Rate of rise (RoR) is the single most useful metric for understanding what's happening inside a roast while it's in progress. It tells you not just where you are, but where you're going — and whether you're getting there too fast, too slow, or on track.
This guide covers what RoR is, how it's calculated, what a good RoR curve looks like, common problems and how to recognize them, and why tracking RoR across roasts is worth the effort.
What Rate of Rise Measures
Rate of rise measures how quickly bean temperature is increasing at any given moment during a roast, expressed in degrees per minute (°/min or °C/min).
If bean temperature is 180°C and one minute later it's 192°C, the rate of rise over that interval is 12°C/min.
A roast is never at a fixed rate — the curve is always changing. Early in the roast, after charge, the beans are cool and the drum temperature is high, so RoR climbs steeply. As the roast progresses, the beans absorb more heat and approach equilibrium with the drum, and RoR naturally declines. This declining-RoR profile, where temperature continues rising but the rate of rise falls throughout the roast, is the foundation of good roast development.
RoR is always calculated from the bean temperature curve (BT), not the environmental temperature (ET). Environmental temperature reflects drum or air temperature and is useful for setup and consistency checks, but RoR as a development metric refers exclusively to bean temperature.
How RoR Is Calculated
RoR is a derivative — the instantaneous rate of change of the bean temperature curve. In practice it's calculated as a rolling delta:
RoR = (BT_now − BT_n_seconds_ago) × (60 / n)
Where n is your interval in seconds. A 30-second interval (n = 30) means:
RoR = (BT_now − BT_30s_ago) × 2
The interval affects how smooth or noisy the result is. A shorter interval (15s) responds faster to real changes but amplifies sensor noise. A longer interval (60s) smooths the curve but lags behind what's actually happening. Most roasters use 30 or 60 seconds as a working interval.
First Crack calculates RoR from imported .alog files using the data point series, applying the same interval-based delta method. The RoR curve is recalculated on import — if your Artisan interval was set differently, the First Crack display normalizes to a consistent basis.
What a Good RoR Curve Looks Like
A healthy RoR curve has a few consistent characteristics:
Declining throughout the development phase. After the initial charge absorption period, RoR should be steadily decreasing from first crack through drop. This gradual deceleration indicates controlled heat delivery — beans are receiving enough energy to develop without accelerating into an uncontrolled finish.
Smooth, not jagged. Sharp spikes or sudden drops indicate sensor noise, airflow disturbances, or drum loading inconsistencies. A smooth curve is easier to interpret and react to in real time.
Positive at drop. RoR should still be greater than zero at the drop event. If RoR reaches zero or goes negative before drop, the roast has stalled — beans are still in the drum but no longer gaining temperature, which typically means underdevelopment regardless of elapsed time.
Consistent across batches. For the same bean at the same roast level, your RoR curves should overlay closely. If they don't, something about your setup — charge temperature, batch size, drum speed, airflow — changed between sessions.
Common RoR Problems
The Flick
A flick is a sharp upward spike in RoR near or just before first crack. Bean temperature suddenly accelerates instead of continuing to decline.
First crack is exothermic — the beans release heat as cell walls rupture. A small uptick in RoR around first crack is normal and expected. A large flick is a different problem: it usually means heat input was too high entering the development phase, and the exothermic energy from first crack compounds an already-fast trajectory.
Flicks often produce roasts that move through development faster than intended, resulting in underdevelopment (too little time between first crack and drop) or scorching if the roaster tries to compensate with abrupt airflow changes.
Common cause: Insufficient heat reduction before first crack, or an airflow setting that's too low to clear the heat being released.
Fix: Reduce burner output or increase airflow in the 30–60 seconds before anticipated first crack.
The Stall (or Crash)
A stall is when RoR flattens out or drops toward zero, usually during the Maillard or development phase. The bean temperature curve levels off — the roast is no longer progressing.
A stall almost always means under-roasted coffee. Even if the beans reach a target temperature, a roast that stalled getting there will typically taste baked, flat, and lacking brightness. The beans were held at elevated temperature without the energy flux needed to drive the chemical reactions that create complexity.
Common cause: Reducing burner output too aggressively after charge, or an unusual batch size or ambient temperature that wasn't accounted for.
Fix: Increase heat input before RoR reaches zero. By the time RoR flatlines, you're already behind — watch for a rate of decline that's steeper than expected and intervene earlier.
Excessive Early RoR
Very high RoR early in the roast (above ~15–18°C/min in the drying phase) indicates the beans are absorbing heat too aggressively. This can cause uneven development — the outside of the bean develops ahead of the interior, producing tipping or scorching on the bean surface.
Common cause: Charge temperature too high, or burner output left at full beyond the first minute.
Fix: Reduce charge temperature or reduce burner output sooner after charge.
RoR Targets by Roast Phase
There's no universal RoR number that works for all beans, all roasters, and all roast levels. But these ranges are a useful starting point:
| Phase | Typical RoR Range |
|---|---|
| Charge → Dry End | 10–18°C/min |
| Dry End → First Crack | 8–14°C/min |
| First Crack → Drop | 3–8°C/min |
The key is the direction, not just the number. A declining RoR that ends the development phase in the 3–6°C/min range is generally a well-controlled roast. A RoR that's flat or rising at any point in development warrants attention.
Lighter roasts typically target higher RoR at drop (more energy still moving through the bean). Darker roasts typically target lower RoR at drop (slower finish to avoid over-development of the outer layers).
Why Tracking RoR Across Roasts Matters
Looking at a single roast's RoR curve tells you whether that roast went well. Looking at the same bean's RoR across ten roasts tells you something much more useful: whether you're consistent, and where your variability lives.
If your RoR curves for the same bean overlay tightly, you have a repeatable process. If they diverge — some batches fast, some slow — you have an uncontrolled variable somewhere: batch size drift, seasonal ambient temperature, charge temperature inconsistency, or equipment wear.
First Crack stores the full RoR curve for every roast and displays it in the analytics view alongside drop temperature trends and DTR history. The curves are recalculated from imported Artisan files, so you can build a historical baseline even if you're switching from a different logging setup.
The practical workflow: after each session, import the .alog (or let the native app sync automatically), then check the RoR overlay for that bean against your previous three to five roasts. If the curves match, the roast was repeatable. If they don't, the deviation tells you where to look.
RoR and Development Time Ratio
RoR and development time ratio (DTR) are related but measure different things. DTR measures how long the beans spent in the development phase (first crack to drop) as a percentage of total roast time. RoR measures the trajectory — how the temperature was moving during that time.
A roast with a good DTR but a bad RoR — say, a stalled curve that happens to produce 20% development time — will still taste underdeveloped. The time was spent, but the energy flux wasn't there. Conversely, a fast RoR with a short DTR can produce the same underdevelopment from the opposite direction.
Good roast control requires both: a DTR in your target range and a declining RoR that ends in the 3–6°C/min window. When both metrics are on target, you have strong evidence the roast will cup the way you intended.
See also: Development Time Ratio (DTR) · What Is First Crack in Coffee Roasting? · FAQ