Pretreatment and Primary Treatment

Chapter 3
Pretreatment and Primary Treatment
1. Introduction 79
2. Screening 79
3. Sedimentation 80
4. Flotation 127
5. Neutralization (and Equalization) 133
Problems 162
References 165
1. INTRODUCTION
Selection of a wastewater treatment process or sequence of processes depends

on a number of factors, i.e., (1) characteristics of the wastewater, e.g.,
biochemical oxygen demand (BOD), percentage of suspended solids, p H ,
presence of toxic materials; (2) required effluent quahty; (3) cost and availability
of land, e.g., certain biological processes (stabilization ponds) are only
economically feasible if low-cost land is available; and (4) consideration of a
possible future upgrading of water quality standards, necessitating design of a
more sophisticated type of treatment for future use.
Pretreatment of wastewater implies removal of suspended solids or condition
ing of wastewater for discharge into either a receiving body of water or a
secondary treatment facihty through neutralization and/or equahzation. Types
of primary treatment discussed in this chapter are (1) screening, (2) sedimen
tation, (3) flotation, and (4) neutraHzation and equahzation.
2. SCREENING
Screening is employed for removal of suspended solids of various sizes. Screen

openings range in size depending on their purpose, and cleaning of screens is
done either manually or mechanically. Screenings are disposed of by burial,
incineration, or anaerobic digestion. Screens are classified as fine and coarse.
Fine screens have openings of in. or smaller. They are usually made of steel
mesh or perforated steel plates and are sometimes used instead of sedimentation
tanks. However, whereas they remove from 5-25% of suspended solids, 40-60%
79
80 3. Pretreatment and Primary Treatment
is removed by sedimentation. For this reason, and also because clogging is

frequently a problem, use of fine screens is not very common.
Coarse screens have openings ranging from 1.5-3.0 in. They are used as
protecting devices so that large suspended sohds do not damage pumps and
other equipment.
Sometimes shredders are used instead of coarse screens. These devices tear
down suspended solids, which are then removed by sedimentation.
3. SEDIMENTATION
3.1. Introduction
Sedimentation is utilized in wastewater treatment to separate suspended
solids from wastewaters. Removal by sedimentation is based on the difference in
specific gravity between solid particles and the bulk of the liquid, which results in
setthng of suspended solids.
In some cases, sedimentation is the only treatment to which the wastewater is
subjected. Sedimentation is also utilized in one or more steps of a treatment
sequence. In a typical activated sludge plant, sedimentation is utilized in three of
the treatment steps: (1) in grit chambers, in which inorganic matter (e.g., sand) is
removed from the wastewater; (2) in the primary clarifier, which precedes the
biological reactor, sohds (organic and others) are separated; and (3) in the
secondary clarifier, which follows the biological reactor, the biological sludge is
separated from the treated effluent.
3.2. Types of S e t t l i n g
Three types of settling are recognized depending on the nature of solids
present in the suspension.
1. Discrete settling. Particles being settled keep their individuality, i.e., they
do not coalesce with other particles. Thus the physical properties of the particles
(size, shape, specific gravity) are unchanged during the process. The settling of
sand particles in grit chambers is a typical example of discrete settling.
2. Flocculent settling. Agglomeration of the settling particles is accom
panied by changes in density and setthng velocity. The sedimentation occurring
in primary clarifiers is an example.
3. Zone settling. Particles form a lattice (or blanket) which settles as a mass
exhibiting a distinct interface with the liquid phase. Examples include sedimen
tation of activated sludge in secondary clarifiers and that of alum floes in water
treatment processes.
3.3. Theory of Discrete S e t t l i n g
The fundamental relationship for setthng of discrete particles is Newton's law,

which is based on the assumption that particles are spherical with a uniform
3. Sedimentation 81
diameter. When a particle settles, it accelerates until the forces promoting

settling, i.e., the particle effective weight, are balanced by the drag of frictional
resistance of the liquid. When this equality is achieved, the particle reaches a
constant settling velocity called the terminal or settling velocity of the particle.*
Consider the particle in Fig. 3.1, which has reached its terminal velocity, and
write the appropriate force balance. The force promoting sedimentation, i.e..
Fig. 3.1. Discrete settling of a particle.
the effective weight of the particle, is the difference between its actual weight and
the hydrostatic hft:
(3.1)
where is the particle effective weight; ps the particle density; P L the Hquid
density; g the acceleration of gravity; and ν the particle volume, ^nd^, where of is
the diameter of the spherical particle.
The drag force impeding sedimentation is
/^D = CO^(PLFV2) (3.2)
where is the drag force; Co the drag coefficient; Λ is the projected area of the
particle, A = \nd^\ and V is the relative velocity between particle and fluid.
For the condition defining the terminal velocity, i.e., = Fp, Eqs. (3.1) and
(3.2) yield
{ps- PÚgv = CÂ{p^Vll2)
where
V = V^ = setthng velocity
Substituting υ = ^nd^,A = ^nd^, and solving for the terminal velocity,

1/2
" 4 / g \ p s - PL
(3.3)
L3VCD PL
which is Newton's law.
* This results from force = mass χ acceleration. Thus zero acceleration corresponds to a net
force of zero, i.e., a perfect balance of forces.
For spherical particles, the drag coefficient C D is related to the Reynolds

number jVR defined as
(3.4)
where d is the diameter of sphere, is the terminal velocity (settling velocity),

and P L and are the density and viscosity of liquid. This relationship is shown
in Fig. 3.2.*
10^
Ν
i •
Sto (es's ! Trc nsitior Newtc n's
egion ! re^jion regior
1
1
CD=24.
1
1
1
10
CD=18
1.0 A
¡
0.4
1
0001 001 0.1
Fig. 3.2. Correlation for drag coefficient for spherical particles.
In general, the drag coefficient Co is approximated by

C D = b/Nl (3.5)
where coefficients b and η for the different regions of Fig. 3.2 are indicated in
Table 3.1. The approximate relationship between C D and given by Eq. (3.5) is
written in logarithmic form for the three regions indicated in Fig. 3.2.
TABLE 3.1
Drag Coefficient C D
Region b η Co = b/Nl
Stoke's law
24 1.0 Co = 24/
Transition
2<N^< 500 18.5 0.6 C D = 18.5/iVj-^
Newton's
A^R > 500 0.4 0.0 CD = 0.4
* For nonspherical particles, Fig. 3.2 is plotted as a family of curves, each curve corresponding to
specified value of a parameter defined as sphericity [sphericity, φ = (surface area of a sphere having
same volume as particle)/(surface area of particle)]. See Waddel [ 1 4 ] .
3. Sedimentation 83
Stokes's region:
Co = 24/7VR
. ' . l o g CD = -log7VR + log24
Transition region:
C D = 18.5/Α^^·^
/ . l o g C D = - O.ólogA^R + l o g l 8 . 5
Newton's region:
CD = 0.4
. •. log C D = log 0.4 = 0.0 log A^R + log 0.4

Thus the curve in Fig. 3.2 is approximately replaced by three segments of straight
hne with slopes of - 1, - 0.6, and 0.0, respectively.
Many sedimentation problems in wastewater treatment occur in Stokes's
region. Substituting C D = 24/íVR = 24μ^αν^ρι^ in Eq. (3.3) and simplifying,
Stokes's law is obtained:
For a specific problem in Stokes's region (ps, PL, and μι^ fixed) Eq. ( 3 . 6 ) is written
as
V, = Kd^ (3.7)
(where is a constant), which in logarithmic form becomes
log Ks = 2 log of + log a: = 2 log i/ + C (3.8)
where C is a constant. Thus, a logarithmic plot of versus d yields a straight line

of slope equal to 2.0 for Stokes's region.
For a specific problem in Newton's region, since C D is a constant ( C D = 0.4),
Eq. ( 3 . 3 ) yields
V, = K'd"^ (3.9)
(where K' is a constant), which in logarithmic form becomes
log = ^ log of + log ^ ' = ^ log + C (3.10)
where C is a constant. Thus, a logarithmic plot of versus d yields a straight

line of slope equal to γ for Newton's region.
For the transition region. C D = 18.5/7V^-^ = ( 1 8 . 5 ^ 2 - ^ ) / ( r f ^ - ^ K ^ ^ V L - ^ ) .
Substituting this value in Eq. ( 3 . 3 ) and simpHfying, one obtains
= [ ( 4 3 / 5 5 . 5 ) ( p ° - V X P S - ρύΙρΔ'''''ά'''' = K"d''*' (3.11)

where K' is a constant. In logarithmic form this becomes
logV, = 1.1431og¿/ + logA:" = 1.143 l o g í / + C " (3.12)
where C " is a constant. Thus a logarithmic plot of versus d yields a straight

line of slope 1.143 for the transition region. The logarithmic plot of versus d
for the three regions is shown in Fig. 3.3. Even in the case of grit chambers, the
theory just outlined suffers from two serious limitations: (1) grit particles are
seldom spherical and (2) grit particles d o not have uniform density.
A graph corresponding to Eq. (3.3) is plotted in Fig. 3.4, giving the
relationship between particle diameter and velocity V^. Particles of specific
gravities 1.001, 1.01, and 2.65 are considered in plotting Fig. 3.4. Value 2.65
corresponds to the specific gravity of typical sand. The hquid used is water at
temperatures indicated, corresponding to respective values of pi^, μι^. Values of
CD are obtained from Fig. 3.2 by a trial and error procedure: (1) for specified
particle diameter and temperature ( p l and p l fixed) assume a settling velocity
; (2) calculate based upon this assumed velocity; (3) from Fig. 3.2 read CD ;
log d Fig. 3.3. Logarithmic plot of versus d.
10'' 1.0
Velocity of fall (cm/sec)
Fig. 3.4. Relation between settling velocity and particle diameter [ 4 ] .

3. Sedimentation 85
and (4) from Eq. (3.3) recalculate V^. If it agrees with the value assumed in (1)
calculations are consistent. Otherwise iteration is continued until agreement is
obtained.
Figure 3.4 is constructed in this manner. Since construction is based upon the
actual curve of Co versus (i.e., Fig. 3.2), the lines in Fig. 3.4 exhibit some
curvature, by contrast with the three straight line segments in Fig. 3.3
[construction of which is based on the approximate relationships given by Eqs.
(3.8), (3.10), and (3.12)]. As an approximation, however, curves in Fig. 3.4 are
replaceable by three straight line segments.
3.4. The Ideal S e d i m e n t a t i o n Tank Concept
This concept, developed by Hazen [8] and C a m p [ 2 ] , is the basis for arriving
at relationships utilized in the design of sedimentation tanks. The model
Fig. 3.5. Model of a sedimentation tank with discrete setthng particles (Cases 1 and 2).
t=0
Fig. 3.6. Model of a sedimentation tank with discrete settling particles (Cases 3 and 4).
chosen for a sedimentation tank is a rectangular section basin which consists of

four zones (Figs. 3.5 and 3.6).
1. Inlet zone. Here the flow becomes quiescent. It is assumed that at the limit
of this zone (i.e., along vertical line xt\ particles are uniformly distributed across
the influent cross section.
2. Sedimentation zone. A particle is assumed to be removed from suspen
sion once it hits the bottom of this zone (horizontal line ty).
3. Outlet zone. Wastewater is collected here prior to transfer to the next
treatment.
4. Sludge zone. This zone is provided for sludge removal.
Settling paths of particles entering the sedimentation zone at points χ and x'
for discrete settling are indicated by lines xy and x'y' in Fig. 3.5 and lines xy" and
x'y in Fig. 3.6. These settling paths are the net result of two velocity vector
components:
1. Flow-through velocity V:
y^Q/A' = Q/WH (3.13)
where Vis the flow-through velocity (ft/sec); Q is the flow rate (ft^/sec) ;A'is the
vertical cross-sectional area of sedimentation zone, namely A' = WH({i^) (refer
to Fig. 3.7); IV is the width of sedimentation zone (ft); and Η is the depth of
sedimentation zone (ft).
2. Setthng velocity, indicated by either vectors or in Figs. 3.5 and 3.6,
respectively.
For discrete settling the settling velocity is constant for any specific settling
path, i.e., and Kj do not vary along their respective paths. This is due to the
fact that a discrete particle is unhindered by neighboring ones (no coalescence),
so it settles with a uniform velocity, read from Fig. 3.4, as a function of the
particle diameter.
Fig. 3.7. Geometry of the sedimentation zone.

3. Sedimentation 87
For flocculent settling the situation is different. Figure 3.8 illustrates a typical
sedimentation path of flocculent settling (Section 3.6). As coalescence with
neighboring particles takes place, the effective diameter of the particle increases,
and thus its settling velocity also increases. The net result is that setthng paths
are curved in contrast with straight hne paths for discrete setthng.
Inlet
zone
v;>Vs
Sedimentation zone
Sludge zone
Fig. 3.8. Model of a sedimentation tank flocculent settling.
Consider the following cases with reference to Figs. 3.5 and 3.6, keeping in
mind that a particle is assumed to be removed from the suspension once it hits
the bottom of the sedimentation zone.
Case I (Fig. 3.5). A particle which at time zero (t = 0) is located at point χ
and possesses a settling velocity (and diameter ¿4 read from Fig. 3.4). This
particle is removed since it touches the bottom of the sedimentation zone at y
(path xyl
Case 2 (Fig. 3.5). A particle which at / = 0 is located at x' on the same
vertical line as χ but below x, and has a settling velocity (or greater than V^).
This particle is also removed since it hits the bottom of the sedimentation zone to
the left of point y (at
If the settling velocity is greater than K^, the particle touches the bottom of the
sedimentation zone to the left of y'. Figure 3.6 portrays the case of particles with
a diameter αχ (di < ¿4), possessing a settling velocity Vi (Vi < Vs)- (Settling
velocity Kj is read from Fig. 3.4 for d= di.) These particles are shown separately
in Fig. 3.6 simply to avoid overcrowding of the diagram. In fact they are together
in the slurry with particles of settling velocity V^.
Case 3 (Fig. 3.6). A particle which at ί = 0 is located at χ and has a setthng
velocity Fj (where Vi < V^), This particle is not removed since it does not reach
the bottom of the sedimentation zone (i.e., line ty) in time for removal
(sedimentation path xy").
Case 4 (Fig. 3.6). A particle with settling velocity (where Κχ < V^) and
situated at at ί = 0. This particle is removed (sedimentation path x'y).
Consideration of similar triangles in Figs. 3.5 and 3.6 leads to the following
relationships. F r o m Fig. 3.5
V/V, = L/H (3.14)
or
V/L=VJH (3.15)
F r o m Fig. 3.6
K/Ki = L/h (3.16)
or
V/L=VJh (3.17)
Comparing Eqs. (3.15) and (3.17)
Vs/H= VJh (3.18)
or
VJV, = h/H (3.19)
N o w derive an expression for the settling velocity V^, from the following
considerations. From Eq. (3.13)
Q=VA'= VWH (3.20)
From Eq. (3.15) it follows that
V= V,(L/H) (3.21)
Substitution of V from Eq. (3.21) in Eq. (3.20) and simplification leads to
Q=V^LW= V,A (3.22)
where A = LW = horizontal cross-sectional area of the sedimentation zone (ft^)

(Fig. 3.7).
F r o m Eq. (3.22)
Vs = Q/LW=Q/A (3.23)
F r o m Eq. (3.23) it follows that the setthng efficiency is a function of the

horizontal cross-sectional area, rather than of the depth H. Thus, in principle, it
is advisable to utilize sedimentation tanks of high surface area A and low depths.
The only reasons for using a reasonable depth are (1) to satisfy depth
requirements in order to provide for mechanical rakes utilized for removal of
settled sludge; and (2) the horizontal component of velocity (flow-through
velocity V) must be kept within certain limits to prevent scouring the particles
which have settled.
F r o m Eq. (3.13) it follows that Η should not be too low since V would rise
above the scour velocity. The subject of scour velocity is discussed at the end of
this section. Scour occurs when flow-through velocity F i s sufficient to suspend
previously settled particles. Scour is not usually a problem in large setthng tanks,
but it can be an important factor in grit chambers and narrow channels.
3. Sedimentation 89
The two fundamental postulates of the ideal tank concept are (1) uniform
distribution of particles across the influent cross section (i.e., along vertical line
xt) and (2) assumption that a particle is considered removed when it reaches the
bottom of the sedimentation zone (i.e., horizontal line ty).
Keeping in mind these two postulates, two corollaries follow: (1) All particles
with a settling velocity equal to or greater than are removed; and (2) all
particles with a settling velocity less than (such as Ki in Fig. 3.6) are removed
in a proportion given by the ratio Κ^/Κ^.
The first corollary follows from consideration of Fig. 3.5. Particles with a
settling velocity K^, and which at / = 0, are situated at the surface of the liquid
(i.e., point x\ will touch the bottom of the sedimentation zone at point y, and
therefore are removed according to postulate (2). Obviously, particles with a
settling velocity K^, and which at ί = 0, are situated below the surface of the
hquid (e.g., point x'\ will touch the bottom of the sedimentation zone to the left
of point y (e.g., point ;;') and with greater reason are removed, according to
postulate (2). Unquestionably all particles with a settling velocity greater than
Ks, regardless of their position at / = 0, are also removed.
The second corollary foUows from examination of Fig. 3.6. For simplicity
assume that / / = 100 in. and h = 75 in. Consider the particles with settling
velocity V^. According to postulate (1), 25% of these particles are, at í = 0,
situated between the surface (point x) and the depth ( 7 / - A) = 100 - 75 = 25
in., i.e., point x'. These 2 5 % of particles of settling velocity will not be
removed since they will not touch the sedimentation zone bottom. However, the
other 75% of these particles of settling velocity Ki, which at time / = 0 are
situated below x' will be removed, since they will touch the sedimentation zone
bottom at point j or at its left. In general, the percent removal of particles with a
settling velocity Vi less than V^, will be given by the ratio h/H (in the case just
considered h/H = 75/100 = 0.75 or 75%). F r o m Eq. (3.19) it fohows that this
ratio is the same as Vi/V^, i.e.,
Vi/Vs = h/H
The overflow rate defined as
Q/A = ftVft^ h = ft/h
is defined as the settling velocity of a particle that settles through a distance

exactly equal to the effective depth of the tank during the theoretical detention
period. This results from the definition of detention period:
t = detention period = (volume of t a n k ) / ß = HA/Q (A = LW) (3.24)
F r o m Eq. (3.24) it follows that a settling velocity V, defined as = H/t is

equivalent to the overflow rate, since
η = H/t = Η/{HA/Q) = Q/A
which is Eq. (3.23).

The scour velocity V^^ is the value of the flow-through velocity V:

V= Vsc = Q/A' = Q/WH (3.25)
for which "previously settled" particles are scoured away.*
The scour velocity is estimated by the following empirical equation [ 2 ] :
Vso = imd(s-l)/n"' (3.26)
where V^c is the velocity of scour (mm/sec); i.e., flow-through velocity required
to scour all particles of diameter d or smaller; β is a constant (0.04 for
unigranular sand, 0.06 for nonuniform sticky material); / i s the Weisbach-
D'Arcy friction factor (0.03 for concrete); g is the acceleration of gravity
(mm/sec^) (normal: 9800 mm/sec^); dis the particle diameter (mm) (particles
with diameter d or less than d are scoured away); and s is the specific gravity of
particle.
The fundamental relationship for particle removal, i.e., Eq. (3.19), has been
established from consideration of sedimentation tanks with a rectangular cross
section. It can be readily shown that a similar relationship apphes to
sedimentation tanks of circular cross section. This case is shown in Fig. 3.9.
Fig. 3.9. Model of sedimentation tank for tanks of circular cross section.
In the case of a tank of circular cross section, the horizontal component of the
velocity V varies with the radius since
Q
(3.27)
A' 2nrH
* The words "previously settled" are placed in quotes because a particle which is scoured away
never actually settles. Mentally one separates the processes of settling and scouring and imagines
that a particle settles and subsequently is scoured away. This reasoning is compatible with the
hypothetical resolution of the velocity trajectory into vectors V (flow-through velocity) and
(settling velocity).
3. Sedimentation 91
where R,. Since F decreases with increasing radius whereas the vertical
component ( o r V^) is constant, the paths of the particles are curved instead of
straight lines. The slope of the setthng curve at any radius is
dhldr^ VJV (3.28)
multiplying the numerator and denominator of the right-hand member of Eq.

(3.28) by 2nrH\
dh/dr = InrHVJlnrHV (3.29)
From Eq. (3.27) it follows that InrHV = Q; therefore, Eq. (3.29) becomes
dh/dr = InrHVJQ (3.30)

or
dh/H=(2nrVJQ)dr (3.31)
For a given Kj, the variables in Eq. (3.31) are h and r. Integrating Eq. (3.31)
between the inside radius Ri and the outside radius R^ one obtains
h _2nV,
= ^(nRl-nRf) = Â (3.32)
H'~Q~ R. Q
or
(3.33)
Η QIA
Since from Eq. (3.24), QIA = V,, Eq. (3.33) is equivalent to Eq. (3.19).
Example 3.1. Consider a suspension of sand (s = 2.65) in water at 20°C with a

uniform particle size (d = 0.07 mm). Flow is 1,0 Mgal/day.
1. Calculate the grit chamber surface (horizontal cross section) for obtaining
removal of 70% of the particles.
2. Suppose that instead of a uniform particle diameter, there is, besides
particles oí d= 0.07 mm, another set with a uniformly larger diameter, which
are completely removed in the grit chamber designed for 70% removal of the
particles with d = 0.07 mm. Determine what is the minimum particle diameter
for total removal.
3. For case (2), determine the flow-through velocity V^^ so that all particles of
lower settling velocity than those completely removed are scoured away. What
combination of length, width, and depth for the grit chamber meets these
requirements?
Solution to Part 1
Step 1. From Fig. 3.4 [for d = 0.07 mm and s = 2.65 (at 20°C)] read
V, = 0.45 cm/sec
or
V, = OAS cm/sec χ ft/30.48 cm χ 3,600 sec/h

ν, = 53Λ ft/h = 53.1 ft^ft^ h
The overflow rate in gal/day ft^ is
53.1 ft^ft^ h X 7.48 gal/ft^ χ 24 h/day
. • . F s = 9,533 gal/day ft^
Step 2. The horizontal cross-sectional area is (for 100% removal)
1,000,000 gal/day
9,533 gal/day ft^
With this area, 100% removal is obtained. For 70% removal, the residence time
(and thus the cross-sectional area) is reduced by 30%. The cross-sectional area is
^ = 105 X 0.7 = 73.5 ft^

Under these circumstances, the settling velocity for 100% removal is
1,000,000/73.5 = 13,605 gal/day ft^

Note: Check on percent removal. Since the settling velocity is 9,533 gal/day
ft^, fixed by the particle diameter as determined from Fig. 3.4, percentage
removal is 9,533/13,605 = 0.70 (70%). Referring to Fig. 3.6 the particles
removed are those which at the end of inlet zone, along vertical line xt, are
already at distance h (or less than A) from the bottom of the sedimentation zone,
where [from Eq. (3.19)]
V, = 13,605 gal/day ft^ (100% removal)

Ki = 9,533 gal/day ft^ (70% removal)
. · . A / 7 / = 0.70
Solution to Part 2 If there is a distribution of particle diameters instead of
uniform diameter d = 0.07 m m , one reads from Fig. 3.4 the diameter, which is
larger than 0.07 m m , for which the settling velocity corresponds to 13,605
gal/day ft^. Therefore the abscissa in Fig. 3.4 is
13,605/9,533 χ 0.45 cm/sec = 0.45/0.7 = 0.642 cm/sec

F r o m Fig. 3.4 [for V, = 0.642 cm/sec and s = 2.65 (/ = 20°C)], read d = 0.085
m m (100% removal). If distribution of particle diameter in the influent is known,
one can calculate the percent removal corresponding to each group of particles
for a given diameter (Example 3.2).
Solution to Part 3 The scour velocity to sweep all particles of lower settling
velocity than those to be completely removed is calculated from Eq. (3.26):
= [8 X 0.04 X 9,800 χ 0.07(2.65 - l)/0.03]^/^ = 110 mm/sec
3. Sedimentation 93
Assuming that the sand contains only two particle sizes, e.g., 0.07 and 0.085 mm,
the scour velocity V^^ = 110.0 mm/sec sweeps away all particles oíd = 0.07 mm,
leaving behind those oí d = 0.085 mm. The value of K^c in practical units is
= 110 mm/sec χ ft/304.8 m m = 0.36 ft/sec
The vertical cross-sectional area is calculated from Eq. (3.13):

1,000,000 gal/day χ day/86,400 sec χ ftV7.48 gal
A' = = 4.3 ft^
0.36 ft/sec
Any practical combination of length, width, and depth is used to satisfy the
requirements A = 73.5 ft^ = LW and A' = 4.3 ft^ = WH.
Example 3.2. Suppose that for Example 3.1 instead of a uniform particle size,
there is a distribution of diameters. Assume the same specific gravity and
temperature as in Example 3.1, i.e., 5 = 2.65 and / = 20°C. Assume that for each
100 lb of grit the following distribution of particle sizes apphes:
(1) (2) (3)

Group lb of each Particle size,
no. particle size d (mm)
50 0.085
20 0.070
20 0.060
JO 0.050
100
The settling velocities for each group of particles are read from Fig. 3.4. This is
indicated in column (4) of Table 3.2. Percent removals are then calculated
[column (5)] and expressed as fractions of unity in column (6). The weight of
TABLE 3.2
Calculations for Example 3.2
(1) (2) (3) (4) (5) (6) (7)

Group lb of Particle (cm/sec) Percent removal Fraction lb removed
no. each size, (Fig. 3.4) removed = (7) = (2) X (6)
particle d (mm) = X 100 (5) - 100
0.642
size
1 50 0.085 0.642 (0.642/0.642)100= 100" 1.00 50.0

2 20 0.070 0.450 (0.45/0.642)100 = 70^* 0.70 14.0
3 20 0.060 0.350 (0.35/0.642)100 = 54.5 0.545 10.9
4 10 0.050 0.220 (0.22/0.642)100 = 34.3 0.343 3.43
Too 78.3
" Already calculated in Example 3.1.

sand removed by settling for each group of particles is computed in column (7).
Therefore 78.3% of the weight of the original particles is removed by setthng.
In this example, if the vertical cross section A' is taken equal to 4.3 ft^ (value
calculated in Example 3.1), all particles of = 0.07 m m and smaller are scoured
away (groups 2 , 3 , and 4). Therefore, the net removal is of only 50 lb. per 100 lb.
of total grit, i.e., the particles with d = 0.085 m m (group 1). This indicates a net
removal by weight of 50%.
If A' is taken larger than 4.3 ft^ the net removal is greater, since there is less
scouring. Examining calculations for Example 3.1, it follows that if A' is taken as
8.6 ft^ (twice 4.3 ft^) the value of V,^ is 0.18 ft/sec (instead of 0.36 ft/sec). This
corresponds t o :
= 55 mm/sec (instead of 110.0 mm/sec)
Since from Eq. (3.26), V^^ is proportional to d^'^, it follows that d is

0.07/4 = 0.0175 mm, so as to yield V,^ = 55 mm/sec (half of 110.0 mm/sec).
Therefore, only particles with i / = 0.0175 m m or smaller are removed by
scouring. Since for the given distribution the smallest particle diameter is 0.05
mm, there is no removal by scouring. Consequently, the net removal is 78.3 lb.
from every 100 lb. of sand, or 78.3% by weight.
If this value is adopted,
A = 73.5 ft^ = LW and A' = 8.6 ft^ = WH
Selecting / / = 4 ft, then
8.6/4 = 2.15 ft and L = 73.5/2.15 = 34.2 ft
This indicates specification of a narrow settling channel 34.2 ft long, 2.15 ft wide,
and 4 ft deep. The hydraulic residence time (or detention time), defined as
Parshall
flume ^ . ^
I ^Grit channel
Influent — ^^î=^^^^^Weir
It-^
Settled grit- Effluent
^ D r a i n to empty grit chamber

(a)
Parshall
J*"'"* f VorltcHannel I
Influent^^ ^ / Jcrit channel 2 Effluent
Gates^ lire ^
Gritchonnel 3 ^ ^
(b)
Fig. 3.10. Diagrams for grit chambers: (a) vertical cross section, (b) horizontal cross section.
3. Sedimentation 95
h = v/Q is
Volume of grit chamber: V = (73.5 χ 4) ft^ χ 7.48 gal/ft^ = 2,199 gal
Flow rate: Q = 1,000,000 gal/day
Then /h = y/Q = 2,199/1,000,000 = 0.0022 day ( = 3.17 min)
Typical detention times for grit chambers are between 20 sec and 1 min. This
example was not typical, because diameters were smaller than usual. Ordinarily,
diameters of sand particles range from 0.1 to 1.0 mm. Typical diagrams for grit
chambers are shown in Fig. 3.10.
3.5. Degritters
The conventional way for separation of grit is by use of sedimentation

equipment as described in Section 3.4. Another possibility is by utilization of
hydrocyclones. Figure 3.11 shows the diagram of a typical model of hy-
drocyclone, manufactured by Dorr-Ohver Inc., bearing the trade name of
D o r r Clone.
DorrClone cyclones are hquid/solid separators. Using energy supplied by the
feed p u m p pressure, centrifugal force separates materials of different specific
gravities. T h e cyclone consists of a stationary cylindrical/conical body with a
tangential feed inlet in the upper cyhndrical section. A n overflow, or hquid
discharge is located at the t o p in the machine axis and an underflow, or solids
discharge is at the opposite or lower end.
As wastewater feed enters the cylindrical feed chamber tangentially, a vortex
or spiral flow pattern develops. Centrifugal action forces the grit to the walls of
RAKE CLASSIFIER MODEL A Β 0 0 L WIDTH DORRCLONE h d

FP 14" 2'6" 0-10'/^" 9'5i/2" Γ9" 12" Size Γ4'Vie 2'2"/.6"
SSFR1'-6"&2O" 5Ό- ΎΑ' 5'2iVi6" 15'6V,6" 2'9"/ 3'3" 18" Size re" 3'5'/fe"
DSFR3O'&4O" 4·10·' ΎΖ" 7'6i/b- 15'6^16" 4'6V5'r
Fig. 3.11. Diagram of a typical model of hydrocyclone (DorrClone). (Courtesy of Dorr-Oliver

Incorporated.)
the cone where the sohds spin towards the apex and discharge through the apex
opening. The hghter, grit-free wastewater (containing the hghter organic sohds)
flows to the center of the vortex and exits through the overflow outlet. The most
commonly used DorrClones in wastewater treatment are either 12 in. or 18 in. in
diameter at the cylindrical section. When flow exceeds the capacity of a single
unit, multiple units in parallel are used.
The sewage grit discharging at the lower end of the DorrClone enters the rake
classifier, which is designed for washing and dewatering the grit material. It
removes the biodegradable materials remaining after cyclone processing and
dewaters the grit so that it can be readily transported for disposal. The rake
classifier consists of a tank, rake, and drive. The tank is rectangular with a long
inclined bottom. The upper end is for grit discharge. The lower end, where the
shmes overflow, is partially enclosed with an adjustable weir over which the
hquid overflows. Feed enters from above into the settling chamber. The raking
mechanism is comprised of one or more mechanically operated rakes swinging
from hangers at the overflow and grit discharge ends of the machine, and
activated by the head motion. Head motion is actuated through a system of
heavy gears, pinions, and cranks and driven by an electric motor, V-belt, drive,
and heavy-duty gearing.
The feed slurry is rapidly separated into two fractions: the fine, slowly settling
material (mostly organics), smaller than the mesh of separation; and the coarse
quickly settling grit, larger than the mesh of separation. The fine fraction is
buoyant due to specific gravity differential and the agitation produced by the
reciprocating action of the rakes. Being unable to settle, the fines pass out with
the liquid as overflow. The coarse fraction, on the other hand, sinks rapidly to
the bottom and is removed by the rake blades. Settled grit advances u p the
inclined tank bottom. The up-and-over motion of the rakes transports and then
releases undersized fines from the coarse material bed and keeps them in
suspension until they overflow. Coarse particles are drained above the liquid
level before being discharged by the last reciprocating blade.
3.6. Flocculent S e t t l i n g
Flocculent settling takes place when settling velocity of the particles increases
due to coalescence with other particles. A diagram of flocculent sedimentation
profiles is shown in Fig. 3.8. The setthng paths of the particles are curves, rather
than straight lines as for discrete settling.
Design criteria for systems exhibiting flocculent setthng are estabhshed by a
laboratory settling analysis. A typical laboratory setthng column is shown in
Fig. 3.12.
Concentration of suspended solids is kept uniform throughout the column at
the beginning of the test by means of a portable stirrer. The depth of the column
is approximately the same as that of the setthng tank to be designed.
Temperature is kept constant during the test. A practical design of a settling
column (Fig. 3.12) is 8 ft deep, with samphng ports at depths of 2 , 4 , 6 , and 8 ft.
3. Sedimentation 97
-O.D.-^
6" I
-I.D.-H
51/2" i
2':0"
Depths
Tap I
counted
in this
2'-0" direction
Tap 2
Testing
Tap 3 cylinder
(plexiglass)
2'-0"
Tap 4
Fig. 3.12. Laboratory settling column. L 3
D a t a taken at 2-, 4-, and 6-ft depths are utilized to determine settling velocity and
detention time relationships. D a t a from the 8-ft port a r e ' u s e d for sludge
concentration and compaction determinations.
Step 1. Fill the column with wastewater, maintaining a uniform con
centration of suspended soUds throughout. A portable stirrer is used for this
purpose.
Step 2. Remove the stirrer from the cylinder. At each sampling port, the
concentrations of suspended solids are measured at predetermined times.
Examples 3.3 and 3.4 illustrate the procedure from data treatment to the
design of a clarifier.
Example 3.3. The suspension being tested has an initial suspended solids
concentration of 430 mg/hter (SSq). The suspended solids (SS) concentrations in
Table 3.3 are measured at the indicated times at the 2-, 4-, and 6-ft sampling
ports. Perform treatment of the data arriving at curves for (a) percent SS
removal versus detention time (min), (b) percent SS removal versus overflow
rate [gal/day ft^], and (c) percent SS remaining (fraction of particles with less
than stated velocity) versus settling velocity (ft/h) for a residence time of 25 min.
Step 1. Calculate fraction of solids remaining in suspension for each sample:
X = SS/SSo
or (in %)
y = SS/SSo X 100 (3.34)
Then calculate for each sample the fraction of solids removed
1 - X
or (in %)
z = m - y (3.35)
A sample of these calculations (for a 2-ft depth) is shown in Table 3.4. Similar
calculations are performed for 4- and 6-ft depths.
Step 2. In order to smooth the experimental data construct a graph of percent
SS removed versus time. This plot is shown in Fig. 3.13 for the 2-, 4-, and 6-ft
depths.
Step 3. F r o m Fig. 3.13 construct a settling profile graph (Fig. 3.14). This is
done by reading from the smoothed curves of Fig. 3.13 the abscissas [/ (min)]
corresponding to selected values of percent SS removed (e.g., 5, 10, 2 0 , . . . ,
TABLE 3.3
Laboratory Sedimentation Data"
SS concentrations
at indicated depths
Time 2 ft 4 ft 6 ft
(min) (Tap 1) (Tap 2) (Tap 3)
5 356.9 387.0 395.6

10 309.6 346.2 365.5
20 251.6 298.9 316.1
30 197.8 253.7 288.1
40 163.4 230.1 251.6
50 144.1 195.7 232.2
60 116.1 178.5 204.3
75 107.5 143.2 180.6
" See Example 3.3.
TABLE 3.4
Calculation of Fraction of Solids Remaining and Removed for a 2-ft Depth
(1) (2) (3) (4)

Time SS remaining Solids remaining (%) Solids removed (%)
(min) (mg/liter) y = (SS/SSo) X 100 z = 100-:^
(Table 3.3)
5 356.9 83.0 17.0

10 309.6 72.0 28.0
20 251.6 58.5 41.5
30 197.8 46.0 54.0
40 163.4 38.0 62.0
50 144.1 33.5 66.5
60 116.1 27.0 73.0
75 107.5 25.0 75.0
3. Sedimentation 99
70, 75%) for each one of the three depths. These values are tabulated (Table 3.5)
and utilized for construction of Fig. 3.14.
90
80
o
70
J
ου
>
/
äS y
ν
150 / jé
o /
Ε
( /
*
-40 / i
ω / / y
/
Γ
/ /
30 / /
/ ¿ /
20
y./
γ
r
10
0 ι0 20 30 40 50 60 7( θ 9C
t (min)
Fig. 3.13. Suspended solids removed versus time.
Fig. 3.14. Settling profile. Encircled numbers are percentage of SS removed.

TABLE 3.5
Values for Plotting Fig. 3.14
t (min)
SS
removed % 2 ft 4 ft 6 ft
5 1.2 2.5 3.7

10 2.5 5.0 6.5
20 6.7 11.0 14.5
30 11.7 19.0 25.0
40 18.0 30.0 39.0
50 27.0 44.0 56.5
60 38.5 61.5 77.5
70 55.0 87.5 -
75 75.0
- -
Step 4. Calculate percent removal of SS and overflow rate (gal/day ft^).

Before the procedure described here is fully understood, some preliminary
considerations must be made. Although for flocculent settling, as indicated in
Fig. 3.8, the diameter of the particles and their settling velocities increase along
the sedimentation path owing to coalescence with neighboring particles; in the
discussion of Step 4 reference is made to effective settling velocities and effective
diameters. By utilizing these concepts, a calculation procedure for flocculent
settling is developed, which is based fundamentally upon the same postulates
studied for discrete settling in ideal sedimentation tanks, in which case diameters
and settling velocities were actually constant along the sedimentation path. An
effective settling velocity is defined as the effective depth (6 ft in this example)
divided by the time (detention time 0 required for a given particle to travel this
distance, i.e.,
= Hit (3.36)
For the discussion in the remainder of this section on flocculent settling,

whenever the terms settling velocity or diameter are utilized, it is understood
they refer to effective values of these parameters.
If a suspension contains particles with different settling velocities, the

efficiency of removal by sedimentation is obtained by performing a settling
column test as just described. Let SS be the concentration of solids remaining for
one specific sample and time. Thus
Xo = SS/SSo = fraction of solids remaining
and
\ - X q = \ - SS/SSo = yo = fraction of solids removed

3. Sedimentation 101
Particles with a settling velocity or higher (where = H/t) are completely

removed. Particles with a lower settling velocity Vi(Vi < V^) are removed at a
ratio given by Eq. (3.19).
A typical graph like the one in Fig. 3.15 is plotted by analysis of data obtained
with the sedimentation column. The details for construction of such a graph
from experimental data are discussed later in Step 4d. However, for understand
ing the calculation procedure described next, it is convenient to assume that this
graph is already available.
Fig. 3.15. Determination of overall removal. = f{x) for a fixed residence time /.
Figure 3.15 is a plot of the fraction of particles with less than the stated
velocity versus the setthng velocity in question. This graph is constructed for a
fixed residence time. Notice that the fraction of particles with less than the stated
velocity (if = H/t) corresponds to the fraction of particles not completely
removed. For example, if 30% of the particles in a specific case are completely
removed, then Xq = OJ is the ordinate corresponding to the settling velocity
Vs = H/t.
F o r particles with settling velocities between and -\- dV (where
y I < K), the fraction removed is Vi/Vs- Similarly, for particles with settling
velocities between + dV and Vi+ldV, the fraction removed is
(Vx^dvyv,.
The overall removal of suspended solids is
Overall removal = (1 - Xo) + (Vi/Vs)dx (3.37)
In Eq. (3.37), is a variable (0 ^ ^ V,) with Kj = / ( x ) portrayed by the

curve in Fig. 3.15. Term (1 - Xq) is the fraction completely removed, cor
responding to particles with velocities ^ V^. The second term in Eq. (3.37), i.e..
1
(Vi/Vs)dx = - Vidx (3.38)
0 f^s
which is the fraction of removal corresponding to particles with velocities less

than Vs (calculated by graphical integration as indicated by the hatched area in
Fig. 3.15). The differential area of width dx, indicated in Fig. 3.15 corresponds to
particles with settling velocities between and Vi -\- dV.
Combining Eqs. (3.19) and (3,37) the final expression for the overah removal
is obtained [Eq. (3.39)].
xo
Overall removal = (1 — Xq) + (V,/V,)dx^{l-Xo) + (h/H)dx
0
(3.39)
After these considerations return to discussion of Step 4.
Step 4a. F r o m Fig. 3.14 for a depth of 6 ft, read the values / (min)
corresponding to 5, 10, 20, 30, 40, 50, and 60% removal, and calculate the
corresponding settling velocities (ft/h). These values are tabulated in Table
3.6.
TABLE 3.6
Settling Velocities
Constant
7„ / (min) Settling velocity (ft/h)
removal // = 6ft V, = H/t = 6/(//60) = m/t
5 3.7 97.2
10 6.5 55.2
20 14.5 24.8
30 25.0 14.4
40 39.0 9.2
50 56.5 6.35
60 77.5 4.64
Step 4b. Calculate percent removal of SS. Calculations for percent removal of
SS and overflow rate for a 25-min setthng time (fourth entries in Table 3.6) are
illustrated next. Similar calculations are also performed for the other settling
times hsted in the second column of Table 3.6. F o r t = 25 min for the settling
depth / / = 6 ft, 30% of the suspended solids are completely removed. Consider
next the particles in each additional 10% range. Start with those in the range
30-40% removal in Fig. 3.14. Particles in this range are removed in the
proportion Fi/F^ or in the proportion of average settled depth (Ai) to the total
settling depth ( / / ) . T h e average settled depth (Αχ) is estimated by drawing (by
interpolation) a curve corresponding to 35% constant removal in Fig. 3.14, and
reading from it the depth hi corresponding to ί = 25 min. Therefore for this first
interval, the percentage of sohds removed is {hJH) χ 10 = (4.2/6) χ 10 = 7.0%.
In a similar manner for succeeding 10% intervals, the curves for constant
percent removal of 45, 55, 65, and 75% are drawn and the average settled depths
of 2.4, 1.4, 0.84, and 0.28 ft are read for t = 25 min. The calculations for / = 25
min are indicated as follows:
Setthng velocity: V, = H/t = 6.0/(25/60) = 14.4 ft/h
Percentage of solids removed (for t = 25 min)
100% removal (g) 30% 30.00

1st interval (35%) (4.2/6.0) X 10 = 7.00
2nd interval (45%) (2.4/6.0) X 10 = 4.00
3rd interval (55%) (1.4/6.0) X 10= 2.33
4th interval (65%) (0.84/6.0) X 10= 1.40
5th interval (75%) (0.28/6.0) X 10 = 0.46
Total removed after 25 min 45.19
Beyond the fifth interval the percent removals are neghgible, so calculations
are stopped at that point. In general, if 10% intervals are selected, the total
percent removal is given by
Total percent removal = A^tot + (hi/H) χ 10 + (Az/^) x 10

+ (h^/H) X 10 + · · · (3.40)
Equation (3.40) is simply an approximation of Eq. (3.39). (1 — Xq) corresponds
to A'tot and the integral \l''{h/H)dx is replaced by a finite summation of terms:
Σ{Κ^Η)χΔχ
where A^vg is the average settled depth for each selected interval. The Δχ in this
example are selected arbitrarily as a 10% range. The smaller the Ax selected, the
closer the approximation between the finite summation and the integral.
Similar calculations are performed for the other residence times hsted in the
second column of Table 3.6. The final results are summarized in Table 3.7. F r o m
Table 3.7 a graph of SS removed (%) versus detention time is prepared (Fig.
3.16).
TABLE 3.7
SS Removed (%) versus

Detention time
(1) (2)
t (min) SS removal (%)
3.7 13.4
6.5 20.1
14.5 33.9
25.0 45.2
39.0 55.0
56.5 64.3
77.5 71.1
80,
70
60
g 50 /
I 40 /
^ 30 /
20 /
Γ
10 f
J
/
/ r «.5 mir
10 20 30 40 50 60 70 80
Detention time (min)
Fig. 3.16. Suspended solids removal (column 2, Table 3.7) versus detention time (column
Table 3.7).
TABLE 3.8
SS Removed (%) versus Overflow Rate
(1) (2)'' (3) (4)''

/ (min) Settling SS removal (%)
Overflow rate
velocity, L(day)(ft2)J
(ft/h) ft 24 h 7.48 gal
Lh day ft^
= X 24 X 7.48 = 179.5F,
3.7 97.^ 17,450 13.4

6.5 55.2 9,908 20.1
14.5 24.8 4,452 33.9
25.0 14.4 2,585 45.2
39.0 9.2 1,651 55.0
56.5 6.35 1,140 64.3
77.5 4.64 833 71.1
" See Table 3.6.

" See column (2) of Table 3.7.
Step 4c. Prepare a plot of SS removed (%) versus overflow rate. Calculations
needed to prepare this plot are presented in Table 3.8.
The plot of SS removed (%) versus overflow rate is presented in Fig. 3,17. All
calculations are performed for an initial suspended solids concentration SSo
2000 6000 10000 14000

Overflow rate (gal/day ft^)
Fig. 3.17. Suspended solids removal (column 4, Table 3.8) versus overflow rate (column 3,
Table 3.8).
t (min) Overflow rate (gal/day ft^)
Fig. 3.18. Suspended solids removal versus Fig. 3.19. Suspended solids removal versus
detention time for different SS concentrations overflow rate for different SS concentrations
ix, < X 2< x^y iX, < X 2< X 3 ).
of 430 mg/liter. If similar calculations are performed for other values of these
concentrations ( A ' l , ^ ^ 2 , ^ 3 » the data plotted in Figs. 3.16 and 3.17 yield
families of curves, as indicated in Figs. 3.18 and 3.19.
Step 4d. Prepare a plot of percentage of particles with less than stated velocity
versus setthng velocity (ft/h) for a residence time of 25 min. Calculations needed
to prepare Fig. 3.20 are presented in Table 3.9. Figure 3.20 is not required for
Settling velocity, (ft/h)
Fig. 3.20. Percentage of particles with less than stated velocity (column 2, Table 3.9) versus
settling velocity (column 4, Table 3.9) (Example 3.3 for t = 25 min).
TABLE 3.9
Percentage of Particles with Less than Stated Velocity versus Settling Velocity
(1) (2) (3) (4)

SoHds removal (%) Particles with A (ft) Vi = VsWH)
Step 4b less than stated Step 4b = 14.4 (A/6.0)
t = 25 min velocity (%) t = 25 min (ft/h)
100 - (1)
30 70 6.0 14.4
35 65 4.2 10.08
45 55 2.4 5.76
55 45 1.4 3.36
65 35 0.84 2.016
75 25 0.28 0.672
" / = 25 min.
the design calculations; it is shown because a typical graph of this type, namely
Fig. 3.15, was utilized in developing Eq. (3.39). The plot is presented in Fig.
3.20.
3.6.1. Design Calculations from Laboratory Data
For purposes of scale-up, the fact that the efficiency of the process in an actual
settling tank is reduced owing to the effect of parameters such as turbulence,
short circuiting, and interference of the inlet and outlet must be taken into
account. The net effect of these factors results in a decrease of the overfiow rate
and an increase in the detention time over values derived from the laboratory
analysis. For design purposes, it is customary to divide the overflow rate
obtained from the laboratory analysis by a factor ranging between 1.25 and
TABLE 3.10
Design Values {Primary Clarifiers)
Depth: 7-12 ft
Detention time: 0.5-1.5 h
Flow-through velocity: V = \-5 ft/min
Overflow rate: 900-1200 gal/day ft^
Efficiencies
SS removal: 40-60%
B O D removal: 30-50%
1 . 7 5 , and to multiply the detention time by a factor in the same range [6,7].
Table 3 . 1 0 presents some commonly used design values.
3.6.2. Materia/ Balance for a Primary Clarifier
Primary clarifiers are designed for a specified removal (usually 40-60%) of the
suspended sohds in the influent. Referring to Fig. 3 . 2 1 , let
Qo: M G D of influent,
Xo: mg/liter of suspended solids in influent,
öe · M G D of effluent (overflow from clarifier),
: mg/liter of suspended solids remaining in overflow,
gu! M G D of underflow from clarifier, and
A'u: mg/liter of suspended solids in underflow.
An overall balance for the liquor yields
00 = Ö e + Qu (3.41)
A material balance for suspended solids yields*
00^0 = QeX. -f QuXu (3.42)
Combining Eqs. (3.41) and (3.42) one obtains
ge = (3.43)
Xu Xe
and
Qu = Q o - ße (3.44)
The concentration of underflow sludge is selected taking into account the

subsequent procedure for its handling and disposal (Chapter 7 ) . Tests can be
* Note on units [Eq. (3.42)]. If Qs are in M G D and A's are in mg/Hter ( = ppm = Ib/Mlb).
Mgal liquor lb solids lb of liquor

Q— - - X X X 8.34 = lb solids/day
day Mlb liquor gal of liquor
Therefore, factor 8.34 cancels out leading to Eq. (3.42) as written.

Overflow
Influent
Qo.Xo
Underflow
Q„.x„
Fig. 3.21. Material balance for primary clarifier.
2.0 Π—I—I—I—I—Γ
S i.oP
1^31.5
_i_Zl 1—I—I—I 1 L_ J I—I—I—I—I—L_
30 40 50 60 90
Settling time, t (min)
Fig. 3.22. Plot of solids in the sludge versus detention time (Example 3.4).
performed to correlate sludge concentration with settling time, by withdrawing

samples from tap 4 of the laboratory setthng column (Fig. 3.12). A plot of
percentage of solids for the compacted sludge versus settling time is then
constructed. A typical graph of this type is shown in Fig. 3.22. Design procedure
for a primary clarifier is illustrated in Example 3.4.
Example 3.4. A raw wastewater for which the flow rate is 1.5 M G D contains
an average of 430 mg/liter of suspended solids. D a t a shown in Table 3.3 are
obtained from laboratory setthng tests.
1. Design a primary clarifier of circular cross section, i.e., determine its
diameter and effective depth to remove 50% of the suspended solids. Use a scale-
u p factor of 1.25.
2. Calculate the daily accumulation of sludge (lb/day), the lb/day of dry solids
settled and the average pumping rate in gal/min. A plot of percentage of sohds
for the compacted sludge versus settling time was determined experimentally
and is shown in Fig. 3.22.
3. For the clarifier designed in Part 1, what would be the percentage of
removal of suspended solids if flow rate is doubled.
Solution to Part 1
Step 1. From Figs. 3.16 and 3.17, for 50% removal, read
Fig. 3.16: / = 31.5 min
Fig. 3.17: 2000 gal/day ft^
Utilizing a scale-up factor of 1.25,
t= 1.25 X 31.5 = 39.4 min

overflow rate (OR) = 2000/1.25 = 1600 gal/day ft^
Step 2. From Fig. 3.22, for ί = 31.5 min, read
X^= 1.07% (10,700 mg/liter)
For 50% removal
Jre = (0.5)(430) = 215 mg/liter
Step 3. Calculate and
From Eq. (3.43)
e „ ( A - „ - Λ · , ) ^ 1.5(10,700 - 430)
X^-X, 1 0 , 7 0 0 - 215
From Eq. (3.44)
Öu = Ö0 - ß e = 1.50 - 1.47 = 0.03 MGD (30,000 gal/day)
Step 4. Calculate cross-sectional area, diameter, and effective height for
clarifier:
^ 1,470,000 (gal/day)
OR 1600 gal/day ft^
H = ^ =^?iiJV 1.500,00θ|ϊ-
^ day
X
1440 mm
X 39.4mm χ -Í^V"!'
7.48 g a l / /
= 5.97 ft
Solution to Part 2
gal
gu = 30,000 of compacted sludge
dav
— of slurry = 3 0 , 0 0 0 — χ 8 . 3 4 — = 250,200 -J^

day day gal day
lb 1.07 lb
of dry solids = χ 250,200 = 2677 of dry solids
day 100 day
gal day
Pumping rate = 30,000 f — χ - — = 20.8 GPM
day 1440 m m
Solution to Part 3 If flow rate is doubled, the theoretical retention time is cut
in half, i.e.,
i = 3 1 . 5 / 2 = 15.8 min
F r o m Fig. 3.16, read: SS removal (%) = 36%.
3.7. Zone S e t t l i n g
Zone settling occurs in clarifiers of activated or chemically coagulated sludge

when the concentration exceeds 500 mg/liter. The sludge blanket exhibits several
distinct zones. Each zone is characterized by a specific sludge concentration and
settling velocity. Consider what happens when a suspension which initially has a
uniform sludge concentration XQ (mg/liter) is placed in a settling cylinder (Fig.
3.23).
Sludge begins to settle out and an interface (interface 1) is established between
the surface of the blanket of settling sludge and the clarified hquid above. The
zone below the clarified liquid is called the interfacial zone. Concentration of the
(b) (d)
Clarified
[water zone
Interface I- Clarified
Iwater zone
4fêrfGCial [Coalescence
- ZQQe__ of interfacesi
I and 2
TwnsiUon
zone
llnterfoce .ly.
BSMPACTION
~:iöjner-
t = 0 t^.>t>0
Uniform sludge Beginning of End of

concentration X„ compaction compaction
Clarification process Thickening process
Fig. 3.23. Zone settling.

sludge in this zone is uniform, and it settles as a blanket with a constant velocity
(KJ. Simultaneously with formation of interface 1 and interfacial zone,
compaction of suspended solids starts at the bottom of the cylinder (compaction
zone). In this zone concentration of SS is also uniform, and the interface
bordering this zone (interface 2) rises in the cyhnder with a constant velocity (V).
Between the interfacial and compaction zones there is a transition zone.
There, the velocity of settling solids decreases owing to increase of viscosity and
density of the suspension. In this same zone, sludge changes gradually in
concentration from that of the interfacial zone to that of the compaction zone.
Consider interfaces 1 and 2 in Fig. 3.23b. Interface 1 moves downward with a
constant velocity K^, whereas interface 2 moves upward with a constant velocity
V. Eventually, interfacial and compaction zones meet, at a critical time (t^) and
the transition zone fades away (Fig. 3.23c). At this time, the settled sludge
exhibits a uniform concentration X^, which is termed the critical concentration.
^—1000 HOOO-
o
σ 500 h
.1
α>
Χ
Fig. 3.24. Sludge versus settling curve.
Compaction starts and the sludge begins to thicken, eventually reaching an

ultimate concentration (Fig. 3.23d). Sedimentation velocity at time
corresponds to a value V^, which is given by the slope of the tangent to the
settling curve at C, as indicated in Fig. 3.24 where < V^, The procedure for
designing clarifiers operating under conditions of zone setthng is as follows:
1. Calculate the minimum surface area required to allow for clarification of
sludge.
2. Calculate the minimum surface area required to provide for thickening of
sludge to the desired underflow concentration.
3. Take the larger of these two areas as the design area for the clarifier.
3.7.7. Laboratory Measurements
To obtain the parameters necessary for design of the clarifier, a settling test for
the sludge is performed in the laboratory using a 1000-ml graduated cyhnder (a
standard graduate cylinder has a height of 1.12 ft). The cylinder is filled with the
slurry to be studied. At the beginning of the experiment (/ = 0), slurry
concentration is uniform throughout the cylinder. Height of interface 1 is
recorded at selected time intervals. This yields the type of settling curve shown in
Fig. 3.24.
It is important to stir the suspension at a rate of about 5 rph. This stirring
simulates the action of the mechanical rakes utilized in sludge removal and
prevents stratification of the sludge.
Figure 3.24 shows that from the start of the experiment u p to a time ti,
interface 1 falls with a constant velocity given by the slope of the tangent,
which essentially coincides with the settling curve from t = 0 to t = t^. After
t = ti, this velocity decreases appreciably. At time t = t^, the velocity is given
by the slope of the tangent at C. At t^ compaction starts and the velocity is
further reduced until it becomes essentially zero, the tangent being parallel to the
abscissa.
Zone settling velocity (ZSV) corresponds to the velocity at which the
suspension settles prior to reaching the critical concentration X^, and is given by
the slope of the tangent AB in Fig. 3.24:
V, = OA/OB = Ho/t = 1.12// ft/min (3.45)

The laboratory determination of the zone setthng velocity just described
should be repeated with slurries of different initial concentrations XQ (mg/liter)
of suspended sohds. This leads to a family of setthng curves. Fig. 3.26 presents
the results of such experiments for an activated sludge at several concentrations
Xo ranging from 589 to 11,775 mg/hter (Example 3.5). The values of V-, (ZSV) are
obtained by drawing tangents to the initial section of each of these curves and
utilizing Eq. (3.45).
3.7.2. Determination of Minimum Surface Area Required

to Allow Clarification of the Sludge
Minimum surface area required for clarification depends on velocity at

which the suspension settles before reaching the interfacial critical concentration
X^. Under continuous flow conditions, velocity of water over the overflow weir
cannot exceed if clarification is to take place. This follows directly from the
basic concept of the ideal sedimentation tank [Section 3.4, Eqs. (3.23) and
(3.24)].
Therefore,
^ci = ß e / n (3.46)
where ße is the effluent flow rate (ft^/min); is the zone settling velocity
(ft/min); and Aî is the minimum surface area required for clarification (ft^). The
material balance equation for obtaining [Eq. (3.43)] was presented in Section
3.6.2.
The value of the zone settling velocity is determined as shown in Fig. 3.24
for the curve corresponding to the influent solids concentration XQ. Value of / is
read directly from the abscissa (point B). is then calculated from Eq. (3.45)
and Aci obtained from Eq. (3.46). This procedure is illustrated in Example 3.5.
3.7.3. Determination of the Minimum Surface Area Required
for Thickening of the Sludge
The fact that the cross-sectional area of the clarifier may be calculated to
insure that clarification is achieved, as described in Section 3.7.2, does not
necessarily mean that the desired underflow solids concentration X,, can be
reached. In fact, the cross-sectional area required for thickening is usually larger
than that required for clarification.
The procedure developed by Yoshioka [ 16] and Dick [4] for determination of
the minimum cross-section area required for thickening will be described in this
section. This procedure is based upon the following considerations.
Consider first the batch sedimentation carried out in the laboratory settling
cyhnder (Fig. 3.23). The capacity of the clarifier for carrying solids downwards
at a concentration X,, by gravity, under a batch situation is*
G^ = Xy, (3.47)
where
G B = batch solids-flux (lb solids/ft^ day),

X; = sludge concentration (mg/hter), and
F¡ = zone settling velocity at concentration Χ-, (ft/min).
From Eq. (3.47) a batch-flux curve can be constructed. Figure 3.25 presents a
typical batch-flux curve G B versus X,. Notice that the curve goes through a
maximum. The reason for this behavior is as follows. Remark that as Χ-,
increases the zone settling velocity decreases. This can be seen from inspection of
Fig. 3.26 (Example 3.5), which presents zone setthng data for sludges of
increasing concentrations Χ-,: as Z¡ increases the slope of the tangents to the
initial section of the settling curves, i.e., the zone settling velocities F¡, do
decrease. At very high concentrations when the suspension approaches an
equihbrium position F¡ 0 and thus, from Eq. (3.47), C B 0. On the other
* Note on units for Eq. ( 3 . 4 7 ) . For Χ-, = mg solids/liter and F¡ = ft/min, then
lb solids lb liquor ft min

G b = X; X 62.4——^ χ χ 1440
Mlb liquor ft^ Hquor min day
G b = 0 . 0 8 9 8 5 6 X, V, lb solids/ft^ day
If Xi is in percentage of solids,
G b = 8 9 8 . 5 6 X, V, lb solids/ft^ day
hand, at very low concentrations as Xi 0, Eq. (3.47) indicates that Gß also

approaches zero. This suggests that a maximum for Gß must occur at some
intermediate concentration Xi, thus explaining the shape of the batch-flux curve.
Consider now the sedimentation carried out in a continuous clarifier. The
solids are transported downwards by both gravity and the downward movement
resulting from the removal of sludge from the clarifier bottom. The flux equation
then becomes
G T = G B + Gu (3.48)
where
G T = total solids flux (lb solids/ft^ day),
G B = batch solids flux (lb solids/ft^ day), and
Gu = underflow solids flux (lb solids/ft^ day).
The term Gu can be written as
G, = X,V, (3.49)
where Κ„ is the sludge velocity due to underflow removal (ft/min). The total flux
G T can be varied by the designer by controlling since this is determined by the
underflow pumping rate.
Substitution into Eq. (3.48) of the values of G B and Gu given by Eqs. (3.47) and
(3.49) yields
Gj = X,V,X,V, (3.50)
If the value of Gj could be determined so that the desired solids concentration

is obtained at the clarifier underflow, the minimum cross-sectional area of
the continuous clarifier required for thickening the sludge would be readily
calculated from
Μ lb solids/day
-ft- (3.51)
G T ~ lb solids/ft^day
where*
M=QoXo (3.52)
Qo is the influent flow rate and Xq is the solids concentration in influent.
Consider the batch-flux curve in Fig. 3.25. Let X^, be the desired solids
concentration in the clarifier underflow, specified by the designer. M a r k this
value on the abscissa of Fig. 3.25. The graphical constructions corresponding to
* N o t e on units for Eq. (3.52). If Q o is in millions of gallons per day ( M G D ) and Xq is in mg

solids/liter, then
Mgal liquor lb solids lb liquor

M = Qo X ^ o „ . . ...... X 8.34-
day Mlb hquor gal liquor
Μ = S34QoXo lb solids/day
Solids concentrations
Fig. 3.25. Graphical procedure for determination of Gj.
this value are indicated in Fig. 3.25 by the fuh lines ( - ) . F r o m Λ^^ draw a
tangent to the batch-flux curve as indicated. Τ is the point of tangency. It will be
shown next that the ordinate intercept of this tangent, i.e., distance OB,
corresponds to the hmiting solids flux Gj that can be allowed in order to obtain
the underflow concentration X^,. This will be shown by following the two steps
described next.
Step 1. Consider the point of tangency T, corresponding to the abscissa X,.

The ordinate of the point of tangency (i.e., distance OA) is equal to Cß. If a
straight line is drawn connecting the origin 0 to the point of tangency T, the slope
of this line equals K¡ since
OA = OX, tan a, Cß = Χ·, tan α (3.53)

and since from Eq. (3.47)
Gß = XiKi
then, slope = t a n a = V\.
Step 2. F r o m Fig. 3.25 it follows that
OB =OAûtanjS (3.54)
Consider now the underflow velocity K^,:

V. = QJA, (3.55)
Multiplying simultaneously the numerator and denominator of Eq. (3.55) by

we obtain
Vu = QuXu/Α,Χ, (3.56)
On the other hand, a material balance for suspended solids around the clarifier
yields
QoXo = M= Q,X, + Q,X, (3.57)
Since for a well-designed clarifier, the loss of solids in the overflow (Q^Xc) should
be negligible, an approximate form of the material balance equation is
QoXo = M^ Q,X, (3.58)
Utilizing Eq. (3.58), Eq. (3.56) yields
V. ^ QoXo/AX, = M/A,X, (3.59)
F r o m Eq. (3.51) it follows that
Μ = A,Gj (3.60)
Substituting this value in the numerator of Eq. (3.59) and simplifying,
Vu = Gj/X, (3.61)
or
Gj = X,V, (3.62)
Comparing Eqs. (3.54) and (3.62) it can be concluded that the slope of line
in Fig. 3.25 equals to underflow velocity Κ„, and that the tangent intercept
OB equals the total limiting flux GT. Since, from Eq. (3.48), GT = GB -f G « , it
follows that vertical distance AB in Fig. 3.25 must equal the value of the
underflow solids flux G^,.
Summary The graphical procedure to determine the minimum cross-

sectional area Ai required for thickening is as follows:
Step 1. Construct the batch solids flux curve from Eq. (3.47) utilizing the ZSV
(Fi) obtained from the batch laboratory settling tests at several solids
concentrations X,.
Step 2. F r o m the specified value of the underflow concentration X, on the
abscissa, draw a tangent to the solids-flux curve. The intersection of this tangent
with the ordinate axis is the hmiting total solids flux G T .
Step 3. The minimum cross-sectional area required for thickening is then
obtained from Eq. (3.51) and (3.58), i.e.,
A, = M/Gj = QoXo/Gj (3.63)
F r o m Fig. 3.25 one can also draw the foUowing conclusions:

1. If for the clarifier which has been designed with a cross-sectional area A^,
corresponding to the total flux Gj, i.e., distance OB, and underflow con-
centration X ^ , i.e., distance OX^, one would increase the total solids flux by
increasing the underflow pumping rate, it would not be possible any longer to
achieve the underflow concentration X^^. Indeed, if this new increased total solids
flux was GJ (corresponding to distance OB'), the underflow concentration X'^,
which could be reached would be obtained by taking from point B' a tangent to
the batch-flux curve. In this case the point of tangency would be T' and the
underflow concentration X'^ would be less than X ^ . The values of the solids
batch-flux Gß and the solids underflow flux G[^ are also indicated in Fig. 3.25 for
this situation. This corresponds to the graphical construction indicated by the
dotted lines (---).
2. On the other hand, if for this same clarifier which has been designed with a
cross-sectional area ^t, the underflow pumping rate is decreased, so that the new
value of the total solids flux is G'j (corresponding to distance OB"), the underflow
solids concentration X'¿ would be obtained by taking from point B" a tangent to
the batch-flux curve. In this case the point of tangency would be T" and the
underflow concentration X'¿ will be greater than X^,. The values of the solids
batch-flux Gß and the solids underflow flux G'¿ are also indicated in Fig. 3.25 for
this situation. This corresponds to the graphical construction indicated by the
dotted hnes (· · ).
Typical detention times for secondary clarifiers following aeration tanks in

activated sludge plants are 2.0-2.5 h.
Example 3.5. Setthng data for an activated sludge were obtained in a

laboratory utilizing standard 1000-ml graduated cylinders. A range of eleven
concentrations of suspended solids from 589 to 11,775 mg/liter was studied and
the results are presented in Fig. 3.26.
For the plant to be designed the concentration of suspended solids in the
influent to the secondary clarifier is taken as 3533 mg/liter. The zone setthng
data for this concentration is as follows (plotted as curve 4 in Fig. 3.26):
t (min) Interface height (ml)
0 1000
2.5 850
5.0 710
7.5 600
10.0 500
12.5 430
15.0 365
20.0 265
25.0 200
30.0 165
35.0 135
40.0 120
45.0 115
1000
800
- 600
«o
o 400
c
200
20 30 40 50 60
Fig. 3.26. Zone settling data (Example 3.5).
Curve Χ; (mg/liter)
1 589
2 1,178
3 2,355
4 3,533
5 4,710
6 5,888
7 7,065
8 8,243
9 9,420
10 10,598
11 11,775
The total flow to the clarifier is 2.0 M G D . The desired underflow sohds
concentration is specified as 11,765 mg/liter. Neglect the solids concentration in
the clarifier overflow. Design the secondary clarifier for this service.
Solutior)
Step 1. Calculation of the ZSVs at the several solids concentrations.

A sample calculation is shown here for XQ = 3533 mg/Hter (curve 4, Fig. 3.26).
The curve is plotted from the laboratory data and a tangent is drawn to the initial
section of the curve.
Abscissa intercept of the tangent (point B, Fig. 3.26): 17.8 min. Then from Eq.
(3.45)
V, = Ho/t = OA/OB = 1.12/17.8 = 0.063 ft/min

10
1.0 h
0.5 μ
0.1
100 1000 10000
X, (mg/liter)
Fig. 3.27. Graph of ZSV versus X, (Example 3.5).
In the same manner ZSV values are determined for the other setthng curves
shown in Fig. 3.26. The corresponding ZSV values are tabulated [Column (3) of
Table 3.11]. A log-log plot of versus Χ-, is presented in Fig. 3.27.
Step 2. Determination of the minimum surface area required for clarification
(^cl).
1. Determine from Eq. (3.43):
ß e = Ö o ( ^ u - ΧώΙ(Χ. - Χ.) = 2.0(11,765 - 3533)/(11,765 - 0)

= 1.4 MGD
Also, from Eq. (3.44)
ßu = Ö0 - ße = 2.0 - 1.4 = 0.6 MGD

2. Calculate Α^χ from Eq. (3.46):
^ i = ße/Ks = f
\
1,400,000^
day
X
ft^
7.48 gal 1440
day
min^
\/
0.063
ft
mm
= 2063 ft^
Step 3, Construct the batch-flux curve. The calculations are presented in

Table 3 . 1 1 .
Step 4. Determination of the minimum surface area required for thickening.
1. Plot the batch-flux curve (Fig. 3 . 2 8 ) . This is a plot of G B [Column ( 4 ) , Table

3 . 1 1 ] versus X, [Column ( 2 ) , Table 3 . 1 1 ] .
2 . Mark X^=^\\ , 7 6 5 mg/Uter at the abscissa axis of Fig. 3 . 2 8 . F r o m X^ draw
a tangent to the batch solid-flux curve. This tangent cuts the ordinate axis at
4000 8000 t Xi 12
p0
o0o0 Τ 16000
y- χ y' yu γ"
^ Xu ^ ^û
SOLIDS CONCENTRATIONS
Fig. 3.28. Graphical solution (Example 3.5).

TABLE 3.11
Calculation of ZSVs and Batch-Solid Fluxes G b at Several Suspended Solids

Concentrations^
(1) (2) (3) (4)

Settling V, = Holt = V, G b = 0.089 856 AîFi
curve no. (mg/liter) 1.12/Í (4) = 0.089 856(2)(3)
(Fig. 3.24) (ft/min) (lb solids/ftMay)
1 589 0.167 8.84

2 1,178 0.159 16.83
3 2,355 0.096 20.31
4 3,533 0.063 20.00
5 4,710 0.040 16.93
6 5,888 0.024 12.70
7 7,065 0.0148 9.40
8 8,243 0.0095 7.04
9 9,420 0.00626 5.30
10 10,598 0.00402 3.83
11 11,775 0.00278 2.94
" See Example 3.5.
Gj = 23.25 lb solids/ft^ day. The graphical constructions corresponding to

X, = 11,765 mg/liter are indicated in Fig. 3.28 by the full lines ( ), and the
construction points by letters A, B, T, X„ and X,. Solid fluxes are denoted as
GB, G u , and Gj.
3. Calculate A^, Eq. (3.63):
A, = M/Gj = QoXolG^
where
Mgal lb lb
Μ = QoXo = 2.0 — ^ X 8.34 — - χ 3533 — - = 58,930 lb solids/day
day gal Mlb
GT = 23.25 lb solids/ft^ day
. • . Λ = 58,930/23.25 = 2535 ft^
Since Ac\ (Step 2) was 2063 ft^ base design upon A^. Consequently the cross-
sectional area of the clarifier will be 2535 ft^, corresponding to a diameter of
D = {4Α/πΥ^^ = ft
The values for the fluxes GB and Gu are read from Fig. 3.28. The application in
Example 3.5 (X^^ = 11,765 mg/hter) is hence referred to as Case 1. All pertinent
values are presented in Table 3.12.
If a detention time of 2 h is adopted, the clarifier volume will be
V=Qot = 2,000,000 1^ X 2 h X ^ = 166,667 gal

day 24 h
TABLE 3.12
Summary of Results"
Case 1 Case 2 Case 3

(Example 3.5) (Example 3.6) (Example 3.6)
Construction lines (see Fig. 3.28)

Construction points A, B, T, X,, X, A', B', T , z:, X'^ A", B", T", X'/, X';
(see Fig. 3.28)
Qo (MGOf 2.0 2.0 2.0
Xo (mg/liter)" 3,533 3,533 3,533
F„ Eq. (3.45); (ft/min)'' 0.063 0.063 0.063
X^ (mg/liter) 11,765 9,420 14,130
X; (mg/liter) (see Fig. 3.28) 7,654 5,280 10,244
G B (Ib/ftMay) (see Fig. 3.28) 8.24 14.7 4.2
Gu (Ib/ft^day) (see Fig. 3.28) 15.01 18.5 11.0
G T (Ib/ftMay) (see Fig. 3.28) 23.25 33.2 15.2
Vi = GB/0.089 856A'i, Eq. (3.47) 0.012 0.0310
(ft/min) 0.00456
V^ = Gu/0.089 856X¡, Eq. (3.49) 0.022 0.0390
(ft/min) 0.0120
öe, Eq. (3.43) (MGD) 1.4 1.25 1.5
ρ„, Eq. (3.44) (MGD) 0.6 0.75 0.5
Eq. (3.46) (ft^) 2,063 1,842 2,210
A, = 58,930/GT, Eq. (3.63) (ft^) 2,535 1,775 3,877
Din) 56.8 (ê) 48.4 (^ J 70.3 (Ad
Thickening ratio: y = XJXo 3.33 2.67 4.0
" See Examples 3.5 and 3.6.

Common values to Cases 1, 2, and 3.
or
ft^
166,667 gal χ = 22,282 ft^
7.48 gal
The clarifier depth will be
H=V/A = 22,282 ftV2535 ft^ = 8.79 ft
Example 3.6. For the application in Example 3.5 calculate the cross-sectional
areas required, solid fluxes, and flow rates of overflow and underflow for two
cases, hence referred to as Cases 2 and 3 :
Case 2. For a desired underflow solids concentration
= 9,420 mg/liter
Case 3. For a desired underflow solids concentration
Aû= 14,130 mg/liter
Solution The graphical constructions for Cases 2 and 3 are indicated in Fig.
3.28 by the hnes (---) and ( · )» respectively. The graphical and calculation
procedures are identical to those for Case 1 ( X , = 11,765 mg/liter) considered in

Example 3.5. The final results are summarized in Table 3.12.
Inspection of Table 3.12 leads to the conclusion that only for Case 2 when a
relatively thin underflow was obtained (X'^ = 9420 mg/liter and ratio
XJXQ = 2.67), the cross-sectional area was controlled by the clarificafion
requirements. For most typical operations, like Cases 1 and 3, the sizing of the
vessel is governed by the thickening requirements. General considerations about
the relative magnitude of areas and A^ are discussed in the next section.
3.7.4. Comparison of Areas Aî and A^
The minimum surface area required for clarification was calculated from Eq.
(3.46) and is based upon the consideration that under continuous flow
conditions the velocity of the water over the overflow weir cannot exceed if
clarification is to take place.
Examine now the processes of clarification and thickening taking place in the
laboratory cylinder (Fig. 3.23). At start of the experiment, let X^ be the uniform
sludge concentration throughout the cylinder. Total weight of solids in the
cylinder is XQAHQ where A is the cross-sectional area of the cylinder. Let be the
critical time counted from the beginning of the experiment, when interfacial and
compaction zones merge together (Fig. 3.22c). Let X^ be the critical con
centration which is uniform throughout this sludge zone formed by the merging
of interfacial and compaction zones.
A graphical procedure has been proposed [13] for determining t^. Consider
the settling curve in Fig. 3.29. D r a w two tangents (AA' and BB') to the two
branches of the curve. Tangent A A ' corresponds to the constant velocity of
settling for the interfacial zone [zone settling velocity (ZSV)], and tangent
BB' corresponds to the settling velocity for the compacted sludge. Point C
(corresponding to time is obtained by bisecting the angle formed by tangents
AA' and BB'. The abscissa of the point where the bisecting line cuts the settling
curve corresponds to the desired value of t^.
Consider now the thickening process.
^/Bisecting line
« zsv\ /
/
O /
/
σ \ /ά
\ /
c B'
Β
Fig. 3.29. Determination of Re settling time, t

1. Start of thickening (Fig. 3.23c). Time t^; concentration of SS in sludge zone

X^; and height of sludge zone H^.
2. End of thickening (Fig. 3.23d). The compacted sludge reaches desired
underflow concentration X^. The time at which this occurs is designated as
Height of the sludge zone is H^.
-j—Volume
of water
squeezed out
Λ—1==^=^—
(a) (b)
Fig. 3.30. Thickening process: (a) Άί t = and (b) at r =
Consider separately the sludge zone at the start and end of thickening (Fig.
3.30). Since the total mass of sludge in the cyhnder is constant, the following
material balance equation [Eq. (3.64)] is written, neglecting the a m o u n t of
suspended solids in the clarified water zone:
XoAHo = XcAH, = XÂH^ (3.64)

or
(3.65)
Consider Fig. 3.30. The volume of water that is squeezed out and discharged
over the overflow weir during the thickening process is calculated from
V = AXH, - Η J (3.66)
The time interval required to discharge this volume of water is t„ - t^. Average
rate of flow ö a v g (ft^/min) over the weir due to the thickening process is
Öavg = t^/(íu - ίο) = A¿H, - HMtu - Q (3.67)
Settling time (min)
Fig. 3.31. Determination of velocity V^.

Consider now the setthng curve and determine graphicaUy the setthng velocity
Kc at time t^ (tangent at point C ) . This is shown in Fig. 3.31.
Let Hi be the intercept of this tangent. Settling velocity at t^ is
K, = t a n a = (//i - / / J A c (3.68)
Under continuous flow conditions the velocity of the water over the weir cannot
be greater than K^, if thickening is to take place. Evidently the minimum surface
area required for thickening depends upon the volume υ of the water squeezed
out during the thickening process (Fig. 3.30). If the required degree of thickening
is considerable, i.e., Xy, » X^, this means that //„ « H^. F r o m Eq. (3.66), ν is then
a large volume and area A^ is likely to be greater than Α^χ, As indicated A^ is the
required cross-sectional area to allow overflow of this volume of water i;, in time
interval (/„ - t^, without exceeding the velocity V^. In this case the design is
governed by thickening considerations rather than by clarification require
ments. This corresponds to situation for Cases 1 and 3 (Table 3.12).
On the other hand, if a hmited degree of thickening is desired, the area Α^χ
required for clarification can be greater than area A^ for thickening. In this case
the design is governed by clarification requirements. This corresponds to the
situation for Case 2 (Table 3.12).
3.8. Types of Clarifiers
Clarifiers are classified according to geometry of their horizontal cross section

as (1) rectangular and (2) circular: (a) center and (b) peripheral feed. Sketches of
typical clarifiers are shown in Figs. 3.32a-3.32c.
1. Rectangular clarifier (Fig. 3.32a). In the type shown in Fig. 3.32a, scraped
sludge is moved toward the inlet end of the tank. Some other designs move
sludge toward the effluent end of the tank.
Scraping mechanism shown is of rotary-hoe type, consisting of a series of
short scrapers mounted on an endless chain, which make contact with the
bottom of the tank. It moves slowly at speeds of approximately 1 ft/min.
2a. Circular clarifier with center feed (Fig. 3.32b). Feed is at the center and
clear solution overflows to a collecting channel at the periphery. The bottom of
the clarifier has a minimum slope of 1 in./ft. Scraping mechanism is of plow
type to overcome inertia and prevent adherence of sludge to the bottom of the
tank.
2b. Circular clarifier with peripheral feed {Fig. 3.32c). The feed is at the
periphery and the clear solution overflows to a collecting channel at the center.
The other details are similar to those for the type shown in Fig. 3.32b.
The inlet section should be carefully designed for a uniform flow distribution
across the width and depth of the tank. Similarly, the outlet section should be
designed to collect the effluent uniformly.
A good design of inlet and outlet sections reduces possibilities of flow short
circuiting, which lead to poor performance of the clarifier. Proper positioning of
weirs and baffles, as indicated in Fig. 3.32, prevents short circuifing.
Tôtory-hoe type scraping mechanism
^7t
-Influent
2
Clear zone Í L·
-Baffle
Clear 'J
solution
overflow Discharge of
thickened
—^ sludge
Rotating mechanism
Feed
Overflow
piping
weir
Blade Άι
Plow-type
scraping mechanism Discharge of thickened sludge
(b)
Clear solution ng mechanism

outlet
^ ^ ^ 5 S ^ channel
-Influent
_ ^ Plow-type
LJ| LJ I I LJ 1 J L J ""J scraping

mechanism
lade Arm
3 - ^ D i s c h a r g e of thickened sludge
(C)
Fig. 3.32. Types of clarifiers: (a) rectangular, (b) circular (center feed), and (c) circular (peripheral
feed).
4. Flotation 127
4. FLOTATION
4.1. Introduction
Flotation is a process for separating low-density solids or liquid particles from
a liquid phase. Separation is brought about by introduction of gas (usually air)
bubbles into the liquid phase. T h e hquid phase is pressurized to an operating
pressure ranging from 30-60 psia (2-4 atm) in presence of sufficient air to
promote saturation of air in the water. Then this air-saturated hquid is
depressurized to atmospheric pressure by passage through a pressure-reducing
valve. Minute air bubbles are released from the solution because of depres-
surization. Suspended sohds or hquid particles, e.g., oil, are floated by these
minute air bubbles, causing them to rise to the surface of the tank. Concentrated
suspended solids are skimmed off by mechanical means from the tank surface.
Clarified liquor is withdrawn near the bottom and part of it may be recycled
Air injection
Pressurizing Thickened
pump sludge
Q(MGD) (negligible
Influent "TTHvolume)
(wastewater) C| (mg/liter
of dissolved air) Q(M60)
Effluent
C2 (mg/liter of dissolved air)
(a)
O(MGO)
Influent
Thickened
idge
(negligible volume)
Gross effluent
0 + R (MGD)
îeTeffluent
Q (MGD)
(b)
Fig. 3.33. Flotation systems: (a) without recycle, (b) with recycle.
1 28 3 Pretreatment and Primary Treatment
(Fig. 3.33b). A flotation system without recycle is shown diagrammatically in

Fig. 3.33a.
In the field of wastewater treatment, flotation is used for the following
purposes: (1) separation of greases, oils, fibers, and other low-density solids
from wastewaters; (2) thickening of the sludge from the activated sludge
process; and (3) thickening of flocculated chemical sludges resulting from
chemical coagulation treatment. Superior effluent quality, i.e., effluent contain
ing lower percentage of suspended soUds, and economy in power are achieved by
flotation systems with recycle.
Basic components of a flotation system are (1) pressurizing p u m p ; (2) air
injection facilities; (3) retention tank (to provide air-liquid contact); (4)
pressure-reducing valve; and (5) flotation tank.
4.2. Evaluation of Flotation Variables for Process

Design
4.2.7. Parameter A/S
For design of flotation systems, a fundamental parameter commonly utihzed

is a dimensionless air to solids ratio (A/S) defined by
lb/day of air released by depressurization

lb/day of solids in the influent
This parameter is estimated from studies with a laboratory-scale flotation cell

or pilot-plant data. The value of A is obtained from determinations of dissolved
air (mg/liter) at sampling locations indicated as (1) and (2) in Figs. 3.33a and
3.33b. Thus
A = Ai-A2 (3.70)
where A is the lb/day of air released by depressurization; Αχ is the lb/day of

dissolved air at ( i ) (Figs. 3.33a and 3.33b); and ^2 is the lb/day of dissolved air
at (2) (Figs. 3.33a and 3.33b).
F o r flotation systems without recycle*
Mgal liquor lb air lb hquor lb air

day M l b hquor gal hquor day
(3.71)
Similarly [Eq. (3.72)]
A2 = S34QC2 lb air/day (3,72)
Therefore
A=Ai-A2 = 8 . 3 4 ö ( C i - C2) lb/day of air released (3.73)
* Notice that Ci mg of air/liter of liquor = Ci lb of air/Mlb of liquor.

4. Flotation 129
For flotation systems with recycle, the corresponding equation is

A = A,-A2 = 8.34i?(Ci - C2) lb/day of air released (3.74)
If XQ is the concentration of suspended solids (mg/liter) in the influent, the value
of S [denominator of Eq. (3.69)] is
Mgal hquor 8.34 lb hquor lb SS lb SS
^ = Q T 7 .7 ^ „ X Xo = 8.34 QXQ
day gal hquor Mlb hquor day
(3.75)
Substitution of Eqs. (3.73) [or Eq. (3.74)] and (3.75) in Eq. (3.69) leads to the
following:
Flotation systems without recycle:
AIS = (Ci - €2)1X0 (3.76)

Flotation systems with recycle:
AIS = {RIQ){C,-C2)IXQ (3.77)
4.2.2. Correlation of Flotation Variables to Parameter AIS

By use of a laboratory flotation ceh or pilot-plant data, it is possible to
correlate mg/liter of suspended solids contained in hquid effluent to parameter
AIS. A typical correlation curve for a wastewater has the shape indicated in Fig.
3.34.
For a given influent, the lb/day of suspended solids (term S in ratio AIS) is
fixed. Ratio AIS increases by operating at higher air rates, which results in
increase of air released {A). F r o m Fig. 3.34 it follows that a higher quality
effluent is obtained.
0.06 Γ
0 50 100
SS in effluent (mg/liter)
Fig. 3.34. Typical correlation of parameter A/S versus concentration of SS in effluent.

1 30 3. Pretreatment and Primary Treatment
G r a p h s hke Fig. 3 . 3 4 permit selection of the A/S ratio for a required degree of
effluent clarification. These curves indicate that increasing the A/S ratio beyond
an optimum value does not result in substantial reduction in effluent suspended
solids.
Judicious selection of the A/S ratio involves an economical balance between
equipment and maintenance costs and desired effluent quality. Typical range of
A/S ratios for thickening of sludges in wastewater treatment is 0 . 0 0 5 - 0 . 0 6 0 .
4.2.3. Alternative Expressions for Parameter A/S*
Consider Eq. ( 3 . 7 2 ) for A2, where C2 is the solubility of air in water in mg/hter.
Frequently the solubility of air is expressed in terms of cm^ of air/liter of water. It
is assumed that conditions at ( 2 ) (Figs. 3.33a and 3 . 3 3 b ) are atmospheric
pressure and ambient temperature. Solubility of air in water in cm^ of air/hter of
water [hence denoted as S^'] is presented in Table 3 . 1 3 for atmospheric pressure
at several temperatures.
This concentration of dissolved air in cm^ of air/hter of water is converted to
the value C2 (mg air/liter of water):
Sa cm^ air/hter water χ mg air/cm^ air = C2 mg air/hter water (3.78)
Utilize for the density of air a mean value of 1.2 mg/cm^. (This corresponds to
the value at 1 atm and 20°C.) Then
C2 = 1.2Sa (3.79)
Consider now Eq. ( 3 . 7 1 ) , where Ci is the solubility of air in mg/liter. Define a

coefficient φ as the ratio of the solubihties of air in water at the retention tank
pressure to that at atmospheric pressure, i.e.,
φ = CJC2 (3.80)
Assuming that the water is saturated with air and that Henry's law is applicable,
one can write
φ = €,/€2 = Ρ/1=Ρ (3.81)
T A B L E 3.13
Solubility of Air in Water at Atmospheric

Pressure at Several Temperatures"
Temperature (°C) 5a (cm^/liter)
0 29.2
10 22.8
20 18.7
30 15.7
" From Ref. [ 1 1 ] .
* See Ref. [ 5 ] .
4. Flotation 131
where Ρ is the operating pressure in atm. (For C2, the pressure is taken as 1 atm.)
In practice a factor / is included in Eq. (3.81) to account for incomplete
saturation. Thus
Φ=/Ρ (3.82)
The value of / depends upon the design of the retention tank. At 20°C, for
pressure range 30-60 psia utilized in most flotation systems, the value of/ i n Eq.
(3.82) is usuahy between 0.5 and 0.8. Adopting the more conservative value of
/ = 0.5, the values of φ vary from
30 psia ( = 30/14.7 = 2.04 a t m ) : 0.5 χ 2.04 = 1.02
60 psia ( = 60/14.7 = 4.08 a t m ) : 0.5 χ 4,08 = 2.04
Therefore for the pressure range from 30-60 psia at 20°C, solubility of air in
baffled retention tanks varies from 1.02-2.04 times its saturation value at 20''C
and 1 atm. Since from Eq. (3.81)
Ci = φ€2 (3.83)
substitution of φ and C2 by their values given by Eqs. (3.82) and (3.79) yields
Ci = / P ( 1 . 2 S J (3.84)
Substitution of values of Ci and C2 given, respectively, by Eqs. (3.84) and (3.79)
in Eqs. (3.76) and (3.77) leads to the following:
Flotation systems without recycle:
A/S=\2S,(fP-\)/Xo (3.85)
Flotation systems with recycle:
A/S = (R/Q)\.2S,{fP - \)/Xo (3.86)
4.3. Design Procedure for Flotation Units W i t h o u t

and W i t h Recycle
4.3.1. Flotation Systems Without Recycle
From Eq. (3.85) it follows that if a ratio A/S is s e l e c t e d , / b e i n g fixed for a

selected type of retention tank and Xq from characteristics of the influent, this
amounts to specification of operating pressure P, which is calculated from Eq.
(3.85) as
Ρ = (\/mA/S)Xo/\2S, + 1] (3.87)
Design of flotation systems without recycle involves calculating the required
operating pressure [Eq. (3.87)] and determining the cross-sectional area of the
flotation unit. This area is calculated from a selected value of the overflow rate,
usually a value between 2 and 4 gal/min ft^.
Example 3.7. Laboratory flotation tests for a given wastewater indicate
optimum air/solids ratio (A/S) as 0.04 lb air/lb solids. Flow of wastewater is 1.0
M G D and it contains 250 mg/liter of suspended solids. Laboratory flotation

tests (without recycle) indicate for a ratio A/S = 0.04 an optimum effluent
containing 25 mg/hter of suspended solids. T a k e / = 0.50 for retention tank and
an operating temperature of 20°C. Design a flotation system without recycle for
this service.
Solution
Step 1. Select A/S = 0.04, as stated.

Step 2. Calculate Ρ from Eq. (3.87).
Ρ = (l/0.5)(0.04 X 250/1.2 χ 18.7 + 1 ) = 2.9 atm
Step 3. Select an overflow rate ( O R ) :
OR = 3 gal/min ft^
Step 4. Calculate required surface area:
ρ =1.0 MGD
or
Q = 1,000,000 gal/day χ day/24 h χ h/60 min = 695 gal/min
695 gal/min
Surface area (ft^) = ρ / O R = . Γ , Γ , · . 2 = ^32 ft^
3.0 gal/min ft^
4.3.2. Flotation Systems With Recycle
It follows from Eq. (3.86) that for a specific application (i.e., for fixed values of
/ a n d XQ), one must specify not only the A/S ratio, but also the recycle before the
operating pressure Ρ becomes fixed. The usual procedure is to specify an
operating pressure Ρ and an A/S ratio and calculate the required recycle from
Eq. (3.86), which solved for R yields
R = {A/S)QXo/\2S^fP - 1) (3.88)
The design procedure for flotation systems with recycle is illustrated by
Example 3.8.
Example 3.8. F o r the application in Example 3.7 design a flotation system

with recycle, taking an operating pressure of 2.9 atm.
Solution
Step 1. Select A/S = 0.04, as stated.

Step 2. Calculate R from Eq. (3.88).
R = (0.04)(1.0)(250)/1.2 χ 18.7(0,5 χ 2.9 - 1) = 0.99 M G D ^ 1.0 MGD
This means that recycle ratio R/Q is approximately unity.

Step 3. Select an overflow rate (OR)
OR = 3 gal/min ft^
5. Neutralization {and Equalization) 1 33
Step 4. Required surface area is defined as
ρ + 2.0 MGD
or
2,000,000 gal/day χ day/24 h χ h/60 min = β + = 1390 gal/min
Surface area (ft^) = (Ö + R)IOK = 1390/3.0 = 464 ft^
Surface area is twice as large as for the unit without recycle. However, an
effluent of superior quahty (i.e., SS < 25 mg/liter) is obtained.
4.4. Comparison of Flotation W i t h Gravity

Sedimentation
Capital costs are lower for flotation units than for gravity sedimentation. On
the other hand, operating costs are usually higher mainly owing to the cost of
power for air compression. F o r the case of activated sludge, allowable overflow
rates for flotation are about double the values for gravity sedimentation, thus
resulting in lower capital costs. This happens because the rise velocity of
activated sludge by flotation exceeds the setthng velocity by sedimentation.
Higher effluent quality is obtained from flotation units, where solids removal of
the order of 95% or higher are common. Flotation yield can be considerably
improved by addition of coagulants. The most c o m m o n coagulants utilized are
alum, ferric chloride, and polyelectrolytes.
5. NEUTRALIZATION (AND EQUALIZATION)
5.1. Neutralization in t h e Field of W a s t e w a t e r

Treatment
Neutralization treatment is often utilized in the following cases arising in

wastewater treatment.
1. Prior to discharge of the wastewater into a receiving water. The justifi
cation for neutralization is that aquatic hfe is sensitive to p H variations beyond a
narrow range around p H = 7.0.
2. Prior to discharge of industrial wastewaters to the municipal sewer system.
Specification of the p H of industrial discharges into municipal sewer systems is
frequently made. It is more economical to neutralize industrial wastewater
streams prior to the discharge into the municipal sewer, rather than attempting
to perform neutralization of the larger volume of combined domestic and
industrial sewage.
3. Prior to chemical or biological treatment. For biological treatment, p H of
the system is maintained within the range 6.5-8.5 to ensure optimum biological
activity. The biological process itself provides a neutralization and buffer
capacity as a result of production of CO2, which forms carbonates and
bicarbonates in solution. The degree of preneutralization required for biological

treatment depends on two factors: ( 1 ) the alkahnity or acidity present in the
wastewater and (2) the mg/liter B O D to be removed in the biological treatment.
The latter is related to the production of CO2, which may provide for partial
neutralization of alkaline wastes. This matter will be further discussed in
Chapter 5, Section 7.7.
5.2. M e t h o d s for N e u t r a l i z a t i o n of W a s t e w a t e r s
Methods for neutralization of wastewaters include ( 1 ) equalization, which
consists of mixing acidic and alkahne waste streams available in the plant, and
(2) direct p H control methods, which consist of addition of acids (or bases) for
neutralization of alkaline (or acidic) wastewater streams.
5.3. Equalization—Introduction
When utilized for purpose of neutralization, equalization involves mixing
wastewater streams of acidic and alkaline nature in an equalization basin.
Equalization is often used for purposes other than neutralization such as ( 1 ) to
smooth out individual wastewater stream flow variations, so that a composite
stream of relatively constant flow rate is fed to the treatment plant and (2) to
smooth out variations in influent B O D to the treatment facihties. Constant and
variable level equalization basins are utilized.
1. Constant level equalization basins. This arrangement is illustrated in Fig.

3.35. The level in the equalization basin is held constant. Therefore as the rate of
Mixer
Influent Effluent
Qe = Qrf(t)
rh=const.
Acid and alkaline streams
Fig. 3.35. Constant level equalization basin.
flow of influent varies, that of effluent is equally affected. Consequently this is

not a technique of flow equalization, simply a method of neutralization. If
fluctuations in the flow rate are too great, the effluent from the constant level
equahzation basin is fed to another equalization basin having as objective flow
equalization.
2. Variable level equalization basins. In this method of equalization, the
effluent is taken out at a constant rate and since the flow rate of influent varies
with time, the level of the equalization basin is variable. This method is also
utilized for the purpose of flow equalization, as well as providing for
neutralization. A diagram of a variable level equalization basin is shown in Fig.
3.36.
Influent Effluent
Qj = f ( t ) Q. ^constant
- o o
- h=variable-
Acid and alkaline
streams
Fig. 3.36. Variable level equalization basin.
Another method of equahzation consists of diverting the "excess" of the

incoming stream to a holding pond, from which a bleed stream is fed to the
equalization tank. This method is not used for neutralization purposes, but only
for equalization of B O D content or flow rate. This is illustrated by Fig. 3.37.
Influent
Excess
Holding pond
Bleed stream
Equal izafionlöTik
Fig. 3.37. "Holding pond" method of equalization.
5.4. Flow Equalization

The design of a variable level equalization basin is illustrated by Example 3.9.
The design procedure is that described by Hummenick [ 9 ] .
Example 3.9. A sampling study has been used to estabhsh the diurnal flow
pattern of wastewater to a treatment plant. A plot of instantaneous flow
(gal/min) versus time (0 to 24 h for a diurnal cycle), obtained in this study, is
shown in Fig. 3.38.
1. Determine the total daily volume of wastewater to be treated.
2. Calculate the average flow in gal/min.
3. Plot in same graph for which the ordinate is the cumulative flow in
thousands of gallons and the abscissa is time (0 to 24 h ) :
(a) A straight line (i) representing the constant rate of equalized flow leaving
the basin (i.e., cumulative outflow in thousands of gallons versus time).
700
Fig. 3.38. Graph of instantaneous flow versus time (Example 3.9).
600 Γ
Fig. 3.39. Graphical solution (Example 3.9).
(b) A curve (ii) representing the cumulative inflow to the basin in thousands
of gallons versus time.
4. Plot in this same graph another curve (in), for which the ordinate is the
basin contents in thousands of gallons and the abscissa is time. Explain how this
curve is constructed from line (i) and curve (ii). Specifically,
(a) What should be the basin contents in thousands of gallons at the start of
the operation (time 0) so that the lowest level ever reached is zero (but never
negative) at the most unfavorable time. W h a t is this time (hours)?
(b) What is the maximum of curve (iii)? Notice that this maximum
represents the minimum required storage volume for the basin.
Solution
1. Total daily volume of wastewater. The total daily volume of wastewater is
given by the area under the curve in Fig. 3.38, i.e.,
i=8
Total daily volume = ^ .4, = -f ^2 + ^3 + ^4 + + ^6 + ^ 7 +

i= 1
The area is calculated by adding the area of the eight trapezoids:

A, = 5 h X 60 min/h X i(200 + 260) = 69,000 gal
A2 = 3 h X 60 min/h X ^(260 + 700) = 86,400 gal
A3 = 2 h X 60 min/h X ^(700 + 650) = 81,000 gal
A4 = 2 h X 60 min/h X i(650 + 640) = 77,400 gal
As = 2 h X 60 min/h X f(640 + 600) = 74,400 gal
Ae = 4 h X 60 min/h X ^(600 + 100) = 84,000 gal
An = 5 h X 60 min/h X §(100 + 300) = 60,000 gal
As = 1 h X 60 min/h X i(300 + 200) = 15,000 gal
Total daily volume = 547,200 gal
2. The average flow (in gal/min) is
gal
547,200^ X ^ X = 380
day 24 h 60 min mm
3a. T h e constant rate of equalized flow is represented by a straight line from
he origin of Fig. 3.39 to a total flow of 547,200 gal after 24 h [line (i)].
3b. The cumulative inflow to the basin is found by determining the area under
he curve in Fig. 3.38 for the time in question.
= 5 h; Ai = 69,000 gal
= 8h; ^ 1 + ^ 2 = 69,000 + 86,400 = 155,400 gal
= IOh; (^1 + ^ 2 ) + /l3 = 155,400 + 81,000 = 236,400 gal
= 12h; ( ^ 1 + ^ 2 + ^ 3 ) + ^4 = 236,400 + 77,400 = 313,800 gal
= 14 h ; (^1 + A2 + A 3 + A 4 ) + As = 313,800 + 74,400 = 388,200 gal
= 18 h ; (Ai + A2 + A 3 + A 4 + As) + = 388,200 + 84,000
= 472,200 gal
/ = 23 h ; {Ai + A2 + A 3 + A 4 + A5 + A^) + ^ 7 = 472,200 + 60,000
= 532,200 gal
/ = 24 h ; {Ay + ^ 2 + ^ 3 + ^ 4 + ^ 5 + ^ 6 + Αη) + A^
= 532,200 + 15,000 = 547,200 gal
A plot of these values yields curve (ii) in Fig. 3.38.

4. Basin contents (gal) versus time.
(a) Basin contents at t = 0 and time for zero level. Construct Table 3.14 in
order to determine accurately the maximum positive and maximum negative
TABLE 3.14
Calculations for Example 3.9 (Refer to Fig. 3.39)
(1) (2) (3) (4) (5)

/(h) Ordinates of Ordinates of = (2) - (3) = 44.0 - (4)
line (i) curve (ii) [(iHii)] {(iii) = 44,0 -
(thousands (thousands (thousands [(iHii)]}
of gallons) of gallons) of gallons) (basin contents -
thousands
of gallons)
0 0 0 0 44
1 23 10 13 31
2 45 20 25 19
3 69 35 34 10
4 90 50 40 4
5 113 69 44 0
/ max. \
Vpositive/
6 136 93 43 1
7 160 120 40 4
8 182 155.4 26.6 17.4
9 205 190 15 29
10 228 236.4 - 8.4 52.4
11 250 280 -30 74
12 273 313.8 -40.8 84.8
13 296 352 -56 100
14 320 388.2 -68.2 112.2
15 340 410 -70 114
15.5 352 423 -71 115
16 363 434 -71 115
/ max. \ /max.\
Vnegative/ Vleveiy
16.5 376 445 -69 113
17 386 455 -69 113
18 410 472.2 -62.2 106.2
19 430 486 -56 100
20 455 500 -45 89
21 480 512 -32 76
22 500 522 -22 66
23 523 532.2 - 9.2 53.2
24 547.2 547.2 0 44
deviations between cumulative equalized outflow [line (i)] and cumulative

inflow [curve (ii)]. The maximum positive deviation occurs at 5 h
[(i) - (ii) = 44,000 gal] and the maximum negative deviation at 16 h
[(i) — (ii) = - 71,000 gal]. At the start of the operation the basin contents must
be at least 44,000 gal, so that after 5 h of operation the level will d r o p to exactly
zero. Then the level will start rising again to reach a maximum after 16 h.
In general the basin contents at any time will be equal to 44,000 gal minus the
deviation between the cumulative equalized outflow and the cumulative inflow
curves. These are the values shown in column 5 of Table 3.14 in thousands of
gallons and plotted in Fig. 3.39 [curve (iii)].
(b) Maximum of curve (iii). The maximum of the basin contents versus
time curve will occur after 16 h of operation. The basin contents then will be
44,000 + 71,000 = 115,000 gal
After 16 h the level drops again and at 24 h it wih be back to 44,000 gal. A new
cycle is then started, etc.
In summary, the minimum required storage is the sum of the maximum positive
and the maximum negative deviations between the cumulative outflow and
inflow curves (i.e., 44,000 + 71,000 = 115,000 gal). In practice a somewhat
higher storage capacity should be recommended so that the level will never d r o p
entirely to zero.
5.5. B O D Equalization
The design of an equalization basin requires the application of statistical
principles. The design procedure described in this section permits the de
termination of the basin size required to maintain the effluent B O D con
centration from the basin within a certain range, based on the variabihty of the
influent wastewater B O D and the level of confidence which the designer wishes
to place on the performance of the system.
Whenever possible it is recommended that the flow to a biological system,
specially activated sludge, be constant so that the only variabihty to the
biological process results from the influent wastewater quality.
In order to develop the design procedure for constant-level completely mixed
equalization basins for B O D damping certain statistical principles and defini
tions will be briefly reviewed.
5.5.1. Variarice
Consider a constant-level completely mixed equalization basin under con

stant flow conditions (Fig. 3.40).
The hydraulic residence time for the liquid in the basin is
/h = V/Q (3.89)
where V is the hquid volume in the basin (ft-^) and Q is the flow rate (ft^/h).
Fig. 3.40. Diagram for an equalization basin (effluent B O D equalization).
Let X; be the B O D values for the influent. Assume that these values fluctuate
vîth time; a typical graph depicting such fluctuations being presented in Fig.
3.41 for Example 3.10. Be jc¡ the mean of these values x , for the time period
selected for study.
1300
i
1200
1100
mo
^ 900
φ
\ 800
ε
Q ^
^ 600h
I 500|-
^400
300
200h
100
0
Κ) 12 i4 16 18 20 22 24 26
TIME (doys in Febmory)
Fig. 3.41. Chronological variation in influent B O D concentration [ 1 ] .
The effect of the equahzation basin will be to dampen these B O D fluctuations

about the average value, for the effluent. Obviously the mean value for the
effluent B O D , i.e., JCe, wih still be equal to x¡ (i.e., jc¡ = x^). It should also be
remarked that for a completely mixed basin, the instantaneous value of effluent
BOD, JCe, equals that for the hquor in the basin.
5. Neutralization {and Equalization) 141
Hummenick [9] considers the case when the B O D fluctuation of the influent
about the mean value χ-, is given by the function
x¡ = x¡ + kx; sin wt (3.90)
where /: is a constant and w = 2π/Γ, where Tis the period for one cycle of the sine
function. Consequently the peak value of χ-, corresponding to sin wt = 1.0 will be
î,max = î + ^ ^ i = (l +k)x, (3.91)

If, for example, k = 0.7, this means that the peak influent B O D is 70% greater
than the average B O D value x,. F o r the case when the influent B O D is given by
Eq. (3.90), it can be shown [9] that the effluent B O D is approximated by the
relationship:
Xe k úniwt - tan " ^ wtu)
If, for example, Γ = 12 h for the same value of A: = 0.7, the peak value for the
effluent B O D , X e , max» corresponding to sin(w/ - tan~^ wt^,) = 1, will be, for a
hydraulic residence time t^ of 8 h, calculated from Eq. (3.92) as
Xe..a>_ (0.7)(1.0)
X; ^[1+(8)^(2π/12)^]'/^ '
Thus the equalization basin has damped the peak influent concentration from
70% greater than the average jc¡ down to 16.2%.
Reynolds et al. [12] have proposed an iterative procedure which can be
utilized for calculation of values of effluent BODs for a constant-level
completely mixed equalization basin under constant flow conditions. The
equation to be solved is
x.,tÂt = î,r(l - e-^^'^'^') + x,,e-^^'''^^ (3.93)
where
ê,í + j í = effluent B O D after addition of flow for a time interval At, with a
concentration x-^j,
Xit = influent concentration averaged over At, and
Xe,f = effluent concentration before addition of flow at concentration x^î.
The method of calculation [9] is an iterative procedure where the x ^ are

calculated for each selected time increment. If a large number of influent samples
are obtained, it is convenient to use for At the time interval between two
successive composite samples. About 80 samples are the minimum and 150 the
desirable number of samples to characterize a wastewater [ 9 ] .
The variances, which will be denoted as t^j or for the influent and effluent,
respectively, are defined as
Vy = af = ^ ( X i - ^ i Y i ^ - ^ ) (3.94)
and
íê = t 7 , 2 = ¿ ( ^ e - X e ) ' / ( « - l ) (3.95)
Since the mean value for the effluent B O D (x^) is equal t o the mean value for the
influent B O D (jCi), Eq. (3.95) becomes
v^^tix^-xd'jin-l) (3.96)
where
V are the variances (Vi for influent and for effluent),
σ are the standard deviations (î for influent a n d for effluent),
Xi or x^ are the individual values for a discrete sample (in this case, individual
B O D values for influent or effluent streams), a n d
η is the total number of discrete samples.
5.5.2. Calculation of the Standard Deviation σ-, for the BOD

Values for the Influent Stream
The standard deviations a¡ can be determined numerically or graphically;

these two procedures are described a n d illustrated next.
(i) Numerical calculation of σ,. This numerical calculation is performed

from Eq. (3.97), which results directly from the definition of variance v,, given
TABLE 3.15
Values of BODs in mg I liter {4-h compositesf
Time Feb. 10 Feb. 11 Feb. 12 Feb. 13 Feb. 14 Feb.15 Feb.16 Feb.17 Feb. 18
4 a.m. 717 1135 1035 1174 316 505 329 1019 340
8 a.m. 946 241 265 1105 758 221 371 1023 949
12 noon 623 396 419 659 769 957 1081 1167 940
4 p.m. 490 1070 498 801 574 654 621 1056 233
8 p.m. 666 440 961 720 ]1135 510 235 560 1158
12 midnight 828 534 308 454 1142 1067 993 708 407
Time Feb. 19 Feb. 20 Feb. 21I Feb. 22 Feb. 23 1Feb. 24 Feb. 25 Feb.26
4 a.m. 853 356 454 776 619 722 1110 472

8 a.m. 751 847 1080 502 691 368 374 671
12 noon 207 711 440 1146 416 686 494 556
(min)
4 p.m. 852 1185 872 1054 nil 915 268 672
(max)
8 p.m. 318 825 294 888 973 361 1078
12 midnight 358 618 763 266 892 346 481 -
-
" From Ref. [ 1 ] .
by Eq. (3.94):
1/2
(Xi-x-ylin-l) (3.97)
Obviously, prior to the design of the stabilization basin, the individual B O D

values jCe for the effluent are not available and, consequently, Eq. (3.97) cannot
be directly used for the calculation of σ^.
Example 3,10.* A wastewater with a total flow of 5 M G D was characterized

as shown in Table 3.15 and Fig. 3.41. O n e hundred analysis of composite
samples were performed, at 4-h intervals, covering a period from February 10 (at
4 a.m.) to February 26 (at 4 p.m.). The values in Table 3.15 are reported as B O D 5
(mg/hter) and the peak value is 1,185 (February 20, at 4 p.m.). Calculate
numerically the variance ν-, and the standard deviation σ-, for these B O D 5 values.
Solution
Step I. The average value for the 100 samples is (refer to Table 3.15)
ΣΪ ΣΓ^^^ί 717 + 946 - f · · · + 556 + 672

X, = = — -= = 689.75 mg/liter
η 100 100
^ 690 mg/liter
Step 2. The variance ν-, is calculated from Eq. (3.94):
^ ^ Z i ^ ^ ( ^ i - ^ i ) ^ ^ Z r ^ ( ^ i - 690)^
^' 100 - 1 99
where [refer to Table 3.15]

100
X (x, - 690)2 = (717 - 690)2 ^ ( 9 4 5 _ ^g^î + . . . + (556 _ 690)^
+ (672 - 690)2
100
X (x¡ - 690)2 ^ 8,568,514 mg2/liter2
1
Therefore
8,568,514 mg2
Vi = - - = 86,551
99 hter2
Step 3. The standard deviation is then

= i;//2 = (86,551)^/2 ^ 294 mg/liter
(ii) Graphical determination of σ,. T o apply this graphical method the data
are first linearized by plotting them in probability paper as described in Chapter
* See Ref. [ 1 ] .
2, Section 13. Since a large number of data points is utilized, Method 2 described
in Chapter 2 is indicated, the frequency values utihzed for the abscissa being
calculated from Eq. (2.84).
From statistical considerations, it results that the standard deviation σ-, may
be calculated as equal to one-half of the difference of the x, at the 84.1 and 15.9
percentile levels, i.e.,
i X i = I ( ^ i , 8 4 . 1 -Xi,15.9) (3.98)
Example 3 . 1 1 . * F o r the wastewater of Example 3.10 (data in Table 3.15)

determine the variance ν-, and the standard deviation σ·, for the B O D 5 values by
the graphical method.
Solution
Step 1. The probability plot for the data in Table 3.15 has been prepared in
Chapter 2 (refer to Example 2.10) and is presented in Fig. 2.23.
Step 2. From Fig. 2.23 read
•^1,84.1 = 990 mg/liter and . ^ ¡ , 1 95 . = 400 mg/hter
Therefore, from Eq. (3.98)
990 - 400 ^
σ; = - - = 295 mg/hter
and
V, = = (295)^ = 87,025 mgVhter^

It should be remarked that the standard deviation a¡ determined by the
graphical method (σ^ = 295 mg/liter) agrees closely with the value 294 mg/liter
calculated numerically in Example 3.10. Furthermore, the B O D 5 value cor
responding to the 50.0 percentile in Fig. 2.23 is 690 mg/hter, which is in
agreement with the mean value Jc¡ calculated numerically in Example 3.10, Step
1. T h e interval 2a¡ = 9 9 0 - 4 0 0 = 590 is indicated in Fig. 2.23. Notice that the
difference (84.1-15.9) between the 84.1 and 15.9 percentiles is 68.2. Basically this
indicates that B O D values of subsequent samples, collected and analyzed under
identical conditions, can be expected to have a value between 400 a n d 990
mg/liter, 68.2% of the time.
Since prior to the design of the stabilization basin, the individual B O D values
Xe for the effluent are not available, this graphical method just described, like the
numerical procedure based upon Eq. (3.97), cannot be utilized for the
calculation of the standard deviation σ^. For the case of the effluent, the estimate
of the standard deviation can be obtained by the probabihty table method which
is described next.
* See Ref. [ 1 ] .
5.5.3. Estimate of the Standard Deviation for the BOD

Values for the Effluent Stream: Probability Table
Method
For the application of the probability table method one defines the cumulative
standard normal (Y) as the difference between the maximum effluent value and
the mean value, divided by the effluent standard deviation, i.e.,
Y={x.,m..-x.)K (3.99)
Since x^ = x¡, Eq. (3.99) becomes
Υ=(χ.,η...-ΧΟ/σ. (3.100)
where
Y = cumulative standard normal,

•^cmax = peak (maximum) effluent concentration from basin, and
= effluent standard deviation.
The value of X e , m a x can be arrived at from pilot studies and design calculations
for the activated sludge systems, which indicate that the effluent from the
equalization basin must not exceed a certain value [ X e . m a x ( B O D in mg/hter)] in
order that the activated sludge system can meet the effluent quality criteria of a
specified maximum B O D concentrafion.
Equation (3.100) can be solved for the effluent standard deviation, yielding
σ. = {χ.,η...-ΧΟ/Υ (3.101)
The cumulative standard normal Y is related to the confidence level P{x) for the
case of a normal distribution by the equation
P{x)= * ^^vi^dY (3.102)

In J - 00
A confidence level P{x) of 95%, for example, [i.e., P{x) = 0.95] means that
there is a 95% probability that the equalized effluent concentration will be equal
to or less than the specified X e , m a x value. Probability tables have been developed
which relate the confidence level P{x) to the value of F (Table 3.16). F r o m this
table for a specified value of the confidence level P{x), one can read the value of
the cumulative standard normal Y. The value of the confidence level P{x) is
specified by the designer: the larger the value specified, the more conservative
the design will be, i.e., the volume of the stabilization basin will be larger. The
usual range of P{x) is between 0.9 and 0.99.
In summary, the procedure for determination of the standard deviation and

variance for the effluent of the equalization basin can be outlined as follows.
Step 1. Specify a value of X e , m a x from pilot plant studies or calculations for the
activated sludge system. This is the maximum permissible effluent concentration
from the equalization basin that still results in the activated sludge system
TABLE 3.16"
Cumulative Normal Distribution
Y Pixf Y Pixf Y Ρ(χγ
0.00 0.5000 1.30 0.9032 2.55 0.9946

0.05 0.5199 1.35 0.9115 2.60 0.9953
0.10 0.5398 1.40 0.9192 2.65 0.9960
0.15 0.5596 1.45 0.9265 2.70 0.9965
0.20 0.5793 1.50 0.9332 2.75 0.9970
0.25 0.5987
- - - -
0.30 0.6179 1.55 0.9394 2.80 0.9974
0.35 0.6368 1.60 0.9452 2.85 0.9978
0.40 0.6554 1.65 0.9505 2.90 0.9981
0.45 0.6736 1.70 0.9554 2.95 0.9984
0.50 0.6915 1.75 0.9599 3.00 0.9987
0.55 0.7088 1.80 0.9641 3.05 0.9989

0.60 0.7257 1.85 0.9678 3.10 0.9990
0.65 0.7422 1.90 0.9713 3.15 0.9992
0.70 0.7580 1.95 0.9744 3.20 0.9993
0.75 0.7734 2.00 0.9772 3.25 0.9994
0.80 0.7881 2.05 0.9798 3.35 0.9996

0.85 0.8023 2.10 0.9821 3.45 0.9997
0.90 0.8159 2.15 0.9842 3.55 0.9998
0.95 0.8289 2.20 0.9861 3.75 0.9999
1.00 0.8413 2.25 0.9878 4.00 1.0000
1.05 0.8531 2.30 0.9893 - -

1.10 0.8643 2.35 0.9906
0.8749 2.40 0.9918
- -
1.15
0.9929
- -
1.20 0.8849 2.45
0.9938
- -
1.25 0.8944 2.50
- -
From Ref. [ 1 ] .
P(x) = confidence level.
meeting the effluent quality criteria of a specified maximum B O D concentration

for its effluent.
Step 2. Determine the value of x¡ for the equalization basin either by
numerical calculation (mean of x¡) or by the graphical method (ordinate of Fig.
2.23, corresponding to the 50% abscissa). Notice that the mean jc¡ is the same for
the influent or effluent of the equalization basin (i.e., X j = X e ) .
Step 3. Specify a value of the confidence level P(x), usually between 0.9 and
0.99.
Step 4. For the specified value of P(x), read from Table 3.16 the value of the
cumulative standard normal Y.
Step 5. Calculate the standard deviation from Eq. (3.101).
Step 6. Calculate the effluent variance: = σ].
Example 3.12.* Take the same wastewater flow of 5 M G D that was

considered in Examples 3.10 and 3.11. Pilot studies and design calculations with
activated sludge systems have indicated that the effluent from the equalization
basin must not exceed 896 mg/liter in order to meet the effluent quality criteria of
an average BOD5 of 15 mg/liter and a maximum concentration of 25 mg/liter
from the activated sludge system.
Calculate the standard deviation a n d the variance for the BOD5 con-
centration of the effluent by the probabihty table method. Base calculations on a
95% probability that the BOD5 of the equalized effluent will be equal or less than
896 mg/liter.
Solution Follow the six calculation steps just summarized.
Step L X e . m a x = 896 mg/hter.

Step 2. JCe = x¡ = 690 mg/liter (Example 3.10, Step 1).
Step 3. Select Ρ(χ) = 0.95 (95% probability).
Step 4. F o r P{x) = 0.95, read from Table 3.16
r % 1.65
Step 5. Compute σ,, Eq. (3.101):
Xe,max-î 896 - 690

= - - = — — — = 125 mg/hter
Step 6. T h e effluent variance is then
ι;^ = σ2 = (125)2 = 15,625 mg^/liter^
5.5.4. Calculation of Residence Time for a Constant-Level

BOD Equalization Basin
Under constant flow conditions, for a constant-level, completely mixed

equalization basin, Danckwerts and Sellers [3] have shown that the ratio of the
effluent variance (v^) to the influent variance (Vi) can be related to the hydraulic
detention time (ί^) by the relationship
e-'i'^R{r)drlty, (3.103)
where r is the lag time and R(r) is autocorrelation coefficient for the influent
data, which is defined by
R(r) = (x; - xOtiXi - xdt^M (3.104)
where the numerator is the average value of the product of the deviations from
the average concentration separated by the time lag r. Equation (3.103) can be
evaluated graphically or analytically after R(r) is found. Hummenick [9] has
* See Ref. [ 1 ] .
developed a method for this calculation and presents computer programs for
this purpose.
Adams and Eckenfelder [ 1 ] utilize a simplified version of Eq. (3.103) to relate
the ratio of the effluent variance {v^) to the influent variance (i;¡) and the
hydraulic residence time {t^). This relationship, which is utilized in Example
3.13, is as follows:
iêM = σ Χ = (zl//2)//h (3.105)
where At can be conveniently taken as the time interval between two successive
composite samples when a large number of samples is collected during the period
of study. Solving Eq. (3.105) for the hydraulic residence time ty,
At Vi _ At
^h = - - - = - - - i (3.106)
Example 3.13.* F o r the wastewater in Example 3.12, calculate the hydraulic

residence time t^, and the horizontal cross-sectional area of the stabilization
basin (in acres) assuming a depth of 12 ft. Base calculations on a 9 5 % probability
that the B O D 5 of the equalized effluent will be equal or less than 896 mg/liter.
So/ution
Step 1. Calculate ty, from Eq. (3.106):
At V, ( 4 h ) 87,025
ih = - - - = ^ - T ^ = l l . l h (0.464day)
2 2 15,625
where At was taken as 4 h (sampling interval). The value of ν-, utilized was that
obtained by the graphical method (Example 3.11) and that for v^ was obtained in
Example 3.12 by the probabihty table method.
Step 2. The volume of the basin is
V=Qt^ = 5 Mgal/day χ 0.464 day = 2.315 Mgal
or
2,315,000 gal χ = 309,492 ft^
7.48 gal
Step 3. The horizontal cross-sectional area of the basin is then
or A = 309,492 ftVl2 ft = 25,791 ft^
acre
* See Ref. [ 1 ] .
5.6. Direct p H Control M e t h o d s : N e u t r a l i z a t i o n of

Acidic W a s t e s by Direct pH Control M e t h o d s
The foUowing methods of direct neutrahzation of acidic wastes are the most
commonly employed: (1) limestone beds, (2) slurried hme neutralization, (3)
caustic soda ( N a O H ) neutralization, (4) sodium carbonate neutralization, a n d
(5) ammonia neutralization.
A few specific comments about these methods are as follows:
Slurried hme neutralization is the most common method and is discussed in
Section 5.8. Design of hmestone beds is described in Section 5.7.
Caustic soda ( N a O H ) is more expensive than hme. It offers an advantage with
respect to uniformity of the reagent, ease of storage and feeding, rapid reaction
rate, and the fact that the end products of neutralization (sodium salts) are
soluble.
Sodium carbonate ( N a 2 C 0 3 ) is not as reactive as caustic soda and presents
frothing problems owing to release of carbon dioxide.
A m m o n i a (NH4OH) presents the disadvantage of being a contaminant;
consequently, its use may be ruled out by pollution control standards.
Factors guiding selection of a neutralization reagent are (1) purchase cost, (2)
neutralization capacity, (3) reaction rate, and (4) storage and disposal of
neutralization products.
5.7. Limestone Beds

57.1. Types of Equipment
Both upflow and downflow types of limestone beds are employed. F o r

wastewaters containing H2SO4, hmestone beds should not be used if con
centration of H2SO4 exceeds 0.6%. T h e reason for this hmitation is that the
limestone becomes covered with an insoluble coat of C a S 0 4 , rendering it
ineffective. In addition, evolution of CO2 causes frothing problems.
Influent
Recycle (acid stream)
Effluent Lirn^stone, Feed reservoir

:*iö;bedÄ:i
Sedimen-
tot ion F=_
basin _
Recycle pump
Fig. 3.42. Typical upflow-type limestone bed neutralization unit.

150 3, Pretreatment and Primary Treatment
Upflow type arrangement is preferable to downflow type since in upflow

units, CaS04 tends to be flushed out before precipitation on the hmestone. Also,
escape of CO2 generated by the neutralization reaction is easier in upflow type
units. In the downflow type units since the CO2 escapes in opposite direction to
the hquid flow, flooding conditions are more hkely to occur. For these reasons,
maximum hydraulic rate for downflow systems is hmited to approximately 50
gal/h ft^. Recychng of the effluent can be utilized. Figure 3.42 presents a diagram
for an upflow system for which the upflow movement of the liquid stream is
provided without pumping.
Presence of metallic ions (e.g., AP"^, Fe^^) in the wastewater reduces
effectiveness of the limestone bed owing to coating of hmestone with pre
cipitated hydroxides. Finally, if dilution of the acid in the wastewater is
increased, higher residence times are required for neutralization.
5.7.2. Design Procedure for Limestone Beds
In this section, the laboratory procedure recommended by Eckenfelder and

Ford [6] for obtaining the basic design data is described. A numerical example is
presented to illustrate design of an actual hmestone column. A model of a
Fig. 3.43. Laboratory model of limestone neutralization column.
laboratory limestone neutralization column is shown in Fig. 3.43. Benchscale

columns operate with heights of limestone of 1.5-5 ft, which is the actual range of
heights for plant-scale units. Column diameter is approximately 6 in., and rates
of flow (gal/h ft^) are comparable to those for plant operation.
Step I. Fill neutralization columns with hmestone (after washing and
screening) to depths of 1, 2, 3, 4, and 5 ft (5 columns).
Step 2, Adjust upward flow rate of acid wastewater for each column. Flow
rates varying from 50 to 1000 gal/h ft^ are used.
Step 3. Check effluent p H from each column at each flow rate utilized until it
is stabilized.
Step 4. After each run replace limestone used in the columns.
Step 5. Plot terminal p H as a function of rate of flow (gal/h ft^) for each depth
of Hmestone. A typical plot of this type is shown in Fig, 3.44.
500 1000
Flow rate (gal/ti ft^)
Fig. 3.44. Limestone neutralization data. Numbers in circles are the heights of packing (ft).
Purposes of the design procedure are (1) to select the most economical height
of column for a specified p H of the effluent. [This is the column height
corresponding to a maximum allowable flow rate, expressed in terms of volume
of limestone utilized, i.e., gal of Uquor/(h)(ft^ of limestone)]; and (2) to calculate
annual requirement of hmestone under these conditions, which corresponds to a
minimum requirement of hmestone.
Example 3.14. It is desired to neutralize a wastewater acid stream containing

ύ.ΧΝ HCl to a p H of 7.0 through a limestone bed. Laboratory tests with the
wastewater yield data plotted in Fig. 3.43. Design a neutralization system for 100
gal/min (6000 gal/h) of wastewater and estimate annual hmestone requirement
for the most economical operation.
TABLE 3.17
Calculations for Example 3.14
(1) (2)" (3) (4) (5)

Depth Flow rate Cross section Volume of Flow rate
(ft) (gal/h ft^) limestone (ft^) (gal/h ft^)
(ft')
(3) = (6000 gal/h)/(2) (4) = (1) X (3) (5) = (6000 gal/h)/(4)
1.0 118 51 51 118

2.0 492 12.2 24.4 246
3.0 845 7.1 21.3 282
4.0 1047 5.73 22.9 262
5.0 1200 5.0 25.0 240
" From Fig. 3.44 for pH = 7.

So/ution
Step 7. For pH = 7 read from Fig. 3.44 rates of flow (gal/h ft^) corresponding
to each column depth. Then calculate cross-sectional area, volume of hmestone
bed required, and flow rate in gal/(h)(ft^ of limestone bed) (Table 3.17).
Step 2. Plot flow rates (gal/h ft^) from column (5) of Table 3.17 versus depths
[column (1)]. This plot is shown in Fig. 3.45 and indicates that a 3-ft limestone
bed is the most economical, corresponding to a maximum on the curve. The
cross-sectional area in this case (see Table 3.17 for a depth of 3.0 ft) is 7.1 ft^,
corresponding to a diameter of 3.0 ft.
300
8
/
/ .2 6< )
Ν
I 250
/ >) Ν
)
to
o /
f
/
S200
/ /
//
I '50 1
I
< L
Kl Iii
100
I 2 3 4 5 6
Depth (ft)
Fig. 3.45. Determination of optimum bed depth. Depth (column 1, Table 3.17) versus flow rate
(column 5, Table 3.17).
Step 3. Calculate the pounds of acid to be neutralized per day (acid contained
in the 6000 gal/h of the 0.1Ν solution of HCl). Since this is a dilute acid solution,
calculation is based on the density of water, taken as 8.34 lb/gal. A 0.17V solution
of HCl contains 3.65 g/hter of HCl, or approximately 3.65 lb of acid per 1000 lb
of solution. Therefore,
Mass now rate: 6000 gal/h χ 24 h/day χ 8.34 lb/gal = 1.2 Mlb/day
Acid content: 3.65 χ (1,200,000/1000) = 4380 lb/day
Step 4. Calculate limestone (CaCOa) required. The neutralization equation is
2HC1 + CaCOs CaCl2 + CO2 + H2O

molecular weights: 2x36.5 100
= 73
Therefore hmestone required is
4380 X 100/73 = 6000 lb/day of hmestone

This is the theoretical a m o u n t of limestone assuming 100% reactivity. For design

purposes, assume an 80% reactivity. Limestone required is then
6000/0.8 = 7500 lb/day or 2.738 χ 10^ Ib/yr
5.8. Slurried Lime T r e a t m e n t

5,8.1. Equipment for Slurried Lime Systems
Slurried hme is the most commonly used reagent for neutralization of acid
wastewaters, the low cost of hme being the main reason for its widespread
utilization. A flowsheet of a two-stage slurried hme neutralization system is
shown in Fig. 3.46.
Quicklime Water
R e c i r c u l a t i o n line Slurry
for slurried lime storage
tank
Slurried lime ^ (agitated
vessel)
Recirculation pump''
pH controller^
Neutralized
effluent
lagiTOted
vessel)
B y - p a s s line
*^pH controller
Alkaline wastewater
Fig. 3.46. Flow diagram of a two-stage slurried lime neutralization system.
Stepwise addition of lime is recommended. For highly acidic wastes, a

minimum of two stages is desirable, the first (bulk neutralization) to raise p H to
a value of 3.0-3.5, and the second (fine tuning) to adjust p H to desired effluent
value. Sometimes a third stage is desirable.
Automatic control of this process is not simple because the relationship
between p H and amount of hme added is highly nonhnear, particularly in the
vicinity of the neutralization point (pH = 7). This is appreciated by inspection
of a typical neutralization curve of an industrial wastewater, as shown in Fig.
3.47.
In the vicinity of the neutralization point, the p H becomes exceedingly
sensitive to small addifions of lime, varying in an ordinary operation of slurried
lime systems at a rate as fast as one p H unit per minute. Also fluctuation in flow
rate of influent complicates operation of the process. Use of an equalization tank
is indicated to dampen fluctuations, as shown in Fig. 3.46. A relatively
Í
14
/
^—
12
10
8
7 *
6
1
1
4
1
yι
y\
2 ^ I
ι -3.600
— ι
Ο 2.000 4.000 6.000
mg of lime/liter of wastewater
Fig. 3.47. Typical neutralization curve for an industrial wastewater.
small amount of reagent is thoroughly mixed with a large hquid volume in a

short time interval. Mechanical mixers are provided for this purpose.
5.5.2. Design Procedure for Slurried Lime Neutralization

Systems
The procedure recommended by Eckenfelder and Ford [6] is summarized in

this section and illustrated by a numerical example. Basic information required
is (1) neutralization curve for the wastewater (see Fig. 3.47) and (2) power
consumption data, i.e., a curve of level of agitation versus detention time for a
desired terminal p H (see Fig. 3.48). This basic information is obtained by simple
laboratory procedures [ 6 ] .
Purposes of the design procedure illustrated by Example 3.15 are (1) to select
the number of stages of neutralization and to size neutralization reactors, and (2)
to select appropriate mixing equipment.
Example 3.15. 200 gal/min of an acidic industrial wastewater are neutralized

to pH = 7.0. The neutralization curve is plotted from laboratory tests (Fig.
3.47). Also a curve of level of agitation versus detention time is obtained for this
neutralization (Fig. 3.48). Design a Ume slurry neutralization system.
Solution
Step 1. F r o m the neutralization curve in Fig. 3.47, hme slurry requirement for
neutralizing the wastewater to a p H of 7.0 is read as 3600 mg/liter.
Step 2. Calculate hme slurry requirement for 200 gal/min wastewater flow.
Since
3600 mg lime/liter wastewater = 3600 χ 10"^ lb lime/lb wastewater
1.0
--
\
\
s
—
%0.1
O.Oll
5 10 50 100
Residence time (min)
Fig. 3.48. Level of agitation versus detention time (for pH = 7.0).
Then
200 gal wastewater/min χ 60 min/h χ 24 h/day
X 8.34 lb wastewater/gal wastewater
X 3600 X 1 0 " M b hme/lb wastewater = 8647 lb hme/day
Step 3. Neutralize acid in two steps as indicated by flow diagram in Fig. 3.46,
the first stage for bulk neutralization, and the second for fine tuning.
Step 4. Select a detention time and size the reactors
Volume of reactor (gal) = Q (gal/min) χ / (min) (3.107)
Selection of the optimum detention time is arrived at by an economical balance.
1. Assume a residence time (usually a value between 5 and 10 min).

2. Size the reactor [Eq. (3.107)].
3. F r o m Fig. 3.48 for the assumed residence time, determine power level
required.
4. Select mixers (as shown in Step 5) from knowledge of power level
[item (3)].
5. Estimate total costs (capital and operating) corresponding to this assumed
residence time. Main items in the column of capital costs are the reactors
themselves (and auxiliary equipment) and the mixers. Energy requirement is the
variable item among operating costs.
6. Repeating Steps (l)-(5), a curve of total cost per year versus a series of
selected residence times is plotted. This curve passes through a minimum which
corresponds to the optimum detention time. Capital costs are expressed on a
yearly basis by estimating equipment life and utilizing the current value for
interest rate.
Assume for Example 3.15 that optimum residence time is estimated by this
procedure as 5 min. Then
Volume of each reactor (gal) = 200 gal/min χ 5 min = 1000 gal
or
1000 gal χ ftV7.48 gal = 134 ft^
Selecting a reactor depth of 5 ft, required cross-sectional area is
A r e a = 134 ftV5 ft = 26.8 ft^

corresponding to a diameter of 5.84 ft.
Step 5. Select mixers. F r o m Fig. 3.48, power level required for 5-min
detention time is
0.15 hp/1000 gal
Since each tank has a volume of 1000 gal, specify one 0.15-hp mixer for each
tank.
5.9. N e u t r a l i z a t i o n of Alkaline W a s t e s
In principle, any strong acid can be used to neutralize alkaline wastewaters.
Cost considerations limit the choices to H 2 S O 4 (the most common) and HCl.
Reaction rates are essentially instantaneous. The basic design procedure for
alkaline wastes is similar to that for acidic wastes described in Section 5.8.
Flue gases containing 14% or more of CO2 are used for neutralization of
alkahne wastewaters. When bubbled through the wastewater, the CO2 forms
carbonic acid, which reacts with the base. Reaction rate is slow but sufficient if
p H need not be adjusted below 7 or 8. Either bubbling through a perforated pipe
or using spray towers is satisfactory.
Owing to production of CO2, the activated sludge process (and other aerobic
biological treatment processes for wastewaters) have a hmited capacity for
neutralizing alkaline wastewaters. This matter is discussed in Chapter 5, Section
7.7.
5.10. Buffering Effect for W a s t e w a t e r s

A buffered solution is one which resists change of its hydrogen or hydroxyl ion
concentrations in the event an acid or a base is added to it. M a n y wastewaters
contain bicarbonate ion in solution, resulting from dissolution of carbon dioxide
in the water. The bicarbonate ion reacts either with an acid or a base according to
the following equations:
With an acid:
- H 2 C O
3 (3.108)
With a base:
HCO3- + OH - ^ H2O + CO^ (3.109)
and as a result a buffering effect does occur for such wastewaters.

In the case of Eq. (3.108), the H C O 3 acts as a proton acceptor (i.e., a base). In
the case of Eq. (3.109) the H C O J acts as a proton donor (i.e., an acid). The
carbonate equilibrium in water can be represented by the following chemical
equations:
CO2 + HgO 1 H2CO3 HC03"
C0|"+ H*
Consider separately the two chemical equations:

H ^ C O j ^ H C O j - + H+ (3.110)
HCO-^CO^ -I-H+ (3.111)
For these two chemical equations the thermodynamic equilibrium constants

may be written as follows:
For Eq. (3.110)

[HC03-][Hn
(3.112)
[H2CO3]
For Eq. (3.111)
[CO^-][H*]
K2 = (3.113)
[HCO3-]
where the brackets denote concentrations in mole/liter. F r o m thermodynamic

considerations the equilibrium constants Κχ and K2 can be expressed as function
of the temperature T a s follows [ 1 0 ] :
logA:, = - 1 7 , 0 5 2 / r - 2 1 5 . 2 1 log Γ + 0 . 1 2 6 7 5 Γ + 545.56 (3.114)

\ogK2 = - 2 9 0 2 . 3 9 / Γ - 0.02379Γ + 6.498 (3.115)
where T i s in °K. At 25°C (298.16°K), the values of these equilibrium constants

are = 4 . 3 1 3 χ 10"^ and ^2 = 4.682 χ l O ' ^ ' .
The pH of a carbonate buffer can be calculated from the basic definition:
pH = l o g p l ^ (3.116)
Substituting in Eq. (3.116), [ H ^ ] by its value obtained from Eq. (3.112), i.e.,
one obtains
Letting \og(\/Ki) = pKi
pH=pK. + \os^^\ (3.119)
For a given temperature (i.e., a fixed Ki) and a given p H , the ratio
[ H C 0 3 " ] / [ H 2 C 0 3 ] can be calculated from Eq. (3.119) for the case of a
wastewater for which the buffering is due to the carbonate system.
For most carbonate-buffered wastewaters, the p H is between 6.5 and 8.5. It is
important to be able to calculate the change in this p H resulting from the
additions of an acidic (or basic) solution. This is the problem which arises when
an acidic (or basic) wastewater from an industrial plant is discharged into a
municipal sewer. Since the municipal sewage contains a buffered carbonate
solution, its p H is much less sensitive to the addition of the acid (or base) than it
would be the case for a nonbuffered solution. A quantity jS, denoted as buffer
intensity, is defined as the number of moles of H"^ (or O H " ) required to change
the p H of one liter of a solution by one p H unit. Inspection of the typical
neutralization curve (Fig. 3.47) reveals that jS is a function of the p H . Indeed for a
given addition of H"^ (or O H ~ ) , a much larger p H change occurs in the
neighborhood of the neutralization point (pH ^ 7.0). Therefore, β is generally
written as a differential, i.e.,
β = - dA/d(pH) = dB/d(pH) (3.120)
where
β = buffer intensity;
dA/dipH) = mole of acid (i.e., mole H"^) required to produce a change of
¿(pH) (negative sign indicates that p H is lowered upon acid
addition); and
dB/d{pH) = mole of base (i.e., mole O H " ) required to produce a change of
d(pH) (positive sign indicates that p H is increased upon
addition of a base).
F o r the case of the addition of an acid (or a base) to a carbonate-buffered

sewer, since there is a considerable dampening of the p H change, Eq. (3.120) may
be written utilizing finite differences instead of differentials, i.e.,
β = - AA/A(pH) = AB/A(pH) (3.121)
or
ß = - zl[H + ]/zl(pH) = z l [ O H - ] / z l ( p H ) (3.122)
F o r the case of addition of a strong acid to a carbonate-buffered solution,

Weber and Stumm [ 1 5 ] found that β can be calculated from
'a([Alk.] - [ O H - ] + [ Η η ) ( [ Η Ί + ^ + ^/T^)
• = 2.3 < , ^—^ - + [ H ^ + [OH-] > (3.123)
where
(3.124)
In Eqs. ( 3 . 1 2 3 ) and ( 3 . 1 2 4 ) we note the following:
β = buffer intensity (equivalents/unit p H ) ;

= equilibrium constants defined by Eqs. ( 3 . 1 1 2 ) and ( 3 . 1 1 3 ) .
/:I,A:2
Temperature effect is given by Eqs. ( 3 . 1 1 4 ) and ( 3 . 1 1 5 ) ; a n d
[H"^] = hydrogen ion concentration for the carbonate-buffered solution,
prior to the acid addition, calculated from Eq. ( 3 . 1 1 6 ) as
[ H ^ ] = l0-''» (3.125)
[ O H " ] = hydroxyl ion concentration for the carbonate-buffered solution

prior to the acid addition. Since for an aqueous solution
[ H ^ ] [ O H - ] = K^ = 10-^^ (3.126)
then
1Q-14 ,Q-14
[OH-] = i ^ = i % (3.127)
[Alk.] = stands for total alkahnity (equivalents/hter), which is defined as the

acid-neutralizing capacity of a water sample to p H = 4 . 3 . Usually [Alk.] is
specified in terms of mg/liter of CaCOa. Since the molecular weight of CaCOa is
1 0 0 g and its equivalent weight is 1 0 0 / 2 = 5 0 g, then [Alk.] is
g/liter of CaCOa
[Alk.] - - ( 3 . 1 2 8 )
The buffering effect for a carbonate-buffered wastewater can be visualized by

calculating the change in p H resulting from the addition of an acidic waste t o a
nonbuffered solution and comparing it for the case when the acidic waste is
added to a carbonate-buffered sewage. These calculations are illustrated in
Example 3 . 1 6 .
Example 3.16. An industrial plant discharges an acidic solution of p H = 3.0

at a rate of 0.1 M G D into a municipal sewage system. F o r the municipal sewage
the flow rate is 1.0 M G D and the p H = 8. Take the temperature as 20°C for both
the sewage and the acidic waste. The alkalinity of the sewage is 250 mg/liter as
CaCOa. (See Fig. 3.49 for diagram of flow.)
1. Assuming no buffering effect, calculate the change of the p H for the sewage
after addition of the acidic stream.
2. Assuming that the municipal sewage is carbonate-buffered, calculate the
change of its p H after addition of the acidic stream.
Industrial plant
Acidic waste
0.1 MGD
pH=3
Municipal
Municipal sewage _ , Combined influent wastewater Final
pH=? treatment effluent
1.0 MGD plant
pH=8
Fig. 3.49. Diagram for Example 3.16.
Solution to Part 1 Calculation without buffering effect.
For acidic stream
pH = 3, .*.[H + ] = 1 0 - P " = 10-^ mole/liter
F o r the sewage
p H = 8, . ·. [ H + ] = 1 0 - P " = 10-« mole/liter

F o r the combined influent
(1.0)(10-«) + (0.1)(10-^)
[Hn = - = 9.092 X 10-^ mole/liter
1+0.1
Therefore we set the combined influent:
= 4.04
Therefore the p H would d r o p from 8 to 4.04, or

zl(pH) = 8.0 - 4.04 = 3.96 units
Solution to Part 2 Calculation considering the carbonate buffering.
Step 1. F r o m Eqs. (3.114) and (3.115) calculate and K2 at 20°C (293.16°K):
5. Neutralization (and Equalization) 1 61
Eq. (3.114):
log/:, = - - 215.21 log293.16 + (0.12675)(293.16) + 545.56
log a:, = - 6.3937, . ' . / : , = 4.039 χ 1 0 " '
Eq. (3.115):
2902.39
\o%K2 = - - (0.02379)(293.16) + 6.498
\ogK2 = - 10.3766, .'.ATz = 4.201 χ l O " ' '
Step 2. F o r the sewage stream:
Eq. (3.125):
[H^] = 10-P"= 10-» mole/liter
Eq. (3.127):
[OH-] = ^ = = 10-·^ mole/liter
Eq. (3.128):
[Alk.] = 0.25/50 = 5 X 10"^ equivalents/liter
Step 3. Calculate a, from Eq. (3.124):
4.039 X 1 0 " "

α = = 0.9719
(4.039 X 1 0 " ' ) + (10-») + (4.039 χ 10-^)(4.201 χ 1 0 - " ) / 1 0 -
Step 4. Calculate β, from Eq. (3.123):
(0.9719)[5x 10-«]|lO-«+ ^ + (4)(4.201 χ l O - ^ ^ j
"1 (4.039 X 1 0 - ' ) 1 +

(2)(4.201 xlO"^^)
10"
+ 10-« + 10-
β = 3.278 X 10"^ equivalents/liter
Step 5. Express β in equivalents/day, considering the flow rate of 1.0 M G D .

, equivalents ^ gal hter
= 3.278 X 1 0 " ^ - i ^ X 1.0 X 1 0 ^ ^ x 3 . 7 8 —-
liter day gal
β= 1239.1 equivalents/day
Step 6. Express addition of [H"^] from the industrial plant in equivalents/

day, considering the flow rate of 0.1 M G D .
^ mole , equivalents
[H^] = 1 0 - P " = 10"^ =10"^ —
hter liter
and
10-3 0.1 x i p e j ^ x 3 . 7 8 * ^ ^ 378 ^5^^^^

liter day gal day
Step 7. F r o m Eq. (3.122):
. ^ _ 378 equivaler^ts/day ^ _ ^
1239.1 equivalents/day
Therefore, the p H of the combined influent will be
8.0 - 0.305 = 7.695
Notice that A(pH) calculated without consideration of the buffering effect was
3.96 units, whereas considering the carbonate buffering, it is only 0.305 units.
The ratio is 3.95/0.305 % 13.
PROBLEMS
I. Design a grit chamber for capture of sand for which the particle size
distribution is as foUows:
Percentage by weight d (mm)
50 1.0
20 0.5
20 0.2
10 0.1
Take operating temperature as 20°C and specific gravity of the sand particles as
2.65. Take β = 0 . 0 4 , / = 0.03, and g = 980 cm/sec^ Assume a depth of 6 ft for
the grit chamber. Flow is 20 M G D . Design the horizontal cross secfion for 100%
removal of the 0.5-mm-diam particles. Design the vertical cross section so that
only the particles of d=OA m m are scoured away.
1. W h a t is the percentage of grit removal obtained?

2. Prepare a sketch for the grit chamber indicating all pertinent dimensions. If
necessary, divide the chamber into parallel grit channels of widths not exceeding
4 ft each.
3. Calculate the detention time required (min).
4. If the wastewater influent contains 50 mg/liter of sand, assuming that the
concentrated slurry at the bottom of the grit chamber reaches a concentration of
Problems 163
5 % per weight of sand, estimate the accumulation of this concentrated slurry in

gal/min,
II. Sedimentation {discrete settling). A particle size distribution is ob
tained from a sieve analysis of sand particles. For each weight fraction, an
average setthng velocity is calculated. D a t a [11] are presented in the following
tabulation.
Settling velocity Weight fraction

(ft/min) remaining
10.0 0.55
5.0 0.46
2.0 0.35
1.0 0.21
0.75 0.11
0.50 0.03
1. Prepare a plot of fraction of particles with less than stated velocity versus
settling velocity (ft/min).
2. For an overflow rate of 100,000 gal/day ft-^, calculate overall removal
utilizing Eq. (3.39).
III. Sedimentation {flocculent settling). A laboratory settling analysis gave
the results tabulated below.
Suspended solids removed

at indicated depth (%)
Time (min) 2 ft 4 ft 6 ft
10 40 25 16
20 54 37 28
30 62 47 37
45 71 56 46
60 76 65 53
D a t a for the percentage of solids of compacted sludge versus settling time were
also taken from a tap 8 ft below the surface of the hquid in the settling cylinder.
These data were as follows.
Settling time t Percentage of solids

(min) in sludge
10 0.40
20 0.75
30 0.97
40 1.17
50 1.34
60 1.48
70 1.60
80 1.69
90 1.75
1. Perform analysis of the data and arrive at curves for SS removal (%) versus
detention time (min), and SS removal (%) versus overflow rate (gal/day ft^).
2. If the initial concentration of the slurry is 430 ppm, design a settling tank
(i.e., calculate diameter and effective depth of the tank) to remove 70% of the
suspended solids for a I Mgal/day flow. Utilize a scale-up factor of 1.25.
3. What removal is attained with the settling tank designed under part (2) if
flow is increased to 2 Mgal/day?
4. For the flow of 1 Mgal/day, i.e., conditions in part (2), calculate daily
accumulation of sludge in lb/day and average pumping rate in gal/min.
IV. Sedimentation (zone settling). It is desired to design a secondary
settling tank to produce an underflow concentration of 8000 mg/hter from a
mixed liquor solids content of 2500 mg/liter in the influent. Wastewater flow is
4350 m^/day. The zone settling velocities determined by batch settling tests are
as follows.
Solids concentration Zone settling velocity

(mg/liter) (m/h)
500 6.67
1,000 4.63
1,500 3.11
2,000 2.23
2,500 1.42
3,000 1.00
3,500 0.738
4,000 0.542
4,500 0.408
5,000 0.317
5,500 0.243
6,000 0.188
6,500 0.148
7,000 0.119
7,500 0.100
8,000 0.083
1. Plot the batch flux curve GB (kg/m^ day) versus (mg/liter).

2. Determine the cross-sectional area required for thickening (m^) and the
diameter (m).
3. Check the cross-sectional area required for clarification (m^). Neglect the
sohds concentration in the overflow.
4. Estimate the clarifier height (m), based upon a detention time of 2.0 h.
V. Flotation. A pilot-plant flotation operation indicated optimum
air/solid ratio to be 0.04 lb air/lb of sohds.
1. If a wastewater to be treated has 250 ppm suspended solids, compute the
recycle percentage to be pressurized to 60 psia at 20°C. T a k e / = 0.68.
2. For a wastewater flow of 1.0 Mgal/day and an overflow rate of 4.0 gal/min
ft^ compute surface area required.
References 165
VI. Neutralization. For the 3-ft hmestone bed designed in Example 3.14
prepare a plot of volume of hmestone required versus a range of selected values
of p H for the effluent (select p H = 5 , 6 , 7 , 8 , 9 , and 10).
VII. An industrial plant project involves a proposed discharge of an acidic

wastewater of pH = 2.0 at a temperature of 35°C to a river for which the critical
now is estimated at 20 M G D and the critical temperature is 15°C. The alkalinity
of the river water is 200 mg/liter as CaCOa and the p H is 7.5. Assume that there is
a buffering effect due to the carbonate system. Calculate the maximum
allowable discharge of the acidic wastewater in M G D so that the p H of the river
water downstream of the sewage outfall will never be below a value of 6.5.
REFERENCES
1. Adams, C. E., Jr., and Eckenfelder, W, W., Jr., "Process Design Techniques for Industrial Waste
Treatment." En viro Press, Nashville and Austin, 1974.
2. Camp, T. R., Trans. Am. Soc. Civ. Eng. I l l , 909 (1946).
3. Danckwerts, P. E., and Sellers, E. S., Ind. Chem. 11, 395 (1951).
4. Dick, R. I., J. Sanit. Eng. Div., Am. Soc. Civ. Eng. 96, 423 (1970).
5. Eckenfelder, W. W., Jr., "Industrial Water Pollution Control." McGraw-Hill, N e w York, 1966.
6. Eckenfelder, W. W., Jr., and Ford, D. L., "Water Pollution Control." Pemberton Press, Austin
and New York, 1970.
7. Eckenfelder, W. W., Jr., and O'Connor, D . J., "Biological Waste Treatment." Pergamon,
Oxford, 1961.
8. Hazen, Α., Trans. Am. Soc. Civ. Eng. 53, 45 (1904).
9. Hummenick, M. J., Jr., "Water and Wastewater Treatment." Dekker, N e w York, 1977.
10. Loewenthal, R. E., and Marais, G. V. R., "Carbonate Chemistry of Aquatic Systems: Theory
and Application." Ann Arbor Sei. Publ., Ann Arbor, Michigan, 1976.
11. Metcalf & Eddy, Inc., "Wastewater Engineering: Collection, Treatment, Disposal." McGraw-
Hill, New York, 1972.
12. Reynolds, E., et al.. Trans. Inst. Chem. Eng. 42, T-13 (1964).
13. Talmadge, W. P., and Fitch, E. B., Ind. Eng. Chem. 47, 38 (1955).
14. Waddel, H., J. Franklin Inst. 21, 459-490 (1934).
15. Weber, W. J., Jr., and Stumm, W., J. Am. Water Works Assoc. 55, 1553 (1963).
16. Yoshioka, N., et al., Chem. Eng. (Tokyo) 21, 66 (1957).

Pretreatment and Primary Treatment

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Pretreatment and Primary Treatment

Uploaded by

Copyright:

Available Formats

Chapter 3

Pretreatment and Primary Treatment

Selection of a wastewater treatment process or sequence of processes depends

Screening is employed for removal of suspended solids of various sizes. Screen

is removed by sedimentation. For this reason, and also because clogging is

3.3. Theory of Discrete S e t t l i n g

The fundamental relationship for setthng of discrete particles is Newton's law,

diameter. When a particle settles, it accelerates until the forces promoting

Fig. 3.1. Discrete settling of a particle.

/^D = CO^(PLFV2) (3.2)

Substituting υ = ^nd^,A = ^nd^, and solving for the terminal velocity,

which is Newton's law.

For spherical particles, the drag coefficient C D is related to the Reynolds

where d is the diameter of sphere, is the terminal velocity (settling velocity),

Fig. 3.2. Correlation for drag coefficient for spherical particles.

In general, the drag coefficient Co is approximated by

. ' . l o g CD = -log7VR + log24

. •. log C D = log 0.4 = 0.0 log A^R + log 0.4

(where is a constant), which in logarithmic form becomes

log Ks = 2 log of + log a: = 2 log i/ + C (3.8)

where C is a constant. Thus, a logarithmic plot of versus d yields a straight line

(where K' is a constant), which in logarithmic form becomes

log = ^ log of + log ^ ' = ^ log + C (3.10)

where C is a constant. Thus, a logarithmic plot of versus d yields a straight

= [ ( 4 3 / 5 5 . 5 ) ( p ° - V X P S - ρύΙρΔ'''''ά'''' = K"d''*' (3.11)

where K' is a constant. In logarithmic form this becomes

logV, = 1.1431og¿/ + logA:" = 1.143 l o g í / + C " (3.12)

where C " is a constant. Thus a logarithmic plot of versus d yields a straight

log d Fig. 3.3. Logarithmic plot of versus d.

Velocity of fall (cm/sec)

Fig. 3.4. Relation between settling velocity and particle diameter [ 4 ] .

3.4. The Ideal S e d i m e n t a t i o n Tank Concept

chosen for a sedimentation tank is a rectangular section basin which consists of

Fig. 3.7. Geometry of the sedimentation zone.

Fig. 3.8. Model of a sedimentation tank flocculent settling.

K/Ki = L/h (3.16)

Comparing Eqs. (3.15) and (3.17)

Vs/H= VJh (3.18)

VJV, = h/H (3.19)

Q=VA'= VWH (3.20)

From Eq. (3.15) it follows that

Substitution of V from Eq. (3.21) in Eq. (3.20) and simplification leads to

Q=V^LW= V,A (3.22)

where A = LW = horizontal cross-sectional area of the sedimentation zone (ft^)

F r o m Eq. (3.23) it follows that the setthng efficiency is a function of the

The overflow rate defined as

Q/A = ftVft^ h = ft/h

is defined as the settling velocity of a particle that settles through a distance

t = detention period = (volume of t a n k ) / ß = HA/Q (A = LW) (3.24)

F r o m Eq. (3.24) it follows that a settling velocity V, defined as = H/t is

η = H/t = Η/{HA/Q) = Q/A

which is Eq. (3.23).

The scour velocity V^^ is the value of the flow-through velocity V:

dhldr^ VJV (3.28)

multiplying the numerator and denominator of the right-hand member of Eq.

dh/dr = InrHVJlnrHV (3.29)

dh/dr = InrHVJQ (3.30)

Example 3.1. Consider a suspension of sand (s = 2.65) in water at 20°C with a

V, = OAS cm/sec χ ft/30.48 cm χ 3,600 sec/h

^ = 105 X 0.7 = 73.5 ft^

1,000,000/73.5 = 13,605 gal/day ft^

V, = 13,605 gal/day ft^ (100% removal)

13,605/9,533 χ 0.45 cm/sec = 0.45/0.7 = 0.642 cm/sec