ASCE 12.2.4 vs 12.2.3.1

[sgs114]

Hello,

I have a question in regards to load application for a building with vertical combinations of lateral force resisting systems. My building is 6 stories with buckling restrained braced frames on levels 2-6 and CMU walls at the base. My question is in regards to the transfer of overturning forces at Level 2. Per ASCE 12.2.3.1: For the design of the lower system, the design coefficients for the lower system shall be used. Forces transferred from the upper system to the lower system shall be increased by multiplying by the ratio of the higher response modification coefficient to the lower response modification coefficient.

My interpretation of this is that I multiply the forces to my lateral force resisting system at the first level by the ratio of my R-values (8/5). This would basically include collector beams at the 2nd level, and shear values into the wall.

Pilasters support the columns of the BRBF at the first level. My tension and compression forces would not need to be multiplied by (8/5) because the pilasters are common to the different framing systems. The design of the pilasters is covered by ASCE 12.2.4 which states structural members common to different framing systems used to resist seismic forces in any direction shall be designed using the detailing requirements of Chapter 12 required by the highest response modification coefficient (R) of the connected framing system.

There is an example in the 2015 IBC Seismic Design Manual which has a similar circumstance to what I am describing. They use the omega for the upper system, however there is no mention of the scaling of forces based on the R-values. The fact they omit it, leads me to my line of thinking. Of course it would be ideal if they just stated the scaling of forces by the ratio of the R-values is not required, but that would be too easy.

Anyone encountered this? Anyone have a different interpretation?

Thanks,

SGS

 

[sgs114]

Bump

 

[KootK]

Rationally, I come to the same conclusion: the pilaster lateral forces do not need to be scaled up by the R value ratio. I get there by a different route though.

My tension and compression forces would not need to be multiplied by (8/5) because the pilasters are common to the different framing systems.

I think that should really be:

My tension and compression forces would not need to be multiplied by (8/5) because the pilasters are NOT common to the different LATERAL framing systems.

I believe that ASCE is describing lateral framing systems, not just any old framing system. It seems to me that the lower level lateral framing contributes nothing to the seismic axial load on the pilasters in question and that, rather, they received their entire seismic axial demand from the BRBF above.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[Hokie93]

In my opinion, the pilasters should be designed for 8/5 of the tension and compression forces originating in the upper (BRBF) system. I believe Section 12.2.3.1 is intended to cover this very scenario. If the pilasters are offset (in-plane or out-of-plane) from the BRBF then the provisions of Section 12.3.3.3 would have to be followed based on the structure having a Type 4 vertical or horizontal irregularity. In that event, the pilasters would be designed for seismic load effects including an overstrength factor and the tension/compression forces would be amplified by 8/5.

 

[sgs114]

Hokie,

What do you think the intent of Section 12.2.4 is? It mentions designing members common to the different framing systems using the requirements of the highest response modification coefficient, R. It seems to me like my pilasters would be designed using the requirements (loading) from based on my R=8.

 

[TehMightyEngineer]

My understanding is that ASCE 7 section 12.2.4 is to address things like columns at a building corner that are utilized in two different LFRS systems. Like say you have a corner column that connects to a BRBF on one side and a SCBF on the other. That gives you an R = 8 for the BRBF but an R = 6 for the SCBF. Per 12.2.4 the column needs to be designed for R = 8.

In your project, the pilaster is not common to the BRBF. The pilaster does not directly contribute to the LFRS at the level of the BRBF and thus is not common; merely supporting.

As you noted the SEAOC Vol 1 Seismic Design Manual covers this in example 7. My understanding was that they ignore 12.2.3.1 as they use the two-stage analysis procedure of 12.2.3.2. This essentially breaks the design of the structures into two separate structures. They do increase the forces on the lower structure though per 12.2.3.2.d which requires amplification of the reactions from the upper structure by R/ρ. However, reading ASCE 7 section 12.2.3 it appears they give no exception to 12.2.3.1 when you use the two-stage analysis procedure. I was under the impression that ASCE intended 12.2.3.2 was to be in-lieu of 12.2.3.1 but that doesn't appear to be how it's written. The ASCE commentary sheds no light on this either.

Annoyingly SEAOC's Volume 1 commentary is lacking as they only point you to the Blue Book article 4.02.040 "Combined Systems". [URL unfurl="true"]https://seaoc.site-ym.com/store/ViewProduct.aspx?ID=11238159[/url]. Perhaps the answer is in there.



Ian Riley, PE, SE
Professional Engineer (ME, NH, MA) Structural Engineer (IL)
American Concrete Industries

https://www.facebook.com/AmericanConcrete/

 

[Hokie93]

My understanding is that Section 12.2.4 is applicable to a member that is part of two different LFRS, such as a column that is part of a braced frame in one direction and a moment frame in the other direction. So I don't think Section 12.2.4 is applicable per se for a vertical combination of framing systems. In any event, Section 12.2.4 is concerned with detailing of the members in question. The design forces are determined elsewhere.

 

[KootK]

My belief is that 12.2.3.1 is intended to apply to forces transferred from a high ductility lateral system above to a low ductility lateral system below. Hence the logic in the force amplification. So, here, that would go BRBF Frames--> Transfer Diaphragm --> CMU walls. That chain bypasses the lower level piers in questions which is why I see no rational basis for a force amplification.

I also feel that this condition could be viewed analogously in two additional ways, neither requiring force amplification to my knowledge:

1) Discontinuous CIP shear wall over CIP columns.

2) The piers could be viewed as simply the beginnings of the foundation for the upper level OT forces.

It seems to me that this is a condition that probably wasn't explicitly considered in the development of the ASCE provisions. And that's okay, we're supposed to be thinkers. For both conventional analysis and two stage, I would support amplifying any frame axial forces transferred laterally through the upper level of the base structure. For cases like this where the forces just pass through, however, I see no logical basis for it. No matter how badly the base structure behaves seismically, I'd think the inputs to the piers in questions would still be 100% a function of the BRBF above.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[sgs114]

Hokie

So even if my pilasters are not part of the lateral system on the base level you still think it is required to amplify the forces by my ratio of R-values? It seems odd that I suddenly have to design my base level columns for loads that are 60% larger than the loading on the floor above. Especially since the load could not even be delivered to the system based on the capacity of the steel columns above. Do you think I could justify designing my concrete pilasters on the base level to the nominal capacity of the columns above plus any loading the columns experience as a result of the loading from the 2nd level (gravity loading only, no additional seismic load)?

SGS

 

[KootK]

So even if my pilasters are not part of the lateral system on the base level you still think it is required to amplify the forces by my ratio of R-values?

Hopefully it is clear that my answer to this question is no.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[TehMightyEngineer]

My belief is that 12.2.3.1 is intended to apply to forces transferred from a high ductility lateral system above to a low ductility lateral system below. Hence the logic in the force amplification.

I agree with KootK, the quote above matches my understanding.

In these discussions it is important to remember what the R-value actually does. Our inelastic forces in the structure are much greater than the elastic design forces we use; the R value helps us bridge this gap allowing as to design structures for elastic forces using normal design methods while still considering the inelastic force distribution and energy dissipation of a seismic system.

So, the BRBF has much higher inelastic energy than we use in our models that is dissipated through yielding of the BRBF elements. If we provide an alternative path for that energy to go we must ensure that such a path remains elastic or otherwise can sustain those forces while the BRBF does it's thing during a seismic event.

The pilasters in the lower LFRS are bypassed via whatever transfer mechanism (transfer slab, etc.) you use to get the forces from the BRBF to the CMU shear wall on the first story. The pilasters need to accommodate the reactions from the BRBF but as they're not intended to add to the inelastic force dissipation of a seismic event they shouldn't care what the R value is above them.

Ian Riley, PE, SE
Professional Engineer (ME, NH, MA) Structural Engineer (IL)
American Concrete Industries

https://www.facebook.com/AmericanConcrete/

 

[Hokie93]

SGS - I think your suggested approach is reasonable and it does feel like overkill to design elements not part of the LFRS for the amplified force of Section 12.2.3.1. To me, the detailing of the pilasters is important. That is, I would give strong consideration to detailing the pilasters in a manner that is consistent with an R=8 system since they are providing structural support for a buckling-restrained braced frame system. As KootK noted, this framing arrangement is not directly covered by ASCE 7 and, so, you need to draw on your engineering judgment and 'structural IQ' to arrive at a safe (not overkill) system.

Do you have a copy of the "Guide to the Seismic Load Provisions of ASCE 7-10" document published by ASCE? There is some discussion of this topic starting on page 45. Unfortunately the discussion is not particularly conclusive.

 

[KootK]

For both conventional analysis and two stage, I would support amplifying any frame axial forces transferred laterally through the upper level of the base structure

I retract this unless the transferred axial loads somehow wind up on the lower level LFRS. Don't know what I was thinking there as it's contrary to what I've been pitching elsewhere.

That is, I would give strong consideration to detailing the pilasters in a manner that is consistent with an R=8 system since they are providing structural support for a buckling-restrained braced frame system.

I agree with this 100%. I think that the perfect detailing treatment for the piers would be to treat them as you would boundary elements in a shear wall. Gobs of over strength axial but little potential for bending, assuming that CMU shear wall system is nice and stiff.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[Toby43]

I'm not well versed in ASCE7, but to quote the SEAOC Bluebook Article 4.02.040 "ASCE 7-02 section 9.5.2.2.2.1, ASCE7-05 section 12.2.3.1, and 1997 UBC section 1630.4.2 address this situation by limiting the R value of the lower system so that it does not exceed the R of the upper system. The intent is to delay the onset of yielding in the lower system until the point at which the entire structure will yield together". it goes on.. "In general, an uneven deformation distribution violates the premise of the equivalent lateral force method used for most code-based design."
The Bluebook also points to the case of the Olive View Hospital that had shearwalls over Braced frames and almost completely collapsed.
So I believe that the approaches recommended above regarding applying overstrength to pilaster axial loads is perfectly rational an I would take a similar approach (for what it's worth).

Toby