Automatic Fixture Removal (AFR)


Note: This feature requires Option S9x007A/B or 007.

In this topic:

Overview

Fixtures are often used for DUTs that have non-coaxial interfaces. This feature allows you to mathematically remove, or de-embed, a characterized test fixture from displayed measurement results of the test fixture and DUT.

Before starting the AFR process, Perform a calibration at the connectors of the test fixture (red lines in images below).

The AFR Wizard will guide you through these steps:

  1. Press Cal > Fixtures > Auto Fixture Removal....

  2. Describe your fixturing situation.

  3. Specify How the Thru fixture characterization will occur.

  4. Do characterization.

  5. Remove the effects of the test fixture. Leaves ONLY the displayed results of the DUT.

  6. Touchstone files are saved that characterize the two halves of the test fixture.

Requirements

Single-ended Fixtures

Test fixture

Thru standard

OR

Left and Right halves of Thru standard

 

Supports both 1-port (not shown) and 2-port single-ended DUT configurations.

When used with a 4-port DUT configuration, the trace coupling is not removed.

Differential (Balanced) Fixtures

Test fixture

Thru standard

OR

Left and Right halves of Thru standard

 

Supports 4-port DUT configuration ONLY which includes removing the effects of coupling between the differential traces.

How to start the Automatic Fixture Removal Wizard

With a calibrated measurement of the DUT in the test fixture present:

Click Response, then Cal, then Fixtures, then Automatic Fixture Removal

Note: The dialogs below show images for a Single ended DUT, but Differential works exactly the same, only with Differential S4P files.

Automatic Fixture Removal Wizard

For best results, follow the AFR Wizard tabs from steps 1 through 5 by either clicking Next > or clicking the tabs.

In this section:

  1. Describe Fixture

  2. Specify Standards

  3. Measure Standards

  4. Remove Fixture

  5. Save Fixture


1. Describe Fixture

The choices that you make in the dialog are reflected in the diagram and text (red box in following image).

My fixture inputs are:

  • Single Ended - The fixture and DUT have single-ended inputs and outputs.

  • Differential - The fixture and DUT have differential inputs and outputs.

My measurement is (Single Ended):

  • 1-port - such as S11

  • 2-port - such as S21

  • Multiport / Use THRUs - select up to 64 ports. Check Use THRUs to use THRU fixture.

My measurement is (Differential):

  • 2-port - such as SDD11

  • 4-port - such as SDD21

  • Multiport / Use THRUs- select up to 64 ports. Check Use THRUs to use THRU fixture.

Advanced Settings (click ^ to show and hide)

After fixture removal set Calibration Reference Z0 to: (Choose one of these settings)

  • System Z0 - Sets impedance to the System Impedance setting. Learn how to set System Z0.

  • Measured Fixture Z0 - Sets impedance to the value that is measured during the AFR process. Not allowed when 'band limited' is selected below.

  • <nn> ohms - Sets impedance to an arbitrary value.

  • Set 'System Z0" to Calibration Reference Z0 - When the impedance is measured or set to an arbitrary value, check to also set the System Z0 to the same value.

Select all that apply:

  • Check if this is true: I want to correct for when the match (Return Loss) of Fixture A is NOT equal to the match of Fixture B.

  • Check if this is true: I want to correct for when the electrical length of Fixture A is NOT equal to the electrical length of Fixture B.

  • Check if this is true: My fixture is band limited. Bandpass mode will be used during the Time Domain measurement. If NOT checked, then Lowpass mode is used. Because Lowpass mode includes impedance in the calculation, it renders the best accuracy. Learn more about these settings.  

Note

When using Lowpass mode and an error message appears ("Measurement settings are not adequate..."), change the start frequency and the number of points so that the frequency span between data points equals the start frequency. This can be done by selecting values using the following logic:

Start freq = 10 MHz

then either:

Stop freq = 20 GHz

Number of points = 2000

or

Stop freq = 50 GHz

Number of points = 5000

In either case, the frequency span between data points equals 10 MHz, the start frequency.

AFR Configuration Dialog

The AFR Configuration dialog can be launched by clicking on the AFR Config icon (shown below).

Access the following settings in the General tab of the AFR Configuration dialog.

Set manual time domain start and stop settings for corner cases shown under Time Domain Settings.

The Tolerance (in ohms) and Number of Iterations can be set under Wizard Tab 3, Edit Window, Measured Fixture Impedance Iteration. Check the Auto Iterate box to automatically iterate when Impedance Method Auto is checked.

DUT Gate -  AFR fixture characterization is based on time domain gating. This can cause high frequency rippling of characterized fixture if the impedance of fixture and DUT doesn't match very well. AFR DUT gating can apply gating on the DUT after de-embedding the fixture from fixtured DUT. This option can smooth the high frequency rippling of DUT.

Coefficient -  Times of fixture length which is used to calculate the stop time of gating window.

Only Reflection Parameters - If this option is on, only reflection parameters of DUT will be gated, otherwise all parameters of DUT will be gated.

AFR Mode Conversion - AFR mode conversion will turn on or off automatically to make the AFR result more precise with non-single ended data.


2. Specify Standards

Note: AFR Config and 2X Thru for N is a Licensed Feature. Learn more about Licensed Features.

Selecting 2 single-ended ports, or 4 differential ports allows a 2X Thru choice.

Checking Use THRUs in the 1. Describe Fixture tab allows 2X Thru choice for a Multiport measurement.

If My characterization fixture DUT measurement fixture is checked in the 1. Describe Fixture tab, the Fixtured DUT choice is displayed and checked.

Otherwise, the fixture MUST be characterized using 1-port fixture measurements using either OPEN or SHORT terminations.

Note: The term 'Standards' is used here because this process can be thought of as the second in a '2-tier' calibration. The first tier of the calibration must already be performed (the VNA calibrated) before starting the AFR process. Another way of describing this step would be:

"How will you be measuring or loading the characterization of the Thru standard?"

  • 2X Thru - Both halves together in one fixture.

  • Both 1X Thru halves separately (also known as 1-port AFR). Specify either Open and/or Short at the end of each half of the standard.

Notes:

  • This image:appears as the 2X Thru when you select 1. Describe Fixture, My measurement is 1 port. This means the left and right fixtures are mirror images of each other and have the same S-parameters. The cascaded combination of the two fixtures are Fixture A + Reversed Fixture A, or A'.

  • If both halves are identical and you do not have a 2X Thru, then only Fixture A measurements are necessary.

Advanced Settings

This setting is used to describe any ADDITIONAL length between the halves of the Thru or added to either of the individual halves.

If the electrical length of the Thru standard is identical to the test fixture, then make no changes to the default settings (Known length = 0).

My Thru fixture has:

  • Known thru length - Enter the length in nanoseconds. See a simulated length in the diagram between the two halves of the Thru.

  • Unknown thru length computed using reflects.  This setting requires the two halves of the Thru fixture be characterized separately with a reflect standard.

  • Unknown thru length computed using fixtured DUT measurements.  This setting requires an additional characterization of the Fixture + DUT.


3. Measure Standards

The following dialog is for Multiport 2X Thru measurements when Use THRUs is checked in the 1. Describe Fixture tab.

This step characterizes the 1X or 2X Thru standards. This is done by either performing measurements or by loading one or more *.snp files that describe the characterization of the fixtures.

Note:  When loading standards from files, the typical system characterization impedance (Z0) value is used, which is 50 Ω.

Click Measure to see the following dialog:

For best results, the analyzer should be calibrated. Also, the measurement Start and Step frequencies should be equal. This is necessary for TDR measurements.

Connect the specified standard at the PNA port. Then click Measure.

OR click Load and navigate to the *.snp file that describes the standard.

For 2X Thru standards, the following dialog allows you to optionally remap the fixture ports.

Calculated Fixture Characteristics

Note: The fixture length must be 4 times the rise time. If not, an error message will be displayed. For example, a measurement at 26.5 GHz has a rise time of 37.7 ps. Therefore, the fixture length must be 4 * 37.7 ps = 151 ps.

The loaded or measured Impedance and Electrical Length of the fixture are calculated and displayed here.

The preview button allows you to visualize the time domain characteristics of the fixture model that will be used for de-embedding. By using the editing features noted below, the output match, length, and impedance of the fixture model can be optimized. This allows flexibility for including or excluding physical characteristics such as excess inductance of plug fingers, excess capacitance of receptacle pads or any other anomalies not desired in the fixture model. This advanced feature should be used with care as misuse could cause passivity or causality errors in the resultant fixture model.

Click Edit to start the following, interactive Measure Fixture Impedance TDR Data dialog.

This dialog allows you to change the Impedance and/or Fixture Length of the saved *.snp data by moving a marker to the desired location on the TDR plot.

Tips

  • The TDR X-axis shows the 1-way electrical length of the fixture in time.

  • Right-click the plot, then click Autoscale.

  • Click and drag to zoom the display on specific details of the plot.

Editing Mode.  (For 2X Thru, Fixture Length can NOT be changed.)

Choose one or more of the following to edit the data in resulting *.snp files:

  • Impedance - move the M1 marker to edit the data as though the measurement was made in the specified characteristic impedance. The Z marker is used to measure the fixture impedance.

  • Gate - move the flag to 'gate out' a part of the fixture measurement, usually close to the transition. The Z marker automatically changes to the location of the gate. The gate flag cannot move further than the fixture length diamond marker.

  • Fixture Length - move the diamond marker to change only the electrical length of the fixture. The fixture length cannot be shortened to a value less than the current gate value. In this way, the gate value will never be greater than the electrical length.

Impedance Editing Mode

Select Fixture Impedance - Select the A or B fixture.

Impedance Method - Select Auto then click Iterate to measure the fixture Z0 automatically and return fixture impedance back to 50 Ω. If more than one fixture impedance needs to be measured, check All to measure all impedances listed in Select Fixture Impedance. Select Marker to measure the fixture Z0 manually using the Z marker. The Set At field shows the X-axis position of the Z marker in ns. The Set Marker At Default returns the Z marker and fixture impedance to the default value.

Measured Fixture Z0 - Shows the measured fixture impedance. When Marker is selected as the Impedance Method, this field displays the current Z marker value or a value can be entered directly in this field. When Auto is selected as the Impedance Method, this field displays the fixture impedance value that was found automatically after clicking the Iterate button. In Auto, moving the Z marker will not change this value.

Diff: ZA: - Enter the AFR fixture impedance manually using the text box.

Fixture Length Editing Mode

Select Fixture Impedance - Select the A or B fixture.

Set Marker Position, Gate Position, or Fixture Length depending on the selected Editing Mode. Move the marker (M1, Gate flag, or diamond) to the desired response on the trace.

Set At Default - PLTS chooses an appropriate length at which to set the initial impedance or length. Click to return the marker to this location.

Preview ON/off - Select Off, then ON to preview the new measurement.

Click OK when finished. When saved, the new Impedance and Length results are copied to the *.snp files.


4. Remove Fixture

Note: First choose Select correction method, then make other selections, then click Apply Correction.

Both operations can be performed, but only one at a time.

  • Turning on Fixturing/de-embedding for channels. Usually only one channel is being used on the VNA, so only one channel will appear in the choices of channels to be corrected.  

    • Select the channels to be corrected.

      • Fixturing and de-embedding will be enabled for the selected channels in the VNA. If the new fixturing UI is used, these de-embedding blocks will place to the left of any existing circuit elements. See Using Fixture Simulator for more details.

    • Advanced settings

      • Enable Extrapolation - When fixture data is loaded from a file and the frequency span of the data is not as wide as that of the channel, check this box to calculate and use linearly-extrapolated fixture data.

      • Compensate for power - When checked, test port power is increased to compensate for loss in the fixture.

  • Modifying the calset(s) used on the channels. Usually only one calset is in use on the VNA, so only one choice would be available.

    • Advanced settings

      • Enable Extrapolation - When fixture data is loaded from a file and the frequency span of the data is not as wide as that of the measurement, check this box to calculate and use linearly-extrapolated fixture data.

      • Compensate for power - When checked, test port power is increased to compensate for loss in the fixture.

      • Prompt for new calset name(s) - When cleared, when you apply AFR to a calset, the calset is overwritten. Once done, this process is NOT reversible. When checked, you are prompted to enter a new calset name and the original is preserved. The new calset is written with AFR correction.

      • Apply AFR to other selected Calsets - When checked, allows you to apply AFR correction to other calsets not currently in use on the VNA. Click Browse, then navigate to the calsets to be corrected.


5. Save Fixture

Select File format to save fixture data:

    • Touchstone (*.snp)

    • Touchstone 2 (*.ts)

    • Citifile (*.cti)

Choose port assignment for save fixture files:

The port assignments are interpreted differently when the file is opened in each program.

Choose which program software you will be using to open the saved file: PLTS, VNA, ADS.

Choose the directory and base names for the saved files:

Click Browse to navigate to a directory folder.

With a base file name:  The resulting filename will appear as follows (assuming a Touchstone format):

  • <base file name>_1.S4P - The left half of a Differential fixture.

  • <base file name>_2.S4P - The right half of a Differential fixture.

  • <base file name>_1.S2P - The left half of a single-ended fixture.

  • <base file name>_2.S2P - The right half of a single-ended fixture.

Click Save Fixture Files to save the files to the specified directory.