Mechanical - jointed mechanism - remote loads - problem with convergence

  • 78 Views
  • Last Post 29 June 2018
Erica2018 posted this 21 June 2018

Hi

I am carrying out a transient structural analysis on a jointed mechanism. It is a reclining seat mechanism. Originally I applied the weight of the person as a distributed mass on the two brackets, but now I realise that a more realistic application of the weight is to use a remote force from the centre of mass.  I was able to achieve convergence with my distributed mass but cannot with the remote force.  Does anyone have any thoughts on this?

I have auto time stepping ON

Initial Timestep = 0.1 sec

Min Time step = 1e-2s

Max time step = 0.1 s

Solver Type is Program Controlled

Order By: Standard | Newest | Votes
peteroznewman posted this 21 June 2018

Hi Erica,

Try a remote mass instead of a remote force.

There are several diagnostics to use to understand the reason for a failure to converge. Are you familiar with the Newton Raphson Residual Force plots? You put a number for how many previous increments you want to see in a field under the Solution Information folder. A good number is 6.

Regards,
Peter

Erica2018 posted this 21 June 2018

Thank you for your prompt response Peter.

My plan is to move the remote force so that it mimics the movement of the centre of mass of the body as the chair reclines.  As I understand it using remote mass fixes it at the initial position through all time steps

I have selected the Newton-Raphson Residual Forces (I usually ask for 4), but I'm not really sure what to do with that information when I have it.... other than it shows me where the maximum is occurring. In what way do you normally plot them?

SandeepMedikonda posted this 21 June 2018

Hello Erica,

  Checkout this useful article on Newton Raphson Residuals. This might give you some additional insight.

Regards,

Sandeep

  • Liked by
  • peteroznewman
  • Erica2018
Erica2018 posted this 25 June 2018

That's really useful - thank you................. that gives me some options to try

Just one thing - - In all Newton-Raphson Residual Force contour plots the maximum is shown at the same joint.  Is there a similar way of reducing the stiffness of a joint to aid convergence?.......................... I only have one contact in this geometry, all the other connections are joints 

peteroznewman posted this 25 June 2018

Hi Erica,

Joints have two (or more) options for connection to the Mobile and Reference geometry: Deformable and Rigid. Whatever you had, you can try changing the joint with the large NR Force Residual to the other and see if that helps.

Regards,

Peter

Erica2018 posted this 29 June 2018

All my joints are currently rigid.  I tried making one joint deformable where the Newton-Raphson residuals showed the maximum value but the simulation crashed sooner.

I have now changed the Analysis settings to

Defined by - Substeps

Initial - 100

Minimum - 5

Maximum - 1000

It's progressing very slowly but has successfully passed the point at which it previously failed to converge....................  I live in hope 

 

Close