sabato 14 novembre 2020

DIY RF Network Analyser with a tracking generator from two RTL-SDRs, a PLL synth and a VSWR bridge

Shared Link: https://github.com/CGrassin/rf_network_analyzer

The repository contains files and instructions to build a crude $50 RF Network Analyser (NA) with a tracking generator from two RTL-SDRs, a PLL synth and a VSWR bridge. This device is very simple to put together (no soldering required) and gives quite reliable measurements for custom antennas or filter design, for instance.

It is controlled by a Python script through a command-line interface. It has a dynamic range of about 50dB and a bandwidth of 35 to 1700MHz.

Note: using a noise source instead of the signal generator is a simpler way of getting those measurements. The only drawback is the reduction of the dynamic range of the SDR due to the amount of noise that may overwhelm the receivers.

 
Material

This is the full BOM. Prices are from AliExpress:

Quantity Item Estimated cost/u
1 Arduino Nano $2
1 Bidirectionnal level shifter $1
1 ADF4351 evaluation board $18
1 VSWR bridge $14
1 SMA 50Ω load $5
4 SMA cable $2
1 SMA female-female transition $2
2 RTL-SDR (TCXO/SMA version) $15
 
Build instructions

1. Wire the electronics as shown on the diagram: 

NA circuit diagram

 

2. Flash the "Arduino_Synth" program on the Arduino Nano.

3. Connect both SDRs and the Arduino to the host computer.

 
Usage
  1. Disconnect both DUT ports, and launch an "open" measurement for calibration: python3 NA.py > open.csv
  2. Connect both DUT ports together with the female-female transition, and launch an "thru" measurement for calibration: python3 NA.py > thru.csv
  3. Connect the DUT and perform the measurement: python3 NA.py -o open.csv -t thru.csv | tee output.csv
  4. You can plot the output csv file with LibreOffice Calc or Octave, for instance.

For single port devices (such as antennas), only the "open" calibration is necessary.

For help on the parameters, use python3 NA.py -h

 
Sample output

I used this device to measure my hairpin filters (http://charleslabs.fr/en/project-Hairpin+filter+design). This is a typical output plot:

Hairpin filter S-parameter plot















TAG: #GiammaIoT @GiammaIoT #SoftwareDefinedRadio #SoftwareDefinedRadio #PLL #VSWRBridge #ADF4351 #VSWR #ArduinoNano #Arduino

Nessun commento:

Posta un commento