hi @hebigt!
Having recently done this on a multi-material syringe extruder, I can confirm that M585 is a method that works. Originally, we started down the path with optical sensors but couldn't find something that fit the budget (as cheap and quick as possible).
Our solution:
- Use a USB microscope , similar to TAMV in that it is in the upright position, and hook it up to a PC.
- Take the usb feed and overlay a crosshair on it.
- Using a macro, step over to the center of T0's needle and store X0, Y0 using G10
- Macro should move the nominal amount between needles (calculate in CAD or eyeball it so T1 is in the camera's view)
- Step into the center of T1's needle and store that with G10 as well.
I've seen better than 20um accuracy with this method. If the needle isn't straight, either bend it, reseat it on the luer, or toss it for another one.
For Z, as mentioned by @Phaedrux, we used a simple block of aluminum and alligator clip to the needle to make a conductive sensor. Do the Z before the XY tool offset to ensure both tools are on the same plane.
Happy to dive in a little deeper but unfortunately can't share pictures of our setup.