Stress analysis- odd results

[Rytcd]

Hi there, So I am trying to get stress analysis to work for me, but I keep getting very odd results. for instance, the attached part should hold something like 1000lbs, yet is reading much, much lower. Can someone please tell me what I am doing wrong?

[JDMather]

I do not see an attached file.

[raviburla]

Hi,

Could you please post the model? From what I understand in your post, I would guess that you are running into "Stress Singularity". Please probe the results away from loads and constraints ( Saint-Venant's principle) and check if the results are in accordance to your expectations.

Thanks,
Ravi Burla

[Rytcd]

Odd, I'll try again.

[DRLTKSE]

I think your physical material properties are wrong. 

Try the ones I've used here, or feel free to look them up yourself. 

[raviburla · Accepted solution]

Hi,

Thanks for posting the model. I think there are couple of things that need to be considered for this model.

1. The fixed constraint creates a stress singularity.

2. The applied force was not along the X-direction (perhaps this is in accordance to your design). But, based on the UX for force load - as soon as one clicks a point on a surface, the surface normal is computed at the pick point and load is decomposed to XYZ components. This likely what happened in this model.

I ran few experiments with the model.

1. Run the model as is, but increase mesh density. This shows stress concentrations and safety factor is very low as shown below.

2. Modify the force load to be along the X direction and recompute the solution. This time the results are little better because the bending component was eliminated and hence lower stresses.

3. Remove the constraint and apply a balancing force on the constrained face. Also, use the "eliminate rigid body option".

I would think that that the results from (3) above correspond to the reality (as we tried to eliminate artifical stress concentrations.). Please let us know if you have more questions.

Thanks,

Ravi Burla

[Rytcd]

Helped a ton, Thanks!