PEMB Continuous Perimeter Footing Between Columns

[txa129]

I have a question about the necessity/function of continuous footings on metal buildings. I have seen this detailed a number of different ways. The way I see it most commonly done is a +/- 2ft wide * +/- 1ft thick continuous footing at the same elevation as the column footings with vertical dowels in to a "turndown" which is poured monolithically with the SOG (Type 1). I have also seen many applications where it was only the turndown to frost depth with no separate footing below (Type 2). In a typical PEMB, the gravity and lateral loads carry to the frames/columns and do not apply a linear load at the foundation level. The question is essentially, in Type 1 what does the bottom continuous footing actually do? In a design where the designer is using the slab to resist lateral loads via haripins and the column footings are designed for vertical loads only, depending on the frame reactions, the slab alone is often sufficient to transfer the loads in the frame direction without the need for actual tie beams between frame columns (let's not get into SOG vs structural slab for lateral loads). So, in the longitudinal (braced frame/portal frame) direction, are we saying the slab is not sufficient regardless of reaction level, and the continuous footing supposed to help stabilize the column footing? If that is the case, why would these footings need to cover the entire perimeter including the endwalls and not be isolated to braced frames. Certainly, in a design using moment resisting foundations, the need for the separate footing under the turndown goes away. But if this turndown is essentially a 12x18 GB that is adequately reinforced and is poured/tied continuous with the slab and "pedestal" portion of the column footing - does it not provide some longitudinal capacity itself? Perhaps it is a case of a detail that works when there is masonry, concrete, or tilt walls which impart a gravity load to the slab, and not just metal panel and so it becomes the "standard" detail that gets used for everything. Maybe the broader question is - what is the accepted way of resolving longitudinal column reactions to ground in a foundation design that uses the slab to resist all lateral loads.

 

[driftLimiter]

In a real scenario you also need to consider the vertical forces at the x-braced bays or portal framed bays. Unless you provide a lot of concrete in an isolated footing most likely you will have uplift or stability problems at the footing. A continuous reinforced grade beam can help distribute this overturning over more soil.

For the in plane shear of the frames, its possible to take it into the slab, but the grade beam is a stiff load path that is easy to justify.

Finally, frost depth. I provide a frost depth footing all around the perimeter of buildings so that the interior foundations don't necessary need to go down so deep.

I don't know where you practice but in my region this is a standard practice.

 

[txa129]

Driftlimiter - thanks for the reply. I should have clarified but it was my intent that it was understood when I said footing designed to resist vertical loads that included both gravity and uplift forces including applied uplift forces as well as resultant uplift forces from overturning loads in both lateral (rigid frame) and longitudinal (braced frame) directions. If you are saying the cont footing should be used to resist some amount of uplift, I would argue it is more cost effective and constrict able to increase resisting mass via concrete or soil at the column location.

For frost, I practice in several regions but mostly practice in the SE US and frost depth varies up to 24" but averages around 12-18". In those cases I have often seen just the turndown/trench footing poured with the slab without a separate discrete footing below. But again I see it with the footing as well and I am questioning what the function of that is. certainly a different story up north.

Thanks again for your response

 

[canwesteng]

It's not uncommon here where frost depth is deeper to only provide a grade beam on void form along the perimeter of the building, no perimeter footing. Where frost depth is shallow this may as well be supported on the soil though. I'd expect you'd use a thickened edge slab if there is no frost depth at all.

 

[wth]

For a PEMB in my region you often get net uplift on foundations (single columns) for stability combinations, add the vertical component of a braced frame and you'll need ridiculously large single footings to resist uplift and gliding. By combining both columns in a frame you're "removing" the vertical load from the braced frame, but still need to handle the overturning moment.

Frost depth is ~2 m and more around here, so we almost always insulate the perimeter with XPS with shallow foundations.