Restart + MESH

[Schandra]

Hello all,

I am using a simple example of tension and shear of a unit cube (attached input file 1 - Icube).
Is it possible to use 'SAVE', 'RESTART' / 'MESH' commands for re-applying new boundary conditions in new input file?

i currently enable 'SAVE' (attached input file 1 - I1) and 'RESTART' to impose shear in the second input file (attached input file 1 - I2).
When I try shearing the extended unit cube, the boundary conditions that are imposed in the earlier input file(tension) are disabled and the simulation starts from the initial undeformed configuration.
Can you please suggest me a possible solution strategy for these type of problems involving two different boundary conditions.

Thanks,
Chandra


Attachments : Icube - tension and shear of cube
                         I1 - I run the file and the restart file is R1001.01 (to enable tension)
                         I2 - I run this file after making a copy and renaming of the restart file from R1001.01 to R2001.01 (to enable shear)

[Prof. R.L. Taylor]

You describe a situation we have not considered.  Currently, this implies you need to have the new boundary conditions imposed before the restart.  However, you can also consider editing the restart files to output the boundary condition array and loads existing when you do the save.

[fethio]

I use the RESTART command in order to turn off a material parameter switch in a user material, but I don't use MESH in that case.
MESH is required if you would like to change properties within the same input file.

[Schandra]

I might have mistakenly meant regarding 'MESH'. What I intend to do is to apply new boundary conditions in the second stage. When I do this without using 'MESH', in the second input file, the boundary conditions are automatically reset but the new boundary conditions are not being applied on the deformed model (that I import using 'REST') but on the undeformed mesh. Is there any simple way of doing this.

[fethio]

Are you sure that FEAP is reading the restart file? You can double check this from the Ofile.

[Schandra]

Yes, after I import the file using Restart and I check it with plot/defo/mesh I was able to see that it had imported the file.

[FEAP_Admin]

What about the boundary conditions? do they show in plot?  do they show in the output file?

[Schandra]

I am able to see them both in output file and plot (I use defo plot in second input file)

[FEAP_Admin]

If you are seeing all the correct boundary conditions, then I do not understand your problem.

[Prof. R.L. Taylor]

Can you describe step by step what you expect to happen in each phase of the  analysis.  Boundary conditions are applied to nodes. So I am unclear on what you want to happen between each phase of the analysis.

[Schandra]

Dear FEAP_Admin,

The issue that I am facing is that in the second step after I import and solve the problem, the displacement on nodes is being applied on the initial undeformed configuration instead of the current imported configuration.

Warm Regards,
Chandra

[Schandra]

Dear Prof. Taylor,

I have attached figures describing what happens at each phase of the analysis for your reference.

Warm Regards,
Chandra

[FEAP_Admin]

Your boundary conditions in I2 make no sense.  Also why do you have a 'mesh' command in the middle of the mesh input?

If you want to continue to have the vertical displacement being imposed in I2, then you need to have that BC in I2.

[Schandra]

I want to apply tension (in I1) and then impose shear (in I2) so that I get effect of tension+shear.

[Prof. R.L. Taylor]

You are missing the reactions from the tension load.  When you change boundary conditions from fixed to free the forces on the nodes are loast.  I attach a file which uses the commands you need -- it is not doing exactly the right shear problem as I did not put the shear forces on the left and right sides, only the top.  Be careful to clean your R files between runs if necessary.