Best Material to Have Descending Branch in Pushover Curve

[rmandok]

Hi,
I want to do pushover analysis for a simple one storey and one bay frame. I define material with steel01 material and when the analysis done, I couldn't find any descending branch in pushover capacity curve and it is ascending. Is it because the definition of material? Will the problem be solved if i change the material to hysteretic material?! (I use PDelta transformation but still i can not see the descending branch.

wipe

set file Push
file mkdir $file

model basic -ndm 2 -ndf 3

node 1 0. 0.
node 2 0. 3.5
node 3 3. 3.5
node 4 6. 3.5
node 5 6. 0.

fixY 0.0 1 1 1

mass 2 [expr (400.+0.2*100)*6.*6/3] 1.e-9 1.e-9
mass 3 [expr (400.+0.2*100)*6.*6/3] 1.e-9 1.e-9
mass 4 [expr (400.+0.2*100)*6.*6/3] 1.e-9 1.e-9

#geomTransf Linear 1
#geomTransf PDelta 1
geomTransf Corotational 1

set Fy 2.4e8
set E0 2.1e11
set b 0.01
uniaxialMaterial Steel01 1 $Fy $E0 $b
#set R0 18.5
#set cR1 0.925
#set cR2 0.15
#uniaxialMaterial Steel02 1 $Fy $E0 $b $R0 $cR1 $cR2

set HEBC 180
set IPEB 45
source Wsection.Tcl
Wsection $HEBC 1 0.18 0.18 0.0140 0.0085 40 8 12 16
Wsection $IPEB 1 0.45 0.19 0.0146 0.0094 40 8 12 16

#set h 0.16
#set bf 0.082
#set tf 0.0074
#set dw [expr $h-2*$bf]
#set tw 0.005
#section fiberSec $IPE {
#patch quad 1 3 4 [expr -$h/2] [expr -$bf/2] [expr -$h/2] [expr $bf/2] [expr -$dw/2] [expr $bf/2] [expr -$dw/2] [expr -$bf/2]
#patch quad 1 2 10 [expr -$dw/2] [expr -$tw/2] [expr -$dw/2] [expr $tw/2] [expr $dw/2] [expr $tw/2] [expr $dw/2] [expr -$tw/2]
#patch quad 1 3 4 [expr $dw/2] [expr -$bf/2] [expr $dw/2] [expr $bf/2] [expr $h/2] [expr $bf/2] [expr $h/2] [expr -$bf/2]
#}

set numIntgrPts 10
element nonlinearBeamColumn 1 1 2 $numIntgrPts $HEBC 1
element nonlinearBeamColumn 2 2 3 $numIntgrPts $IPEB 1
element nonlinearBeamColumn 3 3 4 $numIntgrPts $IPEB 1
element nonlinearBeamColumn 4 5 4 $numIntgrPts $HEBC 1

puts "w^2=[eigen 1]
T=[expr 2*4*atan(1)/pow([eigen 1],0.5)]"


pattern Plain 1 Linear {
#load 2 0. [expr -(400+100)*9.81*6.*6./4] 0.
#load 2 0. [expr -(400+100)*9.81*6.*6./2] 0.
#load 4 0. [expr -(400+100)*9.81*6.*6./4] 0.
eleLoad -ele 2 3 -type -beamUniform [expr -(400+100)*6.*9.81]
}

constraints Plain
numberer RCM
system BandGeneral
test NormDispIncr 1.0e-8 6
algorithm Newton
integrator LoadControl [expr 1/100.]
analysis Static
analyze 100
loadConst -time 0.0


recorder Node -file $file/force1.out -time -node 1 -dof 1 2 3 reaction
recorder Node -file $file/force5.out -time -node 5 -dof 1 2 3 reaction
recorder Node -file $file/disp4.out -time -node 4 -dof 1 2 3 disp
recorder Element -file $file/force11.out -time -ele 1 section 1 force
recorder Element -file $file/force12.out -time -ele 1 section 8 force
recorder Element -file $file/force21.out -time -ele 2 section 1 force
recorder Element -file $file/force32.out -time -ele 3 section 8 force
recorder Element -file $file/force41.out -time -ele 4 section 1 force
recorder Element -file $file/force42.out -time -ele 4 section 8 force
recorder Element -file $file/def11.out -time -ele 1 section 1 deformation
recorder Element -file $file/def12.out -time -ele 1 section 8 deformation
recorder Element -file $file/def21.out -time -ele 2 section 1 deformation
recorder Element -file $file/def32.out -time -ele 3 section 8 deformation
recorder Element -file $file/def41.out -time -ele 4 section 1 deformation
recorder Element -file $file/def42.out -time -ele 4 section 8 deformation

pattern Plain 2 Linear {
load 2 250000. 0. 0.
load 3 250000. 0. 0.
}

set xDamp 0.025; # damping ratio
set MpropSwitch 1.0;
set KcurrSwitch 0.0;
set KcommSwitch 1.0;
set KinitSwitch 0.0;
set nEigenI 1; # mode 1
set nEigenJ 1; # mode 1
set lambdaN [eigen [expr $nEigenJ]]; # eigenvalue analysis for nEigenJ modes
set lambdaI [lindex $lambdaN [expr $nEigenI-1]]; # eigenvalue mode i
set lambdaJ [lindex $lambdaN [expr $nEigenJ-1]]; # eigenvalue mode j
set omegaI [expr pow($lambdaI,0.5)];
set omegaJ [expr pow($lambdaJ,0.5)];
set alphaM [expr $MpropSwitch*$xDamp*(2*$omegaI*$omegaJ)/($omegaI+$omegaJ)]; # M-prop. damping; D = alphaM*M
set betaKcurr [expr $KcurrSwitch*2.*$xDamp/($omegaI+$omegaJ)]; # current-K; +beatKcurr*KCurrent
set betaKcomm [expr $KcommSwitch*2.*$xDamp/($omegaI+$omegaJ)]; # last-committed K; +betaKcomm*KlastCommitt
set betaKinit [expr $KinitSwitch*2.*$xDamp/($omegaI+$omegaJ)]; # initial-K; +beatKinit*Kini

rayleigh $alphaM $betaKcurr $betaKinit $betaKcomm;

puts "groundmotion start!.Time: [getTime]"

constraints Plain
numberer RCM
system BandGeneral
test NormDispIncr 1.0e-8 6
algorithm Newton
integrator DisplacementControl 2 1 0.001
analysis Static
analyze 1000

puts "groundmotion done!.End Time: [getTime]"

On the other hand, I have an example that the material is defined as steel01 material (and it is for a free end member) but i can see the descending branch in it. Is it because it's end is free and it the analysis can consider PDelta effect?

wipe
wipeAnalysis

model BasicBuilder -ndm 2 -ndf 3;
file mkdir data1;


node 1 0 0;
node 2 0 5.


fix 1 1 1 1;


mass 2 200. 1.e-9 1.e-9 ;


geomTransf Corotational 1;

#geomTransf PDelta 1

#geomTransf Linear 1


#uniaxialMaterial Elastic 2 2.e11

uniaxialMaterial Steel01 1 2.4e8 2.e11 0.01
#uniaxialMaterial Steel02 3 240.e6 2.e11 0.01 15 .925 .15



section fiberSec 1 {
patch quad 1 2 3 0.0348 0.023 0.0348 -0.023 0.04 -0.023 0.04 0.023
patch quad 1 2 8 -0.0348 0.0019 -0.0348 -0.0019 0.0348 -0.0019 0.0348 0.0019
patch quad 1 2 5 -0.04 0.023 -0.04 -0.023 -0.0348 -0.023 -0.0348 0.023
}



#element elasticBeamColumn 1 1 2 7.4288e-4 2.e11 77.701e-8 1;
#source section.tcl
#set np 10
#element nonlinearBeamColumn 1 1 2 $numIntgrPts $secIDipe80 1

element nonlinearBeamColumn 1 1 2 10 1 1


eigen 1 ;
set T1model [expr 2*3.1416/(pow([eigen 1],0.5))]
puts "T1model=[expr $T1model]sec"




recorder Node -file Data1/DFree.out -time -node 2 -dof 1 2 3 disp;
recorder Node -file Data1/DBase.out -time -node 1 -dof 1 2 3 disp;
recorder Node -file Data1/RBase.out -time -node 1 -dof 1 2 3 reaction;
recorder Drift -file Data1/Drift.out -time -iNode 1 -jNode 2 -dof 1 -perpDirn 2 ;
recorder Element -file Data1/FCol.out -time -ele 1 globalForce;


pattern Plain 1 Linear {
load 2 0. -2000. 0.;
}

constraints Plain;
numberer Plain;
system BandGeneral;
test NormDispIncr 1.0e-8 6 ;
algorithm Newton;
integrator LoadControl 0.1;
analysis Static
analyze 10;
loadConst -time 0.0;

#wipeAnalysis


pattern Plain 2 Linear {
load 2 2000. 0.0 0.0;
}


constraints Plain;
numberer Plain;
system BandGeneral;
test NormDispIncr 1.0e-8 60 ;
algorithm Newton ;
integrator DisplacementControl 2 1 0.001;
analysis Static
analyze 1000;

puts "Done!"

[fmk]

if you are not seeing the descending branch with loadcontrol and then displacement control it probably means you are not pushing the structure over enough.

[rmandok]

Dear Frank, my frame designed only for gravitational load and is weak and i pushed it to 1 meter. Can you help me with this please?

Here is the capacity curve:
http://cdn.imghack.se/images/9ad4ac07b8 ... 114483.jpg

(And the first code, above is the OpenSees code for it.)

when I change the geomtransf to PDelta, again i don't have descending branch

http://cdn.imghack.se/images/363d5c1610 ... 0e5d68.jpg

and when i change the material to Hysteretic with PDelta geomtransf, This happens:

set Fy 2.4e8
set E0 2.1e11
set b 0.01
set epsY [expr $Fy/$E0]
set espsSH [expr 8*$epsY]
set Esh [expr 0.02*$E0]
set Fy1 [expr 1.5*$Fy]
set epsY1 [expr $epsY+($Fy1-$Fy)/($b*$E0)]
set Fu [expr 1.5001*$Fy]
set epsU 0.25
set pinchX 1
set pinchY 1
set damage1 0.0
set damage2 0.0
set betaMUsteel 0.0
uniaxialMaterial Hysteretic 1 $Fy $epsY [expr $Fy+0.02*$Fy] $espsSH $Fu $epsU -$Fy -$epsY -[expr $Fy+0.02*$Fy] -$espsSH -$Fu -$epsU $pinchX $pinchY $damage1 $damage2

http://cdn.imghack.se/images/63e39ae846 ... e17226.jpg