Which means you need to connect them up. I think the idea here is that you don’t need any sort of sensor.
That said, depending on the surface, the ambient lighting conditions and the type of light sensor, your sensors may not work well in all conditions. eg: If Opto’s are mounted so the sensor can be exposed to light, and you get a stray sunbeam, the sensor may not trigger when the opto flag tries to interrupt it. I’ve seen this happen a few times with optos, leading to the machine crashing the opto flag through the opto and beyond. Same sort of thing could happen with an edge sensor if it is light based.
They are using almost the same stepper drivers as on my arm board.
I experimented with the stall detection more then a year ago. For X and Y it works as a replacement for the homing switches.
The stall detection resolution is one full step.
Detection for Z is difficult. I could detect the nozzle touching the build plate. But not accurate enough.
Maybe with some experimentation and a lot of testing it will work for Z.
In theory it can all be done on an atmel. But these drivers go up to 256 microsteps. It makes the printer very quiet. But an atmel can not generate the step frequency needed for 64, 128 or 256 microsteps.
@NathanielStenzel that’s what I was thinking(I watched vid. w/no sound so I wasn’t going to say anything). Bouncing a laser off of the mirrored bed to find the edges.
You could do this with Z if you home to MAX instead of MIN, or even just put some sort of adjustable hard stop in the MIN position. If you make it equivalent to a Z endstop rather than trying to make it equivalent to a Z probe, using stall detection for Z shouldn’t be a problem.