Braced and Moment frames in office buildings 4

[snare281]

Hello,

I am a senior in Structural Engineering (NCSU), we are designing a 5 story office building for our senior project and I am having touble finding much reference material in frame design. Simply information on bracing, bracing placement, and moment frame placement.

With our design we have to include a report w/ several references about the logic that went into our design and examples of other buildings. I feel somewhat prepared to put a design together, but as of now the only reference that I would have to show for it would be trial and error and my own creativity.


Any ideas.. or any Wisdom. Websites or general knowledge would be great! Thanks so much for all your help!


p.s. Anyone looking to hire in January of 2003?

jkennedy12@nc.rr.com

 

[trainguy]

I remember feeling exactly the same way in my first 6 months as a practicing structural engineer.

Tip 1:
In practice, we choose which bays to brace mostly by avoiding windows, and arguing with the architect. So, braced frames essentially go where architectural constraints permit.

Tip 2:
You can't mix moment frames and braced frames in the same building direction, or your moment frames will never "see" any load.

Tip 3:
Try to place your frames as mush as possible on the perimeter of the building, this way they're effective in counteracting torsion, which will be imposed by your building code even for a regularly shaped building.

Tip 4:
Try to avoid braced bays at corner columns because your dead load is relatively small, thereby giving you possible uplift at the column base, which requires special consideration.


I could continue, but no time...

 

[Taro]

trainguy,

Good tips. However, I would add a qualifying remark for #2 if earthquake resistance is a concern. Using moment frames and braced frames in the same building can actually be quite beneficial for seismic design. Although the moment frames might not "see" much load based on the first order elastic analysis, in reality they will pick up quite a bit of load after the braced frames start buckling and degrading. This "belt and suspenders" system adds excellent ductility and energy dissipation. In U.S. codes, if the moment frames are designed for at least 25% of the base shear, the structure qualifies as a dual system and may be designed for lower forces.

 

[trainguy]

Never done it that way, but it sounds like a great idea.

 

[whyun]

nice one Taro!

more specifically, this 25% base shear into the moment frame in a dual system can be found in section 1629.6.5 of the 1997 UBC.

Chapter 22 Division IV has the requirements for various types of braced frames ordinary, special, concentric, eccentric, chevron etc. Also has requirements for moment frames.

Another source for steel seismic design is the "yellow book" called Seismic Provisions for Structural Steel Buildings published by AISC dated April 1997. There had been at least a couple of supplements. All methods presented here are based on LRFD methods.

My guess is NCSU stands for North Carolina State university(?). My guess is that your state uses the BOCA code (unless it adopted IBC200 already). I apologize for the UBC references but thats the code we are used to here in the west coast.


Good luck in your project.

 

[BantrelStructural]

A further comment to Tip #2

While a moment frame and braced frame in series (i.e. the same column line) probably won't work, in parallel they can. While the differential stiffness may make such an arrangement undesirable, particularly for seismic reasons, sometimes architectural and or industrial layouts do not lend themselves to ideal "bracing" system layouts, and you have to work with what you can use, even when less than ideal. If two parallel bracing lines do not match in stiffness, as long as the centre of reaction matches the centre of load application there should be no torsional loads introduced into the structure. Any inbalance will be handled by a coupling effect of the cross direction bracing system.

Recently I was involved in a project where we actually did mix bracing and moment frames in the same column lines. The structure (tall and skinny) was assembled from a series of modules assembled a thousand miles away from the site. For transportation and erection reasons these modules were assembled as moment frames in both directions (i.e. N-S & E-W). Once erected we were able to install some bracing between certain levels in the longitudinal direction (weaker Y-axis moment frames). This bracing was introduced primarily to provide additional deflection stiffness, but because this bracing was discontinuous, we still required moment frames for the overall structural stability.

The stick diagram below does not fully represent the actual framing, but the general concept. The structure consisted of 8 modules stacked in the following sequence: field erected columns to about 10 ft., then a two level piping module, a single level process equipment module, piping module, process module, piping, process, piping & process. The structure while 52 ft. long, was ONLY 14 ft. wide due to both transportation limitations and plot space limitations within the refinery!

|--------|----|--------|
| |\ /| |
| | \/ | |
|--------|----|--------|
| | /\ | |
| |/ \| |
|--------|----|--------|
| | | |
| | | |
|--------|----|--------|
| |\ /| |
| | \/ | |
|--------|----|--------|
| | /\ | |
| |/ \| |
|--------|----|--------|
| | | |
| | | |
| | | |
---------------------------

While the structure was designed for a seismic zone, we were not able to take advantage of the higher R values that we could have achieved with our ductile moment frames, because we could not tolerate the high lateral sway deflections necessary to mobilize this ductility.

Permissible sway deflections on our tall (88 ft. high) structure could have been as high as 21" under seismic (NBC Canada) code requirements, although while not rated under the Canadian code as a post-disaster structure, as a petrochemical facility similar structures designed under UBC/IBC codes would be rated as high risk, which would have a sway limit of around 10-11". We would have limited ourselves to about 10". Our first pass on our structural model produced deflections in the order of 12" to 14". Piping limitations though restricted us to between 3" & 4", and as a result we ended up with a much stiffer structure that had to be designed for much larger loads, due to the low R value that we had to adopt (our connections were detailed as nominally ductile, but we could never achieve the deflections to actually mobilize this ductility - essentially our structure remains in the elatic region under our design earthquake).