Moment Frame Partial Fixity with a hard upper limit

[lexeng18]

Hi all,

I'm having trouble controlling story drift on a steel moment frame structure with a tall first story (18'-0"). I have been reviewing the multitude of other threads about partial base fixity but wanted to compile my understanding here for some additional input. Here's my specific questions/concerns:

1. The structure bears on spread footings, however even a 9'-0" square footing with the subgrade modulus of 125pci results in a fairly stiff rotational spring (~118,000 kip-ft/radian). This is enough such that in my analysis model, even with heavy moment frame columns there's too much moment being attracted to the bases such that the bearing capacity of the footing would be well exceeded.
2. Some analysis software allows the use of non-linear rotational springs that allow you to restrict the maximum spring moment. This seems like exactly what I'd like to do--this will force more of the frame stiffness up to the beam/column connections. However even this partial/limited base fixity drastically decreases my first story drift. Note you could achieve a similar effect by just using a smaller linear spring stiffness until you end up with the moment you feel comfortable designing the foundation/baseplate for.
3. In my mind this is a perfectly acceptable design approach, assuming I take the following precautions:
   • design for strength assuming fully pinned bases, only consider any base fixity for seismic/wind drift
   • Even though this is an R=3, SDC B structure--design the baseplates with the seismic provisions of ACI to ensure that if the baseplate/foundation is stiffer than I anticipate, things remain ductile and my anchorage isn't destroyed before the frames above have the opportunity to fully contribute.
   • Limit the rotational spring stiffness so something like 20% of the EI/L of the columns to stay within currently accepted documentation with this approach (I think Eurocode for example allows something like this). I think I will be well below this to get my drift under control anyways

Anyways, I'm curious if anyone has specific experience with this and can offer some insight.

 

[TRAK.Structural]

I've done a fair amount of steel moment frame design, but never considered partial fixity at the base to make them work. Are larger footings an option, such that you can satisfy the bearing pressure and keep the drift down without doing a complicated analysis with foundation springs?

 

[XR250]

I have used grade beams to connect the columns when I have soil bearing issues

 

[human909]

Most of the time the bearing capacity supplied by a geotechnical engineer is more settlement driven that total failure. I wouldn't fear exceeding it in a single event.

That said like XR250 said, grade beams are great for dealing with fixed base connection.

 

[RWW0002]

Many may get a bit squeamish in assigning a "partially fixed" joint, as it feels like a bit of hocus pocus.. but is assuming a random % fixity to meet frame drift requirements (even if we know this fixity likely does not match reality), and designing for the resulting moment in the foundation system really that much different than assuming a pinned condition (even though, again we know it is not truly pinned) and designing foundation for the resulting forces, zero moment in this case?? Or for that matter, designing as fully fixed at foundation?

I agree that it is likely wise to envelope the strength checks and ensure some additional strength and ductility at critical locations if you go this route.

 

[bones206]

The approach outlined in #3 is what I tend to do. It works out for me on a lot of projects that are drift controlled.

I do think it’s a good idea to explicitly design the anchors and baseplate for the partially fixed moment reaction, even if considering it to be a pinned connection for frame strength checks. It’s easy enough to do in the age of Profis.