As I am getting more familiar with my set-up, I’m now working on developing a more rigorous testing process with several intentions… The most important of which is to be able to sketch with more clarity the potential of a coffee. Cupping is like gesture drawing for me, brief movements meant not to render in high detail, but capture a core and essential trait. Secondarily, though, I want to validate, or make corrections to, my own understanding of the roasting process. These two skills set the foundation for everything else I do in relation to coffee… So, I began.
I use a sort of shorthand for these, so I’ve provided a legend/glossary here that explains how I define each phase of the roast.
For this exercise, I was interested specifically in the Maillard Phase of the roast and the effect of the length of this phase relative to the Warming-Phase (and Caramelization phase, too, though that is minimized as these roasts were dumped quite close to the end of 1rst Crack.). Also, I’m interested in establishing general acceptable boundaries for these limits, i.e. how long is too long, where is ideal and what is too short. Obviously, this is just the beginning, and really only applies to similar coffees, but everything needs to start from somewhere, right?
The set up:
Using a coffee that I know fairly well – one that I know to be excellent – I constructed three variants on one roast profile. W-Phases and C-Phases remained equal in length, heat application and airflow, allowing only the length of the M-Phase to change via heat application. Airflow was kept as consistent as possible, even during the M-Phase. Cup blind for flavor characteristics and relative overall quality.
General Info:
Country of Origin: El Salvador
Farm/Grower: Finca Kilimanjaro / Aida Batlle
Defects: 0 defects / 300g
Processing: Wet Processed
Altitude: 1450m
Grade: SHB
Varietal: 95% Bourbon & Kenyan, 5% Pacas.
Region: Potrero Grande Arriba, Santa Ana Volcano, Apaneca-Ilamatepec Mnts.
Crop: Nov 2007 (vacuum packed)
| Roast #1 | Roast #2 | Roast #3 |
| W - 0:00 – 4:55 M - 4:55 – 8:40 1Cb - 8:40 1Cr - 9:09 1Cf - 10:10 C - 10:10 – 10:32 F -10:32 |
W - 0:00 – 5:02 M - 5:02 – 9:09 1Cb -9:09 1Cr - 9:21 1Cf - 10:01 C - 10:21 – 10:34 F - 10:41 |
W - 0:00 – 5:00 M - 5:00 – 9:19 1Cb - 9:19 1Cr - 9:39 1Cf - 10:01 C - 10:01 – 11:08 F - 11:08 |
| Cupping Notes | Cupping Notes | Cupping Notes |
|
Compressed, but clear, bright acidity – verging on green, but still quite sweet. Light body. smooth where it might seem like it would be a bit gritty from the light roast. Body flavors are not nearly as developed as other cups. seems comparatively simple (though not absolutely simple). |
Nice balance – <scan drawing> where the size of the body is subdued, supports the clean fruit, but does not overwhelm with it’s richness or presence. Dry aroma is rich chocolate floats around somewhere about 62% dark. Limited florals. |
Rich & relatively full. |
| Roast Notes | Roast Notes | Roast Notes |
|
Unsteady ramp 0:00-4:55. |
Drum exterior temp: ~110°C – ~ +25°C from roast #1. |
Early flux in gas – stabilized by 0:30. |
Overall, I am most happy with the development and balance of roast #2. It was clearly better than the others in complexity, clarity and development. M-phase works its magic on the development of both body and fruit flavors. Further to test is where M begins, as water content does both help and inhibit M development… the next question is how much is the right amount to help keep heat distribution even throughout the bean, but not inhibit full development of fruit flavors.
