10

CHAPTER
10
Design of New Gravity Wastewater
Collection Systems
Engineers have designed gravity wastewater collection systems for over a century
without the aid of computer models. Why then has the use of computerized models
become standard practice in sanitary sewer design?
First, computerized models relieve engineers of tedious, iterative calculations, allow-
ing them to focus on design decisions. Second, because models can account for much
more of the complexity of real-world systems than manual calculations, they give the
engineer, and the sewer utility, increased confidence that the design will function
properly when it is installed. Third, the ease and speed of modeling gives engineers
the ability to explore many more options under a wide range of conditions, resulting
in a more cost-effective and robust design. Finally, models can be integrated with
computer-aided drafting (CAD) and geographic information systems (GIS) software,
facilitating the production of construction drawings and making better use of the
available data.
The following is a list of steps for the design of gravity sewer systems.
1. Identify the area(s) to be served.
2. Obtain regulatory codes and design guidelines and set system design criteria.
3. Collect geologic, geographic, and topographic data.
4. Collect information on roads, railroads, population, industry, other utilities and
community planning.
5. Undertake field investigations, including feature surveys and ground truthing
at sites that potentially conflict with other services.
6. Identify the natural drainages, streets, and existing or planned wastewater
inflow points at the boundaries of the area to be sewered. Locate all proposed
sources of wastewater. Identify likely elevations of customer laterals.
File modified 8/26/04

334 Design of New Gravity Wastewater Collection Systems Chapter 10
7. Design the horizontal layout of the sewer, including manholes and possible
pumping station locations. If necessary, prepare alternate layouts.
8. Divide the total area into logical subareas, as needed, and develop design flow
rates for each section in the system.
9. Select pipe sizes, slopes, and inverts. Perform the hydraulic design of the sys-
tem. Revise selections until the design criteria are met.
10. Complete cost estimate(s) for the design and alternate designs.
11. Carefully review all designs, along with assumptions, alternates, and costs.
12. Modify the design or develop alternate designs, or even alternate layouts. This
cycles the designer back to the appropriate earlier step.
13. Complete the plan and profile construction drawings and prepare the specifica-
tions and other bid documents.
The primary focus of this chapter is the hydraulic design of the sewer, including infor-
mation on obtaining the data needed to perform the hydraulic design and a discus-
sion of how models are used in the design process. Hydraulic analysis is critical, since
it is used to select sewer pipe sizes, invert elevations, and sometimes pipe routes.
Additional details regarding sewer design may be found in references such as ASCE
(1982) and Metcalf and Eddy (1981).
10.1 Materials
Although this chapter centers on hydraulic design, other factors, such as pipe material
and manhole construction, must also be considered. This section provides a brief
overview of the physical components of sewer systems.
Pipes
Gravity sewer pipes are available in a variety of materials, including cast and ductile
iron, PVC (polyvinyl chloride; see Figure 10.1), concrete, asbestos cement, HDPE
(high density polyethylene), ABS (acrilonitrile butadiene styrene), FRP (fiber rein-
forced plastic), brick, and vitrified clay. Most pipe has solid walls, although there are
some truss or profile wall pipes available. Most new sewer pipe has a circular cross
section; however, many older sewers, especially those made from brick, have different
cross-sectional shapes.
A wide array of ANSI, ASTM and AWWA standards are available for specification of
pipe. Manufacturers and suppliers are another source of design handbooks and other
information on pipe materials. Manufacturers also specify how pipes should be bed-
ded. Figure 10.2 shows a typical trench box and excavator used for pipe installation.
The weak links in most piping systems are the pipe joints and service connections.
Currently, most joints are made of a flexible elastomeric seals or rings. Many older
pipes have cement mortar or bituminous material for joints. Some pipe materials,
such as polyethylene, rely on thermally fused pipe ends.
It is best to install wye or tee service connections when the main is installed so that the
quality of the workmanship is maintained. Such lines should be run to the edge of the
pavement and plugged if customers are not immediately ready to connect. Later con-
nections need to be made with proper quality control and inspection.
Section 10.1 Materials 335
Figure 10.1 PVC pipe ready for installation.
Figure 10.2 Excavator and trench box for sewer installation project.
Manholes
Manholes are structures designed to provide access to a sewer. Access is required for
visual inspection of sewers, placement and maintenance of flow or water quality mon-
itoring instruments, and cleaning and repair of the sewer. According to ASCE (1982),
a manhole design must pass four major tests:
• Provide convenient access to sewers for observation and maintenance
operations
• Cause a minimum of interference with the hydraulics of the sewer

• Be durable and generally watertight
• Be strong enough to support applied loads.
In the past, manholes were constructed of brick, concrete block, or poured concrete.
Today, most manholes are constructed with a precast or poured-in-place concrete base
and precast concrete rings up to the surface, where a tapered segment holds the cover.
Covers (see Figure 10.3) are typically 21–24 in. (500–600 mm) in diameter. The use of
lined, coated manholes or plastic manholes is increasing for economic reasons and to
provide corrosion resistance and reduce infiltration through the manhole walls.
Shape and Dimensions. Most manholes are round and 4–6 ft (1.2–1.8 m) in diam-
eter, with larger sizes for larger sewers. In small sewers, the manhole is generally cen-
tered over the pipe. For large-diameter sewers, the manhole entrance is often offset,
with a work platform to the side of the sewer. The rungs or ladders used for entering
manholes should be corrosion resistant to provide the long-term durability and
strength needed to prevent failure.
Figure 10.3 Manhole covers.
Other Appurtenances
Gravity sewers typically contain very few special appurtenances. Devices such as
inverted siphons (see Section 10.8) and permanent flow meters such as Parshall
flumes (page 278) may be used in some instances.
Section 10.2 Initial Planning 337
In combined sewer systems, it is sometimes permissible to have some type of flow

diversion structure for flows in excess of the system capacity. These structures range
from simple, passive devices such as weirs and orifices, to more dynamic devices such
as vortex separators and tilting plate regulators, to remotely controlled gates and
inflatable dams. Outlets to receiving streams will usually have a flap gate or duck-
billed valve to prevent surface water from entering the collection system.
Pipe junctions usually occur at manholes; however, other junction types may exist in a
system. Examples are blind connections and underground box structures with no
direct access to the surface.
10.2 Initial Planning

Before a new gravity sewer system is designed, a planning study should be under-
taken to address the issues of whether to provide sewer service, the type of convey-
ance wanted (gravity and/or pressure), and the pros and cons of separate versus
combined systems.
Decision to Provide Sewer Service to an Area

Wastewater disposal in an area may take the form of on-site treatment and disposal,
small decentralized treatment systems, or a large centralized treatment system. Some
of the factors that must be analyzed when considering the construction of a wastewa-
ter collection system are:
• Population growth and housing density
• Amenability to on-site disposal systems
• Environmental impact of habitation
• Pollution problems in the area
• Accessibility to and cost of collection systems and sewage treatment
• Regulatory requirements
• Ease of construction, including the extent of rock excavation required.
An onsite wastewater treatment system that is properly designed, manufactured, con-
structed, operated, and maintained and that has a suitable effluent-receiving body
may be an appropriate means of wastewater disposal, especially when a development
is far from a regional system. However, wastewater collection and centralized treat-
ment systems usually prove to be the most cost-effective solution to providing sewage
services. Public health authorities generally favor collection and centralized treatment
systems over on-site systems because they are much easier to inspect, monitor, and
regulate.
A preliminary investigation is generally conducted to determine the potential area
to be served, the consequences of not serving the area, how the area will be served,
and the economic feasibility of such a project. There is usually some break-even
population above which a centralized sewage collection and treatment system is
desirable, although the exact break-even point varies widely, according to site-specific
considerations.
A common problem is determining how to serve the first areas to be developed in a

drainage basin. In the short term, construction of treatment at the downstream end of
the development is the least expensive option. However, construction of a large
regional treatment system with larger sewers sized for full buildout may be more eco-
nomical in the long run.
Types of Conveyance
Wastewater collection systems use gravity, pressure, vacuum, or some combination of

these methods to convey wastewater. In most cases, lower total life-cycle costs (con-
struction, operation, and maintenance) favor gravity sewers over pressure or vacuum
systems. In gravity systems, pumping may be required to move flows in areas of flat
or uneven topography. In these systems, sewage flows by gravity into a wet well and
is pumped through a force main. The hydraulic analysis of pump stations and force
mains is described in Chapter 12.
Low-pressure sewers differ from force mains in that the pumps are located on the
property of each customer. These systems are used either where direct gravity flow is
not possible or where excavation depths, extensive shallow rock, obstacles, and align-
ments make the smaller and shallower pressure lines more economical. The pressur-
ized lines may discharge into a wet well or into a gravity sewer. These systems are
discussed further in Chapter 13.
Vacuum systems operate on the basis of differential air pressure created by a central
vacuum station. Wastewater from each connection is discharged to a sump that is iso-
lated from the main line by a valve. When the level in the sump reaches a designated
limit, an actuator opens the valve and the vacuum propels the wastewater to the cen-
tral vacuum station. In a good vacuum system, the maximum water-level differential
in each vacuum zone is about 25 ft (7.6 m). This means that vacuum systems are rele-
gated to rather flat areas. Multiple vacuum zones may be placed in series, but operat-
ing complexity and costs increase rapidly. More information on these systems may be
found in US EPA (1991) and Crites and Tchobanoglous (1998).
Separate versus Combined Systems
In the past, conveyance of sanitary flows and storm drainage in the same conduits
(combined sewers) made economic sense. Today, with expensive mandatory treat-
ment requirements in place, the cost of control and treatment for new systems is
almost always lower if the two types of flow are conveyed and treated in completely
separate sewer systems. Currently, most pollution control agencies specifically pro-
hibit combining storm drainage and wastewaters in their regulations governing new
sanitary sewer design, construction, and operation (e.g., GLUMRB, 1997).
However, it is conceivable that future treatment requirements for storm waters in

urban and industrial settings may become so stringent that combined system convey-
ance and treatment may again be the more economical solution. The practice of inte-
grated watershed management, which considers the combined effect on receiving
waters of stormwater and treated wastewater, may result in some portion (such as the
first flush) of stormwater being directed into the sanitary system.
Section 10.3 Preliminary Design Considerations 339
10.3 Preliminary Design Considerations

Design codes and criteria are promulgated by various public environmental and
pollution-control agencies. Codes give information on location and clearances relative
to other utilities, flow-generation rates, peaking factors, and hydraulic guidelines. In
the United States, design codes commonly come from a state or county/regional
agency. Most of these codes can be traced back to the Ten State Standards (GLUMRB,
1997), which were first published in 1951. Codes may reflect the effects of local condi-
tions, such as soil and weather, on the design of sewers. Current codes tend to make
more use of such terms as “recommended” or “guidelines,” rather than “required” or
“mandatory.” Some regulatory agencies have adopted a more performance-based
approach to establish system regulations. Design engineers need to be well-versed in
these codes.
Depending on the regulatory agency evaluating the design, either the sewers must be
able to carry the design flow while flowing full or at some given fraction of the full
depth, or the design flow must be some percentage of the full-pipe capacity (e.g., 75
percent of capacity at design flow). The engineer must determine which is the better
approach for a given case, as the two requirements can produce different pipe sizes.
For example, a 12-in. (300-mm) sewer with n = 0.013 and a 1-percent slope can carry
3.56 ft3/s (0.101 m3/s) when full. Seventy-five percent of the capacity is 2.67 ft3/s (0.075
m3/s), while the capacity at 75 percent of full depth is 3.25 ft3/s (0.092 m3/s). Minimum
slopes to prevent sedimentation and minimum cover to prevent traffic impacts must
also be considered.
Where local codes do not exist or do not address a specific issue, widely adopted stan-
dards, such as the Recommended Standards for Wastewater Facilities (GLUMRB,
1997), should be consulted. References such as Metcalf and Eddy (1981), ASCE (1982),
and Hammer and Hammer (2001) also provide guidance for sewer design.
In addition to technical design criteria and the hydraulic loads, other pertinent con-
siderations include the following:
• Costs – The planning, design, construction, operation, and maintenance costs
should all be considered in assessing the long-term sustainability of the
project.
• Schedule – The time between the initial planning and the operation of a sewer
may be as long as a decade, and the project may need to be constructed in
phases.
• Operation and maintenance – Gravity sewers are not maintenance free. They
require periodic inspections and cleaning, and some components may require
repair or replacement. An organization to operate the system must be in place.
• Environment – Construction of the sewer results in both temporary disruption
of and long-term effects on the environment. Short-term effects include tem-
porary lowering of the groundwater table and release of sediments. Over its
design life, the sewer system will affect the performance of the wastewater
treatment facility and the water quality of the receiving water body.
• Regulatory compliance – The sewer project must comply with all applicable laws
and regulations.
The design criteria should be developed in consultation with the sewer utility and the
financing and regulatory agencies. Consensus on the design criteria before the start of
the design is essential.
Data Requirements
The data requirements for building a hydraulic sewer model are discussed in
Chapter 5. Additional physical data, beyond that necessary for hydraulic analysis, is
required for the design of new sewers. The following list of data requirements has
been adapted from American Society of Civil Engineers (1982):
• Topography, surface and subsurface conditions, details of paving to be dis-
turbed, underground utilities and structures, subsoil conditions, water table
elevations, traffic control needs, and elevations of structure basements or con-
nection points
• Locations of streets, alleys, or unusual structures; required rights of way; and
similar data necessary to define the physical features of a proposed sanitary
sewer project, including preliminary horizontal and vertical alignment
• Details of existing sanitary sewers to which a proposed sewer may connect
• Information pertinent to possible future expansion of the proposed project
• Locations of historical and archeological sites, significant plant or animal com-
munities, or other environmentally sensitive areas.
The data may be obtained from a variety of sources. Maps, aerial photographs, con-
struction drawings, and ground surveys may be queried. Typically, no single data
source is complete, so one must identify inconsistencies and fill in gaps.
Instrument surveys are the most accurate method of obtaining high quality spatial
data. Many vertical-elevation reference and control points should be established sys-
tematically along the route and be accurate to within about 0.01 ft (3 mm) (ASCE,
Section 10.4 Initial System Layout 341
1982). These rather prominent and durable control points are used later as reliable ref-
erence points for the crucial elevation control needed during sewer construction.
Alternatives
Alternate system configurations, alignments, and pipe sizes must be explored during
the design of new sewers. Each alternate design should be considered until it is appar-
ent that it is infeasible or inferior to another design. The point at which an alternative
is dropped from consideration will vary. In some cases, a simple calculation will show
that an alternative does not meet one or more design criteria. In other cases, an alter-
native design must be completed, with a detailed cost estimate, before its economic
ranking is known.
It may be necessary to document alternate designs dropped from consideration and
the reasons for those decisions. Such decisions are made at various levels. The design
engineer will typically make decisions regarding pipe sizes, invert elevations, and
manhole details. However, decisions regarding alignments, stream crossings, and
locations of pump stations often require input from sewer utility management and
permitting authorities.
Hydraulic sewer models are invaluable for managing design alternatives. Most mod-
eling packages have capabilities that facilitate the analysis of multiple alternatives
without building separate models. A final design report usually should include
results of alternate designs that were considered and the justifications for rejecting
them.
10.4 Initial System Layout

Sanitary sewers are almost always laid out in a dendritic (treelike) pattern. Unlike
water distribution systems, which have pressurized flow in all directions, gravity
sewers normally allow flow in only one direction and are rarely looped. Some of the
more prominent factors that affect the layout of the final wastewater collection net-
work serving a designated area are political jurisdictions and boundaries, develop-
ment types and patterns, street and right-of-way patterns, alignment of existing
underground services (particularly stormwater conveyance), land topography, and
geology. When large cost savings are involved, some of these factors can be modified
or changed, most notably the political boundaries. Political boundaries may change
through cooperative agreements, formation of an overlapping special service district,
or establishing service authorities based on basin rather than political boundaries.
Armed with good information, designers can delineate major drainage areas and
begin the process of laying out the wastewater collection system network.
Developers should consider the location of all utilities when laying out new streets.
Substantial cost savings can be realized from simultaneous construction of all under-
ground utilities. When new sewers are proposed for an existing development, the
locations of all underground utilities should be determined before beginning the lay-
out of the sewage collection system.
Generally speaking, it is most economical to plan the network and design pipes for
the service and flows expected at full buildout of the drainage areas. However, the
cost of such piping in areas that remain sparsely populated can be significant, and
much of the loading may not be realized for many years. This results in tradeoffs
between budget limitations and the desire to provide ultimate capacity. Modeling can
assist with evaluating various options to arrive at rational, feasible decisions about the
extent and capacity of the system.
Gravity Sewer Layout

The layout of a gravity sewer system begins with a topographic map of the service area.
The drainage divide lines and natural drainage paths for a new service area should be
drawn on the map. Then, the locations of the sewer outlet(s) and elevations of the
highest services should be identified. The highest manhole is located where it can provide
a connection to a lateral from the highest service. The highest possible invert is deter-
mined either from the elevation of the basement (or lowest plumbing fixture) minus an
adequate drop along the lateral, or the minimum depth that would give a lower elevation.
The sewer should continue along streets or other public rights-of-way at the flattest
acceptable slope, as determined by capacity needs, self-cleansing requirements, mini-
mum depth, or service lateral depths.
When localized high points in the sewer alignment are encountered, a downhill slope
may be maintained by either deeper excavation or an alternate path. Crossing depres-
sions of streams may require an inverted siphon (see Section 10.8). Ultimately, the line
ends at a connection to a trunk sewer, a wastewater treatment facility, or a pumping
station. Sometimes service areas include drainage divides, such that only part of the
development can flow by gravity into downstream sewers. Other parts drain by grav-
ity to a pump station that either lifts the flow to the minimum elevation necessary to
continue by gravity or pushes the flow through a force main (pressure pipe) to the
desired location.
Manhole Location and Spacing

Manholes are generally placed at every change in slope, pipe size, or alignment; where
two or more incoming lines are joined; at the upstream end of lines; and at distances
not greater than about 300 ft (90 m) for diameters less than 6 in. (150 mm), about 400 ft
(120 m) for diameters of 6–15 in. (150–400 mm), and about 500 ft (150 m) for diameters
of 18 in. (450 mm) or greater. These distances are dictated by regulations and by the
reach of cleaning and maintenance equipment. Manhole spacing greater than 500 ft
has sometimes been used for sewers large enough to allow a person to work and walk
in the sewer if needed (Hammer and Hammer, 2001). Additional manholes may be
required to drop the hydraulic grade line, as described in Section 10.7 on page 358.
The following are some additional considerations related to manhole location:
• Street intersections are common locations for manholes (American Society
Civil Engineers, 1982).
• A terminal manhole should be located at the upper end of a sanitary sewer to
provide access for maintenance.
• Manholes should not be placed in a location that allows surface water to enter.
• Inserts may be used to limit infiltration through pick holes.
• Manholes in cross-country settings usually have chimneys raised above the
ground to prevent surface water intrusion, as shown in Figure 10.4.
Photo courtesy of R. D. Zande and Associates
Figure 10.4 Typical manhole in a cross country setting.
Location of Pumping Facilities

Pumping facilities are logically placed at low points to collect gravity flow from an
area and pump it to a desired point. Pumping facilities are best placed on public land,
where they are readily accessible but also as isolated as possible to minimize odor and
noise complaints. Aesthetic or other local considerations may require putting a pump-
ing station some distance from the ideal location as determined from an initial
hydraulic and cost-efficiency point of view.
Sewer Easements
To the extent possible, sewers should be laid in public rights-of-way to avoid ease-
ments across private land. In most cases, placing sewers under public streets is best
because of construction and maintenance issues. A slight modification in the road net-
work in a subdivision can sometimes result in significant cost savings.
Nevertheless, system economy and design feasibility sometimes make it necessary to
consider non-street routes. If not already in place, easements must be obtained early
in the planning and design process. Although obtaining easements may be easy in
most cases, the process can take a long time and cause project delays. Easements must
be sufficiently wide to allow access for construction, inspection, maintenance, and
repair (ASCE, 1982). Designers should also be fully informed about utility prohibi-
tions and restrictions for sewers that run along or across railroads, highways, major
arterial roads, streams, ponds, or lakes.
Example of a Sewer Network Layout

The depth of the upstream end of a gravity sewer is typically governed by the types of
structures to be served. There are several rules of thumb, such as placing the invert at least
3 ft (1 m) below the foundation slab. This results in relatively shallow minimum sewer
depths where there are no basements and deep sewers where structures have basement
drains. Additional factors influencing manhole depth are described in Section 10.6.
If structure elevations are lower than the street, it is usually necessary to either install
sump pumps to convey the wastewater to the gravity sewer or run the sewer lateral
through the back of the property to a lower street.
In new land developments, sites can be graded or sewer lines can be installed so that
flow is directed by gravity to the sewer under the street. In areas with existing struc-
tures, a survey of the existing facilities should be conducted to determine the typical
location of the lowest plumbing fixture in the homes.
If extensive rock excavation is necessary for sewer installation, increasing the depth of
a sewer can greatly increase costs. In new land developments with no rock and a low
water table, cost is less sensitive to depth.
The layout of a sewer system is illustrated in Figures 10.5 through 10.7. The existing
contours, roads, and sewer line and the boundary of the proposed subdivision are
shown in Figure 10.5. The drainage divide and natural drainage paths are shown in
Subdivision Boundary
Contours
Existing Road
Existing Sewer
Figure 10.5 Location of subdivision boundary, existing road, and

existing sewer with topographic contours prior to development.
Natural
Drainage
Drainages
Divide
Figure 10.6 Location of subdivision boundary, drainage divide, and natu-

ral drainageways prior to development and installation of new sewers.
Edge of Right of Way

Pump
N Station
Sewer
Easement
Easement Force
Main
Manhole
Figure 10.7 Location of subdivision boundary with proposed roads,

right-of-way, sewer manholes, and force main for installing new sewers.
Figure 10.6. Note that the divide runs through the subdivision and it is impractical to
service the entire project with a single gravity sewer.
The proposed roadways and the sewer layout are shown in Figure 10.7. The network
consists of two separate sewer lines. The northeast portion of the subdivision will be
served by a gravity sewer that flows to a pump station. This station will discharge into
a force main (dotted line) that runs parallel to the gravity sewer but in the opposite
direction, until it discharges into a manhole in a second sewer line. This second sys-
tem ties into the sewer main south of the property.nt. The layout requires easements
in two locations.
10.5 Flows in Sanitary Sewers

Chapter 6 discussed sanitary design flows in detail. One approach to formulating
design loads is to follow the values recommended in the appropriate code with
regard to flow per household or commercial building. The values found in codes that
can be used to generate average and peak flows are generally conservative and, in
most cases, provide some excess capacity. Nevertheless, the designer must be alert for
the rare situations where these normally conservative values are still not adequate; in
particular, for commercial and industrial areas. The recommended approach is to fol-
low the method described in Chapter 6. Usually, average flows at loading nodes are
determined based on population or a population equivalent, and these flows are
adjusted for design purposes using extreme flow factors (peak and minimum multi-
pliers). Peak design flow is most critical for pipe sizing, while minimum flows are
most important for insuring that slopes are adequate for self cleansing.
As described in Chapter 6, the initial design minimum flow rate is not the smallest
flow rate expected in the pipe, but rather the smallest flow rate for which self-cleans-
ing is expected. Since the absolute minimum flow in some small sewers is zero, the
minimum design flow rate is actually the largest flow rate during the lowest flow
period for the line, perhaps using the lowest flow week for the time period. Chapter 6
and the sidebar on page 347 contain additional information on design flow rates.
The design maximum flow rate is the discharge for which capacity must be available.
The initial and final average flow rates are needed if peaking factors are used to gen-
erate the design minimum and maximum flow rates, which are needed for self-cleans-
ing and capacity considerations during design or analysis.
Low Flows in Early Years

Solids buildup can be a significant problem in sewers when the flows are extremely
small, particularly in areas with very flat slopes. This problem is often found in the
upper reaches of sewers in developing areas with few connections when the sewer is
first placed in service. If increasing slopes (and thus velocities) in these small flow sec-
tions would result in unacceptable depths and/or costs, a more practical solution
might be to use periodic flushing or cleaning until area flows have increased enough
for the sewers to be self-cleansing.
Allowances for Infiltration and Inflow

Non-wastewater flows that enter a wastewater collection system come from groundwa-
ter infiltration or rainfall-derived inflow and infiltration, as described in Chapter 7. Collec-
Section 10.5 Flows in Sanitary Sewers 347
What Are Minimum Flow and Velocity?

Most sewer design is based on providing enough curve for the ratio between minimum and average
slope to achieve a minimum velocity of 2 ft/s (0.6 flows, with the minimum flow being checked for
m/s) at full flow. However, in the upper reaches of minimum velocity. However, the lowest value on
most sewer systems, flows are very low and are that curve, 0.2, corresponds to a population of
frequently zero (intermittent flow). With such low 1000, whereas there are many sewers serving
flows, the minimum velocity necessary for the much smaller populations. While such a value is
pipes to be self cleansing is rarely achieved. probably not correct, the true minimum velocity in
With one or two houses at the end of a sewer small sewers is zero, and there is a statistical dis-
branch, it is unlikely that flows will exceed 20 tribution of flow such that 0.2 times the average
gpm (1.3 L/s), yet most sewers in these situations flow would be extremely small.
perform adequately. This is possible because In small sewers, it may be best to think of self-
velocity does not decrease proportionally with cleaning as an event that occurs only once or
flow. For example, for an 8-in. (400-mm) sewer at twice a day, especially during the early years of a
a slope of 0.4/100 (GLUMRB, 1997), it only pipe’s life. The concept of a minimum-to-average
takes 12 gpm (0.76 l/s) to reach 1 ft/s (0.3 m/s), flow ratio is meaningless in small sewers with
but it takes 120 gpm (7.6 l/s) to reach 2 ft/s (0.6 intermittent flow, so basing slopes on minimum
m/s). This relationship is illustrated in the figure velocity may be impractical. The engineer is then
below. faced with either adopting minimum slopes from
The question remains of what flow an engineer the standards without explicitly basing the design
should use when determining whether the velocity on velocity or, for important sewers, using a trac-
will be high enough to prevent solids deposition. tive force approach as described in Chapter 2.
Standard references (ASCE, 1982) provide a
Low Flow Velocity in 8-in. Sewer
s=0.2/100
3 s=0.4/100
s=0.8/100
s=1.6/100
Velocity, ft/s
0 20 40 60 80 100 120
Flow, gpm
tively, these flows are called inflow and infiltration (I/I). Sources of I/I are illustrated in
Figure 7.7 on page 212.
In modern codes, allowances for I/I are specified by applying a constant unit rate (e.g.,
volume/day/length of pipe, volume/day/length of pipe × pipe diameter, or volume/
day/service area) as described in Chapter 9 on page 301. Designers should review reg-
ulations and guidance in their local areas. The consensus is that good sewer design
and construction, along with tight ordinances and codes and good enforcement, can
keep I/I at very low levels.
Phased/Staged Construction
A service area may have defined boundaries, with an expected rapid buildout and
plans to complete the full system in a single project. However, when development
will take place over decades, completion of the long-term system through a series of
smaller subprojects may save money or make expensive projects financially feasible.
Sewers are often sized for ultimate upstream buildout if the probability is high that
full development will occur within a few decades, since this is almost always more
cost effective than laying parallel pipes in future years to handle upstream growth.
When new services are added over a long period of time, the model can be used to
analyze several projected loading scenarios.
10.6 Horizontal and Vertical Alignment

Section 10.4 presented guidance for the general layout of the sanitary sewer system
that must be performed early in the design process. Once the design flows for these
pipes have been determined, the engineer can proceed with designing the vertical
sewer alignment and sizing the pipes. Issues related to vertical alignment are pre-
sented in this section, and hydraulic design and pipe sizing are covered in
Section 10.7. However, because pipe size can affect invert elevations (due to cover
requirements), and pipe slope affects pipe capacity and thus pipe size, the two tasks
must be performed concurrently in many cases.
Pipe Slopes
A minimum velocity must be maintained to prevent solids buildup during low flows
and is determined by the tractive force method described in Section 2.9. Alternatively,
the minimum slopes specified in Table 2.2 may be used. These minimum slopes vary
according to pipe diameter.
ASCE (1982) recommends that flow velocities be less than 10 ft/s (3.5 m/s) at peak
flow. Higher velocities may be tolerated if proper consideration is given to the pipe
material, abrasive characteristics of the wastewater, turbulence, and thrust at changes
in direction. GLUMRB (1997) states that if velocities are greater than 15 ft/s (4.6 m/s),
special provisions shall be made to protect against displacement by erosion and
impact.
Minimum slopes for house connections are often specified in local codes. In addition
to the real need for somewhat steeper slopes for self-cleansing given the small flow
rates in most laterals, the steeper minimum slopes cover a multitude of shortcomings
Section 10.6 Horizontal and Vertical Alignment 349
commonly encountered with laterals, such as poor slope control, alignment, bedding,
placement, and backfill, leading to pipe differential settlement and joint separation. ASCE
(1982) recommends a minimum slope of 0.01, while Metcalf and Eddy (1981) recom-
mend 0.02.
Curved Sewer Alignment

Typically, sewer piping is installed in straight segments, with changes in the vertical
or horizontal alignment occurring at manholes. However, in some cases, a good solu-
tion may be to use a curved sewer alignment to change direction. For example, a verti-
cally curved alignment might be used to maintain a reasonable depth without the
need for additional manholes where the ground surface makes a fairly large change in
slope over a few hundred feet. A horizontal or vertical curve may be used to avoid
obstructions or give future access for lateral connections or repair. Curved vertical
and horizontal alignments are shown in Figure 10.8
Hydraulically, there is little to prohibit curved alignment as long as the minimum
slope meets capacity and self-cleansing needs. The main objections to curved align-
ment are the extra effort and cost for field survey staking and construction, as well as
the extra location information needed to avoid damage during future digging, con-
nect laterals, or provide maintenance. When the cost savings resulting from fewer
manholes and shallower construction outweighs these problems, curved sewers
should be considered. Vertical curvature is probably of less concern than horizontal
curvature, since future crews may not have immediate access to detailed information
for locating the alignment. Universal use of metallic location tape placed above the
sewer line during construction helps to alleviate reservations about curved alignment.
Burton (1996) generally recommends straight alignment for pipes 24 in. (600 mm) and
smaller. For larger sewers, curvature should have a radius of not less than 100 ft
(30 m). Rigid pipe may be installed in a curve by putting a deflection at each joint. The
maximum permissible deflection depends on the type of joint and the pipe diameter.
Flexible pipe is simply forced into the desired curvature. In either case, the pipe man-
ufacturer should be consulted regarding allowable limits for deflection. The geometry
of curved alignments is described by ASCE (1982).
An example of a curved segment of sewer pipe is shown in Figure 10.8. The radius of
curvature, R, is given by
L
R = -------------------------
- (10.1)
1 ∆
2 tan --- ----
2  N
where R = radius of curvature (ft, m)

L = length of sewer pipe segment (ft, m)
∆ = total deflection angle of curve (degrees or radians)
N = number of sewer pipe deflections
Minimum Depth of Cover

Sewer pipe should be placed as shallow as possible while still being located
• Deep enough to provide gravity service whenever feasible
Horizontal Curve
Gravity Eliminates Two Manholes
Sewer
MH2
3 + 00
5 + 00
4 + 00
2
+
00
1
+
00
MH1
Cul de sac
Plan View
Road Surface
MH1
300.0
Vertical Curve
Sewer Straight
Elevation, ft
MH2
280.0
Excavation Avoided
Manholes
Avoided
260.0
0 + 00
1 + 00
2 + 00
3 + 00
4 + 00
5 + 00
Profile View
Figure 10.8 Horizontal alignment (plan view) and vertical alignment (profile view)
of curved sewers.
• Below the frost line (in permafrost areas, the sewer must be insulated, perhaps
even heated)
• A reasonable distance below other utilities, especially potable water lines,
unless special features are used to protect against contamination
• Deep enough to adequately distribute traffic and other moving surface loads
without causing loading stress breaks in pipes or connections
The cover depth is the distance from the soil surface to the top of the outside surface of
the pipe. Minimum cover in nontrafficked areas is generally 1.5–2 ft (0.45–600 m) and
in trafficked areas is 4–5 ft (1.2–1.5m), depending on the pipe type. In areas with base-
Section 10.6 Horizontal and Vertical Alignment 351
ments needing gravity sewer service, minimum depths have traditionally been 8–9 ft
(2.4–2.7 m). Modern basements are often used as living areas and may be several feet
deeper than they were in the past. Sewers serving these deeper basements may need
to be 11–12 ft (3.4–3.7 m) deep or more. Utilities may have a specification such as “top
of sewer must be at least 3 ft (1 m) below basement floor,” although requiring a ser-
vice line slope of 1 to 2 percent may be a more appropriate way of specifying mini-
mum depth. If very few buildings have significantly deeper basements, individual or
local-area pumps that discharge to the street sewer might be a more economical life-
cycle solution.
A significant cost issue relative to minimum depth sometimes occurs in relatively flat
service areas, where the required pipe slope is greater than the slope of the ground
surface. As the gravity sewer proceeds downstream, it is forced deeper. An alternative
approach, especially where the water table is high or shallow rock is encountered, is
to raise the upstream end of the sewer and service buildings with pumps. In such
cases, the cost of pumped lateral services on the upstream end should be weighed
against the increased construction cost of the deeper system.
A rough approximation of the costs of raising the upstream end of a sewer can be
made by multiplying the difference in depth by the incremental estimated excavation
and pipe placement cost, adding the incremental cost of manholes in the deeper alter-
native, and comparing this value with the construction and operation costs of the
pressurized service line. Situations sometimes occur in which an isolated deep gravity
service can result in several hundred thousand dollars in incremental sewer construc-
tion costs, as compared to pumping to a shallower street main.
For streets that run along contours on the side of a hill as shown in Figure 10.9, it will
be difficult to convey wastewater from buildings on the downhill side of the street to a
main with a typical burial depth. Options include a very deep main, pumping to a rel-
atively shallow main, or running the lateral through an easement to the main in the
next street downhill. The steepness of the area, extent of rock excavation, and distance
between streets will determine which option is best.
Most codes require that the top of the sewer be at least 1.5 ft (0.45 m) below the bottom
of any nearby water main, unless special sewer pipe and/or joints are used (GLUMRB,
1997). The required horizontal separation is often at least 10 ft (3 m) without special
pipe and joint features.
STOP
Road
Pump to
sewer
Gravity to
deep sewer Road
Gravity to
lower street
Figure 10.9 Alternatives for providing sewer service to structures below
Maximum Depth
Maximum sewer depths are often set at 20–25 ft (6–8 m). Common factors that have
historically limited depth are as follows:
• Groundwater makes construction more expensive and hazardous, and makes
high-quality pipe bedding and pipe placement more difficult.
• Soil layers, rock layers, or other subsurface conditions make excavation very
difficult, the cave-in hazard high, or pipe structural loadings too high.
• Trench stability is more difficult to manage with greater depth. Often extreme
(costly) measures are needed for deep sewers, such as special cutoff walls and
piling support walls.
• Maximum depth capability of excavation and other maintenance equipment
available to the owner may be exceeded.
Relatively new construction techniques, such as microtunneling with laser-guided
boring or pipe jacking, are becoming particularly economical for deep sewers in the
1–3 ft (300–900 mm) diameter range.
A limitation on maximum depth can be viewed primarily as a cost issue, to be
weighed against the cost of increased pumping or alternative horizontal alignment.
Since the cost of pumping, including construction and O&M, is very high, even deep
sewer placement may cost less than additional pumping stations. Chapter 13 covers
economic aspects of pumping.
In areas where few or no service laterals are needed, trenchless sewer construction
using horizontal directional drilling or boring and jacking may prove cost effective.
Trenchless construction can also be used (and is often required) for service laterals
where they cross railroads, streams, or major highways.
10.7 Hydraulic Design

After the proposed sewer network layout has been determined, including structure
locations and initial elevations, and the design flows have been computed, the next
major task is the hydraulic design. Hydraulic design primarily consists of determin-
ing what the pipe sizes will be, but, as discussed in Section 10.6, the tasks of setting
invert elevations and determining pipe sizes are interrelated. Therefore, some adjust-
ment of structure elevations will likely take place during the hydraulic design.
Section 10.7 Hydraulic Design 353
Life-Cycle Cost
A typical example of when a life-cycle cost analy- cost is $5000 for the shallow sewers and
sis should be performed is the case in which a util- $20,000 for the pump station.
ity can lay either a very deep sewer in a flat area The interest rate for the analysis is 6 percent. To
with a very high initial cost or a series of shallow convert the O&M cost to a present worth for com-
sewers connected by a force main and pump sta- parison, the series present-worth factor is calcu-
tion. lated as 15.7. To determine the present worth of
The estimated initial cost for the deep-sewer alter- the pump station rehab, a present-worth factor of
native is $825,000, with annual O&M (operation 0.233 is used. The calculation is summarized in
and maintenance) costs of $10,000. The shallow the table.
sewer is estimated to cost $300,000, the force In this case, the deep-sewer alternative has the
main $125,000, and the pump station lowest life-cycle cost and is also likely to be more
$225,000. The design life of the project is 50 reliable. However, environmental, financing,
years, but the pump station will need to be rehabil- social, and political considerations can also factor
itated in 25 years at $50,000. The annual O&M into the design selection.
Present Worth,
Item Cost, $ Factor $
Deep sewer 825,000 1 825,000
Deep sewer O&M 10,000/yr 15.7 157,000
TOTAL: Deep sewer project $982,000
Shallow sewer 300,00 1 300,000

Force main 125,000 1 125,000
Pump station 225,000 1 225,000
Sewer O&M 5000/yr 15.7 78,500
Pump station O&M 20,000/yr 15.7 314,000
Pump station rehab 50,000 0.233 11,650
TOTAL: Shallow sewer project $1,054,150
The following summarizes the basic steps in the hydraulic design process:
1. Make an initial estimate of diameter.
2. Run the hydraulic analysis and check results against design criteria.
3. Make adjustments to pipe sizes, elevations, etc. as needed to meet the design
criteria.
4. Check along the pipeline to ensure that minimum cover is maintained and
burial depths are not excessive.
5. Check the minimum velocity at the design minimum flow rate and the maxi-
mum velocity at the design maximum flow rate.
6. Check situations where the pipe size decreases downstream and determine if
this is acceptable; change the diameter if it is not. Maintain the same diameter if
impacts are not acceptable.
7. Check where the pipe is shallow or above ground and determine whether
deeper burial, an inverted siphon, or an elevated pipe (see Section 10.8) is the
best solution. Check for conflicts with other buried utilities and structures.
8. Check areas with high velocity or slope and determine if a drop manhole is
desirable.
9. Where the slope is minimal, check if lowering downstream manholes relative
to upstream manholes is desirable to reduce pipe size.
10. Where excavation costs and pipe sizes are uneconomical, investigate pressure
sewers or pumping stations and force mains.
11. When planning an addition to an existing system, the capacity of all existing
downstream piping and pumping to the treatment facility must be checked to
determine if bottlenecks exist. If downstream sewers have insufficient capacity,
sewer rerouting or design of a parallel relief sewer may be needed to carry the
new flows.
12. Develop cost estimates.
13. Repeat the hydraulic calculations and cost estimates, and perhaps investigate
alternate routes, until an optimal design is achieved.
This section describes the hydraulic design of sewer piping and manholes. It also dis-
cusses the use of computer models for design and types of hydraulic analysis.
Pipe Sizing
Pipe sizes are selected to convey the peak flow. According to ASCE (1982), it is cus-
tomary to design pipes having diameters of 15 in. (357 mm) and smaller to flow half
full, while larger-diameter pipes are designed to flow three-quarters full. Actual crite-
ria vary among jurisdictions. Commercially available pipe sizes were discussed in
Chapter 5.
Some regulatory agencies are allowing sewers to be designed with little to no excess
capacity such that they will surcharge during design flow events. Surcharging tends
to increase the amount of in-system storage that occurs, and less conservative designs
lower construction costs. However, because such systems have very little safety allow-
ance built in, having an accurate hydraulic model becomes even more important.
Small-Diameter Sewers. The smallest pipe diameter traditionally specified in the

U.S. for gravity sewer mains is 8 in. (200 mm). This requirement was based on difficul-
ties associated with cleaning and connecting laterals to smaller-diameter pipes, as
well as the propensity of small sewers to clog more frequently. However, smaller
diameter sewers are often desirable because they posses higher velocities for a given
slope and flow rate, allowing for better self-cleansing at low flows. Because good con-
nectors and inspection and cleaning equipment are now readily available for diame-
ters 6 in. (150 mm) and smaller, the use of 6-in. or even 4-in. (100-mm) pipe may be
allowed in some cases.
Pipe Size Reduction Downstream. Sewer pipe sizes usually increase in the
downstream direction due to an increasing flow rate. However, if the ground slope of
a sewer route increases significantly in the downstream direction, the pipe size in a
sewer run may actually be able to decrease without decreasing capacity. If the steep
portion of the sewer run is long, significant cost savings may result from using a
smaller pipe.
For smaller sewer sizes, prohibition of sewer-size reduction downstream is common,

with the reasoning that small-diameter pipes have a higher probability of large debris
lodging where the size reduces. For sewers larger than about 18 in. (450 mm), such
clogging is much less likely, and pipe-size reduction is a viable option if an appropri-
ate hydraulic transition is provided. The designer should not switch back and forth
between larger and smaller sizes if the slopes undulate over the route. Increased costs
associated with manholes having different-size connectors and greater pipe ordering
and delivery complexity may negate the anticipated cost savings. However, if size
reduction is possible for a distance of several thousand feet or more, then it should be
considered as a cost-saving feature.
Manholes
A smooth, U-shaped channel should be formed in the manhole base to convey the
wastewater through the manhole with minimal headloss. This practice is referred to
as manhole benching or shaping. If the manhole has multiple inflow pipes, the U-
shaped channels from each inflow should be curved to merge and provide a smooth
transition through the manhole. A small elevation drop is common through manholes
to allow for slight vertical misalignment and compensate for headloss. For manholes
with smooth and smooth-curvature channels, the energy loss should only be about
0.02 ft (6 mm) for manholes on straight pipe runs, with another 0.02 ft (6 mm) for each
additional inflow pipe due to changes in flow direction and increased turbulence.
Figure 10.10 shows a benched manhole with smooth transitions for three inflow pipes.
Photo courtesy of R. D. Zande and Associates
Figure 10.10 Looking down into a benched manhole with three inflow pipes.
Some codes call for matching the crown elevations of all the pipes at a manhole, while
other codes, such as the GLUMRB (1997), call for matching the 80 percent of maxi-
mum depth point of the pipes. Manholes may occasionally be designed to surcharge
at maximum design flows, usually to dissipate energy.
Computer Modeling for System Design

Hydraulic design calculations can be performed manually, with a computer spread-
sheet program, or with a sewer system model with a graphical user interface. How-
ever, the design process usually involves many iterations, which makes manual
calculations impractical for all but the simplest systems. The ability of sewer system
models to display plan and profile views with color coding and annotation makes
them more desirable than spreadsheet models. In addition, spreadsheet and manual
calculations are insufficient for handling the unsteady flow and gradually varied flow
calculations that are required in many cases.
Although engineering judgment cannot, and should not, be removed from the design
process, some software applications provide constraint-based design facilities that can
produce a design that will meet a set of constraints specified by the user (Haestad
Methods, 2001). The factors that are considered by the automated design algorithm
include
• Percent full at design flow
• Range of pipe sizes considered
• Matching of pipe inverts or crowns at manholes
• Use of a drop structure if the maximum slope is exceeded
• Placement of pipe crown below the top of existing structures
• Placement of pipe invert below the bottom of an existing invert
• Ensuring downstream pipe not smaller than the upstream pipe
• Ensuring slope is less than the maximum allowable
• Ensuring minimum cover is provided at each structure.
The design engineer should not simply accept an automated design, but rather should
carefully review it to ensure that it meets design criteria and determine if some adjust-
ments in parameters can yield a better-functioning or lower-cost design.
Steady Flow versus Extended-Period Simulation

(EPS) Analysis
As discussed in Chapter 1, modeling can be useful for evaluating system design.
Models can range in complexity from simple spreadsheets to sophisticated dynamic
solvers. The engineer should choose the appropriate model for the task at hand.
Depending on the problem and complexity of proposed solutions, a steady-state
model, an unsteady (extended-period simulation) model, or a fully dynamic model
with real-time control capability may be the appropriate design tool.
In a steady-state model, flows and boundary conditions are constant in time at any
point in the system, and the steady-state gravity-flow equations described in
Chapter 2 apply. In reality, flows are constantly changing throughout the day, but for
many analysis and design situations only a specific, single flow rate, such as the peak
or minimum, is needed. This snapshot of the flow at the desired time (e.g., time corre-
sponding to design maximum flow rates) is then analyzed; in this case, to determine
needed pipe capacity.
Extended-period simulation (EPS) models show how a sewer network will behave over
time. This type of analysis can be used to examine how wet wells fill and drain; how
pumps toggle on and off; and how pressures, hydraulic grades, and flow rates change
in response to variable loading conditions and automatic control strategies. EPS is a
useful tool for assessing the hydraulic performance of different pump and wet well
sizes.
In wastewater collection systems, unsteady flow is often the result of wet weather
events. Stormwater is supposed to be excluded from new sewer systems, and flow
routing may not be required for new sewer design. Instead, the peak hourly flows are
generally used to determine the sizes of the laterals and smaller sewers (Metcalf &
Eddy, Inc., 1981), and a steady-state simulation with peak hourly flows may be per-
formed to assess the capacity of a designed or existing line. A second steady-state run
with minimum design flow rates can determine if the velocities at these low flow rates
are adequate to prevent solids buildup.
For long-term planning studies, it is unrealistic to try to establish short-term inflow
hydrographs for future conditions. Therefore, steady-state runs are usually conducted
for several different future times to assess the effects of increased flows and new areas
coming into service. These steady-state runs represent the worst-case conditions
expected in the system.
EPS may be used in the design of new gravity sewers to analyze the effects of the fol-
lowing conditions:
• Wet weather peaks as they move downstream in large systems
• Draining of batch tanks by industrial customers
• New sewers tying into existing areas with inflow and infiltration (I/I)
problems
• Worst-case I/I projections for planned sewers
• Effect of wet well level fluctuations on gravity sewers flowing into wet wells
• Fluctuations of flow and depth in sewers downstream of a force main as a
result of pump cycling
• Adequacy of sizing and cycle time for pump station wet wells.
When EPS is used, sanitary loads are typically simulated using 24-hour pattern loads,
as described in Chapter 6. Hydrographs or pattern loads for industrial customers
should be based on site-specific flows. Wet weather flows are assigned to the model
with hydrographs developed using the techniques described in Chapter 7. The
hydraulic time step is typically one hour for diurnal fluctuations, but shorter time
steps may be required when flows are highly variable, such as for short-duration
storms or pump cycling. The sensitivity of the model to changes in the time step can
be explored by comparing the results of the analysis using different increments. The
time step for the input hydrographs (hydrologic time step) used for flow routing
should be less than or equal to the hydraulic time step used for depth calculation. For
cyclic events, the length of the simulation should be at least several loading cycles.
The duration of the analysis should be long enough that results do not vary between
cycles.
Design Maximum Flow Rates with Pumping

The design of gravity sewers downstream of a pumping station is challenging
because the design maximum flow is not the flow upstream of the station, but rather
the maximum pumping flow. Instead of a smooth hydrograph, the pipe experiences a
step flow, as shown in Figure 10.11. In this case, the pump discharge is 500 gpm (32 L/
s), while the average inflow is 150 gpm (9.5 L/s). For open-channel pipes immediately
downstream of a lift station or at the end of a force main, the peak pump discharge
should be used as the design maximum flow rate. However, as the flow proceeds
downstream, pump cycling becomes less of an influence, and the hydrograph is
dampened until the peak flow rate approaches that of the inflow to the pumping sta-
tion (flattest line in Figure 10.11). The most conservative design practice is therefore to
use the peak pump discharge to size downstream mains; however, an unsteady flow
routing model (described in Chapter 3) can account for downsteam dampening
effects and may be a more appropriate analysis tool in many instances.
Downstream
500
Near Pump
Pump
400
Flow, gpm
300
200
100
0
0 1 2 3 4
Time, hr
Figure 10.11 Example of pump station flow attenuated at downstream points in a
sewer.
10.8 Special Installations

Sewers must sometimes be installed in steep terrain or along or across streams or
other obstructions. The subsections that follow describe some special installation
types.
Sewers in Steep Terrain

If steep terrain results in velocities in excess of 15 ft/s (4.6 m/s), the engineer should
consider using more durable pipe materials such as ductile iron and an energy-dissi-
Section 10.8 Special Installations 359
pation manhole at the end of the steep slope. Alternatively, flatter slopes can be main-
tained through the use of drop manholes, as illustrated in Figure 10.12. With a drop-
manhole configuration, pipes are commonly sloped to prevent velocities in excess of
approximately 10 ft/s (3 m/s). The water drop in the manhole dissipates energy with-
out causing extremely high velocities, which are both a hazard to workers in the man-
hole and abrasive to the pipes. However, drop manholes do cause turbulence, which
can release hydrogen sulfide from the wastewater, resulting in odor and corrosion
problems. ASCE (1982) provides additional guidance on drop manholes.
Ground Manhole
Flow
Drops
(typical)
Sewer
Figure 10.12 Application of drop manholes in a sewer crossing steep terrain.
Sewers Along Streams

When sewers must be routed along waterways, extra care is needed to protect the
sewer from erosion and breakage. Sewers placed parallel to streams or drainage chan-
nels are typically constructed of lined ductile iron or plastic pipe. The sewer is nor-
mally still designed as an open-channel flow conduit, but an inverted siphon may be
required when crossing under a waterway. If the probability of breakage and spills is
deemed significant, the streambank may need to be stabilized to prevent washout of
the pipe. Stabilization is preferably achieved using a method based on the stream’s
natural geomorphology. Manholes along streams often have their chimneys extended
above the level of expected flooding to prevent flood waters from entering through
the top of the manhole. In larger streams in cold climates, extended chimneys may
need to be protected from ice floes.
Elevated Crossings
Sometimes the best solution to crossing an obstruction such as a canal, stream, or
gully is use an elevated, above-grade structure, as shown in Figure 10.13. In some
cases, open-channel flow can be maintained. If the flow must be pressurized at the
crossing (i.e., there is a sag in the pipe), then it is an inverted siphon (see next sub-
section).
A true elevated sewer will have the same slope as the upstream and downstream
pipes. The designer should examine the ground profile to identify the segment of pipe
that is above grade. Special pipe materials and construction techniques may be
required for the exposed pipe. For example, some plastic pipes may be damaged by
ultraviolet light, while metal pipe may need a wrap or coating.
Ground
Manhole
Pier
Sewer
Stream
Figure 10.13 Profile of an elevated stream crossing.
Inverted Siphons (Depressed Sewers)

An inverted siphon is a sag in a sewer used to avoid an obstruction such as a gulley,
stream, canal, depressed railway, or roadway. However, this type of structure is not
really a siphon, and the term depressed sewer has been suggested as more appropriate
(Metcalf and Eddy, 1981). Because the pipe(s) comprising the inverted siphon are
below the hydraulic grade line, it always contains sewage, and its hydraulic behavior
is that of a pressure-flow conduit. An example of a depressed sewer is shown in
Figure 10.14.
Minimum diameters for depressed sewers are the same as for gravity sewers: 6 or 8
inches (150 or 200 mm). Depressed sewers require a larger sediment carrying capacity
for self-cleansing than do normal, open-channel sewers, because it is desirable to
scour larger particles out of the siphon to avoid buildup and clogging, and also
because particles must be suspended in the turbulent flow to be successfully
Channel
Slope Slope
Bottom
12 in. cast iron pipe
18 in. sewer 18 in. sewer

8 in. cast iron pipe
Direction of flow
Rim elevation = 64.04
65 Ground Line
Rim elevation = 63.34

60
Elevation (ft)
55
50 S
=0 Channel Bottom 8
.66 81
Invert elevation = 51.94 .66 Invert elevation = 52.74
81
8 =0
45 8 in. pipe invert = 51.94
S
S = 0.00000
12 in. pipe invert = 52.11 8 in pipe invert = 52.44
40 12 in. pipe invert = 52.64
Invert elevation = 41.0
120 130 140 150 160 170 180 190 200
Station (ft)
Graphic reprinted with permission from Pearson Education
Figure 10.14 Plan and profile of a depressed sewer (inverted siphon) used to avoid
obstructions.
Section 10.9 Wastewater Collection System Optimization 361
removed. To make them as maintenance free as possible, velocities of at least 3 ft/s (0.9
m/s) need to be achieved at a minimum of every few days. Higher velocities can be
induced by placing a splitter box with different weir levels at the inlet and using two
or more parallel pipes in the siphon section. All flow goes into the smaller pipe until it
surcharges, then the overflow spills into the next pipe, and so on. Depressed sewers
are usually constructed using lined ductile iron or plastic pressure pipe.
The hydraulic design is best handled by applying the energy equation from the inflow
section at the entrance to the depressed section, through the transitions (bends), to the
outlet where flow enters the open-channel pipe. Headloss resulting from changes in
velocity at the entrance must be accounted for. If bends in the pipe are greater than 10
to 20 degrees, allowances must be made for losses at the bends (Metcalf & Eddy, Inc.,
1981).
To model depressed sewers and obtain good profile maps, it may be necessary to
insert nodes or imaginary manholes along the siphon corresponding to points where
the slope of the pipe changes. The invert elevation at the manhole would be the same
as at the siphon, but manholes would have no or minimal minor loss associated with
them, and the top of the manhole would extend above the expected maximum
hydraulic grade line.
Based on both theoretical and laboratory analyses, May (2003) determined that the ris-
ing leg of the siphon could either be too steep or too shallow, which could result in
accelerated deposition of solids. He recommended slopes between 22.5 and 45
degrees to prevent deposition.
10.9 Wastewater Collection System Optimization

Engineers must constantly weigh the cost implications of project decisions with the
goal of providing a fully functional, high-quality project at the lowest cost. This is
done by analyzing wide range of alternative layouts, pipe sizes, and invert elevations.
Optimization consists of using a computerized method to automatically analyze a
range of alternatives and find the one that accomplishes the project objectives for the
least cost.
Sewer system design is not readily amenable to mathematical programming optimi-
zation techniques such as gradient search, linear programming, or dynamic program-
ming. Most optimization methods applied to sewer system design are a combination
of some mathematical programming technique and a heuristic algorithm to arrive at
the least-cost design.
A great deal of research on the application of optimization to collection systems
occurred in the 1970s, with papers by Argaman, Shamir, and Spivak (1973); Barlow
(1972); Dajani and Hasit (1974); Deb (1974); Deininger (1970); Liebman (1967); Mays
(1975); Mays and Yen (1975); Meredith (1971); Merritt and Bogan (1973); Tang, Mays,
and Yen (1975); and Yen and Sevuk (1975). The most promising of these was Maysʹ
approach, which used discrete dynamic programming to trade off pipe costs and
excavation costs. However, it required data on the sensitivity of costs to excavation
depth, which were not usually available.
Early optimization models were not widely adopted by practicing engineers for sev-
eral reasons, including lack of user-friendly interfaces and difficulties with handling
nonlinear cost and hydraulic equations. In addition, practicing engineers often are not
interested in the least-cost solution as much as they are in providing conservative

safety factors in their design. Optimization tends to remove any conservatism from
design.
Walters (1985) and Tekeli and Belkaya (1986) developed models for sewer system lay-
out which were expanded upon by Lui and Matthew (1990) to include hydraulics as
well as layout. Greene, Agbenowosi, and Loganathan (1999) produced a GIS-based
method to identify the optimal route for sewers given manhole locations in a GIS.
Desher and Davis (1986) published one of the early microcomputer models of a collec-
tion system and showed that costs could be reduced by 20 percent through a sensitiv-
ity analysis of design criteria. Kulkarni and Khanna (1985) developed an algorithm to
optimize force mains. Elimam, Charalambous and Ghobrial (1989) produced a heuris-
tic algorithm to design sewers and used a digitizer to build the network. Charalam-
bous and Elimam (1991) extended this work to include force mains.
More recently, Swamee (2001) proposed a method based on Lagrangian multipliers,

while Vollertsen, Hvitveld-Jacobsen and Talib (2002) developed an approach which
purported to account for water quality transformations in sewers. Gill, Parker, Savic
and Walters (2001) developed a model that minimizes pollution from combined sewer
overflows.
Genetic and evolutionary algorithms have been applied to the optimization of sewer
and drainage systems. Cembrowicz and Krauter (1987) applied the evolutionary strat-
egy to optimize pipe sizes and dendritic layout for a gravity sewer system. Parker et
al. (2000) coupled a genetic algorithm with an artificial neural network (ANN) trained
with a hydrodynamic model and applied it to minimizing combined sewer overflow.
Diogo et al. (2000) developed a fairly comprehensive model integrating a genetic algo-
rithm for general system layout with a dynamic programming and knowledge-based
approach for further optimizing the detailed design of the optimal layout. Savic and
Walters (2001) developed a genetic algorithm optimization that used simplified rout-
ing based on the kinematic wave method to speed the solution, as compared with
solving the Saint-Venant equations.
While automated design procedures (Haestad Methods, 2003) are, strictly speaking,
not optimization models, the results from such procedures can serve as a starting
point for trial-and-error adjustments of pipe sizes and invert elevations to minimize
costs. That is, the engineer
1. Performs the initial constraint-based design

2. Estimates the cost for the initial design
3. Modifies or further constrains the diameters, invert elevations, or other param-

eter and checks hydraulic feasibility with the model
4. Performs another cost estimate
This process continues until the engineer is satisfied that the design cannot be sub-
stantially improved.
Even though optimization models have shown promise, they have yet to reach the
level of being widely accepted in engineering practice.
Section 10.9 Wastewater Collection System Optimization 363
Example of Optimization/Sensitivity Analysis

Formal optimization to minimize sewer construc- ward, but the optimized design, based on essen-
tion costs has not been practiced widely for a tially the same design values and constraints as
variety of reasons (Merritt, 1999). However, case the original design, was priced about 17 percent
studies have shown that optimization cost savings lower—about one million dollars. This 17-percent
of 5 to 10 percent are often achievable, com- cost differential resulted from the optimization
pared to good designs that were not optimized. model's ability to select the best (least-cost) com-
In addition, cost-sensitivity studies on design bination of pipe sizes and slopes that met all con-
parameters such as flow rate, Manning's n-value, straints and criteria established for the design. In
minimum depth, and self-cleansing particle size this case, nearly all of the savings resulted from
often lead to much larger cost savings. The latter pipe-size reductions.
case is actually similar to a value-engineering The optimal design placed the lines at slightly
sensitivity analysis. A sewer design model is used steeper slopes in some reaches to allow smaller-
to determine the impact of adjustments in design diameter pipes to carry the design flows. The cost
parameters on project costs. Curves can be pre- of the increased depths were far smaller than the
pared relating cost and any design parameter. pipe cost savings, yielding a large net cost reduc-
Designers and decision makers then consider these tion. This 17-percent reduction due to optimiza-
results to decide whether changes in the design cri- tion was rather large in this case, but similar case
teria and constraints are desirable in light of the studies typically show a 5 to 10 percent savings
changes in costs associated with different values. from optimization.
A sewer project designed in 1998 and con- Next, the sensitivity of costs to values of Man-
structed in 1999 consisted of replacing a smaller ning's n were determined, and a curve of cost
sewer main with a new 48-in. (1200-mm) inter- versus n was developed. The figure below shows
ceptor sewer, along a 3.1-mi (5-km) route. Two that costs could be reduced by 5 percent by
additional 24-in. (600-mm) replacement mains, reducing Manning’s n from 0.013 to 0.011.
each about 0.5 mi (0.8 km) long, joined the new
Even larger savings were made possible by
trunk line along its route and were part of the
removing some of the conservatism from pro-
project. The project design was done in a “stan-
jected flow rates. By preparing curves relating
dard” fashion by a capable and experienced
values such as design flow to cost, the owner can
engineering firm. The completed sewer was built
gain an appreciation of the cost for reduction of
in accordance with appropriate codes and will
risk. For example, a 20-percent increase in the
serve the area well.
design flow resulted in only a 9-percent increase
An optimization and cost-sensitivity study was in cost, while a 20-percent reduction would result
performed for this project to provide information in a savings of 13 percent. Optimization and
for value engineering considerations. The study cost-sensitivity plots clarify the situation and pro-
was conducted using the Merritt Optimizing vide accurate information from which to make
Design System (MODS) (Merritt, 2000). This final design decisions.
sewer project appeared to be rather straightfor-
6
Cost, Million $
5.5
4.5
0.011 0.012 0.013 0.014 0.015
Manning’s n
References
American Society of Civil Engineers (ASCE). 1982. Gravity Sanitary Sewer Design and
Construction. Manual of Practice 60 (also WEF MOP FD-5). Reston, VA: American
Society of Civil Engineers.
Argaman, Y., U. Shamir, and E. Spivak. 1973. Design of optimal sewerage systems.
Journal of Environmental Engineering Division, ASCE 99, EE6, 703.
Barlow, J. F. 1972. Cost optimization of pipe sewerage systems. Proceedings of the
Institution of Civil Engineers (London), no. 2: 57.
Burton, F. L. 1996. Wastewater collection systems. In Water Resources Handbook, edited
by L. W. Mays. New York: McGraw-Hill.
Cembrowicz, P. G., and G. E. Krauter. 1987. Design of cost optimal sewer networks.
Proceedings of the Fourth International Conference on Urban Storm Drainage
(Lausanne, Switzerland), 367–72.
Crites, R. and G. Tchobanoglous. 1998. Small and Decentralized Wastewater Treatment
Systems. New York: McGraw-Hill.
Dajani, J. S., and Y. Hasit. 1974. Capital cost minimum drainage networks. Journal of
Environmental Engineering Division, ASCE 100, EE2, 325.
Deb, A. K. 1974. Least cost design of branched pipe network system. Journal of
Deininger, R. 1970. Systems analysis for water supply and pollution control. In
Natural Resource Systems Models in Decision Making, edited by G. H. Toebes. Water
Resources Center, Purdue University.
Diogo, A. F., G. A. Walter, E. R. de Sousa, and V. M. Graveto. 2000. Three-dimensional
optimization of urban drainage systems. Computer-Aided Civil and Infrastructure
Engineering 15, 409–426.
Elimam, A. A., C. Charalambous, and F. H. Ghobrial. 1989. Optimum design of large
sewer networks. Journal of Environmental Engineering 115, no. 6: 1171.
Gill, E., M. A. Parker, D. A. Savic, and G. A. Walters. 2001. Cougar: A genetic
algorithm and rapid integrated catchment modeling application for optimizing
capital investment in combined sewer systems. World Water & Environmental
Resources Congress (Orlando, Florida).
Greene, R., N. Agenowosi, and G. V. Loganathan. 1999. GIS-based approach to sewer
system design. Journal of Surveying Engineering 125, no. 1: 36.
Great Lakes and Upper Mississippi River Board of State Public Health and
Environmental Manager (GLUMRB). 1997. Recommended Standards for Wastewater
Facilities. Albany, NY: Great Lakes Upper Mississippi River Board.
Haestad Methods. 2001. SewerCAD Sanitary Sewer Modeling Software. Waterbury, CT:
Haestad Methods.
Haestad Methods. 2003. SewerCAD Sanitary Sewer Modeling Software. Waterbury, CT:
Haestad Methods.
Hammer, M. J., and M. J. Hammer, Jr. 2001. Water and Wastewater Technology, 4th ed.
Upper Saddle River, NJ: Prentice-Hall.
Liebman, J. C. 1967. A heuristic aid for the design of sewer networks. Journal of
Sanitary Engineering Division, ASCE 93, SA 4, 81.
References 365
Lui, G., and R. G. S. Matthew. 1990. New approach for optimization of urban drainage
systems. Journal of Environmental Engineering 116, no. 5: 927.
May, R. P. 2003. Preventing sediment deposition in inverted sewer siphon. Journal of
Hydraulic Engineering 129, no. 4: 283
Mays, L. W. 1975. Optimal layout and design of storm sewer systems. Ph.D. Diss.,
University of Illinois, Urbana.
Mays, L. W., and B. C. Yen. 1975. Optimal cost design of branched sewer systems.
Water Resources Reasearch 11, no. 1: 37-47.
Meredith, D. D., 1971. Dynamic programming with case study on planning and
design of urban water facilities. In Treatise on Urban Water Systems. Colorado State
University.
Merritt, L. B. 1999. Obstacles to lower cost sewer systems. Sewer Design Paper #99-1.
Provo, UT: Civil and Environmental Engineering Department, Brigham Young
University.
Merritt, L. B. 2000. Example cost sensitivity curves from case studies. In Proceedings of
WEAU Annual Conference. Salt Lake City, UT: Water Environment Association of
Utah.
Merritt, L. B., and R. H. Bogan. 1973. Computer-based optimal design of sewer

systems.” Journal of Environmental Division, ASCE 99, EE1, Proc. Paper 9578.
Metcalf & Eddy, Inc. 1981. Wastewater Engineering: Collection and Pumping of
Wastewater. New York: McGraw-Hill.
Parker, M. A., D. A. Savic, G. A. Walter, and Z. Kappelan. 2000. SewerNet: A genetic
algorithm application for optimizing urban drainage systems. International
Conference on Urban Drainage.
Savic, D. A., and G. A. Walters. 2001. Evolutionary computing in water distribution
and wastewater systems. World Water and Environmental Resources Congress
(Orlando, Florida).
Tang, W. H., L. W. Mays, and B. C. Yen. 1975. Optimal risk-based design of storm
sewer networks. Journal of Environmental Engineering Division, ASCE 101, EE3, 381.
Tekeli, S., and H. Belkaya. 1986. Computerized layout generation for sanitary sewers.
Journal of Water Resource Planning and Management 112, no. 4: 500.
US Environmental Protection Agency (US EPA). 1991. Alternative Wastewater Collection
Systems. EPA 625/1-91/024. Washington, DC: US Environmental Protection
Agency.
Walters, G. A. 1985. The design of the optimal layout for a sewer network. Engineering
Optimization, no. 9: 37.
Yen, B. C., and A. S. Sevuk. 1975. Design of storm sewer networks. Journal of
Problems
10.1 You are trying to decide between using 21- or 24-in. pipes for a sewer that will
be laid at a slope of 0.004 with a Darcy-Weisbach roughness height of 0.0002 ft.
Your design standards require that it be able to carry a peak flow of 14 ft3/s, but
the average flow will be about 5 ft3/s and the minimum flow will be 2 ft3/s.
First, determine the full pipe capacity for each diameter. To do this, you will
need to assume a flow, calculate the Reynolds number, determine f, and check if
the hydraulic gradient slope is 0.004. You can use the Moody diagram, Cole-
brook-White equation, or Swamee-Jain equation. Iterate until you reach a solu-
tion. Use a kinematic viscosity of 1.0 × 10–5 ft2/s. (Alternatively, you can use a
program such as FlowMaster or SewerCAD.)
Parameter 21 in. 24 in.

Velocity, ft/s
Relative roughness
Reynolds Number
Friction factor
Slope
Flow, ft3/s
From the answers in the previous table, you might think that a 21-in. pipe would
not work. But remember that a pipe’s peak carrying capacity occurs when it is
approximately 95 percent full. Using a hydraulic elements chart or a computer
program, find the velocity and depth of flow at each of the flow rates given
below. Assume that you can ignore any gradually varied flow effects (i.e.,
assume normal flow exists).
Velocity (21), Depth (21), Velocity (24), Depth (24),

Flow, ft3/s ft/s in ft/s in
2
5
14
Given the points from the previous table (and the full pipe value), draw a graph
with depth on the vertical axis and flow and velocity (for both pipe sizes) on the
horizontal axis.
At the lower flow rate, is there a significant difference in the velocity?
Which size pipe would you recommend and why?
10.2 You are required to size a single-barrel inverted siphon (depressed sewer) under
a small stream. The inverted siphon is 75 m long with a Hazen-Williams C-fac-
tor of 120. You would like to maintain a velocity of at least 0.6 m/s even at low
flow. Important elevations are shown in the following figure.
Problems 367
314.75 m
311.50 m
312.12 m
309.20 m
Try diameters of 600, 750, and 900 mm. You would like to select a pipe that will
not surcharge the upstream manhole, yet will carry a minimum flow of 300 L/s
at a velocity of 0.6 m/s and a maximum flow of 2000 L/s. The downstream grav-
ity sewer has enough capacity to set the water level in the downstream manhole
to 310.00 m.
For each diameter pipe, draw a curve of flow vs. upstream head and flow vs.
velocity. Using the graphs, select the optimal pipe size and justify your decision.
10.3 You must lay out the sanitary sewer system for a new subdivision. You have
been given the road layout and topographic map shown in the figure. Lay out
the sewer system by drawing each pipe with an arrow beside each gravity pipe
showing the direction of flow. Make sure that all gravity sewers flow downhill.
If you must pump, show the pump station and use a dashed line to show the
force main.
Some considerations for design:

• Keep the sewers in the road as much as possible
• Minimize the use of pump stations and force mains
• House laterals must drain to the front of the house
• New sewers must end at the existing sewer in the southwest corner of the
map.
It is suggested that you photocopy the attached map and draw the lines in color.
An important feature in this design is a greenbelt along a stream in the develop-

ment. The land developer wants you to create two sets of plans:
a.The first set is for the existing land-use ordinances, which allow sewers to
cross a greenbelt but not run along the greenbelt.
b.The second set of plans is for use if the developer receives a variance from
the zoning ordinance enabling him to locate the sewer along the length of
the greenbelt.
[Zoning ordinances are laws that restrict the construction activities that can
occur in each zone. Land owners can request variances which enable them to
not comply with a provision in an ordinance.]
368
Design of New Gravity Wastewater Collection Systems
670 ft
660 ft
Greenbelt
650 ft
620 ft 630 ft 640 ft
Chapter 10
Existing Sewer
In Road Road N
Contour 0 300 600
Property 1 in. = 300 ft
Lines
Problems 369
10.4 You are designing a sanitary sewer to flow across a fairly flat floodplain. The
manhole spacing is 400 ft and the ground profile is shown in the following table.
Ground
Station, ft Elevation, ft
0 784.25
400 783.88
800 780.65
1200 781.85
Use a PVC pipe with a 12 in. diameter. The sewer will have Manning’s n = 0.010,
and there is 0.1 ft of head loss in each manhole. The tailwater elevation is 776.85
ft, and the manhole invert matches the inverts of the incoming and outgoing
pipes. You want at least 3 ft of cover above the top of the pipe. Pick invert eleva-
tions and check to see if the pipe is surcharged at a maximum flow of 2.0 ft3/s
and the velocity is still reasonable at a minimum flow of 0.2 ft3/s. Complete the
following tables. If you are using manual calculations or a computer program
based on normal flow, use the normal depth in the pipe table. If you are using an
approach that calculates backwater curves using a gradually varied flow analy-
sis, give both the upstream and downstream depths and the average velocity.
Manhole data
Invert
Station, ft Elevation, ft
0
400
800
1200
Pipe data
Up Down Depth at Velocity at Depth at Velocity at

Station, ft Station, ft Slope 2 ft3/s, ft 2 ft3/s, ft/s 0.2 ft3/s, ft 0.2 ft3/s, ft/s
0 400
400 800
800 1200
When you have performed the hydraulic calculations, draw a profile of the
sewer showing the ground elevation and top and bottom of the pipe. Annotate
the drawing so that it can be understood. It may be drawn manually, with a
hydraulic analysis program, or using CAD.

10

Uploaded by

Copyright:

Available Formats

10

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

10

Uploaded by

Copyright:

Available Formats

CHAPTER

File modified 8/26/04

Figure 10.1 PVC pipe ready for installation.

• Cause a minimum of interference with the hydraulics of the sewer

Figure 10.3 Manhole covers.

In combined sewer systems, it is sometimes permissible to have some type of flow

10.2 Initial Planning

Decision to Provide Sewer Service to an Area

A common problem is determining how to serve the first areas to be developed in a

Wastewater collection systems use gravity, pressure, vacuum, or some combination of

Separate versus Combined Systems

However, it is conceivable that future treatment requirements for storm waters in

10.3 Preliminary Design Considerations

10.4 Initial System Layout

Gravity Sewer Layout

Manhole Location and Spacing

Photo courtesy of R. D. Zande and Associates

Figure 10.4 Typical manhole in a cross country setting.

Location of Pumping Facilities

Example of a Sewer Network Layout

Figure 10.5 Location of subdivision boundary, existing road, and

Figure 10.6 Location of subdivision boundary, drainage divide, and natu-

Edge of Right of Way

Figure 10.7 Location of subdivision boundary with proposed roads,

10.5 Flows in Sanitary Sewers

Low Flows in Early Years

Allowances for Infiltration and Inflow

What Are Minimum Flow and Velocity?

Low Flow Velocity in 8-in. Sewer

10.6 Horizontal and Vertical Alignment

Curved Sewer Alignment

where R = radius of curvature (ft, m)

Minimum Depth of Cover

10.7 Hydraulic Design

Shallow sewer 300,00 1 300,000

1. Make an initial estimate of diameter.

Small-Diameter Sewers. The smallest pipe diameter traditionally specified in the

For smaller sewer sizes, prohibition of sewer-size reduction downstream is common,

Photo courtesy of R. D. Zande and Associates

Computer Modeling for System Design

Steady Flow versus Extended-Period Simulation

Design Maximum Flow Rates with Pumping

10.8 Special Installations

Sewers in Steep Terrain

Sewers Along Streams

Figure 10.13 Profile of an elevated stream crossing.

Inverted Siphons (Depressed Sewers)

12 in. cast iron pipe

18 in. sewer 18 in. sewer

Rim elevation = 64.04

Rim elevation = 63.34

120 130 140 150 160 170 180 190 200

10.9 Wastewater Collection System Optimization

interested in the least-cost solution as much as they are in providing conservative

More recently, Swamee (2001) proposed a method based on Lagrangian multipliers,

1. Performs the initial constraint-based design

3. Modifies or further constrains the diameters, invert elevations, or other param-

Example of Optimization/Sensitivity Analysis

Merritt, L. B., and R. H. Bogan. 1973. Computer-based optimal design of sewer