Hi,

I have some problems on SIMPLE algorithm used in 3D r-\theta-z cylindrical coordinates.

(1) About the SIMPLE used in staggered grid for 3D cylindrical coordinates. I have written a code, but the code is not converged! and I have wasted somedays on it, but I don't know why. In fact, I have implemented a implicit SIMPLE code in collocated grid, and a explicit projection method code in staggered grid, and these two codes is very good. Are there someone who has successed to apply SIMPLE to staggered grid in 3D cylindrical coordinates, and are there some tricks in the code?

(2) About the Krylov method combined with SIMPLE algorithm on collocated grid. I know the combination implementation for staggered grid because of the mass flux can be calculated by the velocity directly, but for collocated grid, I have to using another varibles (FORTRAN) to save the mass flux after pressure correction, which is different from that caculated by the velocity directly! I want to know how to treat the mass flux when the Krylov method combined with SIMPLE algorithm on collocated grid?

Thanks in advance!

zouchu

[Ali Farahani]

Originally Posted by zouchu
;25592

Hi,

I have some problems on SIMPLE algorithm used in 3D r-\theta-z cylindrical coordinates.

(1) About the SIMPLE used in staggered grid for 3D cylindrical coordinates. I have written a code, but the code is not converged! and I have wasted somedays on it, but I don't know why. In fact, I have implemented a implicit SIMPLE code in collocated grid, and a explicit projection method code in staggered grid, and these two codes is very good. Are there someone who has successed to apply SIMPLE to staggered grid in 3D cylindrical coordinates, and are there some tricks in the code?

(2) About the Krylov method combined with SIMPLE algorithm on collocated grid. I know the combination implementation for staggered grid because of the mass flux can be calculated by the velocity directly, but for collocated grid, I have to using another varibles (FORTRAN) to save the mass flux after pressure correction, which is different from that caculated by the velocity directly! I want to know how to treat the mass flux when the Krylov method combined with SIMPLE algorithm on collocated grid?

Thanks in advance!

zouchu

Hi
I have the same problem too. in fact I have descritized momentum eq. in cylindrical coordinates in implicit mode. but my error in pressure correction do not converge any way.
I wanna know may be there is some common mistake that I've forgot to consider.
does implicit or explicit mode of discritization have any special effect of converging the results?

any help would be appreciated.
ali.farahani87@gmail.com