On Roasting Coffee: Charge Temperature
Definitely not recycling generic “roaster” images from Hart Coffee.
If there’s one term in coffee circles that has extreme fluidity (and thus extreme and undue confusion), it is charge temperature.
Sure, most people are referring to the same part of the roasting process; the point at which green coffee is added to the roaster, however, what is meant by this term is vastly different from roaster operator to roaster operator.
To many roasters that I have spoken to, charge temperature simply refers to the moment when the temperature readout on the front of the machine matches some pre-determined value.
But this is not charge temperature. Rather, this is merely the air temperature inside your drum.
Charge temperature is critically tied to the thermal energy of the system and thus, just because your bean probe temperature readout says 400°F (204°C) this does not mean that the thermal energy of the roaster is the same as was during your last 400 degree charge.
The thermal energy of your roaster is dependent on how you get to that temperature mark and, critically, if your thermal energy is different from batch to batch, you will have problems replicating roasts.
But a quick primer; what is thermal energy?
Simply put (very simply put [for a slightly more thorough run-through of thermal energy check out this previous post]), thermal energy in coffee roasting is the retained heat stored in your machine.
Think about it with this analogy; imagine you’re relaxing at the beach on a blustery but sunny day. You decide that the direct sun is too hot, so you put up an umbrella to block the direct rays/energy of the sun.
As you stand in your now shady spot, will the sand burning your feet or the air around you cool quicker?
The answer, as your blistering feet will likely tell you, is that the sand will remain hot longer. This is because of thermal retention. The sand is going to retain its energy/heat longer than the air around you.
Similarly, while the cooling effect of the air inside the drum may tell you that your temp is all set at your 400°F drop point, your thermal energy has the potential to be still too hot (or in some anomalous cases, too cold).
All this is to say that in order to consistently roast batches either in progressive profiling or in production, you must be able to have a charge temperature that effectively resets the bean probe, drum temperature, AND thermal energy of the drum.
“Okay, I get it, thermal energy is important and I shouldn’t just rely on the readout. But how do I ‘reset’ the drum?”
Well, one effective way I’ve found to do this is to bring the bean probe temperature to idle at your intended drop temperature for 2 minutes with airflow through the cooling bin. The tighter you can make your idle, the more easily you can replicate turnaround from batch to batch. I’ve also found that using periodic spurts of airflow through the drum to set your idle point can help to equalize the system. Perhaps going so far as to allow the temp to rise to set point, killing the input energy, then using airflow to “parachute” the momentum back to setpoint before kicking the input back on to min and closing airflow through the drum. Repeat for 2 mins.
This procedure may vary from roaster to roaster, the main point here is that you must be consistent with your inter-batch protocols.
Going further, and to effectively prove that this drum-reset methodology was accurate, I set up an experiment where a proxy for thermal energy could be monitored.
I took an infrared temperature reading of the thickest piece of metal on the faceplate of the roaster and found different ways to approach the drop temperature.
In 50 batches recorded, despite the fact that the bean probe consistently read “400” for every drop, the batches where the “proxy” read higher at the charge point turned around at higher temperatures (and subsequently required less energy throughout the remainder of the roast, and looked to “take off” until thermal energy lowered) whereas batches with lower proxy readings turned around at a lower point.
While not a perfect measurement of stored thermal energy, this proxy reading allows for an interestingly accurate approximation. When I measured in different places around the faceplate, the numeric values of the proxy changed, however, the trends of the data remained constant.
All this to say that if you are experiencing issues with your batch consistency - i.e. fluid turnaround times, momentum shifts from batch to batch with the same coffee, etc - a likely issue point is with your interpretation of the term charge temperature. Perhaps look into your interbatch protocols and make changes to attempt to make more consistent coffees.
