Hello!
I am trying to simulate a simple 1-storey elasticBeamColumn 3D building with OpenSees and I would be grateful for your help.
A 3D picture of the building is given on this link http://i1311.photobucket.com/albums/s66 ... d0b9e6.jpg
A modal analysis of this on SAP2000 returned these results for Natural frequencies Τ (considering 3 DOFs for the rigid diaphragm master node) : T1=0.172sec , T2=0.111 sec , T3=0.022sec
Considering the coordinate system in the picture above, I used the following Linear transformations:
geomTransf Linear 2 0 0 -1 (for beams)
geomTransf Linear 1 0 1 0 (for columns)
The whole code I used is this:
_______________________________
model BasicBuilder -ndm 3 -ndf 6
node 1 0 0 0
node 2 6 0 0
node 3 0 5 0
node 4 6 5 0
node 5 3 0 0
node 6 0 0 3
node 7 0 5 3
node 8 2 0 3
node 9 6 0 3
node 10 6 5 3
node 11 4 0 3
node 12 3 0 3
set RigidDiaphragm ON ;
node 13 3 2.5 3
fix 1 1 1 1 1 1 1
fix 2 1 1 1 1 1 1
fix 3 1 1 1 1 1 1
fix 4 1 1 1 1 1 1
fix 5 1 1 1 1 1 1
fix 13 0 0 1 1 1 0
geomTransf Linear 1 0 1 0
geomTransf Linear 2 0 0 -1
element elasticBeamColumn 1 6 7 0.15 29000000 12083333.3333333 0.00230674790747372 0.0045 0.00078125 2
element elasticBeamColumn 2 6 8 0.15 29000000 12083333.3333333 0.00230674790747372 0.0045 0.00078125 2
element elasticBeamColumn 3 9 10 0.15 29000000 12083333.3333333 0.00230674790747372 0.0045 0.00078125 2
element elasticBeamColumn 4 11 9 0.15 29000000 12083333.3333333 0.00230674790747372 0.0045 0.00078125 2
element elasticBeamColumn 5 7 10 0.15 29000000 12083333.3333333 0.00230674790747372 0.0045 0.00078125 2
element elasticBeamColumn 6 8 12 1000000 29000000 12083333.3333333 1000000 1000000 1000000 2
element elasticBeamColumn 7 12 11 1000000 29000000 12083333.3333333 1000000 1000000 1000000 2
element elasticBeamColumn 8 1 6 0.16 29000000 12083333.3333333 0.00360533333333333 0.00213333333333333 0.00213333333333333 1
element elasticBeamColumn 9 2 9 0.16 29000000 12083333.3333333 0.00360533333333333 0.00213333333333333 0.00213333333333333 1
element elasticBeamColumn 10 3 7 0.16 29000000 12083333.3333333 0.00360533333333333 0.00213333333333333 0.00213333333333333 1
element elasticBeamColumn 11 4 10 0.16 29000000 12083333.3333333 0.00360533333333333 0.00213333333333333 0.00213333333333333 1
element elasticBeamColumn 12 5 12 0.4 29000000 12083333.3333333 0.00499733613333333 0.00133333333333333 0.133333333333333 1
rigidDiaphragm 3 13 6 7 8 9 10 11 12
mass 13 63.098878695209 63.098878695209 0 0 0 55.2115188583079
set lambda [eigen -fullGenLapack 3]
set omega {}
set f {}
set T {}
set pi 3.14159265359
foreach lam $lambda {
lappend omega [expr sqrt($lam)]
lappend f [expr sqrt($lam)/(2*$pi)]
lappend T [expr (2*$pi)/sqrt($lam)]
}
_____________________________________
The returning results are: T1=0.1607, T2=0.0698, T3=0.040 sec
As you can see, natural frequencies from SAP2000 & OpenSees do not converge satisfactorily...
Can anyone spot the bug/mistake in the code? I am not sure about the Geometric Transformation command.
[Sections: Columns 40/40 cm , Beams 25/60 cm , Pseudocolumn 200/20 cm for shear-wall simulation, Rigid beams ]
Thank you for your time.
Simple 3D building model - Linear Transformations problem?
Moderators: silvia, selimgunay, Moderators
Simple 3D building model - Linear Transformations problem?
Last edited by NikosCV on Thu Feb 20, 2014 6:07 am, edited 1 time in total.
Re: Simple 3D building model - Linear Transformations proble
If anyone has a clue, I would really appreciate it.
This is about my post-graduate thesis, which includes concrete shear walls modelling with linear beam elements among others, and is crucial for me.
Thank you.
This is about my post-graduate thesis, which includes concrete shear walls modelling with linear beam elements among others, and is crucial for me.
Thank you.
Re: Simple 3D building model - Linear Transformations proble
Hi,
The following post may be of use. http://opensees.berkeley.edu/community/ ... =2&t=29144
Thanks.
The following post may be of use. http://opensees.berkeley.edu/community/ ... =2&t=29144
Thanks.
----------------------------------
With Regards,
sb1966
With Regards,
sb1966
Re: Simple 3D building model - Linear Transformations proble
sb1966 wrote:
> Hi,
> The following post may be of use. http://opensees.berkeley.edu/community/ ... =2&t=29144
> Thanks.
Thanks a lot for your answer!
Sadly, I haven't managed to find the root of the problem, so I am in a dead end.
The strange thing is that, when I run an eigen analysis of a building with columns as vertical elements only, the eigen results match with the ones from SAP2000.
Only when I insert walls modelled like this : http://i1311.photobucket.com/albums/s66 ... 093c30.png
the natural periods deviations become significant.
Does anyone have a clue?
> Hi,
> The following post may be of use. http://opensees.berkeley.edu/community/ ... =2&t=29144
> Thanks.
Thanks a lot for your answer!
Sadly, I haven't managed to find the root of the problem, so I am in a dead end.
The strange thing is that, when I run an eigen analysis of a building with columns as vertical elements only, the eigen results match with the ones from SAP2000.
Only when I insert walls modelled like this : http://i1311.photobucket.com/albums/s66 ... 093c30.png
the natural periods deviations become significant.
Does anyone have a clue?
Re: Simple 3D building model - Linear Transformations proble
Nikos CV did u solve your problem?
Re: Simple 3D building model - Linear Transformations proble
Basa wrote:
> Nikos CV did u solve your problem?
Thank you for your concern!
No, unfortunately not yet. Strangely enough, the input I insert to both programms is exactly the same, but results differ (~10% divergence for the first natural period of vibration).
It is significant for me to achieve a good convergence in the first natural period, so that I can confirm that the solution with OpenSees is reliable.
Do you have any clue?
EDIT --
I made a mistake in my above statements.
The problem was likely solved for individual walls ( I achieved a 97-98% convergence for the first natural period of vibration of my example models) , BUT it persists when I simulate C-shaped core walls.
The divergence is about 10% for the first natural period. It seems that there is a difference in the calculation of the Torsional Stiffness of the system.
Any help would be of use!
> Nikos CV did u solve your problem?
Thank you for your concern!
No, unfortunately not yet. Strangely enough, the input I insert to both programms is exactly the same, but results differ (~10% divergence for the first natural period of vibration).
It is significant for me to achieve a good convergence in the first natural period, so that I can confirm that the solution with OpenSees is reliable.
Do you have any clue?
EDIT --
I made a mistake in my above statements.
The problem was likely solved for individual walls ( I achieved a 97-98% convergence for the first natural period of vibration of my example models) , BUT it persists when I simulate C-shaped core walls.
The divergence is about 10% for the first natural period. It seems that there is a difference in the calculation of the Torsional Stiffness of the system.
Any help would be of use!
Re: Simple 3D building model - Linear Transformations proble
UPDATE ---
At last , I managed to solve the problem and I feel I should share it with anyone who may be interested.
It seems that elastic BeamColumn element does not account for shear stiffness of members. I concluded this after a series of tests with stiffness properties modifiers on SAP2000 ( I set shear modifiers to 0 and compared the results with OpenSees).
Then, in order to take into account Shear stiffness for a linear elastic analysis in OpenSees, I used nonlinear Beam Column elements with Elastic Section and the response I obtained from Modal analysis matched exactly with the one from SAP. So I believe this is the correct way to implement it.
I thank everyone for your advice so far.
At last , I managed to solve the problem and I feel I should share it with anyone who may be interested.
It seems that elastic BeamColumn element does not account for shear stiffness of members. I concluded this after a series of tests with stiffness properties modifiers on SAP2000 ( I set shear modifiers to 0 and compared the results with OpenSees).
Then, in order to take into account Shear stiffness for a linear elastic analysis in OpenSees, I used nonlinear Beam Column elements with Elastic Section and the response I obtained from Modal analysis matched exactly with the one from SAP. So I believe this is the correct way to implement it.
I thank everyone for your advice so far.
Re: Simple 3D building model - Linear Transformations proble
Nikos, when you used elastic beam column elements (except from the natural period) did you have any convergence regarding the reaction forces of beams and columns between SAP2000 and OpenSees? And if not, did you have convergence when you used non linear elements with elastic sections?
Re: Simple 3D building model - Linear Transformations proble
Basa wrote:
> Nikos, when you used elastic beam column elements (except from the natural period)
> did you have any convergence regarding the reaction forces of beams and columns
> between SAP2000 and OpenSees? And if not, did you have convergence when you used
> non linear elements with elastic sections?
Yes, I definitely had good convergence for reaction forces, but using nonlinear elements with elastic sections, a better convergence was achieved (almost identical results).
> Nikos, when you used elastic beam column elements (except from the natural period)
> did you have any convergence regarding the reaction forces of beams and columns
> between SAP2000 and OpenSees? And if not, did you have convergence when you used
> non linear elements with elastic sections?
Yes, I definitely had good convergence for reaction forces, but using nonlinear elements with elastic sections, a better convergence was achieved (almost identical results).