Nodal DOF not Compatible With Quad Element!!

bayram_aygun · Post by **bayram_aygun** » Thu Sep 20, 2007 5:00 pm

Hello,

When I run the program below this warning message pops up: "Model Dimensions and\or nodal DOF are not compatible with quad element". I pasted the program below I think about this for two days now but I just can't get it because I'm inexperienced I guess. Can anybody help me out please:

wipe
set NumSteps 100
set StepSize [expr 1.0/$NumSteps]

#Build Model

model basic -ndm 2 -ndf 3

#Define Nodes

node 15 0 0
node 16 0 0
node 23 0 1
node 27 0 1
node 31 0 2
node 39 0 2
node 12 0 3
node 17 0 3
node 24 0 4
node 91 0 4
node 7 0 5
node 3 0 6
node 1 40 0
node 28 40 1
node 4 40 2
node 18 40 3
node 92 40 4
uniaxialMaterial Concrete02 1 -27580. -0.003 -5516. -0.01 0.1 3681. 1930544
uniaxialMaterial Steel02 2 460572.64 199949200. 0.01 18 0.925 0.15
geomTransf Linear 1
section Fiber 1 {

patch quad 1 4 16 -0.30 0.3 -0.3 -0.3 0.3 -0.3 0.3 -0.3
layer straight 2 16 0.00145 0.25 0.25 0.25 -0.25; # top layer rein
layer straight 2 16 0.00145 -0.25 0.25 -0.25 -0.25;
element nonlinearBeamColumn 17 24 12 5 1 1 #<-iter $maxIters $tol>
element nonlinearBeamColumn 39 12 31 5 1 1
element nonlinearBeamColumn 2 31 23 5 1 1
element nonlinearBeamColumn 54 23 15 5 1 1
element nonlinearBeamColumn 8 7 24 5 1 1
element nonlinearBeamColumn 5 3 7 5 1 1

nDMaterial PressureDependMultiYield 3 2 1.7000 55000. 150000. 29 0.1 80. 0.5 29 0.21 0. 0 10 0.02 1 0.85
nDMaterial PressureDependMultiYield 4 2 2.1000 130000. 390000. 40 0.1 80. 0.5 27 0.03 0.8 0 0 0 0 0.45

nDMaterial FluidSolidPorous 5 2 3 2.2e6
nDMaterial FluidSolidPorous 6 2 4 2.3e7

uniaxialMaterial PyLiq1 6 2 65. 0.085 0.07 0.0 6.5 11 12
uniaxialMaterial PyLiq1 7 2 80. 0.070 0.07 0.0 8. 13 14
element quad 11 1 16 27 28 1.0 "PlaneStrain" 5
element quad 12 28 27 39 4 1.0 "PlaneStrain" 5
element quad 13 4 39 17 18 1.0 "PlaneStrain" 6
element quad 14 18 17 91 92 1.0 "PlaneStrain" 6
element zeroLength 47 16 15 -mat 6 -dir 1
element zeroLength 41 27 23 -mat 6 -dir 1
element zeroLength 32 39 31 -mat 6 -dir 1
element zeroLength 29 17 12 -mat 7 -dir 1
element zeroLength 14 91 24 -mat 7 -dir 1

Thanks in advance,

silvia · Post by **silvia** » Fri Sep 21, 2007 6:29 am

there is a post on this one on this forum.

mebowers · Post by **mebowers** » Sat Sep 22, 2007 6:42 pm

The quad element is not compatible with a -ndm 2 -ndf 3 system. A quadUP element is compatible, with the third dof as pore water pressure. You may not need the quadUP element however, you might just want to use the quad element with the PDMY coupled with FluidSolidPorous in the -ndm 2 -ndf 2 space.

I have made a post about how to link soil nodes and pile nodes of different -ndm -ndf systems using the equalDOF slaving command. You may run into some issues when you connect your soil nodes to your nonlinear pile nodes as they may require different -ndm -ndf systems.

Matt Bowers
Clemson University

bayram_aygun · Post by **bayram_aygun** » Sun Sep 23, 2007 11:59 am

My soil-pile model is basically made of one pile and two layers of soil strata.
I created a 2DOF Mesh for soil and 3DOF Mesh for nonlinear pile. Then I used
"equalDOF" to connect pile to soil mesh. Since I use nonlinear p-y springs to
model SSI, my zerolength elements have two nodes: one in the 2DOF soil and the
other one 3DOF pile. The problem is OpenSees puts an error sign stating:
"element zerolength has two nodes which are differing in DOF". I probably have
to introduce some dummy nodes but I don't know how. Is there any way that you
can help me?

mebowers · Post by **mebowers** » Mon Sep 24, 2007 6:02 am

Yes, you are right, you need to utilize dummy nodes in your -ndm 2 -ndf 2 space. Put your dummy nodes at the same x,y location as your pile nodes but in the dof 2 space. Connect the zerolength elements from the soil nodes to the dummy nodes, then slave the pile nodes in 3dof to the dummy nodes in 2dof in the x and y directions.

Good luck!

Matt Bowers

bayram_aygun · Post by **bayram_aygun** » Tue Sep 25, 2007 9:49 am

First of all, thank you for your kind e-mail. I followed your instructions and I think that at least the soil-pile model is OK. I couldn't run the code because I can't make OpenSees read the transient load file(EQ-acceleration record in .txt format). I put the .tcl file+opensees.exe+acceleration .txt file in the C:\tcleditor\bin directory and I use these two command lines:

set ground_motion "Series -dt 0.005 -filePath EQ_Rix44.acn -factor 1.00"

pattern UniformExcitation 2 1 -accel $ground_motion

No matter what I did, I couldn't make OpenSees open the EQ_Rix44.acn file which is simply a .txt file.

Do you have any idea about what the problem might be?

Thanks in advance,