XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3 2031
turbulence (Vakamalla and Mangadoddy, 2023). The CAD
model of the cyclone as designed by the manufacturer was
imported into the software and the resulting watertight
interior geometry generated was discretized using a polyh-
excore mesh. These are shown below in Figure 3.
Prior studies have shown that the Reynolds Stress
Model (RSM) provides accurate predictions of the swirl
flow pattern, transport and anisotropy of turbulent stresses
(Gimbun et al., 2005). This was utilized to calculate the
stress components as well as for dissipation transport (3). A
summary of the schemes used is given in Table 1.
u ul u uli
D DL,ij Pij
G Fij
,
k
T ij
ij
i j i j
2t
2_t
2xk
2_tu
Qij fij
+=++
++++
i
(3)
where,
DT,ij =Turbulent energy diffusion
,ij =Molecular viscous diffusion
Pij =Shear stress generation
Gij =Bouyancy generation
Fij =Pressure strain
Fij =Pressure strain
Fij =System rotation generation
Each of these are calculated through separate equations
comprehensively detailed in the conventional RSM formu-
lation employed by the software (Ansys, 2016).
In hydrocyclones, accurately capturing the air core
requires the distinction of two phases along with the behav-
ior of the free surface under turbulent conditions (Wang
et al., 2009). The Volume of Fluid (VOF) model was
employed in this work, which tracks the interface by resolv-
ing the volume fraction of a given phase from a continuity
equation (3) based on the fluid velocity (Manninen et al.,
1996). The density and viscosity of the fluid medium is
calculated by adding the products of the two for each fluid
phase. The model then solves a momentum equation (4) to
determine the velocity that each phase shares.
uda n
n 2t
2a
+=0 (4)
u f 2t
2 d dp de t
n^duh
tg +=-++^duhT ++^^tuu h h o (5)
wvhere,
an =The volume percent of the nth phase
u =Fluid velocity
r =Fluid density
m =Fluid viscosity
f =Variable containing surface tension and fluid-
particle interaction
To represent the distinct flow trajectories of particles with
varying densities and sizes in the flow, the discrete phase
model (DPM) was used with two way fluid coupling (Aketi
Figure 2. Simplified DMS circuit for single stage DMS operation (adapted at Mintek)
turbulence (Vakamalla and Mangadoddy, 2023). The CAD
model of the cyclone as designed by the manufacturer was
imported into the software and the resulting watertight
interior geometry generated was discretized using a polyh-
excore mesh. These are shown below in Figure 3.
Prior studies have shown that the Reynolds Stress
Model (RSM) provides accurate predictions of the swirl
flow pattern, transport and anisotropy of turbulent stresses
(Gimbun et al., 2005). This was utilized to calculate the
stress components as well as for dissipation transport (3). A
summary of the schemes used is given in Table 1.
u ul u uli
D DL,ij Pij
G Fij
,
k
T ij
ij
i j i j
2t
2_t
2xk
2_tu
Qij fij
+=++
++++
i
(3)
where,
DT,ij =Turbulent energy diffusion
,ij =Molecular viscous diffusion
Pij =Shear stress generation
Gij =Bouyancy generation
Fij =Pressure strain
Fij =Pressure strain
Fij =System rotation generation
Each of these are calculated through separate equations
comprehensively detailed in the conventional RSM formu-
lation employed by the software (Ansys, 2016).
In hydrocyclones, accurately capturing the air core
requires the distinction of two phases along with the behav-
ior of the free surface under turbulent conditions (Wang
et al., 2009). The Volume of Fluid (VOF) model was
employed in this work, which tracks the interface by resolv-
ing the volume fraction of a given phase from a continuity
equation (3) based on the fluid velocity (Manninen et al.,
1996). The density and viscosity of the fluid medium is
calculated by adding the products of the two for each fluid
phase. The model then solves a momentum equation (4) to
determine the velocity that each phase shares.
uda n
n 2t
2a
+=0 (4)
u f 2t
2 d dp de t
n^duh
tg +=-++^duhT ++^^tuu h h o (5)
wvhere,
an =The volume percent of the nth phase
u =Fluid velocity
r =Fluid density
m =Fluid viscosity
f =Variable containing surface tension and fluid-
particle interaction
To represent the distinct flow trajectories of particles with
varying densities and sizes in the flow, the discrete phase
model (DPM) was used with two way fluid coupling (Aketi
Figure 2. Simplified DMS circuit for single stage DMS operation (adapted at Mintek)