Discussion Forum

Hi,

Desciption of Issue

I am attempting to model a flexure hinge subject to bending (see attached picture and model) using the Solid Mechanics module with a stationary study and have run into convergence issues.

For the simulation, I am using an auxiliary sweep with a ramped displacement applied to the bottom right face of the flexure hinge (circled in red on the von_Mises_Flexure_Hinge photo). The issue I am having is the simulation fails to converge around a displacement of 4mm, which corresponds to the deformed shape in the von_Mises_Flexure_Hinge photo. Ideally I would like to be able to model the deformation until their is contact between the two faces on the compression side of the flexure hinge.

Questions

Does anyone have any suggestions/ideas as to why the simulation fails consistantly around a 4mm displacement? Does anyone have any suggestions/ideas on how to make the simulation proceed further until the two faces contact each other?

Current Simulation Settings

• Material: Silicone rubber modeled as neo-hookean (Material Parameters: incompressible, mu = 0.27e6 (Pa), Bulk Modulus 13.4e6 (Pa), and Density 965 kg/m^3.)
• Plane Stress: Thickness 3cm and Quasi-Static.
• Fixed at the top right and left boundaries.
• Prescribed vertical displacement at bottom right boundary.
• Including geometric nonlinearity, auxiliary sweep (0.02 step) to 0-1 cm displacement, predictor: constant, fully coupled.

Trouble Shooting Solutions Already Attempted

• Triangular vs Quadrilaterals Elements - Both fail to converge around the same place.
• Refinement of mesh at joint - Ineffective, I have tried between 5 and 100 element refinement along fillet.
• Solver Type - Segregated and Fully Coupled produce similar results.
• Contact Modeling - Creating a contact pair at the compression fillet does not help.
• Refining Mesh Away from joint - Does not help.
• Linear Elastic Material vs Hyperelastic - Changing the material to linearly elastic does not help convergence.
• Fillet type (Spline vs Bezier Curve) - Changing between does not help.
• Incompressible vs compressible material - For linear elastic material modeling incompressible vs compressible setting makes no difference.
• Decrease joint width - decreasing the joint width helps slightly, but still have convergence issues.

I greatly appreciate any help or thoughts people may have.

Best,

Will