Learning Forum

[Ben_Ben]

I displaced some beam elements axially in a first step and now I want them to springback in the second step. I am using the dynain method so I have defined the following card:

*INTERFACE_SPRINGBACK_LSDYNA
$#    psid      nshv     ftype         _    ftensr    nthhsv     rflag   intstrn
         5         0         0         0         0         0         0         0

Here is the material model:

*MAT_SHAPE_MEMORY
$#     mid        ro         e        pr  
        63 6.4500E-9   51700.0       0.3
$# sig_ass   sig_asf   sig_sas   sig_saf      epsl     alpha      ymrt    
     600.0     670.0     288.0     254.0     0.063       0.0   47800.0




The second model seems to be set up correctly. I copied over the *PART, *SECTION, and *MAT keywords and also included the *INITIAL_STRESS_BEAM card:


The model runs successfully. In the second simulation step. But no springback is seen. No deformation of any kind is registered. They should springback to the diameter of the red shell that surrounds the beams. 

Is there a card I need to apply to get the stresses to be applied in the second simulation?

[Reno Genest]

Hello Ben,

Have you used the dynain file in the 2nd analysis? Please review appendix X in the LS-DYNA User Manual Vol I:

Let me know how it goes.

 

Reno.

[Ben_Ben]

Hi Reno, 

Yes I copy the contents of the dynain file to the 2nd simulation input file and then I copy all the part a material data from the 1st sim file. And yes I have seen the appendix X.

 I can see all the initial stresses in my .k file for the 2nd sim, so I can't work out why its not springing back. Are there any limitations to using dyna with beams? Or is there a step I'm missing? 

Cheers, 
Ben

[Reno Genest]

Hello Ben,

In the 2nd simulation, do you see stresses at time zero in the d3plot? If not, something is wrong. Make sure you setup the *INTERFACE_SPRINGBACK_LSDYNA keyword according to the best practice. Please read appendix X and set the keyword according to the recommendations.

You will find more information on springback with example  models here:

https://ftp.lstc.com/anonymous/outgoing/support/FAQ/springback

Study this resources and try some of the models. I would try to simplify your model and see if you can make it work.

Also, make sure you use the latest LS-DYNA sovler (R14.1). 

https://ftp.lstc.com/user/ls-dyna/R14.1.0/windows/

https://ftp.lstc.com/user/mpp-dyna/R14.1.0/windows/

username: user

password: computer

 

Let me know how it goes.

 

Note that I will be on vacation for the next 2 weeks starting today. Hopefully you will be able to figure this out or someone else will be able to help you.

 

Reno.

[Reno Genest]

Hello Ben,

Also, try with the dynain.lsda format.

You could also try with *MAT_024 instead of memory alloy; there might be a limitation with shape memory alloy. 

If your model is small, please post the keyword text in one post and someone might be able to look at it and help you.

 

Let me know how it goes.

 

Reno.

[Reno Genest]

Hello Ben,

Also, check in the messag file (SMP) or the mesXXXX files (MPP) for warning messages. If there is a limitation, it might show up there. I checked in our database and beam elements should be supported. Make sure to use the dynain.lsda format. I didn't find anything about *MAT_SHAPE_MEMORY and I just asked the developer. He is on vacation right now and he will be back next year. I will be back on January 8th.

 

Reno.

[Reno Genest]

Hello Ben,

I got a reply from the developer and there is currently limited support for *MAT_SHAPE_MEMORY and dynain.lsda. You would have to use the latest DEV solver. If you have access to technical support, you can file a case and we will send you the latest DEV solver for you to try.

customer.ansys.com

 

Reno.