International Journal of Hydropower and Civil Engineering 2024; 5(2): 01-06

E-ISSN: 2707-8310
P-ISSN: 2707-8302
www.civilengineeringjournals.c Numerical modelling of axial outlet hydrocyclone
om/ijhce
IJHCE 2024; 5(2): 01-06
Received: 02-06-2024 S Shilpakar and HP Pandit
Accepted: 03-07-2024

S Shilpakar DOI: https://doi.org/10.22271/27078302.2024.v5.i2a.43

Institute of Engineering,
Pulchowk, Nepal Abstract
Nepal possesses huge networks of perennial rivers and together with the steep topography, the rivers of
HP Pandit Nepal prospects enormous hydroelectric potential. However, due to significant amount of sediments in
Institute of Engineering, the rivers, the hydropower plants suffers considerable losses due to damages in turbines and other
Pulchowk, Nepal hydromechanical equipment’s. Since the gravity-based settling basins, designed for handling
suspended sediments has been found inefficient, this study focuses on use of centrifugal separation
method for better handling of suspended sediments. This paper presents the simulation of axial outlet
hydrocyclone using ANSYS Fluent software. The simulation is first validated with the physical model
in terms of both water distribution and sediment throughput. Subsequently, the hydraulic performance
of the device is analysed by studying various velocity profiles and headloss in the device. Finaly, the
device performance is assessed at various angle of inclination of its axis at 90o, 60o, 53o, 45o and 30o to
horizontal with an intent to reduce the overall height of device considering difficulties in excavation in
rugged topography of Nepal. It was observed that the sediment separation efficiency decreased,
particularly for very fine sediments with reduced headloss by 0.071 m as the axis of device was
changed from 90o to 30o to horizontal.

Keywords: 3D Model, ANSYS Fluent, CFD, Headloss, Hydrocyclone, Hydropower

Introduction
Sediment is a mixture of various organic and inorganic materials that is carried by the rivers
along with the water from one place to another. And Nepal’s river system are the ones that
are responsible for transporting highest sediment loads to the ocean. The estimated total
specific yield of the country is about 4240 tons/km2/year (Bajracharya et al. 2008) [1]. Major
rivers like the Narayani often record sediment loads as high as 25,000 ppm (Carson 1985) [5].
Similarly, sediment load as high as 50,000 ppm has been found in smaller river like Jhimruk
(Basnyat 1997) [3]. The reason behind high sediment yield is due to fragile geology of Nepal
as Nepal is situated in seismically active area at convergence of Indian and Eurasian plates,
the steep topography of Nepal together with the combination of heavy monsoon rainfall
spanning between June and September during which the country receives 55-80% of its
annual rainfall (Basnet et al. 2020) [2], and the influence of South Tibetan Detachment
Surface (STDS) which marks the boundary between Indian and Eurasian Plates is considered
as most active tectonic features in the world that is responsible for erosion and transport of
large volume of sediment to the rivers (Pandit et al. 2008) [9].
Because of the sedimentation problem, many hydropower turbines in Nepal suffers severely
causing reduced plant efficiency, unplanned outage and requires frequent repair and
maintenance. For instance, the Jhimruk Power Plant is one of the severely affected power
plants by river sedimentation and the study shows that the turbine efficiency drops by 8%
within just 2 months (Chitrakar and Neopane 2019) [7]. Similarly, the turbine of Kaligandaki
‘A’ HEP has undergone five major maintenances in between year 2002 and 2014 including
replacement of runner (Chhetry et al. 2014) [6]. Generally, Settling Basins are designed for
excluding the suspended sediments coarser than 200 microns in hydropower plants has been
found inadequate in Himalayan regions like Nepal (Pandit et al. 2008) [9]. Khimti
Hydroelectric Project experiences significant turbine wear and tear despite effectively
trapping 97% of particles larger than 200 microns and 85% of particles exceeding 130
microns (Deshar 2007) [8]. A Pelton turbine operating at high head of 920 m showed severe
Corresponding Author:
S Shilpakar
erosion and cavitation after just 600 hours of operation because of abundance of 77%
Institute of Engineering, particles finer than 63 microns and 99% particles finer than 125 microns (Brekke et al. 2003)
[4]
Pulchowk, Nepal .

~1~
International Journal of Hydropower and Civil Engineering https://www.civilengineeringjournals.com/ijhce

An attempt to trap such fine sediments would require large overall height of device considering the difficulties in
settling basin, which will increase the capital cost and also excavation in rugged geography of Nepal.
the space required to house such large settling basin is The model setup in ANSYS Fluent includes following steps:
generally not available in mountainous topography of
Nepal. Since centrifugal separation methods has been Geometry
successfully employed in industries like oil and gas The 3D geometry resembling the test rig of physical model
refineries as well as food and beverages industries, this serves as the foundational step in the research process. The
research focuses on studying application of hydrocyclone dimensions of the test rig are presented in Table 1 and the
for improved exclusion of suspended sediments in 3D model of test rig is illustrated in Figure 2.
hydropower projects.
Hydrocyclone is a device that separates fluid from solid Table 1: Dimensions of test rig of Physical Model (Pandit et al.
particles or separate one fluid from another based on their 2008) [9]
difference in density by the action of centrifugal force. The S.N. Parameter Unit Measurement
schematic diagram of a typical hydrocyclone device is 1 Diameter of hydrocyclone m 0.38
presented in Figure 1. 2 Height of cylindrical part m 0.50
3 First cone angle deg 18
4 Second cone angle deg 6
5 Height of first conical part m 0.40
6 Height of second conical part m 1.35
7 Dimension of inlet m 0.055 x 0.11 (B x H)
8 Diameter of overflow m 0.035
9 Diameter of underflow mm 15 - 60
10 Length of vortex finder m 0.19

Fig 1: Schematic diagram of hydrocyclone (Vega-Garcia et al.

2018)

As the fluid with suspended particles enters the

hydrocyclone rapidly through a tangential inlet, it initiates a
spiral motion. The particles undergo centrifugal force
pushing them towards the wall and drag force pulling them
inward due to the fluid. Heavier particles, experiencing
stronger centrifugal force, hit the wall early and exit through
the underflow outlet after losing momentum. The conical Fig 2: 3D model of test rig
section's decreasing radius raises pressure, pulling in air
from underflow. Swirling air exits through overflow, Meshing
carrying finer particles and fluid, enabling their escape due The generation of mesh for fluid domain is performed in
to local drag. ANSYS workbench for numerical analysis. Tetrahedral
elements are used with a global element size of 20 mm.
Methodology Three layers of inflation are added with the first layer
The flow inside the hydrocyclone is very complex and thickness of 2 x 10-3 m at a growth rate of 1.2 to capture the
turbulent. Therefore, the study of the device is carried out near wall behavior more accurately. The mesh quality
using numerical modeling approach by using ANSYS analysis is performed as per ERCOFTAC 2018 guidelines.
Fluent (Version 2019 R1), a commercial CFD program. The The number of elements in different configuration of device
numerical model is performed to study the behavior of axial is tabulated in Table 2.
outlet hydrocyclone. Initially, the numerical model is
validated with the physical model conducted by H.P. Pandit Table 2: Mesh element numbers for different configuration of
(Pandit et al. 2008) [9], considering both hydraulic and device
sediment trapping aspects. The hydraulic behavior of device
Angle of inclination
is assessed by studying different velocity profiles and 90o 60o 53o 45o 30o
with horizontal
headloss within the device. Lastly, the device’s performance Number of elements 220,912 284,626 259,466 239,314 243,135
at various angles of inclination of its axis (90o, 60o, 53o, 45o
and 30o with horizontal) is studied thereby reducing the

~2~
International Journal of Hydropower and Civil Engineering https://www.civilengineeringjournals.com/ijhce

Boundary Conditions atmospheric pressure. For the walls, non-slip conditions

The hydrocyclone simulation involved setting the inlet were assigned and partial slip condition was assumed for the
boundary as "mass flow inlet." Both outlets were defined sand, with a specularity coefficient of 0.5. The summary of
with the boundary condition of "pressure outlet" at boundary conditions set is shown in below.

Table 3: Summary of Boundary Conditions

Boundary Conditions Settings
Inlet: mass flow rate of water 16.60 l/s
Inlet: mass flow rate of sand 0.028 kg/s
Overflow and Underflow Outlets Pressure Outlet at atmospheric pressure
Walls No slip for water and partial slip for sediments

Numerical Simulation is calculated using the following formula:

For numerical modeling of device, multiphase model was
used in which water is considered primary phase and E= (1)
sediment as secondary phase. The density of water and
viscosity is set as 998.2 kg/m3 and 0.001 kg/m/s. Sediment Where
size ranging from 1 micron to 400 microns with a density qsu = Rate of sediment received from underflow outlet in kg
ranging from 2500 kg/m3 for finer particles to 2680 kg/m3 /s,
for coarser particles were used. qso = Rate of sediment received from overflow outlet in
For the numerical simulation hydrocyclone, the swirl kg/s.
modification RNG (k-epsilon) model is selected as this
model is an improvement over traditional RANS models The comparison of sediment trapping efficiencies for
such as ƙ-ε (k-epsilon) model. Although this model requires various sizes of sediments is illustrated in Figure 3.
high computational resources, it provides more accurate
results in complex flows or near walls (Yakhot and Orszag
1986) [11].
The convection term was treated using an upwind scheme,
and the pressure term was solved using the SIMPLE
algorithm. A total of 20,000 iterations were set to capture
the turbulent behavior of flow and to ensure convergence
criteria of 5 x 10-4 to reach a stable and accurate solution.

Results and Discussion

Validation of Numerical Model
The results of numerical modeling of axial outlet
hydrocyclone is compared with the results of physical
model (Pandit et al. 2008) [9]. The primary focus was to Fig 3: Comparison of sediment trapping efficiencies for various
verify the hydraulic performance for which the continuity of sizes of sediments
flow was compared with the results of physical model. The
results depicting the comparison of flow rate from different To quantitatively assess the numerical model's performance
outlets for different inlet discharge is presented in Table 4. against the physical model, statistical parameters were
checked. The R2 value was calculated as 0.96 while the
Table 4: Comparison of flow rates between physical and RMSE and PBIAS values were computed as 4.407 and -3.40
numerical model
respectively.
Discharge passing Physical Model Numerical Model
S.N.
through (l/s) (l/s) Hydraulic Performance of Device
Test Inlet 17.20 17.20 Velocity Profile
No. Overflow Outlet 14.68 15.35 The post processed result of simulation depicting the
S2-1 Underflow Outlet 2.52 1.84 velocity vector of water is presented in Figure 4. The
Test Inlet 19.40 19.40 velocity streamlines illustrate that the maximum velocity is
No. Overflow Outlet 17.30 17.33 observed near the inlet, where the fluid flow is accelerated
S2-3 Underflow Outlet 2.10 2.06
due to the reduction in cross-sectional area. The velocity
Test Inlet 17.85 17.850
gradually decreases towards the bottom of the
No. Overflow Outlet 15.45 15.972
S2-5 Underflow Outlet 2.4 1.874
hydrocyclone. Additionally, there is an upward flow in the
Test Inlet 16.60 16.60 opposite direction, causing the fluid to exit the hydrocyclone
No. Overflow Outlet 14.33 14.52 through the overflow outlet.
S2-8 Underflow Outlet 2.27 2.07 Different velocity profiles were assessed to better
understand the hydraulic behavior of device. The flow
After achieving the similar continuity of flow, a detailed velocity inside the hydrocyclone can be split into three
numerical simulation was carried out involving the mixture components as shown in Figure 5.
of sediments and water. The sediment trapping efficiencies

~3~
International Journal of Hydropower and Civil Engineering https://www.civilengineeringjournals.com/ijhce

Fig 4: Velocity vector showing magnitude and direction of flow inside device for discharge of 16.60 l/s

Fig 5: Tangential, Axial and Radial velocity profiles inside hydrocyclone for discharge of 16.60 l/s

Tangential velocity 2. Radial velocity

Tangential velocity is the most crucial and important The radial velocity distribution in a hydrocyclone
velocity component in the hydrocyclone. From Figure 5 it typically follows a pattern in which the velocity is
can be observed that the tangential velocity is higher near highest near the outer wall and decreases as you move
the walls at the greater radial distance and minimum at core. towards the center of axis.
Because of this, the particles near the walls are exposed to
higher centrifugal force which causes them to accelerate Headloss in the device
while the middle core area with low velocity allows the The headloss in the device was assessed by measuring the
water (lighter media than sediments) to be carried away difference in total pressure between inlet and overflow
from the overflow outlet. outlet and the headloss of 0.758 m was estimated for the
1. Axial velocity discharge of just 16.60 l/s. Considering the short
Axial velocity refers to the velocity in the direction of conveyance and small discharge, the headloss in the device
cyclone axis. In Figure 5, the outer walls of the is very high which is due to turbulent flow induced by the
cylindrical and conical sections exhibit higher velocity, centrifugal force and longer spiral flow path.
indicated by negative axial velocity, directing flow
downward. Whereas in the middle core the velocity Performance of Device at Different Inclination
decreases and it is directed upward with positive axial The study is carried out to check the sediment trapping
velocity. This downward flow from the outer efficiency for different sizes of sediments at different
circumferential region is crucial for guiding sediment inclination of device axis. The figure comparing the
particles to the underflow outlet while the upward flow trapping efficiency of sediment for different orientation of
near the central axis facilitates the recirculation of clear device is shown in Figure 6.
water towards the overflow outlet.

~4~
International Journal of Hydropower and Civil Engineering https://www.civilengineeringjournals.com/ijhce

Fig 6: Trapping efficiencies for different sizes of sediments at different orientation of device

It is observed that as the inclination of device is more sediments as shown in Figure 7. It is observed that more
towards horizontal, the trapping efficiency of device is sand particles escaped from overflow outlet while tilting the
reduced, particularly in case of very fine sediments. The device axis from vertical to 30o with respect to horizontal.
same can be observed in the velocity streamline of

Fig 7: Velocity streamline of sediment for different inclination of device

The headloss in the device at different orientation of its axis headloss is desirable, however it comes with the cost of
is shown in Table 5. It is observed that as the device is reduced trapping efficiency, particularly for very fine
inclined from 90o to 30o towards horizontal, the headloss in sediments.
the device is reduced by around 0.071 m. This decrease in

~5~
International Journal of Hydropower and Civil Engineering https://www.civilengineeringjournals.com/ijhce

Table 5: Headloss in device at different inclination of device for discharge of 16.60 l/s
Angle of inclination with horizontal 90o 60o 53o 45o 30o
Headloss (m) 0.758 0.709 0.702 0.695 0.687

Conclusion 6. Chhetry B, Thapa B, Thapa B. Assembly design and

In this study, the comprehensive study and analysis of axial turbine maintenance in sediment–laden conditions. Int J
outlet hydrocyclone has been carried out. The numerical Hydropower Dams. 2014;21(2):82-88.
model was first validated with the physical model. The 7. Chitrakar S, Neopane HP. River sedimentation
output of numerical model closely matched with the challenges for hydropower development in Nepal: A
physical model exhibiting similar water distribution and perspective review. Kathmandu; c2019.
sediment throughput. The model results were compared 8. Deshar IOM. Experience in Sediment Management.
both visually and quantitatively. The quantitative Workshop on Operations, Repair and Maintenance of
assessment involved the utilization of various statistical Hydropower Plants. Kathmandu: BPC and Statkraft
tools such as Coefficient of Determination (R2), Root Mean Norfund Power Invest AS; c2007.
Square Error (RMSE) and Percentage Bias (PBIAS), and 9. Pandit HP, Shakya NM, Stole H, Garg NK. Sediment
whose values were determined to be 0.96, 4.407 and -3.40 exclusion in Himalayan rivers using hydrocyclone. ISH
respectively. Furthermore, a fish hook effect can be J Hydraul. Eng. 2008;14(2):118-133.
observed in Figure 3, which further validates the capability https://doi.org/10.1080/09715010.2008.10514909.
of model to accurately capture the distinct behavior of 10. Vega-Garcia D, Brito-Parada PR, Cilliers JJ.
hydrocyclone device. Optimising small hydrocyclone design using 3D
Further the device’s hydraulic performance was studied by printing and CFD simulations. Chem. Eng. J.
analyzing various velocity profiles and headloss within the 2018;350:653-659.
device. The tangential velocity, vital for sediment https://doi.org/10.1016/j.cej.2018.06.016.
separation, increases from the central axis to the walls due 11. Yakhot V, Orszag SA. Renormalization group analysis
to centrifugal force. Axial velocity moves downward near of turbulence. I. Basic theory. J Sci. Comput.
the outer wall and upward near the core, while radial 1986;1(1):3-51. https://doi.org/10.1007/BF01061452.
velocity peaks at the outer wall and decreases towards the
center. The observed velocity profiles matched with the
typical velocity profiles inside the hydrocyclone device. A
headloss of approximately 0.76 m was observed for a
discharge of just 16.60 l/s.
Lastly the device’s performance was assessed under
different angle of inclination of its axis at 60o, 53o, 45o, and
30o with horizontal. The results showed that while adjusting
the device’s axis from vertical to 30o to horizontal, a
reduction in headloss was observed by 0.071 m. This
decline in headloss is desirable, however it comes at the cost
of lowered sediment separation efficiency, especially for
very fine sediments.

References
1. Bajracharya TR, Acharya B, Joshi CB, Saini RP,
Dahlhaug OG. Sand erosion of Pelton turbine nozzles
and buckets: A case study of Chilime Hydropower
Plant. Wear. 2008;264(3-4):177-184.
https://doi.org/10.1016/j.wear.2007.02.021.
2. Basnet K, Shrestha A, Joshi PC, Pokharel N. Analysis
of Climate Change Trend in the Lower Kaski District of
Nepal. Himal. J Appl. Sci. Eng. 2020;1(1):11-22.
https://doi.org/10.3126/hijase.v1i1.33536.
3. Basnyat S. Monitoring sediment load and its abrasive
effects in Jhimruk Power Plant, 2nd International
Course on Small Hydropower Development.
Kathmandu; c1997.
4. Brekke H, Wu YL, Cai BY. Design of Hydraulic
Machinery Working in Sand Laden Water. Ser Hydraul
Mach. Imperial College Press; World Scientific
Publishing Co; c2003 .p. 155-233.
5. Carson B. Erosion and Sedimentation Processes in the
Nepalese Himalaya; ICIMOD Occasional paper No. 1.
Kathmandu: International Centre for Integrated
Mountain Development (ICIMOD); c1985.

~6~