Non-Coiled Spring Optimization Assisted by an Analytical Model

x. zhou[1]
[1]Triple Ring Technologies, USA
Published in 2019

The non-coiled spring, the most ancient and crucial mechanical component, has been widely used in different industry at different scales. During the spring design process, performance, robustness, tolerances, and cost always need to be considered and certain trade-off have to be made. As a result, spring design will always benefit from the advancement of theories and numerical tools that can answer the question of what makes a spring design optimal.

Non-coiled spring problem has large number of multi-variables, non-linear equations, and sophisticated geometry . Traditional optimization techniques either over-simplify the spring shape such that they can apply beam theory or directly apply a numerical method which can be time-consuming and trapped in some local minima.

The purpose of this work is to look into the fundamental issues regarding spring design and develop a new approach which takes both the merits of analytical solution and numerical solution. With our treatment, engineers can start with a curved non-coiled spring in COMSOL Multiphysics® simulation software and FEM environments to identify the spring’s effective spring rate and maximum stress at any thickness and width. With both those identified values and target values, we can plug them into our derived analytical equations and get the predicted dimensions. The performance of new dimensions will be verified in FEM/COMSOL Multiphysics® environments. Only two iterations of simulation are needed to get the ultimate spring performance.