Non-Linear Buckling Analysis

[georgievWLHEB]

Hello,

I have the following problem. I want to run non-linear buckling on a part to make sure it would not buckle out of the plane. The simulation runs and seems to always converge (in each iteration). However, after the last iteration is done it gives me a message that the simulation failed and the error I see is 5004 that the stiffness is singular on not positive definite. On the other hand, when I run a non-linear static simulation is is successful. Does anyone have an idea what might be going wrong?

[shigeaki.k]

Hello @georgievWLHEB,

I moved your post to start its own forum conversation, as the forum topic you posted to was last commented on in 2015. I am hoping that with its own post, your question will get more views.

The error E5004 is often associated with models that are not constrained adequately (see for example this post here). The description of the error can be found in the Nastran document here.

Could I check that you have defined contacts in your model?

Regards,

Shigeaki K.

-----------------

If my post answers your question, please click the "Accept as Solution" button. This helps everyone find answers more quickly!

[georgievWLHEB]

I have attached a picture of my model. This a planar rotary spring. I have constrain the slots at the ends of the four spirals and apply remote force at the holes at the center to simulate applying torque.

Running a nonlinear static stimulation produces results that seem quite reasonable compared to my theory.

Now the reason why I want to run buckling buckling is that if the spring is too thin (much thinner than this one) the nonlinear static analysis fails, which I believe is due out-of-plane buckling, that is the center of the spring buckles out. I am almost certain that this spring because it is so thick would not buckle out but it would be nice to have a good way of checking this.

Now the problem as I mentioned earlier is that when I run the nonlinear buckling it displays the error 5004. I have also attached the log file. Do you have any ideas what could be going wrong? Thanks a lot!

[shigeaki.k]

Hello @georgievWLHEB,

hard to say. Can you run a modal analysis and check that there are no rigid body modes i.e. there are adequate constraints.

Regards,

Shigeaki K.

[shigeaki.k]

Hello @georgievWLHEB,

just wondering if there has been changes with your work.

Regards,

Shigeaki K.

[shigeaki.k · Accepted solution]

Hello @georgievWLHEB,

just found something else. Could you check that the constraints and loads from the 1st subcase is copied into the 2nd subcase. This is required so that model is constrained for the eigenvalue solution. You should be able to drag-n-drop the constraints and loads from the 1st subcase to the 2nd subcase in the tree.

Regards,

Shigeaki K.

[georgievWLHEB]

Hello Shigeaki K.,

Thanks a lot. This was the problem. I did not copy the constraints and the loading.

Nikola