On Sat, Apr 26, 2025 at 02:06 PM, Dean W8ZF wrote:
I'm a little surprised that nanovna-saver depends on the H4's calibration
state
They did that to avoid some problems that require extra coding. And since you can do one calibration from 50kHz to 900MHz, and save it in slot 0, you never have to revisit it. The problem is that nanovna can output S11/S21 values outside the range 0<=x<=1 if it is not calibrated - which causes arithmetic errors with any normal S11/S21 processing algorithm. And if you calibrate it first, the nanovna does not do that (except maybe under rare conditions). And it doesn't matter so much what the nanovna calibration is - it just needs to be there and be consistent - because the 'working' calibration is that done by the nanovna-saver algorithm.
