On Tue, Jun 27, 2023 at 05:54 AM, Mike wrote:
I compared both methods of calibration, which I refer to as "screws" and
"crocs".
At around 72kHz, where the reactance is approximately 50R, both calibration
methods give an inductance of 108uH. However, the SRF was 7.4MHz for "screws"
and 12.9MHz for "crocs".
I then repeated the measurement with a different VNA, an FA-VA5. This gave
L=110.7uH for "screws" and 110.5uH for "crocs". The SRF was 6.38MHz for
"screws" and 6.45MHz for "crocs".
I can accept the difference in L between the two devices but can't decide
which one is giving me the most accurate SRF.
With a 110 uH coil it takes 4.2 pF to resonate at 7.4 MHz. and 1.4 pF to resonate at 12.9 MHz. That is a difference of 2.8 pF which is not much.
One way to get to the bottom of this is to short the terminals of the coil together and then "grid dip" it like the old days. You can do this with the NanoVNA by making a pickup coil at the end of a short connecting cable. Calibrate over 5 to 15 MHz. in the usual manner. Set the trace to S11 Log (Return Loss setting) Then attach the pickup coil to the NanoVNA and place the 110 uh coil and the pickup coil in parallel a short distance apart. Look for a dip in the trace. Keep moving the coil further away until the dip has just about disappeared. Now move a marker to the dip and the frequency should be very close to the SRF.
Roger
Roger Need