Question: What is the spectral resolution of a Raspberry Pi camera spectrometer?

by TC1994 | January 23, 2020 06:13 | #22373

Hello, I have seen some activities on Public Lab on spectrometers using raspberry Pi camera, but I would like to find out about the spectral resolution that it can achieve.

This research article here shows that raspberry pi camera can hit 1nm resolution, but this project only has a spectral bandwidth of 60nm in the UV region.

My questions are:

1. Is there anyone who has the same success with raspberry pi camera spectrometer to get around 1nm resolution? If so, could I have the link please?

2. Is it possible for the raspberry pi camera spectrometer to see the whole visible light spectrum while getting high spectral resolution?

3. If so, could I replicate the design in the article and analyse the visible light spectrum instead, while getting good resolution?

Thank you! Sorry if I asked the wrong questions because I'm pretty new to spectroscopy.


  1. I've had success in getting sub-nanometer resolution by using lenses rather than mirrors. A short writeup is here. I still have a ton of tweaking to do for concrete proof, but it is possible to lower the resolution by swapping the focal length of the secondary lens for something with a higher optical power.

  2. Your main limitation is the physical size of the sensor. The pixels are 1.12 um in pitch which and you only have 3280 pixels to work with so if you wanted the whole spectrum you would need to move the sensor.

  3. Please, by all means if you want to replicate my design I'll post the files onto github, but I have a ton of polishing I need to do. If you need any help sourcing or building let me know!

Hi @wln215, thank you so much for your reply and thanks for sharing your work!

I'm hoping to build a handheld spectrometer (the size of a mobile phone, but can be thicker) with picamera, so I'm wondering if the second lens of your design can be changed to one with a shorter focal length so the whole set up can have a smaller physical footprint?

What are some considerations that I need to have in order to build a high resolution, handheld spectrometer?

Thank you!

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

You certainly can swap the second lens to shrink the thing to increase bandwidth, but decrease resolution because you are limited by pixel pitch. One of the key considerations is the number of slits in the grating you light up. More slits, more interference, the more contrast on the CCD. The laser can be replaced with a slit, and if you use cylindrical lenses it would be even smaller.

Thank you for your answer :)

Reply to this comment...

Log in to comment