Calibrating the AVNA1 Audio Vector Network Analyzer

Initial setup for the Teensy 3.6 based analyzer that measures impedance and transmission from 10 Hz to 40 kHz.

This page covers the calibration steps that are needed at initial setup of the AVNA1. The main page for this project is linked here. Note that as of version 0.61 of the AVNA .ISO program, there is an alternative procedure to what is on this page. It uses the improved TUNEUP command, and the details are at the TUNEUP page. It remains to be seen as to which procedure is best for a particular circumstance, and by use of the PARAM1 and PARAM2 commands, the two alternatives can be mixed.

Calibration Procedure - Start by plugging a USB from the AVNA1 into whatever PC you have the Arduino/Teensyduino IDE running in. This will power up the AVNA. If you go to Tools/Serial Monitor you should see a sign-on as

  Loading EEPROM data
  VNA-LF Ver 0.58
  Voltage Check (expect 145 to 175): 163

Your voltage will vary, but should be in the range shown.

PARAM1 steps - For this calibration activity, we will not use the touch screen. All control and data output will happen with the Serial Monitor. Inputs are typed on the top line, terminated with the Enter key. We will start with the values of the standard 50 and 5000 Ohm resistor. The default resistor values are 50.00 and 5000.0 Ohms. This can be seen by typing "PARAM1 <Enter>" that will return your typed line followed by the current standard values, as

  PARAM1
  Currrent PARAM1 values:
    50 Ohm reference  = 50.00
    5K Ohm reference  = 5000.00

If you know these are wrong, and an Ohmmeter tells you the correct values are 49.9 and 5123 Ohms, you would type "PARAM1 0 49.9 5123<Enter>" and get the response of the same line you entered. Note the '0' after 'PARAM1'. To check the values, type "PARAM1<Enter>" again and see

  PARAM1
  Currrent PARAM1 values:
    50 Ohm reference  = 49.90
    5K Ohm reference  = 5123.00

To restore the default values, type "PARAM1 99<Enter> or re-enter them as we did above."

PARAM2 steps - Now, we will see how to change the parasitic impedances for compensation of impedance measurements. The model, looking into the "Impedance" port, is a capacitor and resistor to ground (Parasitic capacity and R1 in the schematic). These are nominally 41.5 pF and 1.0 Megohm. In addition at the lower frequencies, the series capacitor, C1 (0.22 uF) is seen. Then there is some lead inductance and resistance. All of these components are modelled in the AVNA1 software and de-embedded to greatly increase the range of accurate measurements. The command for entering these values is

             "PARAM2 41.5  1000000.0    0.22    0.07   20.0"
 With units           pF      Ohm        uF      Ohm    nH

In first calibrating the instrument, it is quite adequate to just compensate for the the shunt capacity. The PARAM2 command has provision for a short line and one can enter only "PARAM2 39<Enter>" and that will then compensate for 39 pF

For example, to search out the correct capacity, we will measure the open circuit impedance at the highest frequency, 40 kHz, and using the high reference resistor. The commands are as follows. Set the impedance measurement mode with 5000 Ohm reference by "Z 5000<Enter>". The frequency is set by, "F 40000<Enter>". standard calibration (it compensates for differences in gain and phase between the measurement and reference channels). The command is "C<Enter>". Next we command a single impedance measurement with "R 1<Enter>". My particular unit, with its lead lengths and lead routing, returned

R 1
40000.000 Hz Series RX: R=1725281.927 X=-2754117.271 C=  1.4pF Q=1.60
40000.000 Hz Parallel GB: G=0.000000163 B=0.000000261 R= 6121758.71 C=  1.0pF Q=1.60

We are most interested in the second line that corresponds to a parallel resistance and capacitance (or inductance, if applicable). This shows a bit over 6 Megohms an 1.0 pF. This is moderate de-embedding of the 1 Megohm, and the capacity shows as 1 pF, which is pretty much the desired answer. This is the value that was set a year ago to remove the stray capacity. In order to demonstrate the process, though, we will change the 41.5 pF to 39 pF. We enter "PARAM2 39<Enter>" and then enter "R 1<Enter>" to again measure impedance. The result is

R 1
40000.000 Hz Series RX: R=203948.388 X=-1085725.094 C=  3.7pF Q=5.32
40000.000 Hz Parallel GB: G=0.000000167 B=0.000000890 R= 5983837.06 C=  3.5pF Q=5.32

The 2.5 pF shift in compensation does what we would expect, in that it shows 2.5 pF more, or 3.5 pF. Don't expect that this measurement will hit this 0.1 pF accuracy, but it should be close. We won't show the re-correction here by going back to 41.5 pF, as it is fairly obvious.

The PARAM1 and PARAM2 values are automatically saved in EEPROM and will be used for any follow-on measuremenst.

Summary and Notes - These two settings, PARAM1 and PARAM2, are used here to correct, in software, for the calibration resistor values and the shunt input capacity. The other components listed for PARAM2 can be used to improve the accuracy of the measurements. The general procedure is:

The value for R1 will not generally be the exact DC value. But it will be in that range. Adjust at 1000 Hz so that the measure parallel resistance is as high as possible.
The input capacitor value, around 0.22 uF, should be adjusted at 20 Hz using a 5K Ohm reference. The value should be adjusted so that the series input impedance is resistive .
The series resistance (nominally 0.07 Ohms)should be adjusted at 500 Hz with a 50 Ohm reference. It should be adjusted with a sold short across the impedance measuring terminals. The value should be adjusted to have 0.0 Ohms series impedance.
The series lead inductance should be measured at 40000 Hz and a 50 Ohm reference. With the terminals shorted, adjust the series reactance to about zero.

Bob W7PUA

Back to AVNA Home Page