0
Research Papers: Flows in Complex Systems

# Hydrodynamic Characterization of a Nozzle Check Valve by Numerical Simulation

[+] Author and Article Information
Stefano Sibilla1

Dipartimento di Ingegneria Idraulica e Ambientale, Università di Pavia, via Ferrata 1, 27100 Pavia, Italystefano.sibilla@unipv.it

Mario Gallati

Dipartimento di Ingegneria Idraulica e Ambientale, Università di Pavia, via Ferrata 1, 27100 Pavia, Italy

1

Corresponding author.

J. Fluids Eng 130(12), 121101 (Oct 23, 2008) (12 pages) doi:10.1115/1.3001065 History: Received July 28, 2006; Revised March 14, 2008; Published October 23, 2008

## Abstract

The ability to obtain correct estimates of the hydraulic characteristics of a nozzle check valve by finite-volume numerical simulation is discussed. The evaluation of the numerical results is performed by comparison of the computed pressure drops inside the valve with experimental measurements obtained on an industrial check valve. It is shown that, even with high mesh refinement, the obtained result is highly dependent on the choice of the turbulence model. The renormalization group theory (RNG) $k-ε$ model proves to be the more accurate to describe the flow inside the valve, which is characterized by repeated flow decelerations and accelerations and by boundary layer development under adverse pressure gradient. Pressure-drop and flow coefficients computed by adopting the RNG model agree well with the experimental values at different positions of the plug. The opening transient of the valve is also analyzed by an unsteady flow simulation where the motion of the plug is taken into account. The characteristic curve of the valve obtained in steady flow conditions is finally compared with the transient opening characteristic, highlighting a temporary increase in the pressure drop, which occurs because of a large unsteady separation region downstream of the plug.

<>

## Figures

Figure 1

Experimental setup: (a) nozzle check valve geometry; (b) sketch of the test section with upstream and downstream pressure probes P1 and P2

Figure 2

Computational mesh, full valve opening: coarse mesh (a) and fine mesh (b)

Figure 3

Effect of the inflow boundary condition: axial velocity profiles at the inlet (left) and in the throat section upstream of the plug (right). Uniform flow (solid), 1/7 power profile (dashed); logarithmic profile (dash-dot).

Figure 4

Dependency of the valve pressure-drop coefficient on the turbulence model and on the mesh refinement

Figure 5

Turbulent kinetic energy field. Simulation with (a) standard and (b) RNG k-ε models. Isolines range (a) from 0 to 0.36V2, step 0.03V2; (b) from 0 to 0.12V2, step 0.01V2.

Figure 6

Pressure profile along the external valve wall: comparison between different k-ε turbulence models

Figure 7

Axial velocity profiles of three sections shown by dashed lines in Fig. 5: simulation with different k-ε turbulence models

Figure 8

Turbulent kinetic energy profiles of three sections shown by dashed lines in Fig. 5: simulation with Chen’s and RNG k-ε turbulence models

Figure 9

Flow field in the fully open valve configuration. (a) Pressure: isolines range from −5ρV02 to 2ρV02, step (1/2)ρV02. (b) Velocity magnitude: isolines range from 0 to 3V0, step 0.25V0.

Figure 10

Velocity vector field in the wake region of the fully open check valve. The vector length is uniform to highlight flow direction and recirculation regions.

Figure 11

Check valve characteristic curves: (a) pressure-drop coefficient; (b) flow coefficient. The symbols indicate results corresponding to the plug positions analyzed by numerical simulation; the lines indicate experimental results obtained by increasing the flow rate from zero to the maximum value (solid) and then decreasing again to zero (dashed).

Figure 12

Pressure field in steady flow at different plug positions. (a) 50% travel; (b) 10% travel; detail of the plug region in the circle. Isolines range from −5ρV02 to 2ρV02, step (1/2)ρV02.

Figure 13

Velocity magnitude field in steady flow at different plug positions. (a) 50% travel; (b) 10% travel. Isolines range from 0 to 3V0, step 0.25V0.

Figure 14

Velocity vector field in steady flow at different plug positions. (a) 50% travel; (b) 10% travel. The vector length is uniform to highlight flow direction and recirculation regions.

Figure 15

Computational mesh for ALE simulation: detail at t=2.12R/V0 (50% open, i.e., after 200 time steps)

Figure 16

Time history for plug travel and flow rate in the opening transient simulation

Figure 17

Velocity magnitude fields during plug opening. (a) t=0.25to, (b) t=0.5to, (c) t=0.75to, (d) t=to, (e) t=1.15to, and (f) t=1.4to. Isolines range from 0 to 3V0, step 0.25V0.

Figure 18

Check valve characteristic curves during plug opening: (a) pressure-drop coefficient; (b) flow coefficient. Transient simulation (symbols) compared with steady-state simulations (dashed) and with experimental results obtained by increasing the flow rate from zero to the maximum value (solid).

## Errata

Some tools below are only available to our subscribers or users with an online account.

### Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related Proceedings Articles
Related eBook Content
Topic Collections

• TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
• EMAIL: asmedigitalcollection@asme.org