Shear Wall Highly Stress at Corner

[seb3k]

For the original poster I can recommend to read up on a very good book we have in download section - "Friedel Hartmann, Casimir Katz - Structural analysis with finite elements"
One of them is an author of Sofistik, and explains alot of things FEM can and cannot do correctly - including what zippaton and myself told you already

[parhyang]

Here's quotes from Ed. Wilson,

Many engineers believe that the creation of a two-dimensional finite element mesh, as shown in Figure 7.7b (above), is the best approach to evaluate the displacements and stresses within the shear wall. In the author’s opinion, this approach may not be the best for the following reasons:
1. As previously illustrated, the use of four-node plane elements for frame analysis does not accurately model linear bending. The approximation of constant shear stress within each element makes it very difficult to capture the parabolic shear distribution that exists in the classical frame element.
2. If a very fine mesh is used, the linear finite element solution will produce near infinite stresses at the corners of the openings. Because the basic philosophy of reinforced concrete design is based on cracked sections, it is not possible to use the finite element results directly for design.
3. Using common sense and a physical insight into the behavior of the structure, it is possible to use frame elements to create a very simple model that accurately captures the behavior of the structure and directly produces results that can be used to design the concrete elements.

[allagui10]

Hi,
I purpose using additional stiffeners along the shear wall it reduce enormously stresses at corner( downstream ). these stiffeners will be placed to redistribute forces along the shear wall.

[techno]

parhyang:

what you mean is not necessary to consider the very corner stress?

allagui10 : i agree on what you suggested. just assign pier to it and use the pier axial load + bending to design that particular strip.

[jaks]

As far a transfer beam/slab is concerened we need to remember that it is providing an elastic support to the column from above levels. The beam/slab is under stress and hence deflecting due to the loads from the column. Due to this deflecting behavior there are high stresses at the corners (also can be related to the punching effect). I did a transfer slab, with the same problems of high stresses, did provided sufficient reinforcement and stiffness to counteract the same.

Also I would like to echo the comment made by Zippaton, that instead of using a point load analyse the transfer beam/slab for a line load, which will infact be the actual condition, the results will be quite different.

The high stresses need to be addressed as suggested by allagui10. If allagui10 can provide us with a sketch as to how the stiffners can be provided and the junction detailed.

I have my doubts regarding taking the average of the stresses.

Jaks.

[chrono146]

Hello,

Here is a file for reference:

Code:

***************************************
Content of this section is hidden, You must be registered and activate your account to see this content. See this link to read how you can remove this limitation:

http://forum.civilea.com/thread-27464.html
***************************************

If you have axial stress concentration, provide additional reinforcement. Also notice that FEM can give very high values that aren't completely true, since mathematical model won't reflect distribution stress areas (model is connecting elements with nodes).

Regards,

[allagui10]

Hi,
using
the linear stress method
sigma(max)=N/S +(M*v/I) .
sigma(min)=N/S -(M*v/I) .
I :inertia of the wall
S:area
M:moment
v:dist between GC to left boundary
by increasing area and inertia of the shear wall / additional webs (stiffeners)/ between boundary zones (upstream to downstream) the stress will be reduced.
the positions of these webs depends geometric arrangement of the structure...
ACI 318-02 gives another method

[parhyang]

techno wrote: For example, in a 500x500mm wall element, i found that the value at corner is 3599KN/m but at the middle of the element, it's actually just 2000KN/m.

techno wrote:what you mean is not necessary to consider the very corner stress?

because reinforced concrete is not homogeneous, the materials is high nonlinearity, cracked section so an elastic FE is not applicable. it's mean you ca not use result directly, integrate stress to equivalent force and place concentrate reinforcement far from section cg.

just compare with elastic FE analysis for normal beams and normal column, infinite stress at corner is the same problem.

fig1


fig2


fig3

[jameschan]

Attn: techno

since you mentioned at wall corner the compression force (KN/m) is very high. so, i suggested that you assign pier to the corner element for 1 floor height and etabs will integrate the pier axial and bending for you. Then, you can design and reinforce the pier as a column.