moment anchorage for columns 1

[jay156]

Hi, I'm designing a new 31,000 sf building with no exterior walls. the column bays are 30'-9" oc and the roof is at 23' aff, flat, and supported by k joists and joist girders.

Wind loading isn't a problem because there are no walls, but to resist seismic (5% of DL) I was thinking of embedding the columns into their pedestals. Does that sound like it would work, and how do I determine the embedment depth?

Thanks.

 

[JedClampett]

It's going to be hard to find a method for calculating the required embedment for a steel column into a concrete foundation. Common sense tells you that if you embed the column 10 feet, it would be fixed at the base, but it's hard to prove.
I see other challenges. Concrete is done by one sub. Steel is done by another. It's difficult to even get anchor bolts placed correctly. It's going to be much more difficult to get a steel column delivered, plumbed up and supported while they're pouring concrete around it. And the interface is notorious for corroding.
Another problem is the foundation design. Ther are ways to develop a moment in a foundation, but many of them depend on passive pressures. To get to passive pressures requires a large movement.
I'd consider moment frames that are pinned at the base. You might have to replace some joists with wide flanges, but you'll have less construction problems.

 

[BAretired]

Don't ignore wind loads. Joists can attract quite a bit of wind force.

I would prefer a bolted base detail for the reasons given by Jed, but I would not necessarily design it as hinged.

BA

 

[structuresguy]

IF i really HAD to embed the columns, I would weld shear studs to the flanges, then use the shear capacity of the studs to transfer the bending moment from the column to the concrete. Then it is fairly easy to calculate how many studs and what spacing, and thus embedment length, to transfer any given moment.

But like the others, I would prefer not to embed the column. I would be looking to bolt it down to the foundation. And I would not also just go with a "hinged" design assumption, especially if using tube columns. I like to check both conditions, looking at stress levels and deflection at each. Then knowing that reality is somewhere in between, I come up with my best guess at what reality will be. I like to design the connection to the foundation as fixed, as generally, the connections will achieve a high level of fixity to the base, particularly with tube columns, where the anchors are always outboard of the column flanges.

For very lightly loaded columns, your foundations could get very large to provide adequate over turning resistance. You might consider drilled piers instead of spread footings (which would also work well if you were embeding the column). Might result in less expensive foundations. You could also look at doing grade beams, but I think your column bays are too far apart for these to be efficient.

 

[a2mfk]

You are getting some great advice that I think is spot on, so I will vote to take care of the loading with your steel frame. I live in wind only country, so I am not that up on seismic, but I have to think to satisfy code requirements it will be easiest with a braced or moment steel frame system. Your building sounds like something industrial/storage/agriculture, so maybe you can get away with some full height braces along the column lines (cheapest) depending on the client's space needs.

 

[csd72]

Look up a precast handbook as they often do cantilevered columns in pockets.

It is really a matter of checking the bearing stress with a couple of extra bars at the top to prevent bursting.

 

[OHIOMatt]

Have you considered making moment frames out of the joists and joist girders? Typically you provide a required moment of inertia and moment required at the joint.

 

[rscassar]

I would still be checking the racking forces which would result from having clad exterior walls and comparing that against the seismic forces. If you can design with no base fixity and still keep the drifts within acceptable levels, I would go with that option, otherwise design the base with partial rotational fixity.

If you are embedding the columns, I would detail a pad which you can stand the column on. Have shear studs welded to the column and then place concrete around the column for embedment. Someone did post a detail for this a few months back which I will try to did up.

 

[rscassar]

See attached

thread507-278242