What does a coffee roaster's temperature probe measure? Understanding BT, ET and RoR
On the roasting chart, the BT line is often called bean temperature, while ET is environmental temperature. However, the common BT probe does not plug into the core of each coffee bean. It measures the temperature at the sensing location within the moving bean mass, where the probe is affected simultaneously by the bean and the hot air flow.
This is a crucial difference when reading the roasting chart, comparing two machines or trying to recreate a profile. The numbers on the screen are most valuable when understood in the context of the device, sensing location, batch weight, and specific operating conditions.
BT is a measurement at the probe within the bean mass, not a direct measurement of the core temperature of each bean. ET also depends on the sensing location and sensor design. Therefore, the same displayed number on two different machines does not necessarily represent the same thermal state of the coffee.
1. How do BT, ET and bean core temperature differ?
| Specifications | The sensor reflects what? | Limitations when interpreting |
|---|---|---|
| BT – Bean Temperature | Temperature at the probe located within the bean mass; the signal is affected by the bean in contact and the surrounding air. | Not a direct measurement of the core temperature of each bean. |
| ET – Environmental Temperature | Environmental temperature at the sensing location, often used to monitor the temperature conditions within the roaster. | Not representative of the entire air temperature at all locations within the roasting drum. |
| Bean Core Temperature | Temperature at the inner region of a single bean, requiring specialized methods to determine directly. | Difficult to measure in production as the sensor must be placed within the bean and may alter the measurement conditions. |
2. Why does BT change when adjusting the flame or air flow?
During the roasting process, energy is transmitted from the drum, hot air, and surfaces in contact with the bean. Simultaneously, the bean loses moisture, releases gases, and changes its structure. The probe only records the thermal response at its location; the signal may be delayed relative to the heat source changes due to the thermal inertia of the sensor, its installation location, and contact with the bean mass.
Therefore, when reducing the burner power, the BT line does not necessarily change direction immediately. Similarly, increasing the air flow may affect heat exchange and gas release, but the BT display still depends on the machine's design and sensor placement.
BT at a given time is only part of the data. To understand the process, you need to view it in conjunction with RoR, the time of bean loading, changes in power, air flow, and observable events such as first crack.
3. What is RoR, and why should it not be viewed as an independent measurement?
RoR (Rate of Rise) is the rate of temperature change recorded over time. When calculated from the Bean Temperature (BT) curve, RoR indicates how rapidly BT is increasing or decreasing within the software’s calculation window.
ΔT is the displayed temperature change; Δt is the corresponding time interval.
For example, if BT rises by 6°C in 30 seconds, the average RoR over that period is 12°C/min. This is an illustrative calculation, not a target RoR for every roast batch.
RoR is typically smoothed by software to reduce noise. Different smoothing windows can produce different RoR curves from the same BT data. Therefore, it is essential to understand how the software performs calculations before comparing RoR curves between two systems.
4. Can drop temperatures from one machine be replicated on another?
It is not advisable to replicate roasting parameters based solely on end-of-roast temperature figures. Two machines may differ in drum diameter, batch weight, heating capacity, airflow, probe placement, and sensor latency. The same displayed BT may correspond to different thermal states and levels of bean development.
When transferring a profile to a different machine, use it as a reference for adjustment sequences and timing, then refine through test roasts. Monitoring bean color, weight loss, crack progression, development time, and cupping results is more useful than forcing the new machine to match a specific temperature figure.
5. Data logging procedure for reproducible profiles
- Record the machine model, batch weight, bean type, moisture content, and processing information.
- Note the sensor position and the software used for data storage.
- Maintain consistency in logging the timing of charge, power adjustments, and airflow changes.
- Mark observed events; do not infer all chemical changes solely from BT.
- Compare the profile with roast color, weight loss, and post-roast sensory evaluation.
With SAVINA, this approach helps maintain traceable roast records and reduces reliance on interpreting profiles based on a single temperature curve. An attractive chart does not automatically prove a good roast; final quality must still be verified through the product and sensory assessment.
References
- Kelly, C. B. D. & Scott, J. B. (2014). Online Thermal Analysis of Batch Roasted Coffee Beans. University of Waikato Research Commons. Read document.
- Physical Model of Heat and Mass Transfer in a Spouted Bed Coffee Roaster. Drying Technology (2007). https://doi.org/10.1080/07373930701438683.
- Smart Online Coffee Roasting Process Control: Modelling Coffee Roast Degree and Brew Antioxidant Capacity... Foods (2020). Read study.