Hall effect endstop - sensing direction of sensor ICs is in the wrong direction

[arkeet]

The SS39ET specified in the BOM senses magnetic fields perpendicular to the board. But when used in a hall endstop on VT/V2, the axially magnetized magnets point in directions parallel to the board, so they are unable to sense correctly.

So some users have had to mess with the magnet position to get the Y endstop to work, example image. I assume this works because it tilts the magnetic field to have a significant component perpendicular to the board.

I think it can be solved by using diametrically magnetized magnets, reverting to through-hole sensors, or moving the magnet position on the printer, but none of these are great solutions. After a quick look I haven't been able to find a SOT23 hall sensor that senses in a direction parallel to the board.

[wlhlm]

Previously reported here: #83