Muneeb Badar

Dear Shaker, The better way is to model this retaining wall as vertical beam element for 1 meter width. Apply the appropriate boundary conditions such as fix at bottom and free or pin at top. Apply the soil pressure laterally and then simply take moments and shear. Shear force will be used to decide the thickness. So just use the general shear formula which we normally use for one way shear and decide the thickness. For flexure use beam flexure formula to calculate reinforcement. Dear Waqas, you are right. Actually in basements the seismic shear/forces are transferred fro retaining wall to shear walls. So there will be inplane forces in the retaining wall. For flexural reinforcement you can use earth pressure but as you know there is a distribution reinforcement in the retaining wall as well. This reinforcement will be designed for inplane forces. This is very simple to calculate. Assign the one side of retaining wall as one pier. Extract V2 force and the use the length of retaining wall as depth and use one way shear formula and calculate reinforcement as in case of shear wall. Thanks Muneeb

Dear you can design it on ETABS for flexure and for shear just take shear force and check the capacity fro formulas given in ACI

Dear Khurram, This type of beam is called coupling beam or link beam. These beams connects two legs of shear wall and transfer the shear fro one wall to another and make these shearwalls to act as one unit. if you increase the size then moment and shear will be increased. If l/h <4 then you need to design that beam as link beam and ACI 21.9.7 provides guidance how to design link beam. you need to provide diagonal bars to balance the shear force. You need to design as link beam. But if still you can not satisfy the demands then you can model this beam as simply supported beam and provide the reinforcement as simply supported beam. Actually the structure will behave as we want it to behave. If you provide the reinf. to be simply supported then in case of earthquake these beams will yield and transfer the shear and flexure to the adjacent walls. Secondly if you design these as link beams the only difference is these will dissipate some energy while in the previous case these will not dissipate energy. So you can design this beam as simply supported beam. Thanks Muneeb

Dear in normally earthquake resistant structure, it is not common to have large axia forces in the beams. So you do not need to check all members. Just for your own satisfaction you can check few beams. But if there are Transfer Girders or Beams connecting to tapered columns, beams supporting columns, then you need to check the axial force. In this case, in ETABS, you can just design that beam as column by providing in the section properties P-M2-M3 design Thanks Muneeb

Dear Abdul Qadeer, There are two points 1- You can add the stair in the structure and ramps in the model just to have a bracing in the members like column and retaining walls. This will provide you stiffness as well as the effective length of the other members reduced which is the actual and practical situation. As far as the addition of stairs and ramps for diaphragm concerned, you can not add these in the diaphragm. Basically as we all know diaphragm transfers the lateral forces to vertical members. The diaphragm design is basically the design of SLAB for inplane forces. If you have not design the slab for these inplane forces then you can not assume semi rigid or rigid diaphragm (In case of minor earthquake areas, normally the existing slab thickness will be enough and also not required by code to design diaphragm for minor earthquake areas). So diaphragm assumption will be valid if you provide a path to transfer the forces from Retaining wall to shearwall through slab at basement levels and from shear wall to slab at top levels. If there is an opening or there is a stair or ramp then you do not have any path to transfer the forces to the shearwall and you can not provide the reinforcement for inplane forces. In this case it is not good enough to model these ramps and consider them these are transferring the forces. Thanks Muneeb

Dear Mohammad Shaker I do not know about this and also did not hear about this at all . In my point of view, frame element can be converted to shell elements where there are large transfer of forces at the joints. For example, in case of Transfer Girders (the girders which transfer shear force from shear wall to adjacent columns) must be or can be modeled as shell element as these are directly attached to the member(shearwall) which is transferring heavy shear. The same is the case under Pile Cap. The connection between pile cap and top of pile is critical. You may have seen that the pile near pile cap shows heavily reinforced because we model it node to node so some portion of pile must be inside the pile cap. But when you model frame element it will not catch this effect. So to have this effect we model top portion of pile as shell element. As far as what you are saying is concerned, in my point of view, the absorption/collection of forces is stiffness based. If you model a s column of 300x1500 as frame element then you need to design it as bi axial bending as it will take inplane as well as out of plane bending. The shear reinf. you need to provide must be both sides. But of you provide this 300x1500 member as shell element then you need to skip out of plane bending because we normally neglect the out of plane stiffness of shear wall. So if shell element with proper modifiers it will not absorb the lateral force in out of plane direction. But if you use as frame element it will give you inplane as well as out of plane stiffness and then you need to design as for both biaxial bending. I am not sure I make a proper reply to your question or not......... Thanks Muneeb

Dear for this particular error, I think u r the best person who can search for this as u know the problem in details.......... But kindly tell me how do u know that by converting to shell element torsion will be reduced and your reinforcement will be reduced. How did u concluded that. Thanks Muneeb

Dear I do not know about this error. It did not happen to me before. But why you wished to use shell element instead of frame element? Secondly you can try it in SAP. I have assigned some beams and columns as shell element in SAP. It did not gave any error like this. Thanks Muneeb

In my opinion, the main question is that how you want to proceed with this. Do you want to use normal approach. Or you want to have Push Over Analysis or Non Linear Time History Analysis. After this decision it will be easy for you to have focused what you want to do. Thanks Muneeb

As you know that we allow the yielding in the flexural members. In shearwall, we also allow flexural yielding but do not allow shear yielding. We also do not want the brittle failure we always like to have ductile yielding. In case of shearwall we want the flexural yielding at certain locations. We can not allow the yielding at unwanted locations. When earthquake pushes a shearwall (structure) in one direction the in shearwall there will be Tension on one side (one corner) and Compression on the other side (other corner). This can be reversed. We do not want the shearwall to fail in brittle form and specially at corners. So code required that concrete to be confined by the horizontal reinforcement. In this case the fc' of that corners increased significantly and there will be no yielding at these locations. We need the shearwall to be yielded so we provide minimum flexural reinforcement in the NON BOUNDARY ZONE ELEMENTS. So that the shearwall can yield at the middle locations in case of earthquake. The remaining reinforcement ( Total Flexural Reinforcement - minimum reinforcement) is provided in the BOUNDARY ZONE ELEMENTS. For this actually we need to check the strain limits obtained by Non Linear Time History Analysis but code also provides a method to calculate boundary zone lengths. So thats the reason for confined zones. The length of Boundary zone also depends u[on Thanks Muneeb

Dear Qasim, As per my experience, the reinforcement required (which is minimum governed) is much less than which should be in that high rise building. Kindly tell which zone r u using. What is the R value. What is the system type dual or not.What diaphragm you assigned. What is the base shear percentage by weight. How you have assigned the piers to shear wall. How many shear walls in plan. The length of above mentioned shearwall is 16m?? Thanks Muneeb

I think you are talking about Progressive Collapse Analysis? You should check some details regarding Plastic Analysis not the non linear analysis. Non linear is also continuation of plastic analysis but I think what you are looking for can be found in Plastic Analysis guide lines. Thanks Muneeb

well I am not sure about that how can we calculate the yielding load for a beam. For this you need to dig into the literature. Also please share what you found. I ll also try to find it. Thanks Muneeb

Dear Omer, Do not provide the concealed beam of 9"x9". It will worses the situation. In my opinion 19'x18' is not too big span. If this is a frame structure then you can remove this beam and apply the loading directly on slab as this slab will be having the loading of walls, SDL and any other Live Load. You can increase thickness to 7 inch or or 7.5 inch. In shal ALLAH it will solve your problem. Because 7 or 7.5 inch slab has sufficient resistance to deflection but you can check it in SAFE. If this is a multistory housing (brick masonry) structure, then also use this slab only for the loading of one floor. Try to locate the load bearing points on the upper floor. Apply the loading on the slab and design it as mentioned in above comment. But if you want to provide concealed beam then go for 24"x10.5". In usual construction, the total slab thickness (including concrete slab+ floor finishing ) is 12". So with the mutual understanding with architect you can provide 24" wide beam with 10.5" depth. This will not solve your 100% deflection but it can help. I have provided this kind of beams in too many brick masonry structures. Thanks Muneen

Dear Goher, If you are using prestress or pre-engineered columns, then in that case how you are going to connect the columns with foundation. If you provide some cast inplace pedestals then you can assign horizontal loads on them in addition to vertical loads. This will solve your problem. Thanks Muneeb