Has anyone experimented with putting a piece of conductive foil on the corner of the print bed and wiring up the extruder nozzle so that it triggers the Z-axis home when the nozzle hits the foil? I fear that the nozzle would have enough plastic on it to keep it from making reliable contact.
You could use a threaded standoff that contacts a higher point on the nozzle, and just calibrate it by threading it in and locking it in place, to get much the same effect. In the CNC machining world, people often use microswitches, as well.
Implementation of a microswitch as a tool length probe, including repeatability data: http://softsolder.com/2010/12/06/improved-tool-length-probe-switch/
I’ve often wondered about trying something like this:
Where I’ve seen people try this, whether conductive or pressure, ooze from the nozzle has been a significant problem for anything better than 0.5mm height accuracy. Even hot the plastic causes too much a gap, and requiring the nozzle to apply pressure to move the plastic causes you to be off by in the other direction due to machine flex.
Yes, the plastic needs to be removed to get reliable contact between the extruder nozzle and the foil.
Per @John_Bump 's suggestion, you would need one in each corner to get the orientation of the bed perpendicular to the Z axis.
See http://derrickoswald.blogspot.com/2011/11/z-axis-calibration-movie.html
Rather than use the nozzle couldn’t you have a little pillar machined into the aluminum that sat flush to the tip. It’s offset could be calculated and the foily could be put where it needs to be.