M.tech Thesis Opt Aq
M.tech Thesis Opt Aq
M.tech Thesis Opt Aq
CREATIVITY OF DESIGNERS
S.N.SATYANARAYANA
00012D0122
M.Tech
STRUCTURAL ENGINEERING
CERTIFICATE
This is to certify that the thesis / dissertation entitled “OPTIMISING
AQUEDUCTS -A CHALLENGE TO CREATIVITY OF DESIGNERS” that is
The results embodied in this thesis have not been submitted to any
other University or Institute for the award of any degree or diploma.
Dr.N.V.RAMANA RAO ,
Professor of Civil Engineering
J.N.T.U. College of Engineering,
Hyderabad
ACKNOWLEDGEMENTS
ABSTRACT
Thotapalli barrage project across Nagavali River in Vizianagaram
District to irrigate an area about 1,20,000 Acres under E.P.C turn key
system of worth Rs.450 Crores,
Srikakulam branch canal, to irrigate an area of about 42,300 Acres
takes off at KM 96.000 of Thotapalli barrage right main canal, passes over
local valley in between Km 9.00 to KM 10.00 where a 625m length of
aqueduct is proposed to carry water for irrigation purpose.
Marcus Agrippa had to deal with only simple open channel hydraulics, be
content with the stone and brick masonry as materials of construction and
adopt without option , arches to span relatively small spans. The creativity
and imagination needed to blend and extend the then existing knowledge by
adopting intuitive solutions were still within the limits of what one man
could master - but even then, he was a rare specimen.
Ancient Indian aqueduct in Hampi(figure-2)
Cross Section
(c) Use of continuous spans over three span and four span units reduced
the longitudinal bending effects making the system cheaper as
compared to the simply supported option.
The net savings over the conventional design were of the order of 40%.
Design & Construction of Road Bridge cum Aqueduct across the river
Narmada near Omkareshwar in Madhya Pradesh State(Figure-4)
The Aqueduct was designed with a trapezoidal cross section and the sub structure
consists of RCC circular Piers with hammer head pier cap
2.0General
2. 1 Conveyance Structures for Canal Flows
A canal conveying water from the head works has to run for large
distances and has to maintain the water levels appropriately, as designed
along its length. It has to run through terrains which generally would have a
different slope smallthan the canal . The surrounding areas would
invariably have its own drainagesystem ranging from small streams to large
rivers . The canal has to carry the water across these water bodies as well as
across artificial obstacles like railwayline or roads .
1. Pipe conduits, culverts and inverted syphons to carry flow under railways
and highways .
2. Aqueducts, syphon aqueducts, super-passage, canal siphon or level
crossings
across natural drainage courses or other depressions.
3. Transitions at changes in cross sections.
This topic deals with the concepts of planning, layout and design of canal
structures for flow conveyance across artificial and natural obstacles .
2.2 Structures for crossing canals across roads and railwaylines
The inverted syphon is a closed conduit designed to run full and under
pressure .If made of pressure pipes , they should be able to withstand the
load of cover and wheel from outside and the hydrostatic head from inside .
Transitions for changes in cross sections are nearly always used at inlet and
outlet of a siphon to reduce head losses and prevent erosion in unlined
canals caused by the velocity changes between the canal and the pipe.
2.3 Structures for crossing canals across natural streams
(cross drainage works)
These structural elements are required for conveying the canals
across natural drainage. When a canal layout is planned, it is usually seen
to cross a number of channels draining the area, varying from small and
shallow depressions to large rivers.It is not generally possible to construct
cross-drainage structures for each of the small streams. Some of the small
drainage courses are, therefore, diverted into one big channel and allowed to
cross the canal. However, for larger streams and river, where the cost of
diversion becomes costlier than providing a separate cross-drainage work,
individual structures to cross the canal across the stream is provided.
There could be a variety of combinations of the relative position of the canal
with respect the natural channel that is to be crossed. These conditions are
shown in Figures 5 to 9. The notations used in the figures are as follows:
(a) CBL: Canal Bed Level;
(b) SBL: Stream Bed Level;
(c) FSL: Canal Full Supply Level; and
(d) HFL: Stream High Flood Level
Figure 7 shows a canal with full supply level almost matching the high flood
level of the naturalstream.
Figure 8 shows a canal full supply level and bed levels below the levels of
high flood level and bed level of stream, respectively.
Figure 9 shows the relative position of canal with respect to the natural
stream where the canal full supply level is below the stream bed level.
In general , the solution for all the illustrated conditions possible for
conveying an irrigation canal across a natural channel is by providing a
water conveying structure which may:
(a) Carry the canal over the natural stream;
(b) Carry the canal beneath the natural stream; or
(c) Carry the canal at the same level of the natural stream.
These three broad types of structures are discussed further in this lesson .
2.4 Structures to carry canal water over a natural stream
Conveying a canal over a natural watercourse may be accomplished in
two ways:(a) Normal canal section is reduced to a rectangular section and
carriedacross the natural stream in the form of a bridge resting on piers and
foundations (Figure 10) . This type of structure is called a trough type
aqueduct .
For more abrupt changes, like a normal canal section being changed to a
vertical walled aqueduct, suitable transitions have been designed which
would avoid formation of any hydraulic with consequent loss of energy. A
typical view of transition of a normal canal bank to a vertical walled flume
section is shown in (Figure 20).
As may be observed, the banks of the normal canal section are first changed
to vertical walls keeping the same canal bed width (B c). Beyond this, the
B
curve, that is, an equation deciding the width (B x) at any distance x from
B
the start of the fluming, assuming a length L for the transition. One formula
that is commonly used for this kind of transition is the UPIRI method,
commonly known as Mitra’s transition and is given as follows:
B x = (Bc * Bf *L ) / (L * Bc – X (Bc – Bf)) (1)
B
O. The values for the bed width Bx at any length X from the start of the
transition and the corresponding side slope mx are given by the following
expressions
B = Bc + X/L [1- (1- X/L)n] (Bc -Bf ) (2)
m x = m0 [1 – (1- X/L)1/2] (3)
Where n = 0.8 - 0.26(m0)½ and the length of transition L, is expressed as
L = 2.35 (Bc-Bf) + 1.65 m0 h c (4)
Chapter-3
AQUEDUCTS
INTRODUCTION :An aqueduct is an artificial channel that is constructed to convey water
from one location to another. The word is derived from the Latin aqua, "water," and ducere
("to lead"). The word is also used for any bridge that carries water, similar to viaducts,
though they carry water instead of a road or railway. Sufficiently large aqueducts may also be
usable by boats or ships. While a road bridge often carries the roadway at a more elevated
level than the rest of the road, such a variation of height is not possible for an aqueduct.
Roman aqueduct supplying Carthage, Tunisia
Ancient aqueducts
Roman aqueducts
Roman aqueducts were built in all parts of the Roman Empire, from Germany to Africa, and
especially in the city of Rome itself, where they totaled over 260 miles (416 km). The
aqueducts were important for supplying water to large cities across the empire, and they set a
high standard of engineering that was not surpassed for more than
View from inside a Roman aqueduct from the Pools of Solomon to Jerusalem
Although famously associated with the Romans, aqueducts were devised much earlier in the
Near East and Indian subcontinent, where peoples such as the Egyptians and Harappans built
sophisticated irrigation systems. Roman-style aqueducts were used as early as the 7th century
BC, when the Assyrians built a limestone aqueduct 30 feet (10 m) high and 900 feet (300
m) long to carry water across a valley to their capital city, Nineveh. The full length of the
aqueduct ran for 50 miles (80 km).
In the new world, when the Aztec capital of Tenochtitlán was discovered in the middle of the
second millennium, it was watered by two aqueducts.
Aqueducts in Persia
In Persia from early times[vague] a system of underground aqueducts called Qanat were
constructed, a series of well-like vertical shafts, connected by gently sloping tunnels. This
technique:
Taps into subterranean water in a manner that efficiently delivers large quantities of water to
the surface without need for pumping. The water drains relying on gravity, with the
destination lower than the source, which is typically an upland aquifer.
Allows water to be transported long distances in hot dry climates without losing a large
proportion of the source water to seepage and evaporation
Aqueducts in India
Ancient Indian aqueduct in Hampi
Indian subcontinent was one of the earliest builders of aqueducts. More prominent evidence
can be found at the sites of present day Hampi. The massive aqueducts near river
Tungabhadra supplying irrigation water were once 15 miles long[1].The elegant water ways in
royal center supplied water for royal bath houses.
Design & Construction of Road Bridge cum Aqueduct across the river Narmada near Omkareshwar in
Madhya Pradesh State-E.P.C.CONTRACT.
Construction of Aqueduct cum Road Bridge to carry Kharsia Canal Waters across a valley in Chattisgarh.
Remains of the Schoharie Crossing, an aqueduct that once carried the Erie Canal over the
Schoharie Creek near Amsterdam, New York.
Mathur Aqueduct in Tamilnadu state, India carries irrigation water
In modern times the largest aqueducts of all have been built in the United States to supply
that country's biggest cities. The Catskill Aqueduct carries water to New York over a
distance of 120 miles (190 km), but it is dwarfed by aqueducts in the far west of the country,
most notably the Colorado River Aqueduct, which supplies the Los Angeles area with water
from the Colorado River nearly 250 miles (400 km) to the east, and the 444 mile (714.5 km)
California Aqueduct which runs from the Sacramento Delta to Lake Perris.
[edit] Uses of aqueducts
Historically, many agricultural societies have constructed aqueducts to irrigate crops.
Archimedes invented the water screw to raise water for use in irrigation of croplands.
Another widespread use for aqueducts is to supply large cities with clean drinking water.
Some of the famed Roman aqueducts still supply water to Rome today. In California, USA,
three large aqueducts supply water over hundreds of miles to the Los Angeles area. Two are
from the Owens River area and a third is from the Colorado River.
In more recent times, aqueducts were used for transportation purposes to allow canal barges
to cross ravines or valleys. During the Industrial Revolution of the 18th century, many
aqueducts were constructed as part of the general boom in canal-building.
In modern civil engineering projects, detailed study and analysis of open channel flow is
commonly required to support flood control, irrigation systems, and large water supply
systems when an aqueduct rather than a pipeline is the preferred solution. The aqueduct is a
simple way to get water to other ends of a field.
In the past, aqueducts often had channels made of earth or other porous materials. Significant
amounts of water are lost through such unlined aqueducts. As water gets increasingly scarce,
these canals are being lined with concrete, polymers or impermeable soil. In some cases, a
new aqueduct is built alongside the old one because it cannot be shut down during
construction.
[edit] Notable aqueducts
[edit] Ancient Greek aqueducts
The Eupalinian aqueduct on the Greek island of Samos.
[edit] Roman aqueducts
The Magdeburg Water Bridge seen from the shores of the Elbe.
[edit] See also
Aztec the Aztec Empire also used aqueducts
Drought
Earthquake engineering
Flow
Goldfields Water Supply Scheme
Irrigation
Leat
Pipeline - some used to carry water
Roman aqueduct
List of Roman aqueduct bridges
Roman architecture
Roman engineering
Sanitation in Ancient Rome
Viaduct - a similar structure to carry a road or a railway
Water resources
TYPE OF AQUEDUCTS:
Aqueducts arc classified into 3 Types.
Type 1 : In this type the sides are earthen banks. The length of the barrels through
which the drainage water is passed under the canal must be of sufficient length to carry
the entire section of the canal including outer slopes as shown in Fig.
The selection of the type of aqueduct or syphon aqueduct mainly depends upon the
magnitude of the drainage to be passed and the size of the canal. A very small drainage will
obviously be passed by a work of Type I, which in an extreme case is merely a pipe passing
under the bed of a canal. Over a large river or stream an aqueduct of Type III is obviously
economical. Type II is intermediate. Choice of this type merely depends if it can work out
economical in comparison to Type I. It may be noted that in Type 1 and Type II the canal is
not flumed and therefore is no head loss and Type III needs loss of head due to fluming of
canal cross section.
An Aqueduct consists the following basic structural components
Trough
Beams
Pier Cap
Pier
Toe Wall
The design of a aqueduct like any other hydraulic structures consists of two phases.
Hydraulic Design : The following are the mam features of hydraulic design in respect surface
flow.
Scour calculations
For the given requirement of discharge and the available hydraulic gradient a large number of
cross sectional shapes will be hydraulically satisfactory and acceptable. The range of
solutions are narrowed down by considerations listed below :
(1) For canals of small cross section and of short length it is preferred not to change the
section and take the canal with its embankments unchanged over the cross drainages.
The waterway for cross-drainage may be a pipe-culvert or masonry arch culvert or
RCC box culvert. This has the advantage of not requiring transition sections and thus
reduce entry and exit losses for the canal If the base width of the embankment is large
it can be restricted as shown in Fig.
(2) For longer lengths of aqueducts and for large cross-sections of canals the above
system becomes uneconomical and one has to resort to minimising the weight of the
water carried over the structure by permitting higher velocities of water. This
reduction in area of the cross-sectional shape of the water body necessitates provision
of entry and exit transition zones.
(3) The hydraulic design of transition zones is an important consideration in the overall
design since the major portion of available gradient is consumed by them. Although,
large body of data is available for estimating the losses, there is scope for additional
studies.
The hydraulic design of the aqueduct proper consists of choosing the maximum
permissible velocities which do not damage the material of the water carrying,
conduit while staying within the overall head loss restrictions taking transition and
aqueduct losses together. Within these limits also a number of alternatives will be
acceptable and further narrowing down of options will have to be on structural
considerations.
(4) The most favoured section for aqueducts carrying 'open channel flows' has
traditionally been the rectangular section of single or multiple cells. It is simple to
design hydraulically and structurally — apart from being easy for construction.
However, for pressure conduits circular cross sections have been mostly favoured.
For open sections in certain cases trapezoidal, segmental circular and even parabolic
cross - sections have been used as illustrated by examples later. The hydraulic design
of all such sections has different hydraulic efficiency, whereas from structural point
of view the efficiency is decided by the optimum use of materials in transverse design
as a primary consideration and longitudinal design as a secondary factor.
The final choice of the section is usually a compromise between hydraulic and structural
efficiency.
The total available depth for accommodating water channel and structure is
governed by the elevation of overall energy line for canal water flow at the higher elevation
and by the requirements of clearances above the HFL of the cross drainage and/or canal bed
levels at the lower elevation. This criteria does not rule out the possibility of allowing partial
submergence of aqueduct box under HFL, although it is to be avoided if possible. The
submergence causes problems of raising the upstream flood levels, apart from putting extra
horizontal forces on the structure and exposing it to impact of floating debris. Provision of
special bearings which can withstand submergence are also called for. Provision of a hump in
the canal bed level can be considered to minimise or avoid the submergence.
The following are the critical conditions to be considered:
In other situations the available structural depth can be large, but again 'the amount by which
the canal bed can be locally lowered is limited by considerations of silting. As a general rule,
trying to obtain the maximum possible structural depth within the limits of hydraulically
acceptable range is likely to lead to a more efficient longitudinal design of the structure.
However, the primary sectional choice should be based on the structural design of the cross
section. The use of single or multiple cells is governed by the capacity of the bottom slab to
span horizontally between the vertical webs and, therefore, for large discharges multiple
boxes arc adopted. For a single box section it is common practice to adopt a closed section
where top slab serves to carry the service road. For multiple boxes only one or two boxes
need to have a lop slab to carry the service road and others are left as open troughs. The
compression flanges if located on top may need lateral supports which are provided by the
intermediate struts.
In these designs the elements of transverse section such as bottom slab, side walls, top slab
etc. are primarily proportioned for bending effects. The design criteria given in national
codes for water retaining structures provide the rules for this purpose. If different shapes such
as closed pipes for pressure conduits, or segmental circles for open channel flows or any
other shape is chosen it is normally done to achieve structural efficiency in the design of
members where the membrane forces are exploited as an efficient load carrying' mechanism
taking advantage of the fact that such loads can be efficiently and economically carried by
prestressing. Reference is made to Bhima aqueduct described subsequently. At this stage of
design also, more than one section is found acceptable and are tentatively chosen for further
study.
The Longitudinal Design : The commonly adopted systems for longitudinal spanning are
shown in Fig.
Since the hydraulic continuity of water (low at expansion joints is a functional requirement,
the details of connection of one unit to the other assumes importance. The copper
strip, PVC Joint, neoprene rubble joint etc. arc some of the available solutions. It is obviously
beneficial to make them replaceable. As a first step in the choice of the span and mainly for
achieving economy in the use of materials for the superstructure, an attempt is made to use
the transversely designed section to span along the longitudinal direction. Choosing one of
the solutions indicated in Fig. If a simply supported arrangement is used then closed sections
have an edge over open sections, wherein for the sole purpose of providing compression
flange extra material is required. Spans with overhangs are ideal for open boxes where the
bottom slab carries full longitudinal compressions and need for extra material is avoided.
Closed boxes are also suitable for choice of continuous spans for obvious reasons.
The Spans thus established do not necessarily lead to the most economical solutions but one
gets an important indication about the efficiency of the use of material in the superstructure.
Longer spans than this will call either for extra materials or change in structural shape only
for longitudinal spanning. The shorter spans will indicate that the concrete section provided
for transverse design is not fully exploited for longitudinal spanning. The smaller span does,
however, require less reinforcing steel in a longitudinal direction, but this is only a partial
saving. This knowledge will help in balancing the cost of substructure and superstructure and
achieving overall economy.
Of the arrangements shown in Fig., the simply supported spans are most commonly used.
With the advent of balanced cantilever construction techniques, solutions (c) & (d) are being
increasingly used and are suitable for longer spans. Continuous spans are used in situations
where foundations are the non-sinking type i.e. on rock.
Design of Sub-Structure : The design of substructure involve the design of the bearings, the
piers and the foundations. Of these, the design of piers and foundations lie well within the
experience and expertise of bridge engineers. It is re-emphasised here that a fully balanced
solution achieving overall economy cannot be arrived at without bringing in fully this
knowledge and expertise- the emphasis is due to the fact that it is not proposed to discuss
these aspects at any length in this paper. However, it is not out of place to make a few
observations regarding design of bearings since the bearing for aqueducts need some
additional considerations over bridge-bearing. The differences can be listed as follows :
(a) The normal temperature variation for aqueducts in service are much less severe due to
the presence of water which acts as a moderating medium.
(b) The expected service life of aqueducts can be much longer than bridges where the
changes in traffic needs may make them obsolete earlier. In this increased lifetime if
the cost of replacement of bearings is taken into the economic assessment, the net
cost of bearings will be very high. In fact, some of the studies we have earned out by
amortizing the replacement cost of bearings and comparing the same with solutions in
which bearings are altogether deleted - and designing the structure for such effects - it
is found that it is cheaper to delete bearings altogether.
(c) Situations of bearings getting submerged are more likely to happen for aqueducts
where the raising of structures above HFL is not possible if the canal bed level is low.
For such applications deletion of bearings or use of concrete bearings will be
preferable. Economically also, for very large loads the R.C.C. or Freyssinet type of
concrete bearings are highly competitive.
Some aspects of loading and material properties : It is to be noted that hydraulic structures in
general and aqueducts in particular are subjected to loading conditions which have distinctive
characteristics. The response of materials under these conditions of loading are slightly
different from other applications. Some of the relevant considerations are listed below :
(a) The structures are subjected almost always to the full design load which is a severe
condition of loading.
(b) The fatigue effects are much less severe since the range of live load variation as
compared to total static effects - i.e. dead load plus water load is a much smaller
fraction.
(c) The presence of water has a moderating influence on the range of temperature
variation and hence of temperature stresses. The characteristics of concrete which
depend upon the migration of water from the body of concrete, such as shrinkage and
creep get modified. Also the rate of corrosion of steel can be expected to be different
The presently used design codes in India do not recognize the differences in the strength of
materials for permanent loading and for frequent loading of a relatively short duration - such
as temperature effects, although they differentiate the extremely low frequency phenomena
like earthquakes, However, the international trend is to make such distinction in load
classification and use different levels of allowable stresses/strength as a general design
philosophy for the design of structures.
The use of special protective measures such as water proof paintings, epoxy coated
reinforcements, and many other new ideas are coming up and deserve detailed and careful
consideration to exploit the benefits offered by them.
Table I
Cross Section
(a) Hydraulic efficiency of cross section has permitted higher velocities and brought
down waterway section from 18 sq.mm. to 13.79 sq.mm. a reduction of 23% in
weight of water.
(b) The choice of membrane state of stresses resisted by prestressing cables combined
with the reduction of water load and reduction in self weight led to the reduction in
concrete cross section from 6.37 cu. m/m to 3.45 cu.m/m i.e. 46% saving of concrete
material.
(c) Use of continuous spans over three span and four span units reduced the longitudinal
bending effects making the system cheaper as compared to the simply supported
option.
The net savings over the conventional design were of the order of 40%.
The principles described hither to can be best understood if one makes detailed case studies
of the outstanding examples which are presented in various technical publications from time
to time. For the purpose of illustrating the various points made in this presentation a selected
number of aqueducts are described below :
A number of aqueducts on Narmada Main Canal carrying a discharge of 1130 cumecs are
under construction. The main canal has a trapezoidal section with bottom width of 72.1m and
top width of 108.5 m with IV : 2H side slopes. The depth of water at FSL is 7.6m. This can
with its banking is carried over a number of minor cross drainages without changing its
section. An interesting solution has been evolved for one of the major crossings at Ashwin
river. The canal section has been constructed reducing the bottom width to 50.0m and
maintaining the side slopes at IV : 2H. The sloped sides of the aqueduct structure consist of
RCC continuous slab spanning, over (lie wall type RCC piers. The horizontal portion of the
bed is earned over a series of continuous ROC slab and the arches are cast monolithically
with the pier wall. The span turned out to be an economic span since good rock is available
for foundation almost at the bed level and the pier walls are bout 7.5m in height only. The
foundations are open type strip footings under the walls. The arches are partly submerged
under design flood.
This 4240m long syphon tube working at pressure under 2.0 kg/sq.cm. is a precast structure
with 1.6m dia pipe section spanning 16m and simply supported over RCC piers. The pipe is
longitudinally prestressed. The photograph shows the aqueduct under construction which
also illustrates the method of construction.
This is a relatively small but interesting structure since it makes use of U-shaped prestressed
concrete shell structure carrying the water load mostly by membrane action. The width on the
top is 18m and the total depth is 6m. It is prestressed transversely and longitudinally. It has
40m central span and 14m overhangs on either side, the photograph shows the bridge under
construction, which is by the cast-in-situ method.
Located next (o the existing brick-arch type of aqueduct and carrying 310 cumecs discharge
of parallel Upper Ganga Canal at Roorkee, this transversely reinforced concrete multiple-cell
box section is designed to span over 36.6m. c/c piers as simply supported, longitudinally
prestressed structure of 321m length. It has five rectangular compartments and two end ones
carrying roadway. The three open boxes in the middle have cross ties to act as supports for
compression flanges. The piers are plate type piers supported on well type foundations.
This is an 8280m long aqueduct with RCC trapezoidal section of 5.6m on top, 2.5m at
bottom and 4.75m height. The segments are match-cast and connected by gluing with epoxy
and prestressing. The 40m continuous spans are constructed using 150m long launching truss
and balanced cantilever technique.
Cellular pier of 30m height were cast by slip forming. The photograph shows completed
portion of the aqueduct as well as the construction method.
"Bhima Aqueduct, Maharashtra"
The design economies achieved in this structure have already been discussed. Apart from the
optimised superstructure, the substructure solution is also structurally efficient. Hollow RCC
piers of upto 40m height support 40m spans in groups of 3 span and 4 span continuous units
and total 947m in length. The photograph shows the method of construction. The match-cast
segments are assembled by cantilever construction method, glued by epoxy and prestressed
longitudinally and transversely. The structure has been acclaimed internationally as an
outstanding construction.
wide at the base, has a water depth of
7.6m and sloping sides
References
Asawa, G L (1996) “Irrigation engineering”, Second edition, New Age Publications
• Garg, S K (1996) “Irrigation engineering and hydraulic structures”, Twelfth Edition,
Khanna Publishers
Version 2 CE IIT, Kharagpur
• IS: 11385-1985 “Code of practice for subsurface exploration for canals and cross
drainage works” (Reaffirmed 1990)
• IS: 7784 “Code of practice for design of cross drainage works”, Parts 1 to 6
• IS: 9913-1981 “Code of practice for construction of cross drainage works”
(Reaffirmed 1992)
• Varshney, R S, Gupta, S C and Gupta, R L (1993) “Theory and design of irrigation
structures”, Volume II, Sixth Edition, Nem Chand Publication
Version 2 CE IIT, Kharagpur
PRESSURE AQUEDUCT
ORIGINS OF THE PRESSURE AQUEDUCT SYSTEM
In 1936, the Legislature also approved the construction of the high service
Pressure Aqueduct System to deliver water to the Metropolitan area. Two
aqueducts would carry water from the Wachusett Reservoir to the new
Norumbega Reservoir in the Town of Weston. Downstream from Norumbega,
the aqueduct would split forming a loop around the metropolitan area which
would terminate at a greatly expanded Fells Reservoir in Stoneham. A
branch would be constructed off this loop to Quincy, terminating at a new
Blue Hills Reservoir. The Pressure Aqueduct could be operated at a uniform
rate of flow as the distribuation reservoir storage would be sufficient to
dampen out fluctuations in demand during the day. Work began on
schedule in 1939 and by the outbreak of World War II in 1941, the Fells,
Blue Hills and Norumbega Reservoirs had been completed, and the Pressure
Aqueduct had been built from the terminus of the Wachusett Aqueduct to
the terminus of the Weston Aqueduct. This portion of the Pressure
Aqueduct is the Hultman Aqueduct.
In hilly areas, sometimes very deep ravines have to be crossed. Here the catchments area of
the stream will be small and the bed level and M.F.L of drain will be very much below the
channel to be crossed. In such cases, naturally, the channel will pass over the ravine by
means of a trough or a conduit. But to support the trough pies will have to be unusually high.
This is not desirable.
Figure 16.6 shows a typical layout of a pressure Aqueduct. The channel is let into a
well whose diameter suits the average width of the channel.
The foundations are taken to hard ground. The positioning of the wells on eighter side
such that, they it will not get undermined by the scour at future date by the ravine. Piers are
built in the ravine course such that their tops are sufficiently above the M.F.L of the drain.
They are taken deep to rest on hard foundation. The span is so chosen that the ravine is
crossed with minimum number of spans.
On top of the piers a pressure conduit is run resting on steel trusses, or a concrete T-
beam. The conduit could be either a steel pipe, or R.C.C. pipe capable of taking the pressure
of the flowing water in it. The difference in levels between the bottom of the conduit and the
F.S.L of channel upstream of the aqueduct will be the pressure in the conduit.
This conduit is connected to another well on the downstream side as shown in figure
16.6. The channel after crossing takes off from the top of this well. The difference between
the channel F.S.Ls. upstream-side and downstream-side of the crossing will be the loss of
head in the conduit. This will depend upon the velocities to which we design the cross-
sectional area of the conduit. This loss of head can be calculated by using Bermouli’s
theorem or by ‘Flow Through Pipes’
Figure 16.8 shows, a typical super-passage. In this case, a distributory is taken over a
branch canal with an R.C.C trough supported on piers and abutments.
In case of a drainage crossing with supper-passage, the only difference will be that
the drainage take the place through the R.C.C trough instead of the distributory.
ORIGINS OF THE PRESSURE AQUEDUCT SYSTEM
In 1936, the Legislature also approved the construction of the high service
Pressure Aqueduct System to deliver water to the Metropolitan area. Two
aqueducts would carry water from the Wachusett Reservoir to the new
Norumbega Reservoir in the Town of Weston. Downstream from Norumbega,
the aqueduct would split forming a loop around the metropolitan area which
would terminate at a greatly expanded Fells Reservoir in Stoneham. A
branch would be constructed off this loop to Quincy, terminating at a new
Blue Hills Reservoir. The Pressure Aqueduct could be operated at a uniform
rate of flow as the distribuation reservoir storage would be sufficient to
dampen out fluctuations in demand during the day. Work began on
schedule in 1939 and by the outbreak of World War II in 1941, the Fells,
Blue Hills and Norumbega Reservoirs had been completed, and the Pressure
Aqueduct had been built from the terminus of the Wachusett Aqueduct to
the terminus of the Weston Aqueduct. This portion of the Pressure
Aqueduct is the Hultman Aqueduct.
In hilly areas, sometimes very deep ravines have to be crossed. Here the catchments area of
the stream will be small and the bed level and M.F.L of drain will be very much below the
channel to be crossed. In such cases, naturally, the channel will pass over the ravine by
means of a trough or a conduit. But to support the trough pies will have to be unusually high.
This is not desirable.
The foundations are taken to hard ground. The positioning of the wells on eighter side
such that, they it will not get undermined by the scour at future date by the ravine. Piers are
built in the ravine course such that their tops are sufficiently above the M.F.L of the drain.
They are taken deep to rest on hard foundation. The span is so chosen that the ravine is
crossed with minimum number of spans.
On top of the piers a pressure conduit is run resting on steel trusses, or a concrete T-
beam. The conduit could be either a steel pipe, or R.C.C. pipe capable of taking the pressure
of the flowing water in it. The difference in levels between the bottom of the conduit and the
F.S.L of channel upstream of the aqueduct will be the pressure in the conduit.
This conduit is connected to another well on the downstream side as shown in figure
16.6. The channel after crossing takes off from the top of this well. The difference between
the channel F.S.Ls. upstream-side and downstream-side of the crossing will be the loss of
head in the conduit. This will depend upon the velocities to which we design the cross-
sectional area of the conduit. This loss of head can be calculated by using Bermouli’s
theorem or by ‘Flow Through Pipes’
Figure 16.8 shows, a typical super-passage. In this case, a distributory is taken over a
branch canal with an R.C.C trough supported on piers and abutments.
In case of a drainage crossing with supper-passage, the only difference will be that
the drainage take the place through the R.C.C trough instead of the distributory.
Model of a siphon
HT = Header tank (left)
RT = Receiving tank (right)
VB = Venter bridge (bottom)
G = Geniculus (bend)
In 1936, the Legislature also approved the construction of the high service Pressure Aqueduct System to
deliver water to the Metropolitan area. Two aqueducts would carry water from the Wachusett Reservoir
to the new Norumbega Reservoir in the Town of Weston. Downstream from Norumbega, the aqueduct
would split forming a loop around the metropolitan area which would terminate at a greatly expanded
Fells Reservoir in Stoneham. A branch would be constructed off this loop to Quincy, terminating at a
new Blue Hills Reservoir. The Pressure Aqueduct could be operated at a uniform rate of flow as the
distribuation reservoir storage would be sufficient to dampen out fluctuations in demand during the day.
Work began on schedule in 1939 and by the outbreak of World War II in 1941, the Fells, Blue Hills and
Norumbega Reservoirs had been completed, and the Pressure Aqueduct had been built from the
terminus of the Wachusett Aqueduct to the terminus of the Weston Aqueduct. This portion of the
Pressure Aqueduct is the Hultman Aqueduct.
To pay for the new program, water rates were doubled from $50 to $100 per million gallons.
List of
over
70
siphon
s
Examples: Model of a siphon HWest part of the The receiving Male side of a
Almu HT = Header tank (left) e venter bridge in tank and ramp of stone cut block /
a Beaunant of the the siphon of the pipe of the siphon
necar RT = Receiving tank (right) d Gier aquaduct of Brévenne in the Roman
VB = Venter bridge (bottom)
(Spain G = Geniculus (bend) e Lyon (France) aqueduct of Lyon aqueduct in Cadiz
) r (France) (Spain)
Cadiz t
(Spain a
n
) k
Lyon a
(Franc n
e) d
r
a
m
p
o
f
t
h
e
Y
z
e
r
o
n
s
i
p
h
o
n
i
n
t
h
e
G
i
e
r
a
q
u
e
d
u
c
t
o
f
L
y
o
n
(
F
r
a
n
c
e
)
Pipes
The Greeks often used cut-stone or terracotta pipes and the same applies for the
Romans in Spain. In France in particular the Romans used lead pipes. These pipes had a
small diameter which were easier to produce than the bigger ones: commonly a series of
pipes were applied in siphons with lead pipes. The most striking example is the 8 - 10
pipes parallel in the nine (!) siphons in the four aqueducts of Lyon (France). Another
interesting example is the Greek siphon of the Madradag aqueduct of Pergamon
(Turkey) where only one lead pipe was applied encased in concrete (??). The Roman
Angitia-siphon (Italy) was equipped with one pipe of masonry (!) which is quite
exceptional.
Problems
The application of a siphon was not without problems. One had to be sure that the whole
system was 100% watertight, since leaks and / or air-bubbles in the pipe(s) could harm
the whole process and make it impossible to start the siphon.
A second problem was the high static (and dynamic) pressure on the pipe(s). A third
problem could be caused by the accumulation of debris and calcareous deposits (sinter)
inside the pipes at the bottom of the valley.
Conclusion?
The overview below lists some 60 Greek and Roman siphons. Elsewhere on this website
literature was collected on more than 600 Roman and Greek aqueducts. This leads to
the suggestion that 1 out of 10 aqueducts were equipped with a siphon.
Wilke Schram, March 2006
Notes:
)1 Stricktly speaking the term 'inverted' siphon should be used since a siphon brings
water over a hill or other obstacle, while an inverted siphon leads it through a
depression.
)2 Sometimes the word venter is also used to indicate the complete siphon, see
Kessener2001.
Top
Quotations
6. If the venter were not made use of in valleys, nor the level
substruction, but instead of that the aqueduct were brought to an elbow,
the water would burst and destroy the joints of the pipes. Over the venter
long stand pipes should be placed, by means of which, the violence of the
air may escape. Thus, those who have to conduct water through leaden
pipes, may by these rules, excellently regulate its descent, its circuit, the
venter, and the compression of the air.
For ease of fabrication, all siphon inverts can be located at the elevation of the lowest siphon
invert.
Register to fully enable the "Click to Calculate" button in the calculation below:
Your browser does not support Java, or Java is disabled in your browser. Calculation
should be here.
Units: cm=centimeter, cfs=cubic feet per second, ft=feet, gpm=US gallons per minute,
gph=US gallons per hour, gpd=US gallons per day, m=meters, MGD=Millions of US gallons
per day, s=second
Introduction
Stormwater and wastewater sewers often encounter obstructions such as rivers, other pipes,
subways, tunnels, or valleys. To pass these obstructions, a common method is for the sewer
pipe to drop sharply, then run horizontal under the obstruction, and finally rise to the desired
elevation. The piping going under the obstruction is traditionally called an "inverted siphon",
but since the pipe is not actually acting as a siphon, a better term is "depressed sewer"
(Metcalf and Eddy, 1981).
Unlike the main sewer pipe, the siphon pipe(s) flow under pressure. Special care must be
taken in inverted siphon design since losses are greater for pressurized flow, and the velocity
in each siphon pipe must be at least 3 ft/s (0.9 m/s) for sewage or 4 ft/s (1.2 m/s) for storm
water (Metcalf and Eddy, 1981). Therefore, even if there is only one main sewer pipe,
several siphons may be required.
Compute the maximum flow in the main sewer pipe using Manning's equation for full pipe
flow:
Compute the diameter of each siphon, Di, or the flow through each siphon, Qi, using
Manning's equation for full pipe flow through each siphon:
Compute the wall heights, yj (relative to main invert), in the inlet box. The walls separate the
siphons from each other. The wall heights are the same height as the water depths, yj, in the
main pipe corresponding to the discharge through the siphons. Here, Qj=1 is the discharge
through siphon 1, Qj=2 is the discharge through siphons 1 and 2, and so on. Manning's
equation for a partially full main pipe is used, but is solved backwards (numerically) in order
to compute yj. We allow up to five siphons (four walls).
Compute the siphon invert elevations in the inlet chamber. According to Metcalf and Eddy
(1981), there is no loss in the inlet box for flow going from the main culvert to the first
siphon since the flow travels in a straight path. However, for siphons 2 through n the flow
must turn 90o to go over the chamber wall (a head loss of 1.5 velocity heads) and has an
additional head loss of one velocity head as the flow enters siphon i. Therefore, for i=2 to n
siphons and j=2 to n-1 walls:
where Ei is relative to the invert of the main pipe. Note that for the first siphon, Hi=0, and for
the last siphon yj is replaced by Dm. Often, all siphon inverts are located at the same
elevation (the elevation of the lowest siphon) for ease of construction.
* Corrugated metal pipe n value can vary significantly with pipe diameter and type of
corrugations (values can range from 0.012 to 0.033) - AISI (1980).
"Need 1e-9<Main n<1e9", "Need 1e-9<Siphon n<1e9". The Mannings n values for the
main culvert and siphons must be between these limits.
"Need 1e-9<D1<1e9 m", "Need 1e-9<D2<1e9 m", "Need 1e-9<D3<1e9 m", "Need 1e-
9<D4<1e9 m", "Need 1e-9<D5<1e9 m". If siphon diameters are input, they must be between
these limits.
"Need 1e-9<Q1<1e9 m3/s", "Need 1e-9<Q2<1e9 m3/s", "Need 1e-9<Q3<1e9 m3/s", "Need 1e-
9<Q4<1e9 m3/s". If siphon flows are input, the flows must be between these limits.
Run-time errors. The following messages may be generated after performing some
calculations:
"Need 1e-9<Qm<1e9 m3/s". Discharge computed in main culvert must be in this range for
calculations to continue.
"Need siphon Q>0". If diameters are being computed, the flowrate through the last siphon is
automatically computed such that the sum of the flow through all siphons is equal to the
discharge through the main culvert. If the siphon flows input by the user exceed the
discharge in the main culvert, then the flow in the last siphon will be negative, which will
generate the error message. You should reduce the flows in the siphons so that there is
positive flow in the last siphon. Or, you could reduce the number of culverts.
"Siphons under-designed". Shown only if siphon flows are being computed and
Qs/Qm<0.95. You need to increase the siphon diameters. This message will not be
generated if diameters are being computed - because diameters are computed so that the total
flow through the siphons is exactly equal to the discharge through the main culvert.
"Siphons over-designed". Shown only if siphon flows are being computed and
Qs/Qm>1.05. Since wall heights cannot be computed for flows grossly exceeding that of the
main culvert, the calculation stops. You need to decrease the siphon diameters. This
message will not be generated if diameters are being computed - because diameters are
computed so that the total flow through the siphons is exactly equal to the discharge through
the main culvert.
Chow, V. T. 1959. Open-Channel Hydraulics. McGraw-Hill, Inc. (the classic text)
Hammer, M. J. and M. J. Hammer, Jr. 1996. Water and Wastewater Technology. Prentice
Hall, 3ed.
Viessman, W. and M. J. Hammer. 1998. Water Supply and Pollution Control. Addison-
Wesley, 6ed.
© 2002 LMNO Engineering, Research, and Software, Ltd. (All Rights Reserved)
Pont du Gard, France, a Roman aqueduct built circa 19 BC. It is one of France's top tourist
attractions and a World Heritage Site.
The Aqueduct in Kavala, Greece. 25 meters height with 60 arches. It was built in AD 1530
on a Roman model during the reign of Süleyman the Magnificent.
Water portal
Contents
[hide]
1 Ancient aqueducts
o 1.1 Aqueducts in Persia
o 1.2 Aqueducts in India
o 1.3 Roman aqueducts
2 Modern aqueducts
3 Uses of aqueducts
4 Notable aqueducts
o 4.1 Ancient Greek aqueducts
o 4.2 Roman aqueducts
o 4.3 Other aqueducts
5 Navigable aqueducts
o 5.1 Notable navigable aqueducts
6 See also
7 References
8 External links
[edit] Ancient aqueducts
View from inside a Roman aqueduct from the Pools of Solomon to Jerusalem
Although famously associated with the Romans, aqueducts were devised much earlier in the
Near East and Indian subcontinent, where peoples such as the Egyptians and Harappans built
sophisticated irrigation systems. Roman-style aqueducts were used as early as the 7th century
BC, when the Assyrians built a limestone aqueduct 30 feet (10 m) high and 900 feet (300 m)
long to carry water across a valley to their capital city, Nineveh. The full length of the
aqueduct ran for 50 miles (80 km).
In the new world, when the Aztec capital of Tenochtitlán was discovered in the middle of the
second millennium, it was watered by two aqueducts.
In Persia from early times[vague] a system of underground aqueducts called Qanat were
constructed, a series of well-like vertical shafts, connected by gently sloping tunnels. This
technique:
Taps into subterranean water in a manner that efficiently delivers large quantities of
water to the surface without need for pumping. The water drains relying on gravity,
with the destination lower than the source, which is typically an upland aquifer.
Allows water to be transported long distances in hot dry climates without losing a
large proportion of the source water to seepage and evaporation.
Roman aqueducts were built in all parts of the Roman Empire, from Germany to Africa, and
especially in the city of Rome itself, where they totaled over 260 miles (416 km). The
aqueducts were important for supplying water to large cities across the empire, and they set a
high standard of engineering that was not surpassed for more than a thousand years.
The Central Arizona Project Aqueduct, the largest and most expensive aqueduct ever
constructed in the United States.
Much of the expertise of the Roman engineers was lost in the Dark Ages, and in Europe the
construction of aqueducts largely ceased until the High Middle Ages. An example of an
extant small scale aqueduct system built in 1202 by Cistercian monks is the Spanish Real
Monasterio de Nuestra Señora de Rueda, whose central heating and waste disposal system
relied upon a series of aqueducts originating from a diversion of the Ebro River. Through
most of the Middle Ages and even up to the 19th century, water was instead usually supplied
through the digging of wells, though this could cause serious public health problems when
local water supplies became contaminated. One notable exception was the New River, a
man-made waterway in England, opened in 1613 to supply London with fresh drinking water
over a distance of 38 miles (62 km). The development of canals provided another spur to
aqueduct building.
The 19th century saw aqueduct building resume on a large scale to supply fast-growing cities
and water-hungry industries. The developments of new materials (such as cast iron) and new
technologies (such as steam power) enabled significant improvements to be made. For
instance, cast iron permitted the construction of larger, more highly pressurised inverted
siphons, while steam- and electrically powered pumps enabled a major increase in the
quantity and speed of water flow. England led the world in aqueduct construction, with
notable examples being built to convey water to Birmingham, Liverpool and Manchester.
Remains of the Schoharie Crossing, an aqueduct that once carried the Erie Canal over the
Schoharie Creek near Amsterdam, New York.
In modern times the largest aqueducts of all have been built in the United States to supply
that country's biggest cities. The Catskill Aqueduct carries water to New York over a
distance of 120 miles (190 km), but it is dwarfed by aqueducts in the far west of the country,
most notably the Colorado River Aqueduct, which supplies the Los Angeles area with water
from the Colorado River nearly 250 miles (400 km) to the east, and the 444 mile (714.5 km)
California Aqueduct which runs from the Sacramento Delta to Lake Perris.
Another widespread use for aqueducts is to supply large cities with clean drinking water.
Some of the famed Roman aqueducts still supply water to Rome today. In California, USA,
three large aqueducts supply water over hundreds of miles to the Los Angeles area. Two are
from the Owens River area and a third is from the Colorado River.
In more recent times, aqueducts were used for transportation purposes to allow canal barges
to cross ravines or valleys. During the Industrial Revolution of the 18th century, many
aqueducts were constructed as part of the general boom in canal-building.
In modern civil engineering projects, detailed study and analysis of open channel flow is
commonly required to support flood control, irrigation systems, and large water supply
systems when an aqueduct rather than a pipeline is the preferred solution. The aqueduct is a
simple way to get water to other ends of a field.
In the past, aqueducts often had channels made of earth or other porous materials. Significant
amounts of water are lost through such unlined aqueducts. As water gets increasingly scarce,
these canals are being lined with concrete, polymers or impermeable soil. In some cases, a
new aqueduct is built alongside the old one because it cannot be shut down during
construction.
Wignacourt Aqueduct, Malta. This aqueduct was built in the 16th century to transport
water from the old capital city of Malta, Mdina to the new capital city Valletta.
Today, only part of this aqueduct is visible in the localities of Balzan, Birkirkara and
Santa Venera.
Aqueduct St-Clément, Montpellier, France - 17th century
Águas Livres Aqueduct, in Lisbon, Portugal (built 1731-1748)
Carioca Aqueduct in Rio de Janeiro, Brazil (built 1744-1750)
Aqueduct of Teruel, Spain
Roquefavour aqueduct, France - built between 1842 and 1847
Winnipeg Aqueduct, Manitoba, Canada - built between 1915 and 1919
Canal de l'Aqueduc, Quebec, Canada
Päijänne Water Tunnel is 120 kilometers long underground aqueduct (continuous
tunnel) connecting lake Päijänne to Greater Helsinki.
Wan Mat Saman Aqueduct, Kedah, Malaysia - built between 1900 and 1909
Mathur Aqueduct in Tamilnadu state, India
Surviving Spanish aqueducts in Mexico:
o Aqueduct of Querétaro, Mexico - built between 1726 and 1738, 1.3 km long
and featuring 74 arches.
o Aqueduct of Morelia, Michoacan, built between 1735 and 1738.
o Aqueduct of Acámbaro, Guanajuato, built in 1528 [1].
Levadas, 1350 miles of 17th century aqueducts on the Portuguese island of Madeira.
Espada Aqueduct, built 1735, in San Antonio, Texas, United States.
Quabbin Aqueduct, 24.6 miles long tunnel, in Massachusetts, United States.
Chicopee Valley Aqueduct, 13.1 miles long, in Massachusetts, United States.
Central Arizona Project Aqueduct
California Aqueduct, a 444 miles (approx. 714.5 kilometers) long combination of
canals, pipelines and tunnels, United States.
Delaware Aqueduct, in New York State, United States - at 85 miles (137 km) long,
the world's longest continuous underground tunnel.
High Bridge, part of the former Croton Aqueduct, built in 1848, is the oldest
surviving bridge in New York City.
Navigable aqueducts are bridge structures which carry canals over other rivers, valleys or
railways or roads. They are primarily distinguished by their size, carrying a larger cross-
section of water than most water-supply aqueducts. Although Roman aqueducts were
sometimes used for transport, aqueducts were not generally used until the 17th century when
the problems of summit level canals had been solved and the modern canal system started to
appear.
Pontcysyllte Aqueduct on the Llangollen Canal, Wales. Built for the transport of goods, the
canal is now used for leisure
A canal boat traverses the Pontcysyllte Aqueduct
Early aqueducts such as the three on the Canal du Midi (1683) were stone or brick arches, the
longest span being 18.3m on the Cesse Aqueduct (1686). However the weight of the
construction to support the trough with the clay or other lining to make it waterproof made
these structures clumsy and it was not until 1796 that the first large cast iron aqueduct was
built at Longdon-on-Tern by Thomas Telford on the Shrewsbury Canal. It has a total length
of 57m with 3 intermediate piers. Within 10 years he had completed the far more ambitious
Pontcysyllte Aqueduct over the Dee valley on the Llangollen Canal which has a total length
of 307m. Other cast iron aqueducts followed such as the single span Stanley Ferry Aqueduct
on the Calder and Hebble Navigation in 1839 with its innovative 50m through arch design.
The impact of new materials can be seen in the experience of the Canal latéral à la Loire in
France. It had 2 substantial arch aqueducts on the higher stretches of the Loire, the longest
being 470m completed in 1838, but a river-level crossing was used to cross the Loire to the
Canal de Briare because the consequent obstruction to the river during flooding was
considered unacceptable. This proved troublesome until the 662m long steel Briare aqueduct
was built in 1896, which remained the longest aqueduct in the world until the 21st century
when the Magdeburg Water Bridge in Germany took the title.
The Magdeburg Water Bridge seen from the shores of the Elbe.
[edit] References
1. ^ http://books.google.ca/books?id=_kc3LKI_slEC
Sextus Julius Frontinus, De Aquaeductu Urbis Romae (On the water management of the city
of Rome), Translated by R. H. Rodgers, 2003, University of Vermont
Aqueduct entry from Encyclopædia Britannica Eleventh Edition
Chanson, H. (2002). Certains Aspects de la Conception hydrauliques des Aqueducs Romains.
('Some Aspect on the Hydraulic Design of Roman Aqueducts.') Journal La Houille Blanche,
No. 6/7, pp. 43-57 (ISSN 0018-6368)
Chanson, H. (2008). "The Hydraulics of Roman Aqueducts: What do we know? Why should
we learn ?" in Proceedings of World Environmental and Water Resources Congress 2008
Ahupua'a, ASCE-EWRI Education, Research and History Symposium, Hawaii, USA, Invited
Keynote lecture, 13-16 May, R.W. BADCOCK Jr and R. WALTON Eds., 16 pages (ISBN-
13: 978-0-7844-0976-3)
This example shows how to apply our iterative mesh generator on structural design
optimization problems. In [1] and [2], the level set method was used to track the
interface during structural shape optimization. Here, we use the same technique for
the interface, but solve the physical problem by updating an unstructured mesh and
using the finite element method. Some of the advantages with this approach
includes:
The Neumann boundary conditions are easily implemented by using natural
boundary conditions with the finite element method.
Graded meshes allow for more efficient solvers (but the two examples below
use uniform meshes).
The finite element method is more developed for structural problems, allowing
for example specialized elements and connection to existing software
packages.
Initial Final
MSMpeg4 AVI (2.0MB)
Again we minimize the compliance subject to a minimum-area constraint. Note that
the level set method can merge holes to create different topologies, but as before we
need an initial configuration with holes.
Initial Final
Bric s
Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.
Alternative Proxies: