Question about making the parts (cover, base plate, etc.) & analyzing soil

[Tin314159]

Hi all!

I am a member of a university club that is trying to build a Raman spectrometer to analyze soil samples. The part list on the site is very comprehensive and we really appreciate all the information on the site. Below are some of the questions that we have come up with:

For 3d printing material:
- what is the reason for having all the parts black dyed? Instead of anodizing, could we just paint the part (like using acrylic paint)?
- for parts such as the cover, do you think we could use PLA instead of SLS & FMJ
For analyzing:
- if we want to analyze soil, would you recommend to analyze it in liquid form or in powder form?
- is there any recommendation for how to analyze the spectrum for soil (since soil is a complicated mixture)?

Thanks in advance!

[andy]

Here's my guesses at answers though it might be wrong:

• You don't want stray light, black helps absorb it. Inside the cuvette holder there is a mini beam dump structure. Whatever material you use try to avoid one that glows (floresces)
• PLA is more brittle and less precise. I think there are some photos in the photos thread of someone that successfully used it. I'm not sure if tapping the threads would be harder
• From my experience the solids show up better due to the laser power all being used and not transmitted through clear solutions but if its very concentrated (opaque) vs dried first I'm not sure how much it would matter
• I'm not sure if soil has anything special. But in general the technique for measuring something specific in a mixed solution is described here: https://www.jasco-global.com/solutions/ ... ctrometer/

[Luc]

Spray painting is okay provided you can do it uniformly & where it matters. In fact, I often spray some of my SLS parts because the dye process of Materialize isn't black enough to my taste.

That being said, I have two objections regarding FDM printing in optics. It can be used (we do it often at the office) but:

• it's not as effective in terms of straylight. FDM prints usually have that shiny specular reflection due to the molten extruded filament. It's possible to get something less specular but at the cost of heavy post-processing. Technology such as SLS are much superior in that regards because you get something that's really matte. And even if the part is more "dark gray" than black, it cuts straylight much more efficiently because of that diffusive effect.
• it's usually weaker for structural aspects and tend to catch a lot of vibrations which is something you really want to avoid in optics. For sensitive stuff like interferometry, you won't even get fringes with a FDM baseplate or brackets (I tried long ago...). Note that for such an application I wouldn't necessarily recommend SLS either but at least aluminum 6061 or even stainless steel. For spectroscopy, I'm a bit reluctant to FDM but an user pointed out that he got results using a FDM baseplate -- so it's possible.

It's also a question of what you're trying to achieve at the end. For educational, FDM is fine. For reliable operation, take the SLS, or better Aluminum, way. My prototype baseplate is in SLS and stayed aligned for more than 1 year; it also still look like it has just been printed and it's already ~3 yr old!