Coffee Data Science
Home roasting has been a passion of mine for some time, but I am impatient when it comes to letting coffee rest. I have previously found the best time to wait is 3 to 5 weeks, and this has challenged me. So I developed a technique to degas faster using water or humidity. This reduced optimal use time to 1 to 2 weeks, but coffee often went stale at 3 to 4 weeks vs 6 to 7 weeks without the treatment.
I initially bought a vacuum chamber to see if I could degas faster without water. However, the internal pressures of trapped gas is high (close to 6 bars according to Samo Smrke, and reducing outside pressure from 1 to 0 bars only changes the differential from 5 to 6 bars between the gas inside and the pressure outside.
Let’s merge both:
- Roast coffee
- Add 4% moisture by adding and mixing water in the beans
- Let sit from 12 to 24 hours for absorption
- Put in vacuum chamber
The main challenge was that the degassing causes the vacuum to gain pressure, so I had to turn it on twice a day to keep a good vacuum. My hope was that whatever oxygenation going on by the water would be reduced by the vacuum.
First, I looked at weight loss of the coffee in a vacuum jar vs a chamber. The vacuum jar had a hand pump, but it would not have as good of a vacuum. Sure enough, the chamber reduced the weight more than the jar, and I suspect that weight loss is CO2. Generally there are 6 mg/g of CO2 in coffee or within that range, so for 100g of coffee, there should be 0.6g of CO2.
Coffee: Home Roasted Coffee, medium (First Crack + 1 Minute)
Shot Preparation: Staccato Tamped
Pre-infusion: Long, ~25 seconds
Filter Basket: 20 Wafo Spirit
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).
Intensity Radius (IR) is defined as the radius from the origin on a control chart for TDS vs EY, so IR = sqrt( TDS² + EY²). This metric helps normalize shot performance across output yield or brew ratio.
I pulled 23 paired shots over 4 roasts, and I found an improvement in taste in most cases.
Edit: After publishing, I was reviewing the data, I found some shot pairs were misaligned. I aim to have shots taken at similar times so the time off roast is similar. This change didn’t affect taste scores, but one pair was lost because it was misaligned DE vs Kim. This change does affect EY and IR results, and I changed the text accordingly.
I saw some improvement in EY and TDS, but it was less clear from the graph.
Breaking down the taste components, there was a positive effect. Sour and Bitter did not see statistically significant effects.
From a higher level standpoint, TDS/EY/IR didn’t change in a statistically meaningful way. This could change if more data was collected, and this is an experiment I’m interested in trying again with thermal pre-infusion.
All the taste metrics, except for Sour and Bitter, saw statistically meaningful improvements.