LS-DYNA MAT_108 Help

[chiefchester]

Hi all,

First off, I am new to LS-DYNA so any suggestions as to where to look for help are greatly appreciated.

Here's my question. I want to use the material MAT_108 (*MAT_ORTHO_ELASTIC_PLASTIC) for a 2D shell model (a heavily loaded precast wall). I put in all the parameters but, I just can't get it to work. LS-DYNA errors out and tells me that the material (which I'm 100% sure I've defined) does not exist. Is there anyone out there that has experience with this material? Is there something I'm missing that a more experienced person would easily know? I've looked at the manuals and I've checked the theory and I can't figure it out. Is there another keyword I need to define? Thanks for your help.

 

[gwolf2]

It most probably means that the reference to the material in the shell property definition is incorrect. Check the MAT_108 material ID field on the material line of the input deck and the material ID in the shell property line on the input deck. They should be the same integer.

 

[chiefchester]

Thanks for your quick response.

Unfortunately, I pretty sure this is not the problem. I created the Material with a unique ID and I assigned the material to the model using the *Part keyword.

I've also run the entire model using MAT_001-elastic and it runs fine so I'm convinced that the problem lies with the definition of MAT_108. Any other ideas?

 

[gwolf2]

Not without going through the deck with a fine tooth comb and a DYNA manual - sorry.

 

[chiefchester]

It's appalling that FEA errors are so poorly documented and difficult to understand. I can understand working with a program for a while, getting to understand it, etc. But when a program that costs thousands of dollars outputs text files of gibberish that I have to struggle to decipher, something is very wrong. [end of rant]

Thanks anyway for your help. Anyone else have an idea?

 

[gwolf2]

I sympathise, this is a very common problem with the old NASA freeware programs (like NASTRAN and DYNA) which companies package up and sell on. They have at their heart really old bits of research FORTRAN dating from the 1970's.

If you want a good Explicit code try ABAQUS/Explicit - it was actually designed rather than evolved and the documentation is excellent.

Good luck.

gwolf

 

[chiefchester]

Hey, so the problem was solved (unfortunately not by me). The issue was that I was using LS-PrePost v2.4 and I guess we needed to use v3.0. One of the cards (card 5 and/or 6) was not being defined correctly. Essentially, it was a user input error.

However, I still stand by my rant regarding the ease of use of LS-DYNA. If you happen to read this post and are planning to make a computer program, just know that there's no such thing as too little documentation.

gwolf, thanks for your help.

 

[gwolf2]

Glad you got to the bottom of it. It was roughly what I thought but almost impossible to find unless you are older than 40 and familiar with old-school FE input decks.

Next time you are up for licence renewal check out other programs.