Coffee Data Science
I previously did a comparison between the Kim Express and the Decent Espresso (DE) machine, where I found that the Kim Express had better coffee flavor and extraction. When I dug deeper, I found two main reasons: water temperature and pre-steam infusion (PI).
Increasing the water temperature did not aid to completely close this gap, so I investigated coffee sneeze what turns out to be pre-steam infusionI wanted to replicate this in DE and it was an engaging journey towards better coffee.
The pre-infusion of steam happens in the Kim Express because the water in the tank is over 100°C. My water usually ran between 116°C and 120°C, meaning that when the water starts to enter the chamber, it first turns into steam. This steam passes through the disc before the water does.
I started with a profile I had been working on for a few months. It involved a rapid pre-infusion that was manually stopped after the filter was covered with coffee. Then there was a pause until manual input was needed to start building up the pulsatile pressure.
I started by adding a 10 second PI of steam, using a 0.5 ml/s flow limit. This had an immediate effect on the buds.
I experimented with longer and shorter steam PI. The best flow was 0.2 ml/s because it allowed just the right amount of water to be put into the puck. 0.1 ml/s was too ponderous and 0.4 ml/s was too rapid.
I also looked at the length of time for the steam PI. Ideally between 5 and 20 seconds. This depends on the boiler temperature. The DE measurement definitely measures a higher puck temperature, but it is measured at the top. I also noticed that the water temperature did not drop as quickly as in my initial profile.
I have modified the Steam PI length based on the coffee I am brewing. I am still testing the variable to better understand how to utilize it.
Coffee Maker: A decent coffee maker
Coffee grinder: Zero niche
Coffee: Home Roasted Coffeemedium (First crack + 1 minute)
Shot Preparation: Staccato Tamped and Staccato
Preinfusion: long, ~25 seconds
Infusion: pressure pulsation
Filter basket: 7g/20g VST
Other equipment: Atago TDS meter, Acaia Pyxis scale
To evaluate the differences between techniques I used two sets of metrics: Final Score and Coffee Extraction.
Final result is the average of a scorecard of 7 metrics (pointed, affluent, syrupy, sweet, sour, bitter, and aftertaste). These ratings were of course subjective, but they were calibrated to my tastes and helped me refine my shots. There is some variation in the results. My goal was to be consistent across each metric, but sometimes the granularity was hard.
Total Dissolved Solids (TDS) is measured using a refractometer and this number, combined with the output weight and input weight of the coffee, is used to determine the percentage of coffee in the cup, called Extraction Efficiency (EY).
I looked at some sample data because I don’t feel ready to do a paired set of shots. I’d like to understand the technique better. So I looked at two sample sets that show what I generally experience.
Let’s start with an example from my 5/4/2022 smoke with and without Steam PI. Steam PI seems to be a bit darker in color.
Looking at two smokes, three shots per smoke, we can look at the performance compared to the Kim Express. The Steam PI does better on both smokes, but the 5/4/2022 smoke is able to match the Kim in EY/TDS and beat it in flavor.
