February 20, 2025
Today's coffee that we are going to approach is a Swiss Water process decaffeinated coffee from Rwanda. Prior to decaffeination, this coffee was natural processed. Decaffeination was done at the Swiss Water decaffeination plant in Vancouver, British Colombia in Canada. The coffee is a part of the limited release small batch collection. The goal of sharing this roasting approach is not to give any kind of prescription on how to roast this coffee, but rather to provide you with some ideas of how to approach similar coffees.
Arrival QC:
The Coffee:
This coffee comes from a group of women producers who are members of the Tuvemubukene washing station. The coffee was natural processed prior to arriving at the Swiss Water decaffeination plant in Canada. The coffee has a relatively low moisture content. This low moisture content tips me off that the coffee will likely have a shorter drying time, may be more prone to scorching, and will end with a lower roast weight loss.
The Roaster:
For this roast, we'll be using a Proaster 1.5 kilo drum roaster hooked up to natural gas. Batch size is 500g, approximately 40% of roaster capacity. I've used a small batch size on this roaster to get closer to mimicking the gas power that you might find on a larger production roaster. It must always be mentioned that each roaster is different. The type of probe, placement of probe, and batch size will all influence the way in which bean probe temperatures are registered in a roaster. With that said, don't read too much into the exact temperatures, but look more closely at the theory behind the approach, time spent in different development periods, and rate of rise to get a better idea of how to translate this profile over to your own machine.
The Goal:
In our sample roast of this coffee, we found notes of chocolate, honey sweetness, and a cherry fruit quality. In this roast approach, I'm featuring two different roast profiles. One profile will seek to highlight the bright fruity complexity of the coffee, while the other featured profile will increase caramelization development to amplify the honey sweetness and body. The lighter roast profile (profile A) would be ideal for drip preparation, while the more developed roast profile (profile B) would be excellent for espresso preparation.
You can read about my approach to coffee profiling and roast phases in my previous blog post here.
Roast A: Lighter roast for drip
The Approach:
In this roast, I started with a moderately high charge temp and restricted air flow. Here I used a charge temperature of 400 degrees F. I kept the gas application at 66%, ensuring that the coffee progresses through drying with enough energy to develop complex acids during Maillard. As the goal with this coffee is to make it bright and complex, I began raising the gas application at turn around, with full gas application as I approached Maillard (at 305 degrees). At 280 degrees, I also increased the airflow to 60%.
The goal of this is to create the maximum chemical reactions during this phase where sugars and amino acids are breaking down and recomposing into complex volatile aroma and acid compounds. I am looking for a short period of time in Maillard, followed by a shortened time in caramelization. During caramelization, complex acids break down and form melanoidins. This will reduce acidity and increase body. However, with the goal of this roast being to create a complex fruit forward coffee for drip, my goal is to have a shortened time in caramelization and post crack development, retaining as much of the complex acids as possible.
At 380 degrees, I began to reduce my gas application moving from 100% to 44%, then down to 22% when first crack starts. Additionally, I increased air flow ahead of first crack. When the bean opens up, many gasses will be released from the bean into the roaster. I increase my air flow in order to move those gasses off the beans and improve clarity in the cup.
First crack occurs on my roaster with this machine at 387 degrees. The coffee finished with 1:15 of post crack development. It is important to remember that development is happening throughout the caramelization phase of the roast, not just once the beans begin audibly cracking.
The coffee ended with total roast time of 7:57. Roast weight loss was 13%.
The Cup:
Aroma - Raspberry, strawberry, candy sweetness
Tasting Notes - Bright green apple and limeade acidity. The coffee is very sweet with notes of cherry, honeycomb, raisin, and cocoa. The finish of the coffee had soft orange blossom florals. Coffee had a medium juicy body.
Roast B: Full bodied and sweet for espresso
The Approach:
The goal for this roast is to create a sweet, full bodied coffee that would excel on espresso. From sample roast cupping, we know that this coffee has a rich honey sweetness and juicy body. To bring this character forward, we will be working to extend the time spent in caramelization and development. My goal here was to extend time in caramelization, breaking down acids, caramelizing sugars, and building melanoidins within the coffee.
I started the roast with a lower start temp (380) and moderate gas application (66%), which I maintained through turn around. At 280 degrees, as I approached Maillard reaction (about 305 degrees), I increased the gas application to 100% and increased airflow to 60%. This level of gas application was maintained through 330 degrees, at which point the coffee begins to approach caramelization reactions. In order to assure that the caramelization phase was extended, gas was significantly reduced to 44% of maximum.
This roast is the second iteration of an "espresso roast" that I did. In the first attempt at this roast, I found I did not keep enough heat momentum through the end of the roast, causing the coffee to come off as "baked". The baked roast of this coffee had many of the attributes I was looking for, but a very flat finish. On this baked roast, I had reduced gas to 22% at first crack and finished at a temperature of 398. I had hoped that the coffee would have enough momentum to complete the development to an end temperature of around 405, but it significantly declined in rate of rise and finished the roast at a crawl. To correct this, I continued with the gas application of 44% through first crack and reduced the gas application to 22% at 394 degrees (approximately 10 degrees lower than my intended drop temp). This change provided ample heat momentum to finish the roast. As I approached the end of the roast, I drew many pulls from the trier to assess roast aroma. I wanted to achieve as much carmelization as possible without entering into carbonization. With this as my goal, I was smelling for the aroma of toasted marshmallows from the trier. In my roasting experience, that aroma of toasted marshmallow is quickly followed by aromas of carbonization/burning. I often use that toasted marshmallow aroma as an indicator for the completion of a roast where sweetness and body are the main goal. In roasts where I am looking for bright acidity, I finish the roast prior to the toasted marshmallow aroma being noticeable.
This roast was finished with an end temp of 407 degrees and a total of 2:22 of post crack development time.
The Cup:
Aroma - Blackberry, cocoa, mango, honeycomb
Tasting Notes - Coffee has notes of sweet cooked blackberry, honeycomb, caramelized white sugar, golden raisin, toffee and hints of molasses. The body is full and sirupy, with a lingering sweet sugar finish.
Bringing It Home:
The graph above breaks down each roast into time spent in the different roast phases by seconds. In this graph you can clearly see that both roasts had a similar time spent in drying and Maillard, with Roast B having a bit more time in both of those phases. The most significant differences in these two roasts is the time spent in caramelization and post crack development. The difference in end temp on the roasts was only 6 degrees (401 on Roast A, 407 on Roast B). However, by looking at the graph above, one can clearly see the distinct difference in the path to get there.
Happy roasting!
Comments will be approved before showing up.