An Automated System for Constructing a Database of Leidenfrost Evaporation Curves Using Image Processing Techniques

To analyze the progression of Leidenfrost evaporation, traditional experiments were conducted manually to generate a complete evaporation curve. However, physical constraints render Leidenfrost evaporation experiments inherently time-consuming and susceptible to uncertainty. To address these challenges, this study aimed to develop an automated system using webcams for real-time image acquisition and processing, as well as a syringe pump constructed using an Arduino microcontroller, a stepper motor, and 3D-printed components. In the domain of real-time image processing, the radii of levitated droplets were determined using circular detection techniques. By fitting the droplet radii over hundreds of consecutive frames, it was concluded that the shrinking rate of levitated droplet radii remain constant when the radius exceeds 0.6 mm, and the evaporation time is accurately derived. A moving average algorithm was employed to identify the heat transfer area as well as the evaporation time between the boiling droplet and the hot surface, enabling simultaneous calculation of the heat flux. The automated system was then used to perform Leidenfrost experiments under varying experimental parameters, and was compared to manual methods to demonstrate its superior precision in both the film boiling and nucleate boiling regimes. For example, the automated system was utilized to perform a series of experiments as the Weber number increased from 7.01 to 23.18. The detected Leidenfrost temperature rose from 154 degrees C to 192 degrees C, while the evaporation time decreased from 85.2 s to 78.9 s. These findings were consistent with previous studies and aligned with physical expectations, reinforcing the reliability of the system and its results.