Coffee Data Science
Modifying the Maillard Reaction in Coffee Roasting
Roasting experiments encore.
The maillard reaction is a key part of cooking and what makes things good. So I thought it would be interesting to try to modify the reaction using the fan. This series of tests uses the same Bean Temperature to Inlet Temperature (BT/IT) profile, but the fan speed was modified only for the temperature ranges where the maillard reaction takes place.
I made four roasts using the Roest:
- Baseline
- Fast: gradual increase to maximum fan speed
- Slow: gradual decrease to minimum fan speed
- Super Slow: Immediate decrease to minimum fan speed.
For the super slow, I also dropped the drum speed to the minimum of 20 RPM, which slows the roasting process more.
Roast Curves
The roasting curves for the first 3 profiles looked very similar.
Even the Rate of Rise (RoR) were nearly identical. The Super Slow roast crashes out, but the bean temperature is only 150C at that point.
The same is true for the crack pattern.
Post-Roast Statistics
All four roasts has similar weight loss.
Only the super slow roast had a big change in moisture.
I’m still unsure about water activity and its usefulness as a metric for roasted coffee.
Density was similar across all, but the super slow was slightly higher.
All four roasts were with a few points in terms of color.
Tasting Equipment/Technique
Espresso Machine: Decent Espresso Machine, Thermal Pre-infusion
Coffee: Home Roasted Coffee, medium (First Crack + 1 Minute)
Pre-infusion: Long, ~25 seconds, 30 second ramp bloom, 0.5 ml/s flow during infusion
Filter Basket: 20 Wafo Spirit
Other Equipment: Acaia Pyxis Scale, DiFluid R2 TDS Meter
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 took 4 shots of each roast as an intuition building dataset. I reordered Slow to be after Baseline because it outperformed in taste. Fast was not great in comparison, and super slow was okay. The Fast roast was intended to see if performance goes the opposite direction than theorized improvement of slow, and so it is a helpful data point.
This is each set of shots as another way to compare.
In terms of EY, they all performed similarly.
This was an exciting data set for me because it opens up a new avenue to explore in how I understand coffee roasting. I’m particularly interested how this knowledge will complement what I have learned from thermal pulsing in coffee roasting.
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
