Braced RC Frame Model - Should I Pin the Columns?

[LH1993_01]

Hi - First time posting.

I am modelling a multi-storey building in reinforced concrete using SCIA Engineer 21.

The frame is 20 storeys and around 50m x 17m on plan. The construction is flat slab with blade columns typically 300mm x 10000mm and is stabilised by a series of shear walls and an offset core.

The question I have is, if I am considering my building columns to be braced (ie. not contributing to the lateral stiffness of the building) should they be released at the slab to column interface in my model?

Releasing them would ensure that all lateral load (wind, this is in the UK not seismic) is distributed into my wall elements thus giving me worst case stresses in the walls as well as overall displacement of the buildings. But this would not allow moments in the gravity load case to be transferred into the columns.

What is best practice for this? Is it two models? one to assess the building displacement and one for the actual design?

Thanks....

 

[KootK]

It's two models for me, particularly in a low seismic zone like the UK where I wouldn't be too concerned about drift compatibility in the slab to column joints.

Some folks will model the columns pinned for the slab gravity design under the premise that will exacerbate slab moments and deflections and, therefore, be conservative. It will be conservative for those things in many cases however:

1) It will often be unconservative for punching shear.

2) It tends to push the inflection points away from the columns which makes for an inelegant rebar layout in my opinion.

 

[LH1993_01]

Thanks KootK - always nice to have a sense check and a second opinion.

You raise good points re. the gravity design of the slab too. I always look at punching separately anyway - I find FEM results are generally a bit sketchy.

I have released the columns and the increase in top floor displacement is 80% more. Suppose this is less of an issue if your columns are small and square.. my "wallumns" were clearly contributing quite a bit.

 

[KootK]

Well, whether or not you should be omitting the wallumns from your lateral model if they make that much of a difference is a separate question. Certainly, in a high seismic design, I'd be paying some attention to that.

 

[LH1993_01]

KootK

Displacement is still well within limit with the wallumns ignored. Wallumns will be fixed for moments arising from lateral forces for the actual column design. No seismic design criteria.

 

[KootK]

The dominant issue is neither drift nor the column design I think but, rather, whether or not you initiate punching shear failures at your wallum ends when they attempt to participate laterally.

 

[LH1993_01]

KootK as mentioned - framing moments from lateral loads in the "fixed-fixed" model will be considered in the punching shear design. I generally do this check by hand.

 

[KootK]

KootK as mentioned - framing moments from lateral loads in the "fixed-fixed" model will be considered in the punching shear design.

I don't believe that you did mention that explicitly. I took your earlier comments to mean that you checked gravity load cases for punching shear assuming restraint at the wallumn to slab joints.