0
$\begingroup$

I have used the free, multibody dynamics software MBDyn for a while now. It is a good program and also fits my needs. There are good manuals, basic tutorials and examples to be found. There is also a mailing list. The plotting program I had to write myself but there are even suggestions for how to do that.

But as I get further into MBDyn I find slight problems constructing the models. I can't say if it is only my own lack of background in engineering and algebra that is the problem. (MBDyn interfaces with the graphical environment Blender, both for input and output, but for me a text interface for input is sufficient, so far.)

So my question is: What software for multibody dynamics is available on the market?

I am interested in connecting solid objects by elastic and deformable joints, like hinges, ball joints etc. I wish to see how the bodies move when forces are applied and "springs and damping" interact with the connected bodies.

I have never worked with the Finite Elements Method (FEM). I don't think that is what I need. I think that is more for modeling continuous structures like fluids, gases, solid materials etc. (I wish to model beams, but I think that this also done in multibody dynamics (MBDyn does it well). I don't need any detailed internal information from the beam. I just need to know "a little bit how it bends".) I do not need any information about internal forces in the objects, just how they move.

So please tell, what multibody dynamics software exist? And, if possible, what are their differences?

$\endgroup$
1
$\begingroup$

Your question is really too broad for this site. There are dozens of commercial and open source multibody dynamics codes, many of which can be identified by simple Google search.

Since you didn't restrict your question to open source, I'll mention two commercial packages that have been very widely used for more than 20 years:

MSC/Adams http://www.mscsoftware.com/product/adams

LMS/DADS https://www.plm.automation.siemens.com/en_us/products/lms/virtual-lab/legacy-applications/dads.shtml

I suggest you take a look at the survey paper, Computational strategies for flexible multibody systems, by Wasfy and Noor that focuses on adding flexibility effects to rigid body dynamics codes. It also mentions the two codes I list above. A pdf version of this paper is currently available here: http://hosting.umons.ac.be/html/mecara/grasmech/PaperWasfyNoor.pdf

This site (http://sbel.wisc.edu/Courses/ME451/2014/), contains a course on multibody dynamics taught at the University of Wisconsin. Among other useful documents, it contains a copy of a book on this topic by Edward Haug, one of the original authors of the DADS code. This book discusses different approaches for multi- rigid body dynamics so should help to answer your question about differences among the various codes.

Finally, you say you are not interested in FEM. But you should understand that, from a software point of view, in recent years, the distinction between FE codes that perform nonlinear structural dynamics and multibody dynamics codes that include flexiblity effects has become very blurred. On the one hand, the Abaqus FE code (http://www.3ds.com/products-services/simulia/products/abaqus/) has a wide variety of joints and links for constructing mechanisms. On the other hand, the multibody dynamics code you mention, MBDyn, has a relatively general library of (essentially) finite elements for modeling flexible bodies. You can easily find several papers where MBDyn has been used for solving structural analysis problems that, traditionally, would have been solved by a FE structural code.

| cite | improve this answer | |
$\endgroup$
  • $\begingroup$ Thanks. That's a really good answer. I come from another academic background than engineering so I appreciate pointers. / Getting just below the surface of MBDyn I sometimes feel I don't get enough out of the manuals to apply the components. It takes time-consuming experimentation to get through. Much is due to my inexperience in the field, of course. / If I prioritize simplicity, is there any package you would recommend? $\endgroup$ – ycc_swe Jul 26 '16 at 12:39
  • $\begingroup$ If you don't have experience in the area, a commercial package like MSC/Adams with its superior documentation, tutorials, and support will be easier. $\endgroup$ – Bill Greene Jul 26 '16 at 13:02
  • $\begingroup$ Thanks for your information. Very valuable for me. As I continue with MBDyn I will gradually widen my horizons in this area. / As I introduce a few more elements, there is much trial and error, but it usually works, finally. / Much is due to my inexperience and it is maybe a time consuming task to set up these models, unless you do it full time for long time and really get into it. $\endgroup$ – ycc_swe Jul 26 '16 at 13:10
  • $\begingroup$ Thanks for your reply. I am still checking around, using the information. But I still "fight it out" with MbDyn. For me it takes much time to find the settings. But I must admit MbDyn seems to work well if you give it the chance, read the manual and get the experience. Fortunately I have got the time. I was in contact with MSC Adams. They offered a test period but after that, for my needs, it would cost 12.000 Euro per year. But shorter periods can be obtained. (You better take all the courses first and prepare to work full time with it, when you lease it for a period, then.) $\endgroup$ – ycc_swe Aug 24 '16 at 13:09
1
$\begingroup$

The open source code HOTINT also has the capability to simulate multi-body systems, with and without deformation. This software can also be coupled with the DEM code LIGGGHTS to model the interaction of multi-body systems with particles or fluids (via the SPH approach).

| cite | improve this answer | |
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.