How make a sloped and flat roof diaphragm in SAP2000

[CorporalToe]

Hello, I am trying to make a flat roof and sloped roof diaphragm.

For the flat roof I think I did it right. I made a area thin shell element, I am using CLT wood for the roof, so I set the material as wood. I made the m11, m22, m12, v13, v23 modifiers 0.0001, also made the mass and weight 0, as I will manually add in the self weight loads. I added a diaphragm constraint to the joints of the diaphragm and ensured that the frame elements have auto mesh enabled. I think this is the correct way to do it but if you have some insight or recommendations please let me know. Btw I am trying to model a rigid diaphragm.

For sloped I do not know what to do, I have a 30 degree slope with a ridge beam and 1 beam on either side of the roof. I want it to act as a diaphragm, but my previous method is not working.

I have 2 questions:

Was by method for diaphragm model correct for the flat roof?
How do I model a sloped roof diaphragm?

 

[driftLimiter]

Are you modeling a flexible diaphragm or rigid?

How are the lateral loads being applied to the model? Via diaphragm forces or by some other means?

Lastly, sloped diaphragm modeling can be fraught with problems. In the past, even for steeply sloped roofs, I have seen engineers simply model it as flat, at least for the purposes of vertical and lateral load takedown this seems reasonable to me. If your using the model for design of slender elements, then obviously you will want to model the heights of those elements accurately.

 

[CorporalToe]

I thought by putting in the diaphragm constraint, it will model it as rigid.

The lateral loads would be applied as a linear distributed load on the perimeter beams, which should transfer it to the floor/roof diaphragms, and then the diaphragm will carry it into shear walls.

I am using the model to try to perform linear dynamic seismic analysis, so I want to model any elements that take lateral forces as accurately as possible.

 

[driftLimiter]

Perhaps consider using your shell elements to embody the diaphragm (semi rigid diaphragm).

When you put the diaphragm constraint, it doesn't even need a shell element present. It simply ties all of the nodes with that constraint together rigidly for lateral displacement. I think you will run into trouble using diaphragm constraint on a slope.

Would also consider applying distributed loads to the diaphragm shell elements.

 

[CorporalToe]

Do you suggest I remove the diaphragm constraints full, and just model as a semi rigid diaphragm?

Also since both my diaphragm and my frame elements are meshed, I thought the beam will be able to transfer loads into the diaphragm.

What is your suggestion on the stiffness factor6 for the shell elements to make them action like a diaphragm?

 

[driftLimiter]

Before I get to far into advising you I should qualify that I haven't tried this for CLT panels in SAP.

But in general I wouldn't mix diaphragm constraints and shell elements. For your application I think semi rigid shell elements is the ticket.

Although I'm calling it semi rigid, I would modify the properties to get it essentially rigid.

I think your proposition about loading the edge beams that are meshed should be fine, but I would do some verification steps to ensure you're getting the behavior you want.

 

[CorporalToe]

Well honestly modeling the CLT panels as flexible or semi-rigid is allow by the code, since my LFRS is masonry shear walls. So ideally I can just model them as flexible or semi-rigid.

Also in SAP there is a option to model semi-rigid in the diaphragm constraint what do you think about this?

Also, do you assign diaphragms for shear walls? I currently haven't done anything to them, they are modeled as thin shell area elements.

 

[driftLimiter]

The caution using diaphragm constraints is for nodes not at the same elevation, not sure that the software will even do that, or if it does not sure you will get what you need out of it. Perhaps the ASCE 7 criteria for rigid vs flexible diaphragm could be applied here to compare diaphragm deflection to wall deflection. I don't have a good feel for CLT diaphragm flexibility. What is the most important behavior you are looking to capture in the model? Shear wall loading, modal properties etc, or rather actual diaphragm shears and deflections. Commonly the global FEA model is used for the MLFRS and the diaphragms are checked elsewhere by hand. This is what I recommend. Then you can play with diaphragm flexibility to envelope the forces in the walls. Depending on the configuration of walls and diaphragms, there will be more or less pronounced differences in the wall forces for flexible vs rigid diaphragms.

For walls generally shell elements with stiffness parameters based on the cracked moment of inertia, for CMU walls i believe the standard is 0.5 Ig.
NEHRP Special CMU wall document provides good modeling guidance for the walls.

As with any global model, you want to ask yourself what do I need from this model.. Why am I making it? Significant caution/verification should be used if you are trying to model and design all of the elements in a single model.