Hi all,
I am new to FE modeling with Ls dyna. Is there a way to model a partial hole in a cylindrical solid in ls pre-post? By that I mean the hole does not go through the entire cylinder.
Thanks
Are you talking about geometry creation? Geometry creation/editing has nothing to do with FE modeling or LS-DYNA.
You can use LS-PrePost to create a geometries of various levels of complexity but keep in mind that LS-PrePost isn't among the strongest CAD tools. CATIA, SolidWorks, Creo, etc. are professional commercially available tools. Tools like SpaceClaim are convenient geometry creation/editing tools. There are open source/free tools as well.
You will gain facility from watching YouTube videos or following tutorials in the documentation for a tool of your choice.
@icebreakersours thanks for the feedback I was referring to ls pre-post. I tried looking up videos on YouTube for creating holes in cylinders but could not find any of substance.
You are not going to find exactly what you want. The geometry you spoke of is very simple to make so you definitely need to think and learn how to create the geometry yourself. Again, LS-PrePost need not be the best tool for CAD creation.
@icebreakersours thanks for your help. I was able to create the model. Now I have a follow up question. I am trying to model a pile foundation so that the bottom of the pile is always in contact with the soil during vibration in any given direction without any separation. In Abaqus CAE, I was able to model this behavior using tie-constraint, however, I do not know if there is a similar feature in ls pre-post.
@kingoz79 good to hear you got the geometry created via Pre-Post.
The tie-constraint equivalent in LS-Dyna could either be accomplished through *Constrained_"fill in the blank" or a tied contact.
Is either part rigid and do you want them to have a rigid or breakable connection?
Without knowing more, it is difficult to give a general recommendation. That being said, I would argue it is preferable to have equivalenced nodes across interfaces - pile/foundation in this case. Unfortunately, very often, that is not practical in the sense that it may simply be challenging to make a mesh that has equivalenced nodes at interfaces in a reasonable amount of time given one's background and tools available to them. If you have to then repeat the process to cover a design space, then the expense in terms of man-hours becomes a hard constraint.
Next to that, you want to try to have similar sized element faces opposing one another with similar thicknesses going opposite ways from the interface. This helps with resolving contact appropriately and reducing numerical/discretization/.. errors in the region of interest - which you want to be as far away from the interface as is practically feasible.
As @negativevolume noted tied contacts or constrained cards work out well but, keep in mind, you have to play with special options if you don't have the same element types on either end of the interface or if some entities are rigid.