Paul,
I can't tell from the Smith chart plots you have included what might be wrong What does the 68 pf capacitor show at around 1-10 MHz? At low frequencies, like 50 kHz, the Smith chart should show an open - a dot on the right side of the Smith chart. The nanoVNA is measuring the reflected signal and phase or complex values of those quantities. At very low frequencies, like 50 kHz, the capacitor is so much like an open circuit, it is very difficult to measure with a VNA. The reactance of 68 pf is around 47 k-ohms at 50 kHz and therefore hard to measure in a system calibrated for 50 ohms. However, at 1 MHz, the 68 pf capacitor has a reactance of ~2.34 k ohms and at 10 MHz it is ~234 ohms. So, measurements at 10 MHz with the nanoVNA are very possible.
Attached is a Smith Chart plot of my measurement of a capacitor marked as 68 pf. I did not do a careful calibration for the test fixture I am using; however, this shows what I think is a correct trend for a 68 pf capacitor from 50 kHz to 30 MHz. As you go higher in frequency than 30 MHz a good capacitor should plot a curve which goes further around the Smith chart than shown in the attached image in a counter-clockwise direction. When nanoVNA-Saver estimates the capacitance at 10 MHz, it is saying it is ~74 pf. Some of this error is likely due to my test fixture; however, I don't have a well calibrated measurement of this unused capacitor.
I hope this helps. Good luck with your measurements.
--
Bryan, WA5VAH
