Sitemap

Coffee Data Science

Added Moisture to Green Beans for Coffee Roasting

More data, maybe more control.

--

My initial feasibility showed adding moisture to green coffee improved flavor. This is not unexpected based on previous experiments, but my aim was to have more control over the roast development. Again, these experiments found adding 3% or 6% moisture improved flavor.

Post Roast Metrics

The roasts with added moisture ended up having some differences in density, but I didn’t see a pattern.

Press enter or click to view image in full size
Press enter or click to view image in full size

In terms of roast color, there were some small differences, but the differences were within a few points which I consider good enough for these tests. Each roast ended at 212 C bean temperature.

Press enter or click to view image in full size
Press enter or click to view image in full size

Metrics of Performance

I used two sets of metrics for evaluating the differences between techniques: Final Score and Coffee Extraction.

Final score is the average of a scorecard of 7 metrics (Sharp, Rich, Syrup, Sweet, Sour, Bitter, and Aftertaste). These scores were subjective, of course, but they were calibrated to my tastes and helped me improve my shots. There is some variation in the scores. My aim was to be consistent for each metric, but some times the granularity was difficult.

Total Dissolved Solids (TDS) is measured using a refractometer, and this number combined with the output weight of the shot and the input weight of the coffee is used to determine the percentage of coffee extracted into the cup, called Extraction Yield (EY).

Shots

I pulled 4 shots for each roast resulting in 36 shots where there were 12 pairs of baseline, 3% added, and 6% added.

Press enter or click to view image in full size
Press enter or click to view image in full size

Added moisture consistently outperformed the baseline.

Press enter or click to view image in full size

This was true across individual metrics.

Press enter or click to view image in full size

TDS and EY seemed to have a slight trend, but it turned out not to be statistically significant.

Press enter or click to view image in full size
Press enter or click to view image in full size

From a statistical standpoint, I compared the paired shots using a two-tailed paired t-test, and all the taste metrics showed improvement except Sharp.

Press enter or click to view image in full size

When comparing 3% and 6%, there was not a statistically significant difference across the metrics.

Press enter or click to view image in full size

This set of experiments has helped me establish a baseline for pre-roast processing to improve the roast. I’m curious how it will compare to other optimizations.

One thing to consider is the starting bean moisture. I measured all three to be between 9.7% and 11.1%, and the experiments did not show much of a difference between 3% and 6%, which suggests adding some moisture is good, but more may not help.

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

My Links

Collection of Espresso Articles

A Collection of Work and School Stories

--

--

Robert McKeon Aloe
Robert McKeon Aloe

Written by Robert McKeon Aloe

I’m in love with my Wife, my Kids, Espresso, Data Science, tomatoes, cooking, engineering, talking, family, Paris, and Italy, not necessarily in that order.