Understanding the structural system for a slab on grade with point loads and uplift? 1

[psychedomination]

Hi there, I am designing a slab on grade for a standard iso shipping container to rest on, where it will remain as a storage container.

This should be quite a simple analysis but I am struggling a bit and trying to understand how the structural system will behave so that I can carry out a conservative analysis and design. I’m not sure if I am going about analysing the structure in the right way, so I can use some help. Essentially I am looking for a lower bound analysis approach, where I can quickly and conservatively do the checks by hand. I don’t have access to FEA and for something this small wouldn’t expect to need it.

The container has a dead load of 22kN and is subject to a 66kN normal force from wind (acting on the longer side), which results in a moment about the container minor axis of 101kNm. I am just designing the foundation for the container not doing anything to the container itself.

I was thinking of a connection detail for the container and think twist-locks welded to a base plate anchored into the foundation should work fine.

With this system, there will be concentrated loads from the container at each of the four twist-lock locations.

To get the loading on each twist-lock connection, I split the system in half along the long side (effectively splitting the moment in half)… therefore 101/2 = 50.5kNm per side.

I then converted the moment into a couple to get the compression and tensions forces in the front and back twist-lock connections respectively; using 50.5/3.05 = 16.56kN in tension and in compression. For analysis sake say front twist-lock in compression and back in tension.

Load summary :
- Container is 22kN in total so 5.6kN vertical load in each corner twist-lock location.
- Concrete strip footing has a vertical load of 4.47kN/m
- tension (uplift) on one twist lock is 16.56kN
- Compression on one twist lock is 16.56kN

I sized the footing to prevent sliding and overturning but where I’m kind of stuck is how to conservatively calculate the bending moments to design the strip footing, considering the point loads and uplift? I’m essentially viewing it like a beam? See attached sketches/rough calcs.

 

[jayrod12]

So if your design load is now roughly 33 kN and your dead load is 22kN, using my local code factors that would result in only a 6.6 kN uplift load at each anchor point. That hardly seems ridiculous to be able to resist. The way I figure the numbers you really only need roughly 8-10 kN worth of concrete dead load to be mobilized to prevent it from lifting off the soil at each anchor point. Maybe a 900x900x500 pad at each anchor point. Seems much easier than all this crazy analysis.

Then what I would do after that back of the hand calc, is provide a thickened portion on only the two long sides that is 900 wide x 500 deep since that's where all the actual gravity loads will come down with a thinner slab in between.

To me, the shipping container would be considered a rigid body, so your slab itself doesn't necessarily need to act like a rigid body for overturning. But maybe I'm missing something. The thickened portions on each side would also allow you to use the passive pressure of the soil to limit sliding concerns.

 

[ProgrammingPE]

I was still referring to overturning in the short direction. If you are looking at a section through the long direction of the foundation (22 feet wide), you have uplift reactions of 10.96kN (per your original calculations) at each end. That uplift is resisted by the weight of the slab which is a uniform downward load. For the entire slab to resist uplift, it needs to be capable of spanning the 20 feet between your uplift reactions, otherwise it will crack and the weight of the slab in the middle will no longer contribute to preventing overturning.

Structural Engineering Software:

http://www.structuralcentral.com

Structural Engineering Videos:

http://www.youtube.com/channel/UCOj2mXXf3ZbZtgxTc6BtkGg

 

[jayrod12]

I'm talking about a section through the short direction. He indicated he recalculated his wind load down to 33 kN. Assuming the box is 8ftx8ftx22ft, the 33 kN acts 4 ft up, and the 22 kN self weight acts 4 ft from the corner, therefore the overall uplift at the one side of the 8 foot box is only 16-20(ish) kN after you factor up the wind and down the dead a bit. split that total uplift between the two anchor points and we've got quite small uplift loads that to me are more easily counteracted by shear weight of localized pads.

Again, maybe I'm missing some crucial point, but in my mind, the container itself is a rigid box and therefore we don't need to consider the the stiffness of the slab across the entire width. Heck, if you could get the sliding to work out with only passive pressure on the sides of the pad, you could technically provide the pad only and eliminate the slab altogether.

To me, this scenario is much different than say a retaining wall footing where the moment is being directly imparted to the footing, this is more of providing foundations for dedicated point loads.

 

[ProgrammingPE]

@jayrod12: I completely agree that designing them as localized pads is the way to go. My comment was in regards to the original overturning check that was done which considered the entire weight of the slab to resist overturning. I had pointed out that the only way that this is possible is if the slab can span the full 22 foot width. This is not required if the overturning check is OK if you only consider the thickened slab at the ends.

Their bearing pressure calculations were resulting in negative pressures which is not possible. Their overturning check, though, said that things were OK since they had used the entire slab which was not consistent with their bearing pressure calculations.

Structural Engineering Software:

http://www.structuralcentral.com

Structural Engineering Videos:

http://www.youtube.com/channel/UCOj2mXXf3ZbZtgxTc6BtkGg

 

[SlideRuleEra]

I guess that is what I am struggling to calculate/understand is how the thickened edge strip behaves with the 8" slab on grade when taking the overturning load.

No need to struggle, your approach to a design will work and is easy to analyze. The 8" portion of your proposal contains (structurally worthless) welded wire mesh... ignore that it is even there for either sliding or overturning cals:

Dimensions of the two end foundations may need to be adjusted, but that's not a problem. Make a careful decision on "width", "length" and "thickness" dimensions for those twin foundations. There are pros and cons for each dimension.

Including the middle (currently 8" slab) is now optional. I would either omit it or construct an independent "thinner" version of it as shown in my sketch. The only purpose for the "thin" slab is "housekeeping" (preventing weed growth, drainage, etc.)

 

[SlideRuleEra]

A mat foundation would definitely work as you suggested (albeit would end up being quite a large foundation).

Not necessarily, your proposed 12" / 8" slab contains about 5.9 yd³. A 12" slab has about 8.1 yd³... an "extra" 2.2 yd³ plus some added rebar. For this "added" concrete/rebar, you get:

1) Simpler rebar placement.
2) Simpler excavation and subgrade preparation.
3) No significant change to forming.
4) No significant change to concrete placement.
5) No change to concrete finishing.

All of these together mean that by buying an "extra" 2.2 yd³ concrete (at cost) your client "saves" appreciable, expensive construction labor.

The reason I say a 12", two-mat slab... at our generating stations, based on several few decades of experience, we decided that 12" two-mat slabs (#4 @ 12", each way, top and bottom) are a cost effective compromise for demanding, unpredictable loading. Things like point loads from outriggers of large truck cranes and routine use by off-road loaders and bulldozers (Caterpillar D9). That is one tough-cookie slab design.

 

[psychedomination]

Thanks everyone! You have been extremely helpful.

With the reduced wind force of 33kN, I'll size the twin strip footings to resist the respective sliding and overturning. I will then as SRE pointed out probably put in a 6" slab on grade mainly to prevent vegetation growth and to give a pitch for water runoff.