Agilent Technologies PN 8510-16 Quick start guide

Agilent PN 8510-16
Controlling Test Port Output
Power Flatness
Product Note
UNCORRECTED POWER
CORRECTED POWER OUTPUT POWER
Agilent 8510C Network Analyzer

2
Introduction
Designers and manufacturers of active devices
often need to control the power level at the test
port1of their power-sensitive devices, but find dif-
ficulty in overcoming insertion losses. The insertion
losses occur as a result of connecting components
in the measurement path between the source and
the DUT. The Agilent Technologies 8 10C is a
microwave network analyzer2capable of setting
and controlling the power level at the test port.
This product note reviews the implementation
and operational considerations of the 8 10C’s
test port power flatness-correction feature.
The 8 10C performs a flatness-correction cali-
bration by measuring the test port power level
using an Agilent power meter and creates a table
of power corrections versus frequency. The
power meter then stores the table into an Agilent
8360 series synthesized sweeper.3When the test
port flatness correction is enabled, the source
will adjust its output power to compensate for
path losses at each measurement point in the
frequency span.
1. In this document, “test port” will refer to the point in the system where the test
device is connected. This may be port 1 or port 2 of any 8510 coaxial-based test
set or the end of a cable or adapter that is attached to the DUT.
2. 8510B network analyzers with revision B.06 or later may also be used to achieve
constant test port power levels. To obtain the latest firmware revision order the
11575F upgrade kit.
3. Any 8360 synthesized sweeper with firmware revision of September 25, 1990 or
earlier must be upgraded. To upgrade your 8360 to the latest firmware revision,
order part number 08360-60167.
Power flatness correction is useful for characteriz-
ing active devices by allowing measurements of
absolute power, gain versus input power, and
swept-frequency gain compression. Swept-power
gain compression measurements are possible in
the power domain with the flatness correction
enabled.
The 8 10XF measurement system is capable of lev-
eling and controlling the test port power between
4 MHz and 110 GHz without requiring a power
calibration for each measurement. When power lev-
eling is enabled, the test port power variation is
typically less than ±1 dB over the full frequency
sweep, with 20 dB of adjustable power. For more
information regarding power leveling in the Agilent
8 10XF, refer to product note 8 10XF-1, literature
number 968- 270E.

3
System Configuration
The basic 8 10C measurement system configura-
tion needed for test port flatness correction is
shown in Figure 1. A single channel power meter
such as the Agilent 437B, E4418A, or E4418B is
required, along with a compatible sensor, for per-
forming the power flatness calibration. Power
meters with the 437B command set must be used
for the power level flatness correction.
Figure 1. Basic Agilent 85107B measurement system con-
figuration for test port flatness correction
The Agilent 8 11B frequency converter test
set may be used with the test port power flatness-
correction feature. The range of the test port
power is determined by the available power of
the 8360 source and the insertion loss of the com-
ponents added to the measurement system.
8517B test set
8485A
power sensor
E4418B
power meter
83651B
synthesized sweeper
8510 system bus
85107B Measurement System
8510C network analyzer

4
Flatness-Correction Operation
The basic setup procedure to obtain flat test port
output power is illustrated in Table 1. To simplify
the execution of the procedure, the front panel
hardkeys are enclosed in [brackets]. The softkeys
are enclosed in {braces}. Table 2 provides the pro-
cedure for setting up the power meter. The net-
work analyzer should be reset before performing
any procedures. The flatness-correction calibration
can be performed in step, ramp, or frequency list
sweep mode. Figure 2 illustrates test port power
versus frequency with and without flatness correc-
tion using an 8 107B measurement system.
When the flatness correction is enabled in an
8 10C measurement system the analyzer’s Power
Source 1 softkey controls the test port power level.
The source and test set used will determine the
available test port power range.
The maximum leveled power the source can out-
put at different frequency spans is indicated in
Table 3. Once the flatness correction is enabled,
the test port power level must be set within the
power range presented in Table 4 for common
source/test set combinations.
Figure 2. Comparison of test port power without flatness correction (channel 1)
and with flatness correction (channel 2)

5
Table 1. Flatness-correction calibration procedure
8510C Keystrokes Description
Set up the power meter (see Table 2)
For proper operation, the E4418B must be set up before initiating the flatness-correction routine.
See Table 2 for specific instructions.
Verify the power meter's address on the 8510 system
[LOCAL] When shipped from the factory, the address of the power meter is 13. If this conflicts with another instrument on the system bus,
{POWERMETER} change the address of the conflicting instrument. The power meter must be set to address 13 in order for the firmware to func-
tion properly. Press [SYSTEM/INPUTS], {REMOTE INTERFACE} and {INTERFACE OVERVIEW} on the E4418B to ensure the
power meter’s address is set to 13. Verify that the power meter is connected to the system bus.
Set up the analyzer and measurement
Set up the start/stop frequencies and the measurement type (for example, S11, S12).
Adjust the number of analyzer trace points (if needed)
STIMULUS [MENU] When the flatness calibration is initiated, the analyzer sends the source a list of flatness-correction frequencies equal to the
{NUMBER OF POINTS} number of trace points set on the analyzer.
{POINTS}
Set the source to the maximum leveled power
STIMULUS [MENU] Set the 8360 for maximum specified power (P1) at the highest frequency in the span; see Table 3. At higher power levels, the
{POWER MENU} power meter will take less time to settle between measurement points during the calibration process.
{POWER SOURCE 1}
P1[x1]
Connect the power sensor to the active test port and begin the flatness calibration
Initiate flatness calibration
{POWER FLATNESS} Do not cycle the power on the analyzer or the source during the calibration process. The calibration may be aborted at any time by
{CALIBRATE FLATNESS} pressing any key on the analyzer’s front panel. Aborting the calibration is not recommended, since the command may cause the
system to freeze. By pressing {CALIBRATE FLATNESS} the analyzer will remind the user to zero the power meter and connect the
power sensor to the source. Press {CALIBRATE FLATNESS} again on the analyzer to initiate the calibration. Once initiated, the
analyzer retrieves the power meter measurement data and transfers it to the 8360 where it is processed and stored with the
appropriate correction frequency. During the flatness-correction calibration, the analyzer will display MEASURING DATA, xx%
DONE, PRESS ANY KEY TO ABORT. The xx% DONE indicates the percentage of the total number of correction points that have
been measured; it is not an indication of elapsed measurement time. When the flatness-correction calibration is completed, the
analyzer will automatically store the correction table into register 1 of the source. Any new calibration of the source will overwrite
the flatness-correction table.
Activate the flatness-correction
{FLATNESS ON} The source output power will now be unleveled as it attempts to output the test port power level (Power Source 1) plus the flat-
ness correction for each measurement point in the frequency span. IF Overload or Source 1 Warning—RF Unleveled may be dis-
played on the analyzer until the test power level is reduced.
Set the test port power level
STIMULUS [MENU] Set the test port power level (P2) equal to or below the maximum allowed test port power (see Table 3) for the highest frequency
{POWER MENU} in the measurement span. A constant power level equivalent to P2will now be available at the test port. The flatness-correction
{POWER SOURCE 1} calibration can be varified using the E4418B to measure the power at certain CW frequencies in the frequency range.
P2[x1]
Perform the measurement calibration
Compensate for systematic errors by performing a measurement calibration (if desired). Although a measurement calibration can
be performed before or after the flatness-correction calibration with or without correction enabled, it is recommended that the cal-
ibration be performed after the flatness correction is enabled. The calibration is no longer valid if the source power is changed
after performing the calibration.
Connect the DUT to the test port

6
Table 2. Agilent E4418B power meter setup
E4418B Keystrokes Description
Preset and zero the power meter
[PRESET/LOCAL] Return the power meter to a known state.
{CONFIRM}
[ZERO/CAL]
{ZERO}
Set up the 437B emulation on the power meter
[SYSTEM/INPUTS] The 437B command set must be activated in order to perform the power flatness calibration. Power meters that can be used in
{REMOTE INTERFACE} place of the E4418B are the E4418A and 437B. The E4419B dual channel power meter does not have the 437B command set and
{COMMAND SET} cannot be used to perform the calibration.
{437B}
Connect the sensor to the power reference output on the power meter and calibrate the power sensor
[ZERO/CAL] Choose a power sensor that can perform within the desired frequency span. To determine which power sensors are compatible
POWER REF {ON} with the E4418B, consult the user’s guide. Agilent E-Series power sensors cannot be used for calibration because they are incom-
{CAL} patible with the 437B command set. Additional configuration may be needed to connect the sensor to the power reference;
POWER REF {OFF} this information can also be found in the E4418B user’s guide, part number E4418-90032. After calibrating the sensor, the
power meter should display a reading of 0.0 dBm (or 1 mW) when the sensor is connected to the reference and the power
reference is activated.
Select or create the calibration factor table that applies to the power sensor in use (if needed)
[SYSTEM/INPUTS] The factory enters nine tables of typical calibration factors for nine different sensors in the E4418B. Use the up and down
{TABLES} arrows on the display to highlight the desired table. If tables 2 through 9 are cleared, the data previously stored in those tables
{SENSOR CAL TABLES} cannot be restored. See the E4418B user’s guide, part number E4418-90032, for information on entering custom calibration
HIGHLIGHT TABLE factor tables.
TABLE {ON}
{DONE}
Table 3. Agilent 8360 series synthesized sweeper maximum leveled power (in dBm)
Maximum Frequency 83620B 83621B 83623B 83623L 83631B 83651B
20 GHz 13 10 17 15 10 10
26.5 GHz N/A N/A N/A N/A 4 4
40 GHz N/A N/A N/A N/A N/A 3
50 GHz N/A N/A N/A N/A N/A 0
Table 4. Settable test port power ranges (assuming no test set step attenuation)
RF Source 83620B/83621B 83631B 83651B
8510 Test Set with 8514B with 8515A with 8515A with 8517B with 8517B, Option 007
Frequency Test Port Power Levels [Pmax to Pmin] (in dBm)
0.05 GHz 2.5 to –20.5 –3.5 to –26 –3.5 to –26 –1.5 to –21.5 5 to –21
2 GHz 1 to –22 –6 to –29 –6 to –29 0.5 to –23.5 5 to –21
20 GHz –7.5 to –27 –13.5 to –30 –13.5 to –30 –7.5 to –30 2 to –23
26.5 GHz N/A N/A 25 to –30 13.5 to –30 1 to –24
40 GHz N/A N/A N/A –20 to –30 – 3 to –21.5
50 GHz N/A N/A N/A –27 to –30 –13 to –29

7
The following error messages may occur while
attempting to enable the flatness correction and/or
reducing the test port power level:
• If flatness correction is enabled before reducing
the test port power level, IF Overload or Source
1 Warning—RF Unleveled may be displayed on
the analyzer as the source becomes unleveled.
Unleveling occurs when the source attempts to
output its maximum specified power plus the
flatness correction.
• If the output power is reduced to a suitable test
port power level before turning flatness correc-
tion on, No IF Found may be displayed. There
may not be sufficient output power from the
source for the 8 10C to phase lock to the signal.
These error messages should disappear when flat-
ness correction is enabled with the appropriate
test port power level setting. If the test port power
level is not correctly reduced, the source will
become unleveled and try to output maximum
power. Although flatness correction will be applied
to the unleveled signal, the measurement for the
unleveled portion of the frequency span will not be
valid because the flatness-correction feature can-
not compensate for the inconsistent power varia-
tions that occur.
Power control with flatness correction
The ability to compensate for power variations
across the entire measurement span of the source
will depend on the highest leveled output power
the source can produce and the amount of power
compensation required. The test port power may
be adjusted over the entire Pmax to Pmin range with-
out degrading the flatness-correction calibration.
Table 4 shows test port power ranges for various
source/test set combinations. To determine the test
port power levels for a particular frequency span,
choose a power level between Pmax of the highest
frequency in the measurement span and Pmin of the
lowest frequency in the measurement span. For
example, the power range for a 0 MHz to 20 GHz
span with an 83621B/8 1 A system is –13. dBm
to –26 dBm. A high-power Agilent 83623B adds
7 dBm to Pmax of the 83621B. If a user sets up a
0 MHz to 20 GHz measurement with an 83623B
and an 8 1 A, the test port power range is
–6. dBm to –26 dBm.
Operational Considerations
Calibration time
The user may choose to speed up the calibration process by reducing the number of trace points on the
analyzer. Table provides some typical calibration times for flatness correction over the full frequency
span of the source for different source/test set combinations. Calibration times may be slightly reduced by
increasing the stepped measurement speed with the quick step feature on the Agilent 8 10C.
Table 5. Typical calibration times (in minutes) versus number of calibration points with maximum leveled source power
Number of Points 83620B/83621B 83631B 83651B
with 8514B with 8515A with 8515A with 8517B
801 30 32 40 60
401 15 16 20 30
201 8 8 10 15
101 4 4 5 8
512224

8
Verifying a flatness-correction calibration
To verify the flatness-correction calibration, the
power sensor should be reconnected to the test
port to measure the test port power at individual
frequencies. Since the measurement system is
calibrated in 0W, inaccuracies will occur when
a device is not well matched. Since the flatness-
correction calibration is not a real-time power
leveling feature, it cannot correct for mismatches
that occur between the test port and the DUT.
Adjusting the frequency span after calibration
Once a flatness-correction calibration with the
maximum number of points (801 points) has been
completed, adjustments can be made to reduce the
number of trace points on the analyzer. The user
may also change the measurement frequency span
to a subset of the original calibration span without
invalidating the calibration. The source will output
corrected power at the appropriate measurement
points.
This capability is particularly useful for users who
are testing a number of devices with different
frequency spans at the same test station. In this
case, the user may choose to perform the flatness-
correction calibration with the maximum number
of calibration points across the full frequency
range. Subsets of the original calibration frequency
range can then be used to meet the specific testing
requirements of the individual devices.
When to recalibrate
The flatness-correction calibration does not need
to be repeated unless: (1) the user wants to cali-
brate over a wider frequency range, (2) the meas-
urement path between the source and the test
device changes, (3) the RF source power changes,
(4) the user wants to increase the number of meas-
urement points, or ( ) the environmental condi-
tions under which the original calibration was per-
formed changes dramatically.
Keep in mind that the flatness-correction table
is automatically saved into register 1 of the RF
source. Any new calibration of the source will
overwrite the flatness-correction table. If multiple
flatness-correction calibrations are performed,
only the most recent calibration will be saved for
the DUT.
Using test set step attenuators with flatness correction
If lower test port power levels are desired, test
set step attenuators may be used. The frequency
response of the step attenuators can be eliminated
from the measurement by performing the flatness-
correction calibration with the appropriate atten-
uation enabled. High sensitivity power sensors
(848 D to 26. GHz and 8487D to 0 GHz) are
available for power measurements from –20 dBm
to –70 dBm. The limitation of making power meter
measurements at these low power levels is that the
actual calibration process takes considerably more
time since the power meter takes much longer to
settle at each correction frequency.

9
Using test port 1 calibrations on test port 2
In most cases, port 1 will be the input port of the
DUT. When port 2 must be used as the input port
to the device, the user may choose to use a port 1
flatness-correction calibration on port 2 since the
port 1 and port 2 signal paths are symmetrical.1
Figure 3 illustrates the use of port 1 flatness cor-
rection on both ports 1 and 2. The port 2 measure-
ment with port 1 flatness-correction calibration is
optimized for measurements below 30 GHz.
1. Test sets with option 003 (high forward dynamic range) cannot be used because
the reverse transmission dynamic range is degraded.
Flatness corrections in fixture or wafer-probing
environments
Test port flatness correction may be applied
with any other power function. Power slope may
be used to compensate for the path loss in non-
coaxial environments such as microstrip and
coplanar waveguide measurement systems. The
maximum test port power for any particular fre-
quency span cannot exceed the maximum test port
power level for the highest frequency in the span
(see Table 3) minus the maximum power slope
compensation required.
Figure 3. Corrected test port power using port 1 flatness correction on port 1
(channel 1) and port 2 (channel 2)

10
Practical Application Examples
Absolute output power measurements with
flatness correction
One benefit of the flatness-correction capability is
the ability to measure the absolute output power of
active devices. Since the input power level to the
DUT is kept constant, the magnitude offset feature
of the Agilent 8 10 can be used to display absolute
output power across the entire frequency span of
the device. Table 6 shows the procedure for
absolute output power versus frequency measure-
ments with test port flatness correction.
Table 6. Absolute output power measurements with flatness correction
8510C Keystrokes Description
Set up the source and power meter as illustrated in Figure 2
Set up the measurement
PARAMETER [MENU] Set up the analyzer for a b2measurement.
{USER 2 b2}
Activate the test port power flatness-correction
Perform the flatness correction calibration. Set the test port power level and enable the flatness
correction as explained in Table 1.
Connect a thru
Perform a thru calibration with the flatness correction enabled
[CAL] Perform a thru calibration to eliminate the frequency response errors of the port 2 path in the measurement. Be sure to
{CAL#…} include any attenuators and/or adapters which are part of the measurement in the thru calibration. It may be necessary to
{CALIBRATE:RESPONSE} swap adapters for the thru connection. Any cal set may be selected to access the response calibration in the calibration
{THRU} menu structure
{DONE RESPONSE}
{CAL SET #}
Connect the DUT
Measure the absolute output power
RESPONSE [MENU] When the device is reconnected, the gain will be displayed. Enter a magnitude offset equivalent to the test port power level
{MORE} (P2). Measure the absolute output power at any point in the measurement span.
{MAGNITUDE OFFSET}
P2[x1]