- Email: [email protected]
Design options for self-cleansing storm sewers
8)
Pergamon
Waf. Sci. T~ch. Vol. 33. No.9. pp. 215-220.1996. Copyright C 1996IAWQ. Published by Elsevier Science Ltd Printed In Oreat Britain. All righls reserved. 0213-1223196 $\ HlO + 0-00
PH: S0273-1223(96)OO389-7
DESIGN OPTIONS FOR SELF-CLEANSING STORM SEWERS Chandramouli Nalluri* and Aminuddin Ab. Ghani** • Department of Civil Engineering. University ofNewcastle upon Tyne, Newcastle upon Tyne NEJ 7RU, UK •• School of Civil Engineering, Universiti Sains Malaysia, Perak Branch Campus, 31750 Tronoh. Perak, Malaysia
ABSTRAcr A list of available codes of practice for self-cleansing sewers is presented and a review of appraisals of minimum velocity criterion is summarised. Compansons of newly developed "minimum velocity" criteria and "minimum shear stress" criterion are presented. Some design charts are also given. These charts are applicable to non-cohesive sediments (typically storm sewers). It appears that sediment size and concentration need to be taken into account, and that a limited depth of sediment bed is recommended for large pipes (diameters> 1000 mm) to maximise their transpon capacity. Copyright © 1996 IAWQ. Published by Elsevier Science Ltd.
KEYWORDS Limit deposition; sediment transport; self cleansing; sewer design. INTRODUcnON Old sewerage systems were designed based on empirical rules to minimise sediment problems and these design rules are still in practice. Either minimum velocity or minimum shear which may not allow any deposition at any time or at least over a long period ohime no deposits would build-up (i.e. high flows flush the deposits) govern these design procedures. In the present trend. "no deposit condition" (clean pipe) in each individual pipe is maintained by allowing flushing to take place fairly frequently and minimum velocity to be achieved under full or half-full conditions. Table 1 reproduces some of the available design criteria given in CIRIA (1987). Ab. Ghani (1993) presented a new design criterion for clean pipe (no deposition) and another one for a pipe having loose deposited sediment bed. A summary of the development of the criteria and ranges of data used is given elsewhere (see Nalluri and Ab. Ghani 1994 a,b). Table 2 gives the newly derived criteria; equations 1 and 2 are applicable for storm sewers only (i.e. with non-cohesive sediments). APPRAISALS OF "MINIMUM VELOCITY" CRITERION
=
Ab. Ghani (1993) appraised the "minimum velocity" criterion (half-full flow, clean pipe and V 0.75 mls) used by Ackers (1984) and Nalluri (1986) and in general concluded and confirmed that the velocity criterion 21S
C. NALLURI and A. AB. GHANI
216
overdesigns the slope for small pipe diameters (D < 500 nun) and underdesigns the slope for larger pipe diameters. Table la. Minimum velocity criterion
Source
Country
American Society of Civil Engineering (1970)
USA
British Standard (1987)
UK
Bielecki (1982)
Germany
Sewer Type
Minimum velocity (mfs)
Pipe Conditions
Foul Storm
0.6 0.9
Fuillhalf-full Full/half-full
Storm Combined
0.75 1.0
Full Full
1.5
Full
Table I b. Minimum shear stress criterion Minimum Shear stress (N/m 2) 3.0 - 4.0 1.0 - 2.0
Reference Source
Country
Sewer Type
Yao (1914)
USA
Storm Foul
Maguire (- )
UK
6.2
Bischof (1976)
Germany
2.5
Pipe Conditions
Fuillhalf full
Table 2. New design criteria for self-cleansing sewers Equations
Transport Mode
(1a)
Clean Pipes
(1b)
-;::.;:;V;::::;:;: Jgd(SI-I) Pipes with deposited loose beds
= 1.18~.16 (WYob)-o·11 (~)-o.34
11 = 0.0014<:.-o.G4
y: )0.,. (~)0.24
w (
1;,.31 (2a)
D;.u (2b)
Comparisons of Ab. Ghani (1993) equations (eqns la and Ib) and "minimum velocity" criterion of 0.75 rnls for half-full flow in a clean pipe (Nalluri and Ab. Ghani 1994a) confirm the inadequacy of the present practice to minimise sediment deposition in pipe sewers. The results also show that the presence of a limited depth of sediment deposit reduces the slope requirement over the entire range of pipe diameters.
217
Self·cleansing Slonn sewers
It was then suggested that sewers of diameter 1 m or more may be best designed with some sediment bed deposition and there exists an optimum depth of sediment deposition in a sewer with given flow and transport conditions (Nalluri & Ab. Ghani. 1994b). APPRAISALS OF "MINIMUM SHEAR STRESS" CRITERION Figures I and 2 show the Q-So-O plots using the new criteria (see Table 2) for comparisons with "minimum shear stress" methods; also shown on these figures is the minimum velocity criterion. Figure I demonstrates that a constant shear stress of 4.0 N/m2 in the case of a clean pipe design (eqns. la & b) would result in overdesigning its slope requirement for volumetric sediment concentration of 20 ppm over the range of diameters up to 1.0 m. However. as the level of sediment concentration increases. the constant shear stress method. tends to underdesign the required slope especially for around 0 > 500 mm. When the sediment concentration reaches 1000 ppm. the constant shear stress method underdesigns the slope requirement over the entire range of pipe diameters; it may be observed that a shear stress of I N/m 2 is grossly inadequate over the whole range of pipe sizes.
Cv C. ~ppmlppm = 20 100 Cv ppm - 1000
10
*-•
1.0
4.0
+-+-+-+-+
0.75
-- V
-I
10 ..·-t-....................,,.-..............,,,...-.--rTTTn....1-r-.......""'"
1
10
10' (I/s)
10
10·
Q
Figure I. Q-So-D plot: clean pipe (half-full flow, dso
=I.Omm, ko =O.6mm).
Figure 2 highlights the level of slope requirements using the constant shear stress methods and new design criteria with loose bed deposits (eqns. 2a & b) for large pipes (0 = 1.0 m to 4.0 m). For a given flow condition (Le. flow depth). an optimum depth of sediment could be present over a range of sediment concentrations as illustrated in Figure 3. This is in agreement with the results from the experimental work by Ab. Ghani (1993) and May (1993). In these experiments. the sediment bed had been kept constant over a range of velocities for a constant flow depth. Figure 2 also shows that even with the presence of an optimum bed. the slope requirements are still larger than those obtained by the constant shear stress method. Both figures I and 2 demonstrate the importance of including sediment concentration as an input in designing the slope requirements for the movement of sediments in sewers. Examples of design charts taking into account sediment concentration are shown in figures 4 and 5.
C. NALLURJ and A. AB. GHANI
218
c. c c. . .......
10-' .....
- 20 = 100
~ppm~ ppm ppm
....... _ ..... 1000 - - V
-
/
i"
"
0 ....
.
....
f,
<'
10 ...
/m N/m: mi.)
l
1.0 4.0 0.75
/;/
.....~,/ ,...... . .~, t:/. '
vr
t
..;
fir
/
,
-.
"...
v
.
I
I
j
... •
I
"J/"
~~
10
10
10 • Q
10 •
10·
(1/,)
10 •
Figure 2. Q-So-D plot: pipes with loose bed (half-full flow. d SO = I.Omm. Y/D=
0.001'
QAQAOO •
1
I~%).
1ft
_0-2'" ,.....,D.31ft
uauD.4", UA.lI D • , "'
.,
Q.
o
Vi
o.OOQe8~.lllI~~""':8".I:O:8~"""~8~.~28~""'"8~.T'3II""''''''~.''''.48
yolO Figure 3a. Optimum sediment depth: C y
=20 ppm (half-full flow. d SO = I.Omm).
0.0030
QAQAOD. , '"
:::::8:1::: .......,0 • • 1ft
UAUO - 5 '"
0.0025
.,
Q.
0
Vi
0.0020
~
~~~ .. ..1. ..
0.0015
0.0010
•• 48
yolO
Figure 3b. Optimum sediment depth: C y
=100 ppm (half-full flow,dSO. I.Omm).
Self-cleansing stonn sewers
o.ooeo
219
C>l>D - 1 m IUUUlD 0 • 2 m
~D·Jm
u.a.a.;D-4m u..A..AJD.' m
O.OO~O
.
Q.
~
0.0040
0
iii
0.0030
Y./D
Figure 3c. OptImum sediment depth: Cv = 1000 ppm (half-full flow. dSO = I.Omm).
0.0120
c, !ppmj -co ppm -..... c, ppm -
0.0100
O.DOGO
--._---_.
0.0040
-
0.D020
0.0000
1000
~--.
0.0080
co a. 0 iii
20 100
0.0
----..
0.2
o.e
0.4
0.8
1.0
D (m)
=
Figure 4. Design chart for clean pipes (half-full flow. dS O I.Omm. ko = O.6mm).
0.0040
c, lppml-co ppm ppm --
. . . . . C,
20 100 1000
0.D030
"
0.0020
Q.
.Q
..
Ul
O.OOtO
O.OOOD
1.0
2.0
3.0
4.
5.0
D (m)
Figure S. Design chart for pipes with loose bed (half-full flow. d SO
="Omm. y/D =IS%).
C. NALLURI and A. AB. GHANI
220
CONCLUSIONS The use of present design criteria (Table 1) to minimise sediment depositions in sewers for large pipes (D > 500 mm) is questionable. The new design criteria (equations I and 2) should be used in designing storm sewers carrying non-cohesive sediments. Equation 1 is recommended for clean pipes with pipe diameter up to 1 m (Figure 4). For larger pipe diameters (D > 1 m), equation 2 which allows a limited depth of sediment bed (Figure 3) is recommended (Figure 5). A reduction factor should be applied if the new criteria (equations 1 and 2) were to be used for designing foul sewers carrying cohesive sediments. Table 1 shows that the cleansing velocity or shear stress for designing foul sewers is smaller than the one for designing storm sewer. It has been shown (Figures 1 and 2) that sediment size and concentration are important factors to be taken into account in minimising sediment deposition problems. It should also be mentioned that equations 1 and 2 are able to simulate variations in flow conditions in real sewers (see Nalluri and Ab. Ghani, 1994a). NOTATION
Cv
d,dso D Dgr ko Q R
So Ss
V Wb
Yo Ys 1.0
As
'to
Volumetric sediment concentration Mean sediment size Pipe diameter Dimensionless sediment size (= d (Ss-1 )glv2) 1/3) Pipe roughness height Flow discharge Hydraulic radius Pipe slope Relative density of sediment Self-cleansing velocity Width of sediment bed Mean flow depth Mean sediment bed thickness Clear-water friction factor Friction factor with sediment Mean shear stress REFERENCES
Ab. Ghani. A (1993). Sediment Transport in Sewers. University of Newcastle upon Tyne. UK, PhD Thesis. Ackers. P. (1984). Sediment Transport in Sewers and the Design Implications. Intern. Conf. on Planning. Construction, . Maintenance. and Operation of Sewerage Systems. Reading. England. CIRIA (1987). Sediment movement in combined sewerage and stormwaterdrainage systems. London. England. Nalluri. C. (1986). Sediment Transport in RigId Boundary Channels. Proc. Euromech 192: Transport ofSuspended Solids in Open Channels. A A Balkema, Rotterdam. Nalluri. C. and Ab. Ghani. A. (l994a). Sediment Transport in Sewers and Design Implications. Proceedings. National Conference on Hydraulic Engineering. ASCE Vol 2. pp. 933-938. Buffalo. New York. USA, 1·5 August. Nalluri. C. and Ab. Ghani. A. (1994b). Sediment Transport i/l Sewers with and without Deposited Beds. Proceedings. 9th Congress of APD-IAHR Vol 2. pp. 27-34. Singapore. 22-24 August. May. R. W. P. (1993). Sediment Transport in Pipes and Sewers with Deposited Beds. Report SR 320. Hydraulics Research Ltd, Wallingford. UK.
Pergamon
Waf. Sci. T~ch. Vol. 33. No.9. pp. 215-220.1996. Copyright C 1996IAWQ. Published by Elsevier Science Ltd Printed In Oreat Britain. All righls reserved. 0213-1223196 $\ HlO + 0-00
PH: S0273-1223(96)OO389-7
DESIGN OPTIONS FOR SELF-CLEANSING STORM SEWERS Chandramouli Nalluri* and Aminuddin Ab. Ghani** • Department of Civil Engineering. University ofNewcastle upon Tyne, Newcastle upon Tyne NEJ 7RU, UK •• School of Civil Engineering, Universiti Sains Malaysia, Perak Branch Campus, 31750 Tronoh. Perak, Malaysia
ABSTRAcr A list of available codes of practice for self-cleansing sewers is presented and a review of appraisals of minimum velocity criterion is summarised. Compansons of newly developed "minimum velocity" criteria and "minimum shear stress" criterion are presented. Some design charts are also given. These charts are applicable to non-cohesive sediments (typically storm sewers). It appears that sediment size and concentration need to be taken into account, and that a limited depth of sediment bed is recommended for large pipes (diameters> 1000 mm) to maximise their transpon capacity. Copyright © 1996 IAWQ. Published by Elsevier Science Ltd.
KEYWORDS Limit deposition; sediment transport; self cleansing; sewer design. INTRODUcnON Old sewerage systems were designed based on empirical rules to minimise sediment problems and these design rules are still in practice. Either minimum velocity or minimum shear which may not allow any deposition at any time or at least over a long period ohime no deposits would build-up (i.e. high flows flush the deposits) govern these design procedures. In the present trend. "no deposit condition" (clean pipe) in each individual pipe is maintained by allowing flushing to take place fairly frequently and minimum velocity to be achieved under full or half-full conditions. Table 1 reproduces some of the available design criteria given in CIRIA (1987). Ab. Ghani (1993) presented a new design criterion for clean pipe (no deposition) and another one for a pipe having loose deposited sediment bed. A summary of the development of the criteria and ranges of data used is given elsewhere (see Nalluri and Ab. Ghani 1994 a,b). Table 2 gives the newly derived criteria; equations 1 and 2 are applicable for storm sewers only (i.e. with non-cohesive sediments). APPRAISALS OF "MINIMUM VELOCITY" CRITERION
=
Ab. Ghani (1993) appraised the "minimum velocity" criterion (half-full flow, clean pipe and V 0.75 mls) used by Ackers (1984) and Nalluri (1986) and in general concluded and confirmed that the velocity criterion 21S
C. NALLURI and A. AB. GHANI
216
overdesigns the slope for small pipe diameters (D < 500 nun) and underdesigns the slope for larger pipe diameters. Table la. Minimum velocity criterion
Source
Country
American Society of Civil Engineering (1970)
USA
British Standard (1987)
UK
Bielecki (1982)
Germany
Sewer Type
Minimum velocity (mfs)
Pipe Conditions
Foul Storm
0.6 0.9
Fuillhalf-full Full/half-full
Storm Combined
0.75 1.0
Full Full
1.5
Full
Table I b. Minimum shear stress criterion Minimum Shear stress (N/m 2) 3.0 - 4.0 1.0 - 2.0
Reference Source
Country
Sewer Type
Yao (1914)
USA
Storm Foul
Maguire (- )
UK
6.2
Bischof (1976)
Germany
2.5
Pipe Conditions
Fuillhalf full
Table 2. New design criteria for self-cleansing sewers Equations
Transport Mode
(1a)
Clean Pipes
(1b)
-;::.;:;V;::::;:;: Jgd(SI-I) Pipes with deposited loose beds
= 1.18~.16 (WYob)-o·11 (~)-o.34
11 = 0.0014<:.-o.G4
y: )0.,. (~)0.24
w (
1;,.31 (2a)
D;.u (2b)
Comparisons of Ab. Ghani (1993) equations (eqns la and Ib) and "minimum velocity" criterion of 0.75 rnls for half-full flow in a clean pipe (Nalluri and Ab. Ghani 1994a) confirm the inadequacy of the present practice to minimise sediment deposition in pipe sewers. The results also show that the presence of a limited depth of sediment deposit reduces the slope requirement over the entire range of pipe diameters.
217
Self·cleansing Slonn sewers
It was then suggested that sewers of diameter 1 m or more may be best designed with some sediment bed deposition and there exists an optimum depth of sediment deposition in a sewer with given flow and transport conditions (Nalluri & Ab. Ghani. 1994b). APPRAISALS OF "MINIMUM SHEAR STRESS" CRITERION Figures I and 2 show the Q-So-O plots using the new criteria (see Table 2) for comparisons with "minimum shear stress" methods; also shown on these figures is the minimum velocity criterion. Figure I demonstrates that a constant shear stress of 4.0 N/m2 in the case of a clean pipe design (eqns. la & b) would result in overdesigning its slope requirement for volumetric sediment concentration of 20 ppm over the range of diameters up to 1.0 m. However. as the level of sediment concentration increases. the constant shear stress method. tends to underdesign the required slope especially for around 0 > 500 mm. When the sediment concentration reaches 1000 ppm. the constant shear stress method underdesigns the slope requirement over the entire range of pipe diameters; it may be observed that a shear stress of I N/m 2 is grossly inadequate over the whole range of pipe sizes.
Cv C. ~ppmlppm = 20 100 Cv ppm - 1000
10
*-•
1.0
4.0
+-+-+-+-+
0.75
-- V
-I
10 ..·-t-....................,,.-..............,,,...-.--rTTTn....1-r-.......""'"
1
10
10' (I/s)
10
10·
Q
Figure I. Q-So-D plot: clean pipe (half-full flow, dso
=I.Omm, ko =O.6mm).
Figure 2 highlights the level of slope requirements using the constant shear stress methods and new design criteria with loose bed deposits (eqns. 2a & b) for large pipes (0 = 1.0 m to 4.0 m). For a given flow condition (Le. flow depth). an optimum depth of sediment could be present over a range of sediment concentrations as illustrated in Figure 3. This is in agreement with the results from the experimental work by Ab. Ghani (1993) and May (1993). In these experiments. the sediment bed had been kept constant over a range of velocities for a constant flow depth. Figure 2 also shows that even with the presence of an optimum bed. the slope requirements are still larger than those obtained by the constant shear stress method. Both figures I and 2 demonstrate the importance of including sediment concentration as an input in designing the slope requirements for the movement of sediments in sewers. Examples of design charts taking into account sediment concentration are shown in figures 4 and 5.
C. NALLURJ and A. AB. GHANI
218
c. c c. . .......
10-' .....
- 20 = 100
~ppm~ ppm ppm
....... _ ..... 1000 - - V
-
/
i"
"
0 ....
.
....
f,
<'
10 ...
/m N/m: mi.)
l
1.0 4.0 0.75
/;/
.....~,/ ,...... . .~, t:/. '
vr
t
..;
fir
/
,
-.
"...
v
.
I
I
j
... •
I
"J/"
~~
10
10
10 • Q
10 •
10·
(1/,)
10 •
Figure 2. Q-So-D plot: pipes with loose bed (half-full flow. d SO = I.Omm. Y/D=
0.001'
QAQAOO •
1
I~%).
1ft
_0-2'" ,.....,D.31ft
uauD.4", UA.lI D • , "'
.,
Q.
o
Vi
o.OOQe8~.lllI~~""':8".I:O:8~"""~8~.~28~""'"8~.T'3II""''''''~.''''.48
yolO Figure 3a. Optimum sediment depth: C y
=20 ppm (half-full flow. d SO = I.Omm).
0.0030
QAQAOD. , '"
:::::8:1::: .......,0 • • 1ft
UAUO - 5 '"
0.0025
.,
Q.
0
Vi
0.0020
~
~~~ .. ..1. ..
0.0015
0.0010
•• 48
yolO
Figure 3b. Optimum sediment depth: C y
=100 ppm (half-full flow,dSO. I.Omm).
Self-cleansing stonn sewers
o.ooeo
219
C>l>
~D·Jm
u.a.a.;D-4m u..A..AJD.' m
O.OO~O
.
Q.
~
0.0040
0
iii
0.0030
Y./D
Figure 3c. OptImum sediment depth: Cv = 1000 ppm (half-full flow. dSO = I.Omm).
0.0120
c, !ppmj -co ppm -..... c, ppm -
0.0100
O.DOGO
--._---_.
0.0040
-
0.D020
0.0000
1000
~--.
0.0080
co a. 0 iii
20 100
0.0
----..
0.2
o.e
0.4
0.8
1.0
D (m)
=
Figure 4. Design chart for clean pipes (half-full flow. dS O I.Omm. ko = O.6mm).
0.0040
c, lppml-co ppm ppm --
. . . . . C,
20 100 1000
0.D030
"
0.0020
Q.
.Q
..
Ul
O.OOtO
O.OOOD
1.0
2.0
3.0
4.
5.0
D (m)
Figure S. Design chart for pipes with loose bed (half-full flow. d SO
="Omm. y/D =IS%).
C. NALLURI and A. AB. GHANI
220
CONCLUSIONS The use of present design criteria (Table 1) to minimise sediment depositions in sewers for large pipes (D > 500 mm) is questionable. The new design criteria (equations I and 2) should be used in designing storm sewers carrying non-cohesive sediments. Equation 1 is recommended for clean pipes with pipe diameter up to 1 m (Figure 4). For larger pipe diameters (D > 1 m), equation 2 which allows a limited depth of sediment bed (Figure 3) is recommended (Figure 5). A reduction factor should be applied if the new criteria (equations 1 and 2) were to be used for designing foul sewers carrying cohesive sediments. Table 1 shows that the cleansing velocity or shear stress for designing foul sewers is smaller than the one for designing storm sewer. It has been shown (Figures 1 and 2) that sediment size and concentration are important factors to be taken into account in minimising sediment deposition problems. It should also be mentioned that equations 1 and 2 are able to simulate variations in flow conditions in real sewers (see Nalluri and Ab. Ghani, 1994a). NOTATION
Cv
d,dso D Dgr ko Q R
So Ss
V Wb
Yo Ys 1.0
As
'to
Volumetric sediment concentration Mean sediment size Pipe diameter Dimensionless sediment size (= d (Ss-1 )glv2) 1/3) Pipe roughness height Flow discharge Hydraulic radius Pipe slope Relative density of sediment Self-cleansing velocity Width of sediment bed Mean flow depth Mean sediment bed thickness Clear-water friction factor Friction factor with sediment Mean shear stress REFERENCES
Ab. Ghani. A (1993). Sediment Transport in Sewers. University of Newcastle upon Tyne. UK, PhD Thesis. Ackers. P. (1984). Sediment Transport in Sewers and the Design Implications. Intern. Conf. on Planning. Construction, . Maintenance. and Operation of Sewerage Systems. Reading. England. CIRIA (1987). Sediment movement in combined sewerage and stormwaterdrainage systems. London. England. Nalluri. C. (1986). Sediment Transport in RigId Boundary Channels. Proc. Euromech 192: Transport ofSuspended Solids in Open Channels. A A Balkema, Rotterdam. Nalluri. C. and Ab. Ghani. A. (l994a). Sediment Transport in Sewers and Design Implications. Proceedings. National Conference on Hydraulic Engineering. ASCE Vol 2. pp. 933-938. Buffalo. New York. USA, 1·5 August. Nalluri. C. and Ab. Ghani. A. (1994b). Sediment Transport i/l Sewers with and without Deposited Beds. Proceedings. 9th Congress of APD-IAHR Vol 2. pp. 27-34. Singapore. 22-24 August. May. R. W. P. (1993). Sediment Transport in Pipes and Sewers with Deposited Beds. Report SR 320. Hydraulics Research Ltd, Wallingford. UK.