Problem with bonded contact changing stiffness of non-contacting parts of model

[gordon.fox]

Hello,

I am having a problem where applying a bonded contact to one part of a model is stiffening other parts of the model that shouldn't be affected. I have been able to recreate this behavior is a simplified cantilever beam model. Has anyone else seen this? Is there some way I am setting my model up incorrectly?

The simplified model that shows this problem is a cantilever beam, fixed on the right end, and with a load at a point in the middle. Applying a bonded contact to the "outside" of the load application point drastically changes the deflection of the beam, which it should not. I first modeled the two parts as one solid in model "H Assy.iam". The deflection of H at the load application point is about .087 inches, which matches analytical calculations very well. 

Then I modeled it as two parts bonded together, "F G Assy - Shows Problem.iam", and the deflection at the load application point is only .058 inches, significantly different.

If anyone has seen this issue before, any help would be appreciated!

Thank you.

[gordon.fox]

Hello,

I am having a problem where applying a bonded contact to one part of a model is stiffening other parts of the model that shouldn't be affected. I have been able to recreate this behavior is a simplified cantilever beam model. Has anyone else seen this? Is there some way I am setting my model up incorrectly?

The simplified model that shows this problem is a cantilever beam, fixed on the right end, and with a load at a point in the middle. Applying a bonded contact to the "outside" of the load application point drastically changes the deflection of the beam, which it should not. I first modeled the two parts as one solid in model "H Assy.iam". The deflection of H at the load application point is about .087 inches, which matches analytical calculations very well. 

Then I modeled it as two parts bonded together, "F G Assy - Shows Problem.iam", and the deflection at the load application point is only .058 inches, significantly different.

If anyone has seen this issue before, any help would be appreciated!

Thank you.

[John_Holtz · Accepted solution]

Hi @gordon.fox 

From a forum post last week, use Offset Bonded contact instead of bonded contact when the model has curved surfaces bonded together.

John

[John_Holtz · Accepted solution]

Hi @gordon.fox 

From a forum post last week, use Offset Bonded contact instead of bonded contact when the model has curved surfaces bonded together.

John

[gordon.fox]

That fixed it. Thank you so much!

[gordon.fox]

That fixed it. Thank you so much!