Why does airflow affect heat transfer and bean uniformity?
Two batches with the same probe temperature, roast time, and development level after First Crack can still yield different results. One of the reasons lies in a factor often overlooked when reading roast curves: the airflow inside the roaster.
Airflow is not simply a fan used to exhaust smoke or cool the beans. In a roaster, the air stream participates in heat transfer, transports moisture, and carries chaff and volatile compounds out of the drum. Changing airflow means changing the physical conditions the coffee beans experience throughout the heating process.
To understand why a roast profile changes when adjusting the exhaust fan, one must view the roaster as a heat and mass transfer system, rather than focusing solely on the Bean Temperature curve.
How does heat enter the coffee beans?
In a roaster, heat can be transferred to the beans through three basic mechanisms: conduction, convection, and radiation. The proportion of each mechanism depends on the machine design, heat source, drum construction, rotation speed, and operating conditions.
Conduction occurs when beans come into contact with the hot drum surface or with other beans. This mechanism is particularly significant in drum roasters, where beans are continuously tumbled and come into contact with the drum wall.
Convection is the heat exchange process between the hot air stream and the bean surface. As the air stream moves, it carries thermal energy to the contact area and simultaneously transports water vapor and volatiles out of the roasting chamber.
Thermal radiation originates from hot surfaces and the heat source within the machine. The contribution of radiation depends on the geometry of the roasting chamber, surface temperature, and design characteristics.
There is no fixed ratio among these three mechanisms for all roasters. A drum roaster, a hot-air roaster, and a fluidized bed roaster have different flow fields and heat transfer mechanisms.
How does airflow affect bean temperature?
In convective heat transfer, a commonly used simple expression is:
Q̇ = h × A × (Tkhí − Tbề mặt hạt)
Where h is the convective heat transfer coefficient, A is the heat exchange area, and the temperature difference is the driving force for heat transfer.
Airflow can affect the convective heat transfer coefficient, but not in a simple linear proportion. When increasing air volume, the velocity and flow state around the beans may change, thereby altering heat exchange efficiency.
However, the amount of hot air passing through the roasting chamber also carries an amount of energy dependent on its temperature and mass flow rate. If airflow is changed without adjusting or recording the heating power, the overall energy condition of the system also changes.
This is why increasing the exhaust fan speed and increasing burner power cannot be considered equivalent methods for increasing heat to the beans.
What else does airflow transport away from the beans?
During roasting, coffee beans lose moisture and release many volatile compounds. At the same time, chaff separates from the beans and is carried by the air stream to the collection system or cyclone.
Therefore, airflow performs two simultaneous tasks:
- Delivering thermal energy into the roasting zone via the hot air stream.
- Transporting water vapor, chaff, smoke, and volatile compounds out of the roasting chamber.
If the exhaust system is not suitable for the machine structure and roast load, smoke or volatile products may remain longer in the roasting zone. This changes the environment the beans are exposed to, but it does not mean that simply increasing exhaust will automatically produce a cleaner cup of coffee.
Increasing airflow does not inherently make the beans hotter or cooler. The actual impact depends on inlet air temperature, mass flow rate, heating capacity, heat loss, roasting chamber design, and the air recirculation mode of each machine.
What has simulation research observed?
The study by Chiang, Wu, and Kang, published in Journal of Food Process Engineering, used finite element simulation to investigate fluid dynamics and heat transfer in a practical coffee roaster.
The authors identified heat source power and gas velocity at the exhaust outlet as key parameters affecting temperature and velocity distribution inside the roasting chamber. The model also recorded non-uniform internal bean temperatures during the first 60 seconds. Subsequently, over 100–720 seconds, internal bean temperatures gradually became more uniform under the simulated conditions investigated.
This is a significant finding for roaster designers: temperature measured at a single location cannot fully represent the temperature field inside every coffee bean.
Another research direction by Alonso-Torres and colleagues uses CFD to simulate heat and mass transfer within individual coffee beans. The authors compared computational results with single-bean roasting experiments in a glass tube using a hot air stream. The model showed reasonably good predictive agreement with experimental data for this design.
By 2025, Al-Shemmeri and colleagues developed a batch roasting model for spouted bed roasters, calibrated with experimental data from pilot-scale roasting trials. Input variables included batch mass, airflow, and inlet air temperature.
These studies collectively indicate that airflow is not a secondary parameter. It is an integral part of the energy balance and mass transfer equation in the roaster.
Why can't a specific airflow setting be replicated between two machines?
A fan setting of 50% on one machine does not mean the same air volume flow as 50% on another. Valve opening, pipe diameter, cyclone resistance, roasting chamber design, and fan characteristics all affect the actual flow rate.
| Parameter | Significance in roasting | Notes |
|---|---|---|
| Airflow | Gas flow rate through the system | Distinguish between volumetric flow rate and mass flow rate |
| Air velocity | Gas stream velocity at a specific location | Depends on cross-sectional area and duct geometry |
| Exhaust temperature | Exhaust gas temperature | Not equivalent to bean temperature |
| Pressure / draft | System pressure differential and suction force | Depends on fan, valves, and duct resistance |
| Burner power | Energy input to the system | Must be monitored when changing airflow |
For roasters without a flow meter, the operator can start by calibrating the relationship between the valve position, vacuum pressure, and exhaust gas temperature. When possible, it is recommended to directly measure the flow or velocity of the gas at the predetermined location.
Design a valuable airflow experiment
Instead of changing the fan, burner, and drum speed simultaneously, it is recommended to perform a controlled experiment.
Use the same batch of green coffee beans, the same batch size, initial moisture content, and charging temperature. Keep other parameters constant, then change the airflow to the pre-calibrated levels on the actual roaster being used.
Record at least:
- Bean Temperature and Environmental Temperature over time.
- RoR at critical stages.
- Inlet and outlet gas temperatures if there are sensors.
- Valve position, vacuum pressure, or gas flow measured.
- Roasting time, time after First Crack, and weight loss ratio.
- Bean color after roasting, uniformity level, and cupping results.
The key is to compare the results after controlling the input conditions. If a sample has a shorter roasting time, different color, and different TDS, it cannot be attributed entirely to airflow if other parameters have also changed.
Airflow is a control variable, not a flavor adjustment knob
Roasters often say to increase airflow to make the coffee cleaner or decrease airflow to retain heat. This may be useful experience on a specific device, but it should not be turned into a general rule.
Airflow should be considered in relation to heat source, batch size, inlet gas temperature, system resistance, and bean characteristics. Only when understanding these relationships can the roaster adjust airflow intentionally and consistently.
For roaster designers, airflow is also a critical technical problem in heat distribution, reducing dead zones, controlling smoke, and building stable roasting capabilities between batches.
To delve deeper into the chemical changes that occur after the bean absorbs heat, you can read the article Acid Chlorogenic in roasted coffee on the SAVINA library.
References
- Chiang, C.-C., Wu, D.-Y., & Kang, D.-Y. (2017). Detailed Simulation of Fluid Dynamics and Heat Transfer in Coffee Bean Roaster. Journal of Food Process Engineering, 40(2), e12398. https://doi.org/10.1111/jfpe.12398
- Alonso-Torres, B., Hernández-Pérez, J. A., Sierra-Espinoza, F., Schenker, S., & Yeretzian, C. (2013). Modeling and validation of heat and mass transfer in individual coffee beans during the coffee roasting process using computational fluid dynamics (CFD). Chimia, 67(4), 291–294. https://doi.org/10.2533/chimia.2013.291
- Al-Shemmeri, M., Fryer, P. J., Farr, R., & Lopez-Quiroga, E. (2025). Batch-scale simulation of heat and mass transfer of coffee roasting in spouted bed roasters. Beverages, 11(6), 162. https://doi.org/10.3390/beverages11060162