Intro by Charlie Habegger
Terra Coffee SAS is a young producer group, established in 2016, with a narrow focus on developing high quality coffees alongside select producers in the Huila and Nariño departments, and sharing them with the world. The small company manages one single producer association in each department where they work, “Ecoterra” in Nariño, with 140 producer partners, and “Terra Verde” in Huila, with 120.
For Terra Coffee SAS as a whole, quality in coffee is very rationally understood as a direct pathway to well-being for volume-limited, small coffee farming families. Driving their business model is an understanding that quality results from small harvests have direct impacts on not just the farm owner, but the many dependents on each small farm, including young children, older adults, and the women of the household performing essential labor that often goes unpaid. By increasing quality and placing microlots in the market, Terra Coffee SAS plans not only to increase prices to growers and their families, but also increase their sense of pride in the details of their work.
The 32 producers contributing to this coffee synchronized on a precise fermentation model which, in the own words of Terra Coffee’s founders, was “designed to obtain the best fruit and citric notes of Nariño, as well as the characteristic sweetness of the region” and “leave us with a brilliant, juicy acidity”. Indeed the municipalities of Cartago, Taminango, and San Lorenzo are all just south or east of La Unión, a very well-known corner of the Nariño Andes for big-bodied and sweet coffees. So this is a direct effort to surpass an already high status quo.
The particular processing method devised here involves two fermentation steps at different phases post-harvest. The first occurs in the whole cherry: fresh-picked fruit is bagged and allowed to ferment for 36 hours, for the mucilage fibers to break down and the sugars to peak. The second phase is more traditional but still very specific: the softened cherry is depulped and fermented dry in open tanks for another 60-70 hours. Finally the fermented parchment is lightly washed and dried in the sun. Each of the 32 producers exhaustively monitored temperature throughout to ensure each batch of coffee was homogenous with the others in order to create a focused, precise final profile. The resulting lot is exceedingly clean, complex in acidity, lightly creamy, and with a distinct berry juice to the sweetness. As designed!
Northeastern Nariño is uniformly high altitude, dense, and rugged, with regular ridge tops surpassing 2400 meters. San Pedro de Cartago, Taminango, and San Lorenzo are all technically valley towns, and still hover around 1600 meters each, with farmland increasing in altitude on all sides. The participating farms are only slightly larger than Colombia’s national average, at 2-3 hectares each, and all processing is done at a very small scale.
Green Analysis by Nate Lumpkin
This coffee from Colombia comes to us with below average density, and somewhat above average moisture content and water activity. Its screen size is well sorted, with most of the coffee clustered into sizes 16 through 18, and only small amounts falling outside of that. Lower density coffee like this one may be scorched by high heat, so consider a gentler approach, especially early in the roast. At the same times, its slightly elevated moisture content may cause a longer drying phase, and its elevated water activity can cause a faster Maillard phase. Consider looking at Evan and Candice’s notes for the roast for guidance.
Caturra is a single-gene mutation of Bourbon, first reported in Brazil in 1937. Its mutation characteristic is its short stature, which allows for denser planting and easier picking, and therefore a higher overall yield. After development in Brazil it was released to Guatemala and spread widely throughout Central America. It is one of the parents of the Catimor group of coffees, which were developed to create a dwarf plant with greater rust resistance.
Ikawa Pro V3 Analysis by Chris Kornman
As of September 2020 we are running all Crown Jewel Analysis roasts on an Ikawa Pro V3, using the most recent app and firmware version on “closed loop” setting.
I decided to start off the new year by putting in a little time with the Ikawa. Nate has quickly developed a knack for pithy roast analysis on the machine, but he was on vacation and so I jumped at the chance to reconnect with the little roaster.
I sample roasted the coffee for approval manually, and found that roast to be especially aromatic, and was a little surprised that, a day off roasting, these Ikawa profiles were a little less fragrant. Nevertheless, each roast produced some distinct flavors. Overall I preferred the extended Maillard profile, but had positive experiences with each of the roasts.
The short, hot, and fast profile hit first crack a little late and only managed 38 seconds of development afterwards. It was bright, tart like a green apple or slightly underripe citrus. It lacked depth and didn’t offer much in the way of sweetness. Some interesting banana and cinnamon notes cued me into the probability that it was a little underdeveloped.
The Maillard +30 profile was the strongest performer, though still had a slightly delayed first crack and resulting short development. However, the cup was balanced, and the strong orange flavors evoked marmalade and tang (in a pleasant way) which melded well with a honey-like and butterscotch sweetness. Ample body and a clean finish rounded off the experience. It seems clear that a little bit of extra time during the color change phase goes a long way with this coffee.
Lastly, the low airflow profile despite a nice-looking roast stage ratio balance, ended up a little imbalanced. It had a really interesting flavor profile including cranberry sauce and red grape, but a little nuttiness seemed to undermine the sweet notes found in the Maillard +30 profile. The body was full but clashed a bit with the acidity. Goes to show that roast stage percentages aren’t the only predictor of cup quality.
I’d surmise, based on the quick results here and the unique processing method, that you could conceivably treat this coffee a little more like a pulped natural or honey coffee than a traditional washed in the roaster. Taking it just a little easier during Maillard should give you some solid results in the cup.
You can download the profile to your Ikawa Pro app here:
Roast 1: Crown Standard SR 1.0
Roast 2: Crown Maillard +30 SR 1.0
Roast 3: Crown 7m SR LowAF 2
Diedrich Analysis by Candice Madison
Whew! Well, I can safely say that we aren’t in Kansas anymore. Unless you are – stranger things have happened! For the most part when we talk about roasting coffees for production, we (I mean, me), look briefly at the processing method, but swing firmly to green coffee metrics for clues on how to approach this coffee in the machine. For the most part, this isn’t a bad way to look at things generally. You absolutely must take the processing method into account when you roast, but the green coffee metrics are far more specific when considering the chemistry and thermodynamic behavior of beans in the roaster. All of this were truer in time gone by, by which I mean Ye Olden Times, or, as my niece has informed me, ancient history – anything more than a decade ago.
I’m sure you have seen the work of Evan and Chris regarding a comprehensive flowchart of the different permutations of processing in an industry that has long since outgrown the simplified misnomers of the ‘big three’ – natural/sun-dried, pulped natural/mucilage or fruit-dried and washed/parchment-dried coffees. If not, check out their article here – it’s a great primer and evergreen resource.
I must admit that I was intrigued to taste this coffee. I adore coffees from Nariño – I can’t lie, I love most Colombian coffees, but I haven’t yet roasted a double fermentation from that region. Looking at this coffee’s processing specifics on paper, it may read as a washed coffee, with a couple of extra fermentation steps. However, if you look again, the first step of fermenting the fruit on the seed, is more of what we would see in a naturally processed coffee. The next step, depulping of the fruit and a controlled stint in the fermentation tanks, are more traditional.
If we look at what these do to the green bean’s overall chemical composition going into the roaster, we can see that drying the fruit on the seed ostensibly lends a sweeter characteristic to coffee. Not by imparting sugars from the fruit into the bean, a common misconception, but usually because the fruit is left to ripen longer, allowing more sugars to develop in the bean itself. The long fermentation soak will add moisture to the bean, whilst leaching some of the sugars, and allowing the acidity to be further revealed through the alchemy of fermentation.
Looking at the green metrics confirmed my hypothesis. A higher than average moisture level, and higher than I would have surmised, but lower overall density. I would expect a lower density from a natural coffee and a higher moisture level from a washed coffee. With a tight screen size and some notable water activity, I decided to hedge my bets and treat this coffee as a pulped natural.
With a natural coffee, I usually try for low and slow at the initial charge, a lot of heat for a short amount of time before stage 2, and then use the air at 50% and declining gas applications to eke out the roast before first crack. Trying to combat the high moisture level as fast as I could, I upped my charge temperature to 380 F and started with 90% gas (5 on the dial), I could have started at 100% and not scorched this roast, to be honest, and I should have. I turned the gas up to 100% after the turning point, but it still wasn’t enough to combat the initial heat uptake resistance from the water content of the bean, for my liking. These are issues I’m looking at regarding the regulator, supply and such – as well as user stats – I’ll let you know how I go!
A shorter time in the Maillard stage with much less water and all of that heat, meant that this stage, even with declining gas applications and open airflow, was much faster than I had hoped. By much faster I mean 2 minutes, which on a machine of any size is going to make a difference to your roast. Hoping to develop what I had, after a quiet, but determined first crack, I let the coffee have a much longer time in post-crack roast than I usually do. I wanted to make sure that any acidity left over would be complimentary and not overwhelming, as the Maillard stage had been shorter than anticipated.
The cup did not disappoint. So much so, I filtered all the cupping bowls and drank it gleefully! I can, hand on my heart, say that this is as close to drinking Cherry Limeade, as I have even known! The fruit was overflowing in the cup, cherries, lime, grapefruit zest, and fruit blossoms, as if I were walking through a meadow. Sweet high-toned vanilla bean taste is rounded out by notes of English toffee and rich roasted walnut. The body was silky smooth and coating. All in all an adventure, but no less delicious and inviting a cup of coffee in spite of the wild ride!
Quest M3s Analysis by Evan Gilman
Unless otherwise noted, I follow a set standard of operations for all my Quest roasts. Generally, I’ll allow the machine to warm up for 15 minutes until my environmental temperature reading is at least 250F, weigh out 150g batch size, and begin roasting when I’ve reached my desired charge temperature. Read my initial post here and my updated post here.
Just like my compatriots, I had a fun time roasting this coffee. It is at this point that I should tell you that I consider novelty to be the driving force in what I consider ‘fun.’ This coffee took me way off guard, and I really enjoyed the experience.
Nate was correct in his estimation that a gentle approach is needed with this coffee. For reasons I no longer recall, I went head on and applied high heat directly to the forehead of this roast. Though it was my first roast of the day, and my environmental temperature probe was still reading quite low at charge (230F), I think I could have been a bit more gentle than a 394F charge temperature. Consider charging at the same temperature you expect first crack for this coffee for an easier time.
Starting with 10A and full fan, then cutting fan at turning point, this roast really started to cook, taking all the residual heat from the drum in the process. At 275F / 2:35 I saw just how quickly things were progressing, added airflow to 3 on the dial, and reduced heat application to 7.5A. This worked rather well, and I decided to attempt to really slow things down through Maillard by reducing heat application to 5A at 310F / 3:15, much earlier than usual. This worked too, but not quickly enough. At 375F / 5:10 I turned fan to full, then decided to cut heat entirely at 390F / 6:10, a little after first crack. This did enable me to get ample post-crack development without my temperature going haywire, but I didn’t spend as much time in Maillard as I would have liked – only 37% of the roast.
Upon tasting, I found abundant sugars, and a cup that just kept on improving while it cooled. While I got some berry notes from the break, the cup was all lime, brown sugar, and sweet plum. If anyone out there has tried mamey sapote fruit, I got a lot of that from this cup. Again, a very brown-sugary flavor. I could have dropped this coffee a little sooner, but I am really enjoy it as a drip and as an AeroPress – those extra suspended solids make this a thick and deliciously sweet cup! I would definitely recommend this as an espresso. Read on for more from Nate on brewing this coffee..
Brew Analysis by Nate Lumpkin
Now this coffee was a real pleasure to brew! I reached this time for the Kalita Wave and the Saint Anthony C70, which I thought would both be a good way to showcase this coffee’s sweetness, and I was not disappointed. For my brew analyses lately, I’ve been gravitating for a lower dose and coarser grind, with a ratio of 1:16.6. I find this makes a coffee with a really broad and balanced flavor profile—though many of these coffees are delicious at higher and lower concentrations, I like this ratio a lot when considering its qualities.
I like the C70. It’s a sturdy and handsome device and the depth of the bed it creates can make for some higher extraction rates compared to flat-bed devices. This brew finished at 3:18 with a TDS of 1.3 exactly and an extraction of 19.06%, a little lower than I expected. In the cup though, it was fresh, clean, and refreshing, with notes of red grape, peach, blackberry, and orange blossom, a honeycomb sweetness, a pleasant nuttiness like raw almond, and a fudgy, milk chocolate viscosity. I really loved this brew, and drank the whole thing.
I chose the Kalita figuring it would create a somewhat lower extraction, and because it’s similar to a lot of common brew devices out there, with its nice wide bed and flat bottom. It brewed through a little faster than the C70, at 2:29, but I was surprised to find it had a somewhat higher TDS at 1.44 and an extraction of 20.16%. I suppose the wider bottom let more solids drip through, rather than get stopped at the bottom of the cone. Anyway, its higher extraction was apparent in the cup: I tasted notes of plum, peach candy, bubblegum, rose, caramel, and chocolate sauce. It was a little bit thicker and more candy-like than the previous cup, with a viscous and coating body, and a more developed florality.
These were both great! If you’re looking for a juicy, light cup, try a lower extraction, and if you’re looking for something heavy and candy-like, try to boost the extraction a little bit with a flat-bed brewer. We’re also considering brewing this one on espresso here at The Crown, and I am very excited to see how that turns out.