Contact with a rigid body

[kabur88]

Hi, I'm modelling a square steel beam connected to an end palte being twisted and pushed into a rigid surface.

I've modelled it in shell as my beam is hollow and will be too thin for proper analysis. I'm using a static linear perturbation step for the analysis and I'm not sure how to model the contact surface for the model. I've included a picture of what my model looks like after the job with only the punching force.

As you would see, the model is directly being pushed into the rigid body without and interaction.

I would like to know how to model this surface properly.

Thanks

 

[portliner]

Define contact between the proper side of the endplate and the rigig surface. Here the rigig surface MUST be the master surface in the contact.

 

[kabur88]

That is what I have been trying to do, however after I create the properties and wish to create the interaction I get this message from trying to create a contact. 'Pressure Penetration interaction can only be created in Planar and Axisymmetric models'.

Does this mean I'll have to create the entire model using planar and axisymmetric parts?

 

[bassmanjax]

It's because you seem to have a *PRESSURE PENETRATION defined in your model somewhere. As your original post doesn't mention any fluid pressure loading, then you've probably made a mistake in your contact definitions.

For the model shown, you should be able to do exactly as portliner says above, and make sure that you only define a mechanical contact property.

Regards

Martin

 

[kabur88]

Ok I tried changing the model, I am now using a Static General step.
However there still seems to be a major problem. In this model I am pushing the endplate into the rigid body. I am applying force all along the endplate closest to the rigid body and not the plate at the end of the beam.

I've changed my contact properties, with the rigid surface being the master and the endplate being the slave. It is frictionless,in the normal behaviour there is a penalty and in the damping the coefficient is 0.001 and the clearance is 1. Note that I have played with these values plenty of times with the same result.

This is my result however

As you can see there is no way that this is the correct result as it defies to laws of physics.

Still looking for help and thanks in advance

 

[bassmanjax]

Couple of things to try;

- Remove the contact damping and clearance. Not sure that you actually need these for your model...
- Make sure that NLGEOM is set to ON.

Regards

Martin

 

[corus]

The deformation scale factor is 1E-10 which says to me that there was no real solution obtained. I'm not sure why you apply a force to the plate adjacent to the rigid plate as the load must go directly into the rigid plate and not produce any meaningful results. Apply a fixed displacement in a first step to bring the parts together and then apply the force at the correct location. For shells make sure the normals to the surface point in the correct direction and/or you've chosen the contact surface on the correct surface to the shell.

Tara

http://tinyurl.com/4ydjg7m

 

[kabur88]

bassmanjax: Thanks, your info helped me. Really appreciate it

corus: I was only testing the model to see if it would bend inwards or not.