858

IWA Publishing 2012 Water Science & Technology

66.4

2012

Utilizing settling tests to design a conventional upow

settling tank modied with inclined plates
Noori M. Cata Saady

ABSTRACT
This paper examines the relationships between the turbidity removal efciency (TRE), the surface
overow rate (SOR), and the detention time (Dt) in settling column and jar tests, as well as the
performance of a conventional upow settling tank modied with inclined plates in the upper zone.
The experimental results showed that the SOR obtained from the occulent settling column test can be
increased by 30% and the corresponding Dt can be decreased by 75% with a variation in TRE of less than
7%. The TRE of occulent settling in the jar test coincided with the performance of the modied upow

Noori M. Cata Saady

Department of Civil and Environmental
Engineering,
University of Windsor,
401 Sunset Ave.,
Windsor,
ON, N9B 3P4,
Canada
E-mail: nsaady@uwindsor.ca

settling tank, while the results of the settling column test were slightly different. For plain settling, the
SOR obtained from jar and settling column tests should be divided by 3 and 2, respectively, before
possible use in the design of the modied upow settling tank. Two empirical models with 1.0% error
in the TRE predictions were developed to facilitate the design of the modied upow settling tank.
Key words

| high rate settling, inclined plates settling, jar test, settling column test, settling test,
upow settling tank

INTRODUCTION
Inclined plate (lamella) settlers are widely used in drinking
water treatment (Hassan & Hassan ), domestic and
industrial wastewater treatment (Saleh & Hamoda ;
Kolish & Schirmer ), upow anaerobic sludge blanket
(UASB) reactors (Halalsheh et al. ), activated sludge
reactors ( Jardin et al. ), storm-water treatment (Clark
et al. ), and the segregation of mineral particles (Zhou
et al. ). Inclined plates increase the hydraulic capacity
and improve the efuent quality for existing conventional
settling tanks. In large scale application (2,555 m3 h
1), tube
settlers increased the hydraulic capacity by 40% while they
removed turbidity (<30 nephelometric turbidity units
(NTU)) and phytoplankton algae (4559 cell mL
1) better
than conventional clariocculators (Hassan & Hassan ).
A new design and conguration of inclined plates increased
the capacity of a settling tank which handled water, sewage
and storm water by 300% (Fujisaki ).
Recent studies optimized the performance of settling
through improved designs of inclined plate settlers (Fujisaki
; Shevidi et al. ). For example, Shevidi et al. ()
reported that multistages of tube settlers, with the tube diameter decreasing from stage to stage, performed better than a
single stage design. Computational uid dynamic (CFD) modeling has been used to optimize the effects of different inlet
doi: 10.2166/wst.2012.260

designs on the hydraulic performance and the distribution

of suspensions to the ow channels between the adjacent
plates (He et al. ; Okoth et al. ; Salem et al. ).
Saady () veried experimentally the models used in the
design of inclined plate settling tanks and examined the
effects of the entry transition length on the turbidity removal
efciency (TRE). Sarkar et al. () applied dimensional
analysis to experimental results and showed the effects of geometric parameters such as distance between plates, plate
angle, length of plate, plate roughness, number of plates,
and particle diameter on the efciency of the inclined plate
settlers.
The solids removal efciency of settling tanks depends
on the surface overow rate (SOR). Settling column and
settling in jars are used to determine the properties of suspensions. The settling column test establishes the solids
removals at various times and depths, and enables the
designer to determine the corresponding SOR and detention
time (Dt) for the design of settling tanks. The designers use
SOR obtained from the settling column test to predict the
performance of settling tanks handling a continuous ow.
Practically, the design SOR obtained from a settling
column test is reduced by 6080% (WEF ), and the corresponding Dt is increased by 175200% (Metcalf & Eddy

859

N.M.C. Saady

Utilizing settling tests to design inclined plate settling tanks

); both are then used in the design of conventional

settling tanks. No studies determined any correlation
between the results of the settling tests and the design of
the settling tanks with inclined plates. Moreover, an important application is that when an existing conventional
settling tank is uprated by using inclined plates. How the
results of the settling tests can be used to design such
tanks for occulent and plain settling is not well dened.
This study determines experimentally the relationship
between the results of the settling test (settling column and
jar tests) and the performance (TRE) of a conventional
upow settling tank modied with inclined plates in the
upper zone for plain and occulent settling of raw water
from a river.

MATERIALS AND METHODS

Experimental setup
Experiments were carried out in a laboratory-scale setup
(Figure 1) consisting of a ground-level suspension preparation tank, an elevated tank, a mixing (coagulation
occulation) tank, an alum solution feeding tank, and a
model of an upow settling tank equipped with inclined
plates in the upper zone (Figure 2). The upow velocities
were 1.0, 2.0, 3.5, 5.0, and 7.0 m h
1 and the corresponding
Dt were 26, 13, 7, 5, and 4 min, respectively. The detention
time (Dt) has been calculated from dividing the settling
tank volume by the discharge. The optimum angle of plate
in the system tested was found to be 60 above horizontal
and it was determined experimentally. The turbidity of inuent was about 350 NTU.
W

Water Science & Technology

Figure 2

Schematic drawing of the laboratory equipment.

2012

pycnometer test and their particle size distribution (Figure 3)

was determined using the hydrometer test (Das ). The
relationship (Equation (1)) between turbidity (T ) in NTU
and concentration of SS in mg L
1 was determined experimentally; the relationship was linear with a regression
coefcient (R 2) of 0.9997. This relationship is specic to
the raw water tested and is a function of the solids properties. Aluminum sulfate Al2(SO4) 16H2O (Sigma, USA) was
used to prepare a 1% strength alum solution which was used
as a coagulant at an optimum dose of 45 mg L
1, this optimum dose was determined experimentally for the raw
river water tested.
T 0:30(SS)0:95

(1)

Analytical methods
The turbidity was measured using a turbidimeter (Hatch,
model No 2100 A). SS concentration was determined by ltering a 200 mL sample of suspension using a Buckner
funnel through 45 m lter paper (Whatman No. 40, UK)

The specic gravity of the suspended solids (SS) in the raw

river water was 2.617 and was determined using the

66.4

Details of the settling tank and the compartment of inclined plates.

Materials

Figure 1

Figure 3

Particle size distribution of the SS in the river raw water.

860

N.M.C. Saady

Utilizing settling tests to design inclined plate settling tanks

and drying for one hour at 1035 C. The pH was measured

using a pH meter (Of Pye Unicam model 292). All samples
and measurements were taken in triplicate.
Settling tests
Turbidity and SS were measured for inuent and efuent
samples taken from the inlet of the mixing tank, and the
outlet of the settling tank, respectively. The ow rate was controlled by a valve and measured by a ow rate meter as well
as manually by a cylinder and a stopwatch. The suspension prepared was left 57 min for initial plain settling in a 1.0 m-deep
tank in order to allow large particles (>40 m) to settle, and
to maintain a nal turbidity of about 350 NTU. The temperature
of the suspension in the settling tank was 30 2 C. No attempt
was made to maintain a constant temperature in the experiments. No pH adjustments were made and the pH value of
the treated water ranged between 7.2 and 7.4.
The settling column tests were conducted in a transparent tube of 15 cm diameter and a length of 180 cm. The
column had three sampling points spaced at 50 cm intervals.
After stirring gently for uniform distribution of particles over
the full depth, the tests were started. Samples were drawn
from the sampling points at 10, 20, 40, 60, 90, and
120 min. Jar tests were conducted in a ZETEL sedimentation jar tester (ZETEC Environmental Control Ltd),
samples were taken at a 6 cm depth below the water surface
by a pipette at 1, 2, 3, 4, 5, 7.5, 10, 20, and 30 min. Settling
column and jar tests were conducted for the two cases of
occulent and plain settling. For the occulent settling
both in the column and jar tests, alum was used at the optimum dose (45 mg L
1). Turbidity, SS, and pH were
measured for the samples from the settling column and jar
tests.
W

Water Science & Technology

2012

RESULTS AND DISCUSSION

Settling column test
The TRE versus SOR relationships for upow settling tanks
(with and without plates), settling column, and jar tests are
shown in Figure 4 for occulent settling and in Figure 5
for plain settling. As the SOR decreased, the TRE of plate
settling coincided with that predicted by jar tests and was

Figure 4

TRE versus SOR for occulent settling.

Figure 5

TRE versus SOR for plain settling.

For the quiescent settling column test, the SOR was computed by dividing the distance from the water surface to
the sampling point by the settling time which is the time
of sampling measured from the start of the test. Similarly,
for the jar tests, the SOR was computed in the same way
as for the case of the quiescent settling column. In the continuous settling in an upow settler with inclined plates in
the upper zone, the SOR was computed using Equation (2)
(Clark et al. ).
Vo
sin L cos

66.4

where Vo is the average ow rate velocity inside the ow

cell, L is the relative length (l/d ) where l is the length of
the plate and d is the distance between any two adjacent
plates, is the angle of plate inclination above horizontal.
The average ow rate velocity (Vo) has been calculated by
dividing the discharge on the total cross sectional area of
the ow cells (a ow cell is formed between any two adjacent plates). The discharge was assumed to be uniformly
distributed to the ow cells. The performance of the settling
tank was expressed as TRE.

Experimental data processing

SOR

(2)

861

N.M.C. Saady

Utilizing settling tests to design inclined plate settling tanks

very close to the performance predicted by the settling

column test (Figure 4). At an SOR of 3 m3 m
2 h
1, the
difference in TRE of the settling column and plate settler
was around 4% for the occulent settling. Results of the
settling column for occulent settling (Figure 4) could be
used directly without any modication to design, safely in
the plate settling tank, but it is also possible to increase
the SOR by 30% and decrease the Dt which will be discussed
later; however, it should be noted that the last possibility
depends on the quality of the settling characteristics of the
suspension to be handled. The inclined plates enable us to
increase the SOR because they offer more surface area for
settling, improve hydraulic conditions, and enhance secondary occulation (Yao ).
For a given SOR, the TRE for plate settling is higher than
that of the settling column test (Figure 4) because there is an
additional occulation occurring in the plate settling due to
the velocity gradient of the water across the ow cell. In comparison, the occulation in the settling column is only due to
the difference in the settling velocities of particles. Camp
() showed that the rate of occulation in settling tanks is
directly proportional to the number and concentration of the
particles, the velocity gradient, differences in settling velocities
of particles, and the detention time. The rate of occulation due
to velocity gradient is proportional to the square of the ow velocity in the tank. Moreover, the counter-current ow pattern of
the water and the settled solids between the adjacent plates
enhances further occulation and improves oc settling. Yao
() showed that the efciency of the downward ow unit
relative to that of the upward ow unit follows Equation (3),
indicating that the upward ow always gives better performance than the downward ow. Yao () attributed that
better performance to the further occulation:
SOR(
) L tan

SOR()
L
tan

Water Science & Technology

TRE
8:3235 LnZ 119:39

SSRE
1:38 SOR 98:27

(5)

2012

(6)

Settling in jars
The results of occulent settling in jars (Figure 4) were close
to that of the settling column tests, thus the SOR could be
used directly to design the settling tank with an inclined

The regression equations for the curves shown in this

study are given in the supporting information (Tables s1, 2,
3, 4, available online at http://www.iwaponline.com/wst/
066/260.pdf). The regression relationship between the TRE
and SOR for occulent settling in the inclined plate settling
tank in this study is given in Equation (4), which is similar to
the regression equations that have been reported in other
studies (Equations (5) and (6)).
(4)

66.4

where SSRE in Equation (5) is the wastewater SS removal

efciency (Saleh & Hamoda ). The dimensionless parameter (Z) in Equation (6) represents the effect of
Reynolds and Froude numbers (Sarkar et al. ) on the
SS removal from synthetic tap water and soil particles.
For plain settling (Figure 5), the SOR used to design the
plate settling tanks should be decreased by a safety factor as
in the case for conventional settling tanks, but the magnitude of the safety factor used in the design of plate settling
tanks is not necessarily the same as that used in the design
of the conventional settling tanks. The last case is important
for storm water settling.
The results of the settling column test (Figure 6) showed
that TRE for occulent settling is 80% at Dt of 27 min and
SOR of 2 m3 m
2 h
1. Applying the safety factors to design a
conventional settling tank, SOR of 1 m3 m
2 h
1 and Dt of
60 min were obtained, and a depth of 1.5 m was required.
For a settling tank with plates at 60o, TRE of 85% could be
achieved at an SOR of 2.2 m3 m
2 h
1 (Figure 7), Dt of
5 min, and a settling depth of 0.18 m. Therefore, it is possible
to decrease the Dt by as much as 75% given that the ocs have
been given sufcient time to develop. Increasing the SOR by
30% and decreasing the Dt by 75% means an increase in the
hydraulic capacity of the settling unit and the whole plant.

(3)

TRE
5:5471 LnSOR 89:434

Figure 6

TRE, SOR and Dt relationships for the settling column test.

862

N.M.C. Saady

Utilizing settling tests to design inclined plate settling tanks

Water Science & Technology

66.4

2012

the results of these tests could be used directly to

design inclined plate settling tanks for the case of occulent
settling.
Modeling of plate settling

Figure 7

TRE, SOR and Dt relationships for the upow settling tank with inclined plate
W

at 60 .

plate. However, for plain settling (Figure 5), the SOR

obtained from jar tests should be decreased much more
than the SOR, obtained from the settling column tests
because for the same SOR the TRE in jars is higher than
the TRE in the settling column tests and the inclined plate
settling tank. This situation may be due to the shorter depth
which gives less hindrance and thus higher turbidity removal.
In plain settling, and for a given TRE, the SOR obtained from
jar tests may be used to design inclined plate settling tanks
after dividing it by 3, while the SOR obtained from the settling
column should be divided by 2. For occulent settling, the
results of settling column tests and settling in jars are within
93% of the inclined plate settling tank results. Accordingly,

Figure 8

Two empirical physically homogenous models (Figure 8)

were developed using nonlinear regression of experimental
data. The models relate TRE to SOR, Dt, and the ow rate
velocity (Vo) in the ow cell between any two adjacent
plates. Inuences of the parameters and their interactions
on the TRE were signicant at a 1% level. The models
passed tests for heteroscedasticity in the error term, the
Klein test for multicollinearity amongst the independent
variables, the DurbanWatson ratio test checks for serial
correlation, the F-test for the assumption of normal and
independent distribution of residuals, the assumption of
constant variance, and the lack of t. The statistical test procedures were according to Wesolowsky (). The R 2
values were 0.968 and 0.956 with standard deviations of
0.01 for the predicted TRE. The models did not exceed the
limitations of the standard residual.
The models are valid between the upper and lower limits
of each independent parameter shown in Figure 8 and are
limited to the system tested in this study. A further validation
must be conducted before these models can be applied in
different situations. The characteristics of water SS (particle
size distribution and density), occulant (type and optimum
dose), geometrical aspects and conguration of inclined
plates, and settling tank designs must be considered.

Design nomograph for an upow settling tank with inclined plates (60 above the horizontal) in the upper zone based on (a) who SOR and ow rate velocity (Vo) and (b) SOR and
W

detention time (Dt).

863

Table 1

N.M.C. Saady

Utilizing settling tests to design inclined plate settling tanks

Water Science & Technology

66.4

2012

Reported efciencies of different settling units

Condition

Suspension

TRE or SSRE (%)

at SOR1.1 (m3 m
2 h
1)

Reference

Plates with alum

Synthetic water

83.28

Demir ()

Plates and alum

River water

88.78

This study

Conventional with alum

River water

73.19

This study

Plates

River water

26.99

This study

Conventional

River water

11.38

This study

Square tubes

Wastewater

95

Jimnez & Ramos ()

Plates

Wastewater

95

Jimnez & Ramos ()

Conventional

Wastewater

88

Jimnez & Ramos ()

Tubes and alum

Wastewater

97

Jimnez & Ramos ()

Plates

Wastewater

96.7

Saleh & Hamoda ()

Conventional

Wastewater

94.75

Saleh & Hamoda ()

The optimum inclination angle is a function of the SS

characteristics and the material of the plates. The optimum
inclination angle for activated sludge, rotating biological
disc ocs and trickling biolter were 45 , 45 and 33 ,
respectively (Jimnez & Ramos ), while for synthetic
water with bentonite solid particles it was 50 (Demir
). The models were developed based on the optimum
inclination angle of 60 for the river water used in this study.
The results from this study were slightly better than similar systems handling synthetic water with bentonite-SS
(Demir ) as presented in Table (1). However, the TRE
in this study is slightly less than the values reported for similar systems handling wastewater. This situation is likely to be
due to more occulation of the organic and inert suspended
particles in the wastewater in comparison with river water.
W

ACKNOWLEDGEMENT
This study was nancially supported by the University of
Technology, Baghdad-Iraq. The author acknowledges the
support from Dr Ahmad M. A. Al-Hiti during this study.

CONCLUSIONS
The TRE of occulent settling in jar tests coincided with the
TRE of the upow settling tank with inclined plates in the
upper zone while that of settling column test was very
close. For the raw river water tested in this study, the SOR
obtained from the occulent settling column tests can be
increased by 30% and the corresponding Dt can be decreased
by 75% to design the modied upow settling tanks with a
variation in TRE of less than 7%. For plain settling, the
SOR obtained from jar and settling column tests should be
divided by 3 and 2, respectively, before they can be used in
the design of the modied upow settling tank. The two
empirical models can be used to design the modied
upow settling tank with a prediction error of 1% in TRE.

REFERENCES
Camp, T. R. Sedimentation and the design of settling tanks.
Trans. ASCE 111 (Paper No. 2285), 895958.
Clark, S. E., Roenning, C. D., Elligson, J. C. & Mikula, J. B.
Inclined plate settlers to treat storm-water solids. J. Environ.
Eng. ASCE 135 (8), 621626.
Das, B. M. Principles of Geotechnical Engineering. Nelson,
Ontario, Canada.
Demir, A. Determination of settling efciency and optimum
plate angle for plated settling tanks. Water Res. 29 (2), 611616.
Fujisaki, K. Enhancement of settling tank capacity using a new
type of tube settler. Water Sci. Technol. 62 (6), 12131220.
Halalsheh, M. M., Muhsen, H. H., Shatanawi, K. M. & Field, J. A.
Improving solids retention in upow anaerobic sludge
blanket reactors at low temperatures using lamella settlers. J.
Environ. Sci. Health Part A 45, 0541059.
Hassan, M. S. & Hassan, A. H. The efciency of using tube
settlers technique in comparison with the conventional
treatment in Mosul city water treatment. Al-Radain Eng. J.
19 (1), 2634.
He, C., Wood, J., Marsalek, J. & Rochfort, Q. Using CFD
modeling to improve the inlet hydraulics and performance
of a storm-water clarier. J. Environ. Eng. ASCE 134 (9),
722730.
Jardin, N., Rath, L., Schonfeld, A. & Grunebaum, T. Costeffective upgrading of a biological wastewater treatment
plant by using lamella separators with bypass operation.
Water Sci. Technol. 57 (10), 16191625.

864

N.M.C. Saady

Utilizing settling tests to design inclined plate settling tanks

Jimnez, B. & Ramos, J. High-rate sedimentation for wastewater

treatment processes. Environ. Technol. 18 (11), 10991110.
Kolish, G. & Schirmer, G. Lamella separators in the
upgrading of a large urban sewage treatment plant. Water Sci.
Technol. 50 (7), 205212.
Metcalf & Eddy Wastewater Engineering: Collection,
Treatment, Disposal. McGraw-Hill, New York, USA.
Okoth, G., Centikaya, S., Bruggemann, J. & Thoming, J. On
hydrodynamic optimisation of multi-channel counter-ow
lamella settlers and separation efciency of cohesive
particles. Chem. Eng. Process. 47, 90100.
Saady, N. M. C. Experimental verication of inclined plate
settling models. J. Appl. Sci. Environ. Sanitation 6 (3), 309317.
Saleh, A. M. & Hamoda, M. F. Upgrading of secondary clariers
by inclined plate settlers. Water Sci. Technol. 40 (7), 141150.
Salem, A. I., Okoth, G. & Thoming, J. An approach to
improve the separation of solidliquid suspensions in
inclined plate settlers: CFD simulation and experimental
validation. Water Res. 45, 35413549.

Water Science & Technology

66.4

2012

Sarkar, S., Kamily, D. & Mal, B. C. Effect of geometric and

process variables on the performance of inclined plate settlers
in treating aquacultural waste. Water Res. 41, 9931000.
Shevidi, A., Azimi, A. K., Bidhendi, G. N., Fazeli, M. &
Asadollahfardi, G. Application of multi-stage inclined
tube settlers for water turbidity removal. Water Wastewater J.
77 (2), 1222.
WEF (Water Environment Federation) Clarier Design.
Manual of Practice No. 8, 2nd edition. McGraw-Hill, USA.
p. 152.
Wesolowsky, G. O. Multiple Regression and Analysis
of Variance: An introduction for Computer Users
in Management and Economics. John Wiley & Sons,
New York.
Yao, K. M. Design of high rate settlers. J. Env. Eng. Div. ASCE
99 (EE5), 621637.
Zhou, J., Walton, K., Laskovski, D., Duncan, P. & Galvin, K. P.
Enhanced separation of mineral sands using the Reux
Classier. Miner. Eng. 19 (15), 15731579.

First received 17 January 2012; accepted in revised form 5 April 2012

Copyright of Water Science & Technology is the property of IWA Publishing and its content may not be copied
or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission.
However, users may print, download, or email articles for individual use.