I am a first-year graduate student researching the nature of sand-particle collisions with ductile surfaces. About a month ago, my research group obtained an LS-Dyna license, and since then, I have been toying with the software, trying to create basic collision models (elastic sphere colliding with elastic surface kind of thing). I am the first member in the group tasked with learning how to use the software, and I feel kind of stuck. I have investigated many of the Dyna examples from dynaexamples.com and have watched many YouTube videos, but that has not gotten me to the desired general understanding of the software and its underlying mechanisms.
My advisor recently told me we had enough funding for me to get some official training on the software. Do you guys have any advice on what training is most helpful? Right now, we are considering the training offered directly by ANSYS listed here along with these third-party training groups.
I will add that my background up to this point has been in pure physics, so I do not yet have a deep understanding of FEM or general structural mechanic (though I am currently taking a course in continuum mechanics). If you feel that I will need to do some independent study before any of the Dyna training courses will be of use, that information would be very helpful as well.
@Jacob_22, personally I have not taken any courses on LS-DYNA but rather just struggled through basic examples presented on Youtube and extended that same methodology to the LS-DYNA examples that you mentioned. This youtube channel has been the most helpful for me with regards to setting up a basic model:
https://www.youtube.com/c/UnpopularMechanics/videos
This channel has also been helpful, specifically for the theory behind contacts:
https://www.youtube.com/c/OasysLtd/videos
There is also the theory manual for LS-DYNA that goes into detail about most of the inner workings of LS-DYNA but I will warn you that it is not an easy read:
https://www.dynasupport.com/manuals/additional/ls-dyna-theory-manual-2005-beta
I will point out that ANSYS has some general courses (not specific to LS-DYNA) that cover the basics of FEA:
I personally have used the ANSYS learning hub to take many courses related to ANSYS (fluent, autodyn, mechanical) and they were all very informative and helpful, you can also potentially get six months free for the student ANSYS learning hub if you contact them directly. Most of my learning with LS-DYNA has been through downloading the models from the examples page, trying to re-create them, and changing parameters to see how they influence the model. I will add that if you dont know there is the pre-post processor in LS-DYNA called LS-PREPOST that I have found to be essential for navigating all the options available in LS-DYNA.
I hope some of this information is helpful!
Thank you so much for the received advice! It is nice to know that I am not alone in this journey of learning Dyna.
My research group luckily has some funds for training, so we decided to purchase George Laird's (PredictiveEngineering) instructional package found here. We heard many good things about it. I'll be sure to write some sort of review on this forum after working through it if one hasn't been created already.
@pjay Those YouTube channels are certainly very helpful. I especially relate to your quote:
Most of my learning with LS-DYNA has been through downloading the models from the examples page, trying to re-create them, and changing parameters to see how they influence the model.
Before deciding to seek training, my main strategy was to treat the software as a lab experiment, varying parameters and trying to infer their effects. However, there are so many parameters that I eventually came to the conclusion that the effort was futile :/
@negativevolume I agree that it is important I hone in on the aspects of FEM relevant to my area of application. As I had mentioned, I have not yet been exposed to a survey course of FEM, so I am not aware of the subfields of FEM. It will probably be of use for me to look into that and at least identify the specific concentrations to begin investigating. SPH sounds promising, and the LSTC course you found looks perfect though I'm not sure if it is still being offered. Perhaps I will reach out to the instructors directly.
Hi @jacob_22,
@pjay offered up some good advice. Most of the time, the best way to learn how to use LS-Dyna and the various pre and post-processors like LS-PrePost are thru trial and error with example models and tutorials etc. One thing to keep in mind, things that may seem trivial in real life can be exponentially more difficult in the finite element analysis modeling space. Just a disclaimer to not get too frustrated.
Professional training is a great way to learn the software if you can afford it of course. My suggestion would be to spend some time like you are doing, working through examples and youtube tutorials to get a basic handle on the software itself and the structure of the model infrastructure (keywords). Then once you have that basic understanding, you may have a little bit of a better idea of where you need to focus and where you can best use professional help.
The reasoning behind this: FEA is a bit of a specialized field, but still encompasses an extremely large variety of applications, each with their own "math". There are big differences in how you would approach simulating a car crash to study occupant safety vs studying wave propagation. If you are studying sand particle collisions, you will probably want to focus on something like Smooth Particle Hydrodynamics (SPH) or some other form of studying particle interaction. I found this introduction to particle methods in LS-Dyna that might be perfect for you.