[FLASH-USERS] Multigrid Poisson solver - Boundary Condition Types

[Marco Mazzuoli]

    Dear all,

A brief introduction to my problem:
I have chosen FLASH_4a to simulate a Stokes turbulent boundary layer over a rough wall (provided by Immersed Boundaries method). Actually, I already have a DNS code based on the fractional-step method and a second-order accurate centered finite-difference scheme, but I have to improve the computational efficiency by using an AMR infrastructure (PARAMESH4). To do that, I think to use the FLASH code infrastructure, which provide a good interface with PARAMESH4 and an efficient parallel MultiGrid Poisson Solver in order to enforce the continuity.
However, I do not understand whether it is possible or not to use the MultiGrid Poisson Solver when the following Boundary Conditions (BC) being applied at the coarsest level/block (physical BC):

-x, +x : Periodic
-z, +z : Periodic
-y       : Dirichlet (no-slip)
+y     : Neumann (symmetry)

Indeed, I read at page 127 of the FLASH_4a's User Guide that "the supported boundary condition for the module (multigrid solver) are Periodic, Dirichlet, given-value and isolated" and "due to the nature of the FFT block solver, the same type of BC must be used in all directions. Therefore, only the value of bcTypes(1) will be considered in the call to Grid_solvePoisson" (bcTypes array contains six integer specifying the BC on the six faces of the block).

These are the questions:
- Can I use anyway the MultiGrid Poisson Solver provided by FLASH4 with my BC (listed above)?

- If yes, what did I misunderstand in the User Guide?
  Do you recommend, eventually, to utilize the hybrid PFFT/MultiGrid solver in my problem?
  
- Otherwise, which Poisson solver could I use (which supports PARAMESH4 and my BC) among the ones provided by FLASH_4a?

Thank you.
Sincerely,

    Marco Mazzuoli


Ing. Marco Mazzuoli
Dipartimento di Ingegneria
delle Costruzioni, dell'Ambiente e
del Territorio (DICAT)
via Montallegro 1
16145 GENOVA-ITALY
tel.  +39 010 353 2497
cell. +39 338 7142904
e-mail marco.mazzuoli at unige.it
       marco.mazzuoli84 at gmail.com




 		 	   		   		 	   		   		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://flash.rochester.edu/pipermail/flash-users/attachments/20111011/e30f5bd4/attachment.htm>