Coffee Data Science
Maillard Reaction with Extended Roasting
A side-quest down final bean temperature.
There is good evidence showing that final bean temperature has a tight correlation to bean color. I wanted to do an interesting test with respect to the Maillard reaction to challenge this color idea. My thought was that a longer roast will get darker than expected simply because the Maillard reaction has gone on for a long time. The reaction itself takes longer at lower temperatures.
So let’s play a game: guess the three final bean temperatures?
If you guessed 180 C, 200 C, and 230 C, you would be wrong.
I made three profiles to hold a bean temperature for 20 minutes at 150 C, 180 C, and 200 C. This test also helped me understand how well the bean temperature could be controlled with the addition of heat.
The Rate of Rise (RoR) stabilized, but it took longer for higher temperatures.
I looked at the inlet and exhaust temperature just to compare how they were during the roast.
Putting all the data together for 180C is nice, and it showed the drum having a constant but lower temperature of 160C.
Coffee Color
I measured the final coffee color using the DiFluid Omix. I plotted the three roasts (Hold temperature) against some color data, and there is a major shift by where the color is expected based on the bean temperature. All of this data is from the same coffee varietal.
Looking at this data suggests the color is correlated as long as one isn’t doing a very long roast, and even then, the trend is consistent for a particular roast profile.
If you like, follow me on Twitter, YouTube, and Instagram where I post videos of espresso shots on different machines and espresso related stuff. You can also find me on LinkedIn. You can also follow me on Medium and Subscribe.
Further readings of mine:
My Second Book: Advanced Espresso
My First Book: Engineering Better Espresso
