Question: When planning out a roast profile, how can I go about subdividing the roasting phases to refine my profiling? For example, how can the phase from the start of first crack to the start of second be subdivided according to the results in the cup?

Are there any notable subdivisions to be made with any of the other phases (either after the start of second crack, or from 300 to the start of first crack)?

noah wrote:Question: When planning out a roast profile, how can I go about subdividing the roasting phases to refine my profiling? For example, how can the phase from the start of first crack to the start of second be subdivided according to the results in the cup?

Are there any notable subdivisions to be made with any of the other phases (either after the start of second crack, or from 300 to the start of first crack)?

I'm a strong believer in the "KISS principle," which of course stands for "Keep it simple, stupid " I see no point in introducing needless complexity when none is called for.

Therefore, I would divide up the roast cycle into the following, notable landmarks and periods:

(1) Charge Temperature (e.g. what is the temperature when you drop in the beans?)

(2) Drying phase -- this is the first part of the roast and begins with charging the roaster and ends at roughly 300F, depending on how you measure it

(3) Pre-First Crack

(4) Initiation of First Crack

(5) Period between initiation of First Crack and end of roast, if you don't roast into 2nd Crack

(6) Initiation of 2nd Crack, if you get that far

(7) Duration of 2nd Crack, if you get that far

(8) Termination of Roast, which may follow either 5, 6, or 7, above

With the proviso that stalling the roast at any time is inadvisable, the goal is to increase temperature at a relatively constant speed (which will differ during the phases) during #s 2-7, above. As a general rule, you will go relatively slow in the drying phase, faster in #3, and fairly slowly from the initiation of 1st Crack (#4) until the end of the roast.

In more direct response to your specific question in the first paragraph, I'm unaware of the particular importance of any specific part of the period between the onset of 1st and the onset of 2nd, however the overall duration is very important. With regards to the question in the second paragraph, stuff happens very quickly after 2nd crack begins, and unless you have a particular goal in mind or you like very darkly roasted coffee, you are probably not going to be going very far into 2nd in any event. All you can do is to time how long you have been in 2nd crack, and follow the temperature changes. These two things, easily measured, are the most important.

I have not heard of any subdivisions of the time after the drying phase until the start of 1st crack, although towards the end of this period you need to watch the amount of "heat velocity" you are carrying into 1st crack. If you are still adding lots of heat just before 1st crack starts, the roast will have a tendency to "get away from you," e.g. to progress at a very fast rate making the control of the period between onset of 1st and the end of the roast "too short," and hard to control.

ken

noah wrote:Are there any notable subdivisions to be made . . . . from 300 to the start of first crack)?

Ken Fox wrote:I have not heard of any subdivisions of the time after the drying phase until the start of 1st crack

Well, there's this one that talks about, among other interesting things, this:

ButtWhiskers on greencoffeebuyingclub.com wrote:Conversely, if a slow ramp to the first crack is taken, with a healthy pause in the range of 350-375F, there will be more caramelization and less Maillard and Strecker products. This results in a sweeter roast.

I'm not sure if the 350-375F range refers to internal bean temp or one that's measured with our bean TCs. I have not experimented with such a slow ramp, but that whole forum post is a very, very, good one IMHO.

Thanks. That was a very nice thread. Now if only that info was posted by someone who didn't go by the name "ButtWhiskers"...

Never knew what that name meant (now I googled it!) but what he wrote in that post is indeed interesting for home roasters; that's what I'm interested in(and grateful for him sharing what he knows!); his nickname is his biz.

coffee.me wrote:I'm not sure if the 350-375F range refers to internal bean temp or one that's measured with our bean TCs. I have not experimented with such a slow ramp, but that whole forum post is a very, very, good one IMHO.

Whenever you read something like the referenced post (from "buttwhiskers,") alarms should go off that make you think about the following issues:

(1) What is the "proof" of what is being said?

(2) What is the actual correlation between the chemical and physical changes cited, and what most people could taste in the cup? What sort of beans is this applicable to?

(3) Where is the temperature being measured?

(4) As a homeroaster, do I have equipment that is capable of being finely enough regulated/adjusted that I could even hope to follow whatever suggestions are being made?

(5) How am I going to test this idea personally so that I can confirm for myself whether it works, and if so, whether the time and effort necessary will be rewarded in the cup?

(6) Is such an approach practical for me to even try? If I'm, for example, limited to a batch size of 4 oz, to pick a number, how much time do I want to spend roasting coffee each week? Will the differences I might see as a result be worth the time and effort expended?

There are many beliefs and opinions that get circulated and recirculated in the coffee universe, including especially "online coffee." Some of these are true, but many of them, when subjected to real scrutiny fail even basic confirmatory testing. Three such opinions/beliefs that I have been involved in personally testing include the "deleterious effects of freezing on roasted coffee," "that rotary pumps make better shots than vibe pump driven machines," and that a shorter interval between the onset of 1st crack and the end of the roast makes for an inferior roast product. All of these things were scientifically tested with blind, paired, shots, with the assistance of Jim Schulman and a few other people. Of the 3 widely held beliefs, only the third was confirmed on blind tasting. I would submit that the vaguer the concept, the harder it is to test and to prove.

Therefore, I'd take most of what I read that infers subtle and poorly described differences, based upon a modification of technique (roasting or other) with large boulders of salt unless there is some actual "proof" behind it. Even if "true," the impacts may be small and as such may only be noticeable to a "super-taster."

ken

Thanks for the cautions, Ken.

That same poster, on a later post, goes on to explain:

ButtWhiskers on greencoffeebuyingclub.com wrote:IMO the most crucial things are:
. . . .
2) Once the bean temperature is at ~350, the temperature increase should be slowed to allow about 3 minutes in the 350-375 window. (the 'delta' is change per time unit, as in +7F per minute) Caramelization is when sugars bond with other sugars, which increases sweetness while inhibiting some later reactions that can create meaty or roasty (or even sharp burnt) flavors. Different beans have different amounts of different sugars, so there is no hard and fast rule for specific temperature or time, but each bean will have a sweet spot or two.

Sounds interesting and goes against the "300 to C1 ASAP" guideline I've been trying to follow so far. Does this poster roasting chemistry sound OK? At least in theory?

Buttwhiskers wrote:Conversely, if a slow ramp to the first crack is taken, with a healthy pause in the range of 350-375F, there will be more caramelization and less Maillard and Strecker products. This results in a sweeter roast.

I'm sorry, but I have to jump on this one hard. As far as I know:

-- Maillard reactions in coffee start at around 295F, and caramelization at around 400F. Maillard reactions use up sugars, amino acids, and water, leaving less of them for caramelization. Stalling the roast in the mid 300s would increase Maillard and decrease caramelization products.
-- The Strecker degradation is one of the later phases of the Maillard reaction, breaking down earlier Maillard products. In coffee roasting, it is associated with the production of Furfural and other sulfurous compounds. These are mostly associated with the dry distillate flavors, and therefore are presumed to originate mostly after the first crack (The Maillard reaction is a complex network of reactions with lots of pathways. Researchers knows when it starts, what compounds are in coffee at the end of the roast at various roast levels, but have no direct ways of finding out what compounds are produced at what times in the roast) .
-- Caramelization reduces the sweetness of sugar just as effectively as Maillard reactions. Maillard compounds taste toasty, savory, oaky, and nutty; caramel compounds taste of honey, caramel, vanilla or chocolate. None of these tastes is as sweet as those of the sugar precursors.

Stalling the roast in the mid-300s is a way for baking out the acids and creating predominantly toasty flavors. By flattening the acids out, the brew will taste relatively sweeter.

This recipe is more evidence for my suspicion that there are a some people who home roast because they dislike what most people consider to be good coffee. I sometimes think that they are the ultimate in irony: doing everything their sometime considerable talents allow to roast and brew coffee so it tastes like Postum, or the other roasted grain coffee substitutes, rather than like real coffee.

Thanks, Jim, now we're talking!

So ButtWhiskers is getting a little sweeter, toastier+nuttier, roasts but killing most of the "honey, caramel, vanilla or chocolate". Did I understand that right?

another_jim wrote:I'm sorry, but I have to jump on this one hard... The Maillard reaction is a complex network of reactions with lots of pathways. Researchers knows when it starts, what compounds are in coffee at the end of the roast at various roast levels, but have no direct ways of finding out what compounds are produced at what times in the roast

Well, that officially answers my question. My original post was motivated by reading through an older thread where some different profiles were tossed around by Mike McG, the main differences being how quickly or slowly the roast progressed through different temps between the onset of first crack and the end of the roast, however there was no report of how the different roasts compared. It begs the question of how the cup results differ between roasts that gently ease through first crack, and those that push through it going on to slowly finish the roast in the same total time.

Also, can a roast go too quickly through the 300-400 stage?

While the exact progression of compounds may not be able to be mapped, cant we generalize along the lines of, for example, from 400-430, you have the best chance of maximizing acidic fruitiness, from 440 on, chocolates, etc. The general principle of 300-400 gives bready, nutty, and 400 to second crack gives chocolates is well known. But cant we further break this down, or is it so bean dependent that it is a fools errand?