Elevated Concrete Slab - Garage Over Basement 1

[adamt83]

I'm designing a suspended concrete slab over a concrete basement walls for a house garage. 24'X24' plan dimension.

Ideally I would like to avoid beams, but I know locally (Missouri) it has been done with one or two steel W shape beams to support the slab. However, this is the first time I am designing/detailing one and want to be sure I cover my bases.

2009 IRC is the local code and the live loading for garages is 50 psf live load. Also for elevated slabs a 2000 lb concentrated load applied over 20 in. sq. area. Is it intended that these loads are applied simultaneously? I would think not as the concentrated load is presumably to simulate a jack load but would appreciate clarification.

I plan to use a ledge, say 4"-6" wide?, in the perimeter foundation wall for the slab to sit on. I would also slope the ledge from the back wall towards the front doors so that the resulting slab is sloped towards the doors for drainage. The top of foundation wall would be constant, say 2-3" above the top of the slab on the back wall and say 6" above the slab on front, but notched down so the slab could sit on top of the wall at the garage doors.

As a two way slab, I presume that a simple way to calculate the appropriate shear and moments would be to use PCA's Rectangular Concrete Tanks with the various coefficients for flat plates? I do have that book but don't have access to design software--want to keep this as simple/direct as I can by hand. What would be the best way to detail the connection of the slab to the wall ledge? Bend some vertical wall reinforcement into the slab and use a pinned edge condition for the slab? Or would the reinforcement dowels create more of a fixed condition?

If using a beam at midspan in one direction (or say 2 beams at 1/3 points) would I be able to design as a one way slab since the plan ratio would be 2 or greater? Any special detailing for the slab over the beam? Should I simply pocket the beam into the foundation wall the same 4"-6" as the ledge above for the slab?

I wasn't planning to use galvanized steel decking as a permanent form and for composite slab action, but would appreciate feedback here as well. I'm not all that well-versed on such design/detailing. I do want to keep things though fairly traditional/simple for the typical local residential builder/concrete flatwork contractor.

Finally I think my recommendation would be for sealing or epoxy coating the floor for extra protection to the slab from deicers. Since there is liveable space below I don't want want any problems with slab. Any other thoughts here? Maybe epoxy rebar?

Thanks in advance for the assistance.

 

[jayrod12]

Stiffeners don't provide restraint against buckling of the top flange. At least not a calculable amount. Therefore to prevent the top chord from bucking laterally, you must restrain it with the slab.

The easiest to install in new construction are shear studs welded to the top of the beam. This is done in the shop and is shipped to site ready to go.

 

[jdgengineer]

Seems like welding on some nelson studs in the shop would be pretty easy to do and inexpensive and would help restrain the top flange from buckling. Probably even at 4'-0" oc would be more than adequate. I can't imagine this costing more than a few hundred dollars and seems to be money well spent in my opinion. We would typically avoid this with metal deck due to sequencing issue of installing the deck, but if you are omitting the deck I don't see a problem with these be shop welded.

 

[KootK]

I understand the top flange is in compression, but don't totally get the "slip" that occurs between the beam and slab especially if the top flange is stiffened to prevent rotation.

It's the slip implied by the deflected shape shown below for lateral torsional buckling. As Jayrod mentioned, stiffeners don't do much of anything to help that. In many cases, I suspect just plain old friction between slab and beam would be enough to restrain the top flange but that's not a risk I'd be willing to take myself. Not for the cost a few studs. If the slab soffit is constructed at all lower than the top of the beam, the beam flange may well be keyed into the slab in a way that prevents movement. I'd like get a look at the underside of these other projects that you mentioned to see what's up there.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[adamt83]

Thanks for the responses. As an alternative to the studs would it be possible to use a deeper beam section to increase Lp greater than Lb (the span length)? The basement below is tall (14') so a deeper beam reducing headroom won't be an issue for the owner.

 

[KootK]

That is possible if not particularly efficient in a material sense.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[jayrod12]

It most certainly is an option. However generally not economical. And more often, it's easier to just go to a stouter beam (heavier but same depth). You want something that has enough out of plane bending strength to prevent itself from buckling.

 

[adamt83]

Thanks. Understand that it may not be the most economical solution just wanted to be sure I was thinking through the options properly. Thanks again.

 

[KootK]

You could always just Hilti in some bolts beside the flange after the fact too.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[HouseBoy]

A deeper beam will not get you to a greater Lp as readily as a heavier beam. That is generally why it is not considered "economical". However, I think MOST considerations of economy revolve around the ordinary need for multiple pieces. Since you have one beam only, the trade off of the cost of a heavier beam might not be so bad.

Post installed anchors might be the simplest but I'd also consider just having holes drilled in the top flange and adding bolts with a nut above and below the flange as a way to get the top flange well anchored by the slab. Generally I think 2% to 5% of the flange force is recommended to be sufficient to provide the needed bracing.

Thinking about costs, I'd weigh the cost of the bolts (or any other strategy) verses the additional steel weight assuming something like $2 per pound. That ought to be a reasonable estimate.

 

[XR250]

I vote squat beam and friction. If it has load on it, there should be enough friction to brace it. If it has no load then no LTB can occur.

 

[KootK]

It might pay to ask your builder what they've got in mind for formwork. Some setups lend themselves to an embedded top flange in which case bracing becomes moot.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[adamt83]

I quickly found out that for a beam of this length (24') I won't be able to make it work without bracing the top flange. A ry high enough is literally off the charts. It was a good exercise though to understand the options/limitations.

I'll probably proceed with shear studs. Besides just making an assumption on type/length/spacing, can anyone point me to the appropriate procedure for design?

If I'm still hell-bent against studs, I guess one other possibility I could pursue would be since Lb > Lr is to calculate the critical elastic moment, Mcr. I could then use that Mcr as Mn. Basically, use LTB as the limit state for flexure rather than yielding, correct?

 

[KootK]

can anyone point me to the appropriate procedure for design?

In practice, I'd just provide 3/4" diameter studs at 24" o/c and sleep easy.

In theory, I suppose that one technically ought to:

1) Use AISC appendix 6 to work out the strength demand on the studs such that they can serve as bracing.

2) Design the studs per AISC to meet that strength demand.

Normally you consider stiffness as well but with the stiffness being provided by a monolithic concrete slab here, there's no need/point.

Basically, use LTB as the limit state for flexure rather than yielding, correct?

Yup. That should shake out of the code specified moment calculation procedures however. No need to reinvent anything.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.