I am wondering what is the impact of the magnetic field generated by the BLDC windings being energized on the rotor position measurements with the Hall sensor. I found some unclear explains stating that that impact is irrelevant comparing to the impact of the rotor magnet's field. But why? Aren't the Hall sensors oriented orthogonally to the direction of magnetic field created by the windings?
Depending on where the Hall sensors are placed it could have an effect. Most of the coil's magnetic field is concentrated in the core, but some 'leakage' will occur away from the pole face. I tested a 12V CDROM motor similar to the one below. The normal Hall signal due to rotor magnet movement was 200mV pk-pk. I then passed +-0.25A through the phase next to one sensor, and it registered 26mV pk-pk.
High power motors generally have the Hall sensors placed at the end the rotor away from the stator coils, or even activated by a separate set of magnets on the shaft. With these arrangements interference should be negligible.