Implementación de un prototipo de máquina automática tostadora de café que permita el monitoreo y control a distancia utilizando IoT y una aplicación móvil

Abstract

The present graduation project aims at the design, construction, and implementation of a prototype automated drum-type coffee roasting machine, specifically developed to meet the needs of small and medium-sized producers. This proposal arises as a response to the recurring issue of lack of uniformity in artisanal roasting, where manual methods often generate inconsistencies that degrade critical parameters such as aroma, flavor, and the final quality of the bean. To address these deficiencies, automation and remote monitoring tools have been integrated, enabling more precise control over the system’s physical variables, transforming an imprecise process into a controlled operation that enhances the final product quality. The core of the developed system consists of a rotating drum with a heating element regulated through programming and the use of the hysteresis method to control the switching on and off of the resistance. This is complemented by a high-precision temperature sensor using a type K thermocouple. The rotational movement is provided by a coupled DC motor, whose speed is adjusted according to the load of beans introduced to ensure homogeneous heat transfer. The ESP32 acts as the central processing unit, managing the control logic and serving as a communication bridge under an IoT architecture. Through the implemented control logic, the system responds immediately to temperature variations within the drum, maintaining the setpoint configured by the operator. This automatic regulation eliminates human error and the uncertainty associated with traditional methods. Regarding the software infrastructure, a monitoring ecosystem based on time-series data was implemented. The system captures variables in real time and stores them in an InfluxDB database, while Grafana is used as the visualization engine with enhanced features for analytical supervision of controlled parameters. The project also includes a custom web interface that functions as a control center for the operator; from this platform, it is possible to select standard roasting profiles such as light, medium, and dark, adapted to the specific characteristics of the beans being processed. This helps even inexperienced operators achieve a high-quality finish with minimal intervention. The prototype stands out for its ability to record and store historical data, facilitating the traceability of each roasting session and enabling continuous optimization of operational parameters based on real data. System validation was carried out through experimental tests compared against roasting levels on the Agtron scale. The results obtained demonstrate that the prototype maintains the assigned setpoint temperature and achieves an objective level of quality that meets the standards required by the modern coffee industry.