Lusas warning help

[standingdouble]

Hi!
I have started to use Lusas, in order to do a modal analysis of vehicle-railway interaction.
Since I'm a beginner I started with an easy test: a spring-damping-mass system.I modelled the viscoelastic joint creating a line (of unit lenght) between 2 nodes and choosing "joint no rotational stiffness" as mesh, furthermore in "attributes-material-joint" I set the parameters as mass (at second or first node), spring and damping (ratio relative to critical damping).

When I launch the solver I get this warning:
"the nodes 1 and 2 for element number 1 are not coincident"

I would know what does it means, maybe is there some errors in my model?
How should I model the viscoelastic link?

thank you

 

[GregLocock]

Not being familar with LUSAS, hard to say, but I bet that the deck for that model is less than 50 lines, if you post it here then I'll have a look.

Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[rb1957]

i suspect that the viscoelastic link element is meant to be zero length, the user manual (such as it probably is) may help; maybe the folks at London University could help (but then, maybe not !?)

good luck; oh, and congratulations on approaching FE from a sensible direction (very cautiously)

 

[standingdouble]

-rb1957
you may be right with zero lenght of link element, but I can't set it properly in the mesh properties...Maybe I would ask to Lusas support
-GregLocock
this is the part of the output of the analisys,I removed some useless lines:

E L E M E N T T O P O L O G Y

ELEMENT ELEMENT NODE NUMBERS
NUMBER
JNT3 ELEMENTS
1 1 2 4

N O D E C O O R D I N A T E S

NODE NO. COORDINATE IN COORDINATE IN
X-DIRECTION Y-DIRECTION
M M
1 0.000000E+00 0.000000E+00
2 0.000000E+00 1.00000
4 0.000000E+00 2.00000
-I don't know why here is also node 4,I added only node 1 and 2
TOTAL NUMBER OF NODES = 3
LARGEST NODE NUMBER = 4

J O I N T P R O P E R T I E S G E N E R A L
MATERIAL
1 K1 = 0.2000E+08 M1 = 1600. C1 = 0.0000E+00 00
K2 = 0.2000E+08 M2 = 1600. C2 = 0.0000E+00
MCODE = 2

-here were information about support nodes: node 1 is restrained (without any spring)

***WARNING*** NODES 1 AND 2 FOR ELEMENT NUMBER 1 ARE NOT COINCIDENT (KJNT4 PROCESSOR)

thank you all

 

[crisb]

Its a long time since I used Lusas but can remember getting very frustrated with their error/warning messages.

Is it possible that you have tried to mesh the member (rather than the joint) as a joint element?

 

[johnhors]

standingdouble,

I use the Lusas solver, but not their pre and post processor.

Here is what the online help has to say about JNT3 :-


JNT3 - A 2D joint element which connects two nodes by two springs in the local x and y-directions.
Number Of Nodes = 3.
The 3rd node is used to define the local x-direction.
Freedoms U, V: at nodes 1 and 2 (active nodes).
Node Coordinates X, Y: at each node.


which has some important notes as well:-


The nodes of a joint element need not be coincident, but for correct response the distance between them should be as small as possible. This is particularly important with joint elements which contain rotational degrees of freedom, since the stiffness matrix is not formulated using engineering beam theory. This means that a joint moment is independent of both shear force and its length. For instance, the moment calculated with a joint length of zero will remain the same magnitude at any other joint length.

These effects can be exacerbated significantly in dynamic analyses (e.g. eigenvalue extraction or Hilber dynamics) and, therefore, we recommend that joint elements representing lumped masses are placed between active nodes of the structure. In this way a zero stiffness may be used for all the degrees of freedom of the joint - the mass applied to whichever freedom is appropriate. Non-coincident nodes will lead to additional forces in the solution which are not in equilibrium (usually small and swamped, but could be significant sometimes). We do not recommend having joints “hanging off” the side of the structure, having a large stiffness associated.


I think that says it all !

 

[standingdouble]

thanks johnhors,
ops,the answer is simply on the manual...
Actually I focused my attention on the modal analysis section and didn't read well the joint section.

now I understand why there are 3 nodes in the output.
Furthermore today I made a comparison with some analytical results (from Mathcad) for the same easy example and I realized they are identical.
Then,in spite of the warning, results seem to be right.

once more thanks to all.