ETABS- Modeling Columns landing On Shear Wall

[StarkInc14]

HELLO GUYS.

IM CURRENTLY MODELING A CONCRETE BUILDING AND I HAVE A QUESTION WITH REGARDS TO WHAT WOULD BE THE CORRECT MODELING TECHNIQUE. I FIRST WOULD LIKE TO GIVE SOME BACKGROUND INFORMATION ON MY CASE.

MY CONCRETE BUILDING CONSISTS OF AN 18 STORY BUILDING. DUE TO SOME ARCHITECTURAL CONSTRAINT I CANNOT KEEP A SHEAR WALL THAT STARTS FROM THE CELLAR CONTINUOUS TO THE ROOF. THE SHEAR WALL ENDS AT THE 10TH FLOOR AND THE ABOVE FLOORS ARE SUPPORTED BY 2 COLUMNS THAT PROTRUDE FROM THE SHEAR WALL BELOW AND CONTINUE TO THE ROOF. THE COLUMNS THAT CONTINUE TO THE ROOF ARE THE SAME THICKNESS AS THAT OF THE SHEAR WALL BELOW (18 INCHES THICK).

MY QUESTIONS ARE THE FOLLOWING: WHAT IS THE CORRECT WAY I MODEL THIS SITUATION. SHOULD I CONTINUE TO MODEL THE COLUMNS (THAT TECHNICALLY END AT THE 10TH FLOOR)ALL THE WAY TO THE CELLAR AS A COLUMN MEMBER, AND DRAW THE SHEAR WALL IN BETWEEN THE COLUMNS FROM CELLAR TO THE 10TH FLOOR (OPTION 1)? OR, SHOULD I KEEP THE "COLUMN MEMBERS" FROM ROOF TO 10TH FLOOR, AND THEN BELOW THE 10TH FLOOR JUST DRAW THE SHEAR WALL ONLY BUT MAKING SURE THAT THE WALL IS MANUALLY MESHED WHERE THE COLUMN NODE IS LOCATED ABOVE (OPTION 2)? SINCE BOTH THE COLUMN AND SHEAR WALL ARE OF SAME THICKNESS, THE LOAD FROM COLUMN ESSENTIALLY DISTRIBUTE TO THE WALL I JUST DONT KNOW IF MODELING ONE WAY IS MORE CORRECT THAN THE OTHER. I HAVE ATTACHED AN IMAGE TO HELP VISUALIZE MY SITUATION.

YOUR HELP IS ALWAYS APPRECIATED.

 

[Trenno]

Personally I'd model the columns as frame elements, as per your Option 2.

The load from the columns above L10 will fan out and disperse into the wall below. You'll need to ensure the wall and frame member are meshed correctly.

One thing to check would be the bearing stress of the column onto the wall.

Alternatively, you could model this all as appropriately meshed shell elements. I doubt this will change your answer significantly though.

PS - kudos on the sketches, makes it so much easier to answer people's questions!

 

[rscassar]

Important to keep nodes aligned, if column was transitioning to a wall, I would have node going on column and only wall with shell elements. You can assigned shells as piers but would need to do some localized calculations for dispersion of column load (go to frame forces at station 0) into the wall. Shells I normally like to mesh at approximately 1.0m squares.

 

[PROFR]

Option 2 is the correct one. The columns above the shear wall can be modeled using frame elements. They can be also modeled as small wall elements.

 

[Agent666]

If you do go with option 1 then make sure you reduce the axial stiffness modifier of the columns in the wall section so it's consist with the f22 vertical stiffness modifier for the shell elements.... If I had a dollar for each time I've seen people not doing this ...

 

[JoshPlumSE]

Two thoughts on this.
1) Pure modeling, I would go with option two. Terminate the columns at the wall and let the load spread out. Though verify that the wall mesh is picking up the column points and sub-meshing properly at those locations.
2) From a construction / detailing standpoint, I'd continue those columns down. Essentially those columns become the confined boundary element for the wall.

 

[StarkInc14]

Thank you all for your inputs.

JoshPlumSE, to follow up on your thought #1. I assume that the bottom node of the column landing on the shear wall would need to be released such that the shear wall does not take out of plane bending. Am i correct?

Thanks

 

[JoshPlumSE]

StarkInc14 -

It's a concrete column? If so, I'm not sure it's necessary to introduce the pin for out of plane bending. I don't think it would transfer much force in that direction. But, it could transfer a little. If so, that's a detailing issue to make sure that it can get into the wall.... One of the reasons why I'd continue the column down a ways and use it as a boundary element or such.