Lack of Convergence - Using Remote Force - Ansys Workbench/Mechanical

[EricaD]

I am working on a transient analysis of jointed structure (see file attached) - it is a reclining chair mechanism. I have previously obtained solutions when I apply distributed masses on the seat bracket and footrest bracket. Now I think this is not a realistic method to apply the load as the centre of mass of the body will be above the mechanism and moves as the chair mechanism changes position. I am trying to apply a remote force at the centre of mass in a number of finite steps as the mechanism expands, but I can't get the convergence even on the first timestep. Any thoughts?


I have Autostepping - ON
initial timestep 0.1 secs
minimum timestep 1e-2 secs
max time-step 0.1 sec

Solver Type is Program Controlled

 

[NRP99]

Did you check whether your joint and remote force are conflicting each other or model has Overlapping Contact Conditions and Boundary Conditions Link.

Check this thread569-215994. What are warnings and error for non convergence? Apply your load more slowly by increasing load sub-steps.

 

[scheah]

NRP99 suggestion is very good. May I ask what is the error message when your convergence failed?


Kind regards,
Jason

 

[EricaD]

Thank you.....that gives me some options to move forward with - - I'll let you know how I get on.

Whether I use a distributed mass or a remote force I get a conflicting contact region or remote boundary conditions warning (see file) and yet my joints appear to show no conflict (see file) and in the distributed mass version I don't have a remote boundary condition

 

[scheah]

Hi Erica,

From what I gather, the solver attempts to negotiate and resolve the overconstraint conditions while it issues that warning message. Overconstrained models most often result in nonconvergence of the solution with small solver pivot warnings. Is that the error message you see in the output file?

Also, I would suggest trying smaller steps:
Initial Substep of 100
Minimum Time Steps of 5
Maximum Substeps of 1000



Kind regards,
Jason