McGee's temperatures are right for a stove top pan. But the very high pressures inside the bean's cell walls prior to the first crack changes all the reaction temperatures. The water doesn't actually boil until the first crack. For instance, barley for beer is malted at 130 to 160 (iirc); but the same reactions only begin in a roast when the bean hits 300F. You can smell the caramel and candy aromas, and they're not happening until the first crack starts (good time to smell a first time coffee roast -- the crack is a bit like breaking the crust when cupping, a good aroma forecast). I think the roasting chapter in Illy has some of the changed reaction temperatures; if not, a library trip for the Clarke and Vitzthum edited collections of research papers will do it.

Thank you. With that, and related to the original question... ideally, how should the first crack occur to maximize aroma? I've been testing roasts side by side whereby all other things equal, I'll ease into first (~390-405) @5F/min, or move on through as usual to final degree (425F city) @ 10F/min. Unfortunately, I'm not finding consistent results across all coffees.

Good question; I don;t really know.

I'm always on full brake going into the first crack, since I'm switching from the fast ramp to the first to the slower finishing ramp. I've had nasty, grassy results going through the first crack fast and then braking; so now I always brake before it starts.

endlesscycles wrote: how should the first crack occur to maximize aroma? I've been testing roasts side by side whereby all other things equal

In the Illy book, they suggest that the aroma created is proportional to the pressure build up. An the pressure build up is proportional to the rate of change of BT up to a limit.

They say:

"a build up of pressure inside the bean is important for the generation of sufficient aroma" (page 184).

A faster rate during the First Crack will give you a louder pop (higher pressure). And a lower rate will give you a quieter pop (lower pressure).

They say:

"temperature control of the heating gas, permitting not only temperature profiles with moderate differences, but also a sufficient pressure build up inside the bean, has to be the objective of roasting"

To me, "sufficient pressure build up inside the bean" translates in to a BT rate of change to some degree.

Cheers

Time to experiment with a pressurized roaster

That PhD dissertation referenced earlier was talking about a roaster with an oxygen-free environment later in the roast! That's your next project I think!

This is an article by Swiss Federal Institute of Technology and Nesle published in J Agri Food Chem in 2008, which discussed time/temperature, and published after the Illy book. The beans were roasted in G45 20kg-batch Probat and 100g fluid bed lab roaster. Sixteen flavor compounds were measured. So yes, the rate of change has a definitive impact on flavor.

This is the synopsis:

http://pubs.acs.org/doi/abs/10.1021/jf8...istoryKey=

The article has details about the material and method, result, discussion, etc, etc. For the chemically inclined regarding this article, contact me off line.

"When the drum roaster and the fluidizing bed roaster were operated in the so-called temperature profile mode, that is, along the identical development of coffee bean temperature over roasting time, the kinetics of aroma generation were similar in both processes."

As I would expect, despite lore to the contrary.

This article is flat our wrong. I have done dozens of roasts with identical PIDed bean profiles on the identical airroaster and gotten differences that even non-coffee drinkers could readily identify in triangle tests

All I did was add or remove the insulating sleeve on the roast chamber, so that the ETs changed by a factor of 50F. I have also tasted some differences, but not as reliably, changing the tuning constants on the PID controller, so the the ET either oscillated or ran asympotically (asymptotic is better).

The beans roast on a heat gradient from the highest to the lowest temperature in the roaster. The lowest temperature, the heat sink, is likely to be the bean center. The highest temperature is probably where the heated air enters the roaster or first meets the beans. This can be anything from 50F to 250F higher than the bean center.

The taste of the coffee depends on its chemistry, not on how fast, when, or where that chemistry occurred. You can get the same taste, the same chemistry, with different roast times and different roasters -- my 12 to 13 minute drum roasts are indistinguishable from my 9 to 10 minute air roasts, since I've worked very hard and for many years to get them both exactly the way I like. And you can get different tastes and chemistry with the same roast times -- as in the example of insulated and uninsulated roasters I gave above.

Anybody who has the gear can confirm this in one afternoon's roasting session.

I think the rate of change may not be critical after certain point, but there is a period where certain rate of change is desirable.

The chlorogenic acid for example follows a first degree rate of degradation. With current roasting technology, certain time has to pass for these groups of acids to degrade to decrease the associated astringency. This may be the reason we don't have roasting that finishes the roast in for example, 30 seconds.