internal hinge (moment free)

[javidsharifi]

Dear all,
How can I model an internal hinge? Supposing the intersection of two members of a frame or a structure does not bear any resisting moment and it is acting like a hinge, how may it be modeled in OpenSees? Or in the case of modeling trusses the same thing happens. When Modeling the nodes with the Node command, joints are modeled as rigid. Is there a command to remove the resisting moment from a node? Or in a like manner, to remove the resisting shear or axial resistance of a node in the middle or at an end of an element, there has to be a command or a way to model it. Can anyone please give me a hint? I'd appreciate it.
Regards,
Behtash from SUTech

[javidsharifi]

For a truss this:

element truss $eleTag $iNode $jNode $A $matTag <-rho $rho> <-cMass $cFlag> <-doRayleigh $rFlag>

is used. But still for a hinge among rigid connections e.g. in a frame I am wondering what to do!

[Ameri]

You should use EqualDOF command so as to define simple connection between elements. Have a glance to the following topics:

http://opensees.berkeley.edu/community/ ... =2&t=60432
http://opensees.berkeley.edu/community/ ... =2&t=60032

There are also several examples in OpenSeesWiki such as this one: http://opensees.berkeley.edu/wiki/index.php/SCBF_Model

[javidsharifi]

Thank u so much. I found the answer in topics u sent.
Best regards,
Behtash

[javidsharifi]

Dear Ameri,
I corrected the commands as explained in the conversations. But a problem still persists. The analysis fails at the very first step. The error appearing is this:

WARNING BandGeneralLapackSolver: : solve() -LAPACK routine returned 2
WARNING ModifiedNewton : : solveCurrentStep() -the LinearSys0fEqn failed in solve()
...
and the analysis fails at iteration 0. I tried different systems and integrators and algorithms but it dd not work. What would you recommend?
I appreciate your help.
Behtash

[fmk]

checking your model .. if solver fails at first step it typically means you do not have enough boundary cconditions

[javidsharifi]

It is a simple model! just a small truss. I double checked the supports. What may it be?! Seems something goes wrong with the linear algebra package. But I just cant get what is wrong!

[javidsharifi]

Seems there was sth wrong with truss elements! I did it with elasticbeamcolumn and nonlinearbemacolumn and worked!

[fmk]

there is nothing wrong with the truss elements .. just your use of them

for others here is a simple script for internal end release:
model Basic -ndm 2 -ndf 3
node 1 0. 0.
node 21 5.0 0.
node 22 5.0 0.
node 3 20.0 0.
fix 1 1 1 1
fix 3 1 1 1
geomTransf Linear 1
element elasticBeamColumn 1 1 21 100.0 1000.0 1000.0 1
element elasticBeamColumn 2 22 3 100.0 1000.0 1000.0 1
equalDOF 22 21 1 2
timeSeries Linear 1
pattern Plain 1 1 {
load 21 0 10 0
}
analysis Static
analyze 1
print ele 1 2

[javidsharifi]

I meant so too. I just couldn't have gotten truss elements to work so far. but I finally will!
Thanks for the script,
All the best
Behtash;
SUTech