Long shot: Tight control of Z axis for a probe and microscope?

[Michael Zions]

I'm recording from neurons in a dish with an extraordinarily delicate probe. I've cracked too many of them already when I use regular manipulators; the problem is exacerbated by the oblique approach angle. I'm looking for a way to drop the probe down vertically, sort of a gantry or arm that can swing over the dish and lower the probe very precisely along the Z-axis. Think 10 micron increments or finer. (Raising the dish toward a fixed probe works fine too.) I also need to hang a low-power microscope off the side of the probe mount, so I can see the tip and target. Our current manipulators don't have the clearance or the load capacity we need, and I haven't found a substitute.

I looked at 3D printer chassis but in the end it just comes with too much we don't need and not enough of what we do. Still, that's a good example of the sort of solution I'm seeking, but it's only feasible if it's already close to workable.

Questions:

Does something like this already exist? Maybe an inspection scope for QA? I have a feeling we're just not looking in the right places.

Has someone successfully repurposed a 3D printer for a similar project? Manual control, very fine Z axis?

I can spend $4k (of someone else's money) on a solution; any advice where to look?

Thanks, and hoping for a Hail Mary pass.

[c f]

Do you only need fairly precise movement in 1 direction? (i.e. you're fine with somewhat crudely zero-ing it much higher than your target, and then slowly approaching your target with precision)

I did something similar to scan a disk head across a platter:

http://www.chrisfenton.com/cray-1-digital-archeology/

It was theoretically taking steps of 1/5200-th of an inch (~4.9 microns), and that's more or less what I got (in practice, I would imagine there was a fair amount of variation around that 4.9 microns, but either way it was very, very tiny steps). Accounting for backlash at that level requires super-precision, but doing it in one direction seems feasible. Mine is basically just a cheapo 200 steps/rev stepper motor from a makerbot thingomatic z-stage, with a little laser-cut acryllic frame built around it. I think it's using 1/8 microstepping.

-Chris

--
You received this message because you are subscribed to the Google Groups "NYCResistor:Microcontrollers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to nycresistormicrocon...@googlegroups.com.
To post to this group, send email to nycresistormi...@googlegroups.com.
Visit this group at http://groups.google.com/group/nycresistormicrocontrollers.
For more options, visit https://groups.google.com/d/optout.