Composite floor modeling in SAP2000 1

[jsmatte]

Hi everyone,

I am currently working on a steel building with composite floors. My problem is that I don't know how to model the connection between the steel beam and the concrete slab (i.e. modelling the shear studs). Even if I put any constraint to the floor joints, SAP2000 does not consider the beams to be braced on their entire length (the unbraced length factor defined by the software is 1) and therefore I get an error message when checking the design (lateral-tosional buckling).

I know I can change the unbraced length factor manually in the overwrites, but I wanted to know if there was a more "realistic" way of modelling composite floors.

Thanks

 

[Dart82]

Try the following.

Add joints at the location of the shear studs; these points should be on both the frames and the slab shell. Now select the slab shell and assign automatic area meshing based on the point intersections: Select Slab > Assign > Area > Automatic Area Mesh > Mesh Area Based on Points On Area Edges > Check both intersection of straight lines objects and point objects.

 

[jsmatte]

Thanks for the reply Dart82, but I am wondering, how can you add joints to a frame element without breaking it into multiple elements. If you break a frame element into multiple elements, then sap2000 will consider each little element independently, and not as multiple elements forming a big frame element. if you understand...

 

[MaxJasper]

Use:
Assign >> Area >> Generate Edge Constraints

 

[johnbridge231]

In Sap2000, I would not use AUtomatic Edge Constraints, nor automatic slab meshing, but I would try to manually divide the slab using Edit->Divide Areas at the location of the intermediate joints (if any). Alternatively, you could use the Assign->Frame->Automatic Frame Subdivide option to force the program to internally break the frame into smaller ones.
The probelm with composite slabs would be their design since it is not supported. So you can do that in hand after getting the internal forces at the span of the composite beams, apply moment redistributions etc.

Rgs.

Analysis and Design of arbitrary cross sections
Reinforcement design to all major codes
Moment Curvature analysis

http://www.engissol.com/cross-section-analysis-design.html

 

[stressed]

ETABS offers much better modeling and design options for composite beam floors (vibration checks, composite beam design checks). If you want to account for the deck bracing the girders, select the beams then use Design menu>Steel Frame design>Design overwrites and assign something like .1 as the unbraced length ratio minor direction. Intermediate joints can be added along frame elements without dividing them using Draw>special joints or add the special joints button/icon. Special joints will divide the beams (or columns) for internal analysis only, but the results and design will use the members' physical length how they were originally drawn/modeled.

 

[johnbridge231]

The approach to assign a factor of 0.1 is not correct, since it depends on many factors, as:

[ul]
[li]The shear connection layout. An I mean whther the studs placement pattern and count and able to carry the developed longitudinal shear between concrete and steel[/li]
[li]The slab thickness. Imagine a thin slab and a relatively thick beam. Then the slab will provide a restraint to beam against buckling or vice versa?[/li]
[/ul]

Have a nice day.

Analysis and Design of arbitrary cross sections
Reinforcement design to all major codes
Moment Curvature analysis

http://www.engissol.com/cross-section-analysis-design.html

 

[stressed]

johnbridge, the deck acts as bracing along the length of girders and therefore unbraced lengths on those girders need to be adjusted. SAP2000 doesn't automatically adjust the unbraced lengths automatically.

 

[johnbridge231]

Yes, therefore a factor of 0.1 is irrelevant and depends on the factors I already stated above.

Analysis and Design of arbitrary cross sections
Reinforcement design to all major codes
Moment Curvature analysis

http://www.engissol.com/cross-section-analysis-design.html

 

[path2doyle]

The object is to build a model, not spen eons figuing the unbraced lengths. 0.1 gets the job done. When the beam is not going to buckle so stop quibbling about whether it's 0.1 or 0.2356L