Cup Testing Coffee by Screen Size: Ethiopia Yirgacheffe

An old article concerning cupping separated screen sizes of roasted coffee to see if there are differences in quality related to coffee bean size.

Today I am motivated by the most un-scientific curiosity to screen a roasted Ethiopia Yirgacheffe and cup the different bean sizes separately. The inspiration comes from eating the roasted whole bean coffee, and occasionally finding a seed with incredible flavor. It seemed that these were smaller beans … is that possible? As you know, the traditions in some coffee origins to sort coffee by screen size into various grades makes little sense in terms of cup quality. I mean, it does matter that all the seeds are healthy, correctly processed and non-defective. But larger size does not make better coffee. Supremo and Excelso grades in Colombia, and AA versus AB in Kenya make little sense. In fact, I thought with this Yirgacheffe that it was the opposite case, and I wanted to see if I could reproduce my anecdotal results

To do this, you start with a set of coffee test screens. These are not cheap! But I find them useful and I make it part of all my final green coffee sample evaluations to test the screen size. Now, these screens are really made for green unroasted coffee, but there is nothing wrong with trying to screen a roasted batch … the same principles apply. My screens are very simple hand-shaken square screens. You can get a setup with a fancy test sieve shaker and round screens that are more accurate, but that is not needed really. In the foreground is the bottom collector screen.

I chose to separate the Yirgacheffe using 18, 17, and 15 screens. These are all measured in 64ths of an inch and these are round hole screens, the standard in coffee. I also have a full set of slotted peaberry screens but that is another story.

The results were that an overwhelming amount were 18+ screen, while the 17-18 screen size held 18% of the coffee, with a fair amount of peaberry. The 15/64ths screen held 7% of the coffee, and was almost entirely peaberry.

A closeup of the 15 screen contents, which would actually hold 15-64ths and 16/64ths.

Interestingly, there were a few brokens and “tri-beans” that made it past 15 screen. Tri beans are what they sound like, a coffee cherry with three healthy seeds, which end up sorta pie-shaped. A couple rat-turd shaped peaberries made it through too. Overall, the percentage of brokens was not enough to be measured as a defect in the rating of this coffee.

Okay, set ’em up on the table to cup. (incidentally, what is left over in the trays is roaster tests, not the coffee ground in the cups.)

I made 3 cups of each and evaluated them in 2 rounds with a comparison cup of the blended/unscreened coffee. The results were interesting to me:

This is a Fair Trade, Organic Yirgacheffe from the Oromia group, basically the giant Ethiopian “Co-op of co-ops” . It’s a nice, uniform-processed washed coffee. The green coffee screens as 16-17. The results were as follows:

  • 15/64ths screen: tangy acidity, more “prickly” than the other cups. Lighter body, a little more citrus brightness, a little less deep/dimensional.
  • 17/64ths screen: cupped very close to the unscreened control cups, but with a little flatter acidity.
  • 18+/64ths screen: more discernible roast-oriented flavors here: caramel, very mild milk chocolates, perhaps not more body in terms of quantity but it was more noticeable.
  • Unscreened Control cup: as the coffees cooled, this had the most dimension, and as I keyed in on the taste attributes I could differentiate in the screened cups, i was also able to switch to the unscreened cup and locate them.

Results: well, I debunked my own notion that any particular screen size was the locus of magic flavor crystals in this lot of Yirgacheffe. I also prove to myself again (not that I need to) that bigger is not better. It makes a case for “estate grade” coffee of mixed screen size, provided that the coffee is free from defect, that the processing is exceptional, that the cherries are all ripe, etc etc. No real revelations here, but an easy cupping test to perform with some personally relevant information coming out of it. And that’s how I spent my afternoon… end of coffee Show and Tell! –Tom

Post script:

Incidentally, here is perhaps the worst-ever cupping photo of myself … right up the nostrils. Anyway, I am breaking the crust on the cups here. –Tom

3 Responses

    1. Hello, there are a lot of people offering coffee grading courses that result in certification. Search online for “Q grade courses near me”. We don’t offer one ourselves.

      Best,
      Dan

Leave a Reply

Your email address will not be published. Required fields are marked *

 

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Reddit
Facebook
Twitter
LinkedIn
Email