Another delicious dry processed Ethiopian coffee crossed our cupping table this week, and we jumped at the opportunity to analyze it.
The coffee is from our friends at Hambela. The Buku cherry collection site in the eastern portion of the estate is the source of this coffee, from a specific plot called Mulu growing a single, unidentified variety.
The name Hambela has emerged as a recognizable denomination of origin in recent years, largely due to the efforts of the Adinew family. While Hambela Wamena is the name of a district, it is the Hambela farm, run by Aman, Michael, and Tariku Adinew, that stakes the claim as a center for exceptional coffee production.
The estate was part of a gift of land during the time of Emperor Haile Selassie to Muluemebet Emiru, the first African female pilot. It is her grandchildren that now manage the farmland and the METAD Agricultural Development company. Among the many important pieces of work undertaken by METAD are their commitment to equal employment opportunities for women and education opportunities for the youth of the coffeelands, their early partnerships with Grounds for Health, and their development of Africa’s first SCAA certified lab.
This coffee is everything we’ve come to expect from naturals processed at Hambela. Big ripe fruit notes lead the parade: concord grape, blackberry, and hints of citrus are all present. The coffee is sweet and jammy, with some roasts accenting hints of floral and vanilla notes, others opening up rustic dark cocoa and wine-like character. The one thing this coffee is not: subtle.
A highly reassuring green coffee by the physical numbers, this Hambela hits the usual marks for dry processed Ethiopian coffees. Low moisture content and low water activity accompany a higher than average density and generally small screen size. This should be relatively easy to work with and will likely weather longer storage times in stable environments pretty amicably.
Ethiopia’s genetic diversity of coffee is no secret, but increased attention is being paid to distinguishing cultivars and varieties, thanks in part to the work undertaken by Tim Hill and Getu Bekele. Established in the 1960s, the Jimma Agricultural Research Center was instrumental in selecting, breeding, and distributing scores of cultivars throughout the country in the decades following Haile Selassie’s downfall. These have included region-specific varieties, specialty cultivars, and hybrids and wild selections made for disease resistance.
Most Ethiopian coffees are known to be of a smaller screen size and are often quite dense, which is ideal for a roast that is short and uses a lot of heat. The moisture content of this coffee seems to be average and I was surprised to see the water activity as low as it was in a naturally processed coffee. I like to roast natural coffees a touch longer than their washed counterparts because they often reach first crack much later and at a higher temperature.
On a hunch I decided to use a 5:00 minute profile that I like to use for sample roasting and I was delighted to hear snaps at the 4:00 minute mark. This gave me a full minute for post crack development before the Ikawa started the cool down phase.
On the cupping table this coffee really shined and had a sparkling acidity that was both delicate and vibrant. The fruit acidity was citric in nature, tart and clean. I particularly enjoyed the light and crisp strawberry and watermelon. I have no doubt that if developed further this coffee would display some beautiful dried berry and marmalade flavors.
For this delicate and floral natural coffee, I decided to use a lower charge temperature of 356F and use heat to move the roast through the stages. I turned the heat up to 3 gas just after the Rate of Change peaked on the graph. As the roast slowly climbed towards yellow, I decided to ratch up the heat by another quarter tick, giving the roast a boost of momentum. The acid structure that we tasted in the Ikawa roast was abound, but also very delicate and I was afraid that if the overall roast was too long, it would muted the acidity and give the coffee a “baked” flavor.
Just a minute after yellow, my roast looked to be on target for a 6:30 minute first crack and so I turned the gas back down to 3. This coffee cracked rather late and high in temperature at 399F which is common for naturally processed coffees roasted on the Probatino. I turned the heat down again to 2 gas, which is my minimum gas adjustment. My rate of change was 8.7F/30 seconds and I knew that because this coffee was naturally processed it would not drop in temperature as quickly as a washed coffee.
On the cupping table this coffee has a beautiful bouquet of aromatic fruits and flowers. Apple, strawberry, and plum wine were among the most popular flavor notes. This is a very clean and sweet coffee that is an easy drinker all day long.
This week I tried something a little different. Frustrated with Linux and more than ready to roast some coffee, I installed Artisan on Windows and plugged my Yocto-Thermocouple in. Lo and behold, it worked immediately without so much as a hiccup. I was able to set up a roast, log it using the BT (bean temperature) thermocouple, and save these nifty graphs showing you just how my roast panned out:
For those of you that are visually inclined (like myself), this is an incredibly handy tool, and can give you a feeling for the roast at a glance. I was overjoyed that it was so easy to set up on a system running Windows. Despite being in the Bay Area, the Raspberry Pi taught me that I am certainly not a tech genius.
I approached this roast with the intention of drawing out the fruity flavors with a slow and steady development. I used a slightly lower charge temperature of 355F as read on the BT (bean temperature) probe, and the back of the roaster closed to stymie airflow. Since I reached turning point at almost exactly the same time as my previous roast (1:15/179F), I decided to close the back earlier (2:00/200F) to engage some airflow. At 3:00/229F I increased fan speed to 3, and at 5:45/287F I reduced amperage to 7.5A. Similar to my last roast, I increased fan speed to 6 immediately at first crack (7:00/305F). After 1:30 of development, I dropped this coffee at 8:30/325.
Looking at the above graphic, I can think of a couple points to work on in future roasts. My first inclination in relation to this roast is to introduce stronger airflow earlier in the roast. Furthermore, during our tasting Jen mentioned that a longer drying time might render better flavors. While I did stretch this coffee out a bit, I think it could have spent more time in drying than in sugar browning.
This coffee displayed very clear fruit notes and a juicy mouthfeel – pretty much everything I could have wanted from a fruit-dried Ethiopian coffee. The one sticking point was a slight roasty note at the very finish. In future roasts, I will hone in on airflow control to keep these roasty notes at bay.
This coffee tasted delicious on the first try: at grind setting 8 on the EK 43 and a 1:16 ratio, my cuppers detected craisins, watermelon, blackberry, wildflowers, guava, vanilla, and milk chocolate. It was really delicious, and I had I struggled to think what I could do to improve it. The extraction was mildly high, although well within range for a dense Ethiopia. Still, as the brew cooled it became more drying and mildly gritty, and I began to wonder if lowering extraction a bit would increase sweetness while minimize some of those tannic flavors.
Coarsening the grind up a half notch to 8.5, I brewed the same recipe again. This time, the brew extended 20 seconds longer than my previous brew at a finer grind size. Despite the extra contact time, this one did taste a little lighter and sweeter, with my panel detecting notes of hibiscus, honey, florals, and chocolate. Incredibly, the TDS reading for this brew was almost identical to the TDS of the previous brew; this wasn’t a total surprise because although the grind was coarser, which should limit extraction, brew time ran longer, which increased extraction.