Learning Forum

[minchunh]

Hello,

I'm modeling a tension specimen under displacement control by applying prescribed displacement to the node set, while from the output data, the motion of the node set is not moving as I prescribed. I'm using solid element -1. I wonder if anyone faced this kind of problem before? 

Thanks,

Min-Chun

[Armin Abedini]

Hi Min-Chun,

Could you please provide more information regarding your model?

For example, what LS-DYNA card do you use to apply the motion? Do you use *BOUNDARY_PRESCRIBED_MOTION_SET?
Do you apply displacement to the specimen's end along the right direction? Do you constrain the other end of the specimen to resemble a fixed clamped condition?

[minchunh]

Hi Armin, 

Yes, I'm using *BOUNDARY_PRESCRIBED_MOTION_SET as well as applying fixed-clamped conditions to the other end. Also, at the end of applying the prescribed motion, Boundary_spc is also applied to constrain the DOFs other than the direction under displacement control. I tried applying other magnitudes of the cyclic displacements, but the problem only occurs when certain prescribed displacement curves are used (i.e., this issue does not always occur, which is weird).

[Armin Abedini]

Thanks Min-Chun.

I see. As you indicated, it's odd that the issue does not always occur. I would suggest that you remove the BOUNDARY_SPC card for now to focus on ensuring the *BOUNDARY_PRESCRIBED_MOTION_SET is working properly. Once it's fixed, you can add more cards.
I don't think element type play any important role here as the problem seems to be more relevant to the prescribed boundary conditions.     

[minchunh]

Hi Armin, 

I only applied *BOUNDARY_PRESCRIBED_MOTION_SET at first but the issue showed up so I added BOUNDARY_SPC. Thanks anyway, I will keep looking into it. 

[Armin Abedini]

No problem Min-Chun.

While you investigate the model, it is worth checking that the generated set includes all the nodes at the one end of the specimen. Also, try applying a velocity (instead of displacement) boundary condition by changing VAD from 2 to zero in the *BOUNDARY_PRESCRIBED_MOTION_SET card and providing a velocity curve to the model. I usually get better luck with velocity BCs for tensile test simulations.

[Andreas Koutras]

Hi all,

Some examples that may help with the modeling of coupon tests are available here:

https://www.dynaexamples.com/introduction/intro-by-a.-tabiei/tension