Skip to main content

[SOLVED] 3D crack growth modelling in Abaqus by XFEM

Submitted by Yazri Yaakob on

Good day everyone,

I'm new to iMechanica and look forward to getting to know everyone here.

I'm currently doing analysis of interlaminar crack growth in fibre-reinforced composite by Extended Finite Element Method (XFEM) using Abaqus. I'm a new Abaqus user and therefore I have to familiarise myself by constructing random 2D and 3D models with isotropic materials before jumping onto anisotropic.

Having gone through a few tutorials by Matthew Pais (thank you), I've managed to successfully model a 2D crack propagation model. The crack seems to propagate like I intended it to and it's all good. But applying the same approach to construct a 3D model, the crack remains stationary even with the increase in load magnitude. I've also tried lowering down the Max Principal stress as well as Fracture Energy under Traction-Separation setting but to no avail. I always ended up with the error "too many attempts made for this increment".

Note that I'm using 100k as my increment number with initial 0.01, minimum 1E-09 and maximum 0.01.

So my questions are, what causes the error and how do I fix it? Are there any additional steps required for 3D crack growth propagation compared to 2D? Apart from the magnitude of the load and the Max Principal stress, are there any other parameters governing crack initiation and propagation in Abaqus?

I could provide the CAE or INP files if required. Any help is very much appreciated. Thanks in advance.

 

Regards,

Yazri

Forums
Fracture Mechanics Forum
Free Tags
enrichment
crack growth
3D
XFEM crack propagation
XFEM in ABAQUS
extended finite element method
Zheng Jia

Hi,Yazri

 Try to increase the number of attempts permitted for each increment, IA, to 10 or more, beacuse the default value of IA is only 5. This can be achieved by using "controls" command in the input file.

If you can tell your email to me, I can send you a PDF file about how to improve the convergence in XFEM. Hopefully it can help you.

 Best,

Zheng

 

Mon, 06/14/2010 - 17:06 Permalink
havelmay

Hello Zheng,

I have also convergence problems with using XFEM in Abaqus that are quite similar to the ones of the questioner. Can you send your offered PDF for improving the convergence in XFEM?

My email is: Thomas.Havelka [at] gmx.de

 

Greets,

 

Thomas

Thu, 09/18/2014 - 14:58 Permalink
Renan Junqueira

I am facing convergence problems on abaqus using XFEM (with traction separation law property for the material) and Concrete Smeared Cracking (with traditional FEM). Both studies does not converge, this PDF you said is about improvement of the convergence only for XFEM, or is it more general? Could you please send this PDF file to the e-mail address: renanjunqueira [at] hotmail.com ? It will be very helpful for my master science dissertation.

Thank you.

 

Sat, 09/26/2015 - 22:09 Permalink
Yazri Yaakob

Zheng Jia,

Thank you for taking time helping me with the problem. I'm not certain whether or not iMechanica allows one to post his/her email on the board itself so I sent you a message. Please do send me the PDF file as I'm sure it will be of a great help.

To be honest, I'm not sure how to increase the IA value but I'll look it up in the Documentation first. If I can't find anything useful I hope you won't mind pointing me in the right direction.

 

Thanks very much,

Yazri

Tue, 06/15/2010 - 09:11 Permalink
Jez

Hi Yazri,

 I am also new to XFEM in Abaqus and i have the exact same problem while trying to run a 3D crack simulation. Have you managed to rectify the problem?

 Zheng, would you be able to send that pdf file on how to improve the convergence in XFEM aswell? You can contact me with lui [at] fev.de (lui[at]fev[dot]de)

 

Jez

Wed, 06/23/2010 - 13:33 Permalink
Josu

Hi Yazri,

 I am also new to XFEM in Abaqus and i have the exact same problem. Have you managed to rectify the
problem?

 Zheng, would you be able to send that pdf file on how to improve the convergence in XFEM aswell? You can contact me with lui [at] fev.de (imotz[at]hotmail[dot]com)

 

Josu

Tue, 02/01/2011 - 09:17 Permalink
Nguyen Van Chau

 I am also new to XFEM in Abaqus and i have the exact same problem. Have you managed to rectify the

problem?

 Zheng, would you be able to send that pdf file on how to improve the convergence in XFEM aswell? You can contact me with chaunguyen9911 [at] gmail.com (chaunguyen9911[at]gmail[dot]com)

 chau

Sun, 06/12/2011 - 01:28 Permalink
Himayat Ullah

Hi  Zheng & Yazri

 I am also new to XFEM in Abaqus and i am facing exactly the same problem. 

 Zheng, would you be able to send that pdf file on how to improve the convergence in XFEM ? You can contact me at uhimayat [at] gmail.com (uhimayat[at]gmail[dot]com)

Himayat

Sun, 06/12/2011 - 13:16 Permalink
mohammedyahya

Hi  Zheng & Yazri

 I am also new to XFEM in Abaqus and i am facing exactly the same problem. 

 Zheng, would you be able to send that pdf file on how to improve the convergence in XFEM ? You can contact me at uhimayat [at] gmail.com (mohammed20074[at]yahoo[dot]com)

mohammed

Wed, 02/08/2012 - 12:03 Permalink
mohammedyahya

Hi  Zheng & Yazri

 I am also new to XFEM in Abaqus and i am facing exactly the same problem. 

 Zheng, would you be able to send that pdf file on how to improve the convergence in XFEM ? You can contact me at uhimayat [at] gmail.com (mohammed20074[at]yahoo[dot]com)

mohammed

Wed, 02/08/2012 - 12:02 Permalink
Honggang Jia

I am a new xfem abaqus , I thank  you try  define the type of growth as discrete crack propagation along arbitrary , solution-dependent path, and if you define the crack propagation direction , you can try .this is my own idea .

Fri, 12/16/2011 - 01:40 Permalink
b_m_

Hello 

Yazri

i am new to ABAQUS.doing project in XEMF.cracack propagation in pipe.i am nt getting what to do how to do....

Time is really against me.. Plz help me out..

if u can send me tutorials or PDF on my mail (bhatemayuri [at] gmail.com)



Please help me out..

Thanking you..

REGARDZ

B.Mayuri 

Wed, 12/28/2011 - 06:46 Permalink
Sarah A. Eassa

hello everyone

when i read  ur problems i feel that mine  is silly.i made a 2D crack propagation model using the XFEM. but the crack didn`t propagate .only there was a separation between the two surfaces the crac lie between. can anyone plz help me in knowing why this happened. and how can i have the crack tip propagates?

thanks

Mon, 03/05/2012 - 23:39 Permalink
ziziwei

could abaqus solve the crack problem very well by XFEM? is there any drawbacks?

i saw a lot of research groups have their own codes, and make some improvements. so i wondered is there any problem that abaqus can not solve by XFEM, or the results can not satisfy the accuracy of research requirements.

Sun, 04/08/2012 - 12:23 Permalink
Dipak Rakshit

Dear all,



I am a M.Tech Student doing research on Fracture Mechanics of Fibre Reinforced Composites using XFEM in ABAQUS.



To
achieve the goal, I started from Crack Initiation and Propagation for
isotropic material in ABAQUS using XFEM following this manual

http://www.scribd.com/doc/102150619/Frac-l09-Xfem.



It has two case:



1. Crack initiation and propagation

2. Propagation of an existing Crack.



For the first case, Maximum principle stress criterion has been used
and maximum principle stress has been set as 22e6. In result, in the
crack tip stress become about twice this maximum stress.But,  no crack
has been initiated.







I am begging for your kind attention to help me in this regard.



Waiting for your kind advice,



Thanks & regards,

Dipak Rakshit

Mon, 09/24/2012 - 16:45 Permalink
havelmay

The default parameter of attempts per increment ist 5. You can increase the number in CAE by clicking on:

Model ==> Steps ==> "Step-1" (or how you called the step after the initial step with the loads) in the model tree on the left with double-click

==> the menu bar on the top changes ==> click on "Other" ==> "General Solution Controls" ==> "Edit" ==> "Step-1"(or the other name)

==> "Specify" ==> "Time Incrementation" => "more" (the first of the three) ==> parameter "IA" ==> increase this parameter

Thu, 09/18/2014 - 20:17 Permalink