Kinematic coupling and BCs of a screw in block model

[dw1982]

Hi,

I am working on a model of a 3D screw in a cubic block and I want to make sure it is done properly. I wish to apply a force in the U1 direction so I created a reference point on the screw heard surface and used Kinematic Coupling to tie it to the surface. I only checked U1 because this is the direction of force application, is this correct or should I uncheck all dofs?

Now the block should be immobile so I used symmtery BC's on all surfaces except the surface through which the screw is placed. I also applied zero displacement BC's to all dof's of the reference point except in the U1 direction. Is this the right way to do it?

Thanks a lot

 

[rstupplebeen]

A picture would really help get your point across.

Rob Stupplebeen

 

[dw1982]

The first picture shows the assembly. I set the BCs as follows:
X-SYMM on surfaces orthogonal to x-axis (2 surfaces)
Z-SYMM on surfaces orthogonal to y-axis (2 surfaces)
Y-SYMM on bottom surface
Zero displacement in all dofs except U1 for reference point on screw head

As for the kinematic coupling, I attached the point as shown in the picture to the screw head surface. I only checked U1 under kinematic coupling since I am applying the force in this direction.

http://files.engineering.com/getfile.aspx?folder=a7eea956-c141-444f-adc1-73e5e18c0884&file=rp

[1].JPG

Is this the right way to do it?

Thanks a lot!

 

[rstupplebeen]

If you fix the bolt then the load will use that constraint as a sink and not transfer load into the model.

Your symmetry planes effectively create 4 bolts on the top and 4 on the bottom. Is that what you want?

If you want to use symmetry on a single bolt then you need to only model part of the bolt. Imagine a coordinate system oriented like yours with the origin at RP-1. Only model +Z space and then have the section cut have symmetry.

I hope this helps.

Rob Stupplebeen

 

[dw1982]

Please bear with me... I'm a newbie

What I'm trying to do is to apply a load to the screw which is placed in a block that does not move and then see how much stress is generated in the block. Technically the screw should be able to move in any direction while the block is fixed.
Initially when I created the model I just used zero displacement BC on all sides of the block but someone told me I should use symmetry conditions instead which I am not familiar with.

What I need to know is what dofs should i check in kinematic coupling, U1 or nothing? What are the proper BCs for the kind of model?

Thanks a lot

 

[rstupplebeen]

To use symmetry you have to cut your model to be less than the full thing. It is a method to reduce the computational size of the model.

How you apply the constraints really is the art of FEA. It is very dependent on your specific problem and what you seek to predict. In general I make sure that my constraints are far enough (either art or sensitivity study) away from the regions of interest to not effect the results. Looking at your model and not the real world case I would probably fix the bottom of the part and split the model as I mentioned previously.

I hope this helps.

Rob Stupplebeen

 

[dw1982]

What about kinematic coupling? should I restrain any dofs if I want the force transmitted directly to the scew?

Thanks for your help

 

[rstupplebeen]

If you want to distribute the load across a face of the screw then yes.

If you are concerned with the stresses near the bolt you probably want to define what is loading the bolt. Remember to keep your constraints away from regions of interest.

I hope this helps.

Rob Stupplebeen