CH 11

MWH’S WATER TREATMENT
PRINCIPLES AND DESIGN

3rd Edition
by Crittenden, Trussell, Hand, Howe, and Tchobanoglous
HOMEWORK SOLUTION MANUAL
FOR
Chapter 11
Granular Filtration
Note: If any errors are noted in this solution manual or in the textbook, please notify
Kerry Howe at [email protected]
PROBLEM 11-1
Instructors Note: This problem is similar to Example 11-1. The problem is solved
completely for problem A. The final solutions of the remaining problems are listed at the
end of the solution.
Problem Statement - Samples of filter media were sifted through a stack of sieves and
the weight retained on each sieve is recorded below. For a given sample (A, B, C,
D, or E, to be selected by instructor), determine the effective size and uniformity
coefficient for the media.
Sieve
Sieve Opening,
Designation mm Weight of Retained Media, g
A B C D E
8 2.36 0 4
10 2.00 35 0 11
12 1.70 178 11 0 60
14 1.40 216 315 4 227
16 1.18 242 242 16 343
18 1.00 51 116 0 33 216
20 0.85 12 55 23 75 40
25 0.71 5 26 217 285 16
30 0.60 3 14 325 270 3
35 0.50 e0 2 151 121 1
40 0.425 0 71 21 0
45 0.355 49 8
50 0.300 4 3
Pan -- 20 4
Homework Solution Manual Page 2 of 48

MWH’s Water Treatment: Principles and Design, 3rd ed. Version 1
Chapter 11 - Granular Filtration
Solution
1. Using a spreadsheet, calculate the cumulative fraction passing each sieve. The
cumulative fraction passing is calculated as the total of the mass below the sieve
size divided by the total weight of the sample.
Sieve Opening,
mm Weight Retained, g Cumulative fraction passing
2.36 0 1
2.00 35 0.95
1.70 178 0.71
1.40 216 0.42
1.18 242 0.10
1.00 51 0.03
0.85 12 0.011
0.71 5 0.004
0.60 3 0
0.50 0 0
0.425 0
0.355 0
0.300 0
Total = 742
2. Plot the given data on a graph. In the following graph, the data is plotted on an
arithmetic scale because a log-normal graph does not have sufficient resolution
between the tick marks for 1 mm and 2 mm to identify d10 and d60 accurately.

3. Select the sieve sizes corresponding to the 10 percent size and the 60 percent
size from the data plotted in step 2.
d10 = 1.18 mm and d60 = 1.58 mm.
4. The effective size is the 10th percentile media grain diameter.
ES = d10 = 1.18 mm
5. Calculate the uniformity coefficient using Eq. 11-1.
d60 1.58 mm
=
UC = = 1.34
d10 1.18 mm
6. The effective size and uniformity coefficient for the remaining samples are found
below.
Sample Effective Size, d10 Uniformity Coefficient
A 1.18 1.34
B 0.96 1.48
C 0.45 1.50
D 0.55 1.35
E 1.04 1.30
PROBLEM 11-2
Problem Statement - Explain why a low uniformity coefficient is important in rapid
filtration.
Solution
A low uniformity coefficient is important in rapid filtration because rapid filters are
backwashed in an upflow direction, and filter in a downflow direction. During
backwashing, filter grains are graded according to size, with the largest grains collecting
near the bottom of the bed and the smallest grains rising to the top of the bed. A low
uniformity coefficient means the grains are similar in size, thus there is not a big size
difference between the largest grain and the smallest grain. Particles can penetrate into
the media and the entire depth of the filter bed is then used during filtration to capture
particles. The effectiveness of capturing particles throughout the depth of the media, by
a filter with a low uniformity coefficient, is shown on Figure 11-12b.

If the media has a high uniformity coefficient, a wide distribution of grain sizes is present
and there is a large size difference between the smallest and the largest grains. During
backwashing, the media will stratify and the bed will settle with the small grains at the
top, and progressively larger sized grains below. The small grains at the top of the bed
will cause a large amount of head loss to occur at the top of the bed, leading to short
filter runs before the total available head is reached. (See the solution to Problem 11-8
for an indication of how significant the fines on the top of a bed can be with respect to
head loss.) In addition to head loss, the small grains at the top of the bed will contribute
to straining at the bed surface resulting in an ineffective use of the entire filter bed,
again leading to short filter runs.
PROBLEM 11-3
Problem Statement - Explain (a) the process of ripening, (b) how ripening affects
recovery, or net water production, and (c) how to minimize the duration of ripening.
Solution
a. Ripening is the process of media conditioning, and occurs as clean media
captures particles and becomes more efficient at collecting additional particles.
The particles that are captured during ripening improve the overall efficiency of the
filter by providing a better collector surface than uncoated media grains. Ripening
is characterized by a rise in filter effluent turbidity during the first portion of a filter
run (immediately after backwash), followed by a drop to a low steady-state value.
As much as 90 percent of the total particle passage through a filter over the course
of a filter run may occur during ripening.
b. Modern filtration plants are designed with a filter-to-waste line (see Fig. 11-1), and
the water produced during ripening is wasted. Thus, a long ripening cycle can
reduce the recovery through a filter. For a well-operated plant, the effect of
ripening on recovery should be minimal.
c. The magnitude and duration of the ripening peak can be substantially reduced by
proper backwashing procedures. An important aspect of backwashing is to not
clean the filter excessively. Excessively clean media takes longer to ripen and
more water is wasted. Use of a filter aid polymer in the backwash water can also
minimize the duration of ripening.
PROBLEM 11-4
Instructors Note: This problem is similar to Example 11-2. Head loss through two
layers of media are additive. No pilot or site-specific information is given, so midpoint
values are selected from Table 11-3. The solution is solved completely for problem C
and the solution to the remaining problems are listed at the end of the solution.
Problem Statement - A filter is designed with the following specifications. The

anthracite and sand have density of 1,700 and 2,650 kg/m3, respectively, and the
design temperature is 10ºC. For a given sample (A, B, C, D, or E, to be selected
by instructor), calculate the clean bed head loss.
Item A B C D E
Bed type mono- mono- dual- dual- dual-
media media media media media
Filtration rate (m/h) 8 15 15 10 10
Anthracite specifications:
Effective size (mm) 1.0 1.0 1.2 1.6
Depth (m) 1.8 1.5 1.4 1.2
Sand specifications:
Effective size (mm) 0.55 0.5 0.55 0.55
Depth (m) 0.75 0.3 0.4 0.7
Solution
1. Use Eq. 11-13 to calculate the head loss through the anthracite. Select the
permeability coefficients and porosity from Table 11-3, and
the density and dynamic viscosity of water from Appendix C, Table C-1.
κV = 228, κI = 4.4, and ε = 0.50. From Appendix C-1 at 10 °C, ρW = 999.7
kg/m3 and µ = 1.307 x 10-3 kg/m·s.
2. Calculate the first term in Eq. 11-13.

(1 − ε)2 µLv
hL,1 = κ V
ε3 ρW gd2
( 228 ) (1 − 0.50 )2 (1.307 × 10-3 kg/ms ) (1.5 m )(15 m/h )

=
( 0.50 ) ( 999.7 kg/m3 )(9.81 m/s2 ) (1.0 mm ) (10-3 m/mm ) ( 3600 s/h )
3 2 2
= 0.380 m
3. Calculate the second term in Eq. 11-13.
(1 − ε) Lv 2
hL,2 = κI
ε3 gd
( 4.4 ) (1 − 0.50 )(1.5 m )(15 m/h )2

= 0.047 m
( 0.50 ) (9.81 m/s2 ) (1.0 mm ) (10−3 m/mm ) (3600 s/h)
3 2
4. Add the two terms together.

hL,anth = hL,1 + hL,2 = 0.380 m + 0.047 m = 0.427 m.
5. Repeat the calculations for sand, using κV = 112, κI = 2.2, and ε = 0.42 from
Table 11-3, yields:
(112 ) (1 − 0.42 )2 (1.307 × 10-3 kg/ms ) ( 0.3 m )(15 m/h )
hL,1 =
( 0.42 ) (999.7 kg/m3 )(9.81 m/s2 ) (0.5 mm ) (10-3 m/mm ) ( 3600 s/h )
3 2 2
= 0.339 m
( 2.2 ) (1 − 0.42 )( 0.3 m )(15 m/h )2
hL,2 =
( 0.42 ) (9.81 m/s2 ) ( 0.5 mm ) (10−3 m/mm ) ( 3600 s/h )
3 2
= 0.018 m
hL,sand = 0.339 m + 0.018 m = 0.357 m
6. The total clean bed head loss through the filter is the sum of the head loss
through the two medias.
hL = 0.427 m + 0.357 m = 0.78 m
7. The solution to the remaining problems is shown below (all values in meters
of head loss):
Problem Anthracite Anthracite Sand, Sand, Total
term 1 term 2 term 1 term 2 head loss
A 0.374 0.012 0.38

B 0.456 0.056 0.51
C 0.380 0.047 0.339 0.018 0.78
D 0.164 0.016 0.249 0.01 0.44
E 0.079 0.010 0.436 0.017 0.54
PROBLEM 11-5
Problem Statement - For the media specification given in Problem 7-2 (C, D, or E, to
be selected by instructor), determine if the two media layers are matched to each
other.
Solution
1. For the given sand size, calculate the matched anthracite size using Eq. 11-
25. For problem C:
0.625 0.625
 ρ2 − ρ W   2650 − 1000 kg/m3 
=d1 d=
2  0.5 mm  = 3 
0.85 mm
 ρ1 − ρ W   1700 − 1000 kg/m 
For problems D and E:

0.625 0.625
 ρ2 − ρ W   2650 − 1000 kg/m3 
=d1 d=
2  0.55 mm  = 3 
0.94 mm
 ρ1 − ρW   1700 − 1000 kg/m 
2. The matched anthracite size would be 0.85 mm for 0.50 mm sand and 0.94
mm for 0.55 mm sand. In each problem, the specified anthracite is
considerably larger than the matched size and the media would not be
considered matched.
PROBLEM 11-6
Problem Statement - A filter contains 0.55-mm sand that has a density of 2650 kg/m3.
Calculate the effective size of 1550 kg/m3 anthracite that would be matched to this
sand.
Solution
Matched media can be calculated using Eq. 11-25.
0.625 0.625
 ρ2 − ρ W   2650 − 1000 kg/m3 
=d1 d=   0.55 mm  = 3 
1.09 mm
 ρ1 − ρW   1550 − 1000 kg/m 
2
PROBLEM 11-7
Problem Statement - In dual media filter containing sand and anthracite, which
material will be the top layer? Why?
Solution
The media bed has lower density material at the top, and higher density material at the
bottom. Thus, the top layer is anthracite because it has a lower density than sand.
PROBLEM 11-8
Problem Statement - Using the sieve analysis from Problem 11-1, determine the clean-
bed head loss through a stratified filter bed by calculating the head loss contribution
from each layer of media. Assume that the media stratifies into layers based on grain
size, that the depth of each layer is proportional to the mass of media retained on each
sieve pan, that the grain diameter of the media in each layer is equal to the arithmetic
average of two adjacent sieve pans (i.e., the layer formed by the 178 g of media in
Sample A that passed through the 2-mm pan and was retained on the 1.70-mm pan has
an average size of 1.85 mm), and that the total head loss is the sum of the head loss
from each layer. In addition, assume that the sand that passed through the smallest pan
has an average grain diameter of 0.1 mm. The total bed depth is 0.9 m, the filtration rate
is 10 m/h, and the temperature is 15º C.
a. Calculate the total clean-bed head loss using the entire sand sample in Problem
11-1, including the material smaller than the smallest sieve pan.
b. Calculate the total clean-bed head loss assuming that the top 5 percent of the filter
bed has been scraped to remove fines.
c. Discuss the importance of scraping the surface of rapid-filter beds after media
installation and the impact it has on clean-bed head loss.

Instructors Note: Only L and d will vary between layers of media, and all other
parameters in the Ergun equation (Eq. 11-13) are constant. To simplify subsequent
calculations, a coefficient containing all constants can be calculated for each term.
Because neither pilot nor site-specific information is given in the problem statement,
midpoint values are selected from Table 11-3. Values of ρW and µ are available in
Table C-1 in Appendix C.
The solution is completely worked out for media sample A in Problem 11-1. The
remaining solutions are included in a table at the end of the problem. The importance of
scraping is demonstrated most effectively for the media samples C and D, which have
smaller media and some media that passed through all sieves and was retained in the
pan.
Solution- Part a.
1. Calculate the permeability coefficient for the linear term and the square term of the
nonlinear Forchheimer equation (Eq. 11-12) using Eq. 11-13.
a. Select values for permeability and porosity from Table 11-3, and for the
density and dynamic viscosity of water from Table C-1 in Appendix C.
κV = 112, κI = 2.2, and ε = 0.42
At 15 °C, ρW = 999.1 kg/m3 and µ = 1.14 x 10-3 kg/m·s).
b. Calculate the permeability coefficient for the linear term.
(1 − ε)2 µv
K1′ = κ V
ε3 ρ W g
(112 ) (1 − 0.42 )2 (1.14 × 10-3 kg/ms ) (10 m/h )

= = 1.643 × 10 −7 m2
( 0.42 ) (999.1 kg/m )(9.81 m/s ) (3600 s/h)
3 3 2
c. Calculate the permeability coefficient for the square term.

(1 − ε) v 2
K ′2 = κI
ε3 g

( 2.2 ) (1 − 0.42 )(10 m/h )2
= = 1.355 × 10 −5 m
( 0.42 ) ( 9.81 m/s ) ( 3600 s/h )
3 2 2
2. Calculate the average size of media in the first layer.

d=i′ 1
2 ( di−1 + di =) 1
2 ( 2.36 + 2.00 mm=) 2.175 mm
3. Calculate the fraction of media in the first layer.

35 g
=fi = 0.0472
742 g
4. Calculate the depth of the first layer.
i= 0.0472 ( 0.9 m=
L=i fL ) 0.0425 m
5. Calculate the head loss in the first layer using Eq. 11-13.
Li L
hL,i K1′
= 2
+ K ′2 i
di di
(1.643 × 10 −7
m2 ) ( 0.0425 m )
+
(1.355 × 10 −5
m ) ( 0.0425 m )
( 2.18 mm ) (10−3
2
m/mm )
2
( 2.18 mm ) (10−3 m/mm )
= 0.0017 m
6. Repeat steps 2 through 5 for each remaining layer. The results of these
calculations are shown in the table below:
Sieve Weight Average Fraction of Depth of

Headloss
opening retained size media in layer
(mm) (g) (mm) layer (m) (m)
2.36 0 2.36 0 0 0
2.00 35 2.18 0.0472 0.042 0.00173
1.70 178 1.85 0.2399 0.216 0.01193
1.40 216 1.55 0.2911 0.262 0.02019
1.18 242 1.29 0.3261 0.294 0.03203
1.00 51 1.09 0.0687 0.062 0.00931
0.85 12 0.925 0.0162 0.015 0.00301
0.71 5 0.780 0.0067 0.006 0.00174
0.60 3 0.655 0.0040 0.004 0.00147
0.50 0 0.550 0.0000 0.000 0.00000
0.425 0 0.463 0.0000 0.000 0.00000

0.355 0 0.390 0.0000 0.000 0.00000
0.300 0 0.328 0.0000 0.000 0.00000
Pan 0 0.100 0.0000 0.000 0.00000
Total 742 1.0000 0.900 0.081
Solution - Part b.
1. Calculate the amount of media that will be removed by scraping. Since all the
media has the same density, removing 5 percent of the volume of media also
removes 5 percent of the weight.
( 742 g)( 0.05 ) = 37 g
2. Scraping is done after a filter has been backwashed, and the backwashing
process stratifies the media with the smallest sizes on top and largest sizes on
bottom. Thus, the scraping will remove 37 g of the smallest material. Thus,
scraping will remove all of the media that passed through the 0.85 mm sieve (20 g)
and some of the media that passed through the 1.0 mm sieve (17 g). Calculate
the amount of media that passed through the 1.0 mm sieve remaining:
51 g − 17 g =
34 g
0.045 m − ( 0.004 m + 0.006 m + 0.015 m ) =

0.02 m
3. Calculate the thickness remaining in that layer.

0.062 m ( 34 g / 51 g ) = 0.041 m
4. Repeat the calculations shown in the table above, eliminating the material
removed by scraping.
Sieve Fraction of
Weight Average Depth of Head
opening, media in
retained, g size, mm layer, m loss, m
mm layer
2.360 0 2.360 0 0 0
2.000 35 2.180 0.0472 0.042 0.002
1.700 178 1.850 0.2399 0.216 0.012
1.400 216 1.550 0.2911 0.262 0.020
1.180 242 1.290 0.3261 0.294 0.032

1.000 34 1.090 0.0458 0.041 0.006
Total 705 0.855 0.072
Solution - Part c.
Scraping the top 5 percent removed reduced the total head loss from 0.081 m to 0.072
m, an 11 percent reduction in clean bed head loss. Scraping the bed increases the
available head for particle removal in the filter, and is an important part of media
installation if long filter runs are to be achieved.
Solution to the remaining problem statements:

B C D E
Sieve Original Scraped Original Scraped Original Scraped Original Scraped
opening hL hL hL hL hL hL hL hL
(mm) (m) (m) (m) (m) (m) (m) (m) (m)
2.36 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0001 0.0001
2.00 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0004 0.0004
1.70 0.0007 0.0007 0.0000 0.0000 0.0000 0.0000 0.0032 0.0032
1.40 0.0280 0.0280 0.0000 0.0000 0.0003 0.0003 0.0171 0.0171
1.18 0.0304 0.0304 0.0000 0.0000 0.0019 0.0019 0.0366 0.0366
1.00 0.0201 0.0201 0.0000 0.0000 0.0053 0.0053 0.0318 0.0318
0.85 0.0131 0.0131 0.0050 0.0050 0.0166 0.0166 0.0081 0.0028
0.71 0.0086 0.0010 0.0652 0.0652 0.0877 0.0877 0.0045 0.0000
0.60 0.0065 0.0000 0.1372 0.1372 0.1167 0.1167 0.0012 0.0000
0.50 0.0013 0.0000 0.0896 0.0896 0.0735 0.0699 0.0006 0.0000
0.425 0.0000 0.0000 0.0592 0.0592 0.0179 0.0000 0.0000 0.0000
0.355 0.0000 0.0000 0.0571 0.0350 0.0095 0.0000 0.0000 0.0000
0.300 0.0000 0.0000 0.0066 0.0000 0.0051 0.0000 0.0000 0.0000
Pan 0.0000 0.0000 0.3464 0.0000 0.0709 0.0000 0.0000 0.0000
Total 0.109 0.093 0.766 0.391 0.405 0.298 0.104 0.092
The numbers in the above table indicate that scraping is most important for the smaller
media.

PROBLEM 11-9
Problem Statement - Compare the clean-bed head loss at 15 ºC through a rapid filter
with a filtration rate of 15 m/h to that through a slow sand filter with a filtration rate of
0.15 m/h for media with the following specifications: effective size 0.5 mm, density 2650
kg/m3, depth 1 m, and porosity 0.42. What implications do these calculations have on
the significance of clean-bed head loss in the design of rapid and slow sand filters?
Instructors Note: This problem is similar to Example 11-2. Because neither pilot nor
site-specific information is given in the problem statement, midpoint values are selected
from Table 11-3. Values of ρW and µ are available in Table C-1 in Appendix C.
Calculate the clean bed head loss through the rapid filter at 15 m/h using Eq. 11-13.
1. Select values for permeability and porosity from Table 11-3 and for density and
dynamic viscosity from Table C-1 in Appendix C.
κV = 112, κI = 2.2, and ε = 0.42
At 15 °C, ρW = 999 kg/m3 and µ = 1.14 x 10-3 kg/m·s
2. Calculate the first term in Eq. 11-13.
(1 − ε)2 µLv
hL,1 = κ V
ε3 ρW gd2
(112 ) (1 − 0.42 )2 (1.14 × 10-3 kg/ms ) (1.0 m )(15 m/h )

=
( 0.42 ) ( 999 kg/m3 )( 9.81 m/s2 ) ( 0.5 mm ) (10-3 ) (3600 s/h)
3 2 2
m/mm
= 0.985 m
3. Calculate the second term in Eq. 11-13.
(1 − ε) Lv 2
hL,2 = κI
ε3 gd
( 2.2 ) (1 − 0.42 )(1.0 m )(15 m/h )2

=
( 0.42 ) ( 9.81 m/s2 ) ( 0.5 mm ) (10−3 m/mm ) ( 3600 s/h )
3 2
= 0.061 m
4. Add the two terms together.
hL = hL,1 + hL,2 = 0.985 m + 0.061 m = 1.046 m.

5. Calculate the clean bed head loss through the slow sand filter at 15 m/h using Eq.
11-13. Use the first term for Eq. 11-13 calculated in Part a., step 2 above.
Calculate the second term in Eq. 11-13 and solve the equation.
(1 − ε) Lv 2
hL,2 = κI
ε3 gd
( 2.2 ) (1 − 0.42 )(1.0 m )( 0.15 m/h )2

=
( 0.42 ) ( 9.81 m/s2 ) ( 0.5 mm ) (10−3 m/mm ) ( 3600 s/h )
3 2
= 6.10 × 10 −6 m
hL = 0.00985 m + 6.10×10-6 m = 0.00985 m = 0.985 cm.
6. The total available head for rapid filters is typically 1.8 to 3.0 m (see Table 11-11),
so the clean bed head loss (1.046 m), has a significant impact on the remaining
head available for head buildup during filtration and is an important consideration
in rapid filter design. In contrast, the clean bed head loss of the slow sand filter is
less than 1 cm. Because the total available head for slow sand filters is 0.9 to 1.8
m (see Table 11-11), the clean bed head loss is a negligible portion of the total
available head and can be ignored in slow sand filtration design.
PROBLEM 11-10
Problem Statement - Given a backwash flow rate of 45 m/h and temperature of 20º C,
calculate the largest (a) sand particle (density = 2650 kg/m3) and (b) floc solid
particle (density = 1050 kg/m3) that can be washed from a filter bed.
Solution
1. Determine the largest particle that can be fluidized during backwash.
The largest particle that can be fluidized at a particular backwash velocity can be
calculated by rearranging Eq. 11-14.
 g ( ρP − ρW ) d1.6 
0.714
v= 
 13.9ρW µ
0.4 0.6

is rearranged to:
0.625
 13.9ρ0.4
W µ
0.6 1.4
v 
d= 
 g (ρ − ρ )
 P W 
2. Match the velocity units given in the problem statement to the equation developed
in step 1. Convert the velocity from units of m/hr to units of m/s.
45 m/h
=
45 m/h = 0.0125 m/s
3600 s/h
3. Calculate the largest sand and floc particle using the equation from step 1.
a. Calculate the largest sand particle that can be washed from the filter bed.
From Table C-1 in Appendix C at 20 °C, ρW = 998 kg/m3 and µ = 1.002 x 10-3
kg/m·s.
( ) ( )
0.625
 13.9 998 kg/m3 0.4 1.002 × 10 −3 kg/ms 0.6 ( 0.0125 m/s )1.4 
d= 


( )(
9.81 m/s 2650 − 998 kg/m
2 3
) 

=1.105 × 10 −4 m =0.11 mm
b. Calculate the largest floc particle that can be washed from the filter bed.
( ) ( )
0.625
 13.9 998 kg/m3 0.4 1.002 × 10 −3 kg/ms 0.6 ( 0.0125 m/s )1.4 
d= 


( )(
9.81 m/s 1050 − 998 kg/m
2 3
) 

9.59 × 10 −4 m =
= 0.959 mm
Comment
The largest sand particle that will be washed from the bed is a sand grain 0.11 mm in
diameter. Thus, sand media larger than 0.11 mm will stay in the filter while floc particles
as large as 0.959 mm will be washed from the bed.
PROBLEM 11-11
Problem Statement - Calculate and plot the size of particles that will be washed from a
filter as a function of backwash velocity ranging from 10 to 100 m/h at 20º C for (a) sand
particles (density = 2650 kg/m3), (b) anthracite particles (density = 1650 kg/m3), and (c)
floc solid particles (density = 1050 kg/m3). In addition, calculate the minimum fluidization
velocities for 0.5-mm sand (porosity 0.40) and 1.0-mm anthracite (porosity 0.50) and
indicate these velocities on your graph. What is an appropriate range for the backwash

velocity for dual-media filters? Assuming the backwash troughs are placed high enough,
will any media be lost with backwash velocities in this range?
Instructors Note: Because neither pilot nor site-specific information is given in the
problem statement, midpoint values are selected from Table 11-3. Values of ρW and µ
are available in Table C-1 in Appendix C.
Solution
1. Determine the largest particle that can be fluidized during backwash.
The largest particle that can be fluidized at a particular backwash velocity can be
calculated by rearranging Eq. 11-14.
 g ( ρP − ρW ) d1.6 
0.714
v= 
 13.9ρW µ
0.4 0.6

is rearranged to:
0.625
 13.9ρ0.4
W µ
0.6 1.4
v 
d= 
 g (ρ − ρ )
 P W 
2. Convert the backwash velocity of 10 m/h to units of m/s:
10 m/h
=
10 m/h = 0.00278 m/s
3600 s/h
3. Calculate the largest sand, anthracite and floc particle using the equation from
step 1. From Appendix C-1 at 20 °C, ρW = 998 kg/m3 and µ = 1.002 x 10-3 kg/m·s.
a. Calculate the largest sand particle that can be washed from the filter bed at
10 m/h.
 13.9 ( 998 kg/m3 )0.4 (1.002 × 10 −3 kg/ms )0.6 ( 0.00278 m/s )1.4 
0.625
 103 mm 
d=   m 
 ( 9.81 m/s2 )( 2650 − 998 kg/m3 )   
−2
= 2.96 × 10 mm
b. Calculate the largest anthracite particle that can be washed from the filter
bed at 10 m/h.
 13.9 ( 998 kg/m3 )0.4 (1.002 × 10 −3 kg/ms )0.6 ( 0.00278 m/s )1.4 
0.625
 103 mm 
d=   m 
 ( 9.81 m/s2 )(1650 − 998 kg/m3 )   

−2
= 5.29 × 10 mm
c. Calculate the largest floc particle that can be washed from the filter bed.
 13.9 ( 998 kg/m3 )0.4 (1.002 × 10 −3 kg/ms )0.6 ( 0.00278 m/s )1.4 
0.625
 103 mm 
d=   m 
 ( 9.81 m/s2 )(1050 − 998 kg/m3 )   
−1
= 2.57 × 10 mm
4. Repeat step 3 a, b and c for increments of velocity between 10 m/h and 100 m/h.
The results are summarized in the following table:
Velocity, m/h Velocity, m/s Sand size, mm Anthracite size, mm Floc. size, mm
10 0.00278 0.0296 0.0529 0.2570
15 0.00417 0.0422 0.0755 0.3670
20 0.00556 0.0543 0.0971 0.4720
25 0.00694 0.0660 0.1180 0.5730
30 0.00833 0.0774 0.1380 0.6720
35 0.00972 0.0886 0.1580 0.7690
40 0.01111 0.0996 0.1780 0.8650
45 0.01250 0.1100 0.1970 0.9590
50 0.01389 0.1210 0.2160 1.0500
55 0.01528 0.1320 0.2350 1.1400
60 0.01667 0.1420 0.2540 1.2300
65 0.01806 0.1520 0.2720 1.3200
70 0.01944 0.1620 0.2910 1.4100
75 0.02083 0.1730 0.3090 1.5000
80 0.02222 0.1830 0.3270 1.5900
85 0.02361 0.1930 0.3440 1.6700
90 0.02500 0.2020 0.3620 1.7600
95 0.02639 0.2120 0.3800 1.8400
100 0.02778 0.2220 0.3970 1.9300
5. Calculate the minimum fluidization velocity for sand using Eq. 11-21.
a. Calculate the backwash calculation factor, β, using Eq. 11-20.
gρW ( ρP − ρW ) d3 ε3
β=
µ2

=
( 9.81 m/s )(998 kg/m )( 2650 − 998 kg/m ) (0.5 mm ) (0.40 )
2 3 3 3 3
(1.002 × 10 kg/ms ) (10 mm/m )

2 3
−3 3
= 128.9
b. Calculate the Reynolds Number using Eq. 11-21. Select values for
permeability and porosity from Table 11-3: κV = 112 and κI = 2.2.
−κ V (1 − ε ) 1
κ2V (1 − ε ) + 4 κIβ
2
=Re +
2κI 2κI
−112 (1 − 0.40 ) 1
(112 ) (1 − 0.40 ) + 4 ( 2.2 )(128.9 )
2 2
= +
2 ( 2.2 ) 2 ( 2.2 )
= 1.81
c. Calculate velocity using Eq. 11-8.
ρW vd
Re =
µ
is rearranged to:
=
µ Re
v =
(1.002 × 10 kg/ms ) (1.81) (10
−3 3
mm/m )
= 0.00364 m/s
ρW d ( 998 kg/m ) ( 0.5 mm )3
= =
v 0.00364 m/s ( 3600 s/h ) 13.1 m/h
6. Calculate the minimum fluidization velocity for anthracite using Eq. 11-21.
a. Calculate the backwash calculation factor, β, using Eq. 11-20.
β=
( 9.81 m/s )(998 kg/m )(1650 − 998 kg/m ) (1.0 mm ) (0.50 )
2 3 3 3 3
(1.002 × 10 kg/ms ) (10 mm/m )

2 3
−3 3
= 794.6
b. Calculate the Reynolds Number using Eq. 11-21. Select values for
permeability and porosity from Table 11-3: κV = 228 and κI = 4.4.
−228 (1 − 0.50 ) 1
( 228 ) (1 − 0.50 ) + 4 ( 4.4 )(=
2014 ) 5.71
2 2
=Re +
2 ( 4.4 ) 2 ( 4.4 )
c. Calculate velocity using Eq. 11-8.
=
µ Re
v =
(1.002 × 10 kg/ms ) ( 5.71) (10
−3 3
mm/m )
= 0.00573 m/s
ρW d ( 998 kg/m ) (1.0 mm )3

= =
v 0.00573 m/s ( 3600 s/h ) 20.6 m/h
7. The appropriate range for backwashing velocity is such that the media is fluidized
but cannot be washed out of the filter box. To achieve fluidization for both types of
media, the velocity has to be above the fluidization velocity for each type, hence,
the minimum backwash flowrate for this installation would be 20.6 m/h. At the limit
of this graph (100 m/h), the sand size that would be washed from the filter would
be 0.222 mm and the anthracite size that would be washed from the media would
be 0.397 mm. These values are below the effective sizes of 0.5 mm and 1.0 mm
for sand and anthracite, respectively, so the media would not be expected to be
washed from the filter at this rate. Thus, the appropriate range of backwash
velocity, based solely on this graph, would be 20.6 to 100 m/h, and media would
not be lost if the backwash troughs were above the height of the expanded media.

PROBLEM 11-12
Problem Statement - A monomedia anthracite filter is designed with the following

specifications: effective size 1.1 mm, uniformity coefficient 1.4, and density 1650 kg/m3.
a. Calculate backwash rate to get a 25 percent expansion at the design summer

temperature of 22º C.
b. Calculate the expansion that occurs at the backwash rate determined in part a at
the minimum winter temperature of 3º C.
c. Discuss the implications of these results on backwash operations for plants that
experience a large seasonal variation in water temperature.
Instructors Note: Calculations for the backwash velocity needed to achieve a certain
amount of bed expansion are shown in Example 11-4. Calculations for the backwash
velocity needed to achieve a certain amount of bed expansion are shown in Example
11-5. Because neither pilot nor site-specific information is given in the problem
statement, midpoint values are selected from Table 11-3. Values of ρW and µ can be
interpolated from Table C-1 in Appendix C.
Solution - Part a.
1. Calculate the expanded bed porosity, εE, that corresponds to 25 percent
expansion using Eq. 11-15. From Table 11-3, the fixed bed porosity is εE = 0.50.
L   1.0  
εE = 1 −  F (1 − εF )  = 1 −   (1 − 0.5 ) = 0.60
 LE   1.25  
2. Calculate β using Eq. 11-20. From Table C-1 in Appendix C at 22 °C, ρW = 997.8
kg/m3 and µ = 9.55 x 10-4 kg/m·s.
gρW ( ρP − ρW ) d3 ε3
β=
µ2
β=
( 9.81 m/s )(997.8 kg/m )(1650 − 997.8 kg/m ) (0.0011 m ) ( 0.60 )
2 3 3 3 3
( 9.55 × 10 kg/ms )
2
-4
= 1165

3. Calculate Re using Eq. 11-21. Select the values for κV and κI from Table 11-3. κV
= 228 and κI = 4.4
−κ V (1 − ε ) 1
κ2V (1 − ε ) + 4 κIβ
2
=Re +
2κI 2κI
−228 (1 − 0.60 ) 1
( 228 ) (1 − 0.60 )2 + 4 ( 4.4 )(=
1165 ) 13.40
2
=Re +
2 ( 4.4 ) 2 ( 4.4 )
4. Calculate v using Eq. 11-8.
=
µ Re
v =
( 9.55 × 10 kg/ms ) (13.40 )(3600 s/h
−4
=
)
42.0 m/h
ρW d ( 997.8 kg/m ) (0.0011 m )
3
Part b.
1. Calculate X using Eq. 11-23. From Table C-1 in Appendix C at 3°C, ρW = 1000
kg/m3, µ = 1.62 x 10-3 kg/m·s.
µv  κ ρ vd 
=X κ + I W 
2  V
2g ( ρP − ρW ) d  µ 
  42.0 m/h  
 (
1.62 × 10 −3 kg/ms 

) 
3600 s/h 

X  ×
 3  1.1 mm 
2

( )(
 2 9.81 m/s 1650 - 1000 kg/m  3
2
 
 10 mm/m  
)

 42.0 m/h  1.1 mm  
 ( 4.4 ) 1000 kg/m3 
( )  3 
228 +  3600 s/h  10 mm/m  
 (
1.62 × 10 −3 kg/ms  )
 
= 0.3215
2. Calculate Y using Eq. 11-24.
k V µv
Y=
3g ( ρP − ρW ) d2
 42.0 m/h 
( 228 ) (1.62 × 10−3
kg/ms  )
 3600 s/h 
Y = 2
0.1859
3  1.1 mm 
( )(
3 9.81 m/s 1650 − 1000 kg/m  3
2
 10 mm/m 
 )
3. Calculate porosity using Eq. 11-22.

( ) ( )
1 1
=
ε X + X2 + Y 3 + 3 X − X2 + Y 3
2 2
3
1 1
0.3215 + ( 0.3215 ) + ( 0.1859 )  + 3 0.3215 − ( 0.3215 ) + ( 0.1859 ) 

2 3 2 2 3 2
=εE 3
   
εE =0.65
4. Calculate the percent bed expansion using Eq. 11-15.
1 − εF  1 − 0.50 
=LE L= 1.0  =  1.44
1 − εE  1 − 0.65 
F
Part c.
The target backwash velocity of 42.0 m/h caused the bed to expand by 25 percent in
the summer, but by 44 percent in the winter, when the water is cold. An expansion of
44 percent may be excessive and may lead to media loss or other filtration problems.
For plants that experience large seasonal variations in water temperature, it may be
necessary to adjust the backwash velocity to maintain a constant amount of bed
expansion as the water temperature changes.
PROBLEM 11-13
Problem Statement - Using the Yao filtration model, examine the effect of filtration rate
on filter performance for particles with diameters of 0.1, 1.0, and 10 µm. Assume a
monodisperse media of 0.5 mm diameter, porosity 0.42, particle density 1020 kg/m3,
filtration rate 10 m/h, filter depth 1 m, temperature 20º C, and attachment efficiency 1.0.
Plot the results as C/CO as a function of filtration rate over a range from 1 to 25 m/h.
Comment on the effect of filtration rate and particle size on filter performance.
Solution
1. Calculate the transport efficiency due to interception for 0.1 µm particles using Eq.
11-37.
2 2
3  dP  3  0.1× 10-3 mm 
η=
I   =  = 6.0 × 10
-8
2  dC  2  0.5 mm 
2. Develop an expression for the transport efficiency due to sedimentation as a

function of velocity and particle size using Eq. 11-38.

g ( ρP − ρW ) dP2 ( 9.81 m/s2 ) (1020 − 998.2 kg/m3 ) ( 3600 s/h ) dP2
=ηG =
18µv F 18 (1.0 × 10 −3 kg/ms ) (106 µm/m )
2
vF
dP2
= 4.27 × 10 −5
vF
3. Calculate the transport efficiency due to sedimentation for 0.1 µm particles at a
filtration rate of 1 m/h.
( 0.1 µm )
2
−5
η=
G 4.27 × 10 = 4.27 × 10 −7
1 m/h
4. Develop an expression for the Peclet Number as a function of velocity and particle
size using Eq. 11-40.
3π (1.0 × 10 −3 kg/ms ) ( 0.5 mm ) (10 −6 m/µm ) (10 −3 m/mm )
Pe = dP v
(1.381× 10−23 kgm2 /s2 K ) ( 293.15 K )( 3600 s/h )
= 3.23 × 105 dP v
5. Calculate the Peclet Number for 0.1 µm particles at a filtration rate of 1 m/h.
3.23 × 105 ( 0.1 µm ) (1 m/h ) =
Pe = 3.23 × 10 4
6. Calculate the transport efficiency due to diffusion for 0.1 µm particles at a filtration
rate of 1 m/h using Eq. 11-39.
4 ( 3.23 × 10 4 ) = 3.94 × 10 −3
−2 / 3
η=
D 4 Pe−2=
/3
7. Calculate the total transport efficiency using Eq. 11-41.

η = ηI + ηG + ηD = 6.0 × 10-8 + 4.27 × 10-7 + 3.94 × 10-3 = 3.94 × 10-3
8. Calculate C/C0 for 0.1 µm particles at a filtration rate of 1 m/h using Eq. 11-36.
 −3(1−0.42 )( 3.94×10−3 )(1.0 )(1 m ) 
 
2( 0.5 mm )(10−3 m/mm )
C /=
CO e 
= 1.05 × 10 −3

9. Repeat steps 1 through 8 for particle sizes 0.1 µm, 1 µm and 10 µm, and filtration
rates 1 through 25 m/h. The results of these calculations are shown in the
following tables.
Table 1
0.1 µm particles
Velocity, m/h ηG Pe ηD η C/Co

-7 4 -3 -3
1 4.27 x 10 3.24 x 10 3.94 x 10 3.94 x 10 0.0011
-7 4 -3 -3
2 2.13 x 10 6.48 x 10 2.48 x 10 2.48 x 10 0.0134
-7 4 -3 -3
3 1.42 x 10 9.72 x 10 1.89 x 10 1.89 x 10 0.0372
-7 5 -3 -3
4 1.07 x 10 1.30 x 10 1.56 x 10 1.56 x 10 0.0660
-8 5 -3 -3
5 8.54 x 10 1.62 x 10 1.35 x 10 1.35 x 10 0.0961
-8 5 -3 -3
6 7.11 x 10 1.94 x 10 1.19 x 10 1.19 x 10 0.1256
-8 5 -3 -3
7 6.10 x 10 2.27 x 10 1.08 x 10 1.08 x 10 0.1539
-8 5 -4 -4
8 5.33 x 10 2.59 x 10 9.84 x 10 9.84 x 10 0.1804
-8 5 -4 -4
9 4.74 x 10 2.92 x 10 9.10 x 10 9.10 x 10 0.2054
-8 5 -4 -4
10 4.27 x 10 3.24 x 10 8.48 x 10 8.48 x 10 0.2286
-8 5 -4 -4
11 3.88 x 10 3.56 x 10 7.96 x 10 7.96 x 10 0.2504
-8 5 -4 -4
12 3.56 x 10 3.89 x 10 7.51 x 10 7.51 x 10 0.2707
-8 5 -4 -4
13 3.28 x 10 4.21 x 10 7.12 x 10 7.12 x 10 0.2897
-8 5 -4 -4
14 3.05 x 10 4.54 x 10 6.78 x 10 6.78 x 10 0.3075
-8 5 -4 -4
15 2.85 x 10 4.86 x 10 6.47 x 10 6.47 x 10 0.3243
-8 5 -4 -4
16 2.67 x 10 5.18 x 10 6.20 x 10 6.20 x 10 0.3400
-8 5 -4 -4
17 2.51 x 10 5.51 x 10 5.95 x 10 5.95 x 10 0.3549
-8 5 -4 -4
18 2.37 x 10 5.83 x 10 5.73 x 10 5.73 x 10 0.3689
-8 5 -4 -4
19 2.25 x 10 6.16 x 10 5.53 x 10 5.53 x 10 0.3821
-8 5 -4 -4
20 2.13 x 10 6.48 x 10 5.34 x 10 5.34 x 10 0.3947
-8 5 -4 -4
21 2.03 x 10 6.80 x 10 5.17 x 10 5.17 x 10 0.4066
-8 5 -4 -4
22 1.94 x 10 7.13 x 10 5.01 x 10 5.01 x 10 0.4179
-8 5 -4 -4
23 1.86 x 10 7.45 x 10 4.87 x 10 4.87 x 10 0.4287
-8 5 -4 -4
24 1.78 x 10 7.78 x 10 4.73 x 10 4.73 x 10 0.4390
-8 5 -4 -4
25 1.71 x 10 8.10 x 10 4.60 x 10 4.60 x 10 0.4488
Table 2
1 µm particles.
-5 5 -4 -4
1 4.27 x 10 3.24 x 10 8.48 x 10 8.97 x 10 0.2101
-5 5 -4 -4
2 2.13 x 10 6.48 x 10 5.34 x 10 5.62 x 10 0.3764
-5 5 -4 -4
3 1.42 x 10 9.72 x 10 4.08 x 10 4.28 x 10 0.4750
-5 6 -4 -4
4 1.07 x 10 1.30 x 10 3.37 x 10 3.53 x 10 0.5409
-6 6 -4 -4
5 8.54 x 10 1.62 x 10 2.90 x 10 3.05 x 10 0.5887
-6 6 -4 -4
6 7.11 x 10 1.94 x 10 2.57 x 10 2.70 x 10 0.6252
-6 6 -4 -4
7 6.10 x 10 2.27 x 10 2.32 x 10 2.44 x 10 0.6543

-6 6 -4 -4
8 5.33 x 10 2.59 x 10 2.12 x 10 2.23 x 10 0.6780
-6 6 -4 -4
9 4.74 x 10 2.92 x 10 1.96 x 10 2.07 x 10 0.6979
-6 6 -4 -4
10 4.27 x 10 3.24 x 10 1.83 x 10 1.93 x 10 0.7148
-6 6 -4 -4
11 3.88 x 10 3.56 x 10 1.71 x 10 1.81 x 10 0.7294
-6 6 -4 -4
12 3.56 x 10 3.89 x 10 1.62 x 10 1.71 x 10 0.7422
-6 6 -4 -4
13 3.28 x 10 4.21 x 10 1.53 x 10 1.63 x 10 0.7535
-6 6 -4 -4
14 3.05 x 10 4.54 x 10 1.46 x 10 1.55 x 10 0.7636
-6 6 -4 -4
15 2.85 x 10 4.86 x 10 1.39 x 10 1.48 x 10 0.7726
-6 6 -4 -4
16 2.67 x 10 5.18 x 10 1.34 x 10 1.42 x 10 0.7808
-6 6 -4 -4
17 2.51 x 10 5.51 x 10 1.28 x 10 1.37 x 10 0.7882
-6 6 -4 -4
18 2.37 x 10 5.83 x 10 1.23 x 10 1.32 x 10 0.7950
-6 6 -4 -4
19 2.25 x 10 6.16 x 10 1.19 x 10 1.27 x 10 0.8013
-6 6 -4 -4
20 2.13 x 10 6.48 x 10 1.15 x 10 1.23 x 10 0.8070
-6 6 -4 -4
21 2.03 x 10 6.80 x 10 1.11 x 10 1.19 x 10 0.8124
-6 6 -4 -4
22 1.94 x 10 7.13 x 10 1.08 x 10 1.16 x 10 0.8173
-6 6 -4 -4
23 1.86 x 10 7.45 x 10 1.05 x 10 1.13 x 10 0.8219
-6 6 -4 -4
24 1.78 x 10 7.78 x 10 1.02 x 10 1.10 x 10 0.8262
-6 6 -5 -4
25 1.71 x 10 8.10 x 10 9.92 x 10 1.07 x 10 0.8303
Table 3
10 µm particles.
-3 6 -4 -3
1 4.27 x 10 3.24 x 10 1.83 x 10 5.05 x 10 0.0002
-3 6 -4 -3
2 2.13 x 10 6.48 x 10 1.15 x 10 2.85 x 10 0.0070
-3 6 -5 -3
3 1.42 x 10 9.72 x 10 8.78 x 10 2.11 x 10 0.0254
-3 7 -5 -3
4 1.07 x 10 1.30 x 10 7.25 x 10 1.74 x 10 0.0485
-4 7 -5 -3
5 8.54 x 10 1.62 x 10 6.25 x 10 1.52 x 10 0.0715
-4 7 -5 -3
6 7.11 x 10 1.94 x 10 5.53 x 10 1.37 x 10 0.0927
-4 7 -5 -3
7 6.10 x 10 2.27 x 10 4.99 x 10 1.26 x 10 0.1117
-4 7 -5 -3
8 5.33 x 10 2.59 x 10 4.57 x 10 1.18 x 10 0.1285
-4 7 -5 -3
9 4.74 x 10 2.92 x 10 4.22 x 10 1.12 x 10 0.1433
-4 7 -5 -3
10 4.27 x 10 3.24 x 10 3.94 x 10 1.07 x 10 0.1564
-4 7 -5 -3
11 3.88 x 10 3.56 x 10 3.69 x 10 1.02 x 10 0.1681
-4 7 -5 -4
12 3.56 x 10 3.89 x 10 3.49 x 10 9.91 x 10 0.1784
-4 7 -5 -4
13 3.28 x 10 4.21 x 10 3.30 x 10 9.61 x 10 0.1877
-4 7 -5 -4
14 3.05 x 10 4.54 x 10 3.15 x 10 9.36 x 10 0.1961

-4 7 -5 -4
15 2.85 x 10 4.86 x 10 3.00 x 10 9.15 x 10 0.2037
-4 7 -5 -4
16 2.67 x 10 5.18 x 10 2.88 x 10 8.96 x 10 0.2105
-4 7 -5 -4
17 2.51 x 10 5.51 x 10 2.76 x 10 8.79 x 10 0.2168
-4 7 -5 -4
18 2.37 x 10 5.83 x 10 2.66 x 10 8.64 x 10 0.2225
-4 7 -5 -4
19 2.25 x 10 6.16 x 10 2.57 x 10 8.50 x 10 0.2278
-4 7 -5 -4
20 2.13 x 10 6.48 x 10 2.48 x 10 8.38 x 10 0.2326
-4 7 -5 -4
21 2.03 x 10 6.80 x 10 2.40 x 10 8.27 x 10 0.2371
-4 7 -5 -4
22 1.94 x 10 7.13 x 10 2.33 x 10 8.17 x 10 0.2412
-4 7 -5 -4
23 1.86 x 10 7.45 x 10 2.26 x 10 8.08 x 10 0.2451
-4 7 -5 -4
24 1.78 x 10 7.78 x 10 2.20 x 10 8.00 x 10 0.2487
-4 7 -5 -4
25 1.71 x 10 8.10 x 10 2.14 x 10 7.92 x 10 0.2520
Comment

Filter performance depends on both filtration rate and particle size. Filter performance
decreases as filtration rate increases. At a filtration rate of 1 m/h, the effluent
concentration of the 0.1 and 10 µm particles is nearly zero, and the effluent
concentration of the 1 µm particles is about 20 percent of the initial concentration. As
the filtration rate increases to 25 m/h, the effluent concentration of all particles
increases.
Of the three particle sizes plotted, the best removal efficiency occurs for the 10 µm
particles for all values of filtration rate. The 0.1 µm particles have the second best
removal efficiency, and the 1 µm particles have the worst removal efficiency. Low
removal efficiency of particles in the range of 1 µm has been observed in actual filter
data.
PROBLEM 11-14
Instructor Note: Calculations are illustrated for 0.1 µm size particles in steps 3 through
9. These same calculations must be performed for the 1 and 10 µm size particles. As
noted on page 767 of the text, the calculation of the Hamaker number is beyond the
scope of this text, but the value ranges from 10-19 to 10-20 J. A value of 10-20 J (1 x 10-20
kg·m2/s2) is being used in this solution.
Problem Statement - Using the Tufenkji and Elimelech filtration model, examine the
effect of water temperature on filter efficiency for particles with diameters of 0.1, 1.0,
and 10 µm. Assume a monodisperse media of 0.5 mm diameter, porosity 0.42, particle
density 1020 kg/m3, filtration rate 10 m/h, filter depth 1 m, and attachment efficiency 1.0.
Plot the results as log (C/CO) as a function of temperature over a temperature range of 1
to 25º C. What implications do these calculations have on filtration in cold climates? Is
temperature more important for filtration of certain particle sizes?
Solution
1. Calculate γ using Eq. 11-53.
(1 − ε ) (1 − 0.42 )
1/ 3 1/ 3
γ= = = 0.8340
2. Calculate AS using Eq. 11-54.

( )2 1 − ( 0.8340 ) 
5
2 1− γ5  
=AS = = 33.64
2 − 3 γ + 3 γ − 2γ 2 − 3 ( 0.8340 ) + 3 ( 0.8340 ) − 2 ( 0.8340 )
5 6 5 6
3. Calculate NA using Eq. 11-51. A Hamaker Number of 1 x 10-20 kg·m2/s2 is being

used in this solution.
=
NA
Ha
=
(1× 10 −20
kgm2 /s2 ) = 2.21 × 10 −2
3πµdP2 v 2  10 m/h 
(
3π 1.73 × 10 −3 kg/ms 0.1 × 10 −6 )( m  ) 
 3600 s/h 
4. Calculate NR using Eq. 11-49.
dP 0.1× 10 −6 m
N= = −3
= 2.0 × 10 −4
dC 0.5 × 10 m
R
5. Calculate ηI using Eq. 11-46.
( ) ( 2.0 × 10−4 )
1/ 8 1.675
η
=I = 0.55 ( 33.64 ) 2.21 × 10 −2
0.55 ASNA1/8NR1.675 = 7.32 × 10 −6
6. Calculate Nvdw using Eq. 11-52.
=NvdW =
Ha
=
(10-20 J)
2.641
kB T (1.381×10-23J/K ) ( 274.15 K )
7. Calculate ηG with Eq. 11-47.
ηG =0.22 NR -0.24 NvdW 0.053 NG1.11
( ) ( 2.64 )0.053 ( 5.70×10-8 )

-0.24 1.11
= 0.22 2.0×10-4
= 1.63×10-8
8. Calculate Pe using Eq. 11-40.
3πµdP dC v
Pe =
kB T
=
( )(
3π 1.73 × 10 −3 kg/ms 0.1 × 10 −6 m 0.5 × 10 −3 m (10 m/h ) )( )
(1.381× 10 -23 2 2
)
kgm /s K ( 274.15 K )( 3600 s/h )
= 5.97 × 105
9. Calculate ηD with Eq. 11-48.

ηD =2.4 A1/3
S NR
-0.081
NvdW 0.052 Pe-0.715
( 2.0 × 10−4 ) ( 2.64 )0.052 ( 5.97 × 105 )

−0.081 −0.715
2.4 ( 33.64 )
1/ 3
=
= 1.20 × 10−3
10. Calculate η using Eq. 11-41.
( ) (
η = ηI + ηG + ηD = 7.32×10-6 + 1.62×10-8 + 1.2×10-3 =1.21×10-3) ( )
11. Calculate C/C0 using Eq. 11-36.
 −3 (1 − ε ) ηαL 
C = CO exp  
 2dc 
 −3 (1 − ε ) ηαL 
C / CO = exp  
 2dc 
C / CO =
exp
( )
 −3 (1 − 0.42 ) 1.21 × 10 −3 (1.0 )(1m ) 
  0.121
 (
2 0.5 × 10 −3 m ) 
12. Calculate log(C/C0).

log ( C / C0 ) = log ( 0.121) = −0.915
13. Set up a computation table and repeat steps 3 to 13 to determine particle removal
for temperatures between 1 to 25 °C and for particle diameters 1 µm and 10 µm.
The tabular results for 0.1 µm and the graphs for all 3 particle sizes follow.
FOR 0.1 um PARTICLES

Temp, ρW µ
NG NA Pe NR η C/CO log(C/CO)
°C
3
(kg/m ) (kg/m∙s)
1 999.84 1.73E-03 5.70E-08 2.21E-02 5.97E+05 2.00E-04 1.211E-03 0.121 -0.915
2 999.88 1.67E-03 5.87E-08 2.28E-02 5.76E+05 2.00E-04 1.243E-03 0.115 -0.939
3 999.92 1.62E-03 6.07E-08 2.36E-02 5.55E+05 2.00E-04 1.276E-03 0.109 -0.964
4 999.96 1.57E-03 6.26E-08 2.44E-02 5.36E+05 2.00E-04 1.308E-03 0.103 -0.988
5 1000.00 1.52E-03 6.46E-08 2.52E-02 5.17E+05 2.00E-04 1.341E-03 0.097 -1.013
10 999.70 1.31E-03 7.55E-08 2.92E-02 4.38E+05 2.00E-04 1.510E-03 0.072 -1.141
15 999.10 1.14E-03 8.77E-08 3.35E-02 3.75E+05 2.00E-04 1.684E-03 0.053 -1.273
20 998.20 1.00E-03 1.01E-07 3.81E-02 3.24E+05 2.00E-04 1.866E-03 0.039 -1.410
25 997.00 8.90E-04 1.17E-07 4.29E-02 2.83E+05 2.00E-04 2.054E-03 0.028 -1.552

FOR 1 um PARTICLES
Temp, ρW µ
°C
3
(kg/m ) (kg/m-s)
1 999.84 1.73E-03 5.70E-06 2.21E-04 5.97E+06 2.00E-03 3.889E-04 0.508 -0.294
2 999.88 1.67E-03 5.88E-06 2.28E-04 5.76E+06 2.00E-03 3.948E-04 0.503 -0.298
3 999.92 1.62E-03 6.07E-06 2.36E-04 5.55E+06 2.00E-03 4.009E-04 0.498 -0.303
4 999.96 1.57E-03 6.27E-06 2.44E-04 5.36E+06 2.00E-03 4.069E-04 0.493 -0.308
5 1000.00 1.52E-03 6.46E-06 2.52E-04 5.17E+06 2.00E-03 4.131E-04 0.487 -0.312
10 999.70 1.31E-03 7.55E-06 2.92E-04 4.38E+06 2.00E-03 4.440E-04 0.462 -0.336
15 999.10 1.14E-03 8.77E-06 3.35E-04 3.75E+06 2.00E-03 4.759E-04 0.437 -0.360
20 998.20 1.00E-03 1.01E-05 3.81E-04 3.24E+06 2.00E-03 5.087E-04 0.413 -0.384
25 997.00 8.90E-04 1.17E-05 4.29E-04 2.83E+06 2.00E-03 5.423E-04 0.389 -0.410
FOR 10 um PARTICLES
Temp, ρW µ
°C
3
(kg/m ) (kg/m-s)
1 999.84 1.73E-03 5.70E-04 2.21E-06 5.97E+07 2.00E-02 5.361E-03 0.00 -4.051
2 999.88 1.67E-03 5.88E-04 2.28E-06 5.76E+07 2.00E-02 5.388E-03 0.00 -4.072
3 999.92 1.62E-03 6.07E-04 2.36E-06 5.55E+07 2.00E-02 5.416E-03 0.00 -4.093
4 999.96 1.57E-03 6.27E-04 2.44E-06 5.36E+07 2.00E-02 5.444E-03 0.00 -4.114
5 1000.00 1.52E-03 6.46E-04 2.52E-06 5.17E+07 2.00E-02 5.471E-03 0.00 -4.134
10 999.70 1.31E-03 7.55E-04 2.92E-06 4.38E+07 2.00E-02 5.607E-03 0.00 -4.237
15 999.10 1.14E-03 8.77E-04 3.35E-06 3.75E+07 2.00E-02 5.741E-03 0.00 -4.338
20 998.20 1.00E-03 1.01E-03 3.81E-06 3.24E+07 2.00E-02 5.875E-03 0.00 -4.440
25 997.00 8.90E-04 1.17E-03 4.29E-06 2.83E+07 2.00E-02 6.010E-03 0.00 -4.542

Comments
Particle removal is a function of temperature, and removal increases as the
temperature increases. It is important for water treatment plant operators to be aware
that filtration is not as effective when the source water is cold as it is during summer
months when the source water is warm. Temperature appears more important for
smaller particles because the line representing the 0.1 µm particle size has a steeper
slope than the other sizes on this figure. Because the primary transport mechanism for
small particles is diffusion and for large particles is sedimentation or interception, it
appears that transport by diffusion is more dependent on temperature than transport by
sedimentation or interception. In summary: removal of the 10 µm particles is very good
over the entire temperature range; removal efficiency for the 0.1 µm size particles is
intermediate; and removal efficiency for the 1 µm size particles is the lowest.
PROBLEM 11-15
Problem Statement - Using the Rajagopalan and Tien filtration model, calculate and
plot the concentration profile of 4 µm particles (i.e., the size of Cryptosporidium
oocysts), through monodisperse filter with 0.5-mm-diameter media under filtration
conditions typical of rapid filtration (v = 10 m/h) and slow sand filtration (v = 0.1 m/h).
Assume porosity 0.40, particle density 2650 kg/m3, filter depth 1 m, and temperature 20º
C. Assume an attachment efficiency of 1.0 for the rapid filter and 0.05 for the slow sand
filter. Explain why rapid and slow sand filtration should be modeled with different values
for the attachment efficiency. Plot the results as C/C0 as a function of depth. Using
these results, comment on the methods used to restore the filtration capacity of slow
sand and rapid filters (i.e., scraping vs. backwashing).
Instructors Note: A plot of concentration as a function of depth through the filter is

required to solve this problem. The approach used to prepare this graph is to divide the
filter into increments of depth and calculate the particle concentration at each location.
In this solution, 20 increments of 0.05 m each have been used, but the selection of the
number of increments is up to the student. The calculations are shown for the filtration

rate of 10 m/h, and must be repeated for the filtration rate of 0.1 m/h. Values of ρW and
µ are available in Table C-1 in Appendix C.
As noted on page 767 of the text, the calculation of the Hamaker number is beyond the
scope of this text, but the value ranges from 10-19 to 10-20 J. A value of 10-20 J (1 x 10-20
kg·m2/s2) is being used in this solution.
Solution
1. Calculate γ using Eq. 11-53.
(1 − ε ) (1 − 0.40 )
1/ 3 1/ 3
γ= = = 0.8434
2. Calculate AS using Eq. 11-54.
( )
2 1 − ( 0.8434 ) 
5
2 1− γ5  
=AS = = 37.98
2 − 3 γ + 3 γ 5 − 2γ 6 2 − 3 ( 0.8434 ) + 3 ( 0.8434 )5 − 2 ( 0.8434 )6
3. Calculate NA using Eq. 11-51. A Hamaker Number of 10-20 J (1 x 10-20 kg·m2/s2) is

being used in this solution.
=
NA
Ha
=
(
1× 10 −20 kgm2 /s2 ( 3600 s/h ) )
= 2.38 × 10 −5
( )(
3πµdP2 v 3π 1.00 × 10 −3 kg/ms 4 × 10 −6 m 2 (10 m/h ) )
4. Calculate NR using Eq. 11-49.
dP 4 × 10 −6 m
N= = −3
= 8.0 × 10 −3
dC 0.5 × 10 m
R
5. Calculate ηI with Eq. 11-43.

1/8 1/8
4  4 
ηI A S  NA  NR15/8 = 37.98  2.39×10-5 
3  3 
(8.0×10 )
-3 15/8
= 1.22 × 10 −3
6. Calculate NG using Eq. 11-38. From Table C-1 in Appendix C at 20 °C, ρW = 998
kg/m3 and µ = 1.002 x 10-3 kg/m·s.

g ( ρP − ρW ) dP2
NG =
18µv F
( 9.81 m/s )( 2650 − 998 kg/m )( 4 × 10 m ) (3600 s/h )

2
2 3 −6
=
18 (1.002 × 10 kg/ms ) (10 m/h )
−3
= 5.17 × 10 −3
7. Calculate ηG with Eq. 11-44.
0.00338A SNR -0.4NG1.2 = 0.00338 ( 37.98 ) ( 8.0×10-3 ) (5.17×10 )

-0.4 -3 1.2
ηG = = 1.60×10-3
8. Calculate Pe using Eq. 11-40.

3πµdP dC v
Pe =
kBT
=
( )(
3π 1.002 × 10 −3 kg/ms 4 × 10 −6 m 0.5 × 10 −3 m (10 m/h ) )( )
(1.381× 10 -23
)
kgm2 /s2K ( 273.15 K )( 3600 s/h )
= 1.30 × 107
9. Calculate ηD with Eq. 11-45.
ηD =4A S1/3Pe-2/3 = 4 ( 37.98 )

1/3
(1.30×10 ) 7 -2/3
= 2.44×10-4
10. Calculate η using Eq. 11-41.
η = ηI + ηG + ηD = 1.22×10-3 +1.60×10-3 +2.44×10-4 = 3.06×10-3

11. Calculate C using Eq. 11-36. The initial concentration at the top of the bed is
taken as C0 = 1.0. (In this solution, the depth of the bed has been divided into 20
increments of 0.05 m each.)
 −3(1 − ε)ηαL 
C = CO e  
 2dc 
( )
 −3 (1 − 0.42 ) 3.06 × 10 −3 (1.0 )( 0.05m) ) 
C (=
1.0 ) e   0.759
 (
2 0.5 × 10 −3 m  )
12. Repeat step 11 for each increment of depth. The value of C calculated in each
increment is taken as C0 for the next increment.
13. Repeat Steps 3 through 12 for the filtration rate of 0.1 m/h and α = 0.05.
14. The results of the calculations are summarized in the following table, and shown in
the following graph.
Rapid filter particle conc. Slow sand filter particle conc.
Depth from top of filter, m (v = 10 m/h, α = 1.0) (v = 0.1 m/h α = 0.05)
0.00 1.00000 1.000
-1
0.05 0.75922 1.586 x 10
-2
0.10 0.57641 2.517 x 10
-3
0.15 0.43762 3.993 x 10
-4
0.20 0.33225 6.335 x 10
-4
0.25 0.25225 1.005 x 10
-5
0.30 0.19151 1.594 x 10
-6
0.35 0.14540 2.529 x 10
-7
0.40 0.11039 4.013 x 10
-8
0.45 0.08381 6.366 x 10
-8
0.50 0.06363 1.010 x 10
-9
0.55 0.04831 1.602 x 10
-10
0.60 0.03668 2.542 x 10
-11
0.65 0.02785 4.033 x 10
-12
0.70 0.02114 6.398 x 10
-12
0.75 0.01605 1.015 x 10
-13
0.80 0.01219 1.610 x 10
-14
0.85 0.00925 2.555 x 10
-15
0.90 0.00702 4.053 x 10
-16
0.95 0.00533 6.430 x 10
-16
1.00 0.00405 1.020 x 10

Filtrate quality (C/Co)
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
0.0
0.1
0.2
0.3
0.4
Depth, m
0.5
0.6
0.7
0.8
Rapid filter (v = 10 m/h, alpha = 1.0)
0.9 Slow sand filter (v = 0.1 m/h, alpha = 0.05)
1.0
Comment
Rapid filters and slow sand filters should be modeled with different values of attachment
efficiency (α). Because the feed water for rapid filters is coagulated, the particles are
destabilized and the attachment efficiency is high (α = 1.0). Whereas the feed water for
slow sand filters is not coagulated, the particles are not destabilized and the attachment
efficiency is low (α = 0.05 in this problem).
Even with a substantial difference in attachment efficiency, most of the particles are
captured in the top 0.05 m of the bed in a slow sand filter (which leads to straining) and
particles are captured throughout the depth of the rapid filter (depth filtration), as shown
in this problem solution. Note, this model is a clean bed filtration model, and the initial
capture of particles at the top of the slow sand filter bed reduces the porosity of the top
of the bed and leads to the development of a straining mechanism for smaller particles
as the filter cycle progresses. Based on these results, scraping is an acceptable
method of removing the particles from a slow sand filter because the particles are all at
the top of the bed, but not for a rapid filter. For a rapid filter backwashing is necessary
to remove the particles from the filter depth.
PROBLEM 11-16
Problem Statement - The results of pilot filter experiments are summarized in the
tables below. For each set of experiments, the independent variable was either
the media effective size or the media depth, as given in the second column below.
For a given set of experiments (A, B, C, or D, to be selected by instructor),
determine equations for how the specific deposit at breakthrough (σB) and the
head loss rate constant (kHL) each depend on the independent variable. Also
determine the optimal value of the independent variable and the corresponding
filter run duration. For all problems, assume C0 = 2.0 mg/L and CE = 0 mg/L.
A. Design conditions: vF = 15 m/h, media = anthracite, depth = 1.75 m, max

available head = 2.8 m.
A Media Time to Initial Head Loss When

ES, Breakthrough, Head Loss, Breakthrough
Filter mm h m Occurred, m
1 0.8 112 0.65 4.6
2 1.0 85 0.39 2.9
3 1.1 72 0.33 2.4
4 1.2 71 0.30 2.0
5 1.4 58 0.24 1.5
B. Design conditions: vF = 15 m/h, media = GAC, depth = 2.0 m, max available

head = 3.0 m.
B Media Time to Initial Head Loss When

ES, Breakthrough, Head Loss, Breakthrough
Filter mm h m Occurred, m
1 0.83 54 0.65 6.1
2 1.05 43 0.40 4.3
3 1.25 38 0.33 2.9
4 1.54 32 0.22 2.0
C. Design conditions: vF = 33.8 m/h, media = anthracite, ES = 1.55 mm, max

available head = 3 m (adapted from pilot results for the LADWP Aqueduct
Filtration Plant).
C Media Time to Initial Head Loss When
Depth, Breakthrough, HeadLoss, Breakthrough
Filter m h m Occurred, m
1 0.6 4.0 0.16 1.0
2 1.0 6.7 0.30 1.7
3 1.8 11.9 0.50 3.2
4 2.0 13.4 0.58 3.6
5 2.2 14.5 0.65 4.1
D. Design conditions: vF = 25 m/h, media = anthracite, ES = 1.50 mm, max available

head = 3 m (adapted from pilot results for the Bull Run water supply)
D Media Time to Initial Head Loss When

Depth, Breakthrough, Head Loss, Breakthrough
Filter m h m Occurred, m
1 2.0 41 0.43 1.8
2 2.3 49 0.51 2.0
3 2.5 55 0.51 2.5
4 3.0 65 0.63 2.9
Instructors Note: This solution is completely developed for set of experiments A. The
solutions to the other set of experiments are summarized at the end of the solution.
Solution for Experiment A

Part 1: Equation for specific deposit at breakthrough
1. The relationship between the effective size of the media and the specific deposit at
breakthrough is generally a non-linear relationship, which can usually be described
by a power function with the following form:
k ( d)
n
σB =
To find the value of the coefficients k and n, take the log of this equation:
log ( σ=
B) log ( k ) + nlog ( d)

When log(σB) is plotted against log(d), the slope of the straight line is n and the
intercept is log(k).
2. Calculate the specific deposit at breakthrough using Eq. 11-57 for experiment 1.
v ( C0 − CE ) tB (15 m/h)( 2.0 − 0 mg/L )(112 h)
=σB = = 1920 mg/L
L 1.75 m
3. Calculate log(σB) for experiment 1.
=
log ( σB ) log(1920)
= 3.28
4. Calculate log(d) for experiment 1.

log ( d ) = log(0.8) = −0.097
5. Repeat steps 2 through 4 for the remaining experiments. The results are
summarized in the following table.
Experiment σB, mg/L log(σB) log(d)
1 1920 3.28 -0.097
2 1457 3.16 0.000
3 1234 3.09 0.041
4 1217 3.09 0.079
5 994 3.00 0.146
6. Plot log(σB) against log(d).
3.30
3.25
3.20
y = -1.16x + 3.16
3.15 R² = 0.98
log(σB)
3.10
3.05
3.00
2.95
-0.2 -0.1 0.0 0.1 0.2
log(d)

7. Perform a linear regression of the data and determine the slope and intercept of
the regression line. A regression analysis of the data was performed using the
trendline function in ExcelÒ. From the graph in step 6, n = -1.16 and log(k) = 3.16.
Therefore, k = 1445.44. Thus, the relationship between σB and d is:
1445.44 ( d )
−1.16
σB =
when the units of σB are mg/L and the units of d are m.

Part 2. Equation for head loss rate constant.
1. Calculate the head loss increase rate constant using Eq. 11-64 for experiment 1.
The specific deposit is the value calculated in Part a because the given value for
head loss corresponds to the time at breakthrough.
=kHL
(hL,B − hL,0 ) L
=
( 4.6 − 0.65 =
m ) 1.75 m
0.0021 Lm/mg
νF ( c O − CE ) t 15 m/h ( 2.0-0 mg/L ) 112 h
2. Calculate log(kHL) for experiment 1.

log (kHL ) = log ( 0.0021) = −2.69
3. Repeat steps 1 and 2 for the remaining experiments. The results are summarized
in the following table. For convenience, log(d) from Part a is also repeated in the
table.
Experiment kHL, L·m/mg log(kHL) log(d) (from Part a)
1 0.0021 -2.69 -0.097

2 0.0017 -2.76 0.000
3 0.0016 -2.78 0.041
4 0.0014 -2.85 0.079
5 0.0013 -2.90 0.146
4. Plot log(kHL) against log(d).

5. Perform a linear regression of the data and determine the slope and intercept of
the regression line. A regression analysis of the data was performed using the
trendline function in ExcelÒ. From the above graph, n = -0.89 and log(k) = -
2.7655. Therefore, k = 1.72 x 10-3. Thus, the relationship between kHL and d is:
kHL 1.72 × 10 −3 ( d )
−0.89
=
when the units of kHL are L·m/mg and the units of d are m.
Part 3. Optimal media effective size.
1. Develop an equation for the time to breakthrough using Eq. 11-62 and the
expression for σB developed in step 7 of Part 1.
1445.44 ( d ) 1445.44 (1.75 m )

−1.16
σBL L
( d)
−1.16
=tB = =
v ( C0 − CE ) v ( C0 − CE ) (15 m/h)( 2.0 − 0 mg/L )
tB = 84.32 ( d )
−1.16
2. Develop an equation for the time to reach limiting head loss using Eq. 11-65 and
the expression for kHL developed in step 5 of Part 2.

=tHL =
(HT − hL,0 ) L (HT − hL,0 ) L
kHL v ( C0 − CE ) 1.72 × 10 −3 ( d )−0.89 v ( C0 − CE )
( 2.8 m − 0.65 m )(1.75 m )

( d)
−0.89
=
(1.72 × 10 ) (15 m/h)( 2.0 − 0 mg/L )
−3
= 72.9 ( d )
−0.89
3. Calculate the time to breakthrough and time to reach limiting head for each of the
media effective sizes listed in the problem statement. Clean bed head loss values
can be taken from the problem statement. The results are summarized in the
table below.
Experiment d, mm hL,0, m tB, h tHL, h
1 0.8 0.65 109.2 60.7
2 1.0 0.39 84.3 82.8
3 1.1 0.33 75.5 92.3
4 1.2 0.30 68.2 100.8
5 1.4 0.24 57.1 118.2
4. Plot tB and tHL.

5. From the graph in step 4, the optimal effective size is 1.04 mm, which will result in
a maximum filter run length of 84 hours.
Solution to the other experiment sets
Note that a variety of equation forms can be selected. Students may select different
forms for the equation in order to achieve a linear regression. In the solutions given
here, equations are of the form y = mx + b for experiments C and D.
Equation for Equation for Optimal
Experiment
specific deposit head loss rate constant condition
ES = 1.25 mm,
691.83 ( d ) k HL 5.82 × 10 −3 ( d )
−0.84 −1.01
B σB = =
t = 38 h
L = 1.7 m,
C σB =−2.45 (L ) + 453.5
= k HL 0.00362 (L ) − 0.000422
t = 11.2 h
L = 3.0 m,
D =σB 56.23 (L ) + 930.6
= k HL 0.000814 (L ) − 0.000335
t = 67 h
PROBLEM 11-17
Problem Statement - For the filter design selected in Problem 11-16, what is the UFRV
at the design condition? If the filters are designed to be backwashed at 40 m/h for
15 min, what are the UBWV and recovery, assuming there is no filter-to-waste
period?
Solution for Experiment A

1. Calculate UFRV using Eq. 11-66.
= vt
UFRV = F (15 m/h)( 84 =
h) 1260 m3 / m2
2. Calculate UBWV using Eq. 11-67
= v=
UBWV
( 40 m/h)(15=
min )
37.5 m3 /m2
BW t BW
60 min/h
3. Calculate recovery using Eq. 11-69.
UFRV − UBWV 1260 − 37.5
=r = =× 100 97%
UFRV 1260

Solution for other experiments
Experiment UFRV UFBV Recovery

B 570 37.5 93%
C 379 37.5 90%
D 1675 37.5 98%
PROBLEM 11-18
Problem Statement - A filter has been designed to have a run length of 48 h while
operating at a filtration rate of 12 m/h. The design backwash rate is 40 m/h, backwash
duration is 15 min, and filter-to-waste duration is 10 min. The plant operator decides to
clean the filters more thoroughly and backwashes at a rate of 55 m/h for 25 min. As a
result, ripening takes longer and the filter-to-waste duration is 50 min. Calculate the
UFRV, UBWV, UFWV, and recovery (a) as designed and (b) as operated. What is the
percent increase in the volume of treated water lost as waste backwash water and filter-
to-waste water?
Solution Part a.
= vt
UFRV = F (12 m/h )( 48 =
h) 576 m
2. Calculate UBWV using Eq. 11-67.
= v=
UBWV
( 40 m/h )(15=
min )
BW t BW 10 m
60 min/h
3. Calculate UFWV using Eq. 11-68.
= v=
UFWV
(12 m/h )(10=
min )
BW t BW 2m
60 min/h
UFRV − UBWV − UFWV 576 − 10 − 2 m
=r = = × 100 97.9%
UFRV 576 m

Solution Part b.
= vt
UFRV = F (12 m/h )( 48 =
h) 576 m
2. Calculate UBWV using Eq. 11-67.
= v=
UBWV
( 55 m/h )( 25=
min )
BW t BW 22.9 m
60 min/h
3. Calculate UFWV using Eq. 11-68.
= v=
UFWV
(12 m/h )( 50=
min )
BW t BW 10 m
60 min/h
UFRV − UBWV − UFWV 576 − 22.9 − 10 m
=r = = × 100 94.3%
UFRV 576 m
Solution Part c.
As designed, 2.1 percent of the treated water was lost as waste either as backwash
water or filter-to-waste. As operated, 5.7 percent of the treated water was lost as waste.
This represents a 170 percent increase in the water lost as waste.
PROBLEM 11-19
Problem Statement - Explain (a) the importance of flow control in proper filtration
operation and (b) the main types of flow control systems.
Solution
a. Flow control systems are important in filtration for several reasons. They are used
to apportion the flow between the filters in a plant, they control the filtration rate,
and they provide a mechanism for accommodating the increase in head loss that
occurs as filters capture solids. An important aspect of flow control systems is that
they prevent rapid changes in flowrate through a filter. This is important because
rapid changes can lead to detachment of previously-retained particles, a
degradation of filtrate quality, and a reduction in the length of filter runs.
b. The main types of flow control systems are: (1) modulating control valve to control
level, filtration rate, and flow distribution, (2) influent weir flow splitting with
constant water level, where the level is controlled by a level transmitter and an
effluent valve, (3) influent weir flow splitting with rising water level, and (4)
declining rate filtration.
PROBLEM 11-20
Problem Statement - Discuss (a) the impact of rapid variations in filtration rate on filter
performance, (b) causes of rapid variations in filtration rate, and (c) design features and
operational methods for preventing rapid variations in filtration rate.
Solution
a. The primary impact of rapid changes in filtration rate is detachment of previously
collected particles (see page 780 in the text), leading to a decrease in filter
performance and an increase in effluent turbidity, which leads to shorter filter runs
or possibly breakthrough.
b. A number of factors can cause rapid changes in filtration rate. The filtration rate to
a filter changes when another filter is taken out of service for backwashing. The
total flow to the filters does not change, but now there is one less filter to receive
the flow, meaning the remaining operational filters see an increase in influent flow,
and thus filtration rate (see page 784). Filters experience a step increase in flow
when a filter is taken out of service for this reason, and a step decrease in flow
when a filter is placed back in service.
The filtration rate can fluctuate depending on the method of flow control. This is
particularly evident when the control method uses effluent modulating control
valves that are not capable of fine position adjustments (see page 788). These
valves can repeatedly overshoot the desired position as they continually “hunt” for
the correct filtration rate, causing rapid and undesirable variations in flow rate.
Although not covered in this chapter, filters are affected by changes in plant flow
rate, such as adjustments in production to meet the demand or, more importantly,
when plants are operated intermittently (e.g., operated only 8 hours per day) and
the flow is fully stopped and started on a regular basis.

c. Design features that can prevent rapid changes in filtration rate include proper flow
control. It is necessary to have a sufficient number of filters so that the filtration
rate does not increase by an excessive amount when one filter is taken out of
service for backwashing. The accepted minimum number of filters for this purpose
is four, but more filters are better.
Proper flow control is critical. Influent weir flow splitting provides the most gradual
changes when filters are backwashed, particularly if the influent channel is large
enough that it will take several minutes for the water level to rise to the new level
when a filter is taken out of service. Effluent modulating control valves, if used,
should have fine position adjustment and the instrument providing the signal to the
valve (the level transmitter or effluent flowmeter) should be sufficiently precise and
set to minimize valve hunting.
Correct plant operation is important for minimizing variations in filtration rate.

Plants should not be operated intermittently, or if they are, all filters should be
backwashed as the plant is started. In addition, changes in plant production
should be made infrequently and in small steps rather than large, frequent steps.
Changes in plant production can be minimized by proper plant design, which
includes sufficient equalization volume in the plant and in the distribution system.
PROBLEM 11-21
Problem Statement - Discuss factors that influence the selection of the number of
filters in a treatment plant design.
Solution
The factors that affect the number of filters are discussed on page 784. These factors
include the overall plant capacity, the largest dimensions of a single filter, the effect of
backwashing on filtration rate, and economics. The minimum number of filters to avoid
problems during backwashing is typically four. The maximum number of filters is
typically determined by dividing the plant capacity by the flow to each filter, where the
flow to each filter is determined by the filtration rate and the largest dimensions of a
single filter, which is typically about 100 m2 (1100 ft2). Economics (e.g. construction
costs) generally leads to the selection of a small number of larger filters, rather than a
large number of small filters.
PROBLEM 11-22
Problem Statement - Discuss the benefits and salient features of air scour.
Solution
Air scour is discussed on page 791 in the text. Air scour is used as a supplemental
cleaning mechanism during backwashing. Typically, the water level in the filter box is
lowered to just above the filter media before air scour is initiated, and then air is
introduced into the underdrains as backwash water is introduced at a low rate. The
backwash water is applied at a sub-fluidization rate, and the air forms cavities in the
media that subsequently collapse, called collapse pulsing. This action causes a great
deal of agitation in the bed, which scours solids from the media. The primary benefit of
air scour is to provide effective backwashing and cleaning of media, particularly for deep
bed filters.


CH 11

Uploaded by

Copyright:

Available Formats

CH 11

Uploaded by

Document Information

Original Description:

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

CH 11

Uploaded by

Copyright:

Available Formats

MWH’S WATER TREATMENT

PRINCIPLES AND DESIGN

by Crittenden, Trussell, Hand, Howe, and Tchobanoglous

HOMEWORK SOLUTION MANUAL

Homework Solution Manual Page 2 of 48

Homework Solution Manual Page 3 of 48

Homework Solution Manual Page 4 of 48

Problem Statement - A filter is designed with the following specifications. The

Homework Solution Manual Page 6 of 48

( 228 ) (1 − 0.50 )2 (1.307 × 10-3 kg/ms ) (1.5 m )(15 m/h )

( 4.4 ) (1 − 0.50 )(1.5 m )(15 m/h )2

4. Add the two terms together.

Homework Solution Manual Page 7 of 48

For problems D and E:

Homework Solution Manual Page 9 of 48

(112 ) (1 − 0.42 )2 (1.14 × 10-3 kg/ms ) (10 m/h )

c. Calculate the permeability coefficient for the square term.

Homework Solution Manual Page 10 of 48

2. Calculate the average size of media in the first layer.

3. Calculate the fraction of media in the first layer.

Sieve Weight Average Fraction of Depth of

Homework Solution Manual Page 11 of 48

0.045 m − ( 0.004 m + 0.006 m + 0.015 m ) =

3. Calculate the thickness remaining in that layer.

Homework Solution Manual Page 12 of 48

Total 705 0.855 0.072

Solution to the remaining problem statements:

2.00 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0004 0.0004

1.70 0.0007 0.0007 0.0000 0.0000 0.0000 0.0000 0.0032 0.0032

1.40 0.0280 0.0280 0.0000 0.0000 0.0003 0.0003 0.0171 0.0171

1.18 0.0304 0.0304 0.0000 0.0000 0.0019 0.0019 0.0366 0.0366

1.00 0.0201 0.0201 0.0000 0.0000 0.0053 0.0053 0.0318 0.0318

0.85 0.0131 0.0131 0.0050 0.0050 0.0166 0.0166 0.0081 0.0028

0.71 0.0086 0.0010 0.0652 0.0652 0.0877 0.0877 0.0045 0.0000

0.60 0.0065 0.0000 0.1372 0.1372 0.1167 0.1167 0.0012 0.0000

0.50 0.0013 0.0000 0.0896 0.0896 0.0735 0.0699 0.0006 0.0000

0.425 0.0000 0.0000 0.0592 0.0592 0.0179 0.0000 0.0000 0.0000

0.355 0.0000 0.0000 0.0571 0.0350 0.0095 0.0000 0.0000 0.0000

0.300 0.0000 0.0000 0.0066 0.0000 0.0051 0.0000 0.0000 0.0000

Pan 0.0000 0.0000 0.3464 0.0000 0.0709 0.0000 0.0000 0.0000

Total 0.109 0.093 0.766 0.391 0.405 0.298 0.104 0.092

Homework Solution Manual Page 13 of 48

(112 ) (1 − 0.42 )2 (1.14 × 10-3 kg/ms ) (1.0 m )(15 m/h )

( 2.2 ) (1 − 0.42 )(1.0 m )(15 m/h )2

Homework Solution Manual Page 14 of 48

( 2.2 ) (1 − 0.42 )(1.0 m )( 0.15 m/h )2

Homework Solution Manual Page 16 of 48

Homework Solution Manual Page 17 of 48

Homework Solution Manual Page 18 of 48

(1.002 × 10 kg/ms ) (10 mm/m )

(1.002 × 10 kg/ms ) (10 mm/m )

c. Calculate velocity using Eq. 11-8.

Homework Solution Manual Page 19 of 48

Homework Solution Manual Page 20 of 48

Problem Statement - A monomedia anthracite filter is designed with the following

a. Calculate backwash rate to get a 25 percent expansion at the design summer

Homework Solution Manual Page 21 of 48

4. Calculate v using Eq. 11-8.