Learning Forum

[bl1019]

Hi Community,

I am confused by the definition of the variable fvalmid in the subroutine uctrl1 from the umat dyn21.F. From the description found in the same file, fvalmid is an "array for predicting load curve values at time n+3/2", but what does n+3/2 mean? 

Further to this, I am using uctrl1 to dynamically update the values of a curve, and assign this curve as a pressure boundary condition used in my ICFD simulation (this is due to the dyn21icfd.F has yet come to effect).  However, I am aware that the method does not currectly update the value to the curve (the pressure I wish to use @time n is different from the value in my .out file). Do you have any suggestions for me to implement to fix this issue?

FYI, I am using R14.1.

Thank you.

[Ushnish Basu]

Binghuan,

I see that you are using the workaround suggested in this other thread: /forum/forums/topic/icfd-user-defined-boundary-condition-through-dyn21icfd-f/

You should know that we resolved the packaging issues with dyn21icfd.F, and the updated version works well. I will see how we can get you the updated version.

Also, n+3/2 means one-and-half steps ahead. The explicit solver in LS-DYNA uses velocity at half time steps. 

[bl1019]

 

Hi Ushnish,

Thank you for the umat package. Unfortunately, the package you sent to me was improperly packed – dyn21icfd.F is an empty file! Please see attched evidence below.

Yes, I was trying to implement the idea as suggested by another post. From there, by any chance you could kindly help me to double check whether the subroutines defined in dyn21icfd.F were pre-compiled in a library distributed with the pc-dyna umat? 

For example, although I overwrote the user-defined circuit, and set the WK type to 6 in my .k file, but from the console output, the program was executing the old example subroutine.

Best regards,

Binghuan

 

[bl1019]

Hi Ushnish,

Thank you for the suggestions. Yes, I confirm that the icfd user-defined subroutines worked in R14.1.

Best regards,

Binghuan