The OpenSees Community

vesna wrote:
> You can do this by using Hysteretic material:
> http://opensees.berkeley.edu/wiki/index ... c_Material.
>
> With this material you can define three points for each direction and
> construct the material you want.

Dear Vesna, thanks for quick replay. I tried numerous tests and so far had no success.
Let me clarify my problem. I would like to model a spring which will behave in this matter:
when force is Pmax the relative deformation is 0, as deformation is increasing the force should decrease as well.
The problem in Hyteretic mat. is that it can not be defined (as I see it) like (Pmax,0) for first point (0,6*Pmax, ey) for second etc.

I am trying to find a way to model a spring which will behave in the way such that when strain is zero, the force in the spring is maximal.
As the strain is increasing, the force should drop (for example linear). I really don't have an idea how to model this.

Any news about reliability part in OS?

Strange problem. I cant get force in spring. However, I get spring deformation and when calculate force it it ok so spring is active...

CODE:

model BasicBuilder -ndm 2 -ndf 3

set P 100
set l 121.9;
set h 91.44;
set E 20600;
set a1 1.33;
set a2 1.49;
set k 1;
set I 1 ;


node 1 0.0 0.
node 2 $l 0.0
node 3 $l $h
node 4 0 $h
node 5 0 $h


fix 1 1 1 0
fix 2 0 1 0

equalDOF 3 5 1 3

uniaxialMaterial Elastic 1 $E
puts materijal

geomTransf Linear 1

element elasticBeamColumn 1 1 2 $a1 $E $I 1
element elasticBeamColumn 2 2 3 $a2 $E $I 1
element elasticBeamColumn 3 3 4 $a2 $E $I 1
element elasticBeamColumn 4 5 1 $a1 $E $I 1
element elasticBeamColumn 5 4 2 $a1 $E $I 1
element elasticBeamColumn 6 1 3 $a1 $E $I 1

element zeroLength 7 4 5 -mat 1 -dir 2

pattern Plain 1 Linear {
load 4 $P 0.0 0.0

}


recorder Element -file 7.txt -time -ele 7 localForce
recorder Element -file 71.txt -load -ele 7 deformation
recorder Node -file D.txt -time -node 3 5 -dof 1 2 3 disp

constraints Plain
numberer RCM
test NormUnbalance 1.0e-2 250 0
integrator LoadControl 0.025 1 0.025 0.025
algorithm Newton
system ProfileSPD
analysis Static

integrator LoadControl 0.05 3 0.05 0.05
analyze 16

integrator LoadControl 0.02 3 0.02 0.02
analyze 10

print ele 1 7

I am trying to model spring between the element 6 and element 3 in my model. I added joint 5 and via equalDOF connected it with the joint 3. Then made zerolength element. The spring should carry axial load from element 6. There are several things that bother me.

1. I assigned element zeroLength 7 5 3 -mat 1 -dir 2 but when I print info for element 7 I get dir 1. Am I missing something or there should be dir 2?

2. By looking at the joints 3 and 5 there is difference in vertical displacements which obviously should be, but there is no axial force in element 7???

Here is the code:
---------------------------------------
model BasicBuilder -ndm 2 -ndf 3

set P 100
set l 121.9;
set h 91.44;
set E 20600;
set a1 1.33;
set a2 1.49;
set k 1;
set I 10 ;


node 1 0.0 0.
node 2 $l 0.0
node 3 $l $h
node 4 0 $h
node 5 $l $h

fix 1 1 1 1
fix 2 0 1 1

equalDOF 3 5 1 3

uniaxialMaterial Elastic 1 $E
puts materijal

geomTransf Linear 1

element elasticBeamColumn 1 1 2 $a1 $E $I 1
element elasticBeamColumn 2 2 3 $a2 $E $I 1
element elasticBeamColumn 3 3 4 $a2 $E $I 1
element elasticBeamColumn 4 4 1 $a1 $E $I 1
element elasticBeamColumn 5 4 2 $a1 $E $I 1
element elasticBeamColumn 6 1 5 $a1 $E $I 1

element zeroLength 7 5 3 -mat 1 -dir 2
#orient 0 1 0 -1 0 0

pattern Plain 1 Linear {
load 4 $P 0.0 0.0
#load 4 $G 0.0
}


recorder Element -file 1.txt -load -ele 1 localForce
recorder Element -file 2.txt -load -ele 2 localForce
recorder Element -file 3.txt -load -ele 3 localForce
recorder Element -file 4.txt -load -ele 4 localForce
recorder Element -file 5.txt -load -ele 5 localForce
recorder Element -file 6.txt -load -ele 6 localForce
recorder Element -file 7.txt -load -ele 7 axialForce
recorder Node -file D.txt -time -node 3 5 -dof 1 2 3 disp

constraints Plain
numberer RCM
test NormUnbalance 1.0e-2 250 0
integrator LoadControl 0.025 1 0.025 0.025
algorithm Newton
system ProfileSPD
analysis Static

integrator LoadControl 0.05 3 0.05 0.05
analyze 16
integrator LoadControl 0.02 3 0.02 0.02
analyze 10

print ele 1 7

Working perfectly. Thanks very much.

thanks for replay, I understand what I have done.

...but I am trying to model a kind of bearing in the middle of the beam. so when the axial stifness of the sping is very large it will become support which will "carry" vertical reaction.

How to model this with the zerolength element?

Hi. I am trying to model simple spring in order to verify more complex models. However I cant get a force in the spring. I looked at the file 10x times and can't figure this.

So please if anyone can help!


model BasicBuilder -ndm 2 -ndf 3
set L 200
set A 20
set E 21000
set I 167

node 1 0.0 0.0
node 2 $L 0.0
node 3 $L 0.0
node 4 [expr 2*$L] 0.0

fix 1 1 1 0
#fix 2 1 1 1
fix 4 0 1 0

equalDOF 2 3 1 3

# Define force-deformation relationship for spring

uniaxialMaterial Elastic 1 1050

# id ndI ndJ mat dir

element zeroLength 1 2 3 -mat 1 -dir 2

# Geometric transformation

geomTransf Linear 1

# id ndI ndJ A E I transf

element elasticBeamColumn 2 1 2 $A $E $I 1
element elasticBeamColumn 3 3 4 $A $E $I 1

pattern Plain 1 Linear {

eleLoad -ele 2 -type -beamUniform -5
eleLoad -ele 3 -type -beamUniform -5

}

#recorder Node Z.out disp -time -node 2 -dof 2
recorder Node -file p.txt -time -node 2 3 -dof 2 disp
recorder Node -file px.txt -time -node 3 -dof 1 disp
recorder Element -file ele1.txt -time -ele 1 localForce
recorder Element -file ele2.txt -time -ele 2 3 localForce

constraints Plain
numberer RCM
test NormUnbalance 1.0e-2 250 0
integrator LoadControl 0.025 1 0.025 0.025
algorithm Newton
system ProfileSPD
analysis Static

integrator LoadControl 0.05 3 0.05 0.05
analyze 16

integrator LoadControl 0.02 3 0.02 0.02
analyze 10

Thanks for your replay Hugo. But please can you be more precise?

Suppose I want to model cantilever beam axially loaded. I increase load until the strains (or stresses) reach some point when beam fails.

How to model this?

I tried with hysteretic material without any success. It only extrapolates the stress-strain relationship given by 3 points. However I didn't try pinching and damage options but I doubt that it will do the trick.


So please help.

it is theoretical model. I intend to take n truss elements. Increase load until first element fails, then the redistribution of the forces will take place. After that I have n-1 element and increase force again (if possible).

This won't be static calculation as I am interested in reliability of this system. But when I know how to model this static system problem is solved.

I will take a look at hysteretic material.

thanks Silvia.
so this settles the geometry problem.

but what material to use?
Hysteretic?
how to model failure of an element?

I would like to know if it is possible to make parallel model of n truss elements. I would apply force (and increase) it until first element fails and force redistribution takes place....

Does anybody has idea how to model this?

I am wondering is there a manual or some examples for new command syntax in the reliability part of the opensees? If you take new version and try to run Terje's files you can notice that syntax has changed....

I am wondering is there a manual for new commands in the reliability part of the opensees? If you take new version and try to run Terje's file you can notice that syntax has changed....