I'm interested in the concept I've seen here (especially from Neal/N3Roaster) of breaking down and manipulating the middle phase as two different "yellow to brown" and "brown to first crack" segments, but unsure of how one really judges when exactly the coffee is turning brown.

I feel like this is a similar situation as judging when exactly the coffee turns yellow, where the main thing that matters is that you apply your criteria consistently, but I haven't seen much direction on brown. For yellow, I go off of when I see almost no green anymore as well as when I start smelling the sweet toasty grain smell instead of grassy. But I feel like there's so many shades that the coffee can be called "brown".

I do not know the underlying reason for making that differentiation, yet it seems completely arbitrary to me...I'd rather use somethink like a temperature goal similar to what Artisan does(making 150'C 'dry end').

There's some guidance out there that says there are reactions that create sweetness during the time the beans are yellow, and extending the time while the coffee is yellow can add sweetness to the final product. Following that, while the coffee is brown before first crack there are reactions creating certain kinds of acids, and modulating that phase can increase or decrease acidity.

I personally don't use the 150C/300F dry end marker, I go off of color and smell, which is usually around 320F on my roaster (which kind of makes sense since I also don't get first crack until 400F, 15-20F later than a lot of other people I see)

I don't use 'brown' as a data point; I use yellow, which is usually at about 150 C. However, I have noticed that yellow does not last long. It quickly turns to what I call 'tan'. From tan onward, the Maillard reaction changes the shade of brown continuously. So, if 'tan' is the brown you seek, then that's fairly easy. But I can certainly understand how one would have difficulty when a roast is "Brown"; I know I certainly would be For what it's worth, my usual data points are yellow (Dry, 150), 1C-start (~200), 1C-end (about 1C-start + 17 degrees C, and only a secondary concern for me) and 2C-start (but I usually drop just before this happens).

I have experimented with double roasting (roast until yellow, cool and roast the rest later), and that definitely extends the yellow phase yet I did not notice any difference in flavor profile.

You should be able to use whatever bean temperature equals yellow in your situation and use that as guidance rather than a color that is painfully hard to judge.

There's a fragrance change as well as a color change there, so it might be easier to go with your nose while initially trying to sort out what "no longer yellow" really means, but once you nail that down for one coffee, pick a temperature and use that instead for everything else. On my machines that's 330F. Interestingly, if I take the blanch point on a decaf coffee using visible spectrum measurements (not really something you can do with your eyes, but I have a camera mounted to my roaster that gives me those measurements) I end up with 280-330 as yellow, 330-380 as brown before cracks, and 380-430 as the between cracks range, which all work out to 50F°. Unfortunately, I haven't yet had time to test how significant that 280-300 portion of that yellow range really is, but to answer the skepticism, you do get significantly different flavor changes out of manipulating the duration of each of those ranges (as well as the one from 2nd crack to the end if going that dark) so reasoning about them separately helps me narrow in on how I want to roast things with fewer tests.

Interesting! I do not yet understand how one could manipulate individual sections of a roast in mid flight and maintain a steady RoR decrease but it's an intriguing idea!

N3Roaster wrote:There's a fragrance change as well as a color change there, so it might be easier to go with your nose while initially trying to sort out what "no longer yellow" really means, but once you nail that down for one coffee, pick a temperature and use that instead for everything else. On my machines that's 330F. Interestingly, if I take the blanch point on a decaf coffee using visible spectrum measurements (not really something you can do with your eyes, but I have a camera mounted to my roaster that gives me those measurements) I end up with 280-330 as yellow, 330-380 as brown before cracks, and 380-430 as the between cracks range, which all work out to 50F°. Unfortunately, I haven't yet had time to test how significant that 280-300 portion of that yellow range really is, but to answer the skepticism, you do get significantly different flavor changes out of manipulating the duration of each of those ranges (as well as the one from 2nd crack to the end if going that dark) so reasoning about them separately helps me narrow in on how I want to roast things with fewer tests.

Was hoping you might see this topic and I could get some input straight from the horses mouth. There was a time a few months ago I was trying to play around with this timing after seeing your videos, and I did kind of settle on a temp -- 360F on my roaster while I normally call yellow at about 320F and first crack starts going at 400F. But every time I always second guessed my self thinking "is that actually brown? Is that light brown? Is it dark tan?" I don't think I really have nailed down an olfactory cue yet, but I'm thinking it'd be when it stops smelling as much like toasty grain that I smell when yellow happens.

For what its worth, I happened to take this image last roasting session, sometime after yellow but definitely before I would think about calling it brown

It's hard to tell in photos since lighting/exposure can make things seem not quite right, but I doubt you're far off from where I'd call it. I think you're on the right track with the fragrance, but in some ways this is just one of those things where it's probably easier to learn during a hands on workshop with a trainer. I'm a bit out of touch with the current state of SCA's Coffee Skills Program, but this used to be covered in the RP112 Intro to Roasting class, Diedrich's training that you can get if you buy their machines back in 1999/2000, and it was called out as a distinct thing in Rob Hoos' book which might be helpful (and has the advantage of still existing).

I've been experimenting with tuning the browning phase of my roast profiles, so I'm quite interested in the wisdoms being deposited here. Thanks for raising this question!

I've read a couple articles/postings that claim caramelization of the browning phase starts at about 170C/338F, and I've observed(measured) a thermodynamic 'inflection' at 170C during a roast of 'dried' greens, which seemed to exaggerate the effect. Post #39 here: Green coffee rehydration

I've been using an admittedly funky roast profile that parks the BT at a level temperature during the browning phase. At 160-165C the tasting notes were distractingly grassy/vegetal - no sir, I didn't like it. Lately I've settled into 180-185C for 4-6 minutes before abruptly switching to the development phase.

Based on what others have posted here, I'm wondering if there's some flavor to be mined between 'dry end' at 150C, and 170C where caramelization really kicks in?