Public Lab Research note


Failed 3d print of mobile spectrometer attachment

by warren | July 18, 2012 01:23 18 Jul 01:23 | #2847 | #2847

After the excitement over the mobile (Android) spectrometer prototype, my nice 3d model printed terribly at Shapeways. It was also covered in a kind of black powder rather than being actually printed in black. I'm going back to PVC (see below) for now but would like to try to improve my 3d printing as well if folks have suggestions.

In the meantime, I cut up the 3d printed parts and reassembled/glued them back together as a workable device which leaves your fingers a little sooty.

Phone spectrometer prototype

Update: Here's a link to the Sketchup file I used: https://publiclab.org/sites/default/files/plots-spectrometer.skp

Files Size Uploaded
PLOTSSpectrometer.scad 1.41 KB 2012-08-07 05:15:03 +0000
PLOTSSpectrometer.scad 1.41 KB 2012-08-07 05:12:27 +0000

25 Comments

Too funny! But that's too bad about the materials. When you mention that it was terrible, was it the precision of the shape low or any other problems other than the surface of the material?

Is this a question? Click here to post it to the Questions page.

Reply to this comment...


Sorry, i should've been more specific. It printed each surface, instead of treating my model like a volumetric shape. So the multiple square "tubes" above are actually the inside and outside surfaces of my model, printed as separate parts. The other photo shows that the recessed area to hold a DVD-R slice and attach to a phone is filled in completely solid, rendering it pretty useless. I cut it out and used the flange alone.

Reply to this comment...


The Shapeways website has a bunch of tutorials for perfecting your model all of which seem very complicated. I'm thinking the paper cutout idea might be the way to go.

Reply to this comment...


the first print is rarely right-- I remember buying an expensive stereolithography part several years ago and having it arrive as a thin surface sloshing full of unpolymerized white goop (that I'd paid for!).

have you thought about posting it to Thingaverse.com and asking for help there?

Is this a question? Click here to post it to the Questions page.

Reply to this comment...


This smartphone-based spectrometer is such a great idea. Comments / questions: -- Thingiverse. I second Mathew's suggestion re: thingiverse -- t'd be great to post the design there, and then add some comment as to the problems you had. You'd probably get several people with experience designing for e.g. Shapeways interested enough to post their own optimized verisions on thingiverse. -- Other 3D printing options: Ben Gamari here at Umass has a RepRap [located in his living room, with his bike handlebars serving as a spool for the plastic coil] -- we'll play around with the sketchup file and see if we can generate a nicely-printed version of your design here. -- Paper cutout. I also second Jessi's vote for a parallel verison of the spectrometer design based on a paper cutout -- if such a design would be good enough for several major use-cases, it'd be super-accessible, and easy for anyone to modify. (Shouldn't there be an "orgiami" version of thingiverse?) -- Lens focal length. How cheap are lens attachements for smart phones? I seem to remember seeing "microscopy" lens attachments for smart phones out there. If it's cheap to change the focal length this way, maybe the same lens that would useful for a shorter spectrometer tube could also do double duty as a macro lens, for e.g. a PLOTS microscope tool? Caveat: I flunked optics. -- Mass production of plastic spectrometer. If you're thinking of turning this into a mold in order to have it mass produced (perhaps funding the mold production (expensive) via kickstarter), are there other design constraints to take into account? I'll ask around ... -- Boot2Gecko. Mozilla is supposed be coming out with their own smartphone OS that should be available for Android-compatible devices soon -- all of the apps will be HTML5-ish "web pages", with some API that grants them access to the phone's camera, USB, etc. -- and all the code would be viewable immediately in each app by clicking "View Source" -- so, truly open source. If it takes off, such an OS would really open up development of smartphone-based scientific intrumentation via the browser. I'll see how far along the project is, and whether it makes sense to start developing an app in parallel for that platform ...

REALLY COOL STUFF! -- Don

Is this a question? Click here to post it to the Questions page.

Reply to this comment...


Doing the thingiverse thing now: http://www.thingiverse.com/thing:28083

cool... i wonder if people will actually refine it.

The PVC one is really working well; i made it point upwards rather than sideways, and could def. see a foldable one working. Also putting a tube of black paper in as a liner helped keep light leakage/reflection down.

I'm very tempted to try to make one which can be laser cut from either thick black cardstock or a foldable rigid black plastic sheet.

Reply to this comment...


I'm working on a fix.

Reply to this comment...


Awesome. That was so fast. I take it by your handle that you're maybe interested in using this thing!

BTW I approved your comments but if you register on the site you won't be moderated.

Thanks!

Reply to this comment...


Awesome suggestions, Don!

I share co-working space with a bunch of the Mozilla guys working on MozillaOS (they just ditched the name Boot2Gecko), and their demo hardware is mighty impressive-- it looks like they'll have a foreign launch first, but next spring some time we should be able to start playing with their phones. I hope they succeed because an html5 based system does sound like the bees knees.

Jeff and I have a conversation elsewhere where we were talking about laser cut black styrene sheeting as a mid-level between card stock and 3D printing. just cross posting for the record

Reply to this comment...


Oh, sweet, thanks for the styrene link. MozillaOS should work fine -- however the web based version is fully functional (i'm adding offline caching so it can be used even without a web connection) and i expect as browsers catch up to Opera in implementing WebRTC features, it'll work with stock browsers as well. Luckily that means that when MozillaOS is ready, it should "just work" but we don't have to wait for it to get started.

Reply to this comment...


Mathew, do you have an intuition of which thickness is a good folding thickness, or should i just order a few and try it out?

Is this a question? Click here to post it to the Questions page.

Reply to this comment...


I think the 1/16" stuff-- .060", 1.5mm is pretty stiff but workable-- I'd check out the .040 stuff too, because its half the price. I think the biggest issue is light transmission, but I've only worked with the white stuff, so I can't speak for the black sheet.

if you score the interior of a fold, styrene sheets fold great. it might be possible to lightly zap the surface with a laser to "score".

Reply to this comment...


I just updated my version of the 3D printable spectrometer: http://www.thingiverse.com/thing:28100

I changed the mounting angle to 32 degrees to more closely match the different gratings I have on hand. The 45 degree mount tends to throw the spectrum off to one side making it difficult to get a good photo.

In the end, while it does seem to work very well, the paper, PVC, or styrene versions will likely be more accessible to the majority of users. At least, until 3D printers become more popular.

Reply to this comment...


Wow, that is super cool! A few things i was excited to notice --

so extrusion printing seems not to yield precise "slits" so you cut it with a razor saw? that's very good to know. For the PVC one I used a band saw and it work surprisingly well.

I was wondering about the mounting angle and expect that people may want it to be parametric. I was thinking of making a spectrometer on a kind of swivel joint so that you could collapse it down when not using it, but this could also let you choose an ideal angle. I'll upload a sketch soon.

I love the idea of nubs with rubber bands. I was thinking of adding an elongated tab to keep the device rigid and stop it from wobbling which throws off the calibration. I'll include this in my sketch.

I was also wondering if you'd be interested in reposting your Thingiverse post on the Public Lab site, mainly because then you'd show up in the "contributors" list on the spectrometer page, which seems proper and fair.

Thanks Brad, awesome contribution, and as always thank you for respecting the license and releasing under the CERN OHL!

Is this a question? Click here to post it to the Questions page.

Reply to this comment...


Thanks Warren. This is actually the second spectrometer I have built. The first was part of my thesis and used 1.5" PVC and a hacked point-and-shoot camera. I have another one in the works that is controlled by an Arduino and uses the sensor out of a barcode scanner.

Some printers can do a very good job at printing slits but even the good ones tend to vary a bit which can make the spectral lines fuzzy. Cutting with a saw isn't too bad but the slit is still a bit wider than I like. I found that gluing a pair of single-edged razor blades such that their edges are facing each other and extremely close together can really clean up a spectrum.

I like the idea of a swivel. I may try one out this weekend and see if I can get one to work.

I wanted to add another brace along two of the edges to keep it from sliding around on my phone but worried about it limiting the number people who could use it. We could always make different mounting plates for different devices or make the plate over-sized and include an extra corner piece that could be glued on to make it a custom fit.

Ideas, ideas, and more ideas. This should keep me busy for a while :).

Reply to this comment...


I'll get it added to the the PLOTS WIKI ASAP.

Reply to this comment...


Awsome thing you have created :)

Reply to this comment...


Flip-up spectrometer sketch

That's kind of what I'm thinking for a reinforcing plate and a flip-up tube.

Brad - the wiki is fine too but you can also post a research note if its just something quick -- like a blog post basically. Just so you know.

Reply to this comment...


I'm kind of thinking the longer plate as pictured above could work well, esp. if combined with permanent double-sided foam tape. If you don't want to attach it directly to your phone, anyone can buy a case for their particular phone, and attach to that. I think some careful work on the shape of the joint could prevent light leaks -- maybe some kind of internal flange...

...but i drew the lower design wrong since it couldn't fold backwards past that flat part at the joint.

Reply to this comment...


Jeff-- what about those soft silicone phone cases? They're cheap and I bet that the attachment plate could flange underneath a hole in one of those, so long as the flange extended in all directions (currently it just goes down.)

Is this a question? Click here to post it to the Questions page.

Reply to this comment...


looks really nice and matte black -- but i've been thinking that the more rigid the case, the better -- the PVC spectrometer is mounted to a fairly stiff but still flexible vinyl-looking case, and the case still flexes easily, messing up the calibration. I wonder though -- if you cut the case open and lay the rigid extended flap against the phone, inside the case, maybe it'd be stable enough. I just don't want folks to have to adhere things to their phones directly if possible. I'll try a few different things... i bought some super-cheap pink phone cases on Amazon for $2 each, so i can make some mistakes :-)

Reply to this comment...


Warren I just discovered you guys by the Kickstarter page. I like the work you are doing. You show your fold up design flipping all the way open so I made a few sketches and convinced myself to go the other way. Most cellphone cameras are in the back left corner and on iphone very close to the edge the spectrometer would have to be sticking off the cell phone to get at the right angle. The album is two designs I did in sketchup it still needs work but it gives you an idea what I am working on. It can do various angles to work with different gratings. The design is roughly 4" X 1" X 1". If your interested in it I will send you the model or upload it somewhere when finished.

Reply to this comment...


Hi, MFarmer - sorry your image did not load... can you try again? I'd love to see your design. Consider posting a note at http://publiclaboratory.org/note/add (after signing up at http://publiclaboratory.org/user/register)

Is this a question? Click here to post it to the Questions page.

Reply to this comment...


I signed up yesterday but somehow missed your reply. I posted an image and the album link here: mobile-android-version-spectral-workbench

Reply to this comment...


Fantastic work. This is a seriously cool project. I'm going to fiddle with my design for a telescope eyepiece phone mount http://www.thingiverse.com/thing:39144 . It's in two pieces (a phone case and a short barrel to slide over the eyepiece) that fit together with 2 formed pins. It may not be such a stretch to design the spectrometer tube to fit to the existing case.

Reply to this comment...


Login to comment.