On the Correct Preloading of Nonlinear Flexible Bodies to Achieve Accurate Multibody Simulation Results

This presentation was made at CAASE18, The Conference on Advancing Analysis & Simulation in Engineering. CAASE18 brought together the leading visionaries, developers, and practitioners of CAE-related technologies in an open forum, to share experiences, discuss relevant trends, discover common themes, and explore future issues.

Resource Abstract

Multibody simulation has evolved from rigid body dynamics to the analysis of assemblies with both rigid and flexible bodies. These flexible bodies can have large deflections and contact other bodies or themselves. These sophisticated models, if created properly, produce results with much higher fidelity than models that only consider rigid bodies.
Sometimes the flexible structure has a deformed (pre-stressed) initial configuration that occurs during the manufacturing/assembly of the product. The simulation model must replicate the initial deformed state of the flexible body in order to obtain accurate results. The challenge is that the starting mesh is in a relaxed state. A series of simulation steps are needed to bring the structure into the correct starting state of deformation. This presupposes that the simulation software is able to save the model, including the deformed structure, at the end of each step such that the model can start each new simulation with the model state from the conclusion of the prior step.
An example is the simulation of the flexible boot of a constant velocity (CV) joint assembly that is used on off-highway, all-wheel drive trucks that traverse challenging terrains. In this environment the CV joint can operate with a transmission angle of 40 degrees or more when the steering angle and suspension travel are near their limits. The performance of the CV joint boot is important because it is critical to keep contaminants out of the CV joint. Failures caused by extreme deformation or rubbing on the shaft need to be avoided.
The evaluation of rubbing on the shaft as well as the stress in the boot depends on the dynamics of the operating environment, including the effects of imbalance of the boot. While the boot is well balanced as manufactured, during usage it is possible for a quantity of grease to drop from the CV joint and lodge between the folds of the boot. Given the high rotational speeds of the boot, even the small imbalance caused by the grease can be significant.
For purposes of validation in this study there is a focus on two comparisons between the simulation results and physical testing. First, when the production boot is rotating with a transmission angle that exceeds the design specification, the outer edges of the top two folds collapse at the outer portion of the rotation, forming dimples. An excellent correlation was observed between the simulation animation results and the photograph from physical testing. Note that the dimpling effect occurs during constant transition. Depending on the transmission angle, there can be transitions between 1 and 2 dimples on a fold or an oscillating behavior between the two folds. The second comparison considers the shape of the boot and the contact between folds. There is an excellent correlation between simulation and test.
Given the excellent visual validation there can be high confidence in the other simulation outputs, including stresses and strains. Design insights can result from seeing the evolution of the stress in the boot as the transmission angle of the joint gradually increases. The details of the time and cost savings to the customer from using simulation for CV boot design will be presented.
The deployment of a roll-up solar array provides a second example of the importance of proper pre-stressing of flexible bodies in order to obtain correct results for the assembly. In the case of solar arrays, it is important to simulate the both the manufacturing assembly and stowing operations first before simulating the deployment.
In conclusion, today’s multibody dynamics simulation software provides capabilities to accurately model assemblies of rigid and flexible bodies in motion, including contact. Second, accurate behavior of flexible bodies requires the careful replication of initial, preloaded conditions. Third, substantial savings in cost and time result from the ability to simulate complex assemblies and structures such as can be found in a CV joint boot or a roll-up solar array.

Document Details

AuthorRoss. B
Date 7th June 2018


Back to Search Results