Strange deflection display for simple 2D frames.

[Anonymous]

Hi,

Please could one of the support explain why the simple 2D portal frames I have developed show strange deflection patterns.

I have developed a 2D test model to justify the general behavior of a larger 3d model by doing the following:

I developed 3 identical portal frame models (as separate structures in one rtf file) and applied an identical nodal load to each at the same node. All 3 models have fixed supports.

I then kept all the member joints fixed in the first model. The second model I applied moment releases "fixed-pinned" to both vertical columns to release moment between the beam and column. The third model I applied "pinned-pinned" moment release to the beam. 

Model 2 and 3 should have identical behavior in both load distribution and deflection. However, this is not the case. The bending moments, shear forces and axial forces are identical in both but the displacement plots are both incorrect.

In model 2 it displays the displacement of the columns as being in double curvature which cant be as there is no reversal of bending moment in the column. It displays the deflection of the beam correctly as 0 (only axial force in the beam).

In model 3 it displays the displacement of the columns correctly but displays the beam as having single curvature which cant be the case as there in no bending moment in beam due to the releases.

The correct answer would be a combination of model 2's beam and model 3's columns.

Please confirm the issue is? I have attached the test model.

Regards

Dane

[mustafahesenow · Accepted solution]

1

[Anonymous]

Hi @mustafahesenow

Thank you for the response.

Can you explain why there is a difference?

Regards,

Dane

[mustafahesenow]

1

[Anonymous]

Hi @mustafahesenow

Thanks for the link. I have read this however it doesn't answer my question as to why the difference? Why would I have an option to display deformation if it gives me the incorrect deformation shapes? 

The help file tells me what the two options consider but not why.

I appreciate the assistance.

Regards

Dane

[Artur.Kosakowski]

Hi @Anonymous

As the first method of determining the shape of deformation doesn't include some factors it may happen that this shape is inaccurate. Assuming that you may want to ask why it is needed at all the answer is that the later cannot be used for surface elements.

I hope this helps 

If you find your post answered press the Accept as Solution button please. This will help other users to find solutions much faster. Thank you.

[Anonymous]

Thank you @Artur.Kosakowski !

This has sparked some new question in this regard.

I understand that the two take into account different variables but why is the first option available in the 2D frame design module (where surface elements can't be defined) if it gives the incorrect deflection shape.

Or am I misunderstanding the use of the first option of deformation?

Should I always be using exact deformation of bars when considering frame structures?

Which option would be appropriate if I am working to the shell design or building design modules? If the answer is the first could there be the possibility that the deformation of the bar elements is displayed incorrectly as in my test example?

Regards

Dane

[Artur.Kosakowski]

Hi Dane,

I understand that the two take into account different variables but why is the first option available in the 2D frame design module (where surface elements can't be defined) if it gives the incorrect deflection shape.

As you may want to see the shape of modes for modal or buckling analysis

Should I always be using exact deformation of bars when considering frame structures?

Apart form the above mentioned situation, the answer is yes.

Which option would be appropriate if I am working to the shell design or building design modules?

When you want to see the deformation of the whole model then the first one. If you want to closely examine bars only then the exact one.

If the answer is the first could there be the possibility that the deformation of the bar elements is displayed incorrectly as in my test example?

Yes, but due to the fact that beams are usually under slabs (so they are divided into smaller calculation elements with calculation nodes at their ends) this is unlikely to happen.

If you find your post answered press the Accept as Solution button please. This will help other users to find solutions much faster. Thank you.

[mustafahesenow]

1

[Anonymous]

Hi @Artur.Kosakowski, @mustafahesenow

Perfect, thank you for clarifying. 

Regards,

Dane