Spectroscopy is one of the most useful tools in all of science, and for The Hackaday Prize’s Citizen Science effort [esben] is putting spectroscopy in the hands of every high school student. He’s built a super cheap, but very good spectrophotometer.
The idea of a spectrophotometer is simple enough – shine light through a sample, send that light through a diffraction grating, focus it, and shine the light onto a CCD. Implementing this simple system is all about the details, but with the right low-cost lenses and a 3D printed enclosure, [esben] has this more or less put together.
Of course, lenses and diffraction gratings are relatively simple. You need real data, and for this we can turn to another one of [esben]’s projects in the Hackaday Prize. It’s a breakout board for a linear CCD module, able to capture the spectrum coming off a sample with incredible precision. This is how real spectrophotometers are put together, but because of the difficulties in driving a CCD, not many people have put one of these together.
Both of these projects are finalists for in the Citizen Science portion of The Hackaday Prize. That’s an awesome result for what is a complete system for learning about spectroscopy with a device that’s also able to produce some high-quality data, too.
I usually don’t post useless comments like this, but I swear I just went to sit down at my desk waiting for the last piece of my homemade spectrophotometer to finish printing when I saw this post appear…
I was going to do something like this too but I was going to use a camera lens and an equilateral glass prism with the same sort of linear CCD sensor. I want to scan stuff about a foot away but with a camera zoom lens I thought it should work well at longer distances too. I want to scan things like fruit and see if I can identify the object type by its light signature.
Is there something wrong with the way I was going to do this?
Is the complex optical setup in this article really necessary?
I have an old scanner with a sensor and an expensive camera zoom lens so all I would have to buy if I did this my way is the glass prism and that only costs 5 bucks on ebay.
If you put a prism in front of a lens (or behind it) then you get a spectra of everything mashed on top of one another and offset and it’s a mess. It’s like wearing X-ray specs. Even if what you point it at has a single colour the resulting spectrum would be very fuzzy low resolution.
To get a clean and sharp spectrum you need an optical system that images a slit (or a pinhole – the slit lets a lot more light through and that dimension does not affect resolution) onto a sensor. The dispersive element goes inside that to produce a spectrum. You then need some form of lens to put the light you want onto the slit.