bolted beam connection

[Anonymous]

Hi,

I try to analyse a bolted beam connection. I want to do a linear and nonlinear Analysis on this joint.

I used preloaded bolt connectors to connect the headplates to each other.

I had the problem that the bolts of the lower row transmitted compressive forces so I decided to model the bolts as a solid. I can not find an adjustment to apply a non-linearity to the bolt connector so that it no longer transfers compressive forces. Is there a setting for that?   

To apply the preload I have used the Option 3 from the following post: https://www.google.com/search?q=%C3%BCbersetzer&ie=utf-8&oe=utf-8In this case, I have the problem that the deformation of the screws no longer seems correct. You can see that in the attached pictures. In nonlinear analysis, the model no longer converges and the calculation is aborted with the error E5076. Does anyone have an idea how to solve this? I attached the model with all the analyzes, maybe someone you me can help me. I know the net is not perfect, but to test if the model is running, I think it would be enough

Thanks a lot, best regards

Fabian

[John_Holtz]

Hi Fabian,

Can you be more specific when you write "the deformation of the screws no longer seems correct". What result is incorrect in the image you provided?

In your fourth analysis "plastisch Vz+My - solid bolts", the bolt connectors do not have any preload to put the bolts into tension. Is there a reason you change the preload to 0?

[Anonymous]

Hello John,

first of all thank you for your help.

At the bolt, I would expect a constant curvature. The discernible offset is not what I expected. The strain is also very high and influences the deformation of the head plate. The deformation of the connection is very different from other analyzes such as the analysis "elastisch Vz+My - connectors". Even when analyzing with the stress analysis from inventor, or other FEM programs the deformation of the head plate is different. 

I removed the preload in the fourth analysis for a test and forgot to pick it up again - that was my mistake. Whether with or without preload, the calculation does not converge. 

Maybe something else to my intention. We are an engineering office in the field of lightweight tensile structures. We mainly use Autodesk Inventor^© for the construction. I'm currently investigating how we can use Nastran In-CAD^© instead of other programs for calculating connections.

best regards

Fabian

[John_Holtz]

Hi @Anonymous 

I ran both linear analyses (but forgot to correct the bolt preload ). The deformed shape does not look that dissimilar to be comparing the bolt connector and the solid bolt.

• There is some shear, and since the bottom connection is in compression, the "S" shape that the bolt should have is more apparent than at the top connection where the bolts are separating. (If the beam element used for the bolt connector would have more elements along the length, you would see a similar shape in the beam element as you see in the solid bolt. Since the beam/bolt is only 1 element, it is not possible to visualize the complex bending that can occur.)
• There is a small gap between the two plates (0.1 mm). I do not know what effect that is having on the analysis. It might be a good idea to remove it from the model. (You can see the effect of this gap in the two figures below. It looks like the pre-load analysis removes the gap, and the analysis with 0 preload allows for a larger opening of the plates.)

Figure 1, Analysis 1. Bolt connectors with preload

Figure 2, Analysis 3. Solid bolts with 0 preload.

[Anonymous]

Hi John,

thanks for the explanation to the bolts. I've tried it again with preload, you're right it looks very similar. Only the slightly larger displacement at the end is still present but I do not see that as a problem. I have attached a picture of the real bolt deformation - so I expected it. It was probably only due to the scaling of the deformations which made me think that the analysis is wrong. 

I modeled the gap between the two headplates because the analysis did not deliver a correct result otherwise. I attached a picture of the analysis without a gap. 

Is this possibly the problem, why do I get such a high compression in the bolts on the analysis with connectors? How can I edit the model so that it works properly without a gap? 

Do you already have an idea why nonlinear analysis does not converge?