How does Etabs9 modelate semirigid diaphragms 2

[asterixUC]

Hi, i'm from Chile and i'm designing a 26 storey bulding that needs to be modelated taking in acount the flexibility in plane of the floor. Normally i wold use several (2 or 3)diaphragms per storey when important discontinuities in the floor are present, but i have heard of people using semirigid diaphragms instead to do that. I have read every manual i have found to understand the modelation, but i still don't find a formal explanation that can make me trust in that thing. How is the mass matrix assembled, not lumped? How are lateral forces aplyed (accidental torsion) to a center of mass taking in acount the in-plane deformation of elements in the diaphragm? How are the cinematic ecuations of the constraint, if there are any?

If someone can answer me these cuestions or send a CSI document whit some aclarations, please do so, i'll be deeply thankful.

P.S.: Sorry for my english, i don't write/speak it very frequent

 

[stressed]

You don't have to assign a semi-rigid diaphragm to account for that flexibility in each floor. By default, the program uses the stiffeness and mass of area finite elements and frame elements modeled on each floor, and spreads mass based on material and section properties. You may or may not have to refine your floor area mesh to create more joints in order to obtain reasonable mass distribution.

As I understand it, assignment of area "semi-rigid" diaphragms is a new option and is used for one purpose, and one purpose only: to tell the program where the diaphragm extents are located for automatic generation of wind loads. That's it. Assignment of semi-rigid diaphragm does not create any rigid diaphragm at all. If you're not using auto-wind load generation based on diaphragm extents, you don't need to make the 'semi-rigid' area assignment in order to account for flexibility of the floors, and the program will by default use flexible diaphram based on material and section properties of what you have modeled.

If you have doubts, simply create a simple, but irregular one or two story structure with floor areas, beams and columns, and see for yourself the period and reactions to lateral loads, checking with and without rigid diaphragms. After assigning rigid diaphragm, you probably already know that in order to delete them, you'll then have to assign area rigid diaphragms again, but with diaphragm property = none. Let us know if that answers your question.

 

[asterixUC]

Thanks, but i allready know everything you just explain to me. Thank you anyway. I also readed that this semi-rigid diaphragm was only to automaticaly calculate and assign wind loads, but how are these loads aplied to the structure? then it is no longer posible to do that at the center of mass. Does the program automaticaly calculate how much load must be assigned to each node of the finite element diaphragm in order to represent te total lateral force? How does it? Does it also work for static torsion loads? Diaphragms are very usefull to assign loads representing sismic accidental torsion, thats why i'm interested in these new diaphragms.
I want to know if someone has allready experimented with these. I will build some simple models to experiment my self, but it's allways interesting and usefull to hear other opinions.
thank you again for answering

 

[WillisV]

I have discussed this same issue with CSI - their answer being:

"Yes, lateral loads can be applied to semi rigid diaphragm and the load will be distributed to points based on their stiffnesses and point locations on slab."

 

[asterixUC]

That's the answer i wanted to hear, thank you. I would like to have a complete explanation of how the program does it, but this is enough for me to start testing these in simple models.

 

[WillisV]

I would too but that's about all I could get out of them for now.