Thanks Steve,
looks like I'm a bit premature with the software though: there's a bug in the Z output (actually , less of a bug and more of a whoops, forgot to actually implement the function !)
I finally found it warm enough to fit the knee scale . It was a bit of a PITA because nothing is vaguely flat on the side of the column - I had to print a couple of spherical fitting to allow the scale mounting with just two screws .
Apart from the Z count direction bug , There are a few things that need improvement :
The I2C display is quite slow , compared to the SPI used in the previous dro tests, so the update feels laggy . I'll have to see how fast I can over-=clock the I2C
The Z output really need to be handled by a PIO . ATM it is possible to move the Quill faster than the Z can output resulting in a off-putting run-on/catch-up of the Z display.