Learning Forum

[Webster]

Hi Community,

I am using LS-DYNA ICFD *ICFD_BOUNDARY_WINDKESSEL model to prescribe the outlet pres. conditions to the branches of my vessel geometry. Anyone knows how the pressure is calculated from the circuit? Does ICFD automatically retrieve the flux and pres. from the prev. iteration to calculate the current pres. on the surface?

I compared the model deployed in ICFD with a prev. in-house model in CFX, ith the same geometry and BCs, yet the solutions are very different. I consulted the user mannual, but unfortunately the description was quite vague...

Thanks.

[igandiko]

Hi, I checked with ICFD developers on this query. It seems this keyword was recently changed, so the pressure update is slightly different in current dev version compared to older versions. The updated manual description of this card is here (https://ftp.lstc.com/anonymous/outgoing/imtiaz/ICFD_BOUNDARY_WINDKESSEL.pdf).

Please see full response from the developer below. 

"The *ICFD_BOUNDARY_WINDKESSEL is a keyword that was recently changed in current dev version. In all cases, the windkessel condition is akin to a pressure boundary condition, we end up imposing a pressure on the boundary. This pressure is provided by the windkessel circuit.

To sum it up :

Pre current dev version: the new pressure would be updated function of quantities coming from the previous timestep like the flux (also output if you use ICFD_DATABASE_FLUX) and the pressure at the previous timestep (beware of the sign that is applied on the flux quantities, can be inverted with the choice of wtype). This update simply uses the circuit equation in a basic explicit way to calculate :

Pressure_new = f (Pressure_old, flux )

Current Dev version : New options for circuits have been added. Pressure update is now also done in a more robust way that still uses the circuit equation as a basis but rearranges the term so that the new pressure is no longer function of pressure at previous step, thus making it more robust. The keyword description will be also be updated and made clearer. This new description is not yet available in the general manual but I attach a PDF copy of what it will soon look like for the user’s benefit. Remarks 1 and 2 are still valid for pre-current-dev versions of the solver as the equations solved are the same, only the way we are solving them internally has been modified. Remark 3 and 4 pertain to two brand new circuits that have been added recently."

[Webster]

Hi,

This is great news. The new Windkessel implementation looks very robust, and the manual descriptions are very clear.

Yet, I wonder whether these features have been released and come into the effect? I am currently depolying my model with LS-PrePost 4.10 (the December release), but could not locate the aforementioned features in the Keyword Manager.

Any help is greatly appreciated.

[Webster]

[Webster]

Dear ICFD Dev,

 

Further to my question above, I wonder if you could clearify the use of notations in the manual. I was particularly confused by the meaning of P', Q', and Q'' in the equations. Thank you.

[igandiko]

Those notations are time derivatives. The new options were implemented recently in the dev version. You will need dev builds dated 12/25/2023 or later.

These cards may be supported in LSPP v4.11 or newer versions. v4.10 will not support it. 

[Webster]

Dear ICFD Dev,

Thank you for your reply. I am about to use icfd_usercircuit user subroutine with WTYPE = 6. Does this user defined feature only supported in current dev package (R14.0 later)? I noticed in the previous object releases, dyna21icfd.F is available but yet being included in the makefile.

I retrieved the current dev object package (ls-dyna_smp_d_Dev_106853-g20403fda1b_winx64_ifort190_lib.zip, downloadble from lstc) but found the .F files are currently empty. Hence, I wonder if it is safe to migrate the dyna21icfd.F from the prvious release to the current dev folder for compilation. Or, could I simply compile this user defined subroutine with the previous object builds?

Thank you.

 

[Webster]

I also suspect the subroutines defined in dyn21icfd.F were precompiled in a certain library, hence, any user modified changes could not be reflected during the execution. Probably the simplist example from my side is I can see the same characters got printed out in my history file, although I have modified my write statements and re-compiled the code.

[Webster]

Dear Imtiaz,

By any chance there is any response from our ICFD expert(s)? Thank you for helping out!

Best,

B