Public Lab Research note


The Evolution of the PLOTS Smartphone Spectrometer

by BradDudenhoffer | February 12, 2013 01:10 12 Feb 01:10 | #5949 | #5949

Back in August 2012 Jeffrey designed a version of the smartphone spectrometer which he had 3D printed, but unfortunately the results were less than ideal. He made a post on Thingiverse looking for some help with the design.

I am a research chemist and love making tools and instruments for use in the lab. I also hate that most scientific instruments are priced way over the budget of many citizen scientists. I had also recently finished building my MendelMax 3D printer and was itching to put it to use For Science! I was able to modify his design and make a functional spectrometer which is printable on the fused powder printers used by Shapeways as well as the extrusion printers used my most hobbyists. I shared it with the community, made a few improvements, and thought I was done with it.

A little while later I was contacted to see if I would help design an official molded spectrometer. I didn’t have much experience with molded manufacturing but jumped in head first. So far I have been through 19 iterations of the design. We tried fold-down, flip-out, and fold-flat designs before settling on a much more robust rigid backpack design. With a lot of input from the great PublicLab team we finally refined the design to something that looks like it belongs on the back of a smartphone. I recently sent the 3D model to our manufacturer and am waiting for some final feedback so we can get these things out there to the citizen scientists who have been patiently waiting.

Here is what we have come up with.

It is a three piece design and will be molded in black high strength plastic.

The slit-end will work just like the paper model but also utilizes a hood to cut down on stray reflections. The hood is sized to hold a standard 1cm square cuvette so you don’t have to worry about your sample moving and changing the spectrum. A pair of magnets recessed in the hood allows for the attachment of future accessories such as an LED light source, a laser, and who knows what else (microscope/telescope eyepiece?).

The square recess on the side is the mounting location. It is 20x20mm square and has a recessed magnet on the opposite side to hold the spectrometer to the mount. The first mounting plate will be for a standard ¼” photography/video tripod screw. The entrance slit is set at about 1mm which is very functional but a bit wider than ideal. There is plenty of room to overlay your own narrower entrance slit if you want to sharpen the peaks a bit at the loss of a bit of light intensity.

I hope to hear back from the molders this week and have any necessary corrections finish very soon.


10 Comments

That looks beautiful. Also very clever.

Will there be any provision to alter the geometry among the slit, the grating and the camera lens? Assuming that the grating is perpendicular to the lens axis, what is the default angle to the slit?

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

Reply to this comment...


In order to make it as sturdy as possible we decided against making the angles adjustable. The grating by default set to be perpendicular to the lens but there is plenty of room inside to shim it to whatever angle you want. The grating-slit angle is set at ~40 degrees. I can't give a more exact number than that since the grating opening is extra large to allow the spectrometer to accommodate just about every smartphone on the market.

Reply to this comment...


So glad you posted this -- i'm going to link heavily to it in the KS update I'm writing.

By making the back plate a separate part, will we get to redesign it and remold it without the extra expense of redoing the whole thing?

Also -- chris, if you can definitively demonstrate that with a common-ish smartphone camera and an otherwise identical geometry, there is a much better but not 90° angle we should go for, i think there is still time to make modifications to the design before submitting it for molding.

Brad - one more thing -- i was thinking the camera mount point might usefully be extruded and made parallel with the top, and notched on the inside to precisely fit a 1/4" bolt, such that the bolt would be held in place as you screwed the camera mount bolt in:

Camera mounting options

What do you think?

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

Reply to this comment...


The base plate was made a separate piece for exactly that reason. It shouldn't be too hard to design a variety of them for different applications; a webcam or DSLR for instance.

Changing the default grating/camera angle would be pretty easy to do. Just let me know what angle you want use.

I can change the tripod point to a recessed nut. I went the route I did because I'm not sure there is room for a regular tripod plate to mount there unless it is flush with the case. For example, my smallest tripod plate is about as wide as an iPhone. If you plan to use one of those smaller smartphone tripods then it wouldn't be a problem. I will check with the molders and see how much trouble it would be to add the hole.

Reply to this comment...


Jeff, the critical angle seems to be the grating angle in relation to the light path from the slit. For 1000 lines/mm grating film, this angle is apparently greater than 50° for slit-grating distances less than 30 cm. I can try to determine in the next couple of days what this angle is for a DVD grating and how critical it is. What is the working slit-to-grating distance in your model?

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

Reply to this comment...


The nearest edge is 59mm and the farthest is about 80. Are you measuring the angle from the line normal to the sensor surface or parallel? If from the normal, the angle range is from 46 degrees at the closest edge to 59 degrees at the furthest edge. My previous designs used a slit-grating distance of 75mm and I found that 58 degrees was about ideal. Since this design is "universal" in order to fit as many phones as possible, the angle is a compromise of the various camera positions.

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

Reply to this comment...


Since this is a relatively open endeavor, I can haz STL since I haz my own uPrint?

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

Reply to this comment...


I just finished posting the previous (somewhat boxy) version to Thingiverse: http://www.thingiverse.com/thing:49934

The most recent model with all those sexy curves should be up by this Sunday :)

Reply to this comment...


Mmmmmmm.... sexy curves.... glglgglggll.

Excellent! Thanks

Reply to this comment...


So this was very interesting 3 months ago, and I decided to wait for version 4 to be posted. Am I watching in the wrong place? What's going on with this?

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

Reply to this comment...


Login to comment.