Discussion:
Question about Submodeling
Alberto B.
2003-06-09 23:02:58 UTC
Permalink
Dear all,
I am trying to perform a submodelling analysis and I get an error mesage on the .dat file when I run the submodel job.
The global model analysis worked fine, but after that, when I run the submodel analysis,
I get the following :

***ERROR : ERROR - * SUBMODEL CARD IS MISSING

that message it's right after the boundary dataline of the .dat file so I thought I was mistaking something in the command line...
but I checked the manual and all the messages posted here without being able to figure out what it's causing that problem.
I tried it both with the 6.2 and the 6.3 version, but I got once more that same error message.
I also thought that something concerning Abaqus' installation here at the supercomputer center of my university could be the reason of that,
but no one of the guys of the suoercomputer center was able to answer my question.
Is there an additional submodeling package that has to be installed with Abaqus Standard in order to perform a submodeling analysis??

Please is there anyone who can help me with this or who might have a sort of idea about what is causing that problem?

Many thanks in advance for taking time in reading this and perhaps in answering me.


Yours

Alberto Bassanese

Department of Aerospace Enginerering
Texas A&M University
College Station
Texas


[Non-text portions of this message have been removed]



------------------------ Yahoo! Groups Sponsor ---------------------~-->
Get A Free Psychic Reading! Your Online Answer To Life's Important Questions.
http://us.click.yahoo.com/Lj3uPC/Me7FAA/ySSFAA/PMYolB/TM
---------------------------------------------------------------------~->
JMZ
2003-06-10 19:04:35 UTC
Permalink
Probably, you forgot the keyword '*submodel' in the *.inp file (before the
keyword '*step') by which you have to specify
the driven nodes, previously defined by '*nset' keyword.
Ciao.
Jacopo Zappettini
Alberto B.
2003-06-11 22:22:19 UTC
Permalink
Dear Massimo Zappettini and Dear All,

Many thanks for answering me,
I left "**SUBMODEL" instead of "*SUBMODEL" in my input file.....a trivial
mistake...
at lesat more trivial than I aspected to be...
I would like to ask you something else about submodeling:
my global model consists in two parts connected to each other through the
use of the "tie" costrain ( so one surface of the first part is connected to
one surface of the second part at the assembly level ); the global model
(assembly level) has a hole that crosses throughout both the parts.
I am interested in the stress-strain field around the hole so my submodel
represents a portion of the global model near the hole and it consists in
two parts as well as the global model.
I am not confident about how to manage the link between the two parts in the
submodel.
I already defined as "driven nodes" all the nodes belonging to the external
surface of my submodel so as to apply the right boundary conditions taken
from the global analysis to the submodel and now I don't know how to do to
define the right link beteewn the two parts:
at the beginning I used the "tie" costrain also for the submodel (as I did
for the global one) , but Abaqus reconizes my submodel as overcostrained,
so I am considering the opportunity of removing the tie costrain and then
adding all the nodes belonging to the surfaces that I tied before to the
node set of the "driven nodes" ....so as to let Abaqus drive them with
results of the global analysis...
do you think that if I do so Abaqus will be able to assign the right
boundary condition to both of those surfaces that I need to have "tied"
together ?

Please two last very short points:

-I am dealing with a linear elastic analysis, without time dependences so I
think that the ZERO INCREMENT parameter it's not important ..., do you agree
with me or do you think that there might be some problems without that
parameter?
-to perform a submodel analysis Abaqus requires me to name the step both in
the global and the local analysis as numbers,
so for as far as you know do you think that both the name of the steps (in
the global and in the sub-model) have to be the same if I just want to
perform the same analysis with the submodel?

Many thanks in advance for taking time in reading this and perhaps in
answering me.

Yours

Alberto Bassanese

Department of Aerospace Enginerering
Texas A&M University
College Station
Texas




------------------------ Yahoo! Groups Sponsor ---------------------~-->
Get A Free Psychic Reading! Your Online Answer To Life's Important Questions.
http://us.click.yahoo.com/Lj3uPC/Me7FAA/ySSFAA/PMYolB/TM
---------------------------------------------------------------------~->
Loading...