This is a spectrograph capable of resolving the sodium D-lines. A spectrograph differs from a spectrometer in that it literally takes a photograph of a spectrum. Analysis through software packages like imageJ (it's free and easily findable with a google search) allows users to create an instrument intensity vs. wavelength graph like a spectrometer.

This is a design that was used in a upcoming American Journal of Physics article (to be officially published ~ March/April of 2024). This design specifically requires the following additional pieces: a diffraction grating (we used an Edmunds Optics 32-215 reflection grating ~ $131), a lens (Thorlabs AC050-0150-A achromat doublet lens ~ $52), and a webcam (Logitech C310 ~ $27). These parts could be exchanged with similar items, but it could necessitate design changes.

In the pictures, there are example spectra (attained with this device) and detailed assembly instructions. Below, I will describe how to calibrate the "pixel axis" so that wavelengths in an unknown spectrum could be measured.

Calibrating your spectrograph pixel axis:
To calibrate your pixel axis so that you can measure wavelengths of unknown light sources, you should start with a known light source. This light source should have a discrete spectrum that is well documented; i.e., you should typically use an atomic spectral tube. One can get the associated wavelengths of atomic sources by going to the National Institute of Standards and Technology (NIST) website; try a google search using “NIST strong lines of atom_name.” If you are operating in a low-budget environment where these atomic spectral sources are unavailable or grossly limited, one can get by using a fluorescent light (the long tubes that are often used in large buildings, schools, and sometimes bathrooms). Also bear in mind that using a fluorescent light source will effectively reduce overall accuracy as many of the line wavelengths are not well documented.
Here are the “Typical” wavelengths found in fluorescent lamps
Wavelength (nm) Approx. color description
404.656 violet Mercury
435.833 Blue-violet Mercury
487.7 blue Terbium (Tb3+)
542.4 green Terbium (Tb3+)
546.074 green Mercury
~577.7 green ? (Hg @ 576.960 or Tb3+ @ ~578)
~580.2 green ? (Hg @ 579.066 or Tb3+ @ ~580)
~587.6 orange Probably Europium (in Eu+3:Y2O3)
~593.4 red-orange Probably Europium (in Eu+3:Y2O3)
~599.7 red Probably Europium (in Eu+3:Y2O3)
~611.6 red Europium (in Eu+3:Y2O3)
~631.1 red Probably Europium (in Eu+3:Y2O3)

Now aim your assembled spectrograph at the known light source and adjust the alignment to maximize the brightness of your spectrum. If this is the first time you have used it, it would be wise to make small adjustments of the focusing lens to sharpen the focus. That is, remove the lid of the spectrograph, clamp down the main box, and carefully rotate the lens_adjust1 piece to optimize the focus. When the best focus is attained, put the lid back on. Now take a picture of the spectrum trying to maximize the brightness of the lines.

To analyze your spectrum, you can use a variety of tools, but a nice open source tool (i.e., FREE) is imageJ. Go to https://imagej.net/ij/ for more information. Use the imageJ rectangle tool (the rectangle icon roughly below the File menu), to select the part of the image you wish to analyze. You should select a horizontal range from the left extreme to the right extreme as pixel 0 is defined as the pixel on the left edge. If you do not do this, subsequent analysis of other images will have a varying location of pixel 0. The vertical extent should cover a range of the spectrum where the lines appear to have the same brightness. Once your region of interest is selected, go to Analyze -> Plot Profile. This should give you a graph similar to what you would get from a spectrometer (an instrument intensity vs. pixel graph). The data for this graph can then be copied and pasted into any spreadsheet program for further analysis.

To complete your calibration, you must now match intensity peak pixel locations with known wavelengths. If done properly, a linear fit between pixel location and wavelength should be highly accurate (a picture of such a graph is available in the photos).
Once the calibration is complete, you can measure the wavelengths of unknown light sources. Simply aim your spectrograph at the new source and maximize the brightness of the spectrum. The linear fit you found in the calibration should still apply to this new light sources with the following caveats:
• Make sure all optical elements are held tightly. If not consider gluing them down.
• It is vital that the spectrograph is aligned to maximize the brightness of the spectrum. If the light source is very bright to the point where the spectral lines are saturating the detector, you cannot misalign the device to reduce the light levels. Either use a neutral density plate or a pair of polarizers to reduce the light level, but still adjust the alignment to maximize the spectrum’s brightness.