Coffee Subscription

Journal #4 - Urubyiruko Trials

More trial and error, more learning!

So here is a concept I have been thinking about the past few weeks:

So, in roasting, double walled drums do a good job of allowing you to really increase the flame without scorching or tipping the coffee since the drum wall doesn’t get as hot as quickly. Now the Giesen drum wall is this hybrid drum that has two layers, one being stainless steel, one being cast iron. I have heard that tipping is really hard in a Giesen so I thought maybe I could see how much energy I could put into the first part of my roast and see what happens.

I also have been struggling with these two Rwandan coffees I have. I know the roasts can be great, the sample roasts I received were really delicious, but not translating well to my roaster. Both being moderately dense, I knew they could take the heat and I should see what that will do.

On most attempts, you definitely get underdeveloped notes. Some are VERY pronounced, others just have a drying note on the finish.

I also saw someone on Instagram post a curve from a Loring where they had a very short first phase and a drawn out maillard. I have never been great at achieving something like that, but hey, if at first you dont succeed…

Here is my progressing through the Rwanda Fugi Urubyiruko roasting. I put three roasting graphs from my Giesen on here which you will see below.

rwanda roasting curves giesen

1061 - Blue
1095 - Yellow
1104 - Red

These roasts happened over different roast weeks. Originally I wanted to push tons of heat and go for acidity and unrefined sugar so thats what I was attempting with 1061. It was not pleasant. It wasnt so underdevloped it broke itself on the table while cupping, but it was thin and had a very unrefined acidity with good amount of dryness. That is when I went the opposite direction and decided to drag out the first phase and try to get some heat into the center of the bean before maillard. And that was slightly better, but honestly that could also just be attributed to a longer maillard phase. With that, I still had the dryness, still a little thin, but not as bad as 1061. So going off my thoughts mentioned above, and trying to have short first phase and long maillard, I went back to 1061 and just cut all the heat I could to let it coast through all of maillard and see if I could drag it out. You’ll see the first phases are very similar, but the Maillard phase in 1104, is a 1:06 longer than 1061! And you could really taste it in the cup. Definitely not white sugar sweetness, but much more pleasant mouthfeel and not as much drying on the backend. Overall a much more enjoyable drinking experience. And a good step in the right direction for nailing better profiles.