iMechanica - Sharing ABAQUS UMAT and VUMAT subroutines - Comments

It's an old topic.

Shunlai.Zang — Sun, 24 Aug 2008 09:05:30 -0400

Just as open source vs. closed source

But I think sharing knowledges is a good thing.

Shunlai

UMAT for Shape Memory Alloys

msimsek — Wed, 20 Aug 2008 12:40:21 -0400

Hi,

I am an M.Sc. mechanical engineering student and I am trying to model super elastic property of shape memory alloys in ABAQUS. I am new to ABAQUS and trying to learn it by reading the tutorials. It seems I have to use UMAT but I dont know how to use it or how to create a subroutine for my case. If anybody helps, makes contribution it would be appreciated.

Look forward to getting some responses.

Thanks

Miray

searching Umat Vumat

Tosche96 — Sun, 20 Jul 2008 15:59:56 -0400

Hello,

I'am looking for a vumat or umat file for thixotropic material! Has somebody already written such kind of file for abaqus! I want to simulate thixoforming / semi-solid process!

thx

Request for UEL or UMAT in ABAQUS or LS-DYNA or a source code

Babak — Wed, 25 Jun 2008 14:04:21 -0400

Dear friend

I am Babak, a Ph.D. candidate in geotechnical engineering, school of civil engineering in university of Tehran (UT), Iran. (www.ut.ac.ir). My research topic is about “Capturing shear band in slope and excavation in static and seismic conditions”. I look for a model which can handle shear banding and strain localization in geological materials specially soils. I really appreciate your favor whether help me in this regard. Any helpful comments about source code, open source software or even a hint which can help me to follow it are appreciated. I am completely ready to do joint research with someone else.

Truly yours,

Babak Ebrahimian

Ph.D. Candidate in Geotechnical Engineering

University College of Engineering

School of Civil Engineering

University of Tehran

Tehran, Iran

Postal Code: 11155-4563, Enghelab St., Tehran, Iran.Mobile Phone: (0098-912)5506553Phone Number: (0098-311)4416302Fax: (0098-311)4343275

Email: bebrahimian@ut.ac.ir

Dear Mr.

ho1mo — Fri, 20 Jun 2008 17:34:14 -0400

Dear Mr. Azadpoor,

Hi,

I am Hojat Mohamadi from Iran.

I and my team work for generation a book that completely explain modeling and analyzing the mechanical events by ABAQUS, step by step, and consists of fundamental of /CAE, creating input files and ABAQUS programming .

I have no problem in fundamental of /CAE and input files because of my experience, but I have no reference in ABAQUS proggraming and will be happy if you guide me about it.

Therewith, may be explaine me about ABAQUS programming(specially UMAT) and is the ABAQUS Documentation enough?

My e-mail: Ho1mo@yahoo.com

Thanks a lot for your kindly response.

Hojat Mohamadi

Compiling UMAT as free f90

alankar — Tue, 06 May 2008 17:48:45 -0400

Alankar
Ph.D. Student
School of Mechanical and Materials Engineering
Washington State University, Pullman

re: running free format fortran

Arun Prakash — Fri, 18 Apr 2008 17:08:54 -0400

Hi Jianhong,

I recently tried programming a basic UMAT in f90. After a whole deal of struggling, i could get it to run. Here are some pointers.

a) extend the compiler options in the environment file to include free format fortran - with the intel fortran compiler these options would be -FR or -free.

b) be careful on number off bytes used for floating point variables (double and single precision)

c) it always helps to first compile the UMAT and then run it using the object file. use the compiler command in the env file and compile your UMAT.f90 (u may have to copy the aba_param.inc file locally) to a UMAT1.o file. Now run your job as before but use UMAT1 as the subroutine file rather than UMAT.f90

e.g.

for run with fortran source

abaqus -j jobname user=UMAT

for run with fortran object code

abaqus -j jobname user=UMAT1

make sure that your source and object files have different names.

Arun

A question for UMAT

Jianhong Jiang — Sat, 22 Mar 2008 23:41:00 -0400

hi,

I have the f90 code for the constitutive relationship for some complex plasticity material, and need to shape it be a Abaqus/Umat subroutine. But I see that the Abaqus only recognize umat.for format, not umat.f90 format. The manual only gives example with fixed fortran format.

1) Does anyone ever try the free fortran format?

2) Can we compile the fortran umat subroutine with some tool before run it with Abaqus? By compiling, we can get umat.obj, then run it with Abaqus,

Abaqus job=job.inp, user=umat.obj

Is there any such tools, and how to compile it fisrt?

Thanks a lot!

Jianhong Jiang

Sharing ABAQUS UMAT and VUMAT subroutines

aamirmub — Sun, 02 Dec 2007 16:54:20 -0500

Hi,

Thanks for your detailed remarks. I think my point is put forward when you acknowledge that development of material models take long time and effort unless changes are being made in available code. As you pointed out, some implementations of material models are available freely but I think implementation of models still takes more effort than just re-arranging a few lines of already available code. If this has been the case, there would be no requirement of sharing anything as the freely available code could be re-arranged easily.

As far as the commercial software is concerned, no one expects them to make the code available. But after being implemented in a few releases, especially the cohesive zone model, is in reasonable state for implementation. The research aspect would always require the development of material subroutines like umat.

Anyways, good luck in your quest for umat.

Regards

Aamir Mubashar

Hi Aamir

azadpoor — Sat, 01 Dec 2007 14:33:18 -0500

Hi Aamir Mubashar

Thanks for your thoughtful comment and a few remarks.

"Successful implementation of complex material models is a result of,
if not years, months of research work"

That is true if it is gonna be from scratch. But I think most of subroutines
can be developed simply by rearranging and converting the available codes to make
FORTRAN subroutines. For example, for crystal plasticity, a handful of models
are available in Tahoe along with the source codes and one just has to bring it
to the UMAT format. For Gurson-type and Cohesive zone elements, the source
codes can be found in WARP3D.

"Cohesive zone models are now available in most of the commercial FE
packages along with different cohesive laws. Gurson model is also implemented
with damage and failure simulation capabilities."

That is true but as you know the source code is not available and the flexibility
while dealing with these models is absolutely limited.

"Also sometimes the subroutines are written for particular class of
problem and not applicable generally."

This is also true but you know that modifying, updating, or improving is
much easier if a basis is already available. Therefore, I think sharing
subroutines would be useful in the sense that other people can build their own
codes on top of the already available subroutines.

"I think that sharing knowledge of development of such routines is more
important. Experienced users can help others by sharing difficulties, hints,
tips, entry level models or tutorials."

I agree. Tutorials, manuals, and examples are particularly useful.

Amir Abbas Zadpoor Faculty of Aerospace Engineering, Delft University
of Technology (TU Delft)

Sharing ABAQUS UMAT and VUMAT subroutines

aamirmub — Fri, 30 Nov 2007 19:45:49 -0500

Hi,

Its a good idea to share the umat and vumat subroutines but we have to be mindful of the effort put in by the author. Successful implementation of complex material models is a result of, if not years, months of research work. Also sometimes the subroutines are written for particular class of problem and not applicable generally. Therefore, I think that sharing knowledge of development of such routines is more important. Experienced users can help others by sharing difficulties, hints, tips, entry level models or tutorials. iMechanica provides a good plateform for this.

Cohesive zone models are now available in most of the commercial FE packages along with different cohesive laws. Gurson model is also implemented with damage and failure simulation capabilities.

Regards

Aamir Mubashar

Wolfson School of Mechanical and Manufacturing Engineering
Loughborough University
Loughborough
United Kingdom

Could you share them?

azadpoor — Fri, 30 Nov 2007 05:43:18 -0500

To Huapei,

Is it possible for you to share the subroutines you have developed? If yes, please let me know.

Amir Abbas Zadpoor
Faculty of Aerospace Engineering,
Delft University of Technology (TU Delft)

To Azadpoor, Good idea. I

Huapei Wan — Thu, 29 Nov 2007 15:46:34 -0500

To Azadpoor,

Good idea. I wish I had the first and the third ones when I was developing them.

Huapei

Sharing ABAQUS UMAT and VUMAT subroutines

azadpoor — Thu, 29 Nov 2007 08:01:52 -0500

I have noticed that a lot of people are using UMAT and VUMAT material model
subroutines for their researches. However, most of these subroutines do not
find their way to free code market and remain in the hands of the developers.
Therefore, a lot of wheel re-inventing has to be done by other people. Since
development and testing of a material subroutine takes a damn long time, I
think it is a good idea if each of us can share the subroutines she has
developed with the others. I know that a couple of subroutines are publicly
available and many other people also know that. Please comment about less-known
freely available subroutines.

To be more specific, I think it would be useful if people can comment about
following highly demanded UMAT subroutines:

1.Cohesive zone elements models

2.Polycrystalline plasticity

3.Gurson-type ductile fracture models

Note added by admin: This topic has also been discussed in another thread: http://www.imechanica.org/node/2376