Contact in Abaqus Explicit

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

Contact in Abaqus Explicit

Abaqus Users mailing list




Dear Group,






I kindly ask for assistance regarding contact.




I am running a simulation in ABAQUS 6.13-1 Explicit. Three cylinders,
all composed of elements CAX4R (the outer cylinders) and CAX4RT (the
central cylinder), have contact at their flat interfaces from the very
beginning and are compressed along their common axis.




The surfaces are defined like:
*Elset, elset=InputRear, internal, instance=Inputbar, generate
22471,  22480,      1
*Surface, type=ELEMENT, name=InputRear
InputRear, S3




Contact is defined by:


*Contact Pair, interaction=IntProp, mechanical constraint=kinematic,
cpset=Int-1
InputRear,SpecimenFront
*Contact Pair, interaction=IntProp, mechanical constraint=kinematic,
cpset=Int-2
SpecimenRear,OutputFront




The results of the simulation are in harmony with the experiment.






The sta-file reports:




1)
***WARNING: Some elements underlying deformable surface
ASSEMBLY_SPECIMENFRONT
             are referenced by a *CONTACT PAIR option. In addition these
             elements have material properties that allow the elements
to fail.
             For elements that can fail, contact should be defined using the
             *CONTACT option. A element set named
"WarnElemFailElemInContact"
             containing these elements is created to help the user
locate the
             elements.


and equivalently for the other pair of interfaces, and:


2)
***WARNING: Each of the nodes listed below (and included in node set
             "WarnNodeBcIntersectKinCon") participates in a boundary
condition
             definition (*BOUNDARY) as well as being part of a kinematic
             contact definition (*CONTACT PAIR, MECHANICAL CONSTRAINT =
             KINEMATIC). If a degree-of-freedom participates in both
types of
             constraints, the boundary condition will override the kinematic
             contact constraint. In case of a conflict, it is suggested that
             the penalty contact algorithm be used instead.


Intersecting Nodes:
node: 24729     instance: INPUTBAR
node: 1     instance: OUTPUTBAR
node: 1     instance: SPECIMEN
node: 177     instance: SPECIMEN


to 1):
I had the same warning in three-dimensional simulations and got rid of
it by modifying contact to
*Contact inclusions
InputRear,SpecimenFront
SpecimenRear,OutputFront


But, see User's Guide chapter 36.4.1 Defining general contact
interactions in Abaqus/Explicit:
Abaqus/Explicit provides two algorithms for modeling contact and
interaction problems: the general contact algorithm and the contact pair
algorithm.
The general contact algorithm in Abaqus/Explicit:
can be used only with three-dimensional surfaces


and 'contact pair' produces the first warning. How can this inherent
contradiction be solved ?




to 2):
I changed the initial definitions of contact to 'mechanical
constraint=penalty', and this succeeded in avoiding the second warning.
But now the results of the simulation for the stress in the central
cylinder are poor.


Is there a more appropriate method to avoid the second warning ?




Thank you for sharing your expertise


Frank




Reply | Threaded
Open this post in threaded view
|

Python script is mixing up data

Abaqus Users mailing list


Dear Group,


I kindly ask for assistance as my Python script is mixing up data.


Below is an excerpt from one of my Python scripts to access data from
the odb file, for 200 frames of a simulation.


I have defined parameters varL and outname to keep the script concise,
as of now only for strain E33 and S33 along the specimen axis, possibly
for more data later on.


Observation: Python is mixing up data, meaning the stress data end up in
files for the strain. The files for the stress are correct. This also
occurs when splitting this code, i.e. all E33 first, then all S33.


The problem is clearly in the extraction of the data by Python.


How can this script be made to open, write and close a file before this
process is initiated for the following frame and output quantity ? It
does not help to spell out the lines in the frames-loop for strain and
stress separately.


I observe the same phenomenon with other code sections when placed
inside a loop over all individual odb-files from a parametric study,
looping over the index of the simulations. The rpt-files written for all
simulations in this loop are all alike.


I am accessing an odb-file from ABAQUS 6.13-1 Explicit, using
ABAQUS/Viewer which comes with Python release 2.6.2.


Thank you for sharing your expertise


Frank


-------------------------------------------


# define a path along the axis of the specimen
session.Path(name='SpecimenAxis', type=NODE_LIST,
expression=(('SPECIMEN', (    '5:40005:1000', )), ))
pth = session.paths['SpecimenAxis']
for out in range(0,2):
     if out==0:
        varL = 'E'
        outname = 'E33'
     if out==1:
        varL = 'S'
        outname = 'S33'
     print out
     print varL
     print outname
     session.viewports['Viewport: 1'].odbDisplay.setPrimaryVariable(
         variableLabel=varL, outputPosition=INTEGRATION_POINT,
refinement=(COMPONENT,
         outname))
     for fr in range(0,201,1):
         session.viewports['Viewport: 1'].odbDisplay.setFrame(step=0,
frame=fr)
         session.XYDataFromPath(name='SpecimenAxis-' + outname +
'-frame=' + str(fr), path=pth,
            includeIntersections=False, pathStyle=PATH_POINTS,
numIntervals=10,
            shape=DEFORMED, labelType=TRUE_DISTANCE)
         x0 = session.xyDataObjects['SpecimenAxis-' + outname +
'-frame=' + str(fr)]
         fn = td + stem + str(V) + '_FEM_' + vs + str(fem) +
'-SpecimenAxis-' + outname + '-frame=' + str(fr) + '.rpt'
         session.writeXYReport(fileName=fn, appendMode=OFF, xyData=(x0, ))