Steel beams supporting hollow core

[bsh117]

The architect has asked me for an alternate design for the standard composite metal deck framing.

I have a floor system consisting of hollow core plank bearing on steel beams. Can the steel beams be considered composite if I size headed studs on them?

Thank you!

 

[mike20793]

I would not consider any composite action. The planks are precast and the topping slab is cast in place and often have different compressive strengths. Precast planks make it impossible to get studs in them and the detail at beam bearing is not conducive to consider any appreciable effective flange width based on topping slab alone.

 

[DETstru]

You can't get any composite action between hollowcore and steel beams. Typically there is just an embed set into the hollowcore and you weld it to the steel beam. As mike20793 said, you can't get studs into the hollowcore because it's precast in the shop and simply put together in the field. If you look at any hollowcore manufacturers catalog you can get a sense of what the typical connection details are like (see this one:

http://goo.gl/2O4Iwr)

 

[bsh117]

Thanks!

 

[Agent666]

I've done it on numerous projects. The main issue is with hollowcore that you cannot get the typical post splitting reinforcement to the bottom of the slab in the region of the high bearing forces at the base of the studs, this reinforcement is required in metal decks to ensure the studs have some dependable strength after the concrete splits (often the continuous deck provides this function in metal decking slabs).

Another semi related issue is that the hollowcore can continue to shorten under the prestressing force (via creep/shrinkage mechanisms), this means a crack can form at the end of the unit which means that the limited concrete around the stud between two units on either side of the beam is unconfined by the units, so it's like putting studs right on the edge of a free slab edge and there is a corresponding reduction to stud capacity.

You have to really go back to first principles in terms of fundamental stud behaviour, a good text that explains the fundamentals is
Oehler, D.J. and Bradford, M.A. (1995). “Composite Steel and
Concrete Structural Members – Fundamental Behaviour.” 1st
edn Elsevier Science.

Also I've found to suppress brittle failure modes, such as longitudinal splitting you really needs two studs across the flanges, this pretty much ensures the studs can yield prior to splitting (because you cannot get any post splitting reinforcement in where it's required this the only ductile mechanism for failure of the studs). We also put about 33% more studs than required (it's based on some statistics around only having the one failure mechanism) and it helps us sleep at night!

Good luck.

 

[Agent666]

One more thing, for beams only supporting HC from one side, we always span the HC past the centreline of the beam to avoid torsion on the beams. Often HC spans are quite long and this would be an issue, so we generally avoid it altogether.

 

[hokie66]

Agent666,
Where do the studs go? I assume in the joint between slabs which are grouted, but that seems an onerous and unrealistic coordination job.

 

[JasonMcKee71]

You should also check the longitudinal shearing in order to confirm that composite action develops.

Jason McKee
proud R&D Manager of
Cross Section Analysis & Design
Software for the structural design of cross sections
Moment Curvature Analysis
Reifnorcement Design etc.

 

[cooperDBM]

You should be aware of recent European research showing a potentially substantial decrease in shear strength (web shear) of hollow-core supported on flexible supports. I believe it relates to horizontal shear in the plank ends in the transverse direction caused by the beam deflection. It adds to the end web shear stress caused by the span of the planks. The Eurocode recognizes the issue but doesn't give any guidance. I haven't heard much here in North America but it was brought to my attention by Peikko who makes Delta beam (a composite hollow-core support beam - not a plug). If you're interested google "hollow-core on flexible supports" and Matti Pajari. Not sure if there have been any failures outside of the lab but it might be worth being conservative.

 

[Agent666]

Hokie66, you need a beam with a wide enough flange to accommodate the seating plus a minimum 190mm wide concrete haunch (based on 2 rows of studs at 5 stud diameter spacing plus 2.5 stud diameters each side) between the units. This concrete is just poured with the topping. We usually specify a 370mm wide flange on a welded beam.

We also are required to feed bars into the core and concrete fill at least 2 cores 800mm per unit in this part of the world to address some of the issues cooperDBM is referencing. We've had to do this for 10 plus years, mainly to address frame elongation issues which can cause a loss of seating & the end of the units to fail due to insufficient development of the strands right at the ends of the units.

It's important to check the longitudinal shear, especially if your studs don't extend above the hollowcore. You may need some stirrups in the gap between the units that extend into the topping. Check multiple critical planes, not just the thinnest topping section at the ends of the unit. If you have sufficient space between the units you can run some larger bars along the beam at the base of the studs to enhance the longitudinal shear capacity. We often drop some 25mm bars in here to enhance the capacity in longer more heavily loaded spans.