Panel calculation model - modelling rigid diaphragm

[skellyRSA2023]

Hi,

i am currently modelling a precast concrete wall frame. the model consists of concrete wall panels and concrete floor slabs. I have ran calculations for the model with two different diaphragm configurations for comparison of results. The panel calulation model configuration for each one is noted below. I am trying to model a typical rigid diaphragm using the concrete floor deck to transfer all the lateral loads to the shear walls of the building which i would have thought option 1 below would be the correct model to use.

1. Elastic stiffness - no finite element; stiffening diaphragm - partial stiffening xy plane; transfer of loads - simplified one way. 

2. Elastic stiffness - no finite element; without stiffening; transfer of loads - simplified one way. 

Thus difference between the two calculation models is partial stiffening and without stiffening. However when comparing the results for moments applied at the base of the walls in plane between the two calculation models the wind loads are significantly different. Option 1 above which has partial stiffening xy gives a significantly less moment at the base when compared to running the same model with the deck having without stiffening. Does anyone know why this is? I would have thought it would be the opposite way round? 

I am trying to find the best way to model a rigid diaphragm to give me the correct base loads based on the in plane stiffness of the walls, i.e. Considering the deck as rigid. Could anyone give me some advice on how to do this? And why there is significantly different lateral load results at the bases when running the two scenarios on the exact same model layout. 

Also, when using the without stiffening option, how does this transfer loads to the vertical elements of the frame?

Thanks,

simon. 

[Mohamed.E.Salem]

Hi @skellyRSA2023

as i know 

without stiffening means no additional forcing constraints will be applied to the shell element .it will be calculate as  a finite element . if 'no finite element ' option is set ,the shell will act as a cladding and .vertical element will be independent of each other.

 stiffening diaphragm means additional forcing constraints will be applied to the shell element and vertical element will be dependent of each other( slab can have flexable or full body action)

[skellyRSA2023]

Ahmed,

Many thanks for the response. when you say forcing constraints what exactly do you mean by forcing constraints?

Basically I am trying to recreate a calculation for lateral load distribution based on the in plane stiffness of the individual walls only as per say a typical manual hand calculation for a horizontal load diaphragm distribution thus assuming that the floor deck is a stiff diaphragm and transfer the horizontal load to walls based on the in plane stiffness of each of the walls. For this I assume stiffening diaphragm as per the first sheet on the attached pdf? is this the correct way to model this?

Simon.

[Mohamed.E.Salem]

 forcing constraints means inplane movement (XY) and z rotations 

in your case,

 - if you want slab act as cladding and it will be  ignored in analysis beyond distributing loads to the supports , choose without stiffening option .

 -if you want to apply constrains that xy movement and z rotaion  of nodes connected to the slab must be fixed to each other  , choose partial stiffening in XY option .

[skellyRSA2023]

Ahmed,

Thanks again. I have created two models for a comparison study and applied a lateral load of 2kN/m2. one model has partial stiffening xy and the other is without stiffening.

The partial stiffening model gives me results I would expect for shear and moment at the base of the wall running in the plane of the two bracing walls (2kN/m2 * 6m * 3m high)/2/2 = 9kN shear at deck level * 3m high = 27kNm moment (RSA = 25.4kNm) which is close enough.

The model without stiffening gives me zero for the moments on the base of the walls.

Thus I assume then for this type of structure I should be using Partial stiffening in xy. See two sample models attached.

Also,"if you want slab act as cladding and it will be  ignored in analysis beyond distributing loads to the supports , choose without stiffening option" does this mean it will ignore all lateral loads applied and distribute vertical loads applied to the supports only or does it distribute lateral loads through the deck to supports also?

Apologies for all the questions, its just that either one gives me significantly different results.

Thanks.

Simon.

[Mohamed.E.Salem · Accepted solution]

HI @skellyRSA2023 

The model without stiffening gives me zero for the moments on the base of the walls.

 this is expected this mean no stifiness in plan it will ignore all lateral loads applied and distribute vertical loads applied to the supports only.

to have the same manual result , you may decrease mesh size or use full stiffening (rigid body) optioin

this link may be helpfull---https://forums.autodesk.com/t5/robot-structural-analysis-forum/slab-rigid-or-flexible-diaphragm/td-p...

[skellyRSA2023]

Mohamed,

many thanks for this and gives a good insight. I think for the precast frame the partial stiffness in the xy is maybe the best option. However you do get less moments on the wall bases with this rather than manual calculations, however it may be due to the fact that I have used the wind simulation in RSA rather than a applied a uniform load directly to the panel face. The reason I originally asked was because I took a model and used a "partially stiffened xy" diaphragm to model deck then ran the same model using the diaphragm modelled as "without stiffness". When modelled with the partially stiffened diaphragm the results at the base are reasonable however less than say manual calculations. when the model is modelled without stiffness the results for the in plane moments are massive and some give me zero thus there must be an issue with running the same model without stiffening for the diaphragm.

Rgds,

Simon.  

[pics.munmun]

@Mohamed.E.Salem  if i select without stiffening, then vertical members will be independent of each other. Does this means that all the load applies on the panel will be resisted by panel/slab itself and it does not transfer any loads to adjoining bars??

Please correct me if I am wrong