0
Research Papers: Fundamental Issues and Canonical Flows

# Experimental and Numerical Investigations of a Turbulent Flow Behavior in Isolated and Nonisolated Conical Diffusers

[+] Author and Article Information
F. Aloui

GEPEA UMR-CNRS 6144, École des Mines de Nantes, Département Systèmes Énergétiques et Environnement, 4 rue Alfred KASTLER, BP20722, 44307 Nantes Cedex 03, France; Département de Physique, Faculté des Sciences et des Techniques, Université de Nantes, 2 rue de la Houssinière, BP 92208, 44322 Nantes Cedex 03, FranceFethi.Aloui@univ-nantes.fr

E. Berrich

GEPEA UMR-CNRS 6144, École des Mines de Nantes, Département Systèmes Énergétiques et Environnement, 4 rue Alfred Kastler, BP20722, 44307 Nantes Cedex 03, France; Département de Physique, Faculté des Sciences et des Techniques, Université de Nantes, 2 rue de la Houssinière, BP 92208, 44322 Nantes Cedex 03, France; Centre Technique des Industries Mécaniques de Nantes (CETIM), 74, route de la Jonelière, 44300 Nantes, FranceEmna.Berrich@univ-nantes.fr

D. Pierrat

Centre Technique des Industries Mécaniques de Nantes (CETIM), 74, route de la Jonelière, 44300 Nantes, FranceDaniel.Pierrat@cetim.fr

J. Fluids Eng 133(1), 011201 (Jan 13, 2011) (10 pages) doi:10.1115/1.4003236 History: Received April 22, 2009; Revised November 12, 2010; Published January 13, 2011; Online January 13, 2011

## Abstract

In some industrial processes, and especially in agrofood industries, the cleaning in place mechanism used for hydraulic circuits plays an important role. This process needs a good knowledge of the hydrodynamic flows to determinate the appropriate parameters that assure a good cleaning of these circuits without disassembling them. Generally, different arrangements are present in these hydraulic circuits, such as expansions, diffusers, and elbows. The flow crossing these singularities strongly affects the process of cleaning in place. This work is then a contribution to complete recent studies of “aliments quality security” project to ameliorate the quality of the cleaning in place. It presents experimental and numerical investigations of a confined turbulent flow behavior across a conical diffuser $(2α=16 deg)$. The role of a perturbation caused by the presence of an elbow in the test section, upstream of the progressive enlargement, was studied. The main measurements were the static pressure and the instantaneous velocity fields using the particle image velocimetry (PIV). Post-processing of these PIV measurements were adopted using the $Γ2$ criterion for the vortices detection and the proper orthogonal decomposition (POD) technique to extract the most energetic modes contained in the turbulent flow and to the turbulent flow filtering. A database has been also constituted and was used to test the validity of the most models of turbulence, and in particular, a variant of the shear stress transport (SST) model.

<>

## Figures

Figure 1

Experimental installation

Figure 2

Sketch of the test section (dimensions in mm)

Figure 3

Synthesized image of the vein with the pressure taps

Figure 4

Principle of measurements using the PIV system

Figure 5

Measurements zones of the velocity fields by PIV

Figure 6

Static pressure distribution along the diffuser on each generating line (A, B, C, and D) and for Re=77,000

Figure 7

Velocity profiles at the inlet of the conical diffuser for the Reynolds numbers: Re=37,000 and Re=77,000

Figure 8

Experimental installation and Dean vortices, according to the calculation (14)

Figure 9

PIV primary results for isolated conical diffuser and for the Reynolds number Re=37,000: (a) Global velocity field; (b) zoom on the region near the wall

Figure 10

Effect of the presence of the inlet elbow at 10Dh upstream of the conical diffuser and for the Re=37,000: (a) Global velocity field; (b) zoom on the region near the wall

Figure 11

Sampling of histograms of the axial velocity in the isolated conical diffuser for the Re=37,000 and X/Dh=1.41: (a) Y/Dh=0.66; (b) Y/Dh=−0.66

Figure 12

Sampling of histograms: Effect of the presence of the inlet elbow at 10Dh upstream of the conical diffuser for Re=37,000 and X/Dh=1.4: (a) Y/Dh=0.66; (b) Y/Dh=−0.66

Figure 13

Effect of Reynolds number: Velocity profiles at X=1.4Dh in the isolated conical diffuser for the different Reynolds numbers

Figure 14

Effect of the presence of the inlet elbow at 10Dh: Velocity profiles at X=1.4Dh for the Reynolds number Re=37,000, respectively, for isolated conical diffuser and nonisolated conical diffuser

Figure 15

Tensor of Reynolds in the nonisolated diffuser for Re=37,000: (a) Diffuser preceded by an elbow at 20Dh and (b) diffuser preceded by an elbow at 10Dh

Figure 16

Example of an instantaneous vorticity field (ω) in the isolated conical diffuser for Re=37,000: (a) ω obtained from the PIV measurements and (b) ω obtained by POD (11 modes) after post-processing

Figure 17

Instantaneous vorticity field ω obtained in the nonisolated conical diffuser for Re=37,000: (a) Directly from the PIV measurements and (b) by POD (11 modes) superimposed with the streamlines

Figure 18

Instantaneous Γ2 criterion obtained from PIV measurements after post-processing and for Re=37,000: (a) For isolated conical diffuser (2α=16 deg) and (b) for nonisolated conical diffuser (2α=16 deg) proceeded by an elbow located at 10Dh upstream of the diffuser

Figure 19

Velocity profiles for an isolated diffuser for Re=37,000 at X=0.6Dh

Figure 20

Axial velocity profiles in the nonisolated diffuser for Re=37,000 at the inlet of the conical diffuser: (a) Elbow situated at 20Dh; (b) elbow situated at 10Dh

## Discussions

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.

Topic Collections