difference in first period

[hoseindnn]

hi. i have a simple model 1 bay 1 story steel frame . opensees give me T1= 0.232 S . i have same model in Etabs with T1= 0.242 S .i want to know where can be this problem ?
when i increase number of bay and story this difference is higher.please help me to know why this two software have this difference.
my model :
wipe;
model BasicBuilder -ndm 2 -ndf 3;
source WSection.tcl;
source LibUnits.tcl;
source DisplayModel2D.tcl
node 1 0.0000000000 0.0000000000;
node 2 358.2000000000 0.0000000000;
node 3 0.0000000000 156.0000000000 -mass 0.05686300000 0.0058350000 0.0000000010;
node 4 358.2000000000 156.0000000000 -mass 0.05686300000 0.0058350000 0.0000000010;
fix 1 1 1 1;
fix 2 1 1 1;
uniaxialMaterial Steel02 1 50.0 29000.0 0.003 20 0.925 0.15 0.0005 0.01 0.0005 0.01;
WSection 1 1 14.200 10.100 0.785 0.450 8 1 1 4;
WSection 50 1 16.300 10.200 0.665 0.395 8 1 1 4;
geomTransf PDelta 1;
geomTransf Corotational 2;
element forceBeamColumn 1 1 3 4 1 1 -iter 10 0.0000000100;
element forceBeamColumn 2 2 4 4 1 1 -iter 10 0.0000000100;
element forceBeamColumn 200 3 4 3 50 1 -iter 10 0.0000000100;
DisplayModel2D "DeformedShape" 1 50 50 500 500 0;
set pi [expr 2.0*asin(1.0)]; # Definition of pi;
set nEigenI 1; # mode i = 1;
set nEigenJ 2; # mode j = 2;
set lambdaN [eigen [expr $nEigenJ]]; # eigenvalue analysis for nEigenJ modes;
set lambdaI [lindex $lambdaN [expr 0]]; # eigenvalue mode i = 1;
set lambdaJ [lindex $lambdaN [expr $nEigenJ-1]]; # eigenvalue mode j = 2;
puts "eigen11 = $lambdaI s";;
set w1 [expr pow($lambdaI,0.5)]; # w1 (1st mode circular frequency);
set w2 [expr pow($lambdaJ,0.5)]; # w2 (2nd mode circular frequency);
set T1 [expr 2.0*$pi/$w1]; # 1st mode period of the structure;
set T2 [expr 2.0*$pi/$w2]; # 2nd mode period of the structure;
puts "T1 = $T1 s"; # display the first mode period in the command window;
puts "T2 = $T2 s"; ;

[fmk]

use an elastic section in OpenSees and compare the periods .. if still differerent your models are different, if now the same your WSection script needs work or the material properties. they should give exactly the same.

[rafal]

Check if ETABS includes shear behaviour in the stiffness matrix. I remember having a similar difference in natural frequencies comparing Sap2000 and OpenSees results. Setting shear area to 0.0 in Sap2000 gave the same results as in OpenSees.

[hoseindnn]

hi again
tnx for answer

i replaced all element with elastic element in opensees and my answer get closer to etabs but there is a diffrence yet.
i set shear area modifier zero in etabs and didnt get any changes in period.

i have 10% error for nonlinear element and 6 % error for elastic element. is it okey?

what can be caused this difference except above reasons

[fmk]

the fiber section and elastic section should be v similar .. what are the elastic section properties

[hoseindnn]

it is not about fiber or wsection it is about forceBeamColumn element.
i use elastic element with w section properties (area and moment of inertia) same in etabs. but i have diffrence

[fmk]

it is about the section definition as you are messing up at the point. it is not the forceBeamColumn. if you have it correct. elasticbeam, force beam and even disp beam will give the same answer. they will be the same as etabs if you can get the models the same. here is the opensees script for different ele types. you can modify it to switch between WSection and Elastic section by adding another foreach loop.

foreach eleType {dispBeamColumn forceBeamColumn Elastic} {

wipe;
model BasicBuilder -ndm 2 -ndf 3;
node 1 0.0000000000 0.0000000000;
node 2 358.2000000000 0.0000000000;
node 3 0.0000000000 156.0000000000 -mass 0.05686300000 0.0058350000 0.0000000010;
node 4 358.2000000000 156.0000000000 -mass 0.05686300000 0.0058350000 0.0000000010;
fix 1 1 1 1;
fix 2 1 1 1;
geomTransf PDelta 1;
# geomTransf Corotational 2;

set A 100.; set E 29000.; set I 1000.
section Elastic 1 $E $A $I
section Elastic 50 $E $A $I

if {$eleType != "Elastic"} {
element $eleType 1 1 3 4 1 1
element $eleType 2 2 4 4 1 1
element $eleType 200 3 4 4 50 1
} else {
element elasticBeamColumn 1 1 3 $A $E $I 1
element elasticBeamColumn 2 2 4 $A $E $I 1
element elasticBeamColumn 200 3 4 $A $E $I 1
}

set pi [expr 2.0*asin(1.0)]; # Definition of pi;
set nEigenI 1; # mode i = 1;
set nEigenJ 2; # mode j = 2;
set lambdaN [eigen [expr $nEigenJ]]; # eigenvalue analysis for nEigenJ modes;
set lambdaI [lindex $lambdaN [expr 0]]; # eigenvalue mode i = 1;
set lambdaJ [lindex $lambdaN [expr $nEigenJ-1]]; # eigenvalue mode j = 2;

puts $eleType
puts "eigen11 = $lambdaI s";;
set w1 [expr pow($lambdaI,0.5)]; # w1 (1st mode circular frequency);
set w2 [expr pow($lambdaJ,0.5)]; # w2 (2nd mode circular frequency);
set T1 [expr 2.0*$pi/$w1]; # 1st mode period of the structure;
set T2 [expr 2.0*$pi/$w2]; # 2nd mode period of the structure;
puts "T1 = $T1 s"; # display the first mode period in the command window;
puts "T2 = $T2 s"; ;
}

[hoseindnn]

you mean when i just change element type (forcebeam to elastic) the first period should not change ?
i think type of element ( linear non linear) can change stiffness matrix and period

[reebt]

I had the same problem and as rafal said, ETABS considers the shear stiffness. Using a shear area of 0 solved the problem.
But now, what if I want to consider the shear stiffness in OpenSees?
Is there a way to add it to a frame model with forcebeamcolumn elements?

[hoseindnn]

reebt wrote:
> I had the same problem and as rafal said, ETABS considers the shear
> stiffness. Using a shear area of 0 solved the problem.
> But now, what if I want to consider the shear stiffness in OpenSees?
> Is there a way to add it to a frame model with forcebeamcolumn elements?

hi
i change shear area modifier to zero in etabs but i get same answer again. did i do wrong ? how can i set shear area 0 ?

[fmk]

the type of element will not change the initial periods, they will change as model goes nonlinear, but initially everything is still elastic and they will all give the same result. if they don't your model is incorrect in either material or section definition.

[satis]

Hi, why don't you try using a linear transformation instead of PDelta and compare the periods. They must be exactly the same.