To apply this information to your dialling in, extracting and explanations in store is all together another big fish that we'll tackle in the not too distant future.
Now, to the green bean..
There are not infinite shapes or sizes when you look at one coffee bean, so we will always focus on that one random coffee bean that you pick out, as it represents the whole.
This coffee bean is a SL28, SL34, and it is washed process, so it's more dense than a natural process coffee. It is, lets say, 1cm squared in surface area, and is more grey green than the usual yellowish green. This is our green coffee.
We have this little humble coffee seed in between our fingers as we hold it up to the light to have a good look at it and we try and think about what we want him to be when he grows up. Will we push him hard consistently through his short life in the drum roaster, giving him a certain crispness and structure, or will we be more lenient with the heat and allow a slow and soft absorbtion of heat in the beginning, gradually increasing the heat as if to push this little coffee seed up the hill towards first crack?
The question is always - 'what would be best for him?' and 'Am I choosing the best path?'
A dense and relatively round bean like a washed SL28,34 would be roasted between the two examples given above - It would be entered into the drum when it's around 220*C with the gas/burner low and the air flow very low.
The beans that are entered into the drum have a moisture content of around 11% and what we are trying to do with the low airflow is allow the 220*C of heat be absorbed by the coffee's moisture and transfer through the bean because of that moisture. Essentially using conduction to initiate convection.
If the airflow was high a lof of heat energy would leave the drum and would not be replaced by heat, as we had the burner on low. The hot moist air going out the exhaust would be replaced by a lot cooler air depeating the purpose of what we're attempting to do - transfer the right amount of heat, the right way, at the right time.
The reason we kept the burner low during this time was because we want to utilize the bracket of time where the coffee is absorbing heat via conduction and is moist. This initial heat is to get into the center of every coffee bean in the drum, and once that has happened we will move to focusing our attention to convection heat transfer. That is, using hot air and hot moisture to cook.
A great example of what I'm talking about here is about roasting vegetables. Who here begins their roasting process in the oven by covering the top of the baking tray with aluminium foil or a lid? Hopefully everyone. This allows the vegetables to essentially steam for a period of time, transfering a lot of heat to the center allowing it to cook close to the same speed as the outside.
(Note; the lid/aluminium foil only stays on for around 10 minutes)
The time in which we keep the gas low and the airflow low is different. Between 30 seconds and 1 minute we will give it a lot of gas, probably as much as we can! Whereas it'll be a few minutes before we even think about increasing the airflow, and that is because it takes a lot of momentum to get coffee heating up at a speed that we want it to. Think about it like this; input affects output.
The speed at which coffee is heating up is called the Rate of Rise which can be and is, abreviated to RoR, which you can see on the roasting curve above as the blue squiggly line. It is related to the numbers on the X axis on the right - these are degrees celcius per 30 seconds (*C/30sec), meaning that 1 minute into the roasting process the beans were heating up by roughly 20*C every 30 seconds and by the time they got to first crack the speed at which they were all heatnig up had slowed to around 11*C every 30 seconds. This decreasing in temperature speed over time leads to a 'decreasing RoR' which is what a roaster usually needs to produce a roasted coffee that has sweetness in it.
As seen in the image above there were two airflow changes in this roast, and the burner was very high for most of the roast. This was becuase we wanted to highlight the floral attributes of this coffee along with the highly delicious and sought after berry cola acidity that comes with Keyan coffees grown in soil with a high level of phosphorus.
The bean curve (other blue curve) goes down, dips, and then starts heading upwards. The tip is becuase this data is coming from termperature probes inside of the drum and they were very hot before we put room temperature beans in there. The turn at the buttom is called the 'turning point' and is where we actually start getting true information from the probes - around the 1 minute mark.
A few minutes into the roasting process you can see the beans begin to turn from green to hay yellow and then brownish followed by brown. This initial change to brown is cause by the maillard reaction and it is amino acids within the bean reacting with reducing sugars because of heat and the biproduct is both a colour change to brown but also new flavors that are born out of this reaction. Note, acid is reacting with sugar, so sugar content is reducing, a necessary evil of roasting coffee. Evil because sugar is excellent, but necessary because in order to make coffee soluble/grindable we must also make it softer and we do that by roasting it.
Around 150*C the beans go from experiencing an endothermic reaction to an exothermic reaction, and this calls for an airflow change usually, and an attempt to slow the speed at which the coffee is increasing in temperature (RoR).
At a point the coffee in the drum has enough pressure within each bean from the heat that the internal moisture has turned to steam and needs to get out, like a bomb. This bomb is harmless, and actually created the platform for a whole miriad of flavours to be formed. This is called first crack and it is where most people begin the 'development' phase. We disagree with this a bit and think that development has more to do with the beginning of the process than the end..
It is the time and temperature after first crack where the visual colour of coffee is made. The 'lightness' or 'darkness' of a coffee is not though, as it is the average colour after it is ground up and measured.
For filter coffees we tend to take the coffees around 8*C to 10*C above the first crack temperature, and we do this in around 1 minute after first crack has begun. For espresso coffees we roast them a bit darker and a bit longer, as they need to be more soluble to get a high extraction and overwhelming acidity is not very pleasurable for most. Espresso roasts will often have between 1:30 and 2:00 in the drum after first crack has begun.
After each roast we take a 20g sample, grind it and then measure it with our colourmeter to make sure that it is within the parameters of quality that we set for ourselves. If its not, we put it aside to wait to be cupped before it goes anywhere.
That little SL28, SL34 bean has gone in, copped some high temperatures to the point of steaming so much he cracked, and came out the other side a more tanned, delicious SL28, SL 34 than before. He grew up into an excellent little coffee bean.
How we approach roasting..