Hi, for a specific application I would like to increase the Z travel range, to be able to focus on a microfluidic structure ~ 3 mm above the usual focal plane, i.e. move forth and back from the slide.
Seemingly, I can only get ~2.5 mm of Z travel with the objective - is there a way, e.g. to somewhere add more clearance for the actuator arm, so it gets stuck at a later point? A single mm or two of additional travel would already help me. When I move the objective directly up and down by hand, I have the feeling that a little more travel should actually be possible.
Alternatively, is there a parameter in OpenSCAD to tweak the construction of the Z axis, without affecting the other parts of the microscope too much?
There might be a parameter for the Z axis, but it probably needs to be adjusted in conjunction with various other things - there are quite a few constraints that would matter. In particular, the limit of travel is set based on the length of the lever in the Z axis - if we increas it without a longer lever, we’d probably damage the flexures, as they will go beyond their elastic limit.
IIRC, the lever in there is about 20mm, which ought to give you +/-2mm from the “neutral” position. If that range is enough but you just need to shift it up, perhaps it’s sufficient to loosen the optics module mounting bolt and slide it up. If you’ve already hit the top, perhaps the easiest modification would be to move the mounting screw down on the optics module - that will shift the objective up slightly, and might enable you to do what you need to.
Hm, I only got ~2.5 mm travel in total, as measured with a digital dial gauge. If it was 4 mm, I would probably be fine already.
At one end, the screw/nut becomes loose (motor turning but no z motion), at the other end, the lever gets mechanically stuck, I suppose. I am not sure where exactly that would be, i.e. whether it would be sufficient to just add a little more clearance where it comes into contact with the microscope body.
It is not so easy to see how the z-axis moves as it is all inside the microscope, and I have not looked at the range of travel. I am not able to get to the lab for a while now, but I would suggest printing the separate z-axis that is for the upright and operating manually. Then you can move faster and see more inside and check what is hitting where.
One thing that has just occurred to me is stringing inside the z-axis. The roof above the z-axis in v7.0.0beta1 has unsupported bridges. Depending how that printed, it might leave strands that block motion in one direction. Is the motion equal up and down from the nominal neutral position? neutral is as printed with the base of the z-axis optics mount level with the base of the microscope. I thought this was just a cosmetic problem, it has been addressed anyway in Z axis printability enhancements (!335) · GitLab. The current master build has the changes incorporated (thread Openflexure current build is here! - Announcements - OpenFlexure Forum)
The other common issue is the O ring not providing enough pulling force downwards - if your travel is limited at the bottom (i.e. when the screw becomes loose) it suggests the viton band is not tight enough (e.g. it’s too weak - there’s no tension adjustment).
Thanks for you input, I’ll check! At the moment I have those: O-Ring, 30x2 mm, FKM (80A)
Perhaps I’ll order stronger rings to try, there are also 30x2.5 or 30x3 available, in NBR 70A and, FKM 80A and EDPM 70A, and have a closer look at the other z-axes I already have available, including a single one I printed earlier. Will get back with my findings…
Thanks - I have no idea if fatter o rings will fit, but may be worth a try. We’ve seen variation between suppliers, but (despite considerable research by Julian) it was tricky to provide a tighter spec than we already have. The main issue is these O rings are rarely specced for use as tension bands.
One of these days, someone will find time to replace the bands with springs. That could solve a lot of problems, but it never seems to be the top priority as the bands generally seem to do a decent job, and springs are never quite as easy as I hope.