zero-length:2 different nodes at the same coordinate

Jfoyian · Post by **Jfoyian** » Tue Apr 28, 2009 9:57 am

I'm trying to assign 2 different nodes at the same coordinate (as is done on the examples manual), but the program doesn't allow. (doesnt run) I was trying to put zero-length elements in these nodes. Does anyone have an idea why I can't define the 2 nodes at the same coordinate?

silvia · Post by **silvia** » Tue Apr 28, 2009 11:39 am

do they have different node tags?
please post the part of the script that is of relevance.
what error does os give you???

Jfoyian · Post by **Jfoyian** » Tue Apr 28, 2009 4:30 pm

yes, the node tags are different as follows;

node 101 0 0;
node 201 0 $h1;
node 1001 0 0;
node 2001 0 $h1;

uniaxialMaterial Hysteretic $IDshear12 1342 0.04 2721 0.624 2721 1.7 -1300 -0.04 -2700 -0.6 -2700 -1.5 0.5 0.5 0 0 0;

element zeroLength 2001 201 2001 -mat $Secshear12 -dir 2

Actually it doesnt give any error, but doesnt converge. Even if I dont include the shear spring and just run the model itself with the previously mentioned nodes (101, 1001, 201, 2001) it doesnt converge. So I guess the addition of the nodes 1001 and 2001 confuses the program. How I can solve this?

silvia · Post by **silvia** » Tue Apr 28, 2009 6:59 pm

likely, there are not enough constraints in your system.

Jfoyian · Post by **Jfoyian** » Tue Apr 28, 2009 7:36 pm

Thanks Silvia but I am not sure I get it clearly. Should I be using fix command to assign fixity to the new nodes as well or do you mean to use equalDOF or any other command? Could you explain in more detail?

msa_imeg · Post by **msa_imeg** » Tue Apr 28, 2009 11:36 pm

wipe;
model BasicBuilder -ndm 2 -ndf 3
node 1 0 0
node 2 0 0
fix 1 1 1 1
fix 2 0 1 1
mass 2 5.0e3 0. 0.
uniaxialMaterial Hardening 1 $Es $Fy $Hiso $Hkin
element zeroLength 1 1 2 -mat 1 -dir 1
recorder Node -file node_disp.txt -time -node 2 -dof 1 disp

set accelSeries "Series -dt 0.01 -filePath BM68elc.acc -factor 40"; # define acceleration vector from file (dt=0.01 is associated with the input file gm)
pattern UniformExcitation 1 1 -accel $accelSeries; # define where and how (pattern tag, dof) acceleration is applied
#D = $alphaM * M + $betaK * Kcurrent +$betaKinit * Kinit + $betaKcomm * KlastCommit
rayleigh 0.0 0.0 0.0 0.00

# create the analysis
wipeAnalysis; # clear previously-define analysis parameters
constraints Plain; # how it handles boundary conditions
numberer Plain; # renumber dofs to minimize band-width (optimization), if you want to
system BandGeneral; # how to store and solve the system of equations in the analysis
algorithm Linear; # use Newtons solution algorithm: updates tangent stiffness at every iteration
integrator Newmark 0.5 0.25 ; # determine the next time step for an analysis
analysis Transient; # define type of analysis: time-dependent
analyze 4000 0.01;

silvia · Post by **silvia** » Wed Apr 29, 2009 5:15 am

don't know what the post above is,
but you should be using whatever constraints you need in your model, whether it is the fix command or the equaldof.
you just have to make sure you don't have a singular stiffness matrix.

Jfoyian · Post by **Jfoyian** » Wed Apr 29, 2009 10:42 am

Thanks Silvia!
Now it works after using fix command for that nodes, although I still cannot get what I want. It still does not understand shear behaviour of my spring. Is it a good idea to use zerolenght element to represent shear behaviour? And as a general question, since I have faced that problem many times, how can we prevent singular stiffness matrix?

silvia · Post by **silvia** » Wed Apr 29, 2009 12:14 pm

isn't direction 2 vertical????