CFD Simulations AC3-10
Combining/dividing flow in Y junction
Application Challenge 3-10 © copyright ERCOFTAC 2004
Overview of CFD Simulations
The commercially-available CFD code CFX4, version 4.3, has been used.
The boundaries of the CFD model are the pressure tapping planes on each leg, located 0.457 metres from the ‘eye’ of the Y-junction. The geometry is available as an IGES file Yjunc.igs, and is also shown in plan view in Figure 6.
CFD calculations have been performed at a Reynolds number of 1.2x107, for both converging and diverging flow (corresponding to rig tests ‘D’ and ‘H’).
Two meshes were constructed, differing only in the mesh resolution near the wall. These are illustrated in Figures 7 to 10.
In the ‘coarse’ mesh, the 2.176 mm thick layer adjacent to the wall was divided into three equal cells of thickness 0.725 mm. The total number of cells in this mesh was 140,544.
In the ‘fine’ mesh, this 2.176 mm thick layer was divided into 8 cells, with an expansion ratio of 1.29171, so that the thickness of the cell nearest the wall was 0.1 mm, and the thickness of the eighth cell was 0.6 mm. The total number of cells in this mesh was 179,584.
Simulation Case
Solution strategy
The default settings of the CFX4 solver were used throughout, as listed below:
| Parameter | Equation Solver | Differencing Scheme |
|---|---|---|
| u velocity | Block Stone | Hybrid |
| v velocity | Block Stone | Hybrid |
| w velocity | Block Stone | Hybrid |
| pressure | Pre-conditioned Conjungate Gradient | Central |
| k | Line Solver | Hybrid |
| epsilon | Line Solver | Hybrid |
Computational Domain
The ‘coarse’ and ‘fine’ meshes were used, and the computational domain extended for a distance of 0.457 m from the eye of the Y-junction, along each branch. Upstream geometry was not modelled, and its effects were ignored.
Boundary Conditions
‘Mass flow boundaries’ were employed at all three legs of the Y-junction, and the flow rates were specified, i.e. any flows entering the domain were assumed to be fully developed (Neumann boundary condition). The inlet velocity profiles and turbulence parameters were therefore not specified directly – they were calculated by the code, and were those appropriate to fully developed flow in a circular pipe.
The walls of the pipe were assumed to be perfectly smooth, and standard wall functions were employed. Both the ‘k-epsilon’ and ‘differential stress’ turbulence models were examined.
Application of Physical Models
Numerical Accuracy
The computer runs were continued until the solution residuals were no longer decreasing.
Reductions of 4 or 5 orders of magnitude were obtained.
CFD Results
‘coarse’ mesh with ‘k-epsilon’ turbulence model:
case P1 (Pa) P2 (Pa) P3 (Pa)
D1 79.3 34292.9 34292.9
D2 83.8 37514.7 29704.3
D3 111.1 39869.0 23799.6
D4 156.6 41022.7 15315.8
D5 201.9 40188.5 3039.3
D6 241.0 35193.8 -15266.5
D7 412.3 24102.3 -43125.9
D8 586.2 5315.6 -80876.0
D9 669.1 -25969.7 -141872.3
H1 -10.0 3088 3088
H2 -10.0 11089.5 -5635.2
H3 -10.0 18941.8 -16041.8
H4 -10.0 27100.8 -29534.6
H5 -9.9 34728.1 -46696.3
H6 -9.9 41525.8 -69179.9
H7 -9.9 45737.9 -99184.1
H8 -10.0 45761.5 -141445.2
H9 -9.9 39479.6 -201817.3
‘fine’ mesh with ‘k-epsilon’ turbulence model:
case P1 (Pa) P2 (Pa) P3 (Pa)
D1 65.4 34592.1 34592.1
D2 68.6 37767.6 30047.6
D3 93.7 40018.7 24246.7
D4 137.4 40986.8 15936.5
D5 176.6 39773.6 3951.6
D6 213.2 34421.6 -14040.4
D7 364.9 22843.1 -41507.4
D8 527.0 3183.8 -79285.7
D9 624.3 -28783.6 -140314.4
H1 -10.1 2823.1 2823.1
H2 -10.1 10815.4 -5891.2
H3 -10.1 18664.6 -16311.6
H4 -10.1 26814.8 -29821.4
H5 -10.0 34415.0 -46999.5
H6 -10.0 41140.9 -69512.1
H7 -10.0 45292.9 -99600.5
H8 -10.1 45228.9 -142007.1
H9 -10.0 38918.6 -202729.4
‘coarse’ mesh with ‘differential stress’ turbulence model:
case P1 (Pa) P2 (Pa) P3 (pa)
D1 36.2 40285.3 40285.3
D2 35.1 43565.7 35536.3
D3 74.9 45817.0 29651.8
D4 157.5 46790.8 21241.4
D5 275.1 45653.5 9137.9
D6 375.0 40520.1 -8844.4
D7 523.3 29599.0 -35945.7
D8 718.8 10976.4 -72959.6
D9 888.9 -17818.0 -131989.0
H1 -145.0 -1614.0 -1614.0
H2 -145.5 6734.7 -10751.2
H3 -145.2 14984.9 -21617.0
H4 -145.6 23625.0 -35743.8
H5 -144.5 31899.1 -53636.4
H6 -144.7 39541.4 -77114.4
H7 -144.4 44924.1 -108445.8
H8 -145.2 46469.6 -152640.8
H9 -144.3 41696.3 -214698.4
‘fine’ mesh with ‘differential stress’ turbulence model:
case P1 (Pa) P2 (Pa) P3 (Pa)
D1 89.3 39780.1 39780.1
D2 88.8 42803.3 35309.6
D3 128.4 44798.6 29755.0
D4 213.0 45525.1 21753.9
D5 332.0 44189.1 10198.0
D6 399.2 38997.6 -6925.8
D7 537.2 27815.0 -33068.9
D8 747.4 8670.7 -69283.4
D9 918.2 -20644.6 -127872.3
H1 -145.5 -1648.1 -1648.1
H2 -145.9 6652.6 -10735.2
H3 -145.6 14791.9 -21513.9
H4 -146.1 23284.9 -35535.7
H5 -145.0 31363.9 -53325.9
H6 -145.2 38668.3 -76656.2
H7 -144.9 43462.0 -107774.1
H8 -145.6 44415.9 -151711.1
H9 -144.9 39438.9 -213667.7
© copyright ERCOFTAC 2004
Contributors: Alan Stevens - Rolls-Royce Marine Power, Engineering & Technology Division
Site Design and Implementation: Atkins and UniS