The method you propose sounds good and I would like to hear how it works out.
Since I first came across piezo bed leveling, I have experimented with various methods of utilizing a piezo sensor attached to the hotend. So far, I have not come across a method that I am completely satisfied with. I have tested mountings similar to those used on the PZ Probe, but have consistently encountered issues with artifacts from filament extrusion pressure and unexpected "popcorn" noises from the hotend.
Currently, I am employing a single underbed sensor to measure the Z-offset and detect any debris on the nozzle. As for the bed mesh, I am utilizing a piezo touch sensor, but I am also investigating the use of an acoustic range finder for conducting a fast multipoint bed scan.
Mike