Check moment frame in Dual system 2

[HuynhKeshley]

For lateral design, the structure is composed of a dual system with moment frames with special reinforced concrete shear walls. The moment frames must be capable of resisting at least 25% of prescribed seismic forces. In ETABS I should model my structure without the shearwalls and apply only 25% of the lateral load in each floor. However, some of my beams are supported at their ends by the same shearwalls I am removing. What should I do in this situation?
Could you please advise for 2 cases:
Case 1: Beams are fixed to shear wall in the combined model
Case 2: Beams are pinned to shear wall in the combined model

Thanks.

 

[r13]

I would keep the shear walls in the model and reduce its stiffness. You can shorten the walls until they account for no more than 75% of the prescribed seismic force.

 

[HuynhKeshley]

Thank you Retired13.

However, the distribution of story shear to the moment frame at each level are not the same.
So, do you modify the stiffness of the shear wall at each level separately?

 

[r13]

I would think so. The idea is to adjust the stiffness of shear walls in each story to meet the 75% limitation, and design the concrete frames for the remaining (25%). Then, when rendering project drawings, you can turn the shear walls back to the desirable dimensions, so in reality, you now have a duel system with a combined strength somewhat greater than the prescribed seismic force.

 

[HTURKAK]

Both of the approaches are not reasonable..
The dual system requires that the moment frame portion of the system SHALL be designed to resist at least 25 percent of the total seismic force.

This reequires that a separate analysis of a frame-only system SHALL be carried out with the panels of the structural walls were removed but the boundary elements of the walls will be kept so that the behavior of connected beams to walls could be analyzed in a reasonable way...

 

[JLNJ]

Maybe you can factor up the forces until you get the 25% of the original total shear in the moment frames.

I would be careful about modifying stiffnesses due to the fact that you could change the entire behavior of the structure.

 

[KootK]

However, some of my beams are supported at their ends by the same shearwalls I am removing. What should I do in this situation?

I might replace the lost wall supports at the ends of the effected beams with dummy, gravity only columns. Keeping with the spirit of the dual system methodology, I don't think that you'd want to rely upon a beam / wall joint for lateral resistance anyhow.

 

[HuynhKeshley]

Dear HTURKAK, Dear KootK,

I have tried, but found another issue.
The building includes tower part and podium part.
At podium part (above ground level), there are a lot of gravity columns.
I have seen that the story shear also distributes to the gravity columns as well.
At tower part, where there are few dummy gravity columns, the distribution of story shear to these dummy gravity columns is small.
However, at podium part, the the distribution of story shear to real gravity columns is huge. Consequently, the distribution of story shear to the columns of moment resisting frame is reduced, and shall be less than 25% of the original story shear.
I have tried to reduce the stiffness modifier of the gravity columns, or used small columns section, but it results in non-convergence analysis.

Please advise.

Thanks.

 

[HuynhKeshley]

Maybe you can factor up the forces until you get the 25% of the original total shear in the moment frames.

I think that the portion of story shear to the moment frame at the every floor levels are not the same, so using one scale factor will lead to over-design at some floor levels.

 

[r13]

I have tried to reduce the stiffness modifier of the gravity columns, or used small columns section, but it results in non-convergence analysis.

Have you tried to upsize the moment resisting frame columns?

 

[HTURKAK]

Dear HuynhKeshley (Structural)(OP);

The use of dual system is a preference rather than a requirement.. That is , you are free to use SRCW with R=6.

For example ,by visual inspection, the frame system could resist say 10 % of total base shear and you may still utilize dual system providing the moment frames themselves are designed to resist at least 25 percent of the total base shear.
The requirement at , Standard Section 12.2.5.1 is to ensure that the dual system is capable and has enough redundancy to justify the increase from R = 6 ( for SRCW ) to R = 7 ( for a dual system SRCW+ SMRF )..(I took the R values from Standard Table 12.2-1).

I want to define the procedure (for MRSA ) to my IMHO,

- Run the analysis for combined system and get total Vbase,
- Run a separate analysis for frame only system with global scaling the RS with 0.25 * Vbase. The global scaling may result in story forces not exactly equal to 25 percent of the story forces from the MRSA of the dual system ,since the frame only model response could be different..
- The frame system SHALL BE designed and detailed to resist AT LEAST justify the frame only system .In general , the 25 percent rule controls only at the lower levels of the building (in your case podium levels).That is, the design of the beams and columns at the lower levels CAN be based on 25 % frame only analysis.

Regarding the modelling of frame only system you have two options,

i= Remove the panels of the structural walls but keep the boundary elements of the walls so that the behavior of connected beams to walls could be analyzed...

ii= Remove the walls completely and provide rigid beams (span will be width of wall ) at walls and nominally pinned columns at both sides of walls.

Regarding gravity columns, i will recommend to keep them as a part of MRF. I do not know your understanding of gravity column. If you want some columns to resist gravity only loads, run the analysis with providing hinges at both ends of these columns..But , still you MUST detail these columns for drift requirements and still follow the detailing requirements of building LFRS.

I hope my respond makes sense.

 

[r13]

Run a separate analysis for frame only system with global scaling the RS with 0.25 * Vbase.

That's exactly in my head last night. From this analysis, you can weed out some hidden inefficiencies, and improve the moment framing system.