Verification of the Part-Composite Approach for Modeling the Multi-Layered Structure of a Rolling Truck Tire

Accurate modeling of a truck tire for predicting dynamic characteristics requires an adequate representa- tion of its composite plies. This study compares two approaches in modeling the multi-layered structure of a rolling radial-ply truck tire using LS-DYNA. In the first approach, different layers in the tire structure are modeled using individual elements; whereas, in the second, all layers are represented by a single element with layered configuration managed by the PART_COMPOSITE keyword. Hence, in this article, these tire models are named as the Individual-Element (IE) model versus the Part-Composite (PC) model. In the Individual-Element (IE) model, the carcass and belt plies are formulated using layers of isotropic solids representing rubber matrix, together with separate layers of orthotropic shells describing fiber-reinforced composite plies. In the Part-Composite (PC) model, however, the carcass and belt plies are simplified to a single layer made of shell elements with layered configuration using the PART_COMPOSITE keyword. This keyword allows to define stacks of plies with pertinent material properties, thicknesses and fiber orientations to manage the integration rule through the thickness of a composite part. The Part-Composite (PC) model was then verified to be in reasonable agreement with the Individual-Element (IE) model in predicting the load-deflection, cornering force/moment and modal characteristics of a truck tire. The validities of the tire models are also evaluated by comparing their simulation results with the reported experimental data. The Individual-Element (IE) model has the potential to be applied as a reliable virtual tool to study the influences of various operating and design parameters upon the tire dynamic characteristics. However, the large num- ber of elements in such a detailed model makes it computationally expensive. With the Part-Composite (PC) model, the computational efficiency was remarkably improved since replacing several individual elements with one layered element reduced the total number of elements and thus the computational demand.