Elmer is a computational tool for multi-physics problems that has been developed in collaboration with Finnish universities, research laboratories and industry over the last 10 years. The software has been recently released under the GPL license. Source code, binaries, examples and tutorials can be found at:

http://www.csc.fi/elmer/

Elmer includes physical models of fluid dynamics, structural mechanics, electromagnetics and heat transfer. These are described by partial differential equations, which Elmer solves by the Finite Element Method (FEM).

Elmer comprises of several different parts: The geometry, boundary conditions and physical models are defined in ElmerFront. The resulting problem definition is solved by ElmerSolver. Finally the results are visualized by ElmerPost. Additionally a utility ElmerGrid may be used for simple mesh manipulation.

The different parts of Elmer software may also be used independently. The strongest of the components is ElmerSolver which includes many sophisticated features. For pre- and postprosessing the users may find also other alternatives.

The software compiles on a large range of platforms, amongst others: Windows, Linux, Solaris, AIX and MAC. There is also support for parallel computation via MPI.

Best regards, Elmer team

From the website, this package looked pretty good. Has anyone installed this and tried it out yet? Any feedback? Thanks!

phsieh

I had a go but failed - there appears to be some include files missing which prevents compilation of the frontend (specifically frontlib.h and probably others).

Some other minor problems I found were that under the intel fortran compiler 7.1 I got an internal error (bug in the compiler) building the fem archive and under v9.0 of the compiler the link option "-Wl,-rpath=/opt/elmer" has the = sign missing in the make file.

To install Elmer you have to make some changes to file ELMER4.0/bin/ElmerFront as per the location of your directory.

Vinod Dhiman

I've just downloaded the binaries and they do work as you say (although not much use on my home computer using SUSE 9.2 and intel fortran/c compilers at v9). My problem was with compiling the codes from scratch - hence the reference to the missing include files.

I've also just attempted to compile it using pgf90 (4.1-2) and this also fails due to an internal compiler error (as well as some minor problems with nullifying pointers; e.g. M => NULL() is f95 and is not valid f90 and so should be either deleted ore replaced by the command nullify(M) ).

Hi,

First of all, thanks for your interest in Elmer. If you have more compilation problems you can post them in this thread, and I will try to help out.

> I've also just attempted to compile it using pgf90 (4.1-2)

Earlier versions of portland group compilers had numerous problems eg. with recursive subroutines -- and there are still some problems. Still, Elmer does compile with pgf90 6.0-4 (this is on x86-64).

> Some other minor problems I found were that under the
: intel fortran compiler 7.1 I got an internal error (bug in
: the compiler) building the fem archive and under v9.0 of

I just fixed a minor problem with 7.1 and I have now successfully compiled on 7.1, 8.1 and 9.0. The fem tarball is updated at the ftp site. The exact compiler versions are:

Intel(R) Fortran Compiler for 32-bit applications, Version 7.1 Build 20030307Z

Intel(R) Fortran Compiler for 32-bit applications, Version 8.1 Build 20050207Z

> the compiler the link option "-Wl,-rpath=/opt/elmer" has
: the = sign missing in the make file.

We use the C++ compiler for linking on linux, so obviously the problem is there. What C++ compiler are you using (version and vendor)? Atleast icc 8.1 and gcc 3.3.5 work. I will try to come up with a solution to fix this.

> I had a go but failed - there appears to be some include
: files missing which prevents compilation of the frontend
: (specifically frontlib.h and probably others).

I will also look into this. Actually you could leave out front, because it is a bit outdated. None of the examples or tutorials require ElmerFront. ElmerPost on the other hand is quite useful (it even has a built-in matlab:ish scripting language), even though it looks a bit rough on the outside.

Best Regards, Juha Vierinen

HI,

I joined the discussion group and then tried to post a few questions, but, it bouncd back. So, I am wondering if you can help me with these questions:

1. elmer 4.0 supports 3D (extruded 2D mesh to 3D). Can it handle complex 3D mesh? Can I use GiD for this? 2. What method is used for free surface (VOF, level set, front tracking...etc.)? 3. Is air fraction of the free surface compressible or incompressible or both? 4. Is the mailing list for discussion operational? It might be a better place for asking questions - sourceforge.net or Yahoo discussion group might be useful.

phsieh2005

Hi

>I joined the discussion group and then tried to post a few >questions, but, it bouncd back. So, I am wondering if you >can help me with these questions:
:
:1. elmer 4.0 supports 3D (extruded 2D mesh to 3D). Can it
:handle complex 3D mesh? Can I use GiD for this? 2.

Yes, Elmer supports general 3D geometries. Available element types include tetrahedra, hexahedra, pyramids and wedges, all with various number of nodes. You can use GiD to create the geometry and the mesh. For transforming GiD data to Elmer mesh format, please check instructions and files under the title "Elmer templates for GiD" on the Elmer download page http://www.csc.fi/elmer/download

>2. What method is used for free surface (VOF, level set, >front tracking...etc.)? 3. Is air fraction of the free >surface compressible or incompressible or both?
:4. Is the mailing list for discussion operational? It might >be a better place for asking questions - sourceforge.net >or Yahoo discussion group might be useful.
:
hsieh2005

The mailing list is operational. It is also a good place to ask questions on the numerical methods etc.

Best Regards, Antti Pursula

Thanks for your reply.

The C/C++ compiler I was using was intel v9.0 (icc/icpc). I'll look more closely at the linking problem over the weekend if I have time and let you know what I need to change to get fem to compile and pass the "make check test".

Tom.

Hi,

> The C/C++ compiler I was using was intel v9.0 (icc/icpc). I'll look more closely at the linking problem over the weekend if I have time and let you know what I need to change to get fem to compile and pass the "make check test".

I fixed some problems with meshgen2d, front and fem -- there were some missing files in the tars, and some missing linker flags for icc. You will find the packages updated in the ftp server.

juha

[mtouil]

hi
after simulate a different phisics phenomena of flow i want to generate a plot from this sofwar but i have a difficut and i dont now how to make it.

[baal]

test

1. The message you have entered is too short. Please lengthen your message to at least 10 characters.