Guide
Coffee Roast Curves: Temperature, Rate of Rise and Development
Read coffee roast curves, compare temperature with rate of rise, and calculate development time with clear examples and a measured research profile.
Quick Answer
A recorded coffee roast curve plots temperature against time. Rate of rise (RoR) shows how quickly that recorded temperature changes, while event markers locate charge, yellowing, first crack and drop. Read the temperature channel and axis units before comparing the shapes.
On This Page7 Sections
Read the Channels and Axes First
Find the legend, the time origin and both vertical axes. A temperature line and a RoR line can cross on the page without their values becoming equal: they measure different things and may use different scales.
| Signal | What to establish |
|---|---|
| Bean temperature, BT | Which sensor follows the bean mass, and how and where it measures |
| Environmental temperature, ET | Which air-temperature position is measured, such as above the beans or in the exhaust |
| Rate of rise, RoR | Which temperature channel it follows, its time unit and any filtering |
Artisan's two-probe example places BT within the bean mass and ET in the air above it. Another machine may use a different arrangement. Probe placement, diameter and batch size can affect the record, so keep those details with a comparison.
Before choosing the smoother-looking graph, I would check what was measured and where the timer began. A clear label is more useful than a familiar-looking shape.
A Falling RoR Can Still Mean Rising Temperature
Positive RoR means the recorded temperature is rising. If a positive rate declines, the temperature can keep rising more slowly. A negative rate indicates a falling temperature trend; zero indicates no change at the displayed rate's time scale.
For example, a hypothetical decline from 6 to 3 °C/min (10.8 to 5.4 °F/min) still describes warming. The rate has halved; the temperature has not fallen.
Check the time unit before comparing numbers. 6 °C per 30 seconds equals 12 °C/min (21.6 °F/min). For temperature differences and rates, multiply the Celsius value by 1.8 to obtain the Fahrenheit value, without adding 32.
RoR also depends on how the software processes the readings. Smoothing can make a noisy line easier to follow while making it less responsive to changes. Record the sampling and filtering settings separately from the displayed unit. A smooth line alone does not establish that two roasts followed the same process.
Read the Event Markers Before the Phase Percentages
Check what each marker records and when the clock starts. A timer may run before the beans are loaded, while a graph may display charge as zero.
| Marker | Meaning in the record |
|---|---|
| CHARGE | Beans enter the roaster; a common start for total roast time |
| DRY or yellow | A recorded color-change landmark |
| FCs | The marked start of first crack |
| DROP | Beans leave the roasting chamber for cooling |
Note whether an event was entered by the operator or detected automatically. For audible first and second crack, the roast-sounds article gives listening examples and explains the changing pattern.
Yellowing and crack markers divide the record into useful intervals. They do not make all the chemical reactions start or stop at those boundaries. Keep the event definitions with any phase names or percentages.
Calculate Development Time and DTR
Development time is the interval from first-crack start to drop. Development time ratio, or DTR, expresses that interval as a percentage of the total roast time from charge to drop.
- Development time = drop time minus first-crack start.
- Total roast time = drop time minus charge time.
- DTR = development time divided by total roast time × 100.
These two hypothetical records show why duration and percentage belong together:
Swipe to see all columnsScroll to see all columns
| Recorded event or result | Example A | Example B |
|---|---|---|
| Charge | 0:00 | 0:00 |
| First-crack start | 8:30 | 12:45 |
| Drop | 10:00 | 15:00 |
| Development time | 1 min 30 s | 2 min 15 s |
| Total roast time | 10 min | 15 min |
| DTR | 15% | 15% |
Both have a DTR of 15%, but Example B continues for 45 seconds longer after first-crack start. The percentage is a description of timing, not a flavor score or a universal target. Compare the actual duration, end result and cup alongside it.
What a Measured Research Curve Shows
In a 2024 study by Anokye-Bempah and colleagues, researchers at UC Davis used a natural-gas Probat P5 with 4 kg (8.8 lb) loads. Their 16-minute research roasts continued into burnt or charred stages to investigate acidity development. Each profile was repeated three times.
In the study's Figure 2A, after the initial temperature dip, the temperature rises while positive RoR generally declines. Panel F shows a temporary temperature dip when RoR becomes negative. The temperature and rate lines tell related, different parts of the story.
The published RoR axis uses °C per 30 seconds. The paper does not specify the physical probe or filtering settings, so read this as its recorded in-drum temperature rather than assuming an identical measurement on another machine.
The strip above each plot marks four study phases: green before color change; yellow from color change to first crack; tan from first to second crack; and brown after second crack.
Its phase labels separate first-to-second crack from the remaining roast. The DTR arithmetic above uses the full first-crack-to-drop interval.
Keep a Comparison Record You Can Use
Before overlaying two curves, collect the details that make their differences interpretable:
- Coffee and batch: coffee identity and green batch mass.
- Machine and measurement: roaster model, temperature channels, probe identities and positions.
- Clock and events: time origin, charge, yellowing, first-crack start and drop, including how markers were recorded.
- Display: temperature units, RoR time unit, sampling and filtering settings.
- Controls: recorded heat, airflow and drum-speed changes where available.
- Outcome: roast color and mass loss if measured, plus the brewing method and tasting observations.
Use the curve to ask a specific question, such as whether a temperature difference appeared before or after first crack. Then compare the record with the finished coffee. The roast-level guide covers taste and buying choices; the resting guide helps plan what happens between roasting and brewing.
Sources
- Artisan: Two-probe setup: an actual BT/ET placement example.
- Cropster: Temperature deviations: factors affecting probe readings.
- Cropster: About Rate of Rise: rate units and the responsiveness/smoothing tradeoff.
- Artisan: Phases: event landmarks and phase displays.
- Artisan: Displaying Development Time Ratio: first-crack-to-drop timing relative to total roast time.
- NIST Special Publication 1038, printed page 13: temperature-interval conversions.
- Anokye-Bempah et al. (2024), The effect of roast profiles on the dynamics of titratable acidity during coffee roasting, Figure 2 and methods. The article is published under CC BY 4.0.