Rigid Diaphragm Analysis

[coolad2007]

Hello everyone!

I am currently designing a 35 foot high warehouse with LFRS system consisting of precast wall panels and with roofing made up of precast/prestress double T joists which would be interconnected along joint lines as per MFR design guidelines. I have assumed that the diaphragm is going to be rigid but the catch is that there is no topping slab over it. Now my question is how to ensure that this double T joist slab would behave as a rigid diaphragm. These joists run from wall to wall (almost 88’-0” long) with joints along parallel line between them. I have never encountered this situation before. Am I supposed to give MFR connection forces (it may be either due to drag, shear or bending) to connect these joist along panels joint lines to make sure that the rigidity behavior holds true. In general these connections are by MFR but not sure what is the industry standard to approach this type of problem. I am not sure how to check deflection of such diaphragm to see whether this would be rigid or flexible.

The reason for this concern is due to the fact that I have varying rigidity along shearlines. There is lot of load getting dragged into stiffer walls and if this rigidity concept won’t hold good I would end with lot of shear in other walls which are relatively less stiffer but have a very large trib area.

Any reference material or theory guidelines would do. I did talk to my principal about it but was not convinced with his response so wanted to check opinions of rest of the engineers.

Thanks in advnace!

--ad

 

[LonnieP]

Coolad, I suggest you talk to the 'T' manunfacturer's engineer first thing. He has probably seen your situation before and can offer suggestions, AND he wants to make a sale so he should be more than helpful. If you're in a high seismic area don't forget about wall anchorage and cross building ties.

LonnieP

 

[JedClampett]

For double tees, they usually embed wires or weldable reinforcing in the top flanges at the edges in a little pcoket. You weld plates to these embedment pieces or weld them together if they're close enough. I've linked to a manual that shows these type details on sheet 9.
You could indvidually design them, but most of the time they're designed for the transfer load at the walls and called good. You would give that load to the precaster and he would supply the number and type of details required.

 

[KootK]

I dug into this issue a bit for an untopped TT parking garage that I worked on over the summer. All of my references, including the PCI handbook, treat TT diaphragms as rigid. Spancrete has a downloadable tech note on the issue that cites some relevant research.

TT flange slabs are obviously pretty rigid internally. It's the connections between TT's that introduces flexibility. I reconciled myself to the rigid diaphragm assumption by remembering that deflection is, mathematically, the integration of M/EI along the span of your diaphragm (similar for shear deformation). Consequently, local soft spots don't add as much to overall deflection as you might think.

You could specify an effective diaphragm stiffness or a maximum deflection in your drawings to ensure that you get a diaphragm that's rigid enough to suit your design assumptions.

The greatest trick that bond stress ever pulled was convincing the world it didn't exist.

 

[coolad2007]

Thanks a lot everyone on your inputs. I think for now I would go with the assumption of considering it rigid. I did find few other references and research articles to support this and I think i should be fine with it for now.
No specific codal requirements like chapter 21 of aci318 cover this situation directly which i think should be done in future editions to clear the air. Thanks a lot Jed for that reference.

 

[mike20793]

A rigid diaphragm assumes that any two points will deflect laterally the same amount. Of course this isn't always the case in real life, so ASCE 7 has limits about the differential deflection to determine whether the diaphragm can be idealized as rigid or flexible. Double tees are always going to be rigid diaphragms whether they are topped or not. Think about the diaphragm as a deep beam loaded laterally and you can visualize the deflection in the "mid-span" of the diaphragm isn't going to much different than the deflection at your lateral force resisting system. Typically, most wood diaphragms and some untopped, metal deck diaphragms are considered as flexible. As always, the diaphragm must be able to resist the lateral forces; so the transfer forces between double tees are usually provided so the connection Jed referred to. But, if you are ever uncertain of the flexibility of your diaphragm, just follow the rigid diaphragm check in ASCE 7.

Also, when I have a code permitted "flexible" diaphragm, I will still design my lateral system for the envelope between flexible and rigid. Some programs will allow you to design for this using a "semi-rigid" diaphragm.

 

[coolad2007]

Thanks Mike for your inputs. I am already very clear about flexible vs rigid diaphragm and governing codal clauses. However, my only concern were the joint lines long the double tee spans. Inherently, the entire rigidity of the diaphragm would depend upon the embed connections that are joining these Ts along their joint line. And my only concern was how to calculate the forces to make sure the diaphragm behaves as rigid one without any slip between joint lines. Anyhow thanks to Jed i referred the PCI Design Handbook on precast design which has cleared up pretty much all my doubts about this scenario.

--ad