Single Sideband Phase Noise

This test applies to the following models:

This manual test uses the N5511A Phase Noise Test System (PNTS) to verify that the M9484C VXG signal generator meets the single-sideband phase noise specifications.

Test Method and Setups

The cross-correlation measurement method is used to test phase noise on the M9484C. The frequency test points being tested will determine the setup and test equipment required.

Using the cross-correlation method, the test system may achieve measurement sensitivity greater than the phase noise specifications of the reference sources as well as the single channel residual phase noise specification of the N5511A Phase Noise Test System (as specified in their respective datasheets). For an explanation as to why this can occur, Closedclick here.

The N5511A PNTS, with or without external mixers, uses two independent phase detector modules with a data processing technique called cross-spectral averaging (colloquially known as “cross-correlation”) that enables the system to remove any uncorrelated noise (noise not common to both phase detector “channel” inputs) at the rate of 5 dB for every 10× increase in correlations (this relationship is known as correlation gain). The DUT signal is split to provide the RF input signal to each of the phase detector modules and two independent reference sources are used to provide the LO references.

Because there are two fully independent phase detector modules and two fully independent reference sources that produce totally uncorrelated noise with respect to each other, the system can remove this uncorrelated noise with cross-correlation. With sufficient cross-correlations, the system can achieve measurement sensitivity greater than the phase noise specifications of the reference sources as well as the single channel residual phase noise specification of the N5511A Phase Noise Test System (as specified in their respective datasheets).

The phase noise of the reference sources and the PNTS single channel residual noise floor are initial noise floors that will be reduced by cross-correlations until the phase noise of the DUT—which is correlated in both channels due to the RF power splitter—is reached.

Required Test Equipment

Test Equipment

Recommended Model

Alternate Model

Phase noise measurement system

N5511A Options 540, CH2

None

Phase noise reference source 1

E8257D1 Options 5xx2, UNT, UNY,1EU

None

Phase noise reference source 2

E8257D1 Options 5xx2, UNT, UNY,1EU

None

Low frequency power divider

Mini-Circuits ZFSC-2-5-S+

None

High frequency power divider

87303C

None

mmWave power divider

  • Not required for frequency options < 532

Marki PD-0465

None

mmWave phase detector

  • 2 required

  • Not required for frequency options < 544

Marki MM1-2567LS

None
  1. The E8267D is not allowed because it does not support low phase noise mode.

  2. Select a frequency that is ≥ the maximum frequency of the DUT.

Test Procedures

TME data entry form

The values required for this test can be found on the TME data entry form. The form is generated in TME when the Single-Sideband Phase Noise test is started.

The table below explains the terms used in the test and their corresponding Closedcolumns in the data entry form.

Test Value

TME Data Entry Form Column Header

Carrier (DUT) frequency

Frequency

Carrier (DUT) power

Power (dBm)

Measured offset frequencies

Offset Frequency

Connection Setups

Click on the links below for the frequencies being tested.

 

This setup uses:

PNTS Channel 1

PNTS Channel 2

 

This setup uses:

PNTS Channel 1

PNTS Channel 2

 

For frequency options 532, 544, and 554.

This setup uses:

PNTS Channel 1

PNTS Channel 2

 

For VXGs with frequency option 544 or 554.

This setup uses:

PNTS Channel 1

PNTS Channel 2

 

Test Procedures

  1. Connect all test equipment as shown for Setup 1. There are a total of four different connection setups depending on the carrier frequencies being tested. These setup changes will be explained in steps 7, 8, and 9 in the Making the Measurement section.

  2. Connect GPIB cables to all GPIB-controlled test equipment.

  3. Preset all test equipment.

  4. Preset the UUT. From the front panel, press the green PRESET button in the upper-right corner, then select Preset.

  5. Obtain the TME data entry form. This form provides the values for the test which will be required later.

    1. Launch TME with the UUT connected to the test system.

    2. Start the Single-Sideband Phase Noise test. This will bring up the data entry form.

    3. With the form selected, press Ctrl-A, Ctrl-C on the keyboard to collect the data in the form.

    4. Open a text editor such as Microsoft Notepad and press Ctrl-V to copy the data entry form into this application.

    5. Print out this data and use it to complete this test.

  6. Configure the Phase Noise System:

    In the N5510A PNTS user interface, change the following parameters:

    1. Select the View pull-down menu.

    2. Select Display Preferences and un-check the Spurs box.

    3. Select OK to close Display Preferences.

    4. Select the System pull-down menu and select Server Hardware Connections.

    5. Select or ensure the following:

      • Test Set has Keysight M9550A selected.

      • FFT Analyzer has Keysight M9551A selected.

    6. Select Close to close Server Hardware Connections.

    7. Select the Define pull-down menu, select Measurement, and select the following settings. Note that settings in bold differ from the preset values.

Parameter

Setting

Type and Range Tab:

Measurement Type

Absolute phase noise (using a phase locked loop)

Measurement Method

Fast

Start Offset Resolution Expansion

10

Channel Setup

Dual

Delay Before Measurement Initiates

0 sec

Start Offset Frequency

0.7 Hz

Stop Offset Frequency

20 MHz (100 MHz carrier frequency)
110 MHz (all other carrier frequencies)

FFT Analyzer Minimum Number of Cross-correlations

50 (Minimum. See notes on measurement settings.)

RBW %

7.84 (256 point FFT)

FFT Overlap %

75

FFT Window Type

BlackmanHarris4

Couple Graph to Measurement Type

Checked

Uncorrected System Noise

Unchecked (N5510 firmware revision 1.00.06.06 and later)

Sources CH#1 Tab:

Apply channel #1 parameters to channel #2

Checked

Carrier Source Frequency

Frequency test point value

Carrier Source Power

+10 dBm

Detector Input Frequency

Same as carrier source frequency

Reference Source Power

≤ 1 GHz: +15 dBm
≥ 2 GHz: +10 dBm

See notes on measurement settings.

VCO Nominal Tune Constant

Set to 0.1 ppm x Carrier source frequency

For example, set to 1000 Hz/Volt for a carrier frequency of 10 GHz

See notes on measurement settings.

VCO Tune Range

1 Volt

VCO Input Resistance

50 ohms

Center Voltage

0

Cal Ch#1 Tab:

Apply channel #1 parameters to channel #2

Checked

Phase Detector Constant

Derive detector constant from measured beat note

VCO Tune Constant

Measure VCO tune constant

Verify calculated phase lock loop suppression

Unchecked

Apply anti-correlation correction

Unchecked

Block Diagram Tab:

Carrier Source

Manual

Use single reference source for both channels

Unchecked

Reference Source Channel #1

Asset name for reference source #1

Reference Source Channel #2

Asset name for reference source #2

Phase Detector

Carrier frequencies < 40 GHz: Automatic Detector Selection
Carrier frequencies ≥ 40 GHz: External Detector

Test Set Tune Voltage Output

Front Panel

Test Set Tune Voltage Destination

Reference Source

VCO Tune Mode

DC FM

Test Set Tab:

LNA Low Pass Filter

Internal — Selected

Auto — Checked

DC Block

Unchecked

LNA Gain

Auto Gain — Selected

PLL Integrator Attenuation

0 dB

Ignore out-of-lock conditions

Unchecked

Pulse Carrier

Unchecked

Graph Tab:

Title

User specific

Graph Type

Single-sideband phase noise (dBc/Hz)

X Scale minimum

1 Hz

X Scale maximum

100 MHz

Y Scale maximum

0 dBc/Hz

Y Scale minimum

–170 dBc/Hz

Normalize Trace Data to

1 Hz bandwidth

Scale trace data to a new carrier frequency of

1 times the current carrier frequency

Shift trace data by

0 dB

Trace Smoothing Amount

5

See notes on measurement settings.

Power present of input of DUT

0 dB

Making the Measurement

  1. From the UUT front panel, set the UUT frequency and power to the first value in the TME data entry form. These settings are done within the RF Output dialog of the UUT.

    Select the RF Output block for the channel being tested.

    In the dialog box, make the following settings as shown in the image below.

    1. Select Enable RF Output.
    2. Set the Frequency value as shown in the TME data entry form.
    3. Set the Power value as shown in the TME data entry form.
    4. Un-select both ALC settings. ALC must be turned OFF.
    5. In Optimizations, ensure Harmonic Filters is set to OFF.

    6. Press the Home icon to save settings and close the dialog box.

    Verify your settings in the main home screen.

  2. Select the Measure pull-down menu and select New Measurement.

  3. When the measurement completes, select the Markers icon and place markers at specified offsets to determine the measured phase noise. When specifying a marker offset, the marker will be set to the closest frequency on the graph as indicated in the following table:

    Nominal Offset

    Actual Offset

    1 Hz

    1.0245 Hz

    10 Hz

    10.058 Hz

    100 Hz

    99.838 Hz

    1 kHz

    1.0014 kHz

    10 kHz

    9.9182 kHz

    100 kHz

    100.71 kHz

    1 MHz

    1.001 MHz

    10 MHz

    9.9609 MHz

    100 MHz

    100 MHz

     

  4. Transfer the marker amplitude values to the TME data entry form for the corresponding value of Frequency (carrier source frequency) and Offset Frequency (marker offset frequency).

  5. Print out a phase noise plot.

  6. Select the Measure pull-down menu and select New Measurement.

  7. For carrier frequencies 2 GHz to 26.5 GHz, the following connection and settings changes are required:

    1. Replace the Mini-Circuits power divider with the Keysight 87303C power divider.
    2. Change the connections on the PNTS to the 1.2 GHz-to-40 GHz Inputs. Click here to go to connection image for Setup 2.

    3. Make the following change in Measurement Settings:

      In the Sources CH#1 tab, change the Reference Source Power to +10 dBm. This is to avoid issues with the N5510A software when switching to the 1.2 GHz-to-40 GHz Inputs on the PNTS.

  8. For carrier frequencies above 26.5 GHz to 40 GHz, make the following connection changes (VXGs with frequency options 532, 544 and 554):

    Replace the Keysight power divider with the Marki PD-0465 mmWave power divider. Click here to go to connection image for Setup 3.

  9. For carrier frequencies above 40 GHz, make the following connection and settings changes (VXGs with frequency options 544 and 554):

    1. Add the two Marki MM1-2567LS mmWave external detectors. Connect to the PNTS as shown in this image for Setup 4.

    2. Make the following change in Measurement Settings:

      In the Block Diagram tab, change Phase Detector to External Detector.

Measurement Settings Notes

Minimum Number of Cross-Correlations

The effect of using cross-correlations is to remove the phase noise contributions of the reference sources and reduce the measurement variance due to noise. The default value should be sufficient in most cases, but in some circumstances based on the offset and relative phase noise performance of the reference sources compared to the UUT, it may be necessary to increase the number of cross-correlations to fully eliminate the noise bias and sufficiently reduce the noise variance. Increasing the number of cross-correlations will increase the measurement time.

Reference Source Power

If necessary due to cable loss and/or source power accuracy, adjust the source power so that a signal in the range of +10 to +16 dBm is delivered to the LO port of the phase detector.

VCO Nominal Tune Constant

VCO Tune Constant × VCO Tune Range sets the phase lock loop (PLL) bandwidth. A wider bandwidth makes it easier for the PNTS to automatically establish a beat note, but at the expense of higher close-in phase noise. Reducing the VCO Tune Constant may reduce the measured phase noise at the lower offsets, but this may require you to manually establish the beat note. Conversely if the measurement performance at low offsets is acceptable but you are having difficulty establishing the beat note, you may consider increasing this value.

Trace Smoothing Amount

Trace smoothing averages the measurement over adjacent points, reducing the variance due to noise. In some cases when there is a nearby spur, smoothing will artificially increase the measured value by including the spurious responses. In these circumstances it is acceptable to reduce the smoothing amount to exclude the spur.