Contact Modeling

[JohnKGH]

Hi all,

I'm trying to model bearing effects of 2 bodies composed of elastic materials defined as *Surface. I am placing a displacement control on to one of the bodies and am connecting the two using beam elements. Contact pairs is used with small sliding, surface to surface. When I run the input file however, I get errors of:



MAX. PENETRATION ERROR -3.3011E-005 AT NODE 1515 OF CONTACT PAIR
(SURF_1,SURF_2)
MAX. CONTACT FORCE ERROR -0.00305203 AT NODE 1515 OF CONTACT PAIR
(SURF_1,SURF_2)
THE CONTACT CONSTRAINTS HAVE CONVERGED.

AVERAGE FORCE 3.357E-02 TIME AVG. FORCE 5.730E-03
LARGEST RESIDUAL FORCE 0.612 AT NODE 1501 DOF 3
LARGEST INCREMENT OF DISP. 7.186E-02 AT NODE 3032 DOF 2
LARGEST CORRECTION TO DISP. 7.096E-02 AT NODE 3032 DOF 2
FORCE EQUILIBRIUM NOT ACHIEVED WITHIN TOLERANCE.

AVERAGE MOMENT 6.835E-03 TIME AVG. MOMENT 3.114E-03
LARGEST RESIDUAL MOMENT -0.479 AT NODE 212 DOF 4
LARGEST INCREMENT OF ROTATION 0.108 AT NODE 3013 DOF 4
LARGEST CORRECTION TO ROTATION 0.106 AT NODE 13 DOF 4
MOMENT EQUILIBRIUM NOT ACHIEVED WITHIN TOLERANCE.


***NOTE: THE SOLUTION APPEARS TO BE DIVERGING. CONVERGENCE IS JUDGED UNLIKELY.


***ERROR: TIME INCREMENT REQUIRED IS LESS THAN THE MINIMUM SPECIFIED

Any ideas or suggestions?

I've attached my input file.

Thanks alot.

 

[corus]

In the quick look I made, I see no contact interaction properties defined. You should define normal and tangential properties. In addition I'd narrow down your surface definitions to only those surfaces that will make contact and not the surface of the whole body. It'll run a lot quicker.

corus

 

[JohnKGH]

Thanks alot for replying. Do I define contact properties via *Contact Controls? I want one body to slide frictionlessly against the other once contact has occured with no penetration but I'm unsure how to go about this? My approach so far has been to set absolute penetration to zero with a delay in friction but I don't think thats right.

 

[corus]

Contact controls are used to set the parameters to determine if contact has been made, and nothing to do with properties. It's useful, however, to include '"contact controls, automatic tolerances' in your data to improve convergence.
Check in the interaction module of CAE for Contact Interaction Property as that defines friction between the two, for example.
In addition, I don't understand why you have so many section properties for the model as each appears to be the same definition. It won't stop the job running of course but with only two materials you can easily make a mistake in the assignment of the property. You also appear to redfine the contact for Step 1 but your surfaces change to include surfaces that could never contact?
I'd go through the model and tidy it up and check to make sure you are defining everything properly.


corus

 

[rstupplebeen]

I have seen some front ends (can't remember which ones though) create a new property for each beam orientation so John may not have control over it. Hope this helps.

Rob Stupplebeen

 

[JohnKGH]

Hey guys,

Thanks alot for the help. I was able to resolve the problem via using nonlinear beam section geometry!..=).

 

[JohnKGH]

To clarify, I found out axial stiffness was incorrect along with some parameters in my beam definition.

BTW, when defining nonlinear beam, it seems shear is calculated automatically for the section as opposed to user inputs for bending and axial stiffness. Is this true and if so, how is shear calculated in this case and can I change it? Thanks