Simulation of Drop Collection with Non-Uniform Cloud Distributions for Collection Efficiency Sensor Validation

Large icing wind tunnels typically have sufficient distance for drops from spray nozzles to spread evenly producing small spatial variations of cloud properties at the wind tunnel test section. As the size of a wind tunnel gets smaller, producing clouds with uniform properties becomes challenging because of 1) the reduced distance from the spray bar system to the test section and 2) the spray characteristics of most air-assisted nozzles used for spray generation. For this paper, discrete-phase simulations using FLUENT were used to explore droplet collection on a partial NACA 0012 model at different angles of attack in the Baylor Liquid Film and Cloud Tunnel (LFACT). McClain et al. (2022) used the LFACT to validate a new microwave sensor system to measure collection efficiency variations along the surface of a wind tunnel model. However, the sensors used in the investigation were essentially the same size as the measured non-uniform cloud features in the wind tunnel test section. A convolution approach was used to map the measured Liquid Water Content (LWC) variations onto the sensor areas, but this approach is based on critical assumptions about droplet trajectories. The FLUENT simulations were performed to further validate the collection efficiency sensors and to investigate water capture by wind tunnel models in the non-uniform clouds of the LFACT. The simulations were performed using the Spalart-Allmaras turbulence model, and the droplet injection pattern was generated using Matlab and a two-dimensional Gaussian random number generator. The injection pattern was then simulated using a non-interacting, spherical drop discrete-phase modeling system in FLUENT. The simulation results for two cloud conditions and two angles of attack demonstrate the validity of the convolution approach and further validate the measurements of the microwave sensor system.