Adding in a mezzanine

[ARKeng]

I have a SDC C all-steel 3-story building. The building is an OMF building for the top two stories and OCBF building at the bottom story.

The owner wants to add a mezzanine between the foundation and the first level for part of the building plan (not full). He's very adamant against adding in additional columns or braced frames to make this mezzanine a self-supporting separate structure. We did anticipate that the columns and footings needed to be designed for some extra load for a mezzanine situation so we have some extra capacity, but we'd been expecting to add in braces where needed.

Due to this situation, my boss has asked me to investigate a lateral system where the seismic loads from the mezzanine are applied as point loads at the mezzanine elevation, and the columns span this load between the foundation and the second floor. The loads would then be distributed through the second floor diaphragm to the overall existing building braced frames. He wants us to do this to the extent that he wants us to investigate if we need to reinforce the existing columns. This is of course an iterative nightmare because if I repair a column, the stiffness will change and we'll have to redistribute seismic forces. And of course, the building plan is irregular, so I have different column shapes, sizes, and even orientations to worry about.

I have some serious questions/doubts:
1. Is this system really allowed by code? I don't know that it's specifically prohibited, but I'm having trouble classifying the
"system".
2. What kind of system is it? Do I need to adjust my seismic "R" factor?
3. Any other ideas/questions/advice I need to consider? Anyone done this before?

 

[SethGuthrie]

I work in technical support at Bentley Systems and we see these kinds of models all the time (in STAAD.pro, RAM Structural System, etc.). I can't say for sure what the best code interpretation would be, but I think most users are applying the loads as you described and using the R factor and building principal period (T) based on the primary building and lateral system. Using an R factor of 3 is conservative, but not really necessary where the mezzanine is small relative to the main structure.

ASCE 7-05 Section 12.1.3 gives some information on “smaller portions of the structure tied to the remainder…”.
Furthermore Section 12.2.3.1 lists a few exceptions for the rule about using the smaller R factor, those are:
1. Rooftop structures…, [and]
2. Other supported structural systems with a weight equal to or less than 10 percent of the weight of the structure.
3. Detached ...

 

[ARKeng]

Thanks Seth. You've actually raised another question since I already have a RAM model.

If I add this mezzanine in as a another floor and set the columns I want to use as "lateral" columns with the tops pinned at level 2 and bottoms pinned at foundation and all the Level 1A mezzanine beams as "gravity" beams, will RAM distribute this correctly based on the rigidities of the columns? Or will it just give an error since there's no frame of any kind? I'm just not sure if RAM will properly redistribute part of this load back up into the level 2 diaphragm?

 

[SethGuthrie]

Good question. So long as there is at least one lateral column or wall connected to the mezzanine diaphragm, the program will still see it as a valid diaphragm and apply lateral loads to it. The lateral columns would resist those forces through local shear/bending even though there are no frame beams or frame action involved. Some column shear would go down to the base directly and potentially some would go up into another diaphragm and over to some vertical frame. That distribution is all a matter of stiffness.

Note, these types of structures tend to be very sensitive to P-Delta amplification and I usually recommend no P-Delta analysis, at least in the preliminary design.