CFD simulation based on collocated grid (Rhie and Chow) products non-smooth velocity and pressure distribution in main flow direction. I think the reason of above problem may caused not to satisfy mass conservation of cell centered velocity.

Do you know how to solve this problem?

For solving a continuity equation we need face velocity values. If we calculate face velocity by simple average of main grid point values it gives pressure velocity osicialltions. In order to avoid it, Rhie and Chow relations were used which couples pressure and velocity that satisfies the conservation of mass. Please correct if I am wrong.

Right! The cell-centered velocity components in the Rhie and Chow scheme satisfy the continuity equation on the assumption of linearly varying pressure conrection. In usual case it does not cause proiblem. However, it can result in a physically incorrect solution when the flow involves a large body forece and the grids are not fine enough. See "Use of the momentum interpolation method for flows with a large body force" Numerical Heat Transfer Part B Vol.43 pp 267-287, 2003. This paper shows the cause and remedy of this problem.

Halim

Thank you very much! I tried to calculate the flow in bent pipe, but inconsistent pressure wave apprears in main flow direction.

I will look for the paper you show.

I used the scheme presented by following paper and solved my problem. Lien, F.S. and Leschziner, M.A. (1994b), "A General Non-Orthogonal Collocated Finite Volume Algorithm for Turbulent Flow at all Speeds Incorporating Second-Moment Turbulence-Transport Closure, Part 1: Computational Implementation", Computer Methods in Applied Mechanics and Engineering, 114,123-148.

Check Versteeg and Malalasekerea's finite volume book, they have some stuff on decoupling - Rhie and Chow can be found in the source mentioned in Halim's (I think) message.