Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations IDS on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Crack Propagation with ABAQUS DEBOND

Status
Not open for further replies.

yve

Materials
Apr 2, 2007
2
Hello,

i try to find out the posibilities using ABAQUS *DEBOND function to calculate crack propagation of a CT-Speciem.

I defined an amplitude for the load to simulate one loadcycle in a step (both same time)in the first step my Model debonds as expected (3 Nodes). And in the following it doesnt, i can see that the fracture criterion at the cracktip node (in defined distance tothe node)is fullfilled but nothing happens. It is an linear elastic calculation. It doesnt make a difference if i take one cycle per step or if i define more cycles in my *Amplitude and calculate them in one step. My Question now is: can I use the debond funktion with a cyclic load? I havent found anything in the manual like this.

Anybody an idea? my file is below. All nodes are in the BOND set, every second node is defined as a crack tip i used CPE8 Elements, as I am interested in stress gradients too, i took quadratic elements. I know linear elements are recomended but noone said quadratic elements doesnt work.

Thanks
Yve


**----------------------------------------------------------
**Define Amplitude for Load Cycle 1-1-1-1 from 0 to 100%
**----------------------------------------------------------
*Amplitude, name=Cycle1-1-1-1
0., 0., 1., 900., 2., 900., 3., 0.
4., 0.
**----------------------------------------------------------
** MATERIALS
**----------------------------------------------------------
*Material, name="linear-elastic"
*Elastic
210000., 0.30
**----------------------------------------------------------
** CONTACT DEFINITION
**---------------------------------------------------------
**
** INTERACTION PROPERTIES
**---------------------------------------------
** Contact between LoadPin and Hole
**---------------------------------------------
*Surface Interaction, name=ContactPinHoleProperty
1.,
*Friction
0.,
*Surface Behavior,pressure-overclosure=HARD
**
*Contact Pair, interaction=ContactPinHoleProperty,small sliding, adjust=0.001, no thickness
CT-Speciem-2D-Shell-1.Surf-1, Pin-2Danalyticalrigid-1.Surf-1
**
**---------------------------------------------
**Contact Definition for Debonding Ligament
**---------------------------------------------
*Surface Interaction, name=LigamentPlusContact
1.,
**
*Contact Pair, interaction=LigamentPlusContact, small sliding
CT-Speciem-2D-Shell-1.Ligament, LigamentRigidSurface-1.RigidSurface_
**
** ---------------------------------------------------------
** Initial Condition for Debond
**---------------------------------------------------------
**
*Initial Conditions, Type=Contact, Normal
CT-Speciem-2D-Shell-1.Ligament, LigamentRigidSurface-1.RigidSurface_, CT-Speciem-2D-Shell-1.BOND
**
**---------------------------------------------------------
**Berechnungsschritt
** STEP: Step-1
**----------------------------------------------------------
**
** STEP: Zyklus 1 Last 1-1-1-1
**
*Step, name="Zyklus 1 Last 1-1-1-1", Inc=400
Loadzykle 1-1-1-1 Max Load on Pin 900N/mm²
*Static
0.01, 4., 1e-09, 4.
**
**---------------------------------------------
** BOUNDARY CONDITIONS
**---------------------------------------------
** Name: LigamentRigidSurfaceFixdisplacement Type: Displacement/Rotation
*Boundary
RFnodeLigamentrigidsurf, 1, 1
RFnodeLigamentrigidsurf, 2, 2
RFnodeLigamentrigidsurf, 6, 6
** Pin-2Danalyticalrigid-RefPt_
** Name: Pin-2Danalyticalrigid-RefPt_ Type: Displacement/Rotation
*Boundary
RFnodePin, 1, 1
RFnodePin, 6, 6
**
**---------------------------------------------
** LOADS
**---------------------------------------------
**
** Name: cyclic load on RP of Pin Rigid Type: Concentrated force
*Cload, amplitude=Cycle1-1-1-1
RFnodePin, 2, 1.
**
**----------------------
**DEBOND OPTION
**----------------------
**
*Debond, Slave=CT-Speciem-2D-Shell-1.Ligament, Master=LigamentRigidSurface-1.RigidSurface_, Time Inc=1.E-07, Frequency=1,Output=BOTH
0.,1.
1.E-7,0.,
*Fracture Criterion, Type=Critical Stress, Tolerance=0.01, Distance=0.01
1300.,1.E12,1.E12
**
**------------------------
** OUTPUT REQUESTS
**------------------------
*Energy Print
*Contact Print,Master=LigamentRigidSurface-1.RigidSurface_
CSTRESS, CDISP
DBT,DBSF,DBS
**----------------------------
** HISTORY OUTPUT: H-Output-1
**----------------------------
*Output, history, variable=PRESELECT
*Contour integral, crack name=H-Output-2_Crack-1, contours=10, symm, Output=BOTH
CT-Speciem-2D-Shell-1.CrackTip1, -1.,0.,
*Contour integral, crack name=H-Output-3_Crack-2, contours=10, symm, Output=BOTH
CT-Speciem-2D-Shell-1.CrackTip2, -1.,0.,
*Contour integral, crack name=H-Output-4_Crack-3, contours=10, symm, Output=BOTH
CT-Speciem-2D-Shell-1.CrackTip3, -1.,0.,
*Contour integral, crack name=H-Output-5_Crack-4, contours=10, symm, Output=BOTH
CT-Speciem-2D-Shell-1.CrackTip4, -1.,0.,
**--------------------
** FIELD OUTPUT: F-Output-1
**--------------------
*Output, field
*Contact Output, VARIABLE=ALL
CSTRESS, CDISP
DBT,DBSF,DBS
*Output, field
*Element Output,Elset=CT-Speciem-2D-Shell-1.ALL
S,E
*Node Output, Nset=CT-Speciem-2D-Shell-1.ALL
U
*End Step
 
Replies continue below

Recommended for you

You can try.


Regarding :
"It doesnt make a difference if i take one cycle per step or if i define more cycles in my *Amplitude and calculate them in one step."

It can make a difference if you use automatic time incrementation and you have sudden changes in cyclic load profile.

If I have kinks in the cyclic load profile (non-smooth loading cycle), I decompose the cyclic load in several subsequent steps. Thus, I know exactly the value of the applied load at the kink in the loading cycle since ABAQUS will perform a computation at the end of the step.

Also, in this way I can make the analysis more efficient when some portions of the loading cycle lead to a more non-linear response of the structure (e.g. crack faces come in contact, plasticity etc.) which require smaller time increments.

 
Thank you xerf, ill try this.

 
Status
Not open for further replies.

Part and Inventory Search

Sponsor