Brazil is a force to be reckoned with in the coffee world. The country grows roughly 40% of the global coffee supply and is a juggernaut in the marketplace. A late season rain, a prediction of a bumper crop, or a summer drought can dramatically affect the harvest and throw the futures markets for coffee into turmoil. Weather during the 2014 harvest was poor, and crops across the country suffered. However, large volumes and generally favorable rains prior to the most recent harvest in June – September of 2015 have produced a superior crop across the board. The 36055 – Brazil Nova Aliança 16+ is a perfect example.

Brazil’s longstanding traditional harvest method allows coffee cherries to ripen and then sun-dry in the cherry husk with minimal processing, as opposed to the fermentation and washing process used by their neighbors in Colombia, Bolivia, and Peru. Access to fresh water can be difficult to attain in the coffee highlands (especially during the dry season, when coffee is harvested) which makes this low-water-consumption method of processing coffee attractive to many producers. Brazil is unsurprisingly the country of origin of the preeminent coffee processing equipment manufacturer, Pinhalense, who stake the claim to the invention of a hybrid processing method known locally as “cereja descascado” (literally “de-husked cherry”) and internationally referred to as “pulped-natural” or “semi-washed.”

36055 – Brazil Fazenda Nova Aliança 16+ comes to us from the Paulino da Costa family’s farm in Minas Gerais, Brazil’s large southern agricultural state. The farm, which was established in the 1850’s, is run by the fifth generation of the family, continuing a longstanding tradition of coffee production. Half of the farm’s roughly 210 hectares of land is devoted to coffee trees, while the rest is divided between pasture and preserved forests. The farm’s owner is also an agricultural engineer, whose involvement in both oversight and day-to-day procedures is evident in the quality of the cup.

Processed in the traditional “natural dry” or “cherry-dried” method, 36055 – Brazil Fazenda Nova Aliança 16+ is a prime example of the kind of clean, pleasantly fruity dry-process coffee that can be produced with the watchful eye of an attentive farmer. We’re thrilled to have added it to our lineup for the first time this season.



It should come as no surprise that larger farms in Brazil have processing down to a science, and 36055 – Brazil Fazenda Nova Aliança 16+ is no exception. In fact, this is a really nice example of an exceptionally graded coffee. The screen size, as advertised, is no smaller than 16 which should make for a very even roast regardless of heat application. The beans are of average density and relatively low moisture at 10.4% so they may absorb heat rapidly after first crack. That combination of low moisture and medium density will yield a high solubility once the coffee is roasted and ground for brewing. Try dialing back your dose, or coarsening your grind setting if you find the coffee is brewing too heavy of a cup for your taste. The low water activity reading (0.50 aW) indicates the coffee was evenly dried and should have a very healthy shelf-life as green coffee provided storage conditions are appropriately cool and dry.



The low moisture and medium density of 36055 – Brazil Fazenda Nova Aliança 16+ can make this a fragile coffee to roast. It will have a tendency to dry out faster in the drum and become porous and brittle quicker after first crack. For coffees like these, having a moisture meter on hand can be beneficial. If you do not have a moisture meter, the roaster can look for a delayed and hardly audible first crack. While some might make assumptions that the green coffee is inferior because of its unusual response in the drum, those judgments should be curbed until the coffee is cupped and evaluated for flavor.

In this case, the coffee cupped very well, with a nice balance of fruited acidity and caramel sweetness. In both roasts displayed above the major difference between them is that PR-0122 had less initial energy, resulting in a decline in the rate of rise much sooner than PR-0121. My intention with PR-0122 was to roast a more traditional profile with a longer overall duration time. Many believe that longer profiles create a bigger body and smoother all around flavor. In this case, the coffee retained some acidity, but was accompanied by a dryness on the finish that I associate with a baked coffee.

One major factor in roasting beyond heat transfer, is air velocity. If you are able to control airflow in your machine, I would recommend reducing air flow during the drying process and maillard reaction to get better results in the cup. Reducing Air Flow will help retain humidity and heat in the drum, while reducing the rate of dehydration in the bean. On the Probatino there is a damper on the exhaust stack that is difficult to manipulate while you are roasting and I do not recommend it. If your machine does not have the ability to manipulate airflow, increasing your batch size is a great way to decrease air flow in the drum.



For the 36055 – Brazil Fazenda Nova Aliança 16+, I wanted to try something a little different. As noted above, this is a very consistent coffee, and one that has undergone rigorous quality control. I wanted to see how consistently I could brew this coffee, to continue the chain.

I brewed the two roasts a number of times, and compared the two most similar brews. Using a Bonavita kettle set to 93C°, and an identical dose of 22 grams of coffee for brewing in a Kalita filter drip, I brewed our PR-121 with 330.4g and PR-1222 with 330.5g of water respectively, resulting final brew times of 3:03 and 3:05. Each brew was subjected to 45g preinfusion for 30 seconds, with the first brew (PR-121) being poured at 5.18g/s and the second (PR-122) at 4.75g/s. My differing rate of pour may have had an effect on the final TDS result, which in itself is an interesting thing to note, though this was admittedly not my goal.

The first brew (PR-121) resulted in a higher percentage of dissolved solids, though not by leaps and bounds (1.30 (17.54% extraction) as compared to the second brew 1.25 (16.9% extraction) in the second brew). Looking at the above metrics, the first brew should possibly contain less dissolved solids; shorter brew time (by two seconds) and a faster rate of pour should have assured a weaker brew, intuitively. The potential for so many interfering variables (grind distribution, agitation, etc.) means we cannot truly assume that outcome. Coffee is not always intuitive, after all. I knew there must be something else going on with these two brews, even though the flavors in each were very consistent with one another.

Reading above, you can see that this outcome was presaged by Jen; even the most consistent coffee can be temperamental. The roast represented in my first brew (PR-121) was more soluble, either inherently or through operator error. For all the baristas and brew nerds out there: if you find confounding results, there is always a reason!