I’ve arrived at a point in the process of building a Raman spectrometer where I need to focus the diffracted light from the grating onto the CCD.
Naïve as I were one year ago when I started the project, I figured I would simply be able to print everything to specification and stop worrying and love the 3D-printer.
Close but no cigar.
I need to be able position the TCD1304 accurately. Ideally I would simply buy a linear stage from Thorlabs or Edmund optics, but precision is so expensive. And I’m cheap.
And afterall, how difficult can it be to make a somewhat precise linear stage? Not very difficult it would seem:
The linear stage consists of two 3D-printed parts, a 40mm M3 screw, two pieces of a bicycle spoke (broken bikes are ubiquitous here in Denmark) and a couple of springs from various ballpoint pens.¹
M3 screws have a pitch of 0.5mm and with a precision of let’s say 1/8 of a turn of the M3 screw, it should be possible to position whatever is attached to the stage to a tolerance of less than 0.1mm.
We’ll see how it actually fares.
One more Kodak moment:
Assembling requires a bit of patience, especially when fitting the spokes. The holes are purposely just exactly too small, so you have to use the spokes as a hand drill.
If you manage fitting everything you should have a linear stage with no free play and possibly some blood on your hands. The range is about ±4 mm.
Here are the openscad-files for the linear stage.
¹ I’ve instructed the students at the high school to bring me all their worn out / broken pens. They’re still wondering why I would need dozens and dozens of springs in various sizes.