Problem with zerolength elements in modal analysis

[redwood]

Hi,

I have a simple one bay-one story frame with simply supported base conditions at both legs. Both legs are connected to the roof by equalDOF command in horizontal and vertical directions. The connectivity for the rotational degree of freedom is enforced by zeroLength elements with elastic uniaxial material properties. When I run modal analysis for this structure, the periods are quite large implying that the system is unstable. But if I enforce the rotational connectivity by equalDOF command, the results make sense. I copied the code below. Could you please take a look and let me know where the problem is.

Thanks,

--------------------------------------------------------
wipe;
model BasicBuilder -ndm 2 -ndf 3;

set HF 10.0;
set pX2 5.0;
set pX1 0.0;

## NODES ##
###################
node 1 $pX1 0.0;
node 2 $pX1 $HF;
node 3 $pX1 $HF;
node 4 [expr ($pX1+$pX2)/2.0] $HF;
node 5 $pX2 $HF;
node 6 $pX2 $HF;
node 7 $pX2 0.0;

fix 1 1 1 0
fix 7 1 1 0

equalDOF 2 3 1 2
equalDOF 6 5 1 2

## MASS ##
###################
set Mass 10.0;
mass 4 [expr 1.0*$Mass] [expr 1.0*$Mass] 0.;
mass 3 [expr 0.01*$Mass] [expr 0.01*$Mass] 0.;
mass 5 [expr 0.01*$Mass] [expr 0.01*$Mass] 0.;

## ELEMENTS ##
###################
set CGTag 101
geomTransf Linear $CGTag

set AElem 1.0;
set EElem 29000.0;
set IElem 0.1;

element elasticBeamColumn 1 1 2 $AElem $EElem $IElem $CGTag
element elasticBeamColumn 2 7 6 $AElem $EElem $IElem $CGTag
element elasticBeamColumn 3 3 4 $AElem $EElem $IElem $CGTag
element elasticBeamColumn 4 4 5 $AElem $EElem $IElem $CGTag

uniaxialMaterial Elastic 102 10000.0

element zeroLength 5 2 3 -mat 102 -dir 3
element zeroLength 6 6 5 -mat 102 -dir 3

puts "Modeling Finished"

## Modal Analysis ##
###################
set pi 3.14159;
set nEigenI 1; # mode 1
set nEigenJ 2; # mode 2
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($lambdaJ,0.5)];
set Ti [expr 2.0*$pi/$omegaI];
set Tj [expr 2.0*$pi/$omegaJ];
puts "Ti: $Ti , Tj: $Tj"

[vesna]

You are assigning zeroLength elements in dir 3 which is translation along Z axis. Replace your zero length elements with this:

element zeroLength 5 2 3 -mat 102 -dir 6
element zeroLength 6 6 5 -mat 102 -dir 6

and you will have them assigned to rotational degree of freedom.

[redwood]

Thanks