Stepper resolution vs ball screw accuracy.

So I've narrowed down my z axis to two choices for ball screw:

1) 2 mm lead, c5 accuracy with a 0.005 mm backlash. This screw gives 10 micron full step layers (or 5 micron with 0.9 steppers), and 330-340 mm of usable z travel.

2) 5 mm lead, c3 accuracy with zero backlash. This screw gives 25 micron full step layers (or 12.5 micron with 0.9 steppers), and 315-325 mm of usable z travel.

Both are basically the same price. Which would you choose?

Would the 5 micron backlash be worse than the lower stepper resolution?

For fun, lets add a new third candidate:

3) 2 mm lead, c3 accuracy with zero backlash, 10 micron full step, 280 mm usable travel.

Backlash typically doesn't matter on the Z axis because of the weight of the bed. I'd go for 2mm pitch.

With ball screws i have heard that there is a risk that with power off, the bed will fall under its own weight. You can work out whether this is a risk if you know the detent torque of the motor. The risk is lower with finer pitch.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

My thoughts about the backlash were the same, and so I originally designed my printer around ball screw no 1. I'm starting to second guess myself, but 5 microns of error in Z is really not going to be noticeable anyway, even if the weight of the bed didn't cancel it out, is it? What effect on the extrusion width would 5 microns of vertical error have?

I mean, it gives me the best stepper resolution AND the most build height. Unless anyone can give a compelling reason, I'll probably stick to #1 and hope the minuscule backlash is irrelevant.

Oh, and as for the bed dropping, I haven't accounted for that yet. I probably, won't to be honest. On my delta, since I put the extruder on the effector, it has tendency to drop if the motors cut out, so I'll just keep the motors energized after a print like I do currently on my delta. It might also be worth putting some shock dampers at the bottom.

Backlash could be an issue depending on where the weight of your bed is, if its center of mass is not at the center, then the lighter side of the bed will fall and rise with the backlash. I had this issue on my cartesian with the X axis, as the X axis motor caused it to be heavier on one side and my original brass nuts were not backlash free. I switched to anti-backlash delrin nuts and I have perfect layer registration now.

I assume that most of the weight will be balanced in a build plate, but its worth considering. As for resolution I don't see the need to go past 50 microns. So, based on that go for the longest with least backlash that you can afford.