I've been very interested in the Clover coffee machine ever since I saw one demonstrated
at Gimme Coffee in Ithaca. Shortly thereafter I posted an explanation of the machine
When I posted a link to this explanation on alt.coffee,
several group members were extremely skeptical that coffee could be properly brewed in the Clover's abbreviated time cycle. Specifically, could one extract a good-tasting, SCAA-recommended 18-22% of the coffee solids when limited to a 30-45 sec dwell time? Some folks speculated that Clover coffee would taste weak and insipid. The only possible explanation we could come up with was that the Clover's filtration system (4.875" dia filter, 70 micron screen, ~0.4 bar vacuum) might support a much finer grind than drip, press or vac pot, and this might facilitate a quick extraction.
As a matter of fact, I've always wondered about this point. And when I heard that many Clover baristas were updosing the dry coffee (using a brew ratio of 8-12% instead of the recommended 5.5-6.5%), I assumed they were compensating for underextraction by overdosing.
Well, that was over two years ago, and Clovers are still so rare and so expensive that I don't have one on my kitchen counter to experiment with <insert sobbing sounds here>. But I finally decided to get off my ass (slightly) and run an extraction experiment with the "Poor Man's Clover," an Aerobee Aeropress. Using the inverted technique that I first learned from Scott Marquardt on Coffeegeek.com, the Aeropress does a very good job imitating a Clover brew.
Sometimes the hardest step in doing an experiment is deciding WHAT IT IS that you're going to test. I often get stuck at this point. There are usually dozens of combinations of parameters, but one never has time to do them all. The teeny weeny graph below is the result of today's experiment. Between running the dinky experiment and writing it up, this is almost 6 hours of work!
I left the data in, but the graphical part is the important thing. It shows a series of Aeropress extractions using two different grind settings on my Versalab M3 (fine and very fine) and two different dwell times (45 sec and 180 sec). The dry coffee dose (6.5 g) and water used (109 g @ 200F) were kept as constant as possible. This is close to the "ideal" SCAA 6% brew ratio. The 6.5 g dose was selected because it was the same coffee mass per sq. cm. of Aeropress filter area as a typical 32 g dose would be in a Clover.
Each press was stirred gently for 10 sec at the beginning, left alone, then pressed at the end of the appropriate dwell time. Each cup of coffee was given a quick taste. The pucks were carefully oven-dried and weighed to determine solubles yield (what percent of the puck was washed away into the beverage).In terms of the numbers, I had no trouble extracting recommended soluble yields in 45 seconds.
In terms of taste, they all offered good body and pleasing Yrgacheffe aromas. In general the 180 sec extractions tasted "overextracted": harsh, acrid flavor notes in varying degrees. Next time I will try slightly coarser grind settings.
Note: opinion follows:
This test is just one tiny data point, but after doing this I have no trouble believing that paired with a high quality grinder, Clover brewers can perform adequate extractions with 40-45 sec dwell times.
I also believe that given the excellent control offered by the Clover, motivated baristas will tweak the extraction parameters and find "sweet spots" offering excellent tasting coffee. It may be that a 45 sec extraction can never be tweaked to taste exactly like a 3-4 min extraction, but that doesn't mean it can't taste great.I also believe that people who try this with grinders that are less than top-grade may be disappointed. In other words, like everywhere else in coffee, the grinder is key.