I have decided to keep roughly the same form factor for the State Definition Table input area as I did with the TMD-2 machine.  Mostly that means that I am going to keep the same tile size (but not necessarily the same tile thickness).  The tiles have a good "feel", but I fear that if they were any smaller (in the interest of making a more compact machine) that would not be the case. 

So I designed and printed just the tile input area for a single state to test with.  On the left above you can see the top of this panel with a few TMD-2 tiles for reference. On the right the back of the panel is shown with indentations to accept the SS49E Linear Hall Effect Sensors which will be protruding from the state PCB (which is exactly the same size as this panel). 

I'm hoping that enforcing this precise alignment will ensure that I get consistent measurements from each of the sensors. Only testing will tell. It would be inconvenient but not too terrible if I had to calibrate the sensors for each panel to get good results. We'll see.