Measuring beer colour with Open Colorimeter

A colorimeter is a scientific instrument used to measure the intensity of a colour in a solution. The Open Colorimeter designed by IO Rodeo consists of an Adafruit TSL2591 Light Sensor and an Adafruit PyBadge inside a 3D printed enclosure. The light sensor is attached to a 3D printed cuvette holder, with a white LED light board attached to the other side. The sample is placed between the sensor and the light source, which allows the absorbance or transmittance of different wavelengths of light to be measured.

Colorimeters can be used to measure concentrations of pollutants in water for environment monitoring, or to ensure colour consistency of food, textiles or paint. They can also be used for soil measurements or biochemical assays.

IO Rodeo has a great tutorial on how to use a colorimeter to measure beer colour. We don't have the 430nm LED mentioned in the tutorial, so we tried with a white LED board instead. Here is the distribution of wavelengths for the white LED according the datasheet:

There does seem to be enough luminous intensity around the 430nm mark that this should work. We bought some beers from our local Lidl, going with the supermarket brand for each style to make it easier for others to replicate the experiment. We measured the following absorbance values for four different beers:

To calculate SRM (Standard Reference Method) beer colour for 430nm you would multiply the absorbance value by 12.7. Since we're using a different LED, we compared the values for the lager and IPA to that of the IO Rodeo experiment:

We already know that he slope of the line for 430nm is 12.7, and assuming the SRM for the lagers and IPAs are similar, we can calculate the slope of the line for the white LED source as 61.08. Using that as our multiplication factor, we can then calculate the SRM values:

Our amber ale appears to have a similar SRM to IO Rodeo's Irish red ale (17.4), and the porter as a similar SRM to a Guinness stout (70.6), so it does appear to work pretty well. To be sure, we could use a spectrophotometer, but unfortunately we don't have one of those lying around. We'd like to eventually redo the experiment with a 430nm LED board to see if the results are the same.

This post was first published on gerritniezen.com.