On Flow Physics of Spinning Samaras

Abstract

A particular form of winged seed (samara) dispersal technique adopted by nature uses autorotative (unpowered rotation of the wing generating thrust force against gravity) descent; for eg. in Maple and Mahagony trees. This technique provides the lowest descent velocities among various seed dispersal techniques found in nature ensuring the safety of the delicate seeds. Bio-mimicked solutions to important engineering problems in aerospace as well as in disaster management - air dropping of life-saving packages during floods can be inspired from the samara. The samara is a complex structure having a bluff root containing the seed attached to a three dimensional wing. The dynamics of the samara from the instant of release is entirely unsteady, involving an initial transition phase where the samara tumbles until it achieves autorotation leading to a steady descent velocity. The distribution of mass and aerodynamic forces in this single structure ensures its stability during descent. Studies to comprehensively understand the physics of the samaras are limited. Recently, Leading Edge Vortex (LEV) has been found to be responsible for the high thrust forces achieved during autorotation. The dependence of LEV on the morphology of the seed needs to be understood to design optimal devices for engineering applications. The principal aim of this study is to understand the effect of morphology on the aerodynamics of the samara with a particular focus on the characteristics of the LEV. The flow field around the autorotating samara is experimentally obtained using Particle Image Velocimetry (PIV) in a specially designed vertical wind tunnel. However, since each natural samara is intricate, optimal, and unique, it has limited utility for parametric studies. Therefore, 3D printed models are developed that closely mimic the functions of the natural samara. A new design methodology has been developed to generate autorotating samara models. Drop tests of the natural samara and the 3D printed models show that the dynamics of the models and the samara are similar. Three 3D-printed samara models with different spanwise distributions of chord and mass are considered. For the first time, a complete characterization of the spanwise distribution of LEV has been carried out on the samara models. We show that in the neighborhood of the maximum chord location multiple LEVs are present, which leads to significantly higher local lift forces compared to other cross-sections near the root and the tip. The elaborate spanwise survey also shows that the locations near the wing tip behave similar to a bluff body, while sections near the root undergo a reversal of flow topology. A new non-dimensional parameter has been defined using Buckingham 𝜋 analysis that encompasses the dominant parameters involved in the study. This enabled us to understand the inter-relationship between observed flow physics, morphology and performance parameters.