Walk into a roastery while a batch is running and you will see someone watching a screen with a line on it. Not the temperature line, though that is there too. A second line, usually noisier, that shows how fast the temperature is climbing rather than what it currently is.
That is rate of rise, almost always abbreviated as RoR, and for a modern roaster it is the primary control surface. Two roasts can hit identical temperatures at identical times and produce completely different coffee, because the shape of how they got there was different. RoR is what makes that shape visible.
The concept is worth understanding even if you never roast, because it explains a category of coffee failure that has no visible signs and no obvious name. The bag looks right, the color looks right, and the cup is missing something.
If you want to taste what a well managed curve produces, browse our most popular roasts and pay attention to sweetness rather than roast character.
What Rate Of Rise Measures
A probe sits in the bean mass and reports temperature continuously. Bean temperature climbs from the moment the beans hit the roaster until the moment they drop into the cooling tray.
Rate of rise is the derivative of that curve. It answers how many degrees the beans are gaining per unit of time right now, usually expressed as degrees per minute or degrees per thirty seconds.
Early in a roast, when cold beans meet a hot environment, RoR is enormous, often fifty or sixty degrees per minute. As the beans warm and the difference between their temperature and the roaster's environment narrows, the rate naturally slows. By the end of a roast, a healthy RoR might be five to ten degrees per minute or less.
So a normal RoR curve starts high and descends steadily toward zero. The shape of that descent is what the roaster is managing.
Why The Curve Should Always Be Falling
The central principle in most modern roasting practice is that RoR should decline smoothly throughout the roast and never increase.
The physical reasoning is about consistency of energy delivery. A bean is being chemically transformed, and the reactions driving flavor development, Maillard browning and caramelization, respond to how much energy arrives and how fast. A smooth declining curve means the energy input is tapering in a controlled way and the reactions proceed at a rate that develops sweetness and complexity.
An erratic curve means the energy delivery is erratic. Reactions speed up and slow down, different parts of the bean develop at different rates, and the resulting cup lacks coherence.
A curve that goes back up is the most serious problem, and that is what roasters call a flick.
What A Flick Actually Is
A flick, sometimes called a crash and flick, is a specific sequence that usually happens around first crack.
First crack is when moisture and gas inside the bean build enough pressure to fracture the cell structure, producing an audible popping. It is also an endothermic event, meaning the beans absorb energy to get through it. That absorption pulls heat out of the system, and RoR drops sharply. That drop is the crash.
Then first crack finishes. The beans stop absorbing energy for the phase change, and simultaneously the roast enters an exothermic period where the beans start releasing heat of their own. If the roaster has not reduced the applied heat in anticipation, all that energy has nowhere to go but into raising the bean temperature faster.
RoR spikes back upward. That is the flick.
On a graph it looks like a check mark: the line falls off a cliff and then kicks back up. It is easy to spot once you know to look, and it is one of the most common defects in the roast logs of inexperienced roasters.
What It Does To The Coffee
The crash and the flick both hurt, and they hurt differently.
During the crash, the beans are getting very little energy at a moment when they need it. This is the beginning of development, the phase where sugars caramelize and the flavors you actually want get built. Energy starvation here produces what roasters call baking. The coffee spends time at temperature without the chemistry advancing properly, and the result is flat. Not burnt, not sour, just empty. Papery. The sweetness that should be there never formed.
Then the flick delivers a sudden burst of energy to a bean whose interior is now well behind its exterior. The surface takes the hit. You get uneven development, with the outside pushed further than the inside, and the cup shows a roasty, slightly harsh character sitting on top of an underdeveloped body.
Together they produce a coffee that is simultaneously dull and rough. Drinkers usually describe it as flat, or as tasting burnt without being dark, or as having a strange thin bitterness. It is one of the most frustrating defects because nothing about the beans looks wrong.
How Roasters Prevent It
The fix is anticipation. The heat reduction has to happen before the crash, not in response to it.
A roaster watching the curve knows roughly when first crack will arrive based on the bean, the batch size, and the profile they are running. They begin backing off gas in the thirty to sixty seconds before it, so that when the endothermic event hits, the system is already in a lower energy state and the drop is gentle rather than a cliff. Then, as the exothermic phase begins, there is no excess energy waiting to cause a spike.
Airflow is the other lever. Increasing airflow through the roast chamber carries heat away and also evacuates the chaff and gases released during first crack. Managing airflow in coordination with heat lets a roaster smooth the transition without cutting energy so hard that the roast stalls.
Batch size matters too. A roaster running too small a batch in a large machine has a thermally unstable system that reacts violently to adjustments. A roaster running too large a batch cannot get enough energy into the mass and will crash hard at first crack no matter what they do.
Taste the result of a controlled roast and compare it to a bag that tastes flat.
Where Heat Transfer Method Comes In
How a roaster delivers heat changes how controllable the curve is, and this is where the machine architecture matters.
A drum roaster heats through three paths: conduction from the metal drum wall, convection from hot air moving through the drum, and radiation from the hot surfaces. Conduction is the problem child. It is slow to respond, because the drum itself is a large mass of steel that stores heat. When a roaster cuts the gas, the drum keeps radiating stored energy for a while. That thermal lag makes precise curve management harder and makes flicks more likely for anyone who is not very experienced with their specific machine.
Air roasting, or fluid bed roasting, transfers heat almost entirely by convection. Beans are suspended in a stream of hot air with no significant hot surface in contact. The thermal mass of the system is much lower, which means changes to the heat input show up in the bean temperature quickly.
The practical consequence is responsiveness. When the crash begins, a reduction in energy input registers almost immediately rather than thirty seconds later. That makes the descent easier to keep smooth and the flick easier to avoid. It also means there is no hot metal wall to scorch beans during the moment when the exterior is running ahead of the interior.
None of this makes good roasting automatic. A careless air roast is still a bad roast. But the control loop is tighter, and a tighter loop is easier to keep on target.
Reading Development Time Alongside It
RoR does not get evaluated alone. It sits next to development time ratio, which is the percentage of total roast time that occurs after first crack begins.
A typical range is fifteen to twenty five percent, though there is no universal correct number and it varies by coffee and intent. What matters is that the development phase received appropriate energy, and RoR is how you know whether it did. A twenty percent development time with a crash and flick in the middle of it is not the same as a twenty percent development time with a smooth declining curve, even though the number on the log is identical.
This is why roasters who only track time and temperature are working with incomplete information. The summary numbers can look correct while the underlying curve was a mess.
Why This Matters To Someone Just Drinking It
You will never see a roast curve for the coffee you buy. But you can taste the consequence, and knowing the mechanism makes the tasting more useful.
If a coffee is flat, meaning it has no obvious defect but also no sweetness or dimension, baking is a strong candidate and a crash and flick is a common cause. This is different from stale, which also tastes flat but comes with a papery, oxidized quality. And it is different from under extraction, which tastes sour and thin rather than dull and empty.
The practical test is to brew the same coffee several ways. A coffee that tastes flat as a pour over, flat as an immersion, and flat as espresso is probably a roast problem, because you have varied everything downstream and nothing changed. A coffee that tastes flat one way and fine another is a brewing problem.
If you land on roast, that bag is not going to improve and no adjustment will save it. The information worth keeping is about the roaster, not the brew.
All images shown in this blog are sourced from pexels.com.