Hi all.

Bean size and shape could be important to the development flavor during the roast. A larger bean would have a larger gradient of temperatures between the center and the surface of the bean (measured as BT). This difference could add to a large range of flavors.

Same could be said about its density, which would affect the speed by which the center of the bean heats up. The center of a high density bean would heat faster (higher heat conduction).

These two factors would affect the choice for the best roasting profile. For example, the center of a small bean size would heat up faster and would have a shorter maillard reaction. If the roast time were long, then the flavor could get negatively affected (overcooked). On the other hand, the center of a bigger bean would take longer to heat up and maybe it could take a longer roast time without much penalty. But if a bigger/less-denser bean had a short time, then flavor in the center of the bean would not develop. In between the center of the bean and its surface, there is room for various stages of flavor. The surface temperature can be measured and the inner temperature could be estimated during a roast.

During a roast, there could be a best compromise of surface/inner temperature to give the best overall flavor depending on the density and the size of the bean.

Density can be measured. The size and shape can be categorized. For example, the Kenyan beans designation for size and shape is:

AA
AB
PB
C
E
TT
T
UG

I am not sure if the Kenyan bean system could be used for other beans or not. Is there a system that can be used for roasting?

Cheers

The center of a high density bean would heat faster (higher heat conduction).

Is there evidence of this somewhere?

I am not sure if the Kenyan bean system could be used for other beans or not.

Screen size?


I have thought for quite some time that each bean has a "specific heat capacity" that governs it's optimal roast parameters. I believe how the heat is applied (convection, radiation, conduction) in relation to MET and stage of roast is also very important in optimizing roast parameters.

This thread would be well served if merged with: Better understanding Harder vs Softer bean roasting

~j

jammin wrote: The center of a high density bean would heat faster (higher heat conduction).

Is there evidence of this somewhere?

Yes. there is what is called thermal conductivity of materials.

http://en.wikipedia.org/wiki/Thermal_conductivity

Also see conduction.

http://en.wikipedia.org/wiki/Conduction_%28heat%29

"As density decreases so does conduction"

The method to transfer heat could be important for softer beans. For example, if using convection (air) and choosing a low temperature for the air, then the beans would heat up faster (because of convection) but at a safer temperature, which could avoid damage. I think the damage to soft beans is done when using a high temperature (like a very hot drum).

Cheers

The momentum factor of the large Pacamara is good example of the where size matters.

Arpi wrote:Yes. there is what is called thermal conductivity of materials.

snip "wiki links"

"As density decreases so does conduction"

Cheers

Thanks, but I already took (and aced) 400 level thermodynamics in college. I was hoping you had some studies done directly on coffee beans. It appears there are some available online, but not for free.

~j

Everything else equal, more porosity = lower thermal conductivity. This will hold for any material, since conductivity is just how well the phonons move. A pore = no material = phonons have to go around it or change thermal transfer mechanisms and go through it (maybe via convection). Either way, it will slow down heat transfer.

Hi

There is probably nothing about coffee out there, but maybe wood would be a similar cousin. Hardwood is usually more dense than soft wood. If you look at the R value (the reciprocal of thermal conductivity) of hard wood (denser) versus soft wood (less denser), you would see the difference there.

http://en.wikipedia.org/wiki/R-value_%28insulation%29

softwood = 0.25
hardwood = 0.12

It would get very complicated if we were to consider the bean volume expansion also during roasting.

Cheers

There is probably nothing about coffee out there, but maybe wood would be a similar cousin.

http://www.sciencedirect.com/science/ar...7407004645

http://www.sciencedirect.com/science/ar...741100094X

http://books.google.com/books?id=4gEG96...ns&f=false

Do not forget about water. It has a much higher thermal conductivity than coffee beans. A large factor when considering how best to roast a DP bean.

~j

thanks

Maybe huge roasting operation do something like this but for perfection I feel a sample roast or two is all there is needed to fine tune a profile. The same bean from the same farm will differ year to year depending on weather, soil changes, and whatever else farmers need to think about to keep their crops going so fine tuning a profile is always happening with my roasting year to year.

My experience with island beans, which is not that much, tells me a longer roast time does not help things out. I go against what many say and use a moderately high drop in and a somewhat short drying phase then take my time getting to first crack and stop shortly after 1st finishes and keeping my total time under 16min for my roaster. I have also come to understand that each roaster and how it works varies, so if you find a perfect profile for your quest it would have to be tweaked to do the same thing on my Diedrich and so on.