but about IB is "old" & "new",

i agree that concept IB is old, there are some about this in old books, if we look at old books on FDM we see somethig like interpolation based IB, e.g. Ritchmeyer's book or Zienkiewicz & Morgan's book (Finite Elements and Approximation)

but is a new research area, maybe closed and re-opened recently

rt : I use IBM in a very specific case, I'm not sure my work can help you. I work with High order Compact schemes combined with FFT for Poisson resolution. My research area is around DNS and LES flow at low Reynolds.

It's very singular, but if you are always interesting by my work, I can give you one paper and if you can read French I can give my PhD.

Sincerely,

u r right, ur research area is not interesting for me,

but as an expert people in IB, could you please give your idea (probably don't rely on num. experience) about my questions.

also u solve poisson with FFT, i don't have experience with FFT but guess that it is cpu intensive, especially DNS enforce using fine grid, is it?

and why FFT? how is computational cost?

is your system stiff and classical iterative methods fail to converge?

are sparse direct solvers better condidate than FFT (consider robustness and cost)?

[siamakghh2000]

Does anyone know if it is possible to include the effect of roughness on the solid walls by employing immersed boundary method?
As far as I know none of the methods mentioned below can do this.

1-Continous forcing method.
2-Discrete forcing method.
3-Ghost cell IBM.
4-Cut cell method.

[siamakghh2000]

Does anyone know if it is possible to include the effect of roughness on the solid walls by employing immersed boundary method?
As far as I know none of the methods mentioned below can do this.

1-Continous forcing method.
2-Discrete forcing method.
3-Ghost cell IBM.
4-Cut cell method.

[sbaffini]

Roughness effects can only be included in wall function models, independently from the specific approach (IBM, body fitted, etc.). The underlying reasoning is that if you cannot resolve the wall details
you cannot resolve the wall at all.

As a consequence, if your IBM can handle wall functions (not all of them can) it can also handle wall roughness (if you know how to include it in the wall function model).

However, the Ghost cell method is actually one of several different approaches for the discrete forcing method.

[siamakghh2000]

Thanks. I imagine IBM is more suitable to apply to low Reynolds flows where roughness effect is insignificant.

[sbaffini]

Well, there are a number of reasons for which the IBM is more suitable at low Re. However, high Re applications are possible as well. It is important to choose the most suitable IBM flavor for the given application.

For wall roughness you need a method which allows using wall functions. Thus, i would exclude continuous forcing, penalization approaches (with porous like terms) and VOF/level set methods.

[siamakghh2000]

Thanks. I think it would be easy to implement wall function if the solid body coincides with the grid points completely. Otherwise, do you know any publication which may have tried a similar attempt?

[sbaffini]

Look for the works of Capizzano or De Tullio, i don't remember now the titles (just add IBM to their names).

The group working on the code NASCART-GT has also produced in this field.

However, the idea is trivial. The wall normal velocity gradient computed by the wall function just has to substitute the one entering the the computation of the viscous term at walls.

The specific details obviously depend from the method.

[leilii]

Originally Posted by sfar
;49891

Hi,

I intend to carry my research on Immersed Boundary Method. I have already read lots of articles on IB method. Now what I need actually is a piece of code illustrating implementation of the IB method. Do anyone know any source from where i can get some initial idea about IB coding? I will prefer finite volume discretisation of the governing equations. Thanks in advance.

Sfar

hi mr safar ,can i take your mail? i have some quiztion about immersed boundray

[leilii]

Originally Posted by Junseok Kim
;49929

http://math.konkuk.ac.kr/kwon/board....ter2007&page=1

hi
this link doesnt open for me
can you send me thats contain?
i need ibm code
tnnnxx