no amount of averaging is going to help
My nanoVNA is of the version identified by Hugen as "worse",
and I intend to run more experiments about improving e.g. CH1 isolation.
1) my initial attempts reinforce the proposition that averaging is doomed.
2) my sample tends to have a strong, but not constant,
CH1 logmag peak @ 450MHz (screen shot attached).
3) running from battery improves CH0 logmag, but not appreciably CH1
4) calibrated with 35cm RG316 BNC leads,
CH0 logmag is sensitive to placement, while CH1 is not..
Also attached are scripts (Python, gnuplot and bash glue) for data collection
with edy555 0.3.1 firmware using 'scan' (scanraw is needed for QRP's 0.4.2)
Oristo
