Matt Perger Builds His Own Roaster - Page 5

Discuss roast levels and profiles for espresso, equipment for roasting coffee.

#41: Post by yamhill »

Jake_G wrote:If the pickgaurd is prepped, that shouldn't take too long...

Setup, too? Or just neck, bridge and electronics?
Someone crossed my streams... Either way it shouldn't take long...


#42: Post by yamhill »

What about humidity? Wouldn't an optimal roaster also measure humidity -- possibly in and out of the roasting chamber?

I know that some at least some of the newer production roasters measure temperature and humidity of the incoming air. Moving between humidity extremes could make the "same" profiles quite different -- unless accounted for. I also saw a roaster that measures roaster exhaust moisture.


#43: Post by N3Roaster »

Some of the conversation here around the validity of transferring profiles across machines is pretty disgusting. It's easy to disparage the skills of others and dismiss experimental work if you're not contributing anything constructive and if you've never done similar work yourself. Regarding the exercises in Can You Taste the Roasting System, yes, there are limitations, including some different from the ones already mentioned (though one that's being picked on is, I believe, down to a failure here to understand what the methodology actually was and is being misrepresented in a worse light than how things happened). The real limitations don't take away from the findings, but rather make those findings more encouraging for anybody who needs to put in the work to get that capability across their own (often more similar) machines, that this is something that is broadly possible. Yes, applying this does mean putting in that extra work and getting more rigorous with the exact machines you're trying to work with, but it can be done and has been done successfully at many companies.

Personally, I had already done a more rigorously controlled version just between two machines that I have several years prior. The result of that work was a calibration curve (so I'm not just matching a few points, but rather, matching measurements after that calibration adjustment throughout the whole roast) that allows me to design roasting plans on the smaller machine and get the same results (interior/exterior degree of roast measurements, percent mass loss, 2 of 5 testing with myself, other staff, and customer panels) from a larger production roaster. My own work had its own set of simplifications which I fully expected I'd need to go back and refine if it turned out my approach still seemed like it might work. I wasn't sure this was possible and scrapping the idea was certainly a possibility in my mind at the time. Instead, it worked so well that it's now been automated and baked into my profile design process. Years later I continue to be astonished by just how well that works across a broad range of roasting styles. The work is continuously tested as new coffees come in and as I try different approaches to roasting. It's still holding up as valid and I'd encourage anybody who doubts this to get out of their arm chair and try doing the work themselves before issuing such shallow dismissals.
★ Helpful