more air theory (we’ll move on to liquid at some point)

July 8, 2005 – 1:33 am

ruminating further on the physics of hot-air roasting…

if my i-roast indeed applies heat to green coffee more consistently than any other machine under $4,000, and the general method is also known as “fluid bed roasting” for presumably optimizing the way heat circulates throughout the process, that would indeed support my theory that longer roasts are simply unnecessary and indeed less than optimal. the time it takes for the hot air method to consistently warm the entire bean interior is the rate at which the roast can be accelerated. i’ve heard some gurus say that the disadvantage to hot air (control) has been mitigated with the programmable-stage i-roast. but that’s different from saying that the hot-air method requires a shorter roast period and even results in better-tasting brew. what was thought to be a disadvantage is actually an advantage, in other words, as long as the roast is consistent.

if X is a unit of heat, for example, and Y is a unit of time, a hot air roaster produces a circulating “cocoon” of X enveloping said beans while a drum roaster produces X in a uni-directional trajectory. the more consistent the cocoon in a hot-air roaster, the lower the strength of X (temperature) must be and the less Y (time period) you need to roast evenly. in a drum, meanwhile, the one-way travel of X requires a higher rate (temp) of X and more units of Y so that the gradual application can make up for the lack of immediate consistency. hot-air machines, of course, can produce too efficient a cocoon of X at too high a rate — most, including my old fresh roast did this, giving rise to the hot-air-is-too-fast claims. the beans were burnt outside when they were barely reaching completion inside. essentially, all we’ve needed all along is a programmable hot-air machine such as the i-roast. that allows a roaster to fine-tune the point at which X and Y are adequately leveraged for an even, yet not-necessarily-long roast. the presumed superiority of a longer roast time disappears.

and indeed, if the i-roast is as efficient as some say, such a long roast period only chars the bean’s interior and overly prolongs the stages. not only is there no need, the new evidence supports shorter roasts that play to the strengths of the air process.