Civil Engineering Association

Hi Paladin,

failures during earthquake could happen at column, beam, wall, and beam-column joint. in columns damage, the causes are mainly due to flexural and shear failures at column ends. flexural failure is affected by the level of axial force in column, particularly at the corner exterior columns. the column was experiencing extremely high compression force due to overturning moment of structure, which was leading to flexural compression failure followed by the loss of gravity load carrying capacity. compression failures of shear and flexural could take place near the column ends.

the shear failure at column is propagated because the tensile stress carried by the concrete before onset of significant shear cracking cannot be resisted by shear reinforcement once shear cracks open which then leading to diagonal tension failure.

buckling (flex outward) failure in column is commonly caused by deficiency of ties and excessive compression force. it actually can be occurred at any place of column span where deficiency of ties takes place. earthquake reconnaissance reports show that buckling failures appear at near column ends as well.

for me, it is best to place point of plastic hinge of member at point where the arrangement of longitudinal and shear reinforcements are not identical to get accurately seismic performance result. nevertheless, using member ends (or near) as plastic hinge points are accurately enough based on many modelling studies that referred to experimental testing.

cheers

Since you use Sap2000 or ETABS, reading FEMA356 and its guidelines on static nonlinear analysis would help. But it is up to you to decide how to model the plastic hinges you expect will be formed (and where..). My advice is: run an elastic analysis using the design loads of the structure and see which is the closest to yield: bending moment or shear force. Then you will form the hinge law on the programm accordingly (M3, or V2 for beams, usually PMM for columns). Either way it is explained well enough on FEMA 356.

PS I don't really like pushover analysis when it comes to 3d models with irregular shape and altered stiffness on each floor as higher eigenmodes have significant contribution to mass participation, while the load pattern of the analysis does not correspond to them. If this happens, FEMA predicts measures, albeit, in my opinion, questionable.

Good luck,
Avge

Dear iceman84, adekajeng and avge
For advanced user of pushover analysis is suggested to modify the properties of pushover hinge ( V, M & PMM ).
Does this pocedure ( modify hinge properties ) can enter/consider real ( actual used ) shear reinforcement of beam/column ?
...
Dear avge, I want ask about your answer ( "...see which is the closest to yield: bending moment or shear force ")
For example for a beam has M design & V design ---> result bending & shear reinforcement ---> can be calculated the Mn actual & Vn actual
Mn actual --> bending moment strength provided by the RC beam + used/actual bending reinforcing
Vn actual --> shear strength provided by the RC beam + used/actual shear reinforcing
Then how to know whether the bending or the shear that closest to the yield ?
...
Dear iceman84, adekajeng and avge, thanks for your time & your answers
I really appreciate your help to me.
Regards
Paladin

Dear paladin,

First things first. If your structure is old, therefore analyzed using old design codes, then there are more variables to consider on the form and properties of the hinge type you will use. If the structure is studied using relatively new codes, then the hinge will probably have ductile properties (bending failure at beam end) instead of shear resistance incapacity (brittle failure). Then you use M type hinge.
Answering your question, you need to estimate how much more reinforcement (shear and bending) was used by the first engineer than necessary. If they are close in percentage, see if there are solutions to prevent shear failure at the end of your beams, such as shorter distance between shear reinforcement compared to the mid-span of the beam. If this is the case, then I would use M type plastic hinge. Besides, in FEMA356 (Prestandard and commentary for the seismic rehabilitation of buildings), which I recommended earlier, you do take into consideration the shear reinforcement in an M type hinge. Pushover is relatively new, so what you do is an evaluation of the method itself.
FEMA 356 (2000)

Dear Paladin,

yes, modify the properties of pushover hinge (V, M & PMM) through hinge element in etabs or sap2000 can consider actual shear reinforcement of beam/column. you can use section analysis program (cumbia, response2000, USC_RC) to define force-deformation (yield) condition of sections at any point that you need to consider.

cheers.