixer level
h
At ACPR range of −42 to −48 dBc with
optimum mixer level
BTS10 MHz±1.00 dBAt ACPR range of −47 to −53 dBc with
j
l
BTS5 MHz±0.44 dB
optimum mixer level
At −48 dBc non-coherent ACPR
Dynamic RangeRRC weighted, 3.84 MHz noise
bandwidth
Noise
Correction
Offset
Freq
Method
Off 5 MHzFiltered
ACLR (typical)
−68 dB−8 dBm
m
Optimal ML
(Nominal)
IBW
Off5 MHzFast−67 dB−9 dBm
Off10 MHzFiltered
−74 dB−2 dBm
IBW
Chapter 1 57
Agilent EXA Signal Analyzer
x
–
Power Suite Measurements
DescriptionSpecificationsSupplemental Information
On5 MHzFiltered
IBW
On10 MHzFiltered
IBW
RRC Weighting Accuracy
White noise in Adjacent Channel
TOI-induced spectrum
rms CW error
n
−73 dB−8 dBm
−76 dB−2 dBm
0.00 dB nominal
0.001 dB nominal
0.012 dB nominal
a. The effect of scale fidelity on the ratio of two powers is called the relative scale fidelity. The
scale fidelity specified in the Amplitude section is an absolute scale fidelity with –35 dBm at
the input mixer as the reference point. The relative scale fidelity is nominally only 0.01 dB
larger than the absolute scale fidelity.
b. See Amplitude Accuracy and Range section.
c. See Frequency and Time section.
d. Expressed in decibels.
e. An ACP measurement measures the power in adjacent channels. The shape of the response
versus frequency of those adjacent channels is occasionally critical. One parameter of the
shape is its 3 dB bandwidth. When the bandwidth (called the Ref BW) of the adjacent channel
is set, it is the 3 dB bandwidth that is set. The passband response is given by the convolution
of two functions: a rectangle of width equal to Ref BW and the power response versus fre-
quency of the RBW filter used. Measurements and specifications of analog radio ACPs are
often based on defined bandwidths of measuring receivers, and these are defined by their
−6 dB widths, not their −3 dB widths. To achieve a passband whose −6 dB width is x, set the
Ref BW to be .
0.572 RBW×
f. Most versions of adjacent channel power measurements use negative numbers, in units of
dBc, to refer to the power in an adjacent channel relative to the power in a main channel, in
accordance with ITU standards. The standards for W-CDMA analysis include ACLR, a posi-
tive number represented in dB units. In order to be consistent with other kinds of ACP mea-
surements, this measurement and its specifications will use negative dBc results, and refer to
them as ACPR, instead of positive dB results referred to as ACLR. The ACLR can be deter-
mined from the ACPR reported by merely reversing the sign.
g. The accuracy of the Adjacent Channel Power Ratio will depend on the mixer drive level and
whether the distortion products from the analyzer are coherent with those in the UUT. These
specifications apply even in the worst case condition of coherent analyzer and UUT distortion
products. For ACPR levels other than those in this specifications table, the optimum mixer
drive level for accuracy is approximately −37 dBm − (ACPR/3), where the ACPR is given in
(negative) decibels.
h. The Fast method has a slight decrease in accuracy in only one case: for BTS measurements at
5 MHz offset, the accuracy degrades by ±0.01 dB relative to the accuracy shown in this table.
58Chapter 1
Agilent EXA Signal Analyzer
Power Suite Measurements
i. To meet this specified accuracy when measuring mobile station (MS) or user equipment (UE)
within 3 dB of the required −33 dBc ACPR, the mixer level (ML) must be optimized for accu-
racy. This optimum mixer level is −22 dBm, so the input attenuation must be set as close as
possible to the average input power − (−19 dBm). For example, if the average input power is
−6 dBm, set the attenuation to 16 dB. This specification applies for the normal 3.5 dB
peak-to-average ratio of a single code. Note that if the mixer level is set to optimize dynamic
range instead of accuracy, accuracy errors are nominally doubled.
j. ACPR accuracy at 10 MHz of fset is warranted when the input attenuator is set to give an aver-
age mixer level of −14 dBm.
k. In order to meet this specified accuracy, the mixer level must be optimized for accuracy when
measuring node B Base Transmission Station (BTS) within 3 dB of the required −45 dBc
ACPR. This optimum mixer level is −19 dBm, so the input attenuation must be set as close as
possible to the average input power − (−22 dBm). For example, if the average input power is
−5 dBm, set the attenuation to 14 dB. This specification applies for the normal 10 dB
peak-to-average ratio (at 0.01% probability) for Test Model 1. Note that, if the mixer level is
set to optimize dynamic range instead of accuracy, accuracy errors are nominally doubled.
l. Accuracy ca n be excellent even at low ACPR levels assuming that the user sets the mixer
level to optimize the dynamic range, and assuming that the analyzer and UUT distortions are
incoherent. When the errors from the UUT and the analyzer are incoherent, optimizing
dynamic range is equivalent to minimizing the contrib ution of analyzer noise and distortion to
accuracy, though the higher mixer level increases the display scale fidelity errors. This incoherent addition case is commonly used in the industry and can be useful for comparison of
analysis equipment, but this incoherent addition model is rarely justified. This derived accuracy specification is based on a mixer level of −14 dBm.
m..Agilent measures 100% of the signal analyzers for dynamic range in the factory production
process. This measurement requires a near-ideal signal, which is impractical for field and customer use. Because field verification is impractical, Agilent only gives a typical result. More
than 80% of prototype instruments met this “typical” specification; the factory test line limit is
set commensurate with an on-going 80% yield to this typical.
The ACPR dynamic range is verified only at 2 GHz, where Agilent has the near -perfect signal
available. The dynamic range is specified for the optimum mixer drive level, which is different in different instruments and different conditions. The test signal is a 1 DPCH signal.
The ACPR dynamic range is the observed range. This typical specification includes no measurement uncertainty.
Chapter 1 59
Agilent EXA Signal Analyzer
Power Suite Measurements
n. 3GPP requires the use of a root-raised-cosine filter in evaluating the ACLR of a device. The
accuracy of the passband shape of the filter is not specified in standards, nor is any method of
evaluating that accuracy. This footnote discusses the performance of the filter in this instru-
ment. The effect of the RRC filter and the effect of the RBW used in the measurement interact.
The analyzer compensates the shape of the RRC filter to accommodate the RBW filter. The
effectiveness of this compensation is summarized in three ways:
− White noise in Adj Ch: The compensated RRC filter nominally has no errors if the adjacent
channel has a spectrum that is flat across its width.
− TOI−induced spectrum: If the spectrum is due to third−order intermodulation, it has a dis-
tinctive shape. The computed errors of the compensated filter are −0.001 dB for the 100 kHz
RBW used for UE testing with the IBW method. It is also −0.001 dB for the 390 kHz RBW
used with the Fast method, and 0.000 dB for the 27 kHz RBW filter used for BTS testing with
the Filtered IBW method. The worst error for RBWs between these extremes is 0.05 dB for a
330 kHz RBW filter.
− rms CW error: This error is a measure of the error in measuring a CW−like
spurious component. It is evaluated by computing the root of the mean of the square of the
power error across all frequencies within the adjacent channel. The computed rms error of the
compensated filter is 0.012 dB for the 100 kHz RBW used for UE testing with the IBW
method. It is 0.034 dB for the 390 kHz RBW used with the Fast method and 0.000 dB for the
27 kHz RBW filter used for BTS testing. The worst error for RBWs between 27 kHz and
470 kHz is 0.057 dB for a 430 kHz RBW filter-like spurious component. It is evaluated by
computing the root of the mean of the square of the power error across all frequencies within
the adjacent channel. The computed rms error of the compensated filter is 0.012 dB for the
100 kHz RBW used for UE testing with the IBW method. It is 0.034 dB for the 390 kHz RBW
used with the Fast method and 0.000 dB for the 27 kHz RBW filter used for BTS testing. The
worst error for RBWs between 27 kHz and 470 kHz is 0.057 dB for a 430 kHz RBW filter.
60Chapter 1
Agilent EXA Signal Analyzer
Power Suite Measurements
DescriptionSpecificationsSupplemental Information
Case: Radio Std = IS-95 or J-STD-008
Method
ACPR Relative Accuracy
Offsets < 750 kHz
Offsets > 1.98 MHz
b
c
±0.08 dB
±0.10 dB
RBW method
a
a. The RBW method measures the power in the adjacent channels within the defined resolution
bandwidth. The noise bandwidth of the RBW filter is nominally 1.055 times the 3.01 dB
bandwidth. Therefore, the RBW method will nominally read 0.23 dB higher adjacent channel
power than would a measurement using the integration bandwidth method, because the noise
bandwidth of the integration bandwidth measurem ent is equal to that integration bandwidth.
For cmdaOne ACPR measurements using th e RBW method, the main channel is measured in
a 3 MHz RBW, which does not respond to all the power in the carrie r. Therefore, the carrier
power is compensated by the expected under-response of the filter to a full width signal, of
0.15 dB. But the adjacent channel power is not compensated for the noise bandwidth ef fect.
The reason the adjacent channel is not co mpensated is subtle. The RBW method of measuring
ACPR is very similar to the preferred method of making measurements for compliance with
FCC requirements, the source of the specifications for the cdmaOne Spur Close specifications. ACPR is a spot measurement of Spur Close, and thus is best done with the RBW
method, even though the results will disagree by 0.23 dB from the measurement made with a
rectangular passband.
b. The specified ACPR accuracy applies if the measured ACPR substantially exceeds the ana-
lyzer dynamic range at the specified offset. When this condition is not met, there are additional errors due to the addition of analyzer spectral components to UUT spectral
components. In the worst case at these offsets, the analyzer spectral components are all coherent with the UUT components; in a more typical case, one third of the analyzer spectral power
will be coherent with the distortion components in the UUT. Coherent means that the phases
of the UUT distortion components and the analyzer distortion components are in a fixed relationship, and could be perfectly in-phase. This coherence is not intuitive to many users,
because the signals themselves are usually pseudo-random; nonetheless, they can be coherent.
When the analyzer components are 100% coherent with the UUT components, the errors add
in a voltage sense. That error is a function of the signal (UUT ACPR) to noise (analyzer
ACPR dynamic range limitation) ratio, SN, in decibels.
The function is error = 20 × log(1 + 10
−SN/20
)
For example, if the UUT ACPR is −62 dB and the measurement floor is −82 dB, the SN is
20 dB and the error due to adding the analyzer distortion to that of the UUT is 0.83 dB.
Chapter 1 61
Agilent EXA Signal Analyzer
Fast ACP - Standard Deviat ion vs. Time
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50
Nominal M e asurem ent and Tr ansf er Ti m e ( log)
Standard Devi ation (dB)
5 ms10 ms20 ms
40 ms
Sweep Time = 6.2 ms
Power Suite Measurements
c. As in footnote b, the specified ACPR accuracy applies if the ACPR measured substantially
exceeds the analyzer dynamic range at the specified offset. When this condition is not met,
there are additional errors due to the addition of analyzer spectral components to UUT spectral components. Unlike the situation in footnote b, though, the spectral components from the
analyzer will be non-coherent with the components from the UUT. Therefore, the errors add
in a power sense. The error is a function of the signal (UUT ACPR) to noise (analyzer ACPR
dynamic range limitation) ratio, SN, in decibels.
−SN/10
The function is error = 10 × log(1 + 10
).
For example, if the UUT ACPR is −75 dB and the measurement floor is −85 dB, the SN ratio
is 10 dB and the error due to adding the analyzer's noise to that of the UUT is 0.41 dB.
Fast ACPR Test [Plota]
a. Observation conditions for ACP speed:
Display Off, signal is Test Model 1 with 64 DPCH, Method set to Fast. Measured with an
IBM compatible PC with a 3 GHz Pentium 4 running Windows XP Professional Version
2002. The communications medium was PCI GPIB IEEE 488.2. The Test Application Language was .NET C#. The Application Communication Layer was Agilent T&M Programmer’s Toolkit For Visual Studio (Version 1.1), Agilent I/O Libraries (Version
M.01.01.41_beta).
62Chapter 1
Agilent EXA Signal Analyzer
Power Suite Measurements
DescriptionSpecificationsSupplemental Information
Power Statistics CCDF
Histogram Resolution
a
0.01 dB
a. The Complementary Cumulative Distribution Function (CCDF) is a reformatting of a histo-
gram of the power envelope. The width of the amplitude bins used by the histogram is the histogram resolution. The resolution of the CCDF will be the same as the width of those bins.
DescriptionSpecificationsSupplemental Information
Burst Power
MethodsP ower above threshold
Power within burst width
ResultsOutput power, average
Output power, single burst
Maximum power
Minimum power within burst
Burst width
Chapter 1 63
Agilent EXA Signal Analyzer
Power Suite Measurements
Description SpecificationsSupplemental Information
Spurious EmissionsTable-driven spurious signals;
search across regions
Case: Radio Std = 3GPP W-CDMA
Dynamic Range
1 to 3.6 GHz
a
Sensitivity, absolute
93.1 dB98.4 dB (typical)
−79.4 dBm−85.4 dBm (typical)
1 to 3.6 GHz
Accuracy
Attenuation = 10 dB
Frequency Range
9 kHz to 3.6 GHz
3.5 GHz to 8.4 GHz
8.3 GHz to 13.6 GHz
±0.41 dB (95th Percentile)
±1.22 dB (95th Percentile)
±1.59 dB (95th Percentile)
a. The dynamic is specified with the mixer level at +3 dBm, where up to 1 dB of compression
can occur, degrading accuracy by 1 dB.
DescriptionSpecificationsSupplemental Information
Spectrum Emission MaskTable-driven spurious signals;
measurement near carriers
Case: Radio Std = cdma2000
Dynamic Range, relative
750 kHz offset
a b
Sensitivity, absolute
750 kHz offset
c
Accuracy
750 kHz offset
Relative
Absolute
d
e
20 to 30 °C
64Chapter 1
74.0 dB81.0 dB (typical)
−94.7 dBm−100.7 dBm (typical)
±0.11 dB
±1.05 dB
±0.34 dB (95th Percentile ≈ 2σ)
Agilent EXA Signal Analyzer
Power Suite Measurements
DescriptionSpecificationsSupplemental Information
Case: Radio Std = 3GPP W−CDMA
Dynamic Range, relative
2.515 MHz offset
Sensitivity, absolute
2.515 MHz offset
Accuracy
2.515 MHz offset
Relative
Absolute
20 to 30 °C
d
e
a d
c
76.5 dB83.9 dB (typical)
−94.7 dBm−100.7 dBm (typical)
±0.12 dB
±1.05 dB
±0.34 dB (95th Percentile ≈ 2σ)
a. The dynamic range specification is the ratio of the channel power to the power in the offset
specified. The dynamic range depends on the measurement settings, such as peak power or
integrated power. Dynamic range specifications are based on default measurement settings,
with detector set to average, and depend on the mixer level. Default measurement settings
include 30 kHz RBW.
b. This dynamic range specification applies for the optimum mixer level, which is about
−18 dBm. Mixer level is defined to be the average input power minus the input attenuation.
c. The sensitivity is specified with 0 dB input attenuation. It represents the noise limitations of
the analyzer . It is tested without an input signal. The sensitivity at this of fset is specified in the
default 30 kHz RBW, at a center frequency of 2 GHz.
d. The relative accuracy is a measure of the ratio of the power at the offset to the main channel
power. It applies for spectrum emission levels in the offsets that are well above the dynamic
range limitation.
e. The absolute accuracy of SEM measurement is the same as the absolute accuracy of the spec-
trum analyzer. See “Absolute Amplitude Accuracy” on page 34 for more information. The
numbers shown are for 0 - 3.6 GHz, with attenuation set to 10 dB.
Chapter 1 65
Agilent EXA Signal Analyzer
Options
Options
The following options and applications affect instrument specifications.
DescriptionSpecificationsSupplemental Information
Calibration Cycle1 year
DescriptionSpecificationsSupplemental Information
Temperature Range
Operating5 to 50°CStandard
Storage−40 to 65°C
Altitude3,000 meters (approx. 10,000 feet)
DescriptionSpecificationsSupplemental Information
Environmental and Military
Specifications
DescriptionSpecifications
EMCComplies with European EMC Directive 2004/108/EC
— IEC/EN 61326-1 or IEC/EN 61326-2-1
— C ISPR Pub 11 Group 1, class A
— AS/NZS CISPR 11
— ICES/NMB-001
This ISM device complies with Canadian ICES-001.
Cet appareil ISM est conforme a la norme NMB-001 du Canada.
a
Test methods are aligned with
IEC 60068-2 and levels are
similar to MIL-PRF-28800F
Class 3.
a. The N9010A/N9020A meets CISPR 11, Class B emissions limits when no USB cable/device
connections are made to the front or rear panel. The N9010A/N9020A is in full compliance
with CISPR 11, Class A emissions and is declared as such. Any information regarding the
Class B emission performance of the N9010A/N9020A is provided as a convenience to the
user and is not intended to be a regulatory declaration.
Chapter 1 67
Agilent EXA Signal Analyzer
General
Acoustic Noise Emission/Geraeuschemission
LpA <70 dB
Operator position
Normal position
Per ISO 7779
DescriptionSpecifications
SafetyComplies with European Low Voltage Directive 2006/95/EC
DescriptionSpecificationSupplemental Information
Power Requirements
Low Range
LpA <70 dB
Am Arbeitsplatz
Normaler Betrieb
Nach DIN 45635 t.19
Voltage100 to 120 V
Frequency
Serial Prefix < MY4801,
SG4801, or US4801
Serial Prefix ≥ MY4801,
SG4801, or US4801
High Range
Voltage220 to 240 V
Frequency 50/60 Hz
Power Consumption, On270 WFully loaded with options
Power Consumption, Standby20 WStandby power is not supplied to
50/60 Hz
50/60/400 Hz
frequency reference oscillator.
68Chapter 1
DescriptionSupplemental Information
Agilent EXA Signal Analyzer
General
Measurement Speed
a
Nominal
Standardw/ Option PC2
Local measurement and display update
bc
rate
Remote measurement and LAN transfer
bc
rate
11 ms (90/s)4 ms (250/s)
6 ms (167/s)5 ms (200/s)
Marker Peak Search5 ms1.5 ms
Center Frequency Tune and Transfer
22 ms20 ms
(RF)
Center Frequency Tune and Transfer
49 ms47 ms
(µW)
Measurement/Mode Switching75 ms39 ms
W-CDMA ACLR measurement time
Measurement Time vs. Span
See page 57
See page 26
a. Sweep Points = 101.
b. Factory preset, fixed center frequency, RBW = 1 MHz, span >10 MHz and ≤ 600 MHz, stop
frequency ≤ 3.6 GHz, Auto Align Off.
c. Phase Noise Optimization set to Fast Tuning, Display Off, 32 bit integer format, markers Off,
single sweep, measured with IBM compatible PC with 2.99 GHz Pentium® 4 with 2 GB
RAM running Windows® XP, Agilent I/O Libraries Suite Version 14.1, one meter GPIB
cable, National Instruments PCI-GPIB Card and NI-488.2 DLL.
DescriptionSpecificationsSupplemental Information
Display
a
Resolution1024 × 768XGA
Size213 mm (8.4 in) diagonal (nominal)
Scale
Log Scale0.1, 0.2, 0.3...1.0, 2.0, 3.0...20 dB per
division
Linear Scale10% of reference level per division
UnitsdBm, dBmV, dBm A, Watts, Volts,
Amps, dBμV, dBμA
Chapter 1 69
Agilent EXA Signal Analyzer
General
a. The LCD display is manufactured using high precision technology . However , there may be up
to six bright points (white, blue, red or green in color) that constantly appear on the LCD
screen. These points are normal in the manufacturing process and do not affect the measurement integrity of the product in any way.
DescriptionSpecificationsSupplemental Information
Data Storage
Standard
Internal TotalIntegrated 40 GB HDD15 GB available on primary partition for
applications and secondary data
Internal User6 GB available on separate partition for user
data
With Option PC2
Internal TotalRemovable 160 GB HDD126 GB available on primary partition for
applications and secondary data
Internal User9 GB available on separate partition for user
data
With Options SSDRequires Option PC2
Internal Tot alRemovable 32 GB solid
state drive
Internal User2 GB available on separate partition for user
DescriptionSpecificationsSupplemental Information
Weight
(without options)
Net16 kg (35 lbs) (nominal)
Shipping 28 kg (62 lbs) (nominal)
14 GB available on primary partition for
applications and secondary data
data
Cabinet DimensionsCabinet dimensions exclude front and rear
protrusions.
Height177 mm (7.0 in)
Width426 mm (16.8 in)
Length368 mm (14.5 in)
70Chapter 1
Agilent EXA Signal Analyzer
Inputs/Outputs
Front Panel
DescriptionSpecificationsSupplemental Information
RF Input
DescriptionSpecificationsSupplemental Information
LAN TCP/IP InterfaceRJ45 Ethertwist100BaseT (Standard)
or
1000 BaseT (with Option PC2)
Chapter 1 75
Agilent EXA Signal Analyzer
Regulatory Information
Regulatory Information
This product is designed for use in Installation Category II and Pollution Degree 2 per IEC 61010 2nd
ed, and 664 respectively.
This product has been designed and tested in accordance with accepted industry standards, and has been
supplied in a safe condition. The instruction documentation contains information and warnings which
must be followed by the user to ensure safe operation and to maintain the product in a safe condition.
The CE mark is a registered trademark of the European Community (if accompanied by
a year, it is the year when the design was proven). This product complies wi t h al l
relevant directives.
ICES/NMB-001“This ISM device complies with Canadian ICES-001.”
“Cet appareil ISM est conforme a la norme NMB du Canada.”
ISM 1-A (GRP.1
CLASS A)
This is a symbol of an Industrial Scientific and Medical Group 1 Class A product.
(CISPR 11, Clause 4)
The CSA mark is the Canadian Standards Association. This product complies with the
relevant safety requirements.
The C-Tick mark is a registered trademark of the Australian/New Zealand Spectrum
Management Agency. This product complies with the relevant EMC regulations.
This symbol indicates separate collection for electrical and electronic equipment
mandated under EU law as of August 13, 2005. All electric and electronic equipment are
required to be separated from normal waste for disposal (Reference WEEE Directive
2002/96/EC).
To return unwanted products, contact your local Agilent office, or see
http://www.agilent.com/environment/product/index.shtml for more information.
76Chapter 1
Agilent EXA Signal Analyzer
Declaration of Conformity
Declaration of Conformity
A copy of the Manufacturer’s European Declaration of Conformity for this instrument can be obtained
by contacting your local Agilent Technologies sales representative.
Chapter 1 77
Agilent EXA Signal Analyzer
Declaration of Conformity
78Chapter 1
2 Option B25 (25 MHz) - Analysis
Bandwidth
This chapter contains specifications for the Option B25 (25 MHz) Analysis Bandwidth, and for
convenience, also has specifications for the standard bandwidths of 10 MHz and below.
79
Option B25 (25 MHz) - Analysis Bandwidth
Specifications Affected by Analysis Bandwidth
Specifications Affected by Analysis Bandwidth
Specification NameInformation
IF Frequency ResponseSpecifications presented in the core chapter (“Agilent EXA Signal
Analyzer” on page 13) are redundantly contained within this
chapter.
IF Phase LinearitySee specifications in this chapter.
Spurious ResponsesThe “Spurious Responses” on page 46 still apply. Further,
bandwidth-option-dependent spurious responses are contained
within this chapter.
Third-Order Intermodulation, Displayed
Average Noise Level and Phase Noise
The performance of the analyzer will degrade by an unspecified
extent when using wideband analysis. This extent is not substantial
enough to justify statistical process control.
80Chapter 2
Option B25 (25 MHz) - Analysis Bandwidth
Other Analysis Bandwidth Specifications
Other Analysis Bandwidth Specifications
DescriptionSpecification
IF Spurious Response, 25 MHz IF
Bandwidth (Option B25)
IF second harmonic
Apparent
Freq. (f)
Any on-screen f(f + f
IF conversion image
Apparent
Freq. (f)
Any on-screen f2 × f
−20 dBmHigh−70 dBc (nominal)
a
d
Excitation Freq.
+ 22.5 MHz)/2−15 dBmLow−54 dBc (nominal)
c
e
Excitation Freq.
− f + 45 MHz−10 dBmLow−70 dBc (nominal)
c
Mixer
Level
−25 dBmHigh−54 dBc (nominal)
IF
b
Gain
Supplemental
Information
Preamp Off
c
a. To save test time, the levels of these spurs are not warranted. However, the relationship
between the spurious response and its excitation is described so the user can distinguish
whether a questionable response is due to these mechanisms or is subject to the specifications
in “Spurious Responses” in the core specifications. f is the apparent frequency of the spurious,
fc is the measurement center frequency.
b. Mixer Level = Input Level − Input Attenuation.
c. The spurious response specifications only apply with the preamp turned off. When the preamp
is turned on, performance is nominally the same as long as the mixer level is interpreted to be:
Mixer Level = Input Level − Input Attenuation − Preamp Gain
d. IF second harmonic significant only for Pre-FFT BW ≥10 MHz.
e. IF conver sion image significant only for Pre-FFT BW ≥10 MHz.
Chapter 2 81
Option B25 (25 MHz) - Analysis Bandwidth
Other Analysis Bandwidth Specifications
DescriptionSpecificationsSupplemental Information
SFDR (Spurious-Free Dynamic Range)
Signal Frequency within ±12 MHz of center–75 dBc
Signal Frequency anywhere within analysis BW–70 dBc (nominal)
Test conditions
a
a. Signal level is –6 dB relative to dBfs where: FS = –10 dBm at mixer, IF Gain = 0..
DescriptionSpecificationsSupplemental Information
IF Frequency Response
Demodulation and FFT
response relative to the
center frequency
Freq (GHz)
≤ 3.6≤ 100.40 dB0.12 dB0.100.03 dB
3.6 to 26.5≤ 100.25 dB
a
FFT Width
(MHz)
b
Max Errorc
(Exceptions
th
95
Percentile
Midwidth
d
)
Error
Slope
(dB/MHz)
e
Rms
(nominal)
≤ 3.610, ≤ 250.45 dB0.12 dB0.050.04 dB
3.6 to 26.510, ≤ 250.80 dB
a. The IF frequency response includes effects due to RF circuits such as input filters, that are a
function of RF frequency, in addition to the IF pass-band effects.
b. This column applies to the instantaneous analysis bandwidth in use. The range available
depends on the hardware options and the Mode. The Spectrum analyzer Mode does not allow
all bandwidths. The I/Q Analyzer is an example of a mode that does allow all bandwidths.
c. The maximum error at an offset (f) from the center of the FFT width is given by the expres-
sion
± [Midwidth Error + (f ×Slope)], but never exceeds ±Max Error. Usually, the span is no
larger than the FFT width in which case the center of the FFT width is the center frequency of
the analyzer . When the analyzer span is w ider than the FFT width, the span i s made up of multiple concatenated FFT results, and thus has multiple centers of FFT widths so the f in the
equation is the offset from the nearest center. These specifications include the ef fect of RF frequency response as well as IF frequency response at the worst case center frequency. Performance is nominally three times better at most center frequencies.
d. The specification does not apply for frequencies greater than 3.6 MHz from the center in FFT
widths of 7.2 to 8 MHz.
e. The “RMS” nominal performance is the standard deviation of the response relative to the cen-
ter frequency, integrated across a 10 or 25 MHz span. This performance measure was
observed at a center frequency in each harmonic mixing band, which is representative of all
center frequencies; it is not the worst case frequency.
82Chapter 2
DescriptionSpecificationSupplemental Information
IF Phase Linearity
Relative to mean phase linearity
Option B25 (25 MHz) - Analysis Bandwidth
Other Analysis Bandwidth Specifications
Freq
(GHz)
Span
(MHz)
Peak (nominal)
rms (nominal)
≤ 3.6≤ 10±0.5 deg0.2 deg
3.6 to 26.5≤ 10±1.5 deg0.4 deg
a. The listed performance is the r.m.s. of the phase deviation relative to the a best-fit linear phase
condition, where the r.m.s. is computed over the range of offset frequencies and center fre-
quencies shown.
a
Chapter 2 83
Option B25 (25 MHz) - Analysis Bandwidth
Other Analysis Bandwidth Specifications
84Chapter 2
3 Option EA3 -
Electronic Attenuator, 3.6 GHz
This chapter contains specifications for the Option EA3 Electronic Attenuator, 3.6 GHz.
85
Option EA3 - Electronic Attenuator, 3.6 GHz
Specifications Affected by Electronic Attenuator
Specifications Affected by Electronic Attenuator
Specification NameInformation
Frequency RangeSee “Range (Frequency and Attenuation)” on page 87.
1 dB Gain Compression PointSee “Distortions and Noise” on page 88.
Displayed Average Noise LevelSee “Distortions and Noise” on page 88.
Frequency ResponseS ee “Frequency Response” on page 89.
Attenuator Switching UncertaintyThe recommended operation of the electronic attenuator is with the
reference setting (10 dB) of the mechanical attenuator. In this
operating condition, the Attenuator Switching Uncertainty
specification of the mechanical attenuator in the core specifications
does not apply, and any switching uncertainty of the electronic
attenuator is included within the “Electronic Attenuator Switching
Uncertainty” on page 89.
Absolute Amplitude AccuracyUse “Frequency” specifications from this chapter and the formula
from the ““Absolute Amplitude Accuracy” on page 34 of the core
specifications.
Second Harmonic DistortionSee “Distortions and Noise” on page 88.
Third Order Intermodulation DistortionSee “Distortions and Noise” on page 88.
86Chapter 3
Option EA3 - Electronic Attenuator, 3.6 GHz
Other Electronic Attenuator Specifications
Other Electronic Attenuator Specifications
DescriptionSpecificationsSupplemental Information
Range (Frequency and Attenuation)
Frequency Range9 kHz to 3.6 GHz
Attenuation Range
Electronic Attenuator Range0 to 24 dB, 1 dB steps
Calibrated Range0 to 24 dB, 2 dB stepsElectronic attenuator is
calibrated with 10 dB
mechanical attenuation
Full Attenuation Range0 to 84 dB, 1 dB stepsSum of electronic and
mechanical attenuation
Chapter 3 87
Option EA3 - Electronic Attenuator, 3.6 GHz
Other Electronic Attenuator Specifications
DescriptionSpecificationsSupplemental Information
Distortions and NoiseWhen using the electronic attenuator, the
mechanical attenuator is also in-circuit. The full
mechanical attenuator range is available
1 dB Gain Compression Point The 1 dB compression point will be nominally
higher with the electronic attenuator “Enabled”
than with it not Enabled by the loss
high settings of electronic attenuation
Displayed Average Noise LevelInstrument Displayed Average Noise Level will
nominally be worse with the electronic attenuator
“Enabled” than with it not Enabled by the loss
Second Harmonic DistortionInstrument Second Harmonic Distortion will
nominally be better in terms of the second
harmonic intercept (SHI) with the electronic
attenuator “Enabled” than with it not Enabled by
b
the loss
.
a
.
b
, except with
c
.
b
.
Third-order Intermodulation
Distortion
Instrument TOI will nominally be better with the
electronic attenuator “Enabled” than with it not
Enabled by the loss
high attenuation setting and high signal frequency
b
except for the combination of
a. The electronic attenuator is calibrated for its frequency response only with the mechanical
attenuator set to its preferred setting of 10 dB.
b. The loss of the electronic attenuator is nominally given by its attenuation plus its excess loss.
That excess loss is nominally 2 dB from 0 − 500 MHz and increases by nominally another
1 dB/GHz for frequencies above 500 MHz.
c. An additional compression mechanism is present at high electronic attenuator settings. The
mechanism gives nominally 1 dB compression at +20 dBm at the internal electronic attenuator input. The compression threshold at the RF input is higher than that at the internal electronic attenuator input by the mechanical attenuation. The mechanism has negligible effect
for electronic attenuations of 0 through 14 dB.
d. The TOI performance improvement due to electronic attenuator loss is limited at high fre-
quencies, such that the TOI reaches a limit of nominally +45 dBm at 3.6 GHz, with the preferred mechanical attenuator setting of 10 dB, and the maximum electronic attenuation of
24 dB. The TOI will change in direct proportion to changes in mechanical attenuation.
d
88Chapter 3
DescriptionSpecificationsSupplemental Information
Frequency Response
Maximum error relative to
reference condition (50 MHz)
Option EA3 - Electronic Attenuator, 3.6 GHz
Other Electronic Attenuator Specifications
20 to 30 °C5 to 50 °C
Attenuation = 4 to 24 dB, even
steps
9 kHz to 10 MHz±0.75 dB±0.90 dB±0.32 dB
10 MHz to 50 MHz±0.65 dB±0.69 dB±0.27 dB
50 MHz to 2.2 GHz±0.48 dB±0.60 dB±0.19 dB
2.2 GHz to 3.6 GHz±0.55 dB±0.67 dB±0.20 dB
Attenuation = 0, 1, 2 and odd steps,
3 to 23 dB
10 MHz to 3.6 GHz±0.30 dB
DescriptionSpecificationsSupplemental Information
Electronic Attenuator Switching
Uncertainty
th
Percentile (≈2σ)
95
Er r or r e lat i ve t o ref e ren c e
condition (50 MHz, 10 dB
mechanical attenuation, 10 dB
electronic attenuation)
Attenuation = 0 to 24 dB
9 kHz to 3.6 GHz
See note
a
a. The specification is ±0.14 dB. Note that this small relative uncertainty does not apply in esti-
mating absolute amplitude accuracy . It is included within the absolute amplitude accuracy for
measurements done with the electronic attenuator. (Measurements made without the elec-
tronic attenuator are treated differently; the absolute amplitude accuracy specification for
these measurements does not include attenuator switching uncertainty.)
Chapter 3 89
Option EA3 - Electronic Attenuator, 3.6 GHz
Other Electronic Attenuator Specifications
90Chapter 3
4 Option P03 - Preamplifier
This chapter contains specifications for the EXA Signal Analyzer Option P03 preamplifier.
91
Option P03 - Preamplifier
Specifications Affected by Preamp
Specifications Affected by Preamp
Specification NameInformation
Frequency RangeSee “Frequency Range” on page 15 of the core specifications.
Nominal Dynamic Range vs.
Offset Frequency vs. RBW
Measurement RangeThe measurement range depends on DANL.
Gain CompressionSee specifications in this chapter.
DANLSee specifications in this chapter.
Frequency ResponseSee specifications in this chapter.
Absolute Amplitude AccuracySee ““Absolute Amplitude Accuracy” on page 34 of the core
Does not apply with Preamp On.
See “Amplitude Accuracy and Range” on page 29.
specifications.
RF Input VSWRSee plot in this chapter.
Input Attenuation Switching
Uncertainty
Display Scale FidelitySee “Display Scale Fidelity” on page 39 of the core specifications.
Third Order Intermodulation
Distortion
Other Input Related SpuriousSee “Spurious Responses” on page 46 of the core specifications.
Dynamic RangeSee plot in this chapter.
GainSee “Preamp” specifications in this chapter.
Noise FigureSee “Preamp” specifications in this chapter.
See “Input Attenuation Switching Uncertainty” on page 33 of the core
specifications.
See specifications in this chapter.
92Chapter 4
Option P03 - Preamplifier
Other Preamp Specifications
Other Preamp Specifications
DescriptionSpecificationsSupplemental Information
Preamp (Option P03)
Gain
100 kHz to 3.6 GHz+20 dB (nominal)
Noise figure
100 kHz to 3.6 GHz15 dB (nominal)
a. The preamp follows the input attenuator, AC/DC coupling switch, and precedes the input mixer. In
low-band, it follows the 3.6 GHz low-pass filter.
b. Preamp Gain directly affects distortion and noise performance, but it also affects the range of levels that
are free of final IF overload. The user interface has a designed relationship between input attenuation
and reference level to prevent on-screen signal levels from causing final IF overloads. That design is
based on the maximum preamp gains shown. Actual preamp gains are modestly lower, by up to nomi-
nally 5 dB for frequencies from 100 kHz to 3.6 GHz.
a
Maximum
b
Chapter 4 93
Option P03 - Preamplifier
Other Preamp Specifications
DescriptionSpecificationsSupplemental Information
1 dB Gain Compression Point
(Two-tone)
ab
Preamp On (Option P03)
Maximum power at the
preamp
c
for 1 dB gain compression
10 MHz to 3.6 GHz−10 dBm (nominal)
a. Large signals, even at frequencies not shown on the screen, can cause the analyzer to mismeasure
on-screen signals because of two-tone gain compression. This specification tells how large an interfering signal must be in order to cause a 1 dB change in an on-screen signal.
b. Reference level and off-screen performance: The reference level (RL) behavior differs from some ear-
lier analyzers in a way that makes this analyzer more flexible. In other analyzers, the RL controlled how
the measurement was performed as well as how it was displayed. Because the logarithmic amplifier in
these analyzers had both range and resolution limitations, this behavior was necessary for optimum
measurement accuracy. The logarithmic amplifier in this signal analyzer, however, is implemented digitally such that the range and resolution greatly exceed other instrument limitations. Because of this, the
analyzer can make measurements largely independent of the setting of the RL without compromising
accuracy. Because the RL becomes a display function, not a measurement function, a marker can read
out results that are off-screen, either above or below, without any change in accuracy. The only exception to the independence of RL and the way in which the measurement is performed is in the input
attenuation setting: When the input attenuation is set to auto, the rules for the determination of the input
attenuation include dependence on the reference level. Because the input attenuation setting controls
the trade-off between large signal behaviors (third-order intermodulation and compression) and small
signal effects (noise), the measurement results can change with RL changes when the input attenuation
is set to auto.
c. Total power at the preamp (dBm) = total power at the input (dBm) − input attenuation (dB).
94Chapter 4
Option P03 - Preamplifier
Other Preamp Specifications
DescriptionSpecificationsSupplemental Information
Displayed Average Noise Level
(DANL) − Preamp On
(Option P03)
a
Input terminated,
Sample or Average detector
Averaging type = Log
0 dB input attenuation
Refer to the footnote for
Band Overlaps on page 15.
IF Gain = Any setting
1 Hz Resolution Bandwidth
Preamp On
20 to 30 °C5 to 50 °CTypicalNominal
Option P03
100 kHz to 1 MHz
b
−146 dBm
1 MHz to 10 MHz−161 dBm
10 MHz to 2.1 GHz−161 dBm−159 dBm−163 dBm
2.1 GHz to 3.6 GHz−160 dBm−158 dBm−162 dBm
a. DANL for zero span and swept is normalized in two ways and for two reasons. DANL is measured in a
1 kHz RBW and normalized to the narrowest available RBW, because the noise figure does not depend
on RBW and 1 kHz measurements are faster. The second normalization is that DANL is measured with
10 dB input attenuation and normalized to the 0 dB input attenuation case, because that makes DANL
and third order intermodulation test conditions congruent, allowing accurate dynamic range estimation
for the analyzer.
b. Specifications apply only when the Phase Noise Optimization control is set to “Best Phase Noise at off-
set > 30 kHz.”
Chapter 4 95
Option P03 - Preamplifier
Other Preamp Specifications
DescriptionSpecificationsSupplemental Information
Frequency Response − Preamp On
(Option P03)
Refer to the footnote for
Band Overlaps on page 15.
Maximum error relative to
reference condition (50 MHz)
Input attenuation 0 dB
Swept operation
a
20 to 30 °C5 to 50 °C
th
95
Percentile (≈2σ)
20 to 30 °C
100 kHz to 3.6 GHz
b
±0.28 dB (nominal)
a. For Sweep Type = FFT, add the RF flatness errors of this table to the IF Frequency Response errors. An
additional error source, the error in switching between swept and FFT sweep types, is nominally ±0.01
dB and is included within the “Absolute Amplitude Error” specifications.
b. Electronic attenuator (Option EA3) may not be used with preamp on.
96Chapter 4
Nominal VSWR − Preamp On (Plot)
VSW R vs. F req uency , 3 Units, Preamp On, 0 dB Attenuation
1.0
1.1
1.2
1.3
1.4
1.5
1.6
1.7
1.8
0.00.51.01.52.02.53.03.5
GHz
VSWR
Option P03 - Preamplifier
Other Preamp Specifications
DescriptionSpecificationsSupplemental Information
Third Order
Intermodulation Distortion
Tone separation 5 times IF
Prefilter Bandwidth
a
Sweep type not set to FFT
Preamp On
(Option P03)
Preamp
b
Level
30 MHz to 3.6 GHz−45 dBm−90 dBc0.0 dBm
a. See the IF Prefilter Bandwidth table in the specifications for “Gain Compression” on page 42. When
the tone separation condition is met, the effect on TOI of the setting of IF Gain is negligible.
b. Preamp Level = Input Level − Input Attenuation.
c. TOI = third order intercept. The TOI is given by the preamplifier input tone level (in dBc) minus (dis-
tortion/2) where distortion is the relative level of the distortion tones in dBc.
Chapter 4 97
Distortion
(nominal)
c
TOI
(nominal)
Option P03 - Preamplifier
Other Preamp Specifications
Nominal Dynamic Range at 1 GHz, Preamp On (Plot)
98Chapter 4
5 Option PFR - Precision Frequency
Reference
This chapter contains specifications for the Option PFR Precision Frequency Reference.
99
Option PFR - Precision Frequency Reference
Specifications Affected by Precision Frequency Reference
Specifications Affected by Precision Frequency
Reference
Specification NameInformation
Precision Frequency ReferenceSee “Precision Frequency Reference” on page 18 in the core
specifications.
100Chapter 5
Loading...
+ hidden pages
You need points to download manuals.
1 point = 1 manual.
You can buy points or you can get point for every manual you upload.