Tie beams too long between foundation pads, any good? 4

[IJR]

This one bugs me.

Seismic codes force us to connect column footings by tie beams (you guys may have other names) using some % of footing load as strut force for the tie below grade.

NOW

In metal structures, you have closely spaced footings in the longitudinal direction and very large spacing in the other direction since you are crossing the span with a frame with no intermediat column.

How effective will a tie beam between footings at the other ends of the span be, given all that distance say 70ft?

Regards

IJR

 

[ihs]

I personally have never designed tie beams connecting the footings in the plane of frame as obviously spans are so large.

If the foundation depth is not much, you should think of connecting the pedestal top and ground slab with the 'U' pin. That will also resist the lateral force transferred to footings at the base plate level.

 

[IJR]

Thanks ihs.

Usually except for very large span industrial buildings or warehouses we can handle the lateral force even without utilizing slabs.

So the problem is not statical, but code based. We are told in my area to connect footings to one another by tie beams treating them as struts capable of axial force 30% of the axial force in a pedestal

No commentary to clarify this one and we usually have to argue in circles.

Respectfully
ijr

 

[Boog89]

We prefer not to use the U-bar approach. Assuming that there is a floor slab, it becomes nothing more than a diaphragm. We usually run rebar across the span in a 'thickened' portion of the slab. Generally, the columns are placed at the top of the pedestals and therefore, we utilize hooked rods out of the pedestal in order to provide the tensile splice. Thus, we create a chord member across the span that takes the column thrust.

 

[Gourile]

These kind of tie bars work to resist lateral base reaction of the columns ignoring the passive resistance of the ground. So it makes sence if you add up all the lateral reactions of the columns in a structure and design for as many tension resisting tie beams as needed for it. In this way you may decrease number of tie beams, and probably use them for every other foundation. You may even connect every three foundation by one tie beam using a design force equal to 30% of the total compressive force of three columns pertaining to them, to satisfy your code and decrease the number of beams as well.

 

[IJR]

Thanks Gourile

But my question also remains in that the beams are usually too long and require a large area excavation, because spacing between them is usually about 18ft while their length will be about 60-80ft in a typical warehouse steel frame construction. Code says do it without comments and sometimes you see this was probably meant for reinforced concrete buildings, and clients say "dont put me into unnecessary excavation-your 1ft*1ft beam what good is it huh! ". You know the feeling when you can not defend your design by any rationale

Respects
IJR

 

[JAE]

IJR,
In my area, the UBC 97 (for example) seems to only require lateral interconnection for piles and caissons for any building subject to any seimic forces. (see UBC Section 1807.2) For footings, I don't see any direct reference to interconnections so if you have spread footings, you needn't concern yourself.

Now with the piles, I believe the concern is that in a seismic event, much of the ground around the piles will become quite unstable, eliminating some or all lateral support on the piles. Thus, the piles need some measure of lateral restraint to avoid a sidesway dislocation of the pile or pile group. This would be disasterous for a building.

For your "long" distances across your building, the ties aren't intended to provide a tight restraint against lateral deflection but only a tie providing enough strength to avoid a pile/column buckling. The "stretch" of the tie may allow much larger deflections in the seismic event than you would normally want to restrict under other loads.

 

[IJR]

Thanks all of you guys who have joined this discussion and lots to you JAE for making me understand the rationale behind tying caps to one another.

We seem to agree that tying footings to one another is not an established strict requirement.

Respects
IJR