Content uploaded by Gerrit Basson
Author content
All content in this area was uploaded by Gerrit Basson on Oct 20, 2019
Content may be subject to copyright.
REMOVAL OF FINE NON-COHESIVE SEDIMENT BY SWIRL/
VORTEX SETTLING BASIN AT SMALL RIVER ABSTRACTION
WORKS
Kuria Kiringu1, Gerrit Basson2
Dept. Civil Engineering, Stellenbosch University, South Africa
1 kkiringu@gmail.com, 2grbasson@sun.ac.za
A vortex settling basin (VSB) offers a promising alternative to conventional sediment settling
structures, such as sand traps, for the removal of fine non-cohesive sediment for potable water use at
small river abstraction works of less than 100 l/s pump capacity (7.2 Ml/d at 20 h/d). The hydraulic
design of a suitable VSB was carried out by numerical CFD model. The design was optimized and
validated against two physical VSB models: 0.48 m diameter and 0.7 m high, as well as 0.68 m
diameter and 1.0 m high, in order to optimize the hydraulic design. The simulation results indicate
that a design with the following characteristics works well: inlet velocity =
0.26 m/s,
,
,
,
and deflectors. It was also established that sediment size and
concentration, play important roles in controlling the sediment trapping efficiency. The cone angle
and the angle of the inlet effects are minimal. Two VSBs designs for removal of 75µm sediment
particles at maximum inflow sediment concentration of 10,000 mg/l are proposed: (a) inflow of 5 l/s
with 5% water loss at a 99% trap efficiency and (b) inflow of 10 l/s with 8% water loss at 91% trap
efficiency.
KEY WORDS: Vortex Settling Basin; Swirl separator; Sediment removal; Settling; River
abstraction
1. INTRODUCTION
A Vortex settling basin (VSB) is a cylindrical fluidic device with a conical base where
sediment-laden flow enters tangentially to the flow domain, utilizing gravity and weak
centrifugal forces, more concentrated flow (underflow) exits at the bottom outlet and clear
water as overflow (Chrysostomou, 1983; Paul et al. 1991). VSB’s have small footprints,
no moving parts, no chemical dosing, high sediment removal rates and continuous flushing
of sediment back to the river, which makes them attractive for selection (Mashauri, 1986).
VSB’s have been applied widely in grit removal in wastewater treatment and stormwater
systems for the removal of coarse sediments, but have not often been implemented at river
abstraction works (Andoh and Saul, 2003; Field and O’Connor, 1996). It is the objective
of this study to give a better understanding of VSB separation mechanism for the removal
of non-cohesive sediment particles in the order of >75 µm for utilization in small river
abstraction works for African rivers. Numerical modelling was utilized to design and test
various VSB layouts and validation was undertaken on two physical models.
2. PHYSICAL MODELLING
Figure 1 shows a schematic of the physical model used in this study with the range of
model parameters summarised in Table 1. The inflow was supplied from an overhead tank
regulated by a valve and monitored at a flow meter with all the flow recycled back to the
tank to have a closed system. For each run, the flow was injected in the VSB domain and
allowed to stabilise and sediment particles were injected into the stream at a constant rate
to achieve a predetermined concentration. Both overflow and underflow particles were
captured on a filter, oven dried, the mass determined and the trapping efficiency calculated
as follows:
Figure 1: Schematic diagram of the vortex settling basin
Table 1: Vortex settling basin dimensions
Parameter
Symbol and unit
Model 1
Model 2
Inlet flow
Qi (l/s)
1-3
3-5
Cone height
Hc (mm)
250-1570
330
Cylinder diameter
D (mm)
480-1500
634
Cylinder height
Ht (mm)
240-1530
700-1000
Inlet diameter
Di (mm)
30-100
90-156
Inlet height
Hi (mm)
11.5-525
11.5-800
Outlet height
Ho (mm)
80
80
Inflow sediment concentration
C (mg/l)
10,000-50,000
10,000-50,000
Sediment particle diameter
d50 (µm)
75-112
75-112
Underflow diameter
Du (mm)
0-53
0-53
3. NUMERICAL MODELLING
Modelling of the VSB was undertaken with commercial CFD software FLUENT
ANSYS version19.1. The structured mesh was used with in-compressible continuity,
momentum and energy Navier-Stokes equations, discretized by Finite Volume Method.
Flow in the VSB domain is of moderate turbulent nature and ANSYS, (2013) have
recommended the use of realizable k- ε turbulence model but Griffiths and Boysan, (1996);
Cullivan et al., (2004) noted the results can have a deviation of 12% with the physical
model. Slack et al., (2000); Gimbun et al., (2005) recommended the use of Reynolds Stress
Model (RSM) models which was adopted for this study with standard wall function. To
simulate fluid-particle interaction, Volume of Fluid (VOF) simulated the interaction
between air-water phases and Discrete Phase Model (DPM) between water-sand particles
as by volume the particle fraction were less than 12% (ANSYS, 2003). To investigate the
effects of concentration Euler granular model was utilized allowing full interaction
between all phases. Grid sensitivity analysis was conducted and validated by
corresponding physical model ensuring grid independent results.
4. RESULTS AND DISCUSSION
4.1. INFLUENCE OF UNDERFLOW
From literature, no apparent trend could be observed from data undertaken by previous
studies as shown in Figure 2a. Curi et al. (1979); Mashauri, (1986); Paul, (1988)
underflow(Qu)/inflow(Qi) ratios were between 4% to 16%. This was investigated on
model 1 summarized in Table 1 with only the underflow being varied. Figure 2b
summarizes physical and numerical model results with different sediment particles: 75 µm,
100 µm and d50 = 112 µm. Having a ratio greater than 10% leads to air core formation
decreasing trapping efficiency thus a ratio 5% to 10 % is recommended as to achieve
maximum trapping efficiency with minimum water loss.
Figure 2: Influence of underflow on sediment trapping efficiency: a) investigation by various
authors b) model 1 numerical and physical model results
4.2. INFLUENCE OF INLET VELOCITY AND FLOW
Richardson et al. (2002) and Veerapen, (2003) established VSB is mainly gravity
driven and weak centrifugal forces aid in keeping the sediment particles in suspension near
the wall. Having high inlet velocities increase the centrifugal forces and secondary currents
which are detrimental to removal efficiency. This was investigated by varying the model 1
inflow and the influence of velocity and inflow was investigated for d50=112 µm, 100 µm
and 75 µm sediment particles. The numerical and physical model trap efficiencies results
are shown in Figure 3. Flow velocity and inflow are directly proportional and an inflow
velocity of 0.26 m/s is recommended. A dip in efficiency was experienced at a velocity of
0.20 m/s and it is due to destructive secondary currents/turbulence.
Figure 3: Simulated numerical and physical model impact of a) inlet velocity b) inflow on
sediment trapping efficiency
4.3. INFLUENCE OF INLET POSITION
This was investigated by varying the position of the inlet location relative to the
cylinder height while maintaining the Table 1 parameters constant. The resulting model 1
and 2 numerical and physical model results are shown in Figure 4. Trapping efficiency
increased with the inlet closer to the outlet with a ratio of 0.50 to 0.88 recommended. This
increase is counterintuitive and was experienced due to a higher percentage of secondary
currents moving towards the underflow at the inlet location.
Figure 4: Numerical and physical model impact of inlet position relative to cylinder height on
sediment trapping efficiency: a) model 1 b) model 2
4.4. INFLUENCE OF CYLINDER DIAMETER
Larger diameters of the VSB yield higher trapping efficiency however under a specific
inflow rate and inlet velocity, there exists a specific small diameter where the system will
act like a hydro cyclone or a large diameter where the system behaves like a sand trap. This
was investigated and Figure 5 shows the influence of cylinder diameter on trap efficiency
and residence time. It was concluded a ratio greater than 8.2 does not significantly increase
the residence time and this ratio is thus recommended for removal of fine particles.
Figure 5: Influence of cylinder diameter/inlet diameter on sediment removal efficiency and
residence time
4.5. INFLUENCE OF CYLINDER HEIGHT
Sullivan et al. (1974) and Chrysostomou, (1983) recommended Ht/D > 0.26 for
removal of coarse sediment. This was investigated for fine sediment and concluded the
influence of cylinder height is minor with a ratio Ht/D = 0.5 recommended.
4.6. INFLUENCE OF CONE AND INLET ANGLE
Due to the cohesive nature of washload in African rivers, a cone needs to be provided
to avoid clogging of the underflow. In the design of hoppers, a cone of 2:1 (V: H) has
extensively been used to ensure sustainability and is recommended. Varying the inlet angle
has no significant influence on the trapping efficiency and thus a tangential inlet is
recommended
4.7. PROPOSED LAYOUT
With the recommended parameters numerical model
supervised optimization was undertaken to yield the final model
configuration shown in
Figure 6. Performance evaluation was undertaken over varying sediment sizes,
underflows, inflow and sediment loading and to remove 75µm sediment particles at
maximum inflow sediment concentration of 10,000 mg/l two models are proposed: (a)
inflow of 5 l/s with 5% water loss at a 99% trap efficiency and (b) inflow of 10 l/s with 8%
water loss at 91% trap efficiency.
5. CONCLUSION
The numerical and physical model research presented here has given a better
understanding of the removal of fine sediment particles by VSB. The core findings can be
summarized as follows:
Gravity is the main driving removal mechanism assisted by weak centrifugal
forces.
Particles smaller than 75 µm cannot be effectively removed by a VSB due to
long hydraulic retention time required for gravity-driven mechanism.
An inlet velocity of 0.26 m/s should be maintained to provide adequate
centrifugal forces.
The Qu/Qi= 0.05-0.10 gives maximum trapping efficiency with minimal
water loss.
Having tall cylinders does not necessarily improve the trapping efficiency and
a ratio of Ht/D>0.5 is recommended
VSB’s with large cylinder diameters will behave like settlers and small
cylinders as Hydro cyclones operating under low pressure. A ratio of D/Di
=8.2 is optimum for fine sediment removal
The inlet should be placed closer to the outlet with a ratio Hi/H= 0.50-0.88
recommended. At this critical zone, strong secondary currents flowing
towards the underflow assist in the removal of sediment particles.
Figure 6: Proposed model dimensions
ACKNOWLEDGEMENTS
The authors are grateful to Council for Scientific and Industrial Research (CSIR) for
access to Centre for High Performance (CHPC) and the South Africa Water Research
Commission for their financial support.
REFERENCES
1. Andoh, R. Y. G., Saul, A. J. (2003). The use of hydrodynamic vortex separators and screening
systems to improve water quality. Water Science and Technology, 47(4), 175–183.
2. Ansys. (2003). Fluent 6.1. User’s Guide, (February), 7–24. https://doi.org/10.1111/j.1600-
0447.2011.01711.x
3. ANSYS. (2013). fluent tutorial guide, 3304(November), 724–746.
4. Chrysostomou, V. (1983). Vortex-type settling basin. (Unpublished master’s thesis) University
of Southampton, Southampton, England.
5. Cullivan, J. C., Williams, R. A., Dyakowski, T., Cross, C. R. (2004). New understanding of a
hydrocyclone flow field and separation mechanism from computational fluid dynamics.
Minerals Engineering, 17(5), 651–660. https://doi.org/10.1016/j.mineng.2004.04.009
6. Curi, K. V., Esen, I. I., Velioglu, S. G. (1979). Vortex type solid-liquid separator. Progress in
Water Technology, 7(2), 183–190.
7. Field, R., O’Connor, T. P. (1996). Swirl Technology: Enhancement of Design, Evaluation, and
Application. Journal of Environmental Engineering, 122(8), 741–748.
https://doi.org/10.1061/(ASCE)0733-9372(1996)122:8(741)
8. Gimbun, J., Chuah, T. G., Choong, T. S. Y., Fakhru’l-Razi, A. (2005). A CFD Study on the
Prediction of Cyclone Collection Efficiency. International Journal for Computational Methods
in Engineering Science and Mechanics, 6(3), 161–168.
https://doi.org/10.1080/15502280590923649
9. Griffiths, W. D., Boysan, F. (1996). Computational fluid dynamics (CFD) and empirical
modelling of the performance of a number of cyclone samplers. Journal of Aerosol Science,
27(2), 281–304. https://doi.org/10.1016/0021-8502(95)00549-8
10. Mashauri, D. A. (1986). Modelling of vortex settling chamber for primary clarification of water.
(Unpublished doctoral dissertation). University of Tampere, Tampare, Finland.
11. Paul, T. C. (1988). Designing circulation chamber sediment extractor (OD91). Hydraulics
Research Report, Wallingford, United Kingdom. Retrieved from
http://eprints.hrwallingford.co.uk/489/
12. Paul, T. C., Sayal, S. K., Sakhuja, V. S., Dhillon, G. S. (1991). Vortex‐Settling Basin Design
Considerations. Journal of Hydraulic Engineering, 117(2), 172–189.
https://doi.org/10.1061/(ASCE)0733-9429(1991)117:2(172)
13. Richardson, J. F., Harker, J. H., Backhurst, J. R. (2002). Particle Technology and Separation
Processes. October (5. ed., re, Vol. 2). Oxford: Butterworth Heinemann.
https://doi.org/10.1016/0009-2509(60)80030-9
14. Slack, M. D., Prasad, R. O., Bakker, A., Boysan, F. (2000). Advances in Cyclone Modelling
Using Unstructured Grids. Chemical Engineering Research and Design, 78(8), 1098–1104.
https://doi.org/10.1205/026387600528373
15. Sullivan, R. H., Parkinson, F. (1974). Relationship between diameter and height for the design
of a swirl concentrator as a combined sewer overflow regulator (Vol. EPA--670/2). National
Environmental Research Center, Office of Research and Development, US Environmental
Protection Agency; Washington. Retrieved from
https://nepis.epa.gov/Exe/ZyPURL.cgi?Dockey=9100TORJ.TXT
16. Veerapen, J. P. (2003). Solids removal in recirculating aquaculture: the swirl separator and
double-drain fish tank. University of New Brunswick, Fredericton, Canada.
