R = 1 seismic systems and their load combinations

[Checkmann]

Hello all,

I have a very unique structure that I will give a bit of background on below. But what I'd like to know are your thoughts on what seismic load combinations to use when you have a steel structure designed for R = 1.0. Can I set the redundancy factor to 1.0 and thus not increase my seismic forces by 1.3.?

Some features of the structure:
-steel box girder moment framed canopy attached to a concrete building in a high seismic region (SDC D)
-designed to R=1 because it will not be possible to create box girder connections for Special Moment Frames, thus we have to design it elastically for no ductility
-it's not a redundant structure at all, one massive story (20m or so) with just a few giant columns

My thought is since we are not designing for any ductility or yielding that the seismic load combination provisions can be thrown out the window and I can use a 1.0 factor on my EQ case, no redundancy factor or overstrength factor.

Thoughts?

 

[r13]

Why not use ASD instead of LRFD? I think the entire LRFD universe is ductility based, thus the factors.

 

[phamENG]

retired - redundancy factor (Greek letter rho) is independent of ASD/LRFD.

Checkmann - not sure if you're using 7-10 or 7-16. I have 7-10 in front of me at the moment. If you meet the exceptions in 12.3.4.2, you can use 1.0 as your rho. The redundancy factor, as I understand it, doesn't look at ductility (or lack thereof). It's about how redundant the structure is. If the structure is either low enough risk or sufficiently redundant, you don't have to worry about it. If you don't have design redundancies, then you bump up the design load by 30% to introduce an added "safety factor" to the design that would be there if you had more redundancy. If anything, I'd be hesitant to use less than 1.3 for what sounds like an improbably tall single story building (66ft!) in SDC D, especially if you have "only a few columns." It sounds like your building is anything BUT redundant.

As with most of my responses to seismic questions, I'll add the caveat that I'm an East Coast guy. Wind is my bread and butter - my seismic knowledge is more of personal interest than professional necessity.

 

[r13]

Yes, phamENG is correct, either design method wouldn't let you get away with the redundancy factor and the over strength factor. However, compared to LRFD, ASD likely will lead to a design that remains elastic during seismic event (consider 0.7E load combination, and lowered design strength). (Note, for R=1, over strength factor seems does not apply)

 

[HTURKAK]

Dear Checkmann (Structural),

Refer to ASCE 7-10 or 16 Table 12.2-1 Design Coefficients and Factors for Seismic Force-Resisting Systems. OMF is not permitted for SDC D,E,F.

-steel box girder moment framed canopy attached to a concrete building in a high seismic region (SDC D)

How the canopy structure connected to the main building? What is type of curtain wall ? Glass ?

If you post the structural plan and give info. for the type of cladding , main structure conn.. you may get more meaningful comments.

 

[Checkmann]

Thanks for the responses everyone. To give you a better idea I made the simple sketch attached. I can't share anything from the drawings as it's a government building. As I tried to show, it's a 4 story concrete building with a steel trellis/canopy structure attached to it's front. The concrete building is heavy... thick slabs, big columns and beams, lots of shear walls randomly located, and it was designed for R=5 and special shear walls.

The steel trellis is architecturally governed and I can't change anything about the way it looks. There is a lightweight open steel louver-like system on its exterior, so wind load is really not an issue. But the seismic loads are high.

While it's in SDS D, I cannot use special moment frames. The connection detailing required for that is not acceptable to the client. In addition to that. there is no global precedence for an HSS to HSS moment frame in a high seismic region. It's never been done, as far as my research can tell. As ordinary moment frames are not allowed in SDS D, I am forced to go with a completely elastic design. Hence the R=1 analysis. With R=1 all seismic design criteria and detailing is obsolete. It becomes like a wind design, just design it for the loads. Problem is the loads from R=1 are massive, as the heavy concrete building is imposing a displacement on the steel beam that induces major force into the steel structure.

The steel structure is not redundant at all. There is really no leeway for load redistribution and it doesn't meet any of the checks needed for a rho of 1.0. That being said, I don't believe that I still need to apply rho=1.3 to the seismic load combos. That factor is meant to penalize you when you have a structure that needs to yield but can't redistribute any loading properly. When you design for fully elastic seismic loading though, you don't have any yielding or redistribution. And I don't see a reason the rho factor would apply. It becomes like a wind design. There is no rho or overstrength for wind load combinations. The problem is that the code does not specifically say that. In fact, R=1 is not even mentioned in ASCE7.

The plan is to use R=1 and design fully welded CJP connections. Cost is not a concern from the client so 1" CJP welds seem to be no big deal. But reducing my loading by a factor of 1.3 would be huge.

Thanks for the input.

 

[HTURKAK]

Dear Checkmann (Structural)(OP),

How are the box beams 350X600 and 300X600 beams connected to the RC bldg slabs? The box beam 350X600 implies there is composite deck at this level..what is the construction at roof level?

The sketch that you provided implies, the decks at two level and beams , rigidly connected to the RC bldg and the analysis shall be performed for the whole structure.

Pls provide more info to get valuable comments.

 

[DCBII]

Have you considered using ASCE 7-16 Chapter 15 for non-building structures? The "Detailing Requirements" section of Table 15.4-1 lets you use AISC 360 rather than AISC 341 when using an R of 1, and it allows R of 1 regardless of height and regardless of seismic design category. The redundancy factor would still apply.

 

[r13]

DCBII raise an interesting argument that you might employ. An article from Structure Magazine regarding nonbuilding structure seismic design considerations is linked for your information. Link

 

[Checkmann]

Funny you mention that. Using Chapter 15 was actually one of our first thoughts. But after researching it we concluded that non-building structures can only be used for structures that are unoccupied. Underneath this giant steel canopy is a garden area with outdoor dining, so we don't think it qualifies. That would have made life easier though....

HTURKAK: The connection of the steel structure back to the concrete building is still in consideration. There is a thick concrete curb (800mm thick, 1200mm tall) that we have to connect to at the roof level and a large concrete beam at Level 2. Initial design is a rigid connection, but I would like to get away with a pin connection if I can. We are analyzing as one structure, with R=5, and increasing the loading on the steel elements by 5.0. There is a lightweight steel deck on the roof of the canopy.

 

[HTURKAK]

Dear Checkmann (Structural)(OP);

The codes are not the rules of nature which can not be circumvented. This case is an example to circumvent the codes which are changing every several years.

IMO, if i were the designer; i will rigidly connect the beams to RC bldg at level 2, and provide composite deck at level 2, rigidly connect at roof level or provide horizontal bracing at roof level, provide spandrel tie beam ( 300X 600 box ,having torsional stiffness ) at level 3.

Regarding the seismic analysis, i would perform two tier analysis. The 1st , the columns are pin connected to beams ( assuming the columns gravity columns ) and catch the displacements. In the second analysis, assume columns rigidly connectet to beams and found. and catch the displacements and seismic load effects.

The steel elements shall be designed for gravity and keep elastic at seismic displacements.

One can say this is by-passing of the code..But the answer is , the columns are essentially gravity..