Flexible diaphragm not working correctly:have I missed some settings?

[corne_mo]

I have made a small test file with a 1000x1000 mm panel in the XY plane with all 4 edges simply supported. The thickness of the panel is 20 mm and it has a very low E-modulus. I first did a run with the panel calculation model set to shell. The deformation of the panel in Z direction is 213 mm, thus giving large displacements and requires a diaphragm type solution.

I therefor changed the calculation model to "slab-flexible diaphragm". The results are still exact the same though. 213 mm deflection and no horizontal reaction forces.

It looks like I am missing some settings required for this type of analysis. Can anyone please guide me to make this analysis work?

Thanks in advance for your reply.

[tony.ridley]

Hi,

Not sure how you get the model to even work with flexible diaphragm and supported with pinned supports - I get an error with having a support in a slave node. 

If you want the horizontal reactions, keep it as a shell model and run P-Delta analysis.  This will account for the tension force generated by the plate deflecting (i.e. catenary action). 

Tony

[corne_mo]

Made it running by selecting "RLINK elements (rigid links)" in the Job Preferences --> Structure analysis menu.

I tried the P-Delta analysis, but it doesn't converge. Maybe my model is a bit overloaded?

Any other suggestions to make this small model run OK?

[Rafal.Gaweda]

1. Change E modulus

2. Send us file to look at

[PaulGrimes2282]

Hello

I have a similar problem with flexible diaphragms...

I have been trying to model a CLT floor panel as a felxible diaphragm in order to provide lateral stability to a building.

When the slab is modelled as a 'shell' without stiffening, then by changing the Youngs Modulus of the floor material I can control the deflections of the slab. But I am a bit confused as to why the material propoerties of the slab have an effect on its deflection when the slab is modelled 'without stiffening'.

If I try to model the slab as a 'flexible diaphragm' I get errors about having supports under slave nodes.

It's as if a stiffening diaphragm with 'flexible stiffening' behaves as a rigid link?

[Rafal.Gaweda]

When the slab is modelled as a 'shell' without stiffening, then by changing the Youngs Modulus of the floor material I can control the deflections of the slab. But I am a bit confused as to why the material propoerties of the slab have an effect on its deflection when the slab is modelled 'without stiffening'.

Are you trying to compare these results to simply supported beam results ?

If I try to model the slab as a 'flexible diaphragm' I get errors about having supports under slave nodes.

You can not have supports in slave nodes

It's as if a stiffening diaphragm with 'flexible stiffening' behaves as a rigid link?

Any kind of stiffening = proper rigid links used on model

[PaulGrimes2282]

I am trying to understand what the difference is between the two options under panel calculation model; 'flexible stiffening (in XY)' and 'full stiffening (rigid body)'. Are you saying that both of these will apply rigid links to the model?

The floor slabs are supported by load bearing CLT walls. But I have problems using linear supports with the mesh because it sees this as a support under a slave node. What is the best way round this?

[Rafal.Gaweda]

Make such model \ test models and check results