SIMPLE algorithm for collocated storage
Hi,
I'm trying to write a 2d FV solver but I can't get the SIMPLE algorithm to work. I get the basics of how it should work but it just keeps adding to the pressure at each iteration until pressure=NaN. Do you know of a good reference that describes how to code a SIMPLE solver for a collocated, structured grid arrangement. (has to be collocated as I can get it to work with a staggered grid). Preferably written in as simple english as possible! Thanks Alex 
Re: SIMPLE algorithm for collocated storage
the difference between staggered and collocated is that you should have correction term for calculating fluxes on cell faces. have you read the paper of Rhie and Chow?
Rhie, C.M. and Chow, W.L. (1983). "Numerical Study of the Turbulent Flow Past an Airfoil Trailing Edge Separation", AIAA Journal 21(11), 15251532. 
Re: SIMPLE algorithm for collocated storage
hi, A very good reference for SIMPLE in colocated grid is the book 'computational fluid dynamics' by ANIL DATE. If you cant get this, you can refer to some of the papers published by the author.

Re: SIMPLE algorithm for collocated storage
Hadian, Sudhakar,
Thank you both for your help. The book you recommended by Anil Date was indeed a very good reference. Highly recommended to anyone else interested in SIMPLE for colocated grids. There is one thing I'm a little unsure of though. In implementing the dP'/dn=0 boundary condition, (on the west boundary for example) I set the west coefficient of the pressure correction to zero. However I still need a value of AW for AP(=AW+AE+AN+AS) (where AW & AP, etc. are coefs. of the pressure correction equation). AW involves ap (ap = coef. of momentum equation) at the west face, which is obtained by linear interpolation between nodal values for internal CVS. How do I obtain ap at a boundary? Thanks 
Re: SIMPLE algorithm for collocated storage
hi alex, I think you are little confused. The issue is straightforward. Near the west boundary, aW=0; substitute this in your calculation of aP, then it becomes aP=aE+aS+aN. Once you implement this, you will experience that your iterative solver takes slightly longer time to converge. hope it helps.

All times are GMT 4. The time now is 00:52. 