Coffee Data Science

Moisturizing Green Coffee Before Roasting

Exploring moisture

Robert McKeon Aloe
Towards Data Science

--

Some roasters have been controlling for green bean moisture for a long time. I wanted to try it for myself and collect some data on how it affects taste and extraction. From what I read, 16% moisture content in the green beans is optimal, so that was my aim.

Moisture Protocol

From Chris Feran, he suggested the following:

  1. Measure the moisture of the green beans
  2. Add the required amount of water to get to a desired moisture content (16% recommended)
  3. Let the beans absorb for a maximum of 16 hours, minimum of the water being absorbed
  4. Roast

Adding Moisture

I didn’t want to invest in a moisture meter because they are expensive. I didn’t have data to suggest this would be worthwhile, so I dried out my green coffee beans at 180F until their weight no longer changed.

All images by author. 3 green beans

Drying!

I looked at two blends (four roasts) that were moisturized separately but roasted as a blend:

Here are how the beans looked from the 10/26/2021 roast:

Left: regular roast, Right: moisturized roast

I looked at the particle distributions, and moisturizing coffee causes the distribution to go coarser. So I compensated by grinding finer. I’m not sure how the particles changed in their shapes.

Metrics of Performance

I use two 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).

Equipment/Technique

Espresso Machine: Kim Express

Coffee Grinder: Niche Zero

Coffee: Home Roasted Coffee, medium (First Crack + 1 Minute)

Coffee Preparation: Humidity Treatment for 5 days post-roast.

Shot Preparation: Staccato Tamping

Pre-infusion: Long, 25 seconds

Infusion: Pressure Pulsing

Filter Basket: 20g VST

Input/Output: 21g In, around 24g Out

Other Equipment: Atago TDS Meter, Acaia Pyxis Scale

Performance

For these two roasts, I looked at 22 pairs of shots where the parameters for each pair was the same for regular vs moisturized. I saw a performance improvement in taste and extraction, but extraction was less clear as there were a few shots with a decrease.

I can break these down by roast, but there isn’t much of a pattern. There are a few outliers.

The change in taste appears to be statistically significant in a two-tailed paired t-test, but the EY and TDS just miss the threshold of 0.05. There was also an effect on shot times. The moisturized shots covered the filter (Time to Cover Filter (TCF)) faster and the total shot time was lower.

I plan to continue to study moisturizing green coffee beans before roasting. The main challenge is that between moisturizing them before the roast and humidifying them after the roast, I’ve had to go to the zero setting on the Niche. This has resulted in difficulty with the grinder halting, so I’ve had to slowly pour in the beans into the grinder than all at once.

I also want to study how moisture affects the particle shapes especially for the coarser particles. Controlling for moisture is key for maintaining consistency throughout the year as the moisture in the green beans is impacted by the humidity in the storage conditions.

--

--

No responses yet