Hi everyone,

After switching from a working 2D model to 3D, I'm facing the following error message after about 20 iteration steps:

The system error in std_findcutoneedge3d8_xfem -- nodal level set values might not be correct for element 12345 instance FEM_MODEL_V1

I'm trying to simulate a 3-point bend test using XFEM. The model works and converges fine when I suppress the cracks in the assembly. I also played with various material settings (although the current ones work in the 2D model). From a Chinese forum I found that the issue might be cracks that connect/merge after some propagation, which seems to be a scenario XFEM can't handle. However, I haven't found a solution for this, and it doesn't seem to be such a rare scenario? I'd highly appreciate any input and help!

Thanks a lot!

Hello everyone,

I am currently working on modelling ITZ zones in concrete 2D microstructure. ITZ is the cement paste zone near the surface of aggregate which differs in properties from bulk cement paste. The thichness of ITZ zone ranges from 10-50 micrometer (0.01 - 0.05 mm). The size of RVE I am using is 100X100 mm which makes ITZ a very fine detail in my microstructure. I have created a instance in abaqus merging all parts, Aggregate, ITZ and Matrix( Cement Paste).

I seeded the instance with using these parameter:

Global Size: 2

Maximum deviation factor = 0.0001 (Approximate no of element per circle 7854)

Minimum size value = 0.1

these parameter ensures that I get same size elements at ITZ region regardless of size of perticular aggregate.

Now here is my problem:

This seeding gives me the meshing in Picture 1, in which element size decrease from outer boundary to ITZ zone, but not increases again after passing the ITZ zone and getting into Aggregate region. My required mesh can be seen in Picture 2. Can anyone help me with this?

In abaqus documentation, there is mention of "increase the size of the interior elements" but it says one can only do it when using tetrahedral element. Can I do it 2D case using quad dominated elements?

Thanks.

Vasav Dubey

Graduate Student

Texas A&M University

I will be developing constitutive material models into commercial FE codes for nonlinear elasticity and searching for good books to get started for computational aspects. There are many good books for computational plasticity but I did not find any for nonlinear elasticity. Suggestion for good books or references is welcome.

Hello everyone,

I need to compute the second derivatives of displacement field in ABAQUS, in the integration point of quadratic elements if it is possible. I thought about using USDFLD subrutine, but I am stuck with getting the Jacobian matrix that couples the isoparametric space with the physical space in ABAQUS. GETVRM is what everyone suggests to get variables from abaqus, however I did not find the definitions of JMAC, JMATYP... so I do not know what I can get from that subrutine exactly.

Another way to solve my problem would be accessing the coefficients of the quadratic element equation. Second derivatives of this equation should give me an array of constants for each integration point. But, I do not know how to access this information.

I would appreciate any kind of feedback you can give me, thanks in advance to all the iMechanica community.

Normal
0

false
false
false

EN-US
X-NONE
X-NONE

/* Style Definitions */
table.MsoNormalTable
{mso-style-name:"Table Normal";
mso-tstyle-rowband-size:0;
mso-tstyle-colband-size:0;
mso-style-noshow:yes;
mso-style-priority:99;
mso-style-parent:"";
mso-padding-alt:0in 5.4pt 0in 5.4pt;
mso-para-margin-top:0in;
mso-para-margin-right:0in;
mso-para-margin-bottom:8.0pt;
mso-para-margin-left:0in;
line-height:107%;
mso-pagination:widow-orphan;
font-size:11.0pt;
font-family:"Calibri",sans-serif;
mso-ascii-font-family:Calibri;
mso-ascii-theme-font:minor-latin;
mso-hansi-font-family:Calibri;
mso-hansi-theme-font:minor-latin;}

Hello,
I am a PhD student in Mechanical Engingeering.

I am currently studying the co-rotational formulation, following the theory on Crisfield's books. The next step in my project would be to apply the corotational formulation for the simulation of a beam, and my supervisor suggested me this could be done in ABAQUS through a User Defined Element Subroutine.

Is there anybody who has applied the corotational formulation in ABAQUS before? A similar application in MATLAB would be useful as well; in that sense I found great help in this previous discussion on iMechanica: http://imechanica.org/node/4880  .

Thanks in advance to anyone who would reply,

Dario

Hello,  I am a PhD student in Mechanical Engineering, and at the moment I am studying the co-rotational formulation from Criesfield's books.

I am looking for applications of the corotational formulation in ABAQUS or MATLAB; my supervisor suggested me it could be done in ABAQUS through a User Defined Element Subroutine, but being a new user to this software I don't really know from where I should start.

I found great help in this previous discussion on iMechanica - http://imechanica.org/node/4880 - and the papers it refers to, where the author describes the necessary steps to follow for a corotational-based FEM in MATLAB, but unfortunately no code is provided.

If anybody could help me, it would be much appreciated.
Thanks in advance to anyone who would reply,

Dario

Hello,

I have some troubles in calculating the strain in ABAQUS. I have the following Model with one element (see picture) and a linear elastic isotropic Material (Steel), which is sheared in the x-direction. The analytical result of the strain tensor gives me in xz-direction 0.25 (see picture). The result in ABAQUS is 0.5. I ran the same simulation in CalculiX CrunchiX and got the expected strain of 0.25 in the xz-direction

Can anyone tell me how ABAQUS calculates the strain for a linear static analysis? Why is the ABAQUS result exactly 50% higher?

Thanks and regards

Christof

DECATHLON is a French company which designs and sells sports goods. With about 900 shops all around the world, DECATHLON has more than 60000 employees, all passionate about sports, and generates a 7 billion euros turnover.

The Mouvement Science Lab (MSL), which is part of Decathlon SportsLab, is looking for a Finite Element (FE) engineer who is able to build and lead numerical simulation projects that will improve sports footwear conception. These projects have to be in line with fundamental research studies focused on understanding interactions of different types of footwear with the human body during sports practice.

Required skills for this position :

-          finite element simulation (a previous experience on Abaqus or Ansys sotwares is preferred),

-          good knowledge of materials behaviours, such as viscoelasticity,

-          good knowledge of CAD softwares (NX, Rhino, Solidworks, or any other),

-          a previous experience in musculo-skelettal modelling human movement simulation would be a plus,

-          ability to work in a team,

-          good knowledge of English and French,

-          regular practice of sporting activities or a strong interest for sports.

Needed responsibilities for this position :

-          create and lead applied and fundamental research projects,

-          generate knowledge on the human body in motion as well as innovative concepts,

-          generate guidelines for footwear products across the different brands of the Decathlon company,

-          ensure good collaboration between decathlon and its partners on research projects, and identify other potential partnerships,

-          be part of team composed of biomechanics engineers et help in the development of the team.

If you want to apply for this position, please send your CV and motivation letter to Nils Gueguen (nils.gueguen@decathlon.com) or Lise Sissler (lise.sissler@decathlon.com).

Dear all,

I'm using UEL in ABAQUS. The UEL works perfectly as far as I have 256 time increments (each 1e-6 sec). When I'm trying to increase the time increments (let's say 512 with the same 1e-6 sec per increment) Abaqus gives me error code 1073741819.

I'm running Abaqus 6.14.2 with Intel Fortran Composer XE 2013 on a Windows 7. CPUs are 3rd generation.

I would appreciate if you could help me to figure it out.

Best Regards,

Dear all,

I'm using UEL in ABAQUS. The UEL works perfectly as far as I have 256 time increments (each 1e-6 sec). When I'm trying to increase the time increments (let's say 512 with the same 1e-6 sec per increment) Abaqus gives me error code 1073741819.

I'm running Abaqus 6.14.2 with Intel Fortran Composer XE 2013 on a Windows 7. CPUs are 3rd generation.

I would appreciate if you could help me to figure it out.

Best Regards,

     *** This is a NEW OPEN POSITION - not related to any previous post ***

   A postdoctoral position with primary focus on tissue mechanics is available immediately in the Shenoy Research Group at the University of Pennsylvania. We are looking for a strongly motivated candidate to work on an NIH supported project on the interactions of cells in 3D micro tissues mediated by extracellular matrices using analytical and finite element methods (for example, H. Wang et. al, PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES, 110(52): 20 923-20928 NOVEMBER 2013). 

   The ideal candidate will have a background in solid mechanics with expertise in finite element simulations, in particular numerical implementation of non-linear material models. This individual will have the opportunity to be directly involved in complimentary experimental investigations in the medical school at UPenn and our collaborators elsewhere. 

  Candidates should send their CV with names of three references to Prof.Vivek Shenoy ( vshenoy [AT] seas.upenn.edu )

  If you have any question regarding this position and/or about Prof. Shenoy's group activities, please feel free to contact me at : dibakar_datta [AT] alumni.brown.edu 

 Thank you very much. 

The Corporate CAE group for Becton Dickinson is looking for a CAE Senior Staff Engineer at the Research Triangle Park, NC USA location (Raleigh-Durham area).

The CAE Senior Staff Engineer is an essential part of BD’s Corporate Computer-Aided Engineering (CAE) team and will have the opportunity to contribute to the advancement of BD’s next generation medical products and enhance the current product portfolio. The CAE Senior Staff Engineer reports operationally to the CAE Analysis Manager and serves as an organizational subject matter expert in the technology sensing, development and application of advanced structural mechanics simulation capabilities important to the company. Acts as a strategic advisor to all levels of the BD community to maximize business contributions and has the responsibility to technically lead the implementation and usage of CAE simulation to best impact BDX product development. The ultimate goal is to promote the utilization of CAE simulation to drive innovation and deliver high quality and robust products to market faster.

For more information and to apply for the position, see the link below.

https://jobs.bd.com/job/-/-/159/1022600?src=SNS-102

Dear Abaqus Experts:

I am new Subroutine user and I want to analyze plastic deformation behavior of micro component.

So, I have to link Abaqus,gfortran,and Microsoft Visual Studio.

Itriedseveral methodsby reference to the followinglink,but I have notyetbeen successful inthem.

So,I'ma verynecessaryhelpof Abaqus experts.

Detail

1) I added the command "mingwvars.bat" to Abaqus Command > right click > properties > target

2) I double-clicked "vcvarsall.bat" and "mingwvars.bat".

3) I typed "set >path.info" on Abaqus Command window.

4) I edited environment variables. Control Panel > System > Adbanced tab > Environment Variables

5) I typed "abaqus verify -user_std". But itis displayed as ERROR on Abaqus Command window.

I referenced these links.

http://www-h.eng.cam.ac.uk/help/programs/fe/abaqus/faq68/UserSubConfigureInfo.html

http://imechanica.org/node/11219

http://imechanica.org/node/13370

http://manishkumar.org/manishkumar/resources/softwares/abaqus/AbaqusLinking.pdf

I tell you mycurrentwork environment.

OS:Windows7 Professional SP1

Abaqus:Abaqus standard 6.14-5

Compiler:gfortran

Visual Studio:Microsoft Visual Studio Community 2015

Please help me.

I have set up a vibration problem. I assign a local material orientation by assigning coordinates a and b to define the 3rd material direction. From documentation, I understand that the other two directions are assigned to be orthotropic to the 3rd direction. Beyond that, I think that they are set somewhat arbitrarily. I want to output each component of the local material orientation.   *Orientation, name=Ori-DS_LV_DS_LV_FIBRES, definition=COORDINATES, system=RECTANGULARDS_LV_DS_LV_FIBRES1, 0. I found a way to output the local material orientation to the .dat file with:  *El Print, POSITION=AVERAGED AT NODES,E This gives me the following in my .dat file. (Just the first few lines).   THE FOLLOWING TABLE IS PRINTED FOR ALL ELEMENTS WITH TYPE C3D8H AVERAGED AT THE NODES        NODE  FOOT-   E11         E22         E33         E12         E13         E23             NOTE            1   OR  1.0881E-03 -2.4529E-03  5.5659E-04 -3.6484E-04 -3.3336E-04 -1.3667E-03           1  SSD  3.4338E-04  7.5227E-04 -7.7260E-04 -6.8945E-05 -1.4456E-04 -1.4310E-03           2   OR -1.1083E-03  3.8331E-04  8.6723E-04  1.6965E-05 -5.9742E-04  4.4579E-04           2  SSD  7.3394E-04 -6.7419E-04 -4.6673E-04  6.4805E-04 -7.2251E-05 -1.6647E-03           3   OR -1.2586E-04 -7.8578E-04  9.5575E-04  2.8798E-03  2.1864E-03 -1.3924E-03           3  SSD -8.5076E-04  1.2290E-03 -1.8964E-04 -9.0878E-04 -7.5154E-04 -1.1113E-03           4   OR -9.3005E-04 -2.0169E-04  9.0388E-04 -1.6976E-03 -6.0696E-04  4.8134E-04           4  SSD  2.0887E-04  4.6213E-05 -2.5190E-04  2.5068E-04  2.9128E-04 -1.0801E-03           5   OR  5.3199E-04 -1.9483E-03  4.2709E-05  2.3164E-05 -7.6685E-04 -6.1928E-04           5  SSD  7.9289E-04 -6.5242E-04 -9.3984E-06 -6.1390E-04 -7.8684E-04 -1.2823E-03           6   OR -1.4705E-03  8.0179E-04  9.6646E-04 -2.2979E-04  8.8915E-04  1.9077E-03           6  SSD  7.2084E-04 -9.1113E-04 -3.0886E-04  2.4443E-04 -1.2574E-03 -2.0021E-03           7   OR  9.3148E-04 -2.4115E-03  2.1603E-04  2.3331E-03  2.0623E-04 -2.8699E-03           7  SSD  1.6355E-04  2.9455E-05  5.5929E-04  8.2595E-04  4.0638E-04  8.2047E-05           8   OR -5.3565E-04 -7.1252E-04  8.8212E-04 -1.5246E-04 -7.3226E-04  7.6021E-04           8  SSD -2.1401E-04 -1.2126E-05  6.5431E-05 -2.2339E-04  2.5891E-04 -4.2669E-04           9   OR  8.8023E-05 -1.9433E-03  2.3335E-04 -5.6005E-04 -1.6104E-03 -1.5217E-03           9  SSD  4.5202E-04  1.7247E-04 -1.0855E-04 -1.9394E-04  1.8261E-03 -6.4481E-04          10   OR -1.1895E-03  1.1078E-03  3.2344E-04  6.7851E-04 -6.4467E-04 -7.4874E-04          10  SSD -1.7305E-04  3.0125E-04  1.4873E-04 -1.2751E-04 -9.9659E-05 -3.6749E-04 Looking further into documentation, I understand that OR (orientation) is written out in "Type 85" format.  Record key: 85Record type: Local coordinate directionsAttributes:  1  –  First component of the first direction. 2  –  Second component of the first direction. 3  –  Third component of the first direction. 4  –  First component of the second direction. 5  –  Second component of the second direction. 6  –  Third component of the second direction.I have two problems: 1) When I take the cross product of the these first two directions in the output orientation, I should get my 3rd direction. This, should be the same (or close to), my input fibre field interpolated at each node. However, the fibre directions are not equivalent (or even close) to my input fibre field. I also realise that there will be discrepencies because I am outputting the orientations averaged at nodes. 2) I would have expected the first two directions which were output to the .dat file to be orthogonal to one another, but they're not.  Is there another way to correctly output the local material orientation? Or if this is the true orientation, then I assume that the third direction may be far from my original fibre field due to the averaging of element orientations at the nodes. But then, why are the other two directions not orthogonal to one another?  Thank you for your help.  Also, I have attached both my input file.

I have a vibration problem where I have assigned the local material orientation by assigning points a and b. By assigning points a and b, I define one material direction in my local coordinate system - direction 3. From Abaqus documentation, I understand that the other two directions are assigned to be orthogonal to the 3rd direction but are set somewhat arbitrarily. I want to write out the directions so that I know the three vectors defining all three directions of the local material orientation.

I found that I can write out the local orientation averaged at each node by using: 

*El Print, POSITION=AVERAGED AT NODES

, 

E

This gives me data in my .dat file where every other line has the footnote: OR. When I look further into documentation, I understand that this is written out as Type 85 record type: the first three numbers are the three components of the first direction and the last three numbers are the three componetns of the second direction. 

Therefore, taking the cross product of these two vectors should give me the 3 direction. In addition, these two vectors should be orthogonal to one another. Neither of these two things are true. 

Is there another way to output all three components of the local material orientation? Or is this method correct? And if so, why aren't the first two directions orthogonal to one another?

I have attached both the input file and part of the dat file with the output.

Thank you for your help. 

Renee

In ABAQUS, I have modelled a assembly containing two parts with lot of constraints and other engineering features which are very crucial to my FE analysis. The job ran successfully and I have got satisfactory results. As a next step, I would like to import this model into the Multi Body Software SIMPACK as a flexible body. This would require the introduction of substructuring concepts and I referred to SIMPACK tutorials for exact way of doing it. My question is, can we import an assembly into SIMPACK. Because the tutorials explain only a part being impoted into SIMPACK. SO when I follow the same procedure given in the tutorials to my model, ABAQUS gives me an error message saying SUBSTRUCTURE GENERATE cannot be used for assemblies. It seems it can only be used for parts. So how do I go about this?

I have been trying to model the cyclic behaviour of soil. I got a UMAT. But when I'm using UMAT, it is creating lot many errors (like numerical singularity and excessive distortion). I have constrained it properly. I tried running my model with an inbuilt material (clay plasticity) and I'm getting results. Here I'm not even able to move from the initial geostatic step. Can anyone please help

Hello,
I am looking for an explanation about the formulation used for CONTINUMM SHELL elements in ABAQUS.

The software manual only explains that SC hexaedral elements consist of 4 nodes, each with 3 translational degrees of freedom and no rotations , exactly as the normal CONTINUUM SOLID elements, but that apply a formulation similar to plane SHELL elements (plane stress state). The deflection in the third direction is then evaluated as the difference of displacement along z between elements with the same x,y coordinates, while a normal shell elements would use the rotational degrees of freedom.

Unfortunately Abaqus doesn't really explain what this special formulation is (normal solid or shell elements formulations are easily found in literature).
Does anyone has an in-depth explanation of how continuum shell elements work?

I am simulating the fracture of a metallic thin plate subject to impulsive loading in the third direction.
I managed to calibrate the shear and ductile damage initiation and evolution parameters for the metal (mild steel) by reproducing the tensile test of a dogbone specimen of the same material with continuum shell elements, and I am quite satisfied with the results obtained.
I want to study the formulation of such elements to try to understand why the same simulation running with normal solid or shell elements gives different ,and worse, results.

Thank you

Hi,

I am fairly new to Abaqus and trying to simulate crack propagation in a concrete 3D cube where I have used Matthew Pais' 3d edge crack model as guidance. However i want to use concrete as the material for the simulation and i have searched the net and am having trouble finding any reliable values for the damage criteria. Please could anyone assist me with this problem. As far as i know i need to find values for the Mass Density, Young's Modulus, Poisons ratio, Max principal stress and Damage evolution? is that right or are there other things i should be including when creating the material.

Also another query that i have is, following matthew pais model, how can i get the crack to propagate further than the length of the shell 'crack' part i created. At the moment i have placed the shell crack part on the edge of 1 surface of my cube and it has a length of 2, when i run the job (using material properties for aluminium) the crack only opens until the length 2. It does not propagate any further no matter how much i increase the tensile load being applied to the cube. Is there any way around that?

Thanks in advance and would appreciate any help available.

Date: Thursday, March 17, 2016

Time: 10.00 am – 2.00 pm

Venue: Wyndham Houston West - Energy Corridor, 14703 Park Row, Houston, TX 77079

Fee: free-to-attend

This lunch & learn seminar is aimed at people who have experience and are interested in FEA simulations and would like to extend the scope of their FEA work to image-based digital rock physics and reservoir characterisation. If you are currently working in the Oil and Gas Industry but feel you are limited in your existing capability or simply want to enhance you knowledge of the technology then this is for you.

Topics covered:

Generating multiphase meshes from 3D scan data with Simpleware Software

Quantifying porosity, cracks, effective material properties, absolute permeability and other statistics

Simpleware software compatibility with Abaqus, and solutions for digital rock physics and reservoir characterisation

Essential concepts of advanced simulation with SIMULIA Abaqus

Oil & gas applications using SIMULIA Abaqus and Simpleware software suite

Agenda:

10.00 am – 11.30 am    Simpleware: Linking Abaqus and Simpleware software for oil & gas applications - From core analysis to reservoir characterisation.

11.30 am – 12.30 pm    Lunch

12.30 pm – 1.30 pm     VIASSIMULIA software for oil & gas applications - concepts & typical solutions for advanced FEA analyses. Isight workflow development for BOP optimization.

1.30 pm – 2:00 pm      Simpleware & VIASQ&A

Register here: http://simpleware.com/news-and-events/events/workshops/simpleware-and-vias-houston.html#.Vt7b2JyLTcs