There are unstable measurement on frequency band boundaries (where reset pll is performed).
The root of cause is missing wait for si5351 PLL lock. Also there is minor affect from flashing LED during sweep, it leads to a small voltage deviations and as result higher measurement noise.
The bug can be noticed as a random spikes at frequency boundaries (at begin of sweep, at 300 MHz, etc).
There are unstable measurement on frequency band boundaries (where reset pll is performed).
The root of cause is missing wait for si5351 PLL lock. Also there is minor affect from flashing LED during sweep, it leads to a small voltage deviations and as result higher measurement noise.
The bug can be noticed as a random spikes at frequency boundaries (at begin of sweep, at 300 MHz, etc).
Both issues are fixed in this changeset: https://github.com/ttrftech/NanoVNA/pull/48/commits/23dba7d9b4dce23889ec25e39c922cc9973daf82