Please help: How to generate a node in CAE?

[jihuili]

Dear all:

I am a new user of Abaqus, have a simple question.

I need to add a concentrated force to a point inside of a part. There is no node there, although I know the coordinates. I can add a new node easily in inp file, but import inp file will generate orphan mesh, which can not be remeshed or removed, and there is no seed on the new node. So I want to know if there is a way to generate a node in CAE, for example type in the coordinates. Or are there any other choices to add the force onto a point (no node there)?

Thanks a lot,

James

 

[corus]

Partition the part so that a geometry point is at that position.

Tata

 

[jihuili]

Hi, Corus:

your suggestion worked for me. Thank you very much.

One more question: I added a moment load onto a node, but an error message came out: 1 node have inactive dof on which boundary conditions are specified". The node is the one I added moment. My model has tet mesh (only choice since the model was stl. format). And I have not defined a boundary condition. The only constraint is a contact with another part, which is fixed.

Thanks,

James

 

[corus]

You can't put a moment on to a node of a 3D solid element.
Another reason for getting the inactive freedom message is when you apply a boundary condition to a node that is also in contact, as that freedom will have been removed.

Tata

 

[jihuili]

Thanks, Corus.

I added two opposite concentrated forces at an unimportant area to simulate the moment, looks good.

One more question: I want to simulate a contact on two shell surfaces (composed of many triangles). The shells should have thickness (to simulate 2mm cartilage). Right now I use homogeneous shell and define the thickness on top of the surfaces. So the two surfaces have a 4 mm gap. However, seemed the thickness does not count when calculating the contact. I still have to move the two surfaces attach to each other to let the contact work. My question is do you know any other shell type can work for my case? In other words, the shell elements should have real thickness.

I know a suggestion could be extruding the surface model to a solid model with 2 mm thickness. However, since my geometric model (came from stl format) has a lot of triangles, which normals point to different directions. Simply extrude using UG or Abaqus will not work.

Thanks,

James

 

[rstupplebeen]

Rob Stupplebeen

 

[jihuili]

Hi, Rob:

Thanks a lot for the reply.

I am using a simplied model (a ball and a plate) to represent the real model. Although I used your method to define the contact, still got the same problem (too many attempts).

Please note the Z-direction distance between the ball and the plate is 0.5 mm. I created a shell layer (0.5 mm thick)on both ball and plate to represent the cartilage. I set up the shell offset on the middle surface, so theoretically there is no gap between the two parts. The cartilage layers were tied onto the ball and plate, respectively. A contact was set up between the two cartilage layers. I did not add real material property to the cartilage layer.

Anyway, Please take a look of the attached cae file.

Thanks again.

James

 

[rstupplebeen]

Below is what I have found so far:
1. Turn on Nlgom in Edit Step
2. Suppressed both surfaces (apply contact to the exterior surfaces of the solids)
3. Suppressed both constraints
4. Removed initial gap
5. Quad mesh 1mm resolution C3D8

Use the symmetry plane to reduce model size.

I hope this helps.


Rob Stupplebeen

 

[rstupplebeen]

I forgot to mention I removed the cartilage to simplify the model for error checking.

Rob Stupplebeen