Coffee Data Science
Coffee Grinder Showdown, Particle Distributions
In part 1, we found that the Rok grinder was better for espresso than the Niche. I wanted to understand why, so I collected some grounds particle distributions to search for clues. I looked at a few roasts as well as with and without the flow control disk in the Niche.
For all of these roasts, the grinds from the Rok pulled at a higher extraction yield than from the Niche.
This roast had a shift to the right for the Rok compared to the Niche, so the Rok was giving a coarser grind. I tried going coarser on the Niche, but the extraction yield dropped.
This roast had much similar distributions, but the Rok still was pulling higher.
The story was the same as before. I swept some grind parameters for this roast in part 1, and this was still the best. There is not a clear answer why the Rok would outperform the Niche.
The Rok didn’t have a control disk, so I thought I would check that. The flow control disk didn’t greatly change the distribution either. In this case, the Rok was a lot coarser, but when I tried to match the coarseness with the Niche, my extraction yield went down.
I also checked the distributions for another roast, and they were nearly identical with or without the control disk.
Unfortunately, these distributions confused me more than anything. They imply that the particle distributions only tell part of the story. Maybe, the particles themselves were of different shapes, which is what I explore in part 3.
If you like, follow me on Twitter and YouTube where I post videos of espresso shots on different machines and espresso related stuff. You can also find me on LinkedIn. You can also follow me on Medium and Subscribe.
Continue reading: https://towardsdatascience.com/rok-defeats-niche-zero-part-2-acdbae134018?source=rss—-7f60cf5620c9—4