Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations GregLocock on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Boundary from the results of previous step?

Status
Not open for further replies.

Keiker74

Structural
Jun 8, 2005
8
Hi,

I want to enforce a *BOUNDARY condition (displacement) that is dependent on the results of a previous step.
For instance, if I plastically deform a material and then unload the material, plastic deformations exist. I then want to displace a certain node 1 inch based on the results after unloading.

I have tried to used dummy nodes with the *Equation option, but for this to work, I would have to make the equation active after the previous step. Or does anyone know of a way that I can specify a boundary condition that is a movement from the previous step result and not using the FIXED condition. Any help would be greatly appreciated. Thank you.
 
Replies continue below

Recommended for you

I may be completely missing the point here but isn't it this simple?

*STEP, NLGEOM
[PLASTIC DEFORMATION LOAD STEP]
*ENDSTEP
*STEP
[PLASTIC DEFORMATION UNLOAD STEP]
*ENDSTEP
*STEP
*STATIC
*BOUNDARY
node,dof,,<distance>
*ENDSTEP
 
Displacements are carried on to the next step. gwolf is correct though you might need a OP=NEW option here and there.

corus
 
Thank you for the replies gwolf and corus. You are both correct. The problem is that I don't know what the <distance> is. I know what its magnitude has to be in "comparison" the previous step result.

For instance, I want Node 10 to move 1 inch from where it was after Step 2. The problem is I do not know where Node 10 will be after Step 2. I want to run all steps but right now I have to stop the analysis before programming Step 3.

I currently would run Step 2 with a *RESTART, WRITE option, and then perform a *RESTART, READ and displace the node 1 inch from the result of Step 2. It works but being able to program this for the same analysis over night would be beneficial. RIght now, to run my files overnight, I need to get up about every 3.5 hours.
 
Idea 1: Use a contact definition in your model. Contact area = target position for node

Idea 2:Activate a nonlinear spring in the last step by using *MODELCHANGE,ADD. This spring should act between the node and the target position. Choose nonlinear spring stiffness so that there a huge nodal force if node isn't at target position.

Pam
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor