A simple experiment for observing clustering and dynamics of coalescing particles in air turbulence

A novel experimental platform is developed with a focus on simplicity to investigate the dynamics of inertial particles (micro-droplets) in air turbulence. The ultimate goal of the effort is the observation of particle collision and coalescence in turbulent flows, while the immediate focus is on the influence of collision-coalescence process on other dynamical metrics such as the radial distribution function (RDF) and relative velocity statistics. The main observation tool is a three-dimensional Lagrangian particle tracking (LPT) system, designed to achieve high-resolution measurements at high particle number density and at deep sub-Kolmogorov scales in turbulent flow. The system leverages the simplicity of LED illumination combined with high-speed spinning-disk atomizers, enabling reliable tracking of particles with diameters of approximately \((10 \mu\mathrm{m})\) and larger under controlled turbulent conditions. A minimum resolvable particle separation of \((r/\eta \approx 0.1)\) is achieved, thereby extending the accessible measurement regime for small inertial particles. A central contribution of this work lies in the systematic identification and mitigation of three dominant sources of spurious particles: false stereo-matching induced spurious particle (FMIS), interpolation induced spurious particles (IIS), and threshold induced fragmentation (TIF). An angle-based geometric filtering criterion is introduced, through which the artifact due to FMIS on RDF are strongly suppressed. Together, these procedures establish a validated workflow for obtaining reliable small-scale statistics. Using this framework, the radial distribution function (RDF) and a normalized pseudo-collision rate are measured at near-contact separations for particles with Stokes numbers \((St \approx 0.2)\)--\((1.0)\). A clear enhancement of sub-Kolmogorov clustering with increasing Stokes number is observed, and consistent near-contact statistics are ensured through the proposed filtering strategy. The present study therefore extends the operational limits of LPT and provides a reliable experimental methodology for investigating inertial-particle dynamics at previously inaccessible spatial scales.