Mount Kenya, among the world’s most impressive peaks, sets the backdrop for much of the coffee growing region in central Kenya. In all of Africa, only Kilimanjaro tops the maximum elevation of Mount Kenya.
On its southwestern slopes, Nyeri has emerged as a county well-recognized for its superior quality coffees. This Crown Jewel is one such thrilling Nyeri-grown selection, arriving nice and early to our warehouse.
The coffee was grown by contributors to the Ndurutu factory in Kirimukuyu, about halfway between the county capital city and the county border on the east with Kirinyaga. Ndurutu wet mill was built in 1996 and incorporated into the Rutuma Farmer’s Cooperative Society in 2005. The cooperative society now has around 4,000 members in and around Karatina in Eastern Nyeri, of which about 500 contribute to the Ndurutu factory.
Coffee cooperatives in Kenya have a mixed history. Smallholder farmers under British colonial rule weren’t even allowed to grow coffee on their own land until 1934. A decade later, all smallholders were required to join cooperatives, a move that weakened their standing with the Coffee Board, whose directors were mostly large estate owners. However, in 1963 when Kenya gained its independence; plantation lands were nationalized and redistributed, loans were meted to cooperatives to build washing stations, and the government mandated all coffee to move through the Nairobi Coffee Exchange weekly auction.
Many cooperatives still have little outlet besides the NCE, which can fetch high prices but can also leave smallholders with few alternatives to a convoluted system that guarantees very little money actually returns to the farmer. In 2008, a “Second Window” option opened allowing direct exports. While not without drawbacks and abuses, using this option has proved advantageous and traceable for many growers and buyers alike.
Royal Coffee purchased this delicious Ndurutu, and many of our Kenyan offerings, directly from Kenya Cooperative Coffee Exporters, a farmer-owned company founded by a group of cooperative unions in 2009. We’re pleased to have this sparkling clean Ndurutu Kenya on the menu, and thrilled to be a sustainable part of the supply chain.
A dry coffee by all accounts, even for Kenyan options which frequently read below 10% on arrival, this coffee is also of high density and large screen size. It’s a AA, meaning screen size 18 & 19, and while graders are quick to point out that this is not a quality designation, AA coffees frequently earn the highest prices at auction.
Built on common Kenya cultivars, this lot includes SL-28 and SL-34, produced by Scott Laboratories in Kenya in the 1930s. Regarded as the best of the SLs in terms of quality and resilience, both varieties are Bourbon derivative cultivars, though from different lineages: SL-28 was developed from a drought-resistant variety originally cultivated in Tanganyika, a territory that makes up part of modern day Tanzania; it’s generally considered to be of the highest quality but is not very productive compared to other commercial Arabica varieties. SL-34 is a Kenyan mutation originally found near Kabete, and excels at lower elevations. Both of these SL variants exhibit bronze-tipped leaves on the newest growth.
Joining the classics are two relative newcomers. Ruiru-11 was developed in the mid-1980’s as the result of attempting to make an SL-28 more productive and resistant to Coffee Berry Disease and Leaf Rust by crossbreeding with varieties as disparate as Sudan Rume (for quality) and Catimor (for disease resistance), among others. In response to qualitative feedback, the Coffee Research Institute retraced the steps to creating Ruiru-11, attempting to improve cup quality without compromising disease resistance. Since 2010, the new cultivar – called Batian – has trickled into production.
Here are two roasts on the Ikawa with strikingly different results. Ikawa Roast (1) has a slowly decreasing fan speed until first crack, then the fan speed steadily increases afterwards, through the end of the roast. The slower fan speed (and therefore more heat retained in the roasting chamber) led to a lower First Crack in time and in temperature. Ikawa Roast (2) follows the same temperature profile line, but the fan speed profile is set up in a V pattern, reducing until Maillard Begins then increasing until the end of the roast. The higher fan speed means more heat is applied to reach the set temperature profile. The V form profile also results in a higher and later First Crack that you are more likely to see in an air roaster.
While First Crack occurred with a 5 °F difference, both roasts kept a total of 59 seconds of post crack development time. This also meant that Ikawa Roast (2) was 15 seconds longer than Ikawa Roast (1). On the cupping table Ikawa Roast (1) was bright and showed some signs of savory tomato while Ikawa Roast (2) was jammy and showed lots of dried fruit flavors.
This unusually dry and dense coffee gave me the impression that it would do best with a shorter roast with enough heat to reach first crack at the beginning of the curve. After one minute of post crack development time, the coffee expanded and puffed up and the trier started to release more red fruit aromatics. On the cupping table, this coffee was extremely bright and had similar flavors to Ikawa Roast (1).
The dry nature of this coffee made me hesitant to extend the total time too much longer, but the cupping table tells me that this coffee can take it. If I was to roast this coffee again I would start with a higher charge temperature to reduce my drying time, keep Maillard time the same, and use this added time to extend post crack development time and increase my end temperature as well.
Unless otherwise noted, I follow a set standard of operations for all my Behmor roasts. Generally, I’ll use the 1lb setting, manual mode (P5), full power, and high drum speed until crack. Read my original post and stats here.
This coffee was a peach to roast, and a roasted peach on the cupping table too. Lemon and lime brightness came out quite readily, and nobody was struggling to find tasting notes for this lot.
This week features three very dense coffees, and this is the driest one of them all, with the lowest water activity. This is a great thing for shelf stability, but with an implied detriment to Maillard reactivity. Regardless, I didn’t have any trouble getting sweetness out of this coffee. While it cracked slightly later than average, this coffee had a very distinct start of first crack, and I engaged P4 just a little after first crack.
Rolling through development was no chore, and while my roast loss percentage was at 10.9%, this is due in part to low moisture content. The coffee did not present as underdeveloped on the cupping table, and I think I might just enjoy a cup of it in a few minutes.
If I were to change anything, I would have dropped the power slightly to P3 and continued roasting for a few more seconds. The internal development of this coffee was slightly less than I had expected. The results were nevertheless very enjoyable.
This was, hands down, the most fascinating coffee I brewed this week, but perhaps not for the reason you think. From the get-go this clean Kenya was complex and delicious, full of bright tropical fruits, cola, blood orange, and brown sugar. It even had a certain crispness, like watermelon or cucumber, that was really lovely. I brewed it on a Kalita at a 1:15 ratio and a slightly coarser grind than usual. Upon taking a TDS reading with my VST refractometer I was left dumbfounded. It read: 1.73, incredibly high for a total dissolved solids count, with a correlated percent extraction reaching 27.61%. This reading was so extreme that I re-brewed the coffee to the same parameters to make sure it wasn’t a fluke; when the readings came up the same I checked my refractometer’s calibration and re-calibrated it to 0.0, even though it was still accurately calibrated. Then I brewed a different coffee, last week’s Chiapas from Rosalba Fuentes’ family and measured TDS, only to find that despite my suspicions the refractometer was reading consistently and accurately.
I coarsened the grind up to 10 on the EK43, lengthening the brew just to a 1:16 ratio and cooling my brew water to 195F Again, it had anomalous TDS readings: 1.75 with 29.75% extraction, putting it just on the border of what’s physically possible. More surprising still was that the coffee still tasted delicious despite these crazy reading. In regards to cupper preference this brew was the standout, full of caramelized sugars, apricot, orange peel, rosehip, and melon.
Still suspicious of my equipment, we set up a side by side cupping of this Kenya and this week’s Costa Rica from Coopedota; Chris took a sample of both cups right after the break, around the 5:00 minute mark, and again when the coffee had been steeping for about thirty minutes. We found that the readings for the baseline coffee (the Costa Rica) were within normal range for a cupping and that the Kenya’s reading was inordinately high; more importantly we found that both cups were extracting at a similar rate, so although the Kenya read higher throughout the readings for both went up proportionally. This meant that the refractometer was reading the Kenya Kirimukuyu’s extraction accurately but on a higher scale than we’re used to. In fact, it was reading on a scale that doesn’t fit with what we know to be possible; only about 28-30% of the compounds in roasted coffee are soluble; no matter how aggressively one extracts, water simply can’t get any more than that out. So could I possibly be achieving such high extraction rates?
There was only one option left to me: I had to push extraction well beyond what’s physically possible to prove that we were getting false readings. I tightened all the way to setting 6.5 on the EK, boiled the crap out of my water, and used a 1:18 ratio. I also poured pretty aggressively with the Bonavita kettle (I had previously been using Fellow’s Stagg) which may have led to channeling – my brew time went down significantly. TDS read out at 1.57 with a 30.04% extraction. Though technically impossible, I wanted to be absolutely sure; so I brewed again with an even tighter grind, 6 on the EK, but this time used the Stagg kettle to prevent channeling. Finally, I had radically impossible numbers: a TDS reading of 1.85 and a 35.4% extraction.
Perhaps the most bewildering part of all of this is the fact that the coffee tasted really lovely, even with these ridiculous TDS readings. We’ve speculated a bit about what might have happened; it seems likely that the coffee has something that refracts light differently than how the refractometer expects to read the light bouncing off of particulates. I for one am convinced that the coffee is full of gold, and if we were only to pan the spent grounds we would all be rich.
Though we may not yet know the cause of these incorrect readings, this experience proves yet again that we should rely on refractometers to dial in our coffee.We must instead rely primarily on our palates to find out what is most delicious. And despite all errant readings, this coffee is spectacularly delicious.