CT3330

Hydraulic Structures

Locks

version March 2011 W.F. Molenaar e.a.

Cover photo and artist impression:

IJmuiden lock complex – the Netherlands
Courtesy Rijkswaterstaat
CT 3330 Hydraulic Structures Locks

 Hydraulic Engineering
Faculty of Civil Engineering ii 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

TABLE OF CONTENTS

PREFACE......................................................................................................................................................1

READER TO THESE LECTURE NOTES .....................................................................................................1

1. Introduction to sluices, locks, especially navigation locks ................................................................3

1.1 Dewatering gate (uitwateringssluis) .....................................................................................................3
1.2 Stop lock (spuisluis) .............................................................................................................................5
1.3 Guard Lock (keersluis) / Storm Surge Barrier.....................................................................................6
1.4 Navigation Locks..................................................................................................................................8

2. Navigation lock design .........................................................................................................................14

2.1 Design process in general..................................................................................................................14

2.1.1 Functions of a Navigation Lock ...................................................................................................15
2.1.2 Operational aspects ....................................................................................................................17
2.1.3 Life cycle considerations regarding the Navigation Lock ............................................................20

2.2 Navigation traffic and lock capacity....................................................................................................21

2.2.1 Duration of lock cycle and vessel passage .................................................................................21
2.2.2 Lock Capacity versus Traffic Intensity.........................................................................................24

2.3 Manoeuvrability while navigating through the lock and approach areas ............................................26

2.4 Positioning of the lock complex..........................................................................................................28

3. Overall lock design................................................................................................................................37

3.1 Lock geometry and lock configuration ...............................................................................................37

3.1.1 Typical inland waterway lock lay-out and geometry ....................................................................37
3.1.2 Special lock geometries and/or lock configurations ....................................................................38

3.2 Lock Approaches ...............................................................................................................................40

3.2.1 Lock approach layout ..................................................................................................................40
3.2.2 Alignment requirements with respect to the inland waterway - inland navigation .......................42
3.2.3 Location of the lock and construction ..........................................................................................43

3.3 Horizontal dimensions lock ................................................................................................................44

3.4 Vertical dimensions of the lock ..........................................................................................................47

3.4.1 Top of structure (ToS) and Bottom of structure (BoS) ................................................................47
3.4.2 Ship's draught and keel clearance ..............................................................................................49
3.4.3 Water Levels ...............................................................................................................................50

 Hydraulic Engineering
Faculty of Civil Engineering iii 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

4. Structural elements of a navigation lock ............................................................................................53

4.1 Filling and Emptying Systems (F/E) ...................................................................................................53

4.1.1 Typical F/E alternatives ...............................................................................................................53
4.1.2 Time required for F/E ..................................................................................................................60
4.1.3 Hawser forces .............................................................................................................................78
4.1.4 Specific subjects regarding F/E...................................................................................................84

4.2 Lock Gates .........................................................................................................................................88

4.2.1 Gate function ...............................................................................................................................88
4.2.2 Gate types ...................................................................................................................................88
4.2.3 Transfer of forces ........................................................................................................................97
4.2.4 Sealing.......................................................................................................................................101
4.2.5 Gate selection............................................................................................................................102

4.3 Lock Head ........................................................................................................................................104

4.3.1 Functions and requirements......................................................................................................104
4.3.2 Type of lock heads ....................................................................................................................106
4.3.3 Structural engineering aspects ..................................................................................................107
4.3.4 Construction aspects.................................................................................................................112

4.4 Lock Chamber..................................................................................................................................113

4.4.1 Lock walls ..................................................................................................................................113
4.4.2 Lock chamber floor....................................................................................................................115
4.4.3 Combination of lock chamber walls and floor............................................................................117
4.4.4 Chamber selection ....................................................................................................................119

4.5 Seepage cut-off screens ..................................................................................................................120

4.6 Lock approach structures.................................................................................................................121

5. References ...........................................................................................................................................125

APPENDICES

A1 Inland waterway vessels, CEMT classes & transport capacity ............................................................129

A2 Water Management..............................................................................................................................130
A3 Ice control in locks................................................................................................................................136
A4 Case study:Naviduct in Enkhuizen .......................................................................................................139

 Hydraulic Engineering
Faculty of Civil Engineering iv 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

PREFACE
Are these lecture notes finished for the next years? Probably not; big projects like the Panama Canal Lock
project and the Seine-Nord project are knocking on the door and shall find an entry into the material
written down here. But it is time to stick to the 80-20% efficiency rule.
And why did writing these lecture notes take so long if only an English translation of the already existing
Dutch lecture notes had to be produced in 2004? The answer is that opportunity knocked on the door:
there was a chance to include material from books published e.g. by Rijkswaterstaat, and the new
material that would be gathered by the PIANC Workgroup for the revision of the navigation lock report;
this update was recently published indeed, Of course, from the moment more considerable changes get
into view, minor items like upgrading the illustrations is taken at hand as well. It all boiled down to a very
substantial revision of the lecture notes on locks; it took the time that was needed, too long for the
impatient mind, but it was done with pleasure.
These lecture notes were written upon the backs, shoulders and brains of many predecessors, if anyone,
Kees Bezuyen has to be mentioned because he meant so much for many hydraulic engineering students,
so long. Students, graduate students and young engineers also contributed and I hope that especially
student readers will show the initiative to point the finger at mistakes, but what would be better, possible
improvements in the future.

Delft, March 2011

Wilfred Molenaar

READER TO THESE LECTURE NOTES

These lecture notes on locks are part of the study material belonging to the course 'Hydraulic Structures 1'
(code CT3330), part of the Bachelor of Science and the Master of Science, the Hydraulic Engineering
track, for civil engineering students at Delft University of Technology. Many of the principles and
engineering techniques treated in the BSc curriculum have to be applied when designing a lock.
Everything has to be used, from construction mechanics to fluid and soil mechanics, and anything in
between, or vice versa. A most important design aspect will be added: construction. Unfortunately,
construction differs considerably from site to site, hence this important matter surely cannot be discussed
in all its details and varieties, but an effort to show the most relevant items to be taken into account will be
made. It should be emphasised that the designer has the opportunity, maybe the obligation, to combine all
the imaginative powers, common sense and engineering judgement, to deal with the present day and
future problems related to the project at hand.

In chapter 1 we will provide a general overview of the various sluice types that exist. Functions and their
application in different types of sluices will be discussed. A (navigation) lock has to suffice to most, if not
all the functions that will be reviewed. Therefore it is meaningful to produce a (preliminary) design of a
navigation lock to find out into what trouble the designer gets combining all the functions. In order to be
able to make a design the required theory will be presented in Chapters 2 through 4. Chapter 2 deals with
general design matters and subjects that are especially important in the initiative and feasibility stages of a
project. In following chapters the focus will be more and more on the structures of lock (complex) and the
constituting elements. Chapter 3 the main items to determine the overall navigation lock layout are
described, whilst Chapter 4 will treat the most important structural components (filling and emptying
systems, lock chamber, lock gates, lock head, etc.), in more detail.

For Dutch BSc-students this is probably one of the first courses in the English language, so some very
specific technical terms have been translated into Dutch (indicated between brackets and in italics).

 Hydraulic Engineering
Faculty of Civil Engineering 1 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

 Hydraulic Engineering
Faculty of Civil Engineering 2 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

1. Introduction to sluices, locks, especially navigation locks

According to Webster's Universal College guide sluices are defined as:

1. An artificial channel for conducting water often fitted with a gate at the upper end for regulating the
flow.
2. The body of water held back or controlled by a sluice gate.
3. A channel carrying off surplus water.
4. An artificial stream or channel of water for moving solid matter
The definition of sluices used in hydraulic engineering is often the first. A lock is part of the sluice family, it
consists of an enclosed chamber in a waterway (vaarweg) used to transport vessels from one to another
water level.

The main functions of a sluice, as defined by the Dutch TAW (Technische Adviescommissie
Waterkeringen; this committee was abolished in 2005) are:
• Water retention (keren)
• Water locking (schutten)
• Water discharge (spuien)
• Shore connection (oeververbinding)

Before focusing on navigation locks some other sluices will be briefly described to illustrate the wide variety of
structures used for, on one hand, water retention, on the other hand, passage of the same water and e.g.
ships. The latter last but not least in many situations. Keep in mind that the word sluices is used to indicate
that the lock belongs to a larger family of structures constructed for similar reasons. The following sluices will
be dealt with in different sections:
• Dewatering gate - uitwateringssluis
• Stop lock - spuisluis
• Guard lock - keersluis
• Navigation lock schutsluis

1.1 Dewatering gate (uitwateringssluis)

Large parts of the Netherlands are situated below mean sea level. Therefore, the surplus of water in the
polders (caused for instance by seepage or rainfall) has to be discharged into the channel surrounding the
polder (boezem) or directly into sea by artificial means, e.g. using pump stations (gemaal). For these
water management (waterbeheer) purposes, hydraulic structures such as dewatering gates
(uitwateringssluizen) are built.

In polders the discharge of water (from low to a high) is separated in two steps. First, the water is pumped
in a storage basin or reservoir, or discharge channel (boezem) located at a higher level than the polder.
Next, as soon as the water level in the basin or channel exceeds the water level of the sea or the nearest
river, the water is discharged through a dewatering gate. Typically such a dewatering gate is a duct
(duiker, (omloop)riool)) with a valve that serves as (adjustable) closing element (afsluitmiddel). This
implies that shipping through a dewatering gate is impossible.
Additionally, the dewatering gate may serve as fresh- salt water separator or retain water (for instance
between salt sea water and fresh or potable water of a canal or polder system). In case high external
water levels occur it acts as a barrier.

From the above the functions of a dewatering gate can be deduced, as follows:
- Discharge of water from the area taken into consideration (polder) when the basin level exceeds the
target or allowable level
- Retain water at high external water levels
- Separation of salt water and fresh water

 Hydraulic Engineering
Faculty of Civil Engineering 3 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Single or multiple gates working by force of nature

Water is discharged through a

duct in free flow or submerged
condition. To prevent flow, for
water retention, traditionally
mitre gates are used. The HW HW
Water level polder Water level polder
principle is simple: difference in LW
LW
water level will keep the gate in
a closed position at a higher
exterior water level and open A B
them automatically at high
interior water levels. This
principle is depicted In Figure
1-1. This combination of duct
and mitre gates (duiker) is HW
Water level polder
often used in areas with a low LW Water level polder
risk profile for flooding. In case LW

of a higher risk profile other

solutions (e.g. different type of C D
gates) have to be applied.
Figure 1-1 Mitre gates (puntdeuren) that open or close
depending on the water levels
Automatic dewatering gates

There is an increasing demand for a more accurate control of the storage basin levels. This requirement
results in the application of gates that can be operated by will rather than as a result of coincidental
differences in water level. Moreover, the gates must be able to withstand water level differences in both
directions, generally lift gates (hefdeur) or rolling gates (roldeur) are more suitable than mitre gates. The
whole is referred to as the automatic dewatering gate. An advantage of this solution is that it can be used
for irrigation when the water levels are too low to meet for agricultural demands. Purpose built for that
situation the dewatering is more accurately referred to as an inlet sluice (inlaat sluis). In arid countries
these types of sluices are a common element of irrigation systems.

Figure 1-2 Cross section of the Haringvlietsluizen, the Netherlands

Obviously this doesn't apply to dewatering gates in coastal areas because of the salt content of the
seawater (e.g. the Haringvlietsluizen where segment gates were used).

 Hydraulic Engineering
Faculty of Civil Engineering 4 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

In the design of the Haringvlietdam the following main functions are combined:
• Protection, retain water at high external water levels (protection North Sea)
• Diversion of water, forcing more water into the Nieuwe Waterweg (Rotterdam) to prevent the salt
water tongue to intrude too far landinward
• Discharge of water (river Rhine and Meuse) when the basin level exceeds the external water level

More info on the Haringvlietsluizen; see the website: http://www.haringvlietsluizen.nl (October 2005)

1.2 Stop lock (spuisluis)

The purpose of a dewatering gate is to control the water level and to refresh water in a closed water basin
or channel system. Besides water level control the structure can be used to clear the basin or canals from
sediments, debris and other pollutants while water is being discharged. In this case the sluice is referred
to as stop lock. Although the stop lock resembles the dewatering gate, the main difference is based on the
principle that a stop lock is opened when the reservoir level is (considerably) higher than the water level of
surrounding water areas. The water level difference should result in sufficient water discharge to flush out
the sediments, debris and pollutants.

The higher the water level difference, the higher the flush velocity u, which results in increased sediment
transport s, which is illustrated by the following formula:
n
s = mu
Where:
s = sediment transport
m = ‘garbage’ coefficient or expression, not dimensionless, typically ≈ 0.01
u = velocity of water
n = constant, typically 3 ≤ n ≤ 5

Since the gates have to be opened at a water level difference, mitre gates cannot be applied for a stop
lock. An example of a stop lock in the Netherlands is illustrated in figure. In this particular case the stop
lock is positioned adjacent to a navigation lock.

Figure 1-3 Stop lock, Zandkreekdam Katse Figure 1-4 Stop lock, Afsluitdijk – Den Oever,
Heule, the Netherlands the Netherlands

More detailed information on the design of the Zandkreekdam - Katse Heule stop lock, on the east side of
the ‘Veerse meer’, can be found on the following website:

http://www.dubo-centrum.nl/projecten/project.php?recID=103 (Oktober 2005)

 Hydraulic Engineering
Faculty of Civil Engineering 5 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

1.3 Guard Lock (keersluis) / Storm Surge Barrier

Guard locks and Storm Surge Barriers (keersluis, stormvloedkering) generally have two conflicting primary
functions:
• Passage of vessels
• Retention of water

The gate of the guard lock will be kept open as long as possible to allow ship passage, obviously, when
opened the guard lock can not retain water. A water level being too high or too low, not necessarily
extreme water levels, on either side of the lock may require closure of the gate, preventing further ship
navigation. Conditions for closure may be either extreme low water levels due to draught or extreme high
water levels due to a storm surge. Two types of guard locks exist: one-way retaining or both-ways
retaining.

Figure 1-5 Guard lock Kromme Nol near Heuschden, the Netherlands

One-way retaining guard lock

In normal conditions, from water level point of view, the lock is open, only in certain extreme conditions
(during a storm) it retains water in one direction. This type of guard lock is commonly found in areas with a
small tidal range. The doors are closed when the outside water level reaches a certain predefined level.
When closed, the doors are part of a flood defense system, which also includes adjacent dikes or other
flood defense structures. The design height of the doors may be less than the design height of the
surrounding dikes pending on berm design of the dike and the allowable amount of water overtopping the
structure (wateroverslag).

Both ways retaining guard lock

In areas with large tidal ranges, as for instance along the English Channel, harbour entrances are usually
equipped with guard locks retaining water both ways. For instance, at the harbour side of the guard lock,
the water level is kept between certain boundaries, effectively a harbour basin is created. A minimum
water level is defined to enable a vessel to manoeuvre safely without the fear of running aground.
Obviously, also a maximum water level is defined to prevent flooding of e.g. the port facilities and the
town.

 Hydraulic Engineering
Faculty of Civil Engineering 6 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

The harbour basin level differs from the tidal range. During low tide the water must be kept within the
harbour basin, and during high tide the water must be kept out. This implies that the guard locks are only
opened during the period around the turn of the tide (kentering), between high tide and low tide. Ships can
only pass during this period with, fortunately, limited flow velocities. There may be a more restrictive limit
to the draught of vessels because the high water tidal wave, thus deeper water, cannot be used

As an alternative to double retaining mitre, rolling, or sliding gates, or a vertical lift gate can be used.
These types are able to retain water in both directions. A disadvantage of the vertical lift gate to be taken
into due consideration is the limitation to the air draught of ships.

Storm Surge Barriers - Extreme guard locks

Storm surge barriers are closed to provide protection against extreme water levels, e.g. in storm
conditions in coastal areas. The main purpose of the structure is to protect the land from the sea in
(extreme) storm conditions and to ensure the passage of vessels for the remaining time.

A well known example is the Maeslantkering, the storm surge barrier in the New Waterway the gateway to
the port of Rotterdam.

Figure 1-6 Birds-eye view of a storm surge barrier; Maeslantkering, Rotterdam

For more information on the Maeslantkering, see:

http://www.keringhuis.nl/engels (January 2006)

 Hydraulic Engineering
Faculty of Civil Engineering 7 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

1.4 Navigation Locks

A navigation lock, shortly referred to as lock, is the link between two sections of a waterway (river, sea or
channel) with different water levels. The lock enables the transfer of the ship from one to the other section
of the waterway. Obviously the water level difference between both sides of the lock, the lift, is one of the
main parameters to be considered in the design of the lock.

Navigation locks and the required lift

Comparing rivers or waterways having a steep water profile with rivers or canals having a rather flat water
profile results in the obvious conclusion that either a bigger number of locks, or a lock designed for a
higher lift, will be required for the steep profile. The efficiency of navigation would benefit in general from
locks being spaced as far apart as possible, which reduces shipping delays, but may result in high lifts that
can not be accommodated in most of the (traditional) locks. Positioning a number, typically 4 or 5, locks
behind each other, i.e. construction of a flight of locks or lock ladder, could solve this problem. However, in
these cases the use of alternatives, e.g. a ship lift or pente d'eau, should be investigated. Typically the
water level difference of the latter type of structures is > 25 meter.

Since the lock generally is part of the flood defense system along the river, the decision for the lift of the
lock also has effects on the retaining height of those systems and vice versa. The costs of one single big
lift lock and having to increase the top level of the (whole) flood defense system, to match the retaining
heights, have to be weighed against the costs of a bigger number of locks and smaller or no changes to
the flood defenses.

Locks for inland navigation and for coastal areas

Figure 1-7 shows a typical inland navigation lock in the Netherlands. Due to the Dutch topographical
situation the required lift generally varies from say 3 to 5 meter. Following chapters will deal extensively
with this type of lock.

Figure 1-7 Typical inland navigation lock Figure 1-8 Princess of Scandinavia entering the
Middensluis, IJmuiden, the Netherlands

 Hydraulic Engineering
Faculty of Civil Engineering 8 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Locks in tidal areas differ in many respects from those in inland waterways. The generally much larger
dimensions, due to the larger size of maritime vessels, see Figure 1-8, is the first to be taken into account.
Besides this they operate under water level differences in either direction; may have to be designed to
reduce salt water intrusion and should be able to cope with more adverse weather circumstances,
including wave attack.

The following navigation lock types, which are able to lift the ship over a bigger water level difference from
traditional lock to e.g. the rotating wheel in Falkirk, will be described in the hereinafter:
• (Traditional) lock; for inland navigation and for coastal areas
• Lift lock
• Inclined plane
• Pente d'eau
• The rotating wheel at Falkirk - Scotland

Lift lock

The principle of a lift lock is vertical transport of a vessel with the surrounding water in a movable closed
lock chamber. Some times the lift lock has two movable chambers counterbalancing each others weight,
see the description of the lift lock in Henrichenburg – Germany below, but this system can be considered
as slightly outdated. More often the weight of the moveable lock chamber is balanced by a counterweight
and the chambers are designed and constructed as independent systems. Provided the weights of the
chamber and the counter weight are more or less in equilibrium, a relatively small 'force' is required to put
the system into motion. For the latter mechanical devices, e.g. winch-cable or worm screw-engine
systems, or electro-magnetic engines can be used. Note that the weight of the moveable chamber nearly
always remains the same, since the total volume of water displaced equals the mass of the vessel
entering the chamber.

Figure 1-9 Lift lock Strepy Thieu, Belgium

Old type lift lock – “floaters”

To reduce the total force in the chamber lifting system usually floaters are constructed under the
chambers. These floaters move up and down in separate reservoirs by adjusting the water level. Imagine
a piston moving in a cylinder. When floater and chamber 1 move down, water flows out of reservoir 1 into
reservoir 2, where the increased water level lifts floater and chamber 2. Instead of under barometric
pressure, the water can be kept under high pressure in the cylinder. By linking the cylinders through a pipe
with a valve that can be opened and closed as required an alternative, very accurate motion system is
created.
The chambers have to be closed at each end by a gate or bulkhead to prevent the loss of water during the
lift, the vertical transport. Obviously the adjacent ends of the upper and lower canal reaches have to be
equipped with gates as well to avoid draining the canal when the chambers are on the move. When the
lower chamber is correctly aligned for opening the gates from the lower canal, the upper chamber has to
be in the right position (level/height) for opening the gates from the upper canal as well. Before opening

 Hydraulic Engineering
Faculty of Civil Engineering 9 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

the gates the bottom and walls of the chamber and canal have to be joined with a watertight connection
and water level differences between the chamber and the canal reach have to be equalized as well.

Figure 1-10 Schiffshebewerk Henrichenburg, Germany

Figure 1-11 Lift lock Luneburg – Germany. Areal view and cross sections

 Hydraulic Engineering
Faculty of Civil Engineering 10 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Inclined plane

The principle of an inclined plane is the use of a natural slope in the terrain to overcome a height
difference. A movable lock chamber is moved over a slope. The adjoining waterways must be sealed of
with a water retention structure and gates (e.g. Ronquires 1968). Inclined planes can de divided in two
types:
• Transport in longitudinal direction - in the axis of the vessel (Ronquieres – Belgium; not in these
lecture notes
• Transport in transversal direction - perpendicular to the axis of the vessel (Arzviller – France, see
below)

Depending on local circumstances a choice is made for the one or the other. Some advantages and
disadvantages of these systems are listed below.
• Movement of water in the chamber during acceleration or slowing down, which is especially
important for longitudinal systems
• Amount of required space
• Forces in the cable system
• Positioning in the landscape.

Figure 1-12 Inclined plane in sideways direction, Arzviller, France

Pente d'eau

The principle is the same as the inclined plane; the use of a natural slope in the terrain to overcome a
height difference. In this case water is being moved using a more or less watertight door in a kind of canal
pushing the water up or down a slope. The advantage would be a simpler construction than an inclined
plane. The disadvantage is that more energy needed to transfer water plane and ship. Up to now only a
few have been built (Montech 1974, Béziers 1983).

Figure 1-13 Pente déau, Béziers, France

 Hydraulic Engineering
Faculty of Civil Engineering 11 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

On the following website a nice animation showing the working principle of the pente d’eau in Montech –
France can be found:
http://www.canaux-historiques.com/d2m/ouvrage/montech/fiche_ouvrage/fo_00.html#

For improving the inland shipping connection between Paris and Belgium & the Netherlands, known as Seine-
Nord, the feasibility of a large pente d’eau has been taken into consideration (2006).

Rotating wheel

The original concept of a wheel to act as a boat lift actually dates back to 19th century Europe; it was
seriously considered by British Waterways as the solution for Falkirk in 1994.

Artist impression

Landrot:
boatlift

Schipper / Zeeman:
shiplift
Figure 1-14 The wheel during construction, Falkirk, Scotland

The objective was to create a functional boat lift that could raise and lower boats swiftly, over a height of
24 m, with a structure worthy of a new millennium. The Falkirk Wheel lies at the end of a reinforced
concrete aqueduct that connects, via the Roughcastle tunnel and a double staircase lock, to the Union
Canal.

Ships entering the Wheel’s upper gondola are lowered by a rotational movement, along with the water that
they float in, to the level of the lower canal reach. At the same time, an equal weight is lifted up in the other
gondola, which acts as counterweight. Archimedes Principle of displacement is applicable on the total
weight of the gondolas. The mass of the boat sailing into the gondola displaces a volume of water having
exactly the same mass. The total weight of the gondola remains the same.

The gondolas, or moveable lock chambers, have to rotate in the opposite direction of the wheel, otherwise
the vessel and surrounding water would be thrown out like the water of a bucket being emptied. Therefore
small wheels. running on a single curved rail fixed to the inner edge of the opening, have been assembled

 Hydraulic Engineering
Faculty of Civil Engineering 12 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

to the gondola, at both ends. In theory, the downward weight of the gondola should be sufficient to always
remain horizontal, but any friction or sudden movement could cause the gondola to get stuck or tilt. To
ensure that this could never happen and that the gondola, the water and the boats, always remain
perfectly level throughout the whole rotation cycle, an inventive series of linked cogwheels (tandwielen) is
used as safety system.

Given the precise balancing of the gondolas and the relatively simple, nonetheless clever system of
cogwheels, operating without too much friction, only a small amount of energy is required to turn the
Wheel. In fact, it is a group of ten hydraulic motors located within the central spine that provide the small
amount, just 1.5 kW, of electricity to turn it. The figure below shows their position.

Figure 1-15 Principle of the rotating wheel at Falkirk

More information on the Falkirk Wheel can be found e.g. on:

http://www.falkirk-wheel.com
http://en.wikipedia.org/wiki/Falkirk_Wheel

 Hydraulic Engineering
Faculty of Civil Engineering 13 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

2. Navigation lock design

2.1 Design process in general

Design starts with thorough analysis of the problem or project at hand. Is it really necessary to maintain a
water level difference? Is the shipping intensity indeed high enough to warrant the investments in a
navigation lock? Asking the right questions about the problem, the objectives, functions, operational
aspects, constraints and assumptions should finally result in a good set of design criteria or in the Basis of
Design (Programma van Eisen).
To produce alternatives,
not only a systematic
work method, but
sufficient creativity,
which may be best
served by the opposite
of systematics, is
required as well. Politics
comes into play during
the selection of
alternatives to be
elaborated. Frequently
multi-criteria analyses
are used to weigh for
instance safety, noise or
panoramic pollution,
think about the required
towers for a lifting gate,
against costs. Figure 2-1 Common flowchart design process
Fortunately further
elaboration of the selected alternative(s) is a more straightforward engineering matter.

During the whole design process the work proceeds from more general considerations and rule of thumb
use into more detailed calculations. Results of more detailed calculations are feedback for the previous
level. Use the feedback to decide whether or not design is still heading in the right direction.

At a certain moment the life stages of the project or structure have to be taken into account. It is easy to
imagine that e.g. construction methods or maintenance requirements (operation phase) influence the
design. The following phases should be distinguished in the life cycle of a navigation lock:

Figure 2-2 Life Cycle stages

Design plays its role from the idea/initiative into the reuse or disposal phase, but most dominant in the
design phase itself. In this and the following chapter general and more detailed lock design issues will be
dealt with.

 Hydraulic Engineering
Faculty of Civil Engineering 14 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

2.1.1 Functions of a Navigation Lock

The functions of a navigation lock are analysed in the first place to determine the necessary lock
components. The analysis also contributes to the preparation of the design criteria for the navigation lock
or finding the (quantitative) requirements the lock should suffice to.

The three main functions of a navigation lock:

1. Water retention - maintain a water level difference e.g. for water management or flood defense
purposes
2. Ship passage – the water level difference and the means to maintain the water level difference
require solutions for vertical and horizontal transport of the vessels
3. Water quality management – depending on the surroundings or environment of the lock care may
have to be taken about e.g. sediment and debris discharge or separation of salt and fresh water

Note:
th
As mentioned in the Introduction the TAW defined ‘Shore connection’ as the 4 ‘main’ function of
locks. This is a matter of definition indeed; what is a main function and what is a secondary
function? Often it depends on a particular situation. One could argue that water quality
management is not a main function at all. Another example would be to consider ‘Provision of port
facilities’ as main function of a pleasure navigation lock, whereas this would not be applicable for
commercial navigation.

It is easily observed that (structural) solutions for water retention may get in the way of ship passage and
water management, and vice versa, which is illustrated by the (incomplete) table below.

Function
Retain water Transport vessels
Horizontal Vertical
Dike / levee In water & using ship’s engine Adjusting water level
*
Sheetpile wall Sliding Lifting / lowering *
Door / gate Rolling *
‘Return pumping’
*
with or without the surrounding water or lock chamber
Table 2-1 Functions and solutions
Depending on the point of view the lock will be considered as a flood defense structure or as transport
infrastructure. In the following the emphasis will be on inland navigation, hence the transport function.

Ship passage function

Navigation locks play an important role in the inland waterway transport network for the transport of goods
in Europe and in the Netherlands in particular. More than 35 000 kilometers of waterways connect
hundreds of cities and industrial regions. In European inland waterway transport is a realistic alternative to
road and rail transport. It is energy-efficient; the energy consumption per ton-kilometer is only 17% and
50% of the consumption of road and rail transport respectively. Noise and gaseous emissions are modest.
Additionally, inland waterway transport ensures a high degree of safety, in particular when it comes to the
transport of dangerous cargoes.

For the most common Dutch inland navigation lock the solution developed for retaining water and
transport of the vessel is to use a lock chamber that can be closed by gates to maintain the water level
difference. Within the lock chamber the water level can be adjusted to allow for the vertical transport of the
ships, the horizontal transport is taken care of by the ship’s own propulsion. Generally cut-off screens
below and adjacent to the structure are used to prevent piping. The function analysis for the typical Dutch

 Hydraulic Engineering
Faculty of Civil Engineering 15 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

lock would result in the following lock components being necessary: Gates and housing, lock chamber, a
water leveling solution and cut-off screen.

It is left to the reader to do a more elaborate function analysis.

Note:
A clear cut between the analysis of functions and the analysis of operational aspects is not always
possible, and again, depends on definition. In these lecture notes the lock chamber will be the
result of the functional analysis, whereas the dimensions of the lock chamber will be the result of
operational considerations, see section 2.1.2. The result, a comprehensive list or document of
design criteria, matters much more than determining by which principal method or principal
analysis it has been or should be obtained.

Water retaining function – maintaining a water level difference

A lock is situated in a waterway between two sections with a different water level. Combined with dikes
and other structures it is part of a water defense system. The upper lock head (at the side of the highest
water level) must be able to retain the highest water level. The lower head only retains the maximum
chamber level, which is not necessarily the highest water level. In tidal areas lock heads are referred to as
outer head (on the sea side), respectively inner head (on the land side).

As part of the water defense structure the navigation lock must be able to retain water under all
circumstances. If for any reason (e.g. a ship runs into a gate) the lock is damaged a back-up system, a
second set of gates or stop logs, must be available to take over the water retaining function.

Typically, water levels will be different at both sides of a lock. This will result in a groundwater flow under
and around the sides of the lock structure. At the downstream side where the groundwater flow exits the
soil, erosion will occur in case the flow velocity is high enough to induce soil movement. This is a self
sustaining and enlarging process; since the first erosion will result in decreased flow resistance and
subsequent increased groundwater flow, exit velocity and erosion. The phenomenon is known as seepage
and /or piping, and may undermine the whole lock structure To avoid piping locks are provided with
seepage or cut-off walls, often sheetpiles are used as cut-off screen, acting as an impermeable curtain
perpendicular to the axis of the lock chamber. Alternatively filter layers and scour protection are used
around the structure, provided the resulting seepage or piping remains within acceptable limits. These
limits may be enforced from retention, water management or structural point of view.

Water management function

A lock can also be used for water management, for quantitative as well as qualitative aspects of a water
management, or combinations. Qualitative aspects generally boil down to keeping water masses with
different properties separated, for instance:
• Clean and polluted water.
• Salt and sweet or fresh water.
Quantitative aspects in the water management system are for instance:
• Discharge of a predefined amount of water within a certain period of time.
• Minimizing the fresh water loss from the upper canal reach.

Discharge of water through a navigation lock can be achieved in various ways, e.g. valves in the gates or
a drain system. In tidal areas, the drainage system can be used to dispose salt water that entered the
fresh water. At times of large river discharges (e.g. during a period of heavy rainfall) the full cross section
of the lock may be needed for discharge. Flow velocities will be too high for safe navigation and locking is
impossible in this situation anyway. The lock at Dintelsas, at the mouth of the Mark or Dintel to the
Volkerak – the Netherlands, is an example of this type of lock.

 Hydraulic Engineering
Faculty of Civil Engineering 16 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

In case of smaller discharges the lock's drainage system can be used to get rid of a surplus of water.
However, sluicing and locking cannot take place simultaneously. Preferably the discharge of water is in
"quiet" shipping hours, e.g. at night time or during the weekends. This to minimise hindrance for vessel
traffic. In case of multiple lock chambers both functions are more easily combined. The lock near Wijk bij
Duurstede – the Netherlands, is used to maintain the right waterlevel on the Amsterdam –Rijn canal. The
whole lifting gate is used to manage the water supply; not just valves in the gate.

The amount of water lost per leveling cycle is: A ⋅ z + Wup − Wdown
Where:
A = horizontal area lock chamber
z = difference in water level, water head
Wup = amount of water which is displaced by ships going upstream
Wdown = amount of water which is displaced by ships going downstream

Figure 2-3 Locking cycle and water losses

Generally there is an imbalance in traffic, either a different number of ships sail up or down, using
alternative routes to return to the point of origin or destination, or vessels are loaded in upstream direction
whilst they sail in ballast in the downstream direction, or vice versa. To minimise the amount of water lost,
the use of reservoirs or water saving basins could be considered, or much more rigorously, the amount of
lost water could be pumped back to the upstream river reach.

2.1.2 Operational aspects

After the analysis of lock functions operational aspects are studied. Predominantly this will result in
dimensions of the lock components (preparation of design criteria), however, it will extend the list of lock
components as well. To mention only a few operational aspects:
• Navigation through the lock
• Operation of the discharge system
• Lock maintenance (regular and incidental)

 Hydraulic Engineering
Faculty of Civil Engineering 17 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Operation: Navigation through the lock

Analysing navigation through the lock results amongst other, in the following three design requirements:
1. Lock dimensions must be large enough to accommodate the governing (maatgevend) vessel for
the considered waterway (design ship).
2. Lock capacity, defined as the number of ships passing the navigation lock within a certain time
interval, must match present and/or future demand.
3. Even if the capacity is not fully utilised the time losses during passage should be kept at a
minimum.

The intensity of a waterway is defined as the number of ships that pass a given cross section of the
waterway per time interval. The capacity of a lock is the number of ships a lock is able to process. If the
capacity is too small this will result in long waiting hours, if the capacity is too large this will result in an
expensive navigation lock. For instance:

According to the client 12 800 vessels are expected to pass the waterway in one year. The
navigation lock is available 7400 hrs per year. This results in an intensity of 1.7 vessel per hour. If
the lock capacity is smaller than required the lock will not be able to handle the amount of vessels.
To determine the lock capacity a quick-and-dirty method to estimate a figure would be to use a
multiplication factor of 2 - 2.5. In this case this would result in a required capacity of 3 - 4 vessels
per hour.

Capacity calculations and related subjects are treated more elaborately in the course on Ports and
Waterways (CT 4330).

In the above the focus was on dimensions or the quantitative aspects relevant for lock design, now a brief
illustration of more qualitative results of the operational analysis. In the following table the navigation
process has been described, not even in the greatest possible detail, in the left column, whilst the right
hand column shows required lock components or lock furniture:

Operational activity Required facilities

1. Ship reports in ‘VTS’ room / Lock-Master room
2. Ship decelerates; possibly extra propeller
action required for manoeuvring Bottom and bank protection
3. Ship waits / not Waiting berths
4. Ship sails into lock chamber Guard / guide walls
5. Ship berthing Fendering
6. Adjusting the water lock level Bollards, mooring rings,
– ship moored ‘flexible’ ladders, etc
7. Deberthing
8. Sailing out & reporting out Bottom and bank protection
9. Acceleration ‘at safe distance’
Table 2-2 Operational activities and required facilities

 Hydraulic Engineering
Faculty of Civil Engineering 18 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Operation: the discharge system

Every lock provides some discharge capacity for its surroundings. It depends on the specified function and
the location of the lock whether or not this has a large impact on design of the lock; some examples:

• Lock complex IJmuiden: the Netherlands: the required discharge capacity is large since a number
of polders surrounding the Northsea canal are dewatered by this system. The capacity requirement
resulted in construction of a separate pumping station (gemaal) instead of combining the navigation
and discharge function in one lock.
• Locks in the city of Amsterdam: the canal system in Amsterdam has to be flushed through at
regular intervals (daily) to prevent al sorts of health risks in the city. A number of the stop locks are
also available for navigation. Navigation is not possible when the gates are open for discharge,
generally during night hours.
• The lock in Dintelsas: after large rainfalls the north western part of Brabant is dewatered through
this lock. When the lock is used for dewatering, navigation is interrupted.
• The lock at Wijk bij Duurstede: to let water in into the Amsterdam-Rijn canal the lifting gates of the
lock are used. The whole gate is lifted for 1 – 2 meters for this purpose, instead of using valves in
the gates, or a culvert system around the lock heads.

See the (sub)section on ‘Filling and emptying systems’ in Chapter 4 to determine whether or not the
functions navigation and discharge (in fact the water management function) can be combined in one lock
from an operational point of view.

Operation: lock maintenance (regular and incidental)

To avoid too much interference of regular maintenance (annual) with navigation through the lock it may be
decided to take some extra measures, e.g. providing extra concrete cover or an extra set of gates.
Obviously the costs involved throughout the years would have to be carefully considered.

When irregular maintenance (groot onderhoud) (say once in a decade) is required generally navigation is
influenced, occasionally ships may have to be rerouted when the lock is completely taken out of operation
during the maintenance period. Suppose the irregular maintenance requires the lock chamber to be laid
open (droog gezet), which would imply that a lock with a closed concrete section has to be checked for
floating up. During design it could be decided to prepare the structure for this condition by adding, for
instance, tension piles, extended bottom slabs, extra concrete weight or combinations. Often structural
measures of this kind result in a substantial increase of the initial investment. Alternatively it could be
decided to solve the possible floating up problem if/when it actually occurs in the future if it is ever (really)
necessary to lay the lock open.
Nowadays it is more common practice ‘not’ to design and construct for being laid open. As a consequence
maintenance to the lock chamber walls and floor has to be done underwater.

In the design stage a decision has to be made on the maintenance strategy taking into consideration the
extra initial investments, the future costs and last but not least, the 'negative' revenues. Commercial
navigation pays for the use of the lock. When the lock is not available due to maintenance it does not
generate revenues and on top of this there are extra transport costs due to rerouting (extra costs for the
society; not necessarily for private parties).

 Hydraulic Engineering
Faculty of Civil Engineering 19 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Suppose the evaluation method uses the Net Present Value (NPV) technique for costs, than the costs for
regular, irregular or combinations of maintenance can be ranked as follows (indicative costs only):

Maintenance type / Maintenance and incidental costs

Maintenance strategy as percentage of initial
investment in lock
NPV value ***
Annual Incidental Total initial
discount rate
Initial investment in lock: * investment
8%
€ 100 000 000,- During During**
constructio lifetime
n
Regular + No extra
1: measures for 1% 0 0 € 100 000 000,- € 112 077 074,-
prevention
Regular + Extra
2: measures for 0.5% 2.5% 0 € 102 500 000,- € 108 538 537,-
prevention
Incidental + Prepared
3: structure 0 10% 5% € 110 000 000,- € 114 115 750,-

Incidental + No extra
4: measures during 0 0 15% € 100 000 000,- € 116 462 999,-
construction

5: 2+3 0.5% 12.5% 5% € 112 500 000,- € 122 669 951,-

6: 1+4 1% 0 15% € 100 000 000,- € 128 571 401,-

* this is a rough guestimate for a Class V canal lock with 2 chambers, one for each direction of travel.
It also includes landscaping of the surroundings and other related infrastructural works
** assume a lifetime of 50 year; irregular maintenance in the 10th, 20th, 30th and 40th year
*** the selection of the discount rate is important; the higher the rate, the more attractive it is to reduce
the initial investment as much as possible and move costs as much as possible into the future
Table 2-3 NPV’s of maintenance strategies

For the above example cost figures have used partially from the writers own experience and partially to
arrive at a plausible answer and conclusion. In any way it is left to the reader to dig up appropriate (cost)
figures in real life and ‘nowadays’ design situations.

The above is a mere illustration of the fact that the operational aspect 'maintenance' will require
considerable attention during design, most appropriate in the more detailed design stages. In the
preliminary design stages the designer has to use engineering judgement.

2.1.3 Life cycle considerations regarding the Navigation Lock

Life Cycle Management (LCM) is a management approach to infrastructure construction to achieve
optimum quality and minimum Whole Life Cost (WLC). Whole Life Cost relates not only to the direct cost
of construction, maintenance, etc. of the structure itself but also to indirect costs and probable benefits
related to its use and the environment in which it is located. It also examines appropriate life cycle stages
such as design, construction, operation, maintenance (including inspection, evaluation and repair), re-use
and /or disposal. In principle LCM is aimed at providing minimum Whole Life Cost. This topic is dealt with

 Hydraulic Engineering
Faculty of Civil Engineering 20 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

in detail in the PIANC Report of MarCom WG42, Life Cycle Management of Port Structures:
Recommended Practice for Implementation.

During the last 20 years Life Cycle Management has become of increasing importance. Increased costs of
labour and mechanical parts lead waterway authorities to design more efficient locks (i.e. with a higher
quality/cost ratio). The priorities of different owners of locks varies, with some requiring minimisation of
maintenance effort and cost even if this leads to higher initial construction costs, while others require
minimum construction cost even if this leads to higher maintenance costs. A third category aim to assess
and minimise Whole Life Cost (including operational and maintenance costs) as their main criterion, and
in the absence of other investment or maintenance constraints this should be considered to be the
preferred approach.

For locks, a reduced WLC should imply optimum levels of reliability (which must be determined
specifically for each project), and more efficient maintenance. This does not mean less inspection and
survey or less maintenance. Locks normally have to be designed to allow efficient inspection and
daily/weekly maintenance without stopping traffic. Other means can also be adopted to assist in achieving
the objective of minimum WLC, including quantitative assessment of the costs of closures for inspection
or maintenance, and use of standardization of components as a means of reducing construction and
maintenance costs. Operating costs should also be considered as part of the optimisation process. These
can be reduced by adopting automation or remote control. This subject has been covered by PIANC
InCom-WG18 (1992) and PIANC WG96 (2008)

2.2 Navigation traffic and lock capacity

2.2.1 Duration of lock cycle and vessel passage

Taking the journey of a vessel as a starting point, which begins and ends with the vessel being loaded or
unloaded somewhere, the lock could be considered as a part of the trip where delays are likely to occur.
For the vessel's journey from port A to port B, the time needed to pass through a lock has a great
importance (see Figure 2-4). For inland navigation some of the items mentioned in this figure are not
applicable, as items 3, 4, 9 and 10.

This figure shows clearly that more time is required than just the time to navigate through the lock. What
the figure does not consider is the waiting time induced by a queue. Especially when lock capacity is lower
than the average traffic intensity, queuing time will be a substantial part of the delay faced by the vessel,
which is generally unacceptable. Before getting into the laborious work of optimizing the lock cycle and
reducing its duration, it is absolutely necessary to check, whether or not, the overall capacity of the lock is
sufficient regarding the vessel traffic to be handled.

The duration of lock passage has to be considered in terms of the vessel's journey, which includes looking
at the whole river or canal system, probably including more than one lock or even including series of locks.

Maritime navigation
1. Vessel reports in
2. Vessel decelerates 7. Sailing out & reporting out
3. Pilot boarding 8. Tugs loose
4. Tugs tying up Lock operation 9. Pilot disembarking
5. Vessel standby /not 10. Acceleration at ‘safe’
6. Entering lock chamber distance

Figure 2-4 Vessel passage through the sea lock (with assistance of a pilot)

 Hydraulic Engineering
Faculty of Civil Engineering 21 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

The reduction of the lock cycle of few minutes does not really seem significant with regard to vessel
journeys in the order of days. However, for container lines in ports behind a lock, fast locking and transit is
required. These container lines consider a lock as a general risk regarding the timetable to be respected.
The lock may even be considered as an unacceptable risk with regards of possible strikes or blockades,
or the occurrence of operational breakdowns.

Table 2-4 gives some information on the time required for some typical lock operations. The activities
required for locking have a duration in the order of ten's of minutes, except gate operation.

List of Events Average time % of the Average time % of the Possibilities for
Inland Navig. total Sea lock total optimisation to reduce
time time the total time
TOTAL LOCKING 28 min 100% 45 min (*) 100%
(1/2 cycle) (20 – 40 min) (40 – 90 min)
Entrance / Exit 5 min 18% 15 min (*) 33% Medium
(3 to 10 min) (10 to 20 min)
Mooring 5 min 18% 7 min (*) 15.5% High
(3 – 10 min) (3 – 10 min)
Gate manoeuvring 3 min 11% 3 min (*) 7% Low
(2-4 min) (2-5 min )
Filling / Emptying 15 min 53% 20 min (*) 44.5% High
(8 – 20 min) (10 – 25 min)
(*) For Panama the figures are: 45-60 mins for the existing Panamax locks and 80 minutes for the third
locks (in project)
• Entrance/exit: Existing locks 15 mins. Third locks 20 mins.
• Mooring: Existing locks 2 – 3 mins Third locks 5 mins.
• Gate maneuvering: Existing locks 2 mins. Third locks 5 mins.
• Filling/emptying: Existing 8-13 mins. Third locks 10-17 mins.
Table 2-4 Duration of navigation through a lock (1/2 lock cycle)

Figure 2-5 shows the lock cycle, distinguishing vessel and lock operations, for single and dual direction
use of an inland navigation lock. Looking for possibilities to optimize, read reduce the lock cycle in the
following circumstances and possible (counter) measures will be discussed influencing the duration. In
general, for single lock, dual direction use of a lock chamber is preferred because it maximizes the lock
capacity (compared to a single direction use). For a fleet of locks, as in Panama, single direction is
suitable.

In the absence of vessels in the lock chamber the emptying or filling operation can be executed much
faster because limiting the water movement and resulting mooring forces in the lock chamber is not
necessary.

Some ideas, which are not intended to limit, restrict or even prevent new innovative ideas, to reduce the
duration of the navigation cycle are:
- Pre-positioning of ships in the lock chamber to avoid waiting time – use of radio/radar/VTS : Use of
radio, radar or VTS to make sure the ship is being locked at the earliest possible lock cycle reduces
the waiting time.
Either the vessel may have to increase its speed or the lock operation is delayed until the vessel
arrives. The latter may be more feasible for a single direction lock chamber than for a dual direction
lock chamber.
- Faster movement of gates: It is imaginable that selection of a more innovative type of gates, other
than the traditional mitre, lift and rolling gate, reduces the time needed to open or close the gate.
- New concepts for the filling and emptying system to reduce the filling and emptying time: for instance
a non linear opening of the valves.
Equipment: self releasing mooring hooks, magnetic or vacuum based mooring devices, may be used to
reduce the time needed for mooring line handling and/or tending.

 Hydraulic Engineering
Faculty of Civil Engineering 22 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Point of time: Symbols:

1. Stern of vessel of previous locking passes tdeb = time for deberthing and manoeuvring into lock
gates chamber of the first vessel
2. Stern of first vessel to enter passes gates Σtin = interval between the first and last vessel
3. Stern of last vessel to enter passes gates (sterns) to enter the lock chamber for locking
4. Entry gates closed Σtout = interval between completed exits of successive
5. Exit gates start opening vessels
6. Exit gates open (first vessel starts leaving) tswitch = interval between exit of last vessel of preceding
7. Stern of last vessel leaving passes gates locking operation and completed entry of first
vessel of new locking operation
Figure 2-5 Lock cycle, indication of duration, distinguishing vessel and lock operations
In the above factors influencing the lock cycle and their effect on the lock capacity of an existing lock have
been discussed qualitatively. Combination of the lock capacity and variation of the traffic will result in a
certain performance level of the overall traffic handling system. To arrive at conclusions on the
performance it may be necessary to use quantitative results of a traffic simulation model. In the
Netherlands the Ministry of Transport and Water Management, Rijkswaterstaat, uses the SIVAK (2003)
simulation model for this purpose.

 Hydraulic Engineering
Faculty of Civil Engineering 23 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

2.2.2 Lock Capacity versus Traffic Intensity

Generally there is enough data available to express the traffic intensity, on a river or canal, in terms of
number of ships per hour, for both the upstream and the downstream direction. For the design of a new
lock in the waterway, the rehabilitation or upgrade of an existing lock the following data are required as
well:
• the seasonal, monthly, weekly or even daily variation of the traffic intensity
• the types of ship (ocean going vessel, barge, recreational craft) but also how they are distributed
(mixed)
• the frequency of special transport and vessels with dangerous cargoes
• the (in)balance between upstream or downstream traffic, or even the (in)balance in loaded or
unloaded vessels

Besides gathering the historic data a forecast has to be prepared as well.

Traffic intensity expressed in e.g. numbers per hour could be combined with data on the dimensions or
deadweight tonnage of the vessels. Subsequently the traffic intensity in numbers per hour can be worked
3
into deadweight/hour (DWT/hour) or m /hour. Especially expression in tonnage per time unit would be
useful for economic evaluation of a lock project. For technical design of the lock a requirement for the lock
3 3
capacity expressed in time, m and combined m /time unit is needed. Given the required lock capacity the
lock cycle of an existing lock can be evaluated and, for a new lock, the lock cycle requirements can be
determined. In case of an existing lock, comparison of what is required and what is available, results
either in the positive answer that the capacity is and will remain sufficient in the future or not. In the latter
situation lock cycle requirements have to be determined as for the case of a new lock project.

The factors influencing the lock cycle and their effect on the lock capacity of an existing lock are the
following:

• Type of the vessels and heterogeneity of the fleet:

Different types of ship have been distinguished in this report, i.e. ship, barge, recreational craft. A lock
designed specifically for ocean going ships generally will be less effective if mixed with inland barges or
recreational traffic. This mainly is the result of the dimensions of the indeed very different vessels. The
larger the mix in vessels the less effective the lock can be operated. In many cases the type of fairway or
location of the lock limits the variety of ship types and only a few governing ships are in fact considered for
the practical design

• Dangerous cargoes and special transports:

The transport of dangerous cargo or special transports by ship generally reduce the lock capacity because
of safety requirements. Generally other vessels are not allowed within a certain safety zone around the
considered ship or transport, which effectively prohibits entrance of more than one ship in the lock
chamber.

• Size of the lock chamber:

Given the lock chamber dimensions, it is quite easy to determine the number of ships that can enter the
lock, which depends on the size and number of the ships that arrived to pass the lock. The larger the
chamber the more ships can pass in one cycle, which increases lock capacity. Less effective filling of the
lock by ships, space wise, reduces the lock capacity.
To avoid emptying or filling a lock chamber that is only partially filled with ships, in many locks an extra set
of gates has been provided, say at 2/3’s of the lock’s length. This reduces the amount of water to be
released, thus saves time and increases the lock capacity. Unfortunately such intermediate gate cannot
be easily combined with water saving basins.

• The number of lock chambers:

A lock chamber can be considered as a service point for the ship traffic. The more service points the
larger the amount of traffic that can be dealt with.

 Hydraulic Engineering
Faculty of Civil Engineering 24 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

• The type of gates:

Depending on the type of gates the effective area of the lock chamber increases or decreases. See ‘Size
of the lock chamber’ above for the effects on the lock capacity. If the gate reduces the air clearance of
vessels some vessels require special measures to pass the lock. Although not substantial, this will
negatively influence lock capacity.
Depending on the type of gates installed (e.g. mitre gates versus vertical lift gate), the lock chamber
contains a larger or smaller volume of water. Consequently emptying or filling and duration of lock
passage, thus lock capacity is influenced.

• Gate opening and closing time:

As a separate item within the lock cycle the time needed to open or close the gates of the lock has a direct
effect on the length of the lock cycle. The longer opening and closing times, the longer the lock cycle and
the smaller the lock capacity. Given the dimensions of the lock, the time needed to open or close the
gates is determined by the used mechanical devices. Cautiously it could be assumed that a larger lock
requires larger gates, which would require more time to open and close. This is not a significant
parameter to reduce the locking time. See Table 2-4.

• Water level difference (lift height):

Given the efficiency of the locking system it takes a certain time to adjust the level of the water within the
lock chamber to the required outside water level. The larger the level difference, the smaller the final lock
capacity and vice versa.
In coastal areas and in tidal rivers the absolute and relative variation in water levels will be higher, which
generally negatively influences lock capacity.

• The operating speed of the filling and emptying system:

Usually the locking system has more sophisticated and has a higher capacity for larger water level
differences. Given a certain lift height, the faster the water level can be adjusted, the shorter the required
time and the higher the lock capacity. It is a major parameter to reduce the locking time, see Table 2-1.

• Water motion in the lock chamber:

Due account should be taken of water flow or waves resulting of the filling/emptying process within the
lock chamber. These hydraulic phenomena will induce ship movements, thus may result in unexpected
high mooring forces. To what extend mooring forces should be limited depends on the type of traffic.

• Water management:
Water Management requirements may concern:
• Saving the upstream water,
• Prevention of upstream or downstream pollution,
• Separation of salt and fresh water.

Saving the upstream water: Due to the water management requirements, water from the lock
chamber may have to be stored in saving basins, instead of being discharged directly in the
downstream reach. Depending on the systems used, this may take less or more time. Regarding the
lock capacity similar comments can be made as under “Water level difference”.
Prevention of upstream or downstream pollution: Pollution may be caused e.g. due to a ship-collision
or accidental industrial spillages, resulting in the loss of dangerous cargo, such as oil, gasoline,
chemical fluids. Depending on the kind of pollution the lock can be closed while the pollution in the
upstream or downstream river reach is being removed. This will result in congestion for navigation. It
is also possible to increase the discharge to the sea or downstream reach, in order to speed up the
dissolving or dispersion process of the polluting material and get to an acceptable concentration level
in the water.
Separation of salt and fresh water: Some technical solutions to reduce the salt water intrusion on fresh
water storage basins increase the lock cycle time. In case the entire lock chamber content has to be

 Hydraulic Engineering
Faculty of Civil Engineering 25 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

replaced, providing fresh water instead of salt water or vice versa, the lock cycle time increases very
much.

• Manoeuvring and mooring aids in and around the lock:

The easier the vessels can perform the necessary manoeuvres to sail in and out of the lock and the more
convenient the mooring facilities, the less time will be lost. Consequently the length of the lock cycle may
be significantly reduced by providing sufficient manoeuvring and mooring aids.

• Inspection and maintenance activities:

Insofar inspection and maintenance activities obstruct the effective use of the lock chamber, time wise or
space wise, the immediate result is a decrease in lock capacity.

2.3 Manoeuvrability while navigating through the lock and approach areas
The operational analysis of a ship approaching the lock, being lifted up or down, and continuing it's
journey, reveals that manoeuvrability of vessels is a key issue. Other reasons making it necessary to look
into manoeuvrability are safety and the duration of the lock cycle. Manoeuvrability is influenced by:
• Visibility
• Available water depth
• Cross currents and/or cross wind
• Available water area
• Approach structures
• Tug boats or other means for manoeuvring the vessel
• Twin propellers or bow thrusters

• Visibility:
For safe and expeditious navigation a clear and unobstructed view is important. This results e.g. in
requirements for the alignment of lock approaches and for positions of waiting or overnight berths at
sufficient distance to the track to be navigated by ships sailing in or out of the lock. For night navigation
and bad weather situations (fog and/or heavy rain) sufficient navigation aids and lighting have to be
provided.

• Available water depth:

The smaller the keel clearance of a ship the slower it will sail and the more its maneuverability will be
reduced. Given the draught of the governing design vessels sufficient water depth has to be provided.
Water levels on either side of the lock are variable and it should be kept in mind that erosion and accretion
may occur due to sediment transport. Therefore water depth is quite a dynamic variable in time.

• Cross currents and/or cross wind:

Cross currents and cross winds result in ships sailing at an angle with the axis of the approaches or the
lock itself. Subsequently wider approaches or a wider lock is required to prevent collisions with the
structures or running aground.

• Available water area:

Ship manoeuvring is more difficult in confined water areas. An escalation of manoeuvring problems in
confined waters is caused by the fact that reducing the sailing velocity in itself decreases manoeuvrability
of the ship.

• Approach structures:
Besides protecting the main lock structure for ship collisions, approach walls provide physical and visual
guidance to the ship respectively Master of the ship, hence; have a positive effect on the manoeuvres.
Opposed to jetty type structures, which are structurally light and open, guard walls are generally closed to
prevent cross currents and, to a lesser extent, also reduce the effect of cross winds.

 Hydraulic Engineering
Faculty of Civil Engineering 26 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

A reason not to construct a approach wall at a lock would be the availability of tug boat assistance. This
generally applies to locks in coastal areas and for locking maritime vessels. The extra space required for
the tug manoeuvres pushes the approach wall into an ineffective position for situations where tug boat
assistance is not required.

• Tug boats or other means for manoeuvring of the vessels:

To access approach walls, tug boats and e.g. the locomotives in the old Panama lock complexes enhance
the manoeuvring capabilities of vessels. The manoeuvres of the vessel require less space or are less
risky, but, in case of tug assistance extra manoeuvring space is required for the tugboat.

• Twin propellers and bow thrusters:

Although effective indeed, only a limited number of vessels has twin propellers or bow thrusters. Usually in
design resort to the use of approach walls or tug assistance has to be taken into account.

Poor visibility, small water depths and cross winds and currents negatively influence manoeuvrability,
therefore increase the size of the required water area. Approach walls, tug boats and extra propulsion of
the ship positively influence manoeuvrability. However, compared to a ship navigating through the lock
only using the main propeller, the use of these means does result in a slower process of ship passage,
thus a reduced capacity of the lock.

Other requirements, primarily the result of the ship passage function and important for the lock layout,
might be:

• Lay-by, waiting and overnight berths:

At the lay-by berths, close to the lock chamber, the ship(s) that will be the first to enter the lock are
waiting. Ship(s) that have to wait more than one full lock cycle stay at the waiting berths. At the overnight
berths ship(s) will be moored for the night or for a relatively long period with a duration unrelated to lock
operation. Space for lay-by, waiting and overnight berths, has to be positioned well out of the axis of the
fairway. Where applicable, separate or dedicated berths should be provided for ships carrying dangerous
cargoes or for recreational craft.

• Lightering area:
A zone or area where ships with too high a draught to pass the lock can be lightered. The lightering zone
includes manoeuvring areas around it and shall be at a safe distance from the sailing route in or out of the
lock.

• Dedicated lock chambers:

Provided the intensity of a type of traffic is sufficient it would be warranted to construct a dedicated lock
chamber fort that particular traffic. A dedicated lock chamber, e.g. for recreational navigation, has quite
some impact on the overall lock layout. Not only there is an extra lock chamber to be taken into account,
but the approaches to and the influence of discharging water to the other lock chamber(s) as well. A
sufficiently large apron between separate lock chambers reduces the effects of letting water in, or
discharging it, on navigation to or from the other lock chamber.

A quick look at the functions of a lock, other than ship passage, might result in requirements, subsequent
facilities and space requirements regarding the following:
• Removal of floating objects and ice floes. (See InCom WG23, PIANC 2004 )
• Prevention of sediment settlement or enabling sediment transport through the lock
• Dewatering or discharge flows due to the water management function
• Basins for salt-fresh water separation

From a Life Cycle Management point of view the following should be duly considered regarding the
location of the lock and/or the layout:
• Master plan of the waterway under consideration. What is the traffic forecast, what will be the number
of locks (a few high lift locks or many small lift locks), etc.?

 Hydraulic Engineering
Faculty of Civil Engineering 27 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

• Expansion possibilities of separate lock chambers or the lock complex as a whole

The examples in the above are not intended to be exhaustive, but they are issues likely to be
encountered. Other items or subjects can and should be added, if appropriate for the design under
consideration. Requirements for the dimensions of the lock chamber can be found in Chapter 3. For outer
ports and connected approaches, guidelines to establish layouts, more specific dimensions and
alignments, are included there as well.

2.4 Positioning of the lock complex

Consider a project to design and construct a lock (one lock chamber) or a lock complex (multiple lock
chambers, water saving basins, etc.). Depending on the area available, Greenfield or Brownfield, the
footprint of the required structures or lock facilities, the characteristics of the waterway, and the interaction
between the three previous factors, it has to be determined how and where the lock or lock complex will
be positioned. Amongst others, the following aspects should/could be taken into consideration when
determining the position of the lock or lock complex:
• The number of locks to be constructed in the waterway
• Lock approach layout
• Location of the lock and construction;
See the remainder of this section for the first item and section 3.2 for the last 2 items.

The number of locks to be constructed in the waterway

Provided there is sufficient water depth available for navigation, the lesser locks that have to be passed
the better. However, construction of a lock with a large head (large water level difference), where the
vessels have to be lifted or lowered over the full distance, may be more expensive than construction of a
couple of smaller locks. Considering e.g. the costs of dikes upstream of the lock may or may not swing the
balance. Those costs depend on the height and length of the dike. If h is the head, the height of the dike
2
varies linearly with h, but the costs vary with h ; because the area of the cross section has to be
considered. The length of the dike to be taking into account is related to h (again) and to the slope of the
river or waterway. Of canals without slope the length of the dikes is known (equal to the canal ‘s length).
Often the lock or lock complex is adjacent to weirs or barriers. In that case reasons to select the number
of weirs or the location of the barrier are of importance for lock design.

Often weirs and locks are constructed in

each others near vicinity. If e.g. a weir’s
objective is to produce hydroelectric power, it
will be situated at the location with the
largest difference in water level. In case the
weir’s primary objective is to maintain a fixed
water level at a given location, it will be
situated downstream as close as possible to
the location of interest, so as to minimize the
dimensions (especially) height of the
structure. Determining the position of a weir,
in a river with a small discharge in order to
create navigable waterways, is more
complex. The river will be divided up into a Figure 2-6 Number of weirs and the maximum
number of sections. In each section the water head
water level will be regulated by a
downstream weir. The total number of weirs and the maximum water head over the weir determines the
rough position of the weirs. Figure 2-6 provides some illustration of the above mentioned.

 Hydraulic Engineering
Faculty of Civil Engineering 28 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Intermingled as it may be, the optimum number of weirs and/or locks is determined by a number of
economic factors, such as:
• The construction costs of the weirs, locks and the embankment or dikes etc.
• The ship navigation costs for the time delay having to pass the weirs or locks besides it
• The damage to the surroundings, the environment in general and the (increased) risk of flooding
• The maintenance and operation costs
• The revenue generated by lock operations (toll on ship passage)
• The revenue generated by hydroelectric power
• etcetera
In the following paragraphs we will take a closer look at some of these cost or economic factors in relation
to ‘a number’ of locks to be designed and constructed.

Nowadays Life Cycle Management is being introduced into all sorts of civil engineering projects and into
the design of the project, see section 2.1.3. In agreement with the life cycle stages the following costs
would have to considered:
• Design costs
• Construction costs
• Operational costs
• Reuse and/or Removal costs

Design costs
A rough, over the thumb figure for design costs would be between 5 and 10% of the overall project costs.
This indication is based on experience in the past decades, where design spanned from the idea or
initiative stage to the construction stage, hence, excluding design in or for operation, maintenance and
removal. Obviously within a Life Cycle Management approach of projects these stages have to be
included, increasing the mentioned percentages.
For larger civil infrastructure projects it is advised to take into account the upper boundary, 10%, because
the societal and/or governmental decision making process generally is very complicated, hence takes a
long time and continuously invokes extra design activities. Generally lock and/or weir or barrier projects
fall into this category.

Construction costs
The following costs will be included and described as construction costs:
• Construction costs per weir depending on water head
• Costs of mitigating measures
Obviously construction costs are not limited to the items mentioned.

Construction costs per weir depending on water head:

The total horizontal force acting on a weir is a function of the water level difference over the weir. The
following force function can be written down:

1 1 
F = b  ρ g ( H + d ) 2 − ρ gd 2 
2 2 

The construction cost will be dependent on the total moment forces.

1 1 1 1  1
M = b  ρ g ( H + d ) 2 ( H + d ) − ρ gd 2 d  = c2 ( H 2 d + H 3 + Hd 2 )
2 3 2 3  3
Where: b = width of the weir
H = water head (average in time)
d = downstream water depth relative to the foundation
ρ = water density
c2 = constant (½ ρgb)

 Hydraulic Engineering
Faculty of Civil Engineering 29 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

The weir structure may be a bit smaller or larger but there will be a base sum of money that has to be paid
anyway (Mobilisation of equipment, labour, temporary works). Besides the more or less fixed cost part
there will be a variable part of the costs that will vary with the water head (time variation) over a weir see
formula above. Thus, from a certain base sum the construction costs of the weir are proportional to the
load on the structure. Written down as:
Where:
1 3 c3 = costs per unit of load on structure
cweir = C1 + C2 ( H d + H + Hd )
2 2

3 C1 = fixed costs
C2, = constant
The construction costs of n weirs Cconstr,n for the full length of the river amounts to:
1
Cconstr ,n = ncweir = nC1 + nC2 ( H 2 d + H 3 + Hd 2 )
3
2 3
H tot 1 H tot H
Say that nH is roughly constant, then: Cconstr ,n = nC1 + nC2 ( 2 d + 3
+ tot d 2 )
n 3 n n
In the graph on the right there are
250 000 000
two examples of a function
representing total construction €
200 000 000
graph costs. Observe that the
construction costs in the first
150 000 000
example are minimal with the
construction of 3 weirs and in the
second example with the 100 000 000

construction of 2 weirs. The

difference in both cases is the ratio 50 000 000
of the coefficients C1, C2 and C3.
This ratio is important to determine
the linearity of the function. If the 1 2 3 4 5 6 7 8 9 10

fixed costs are relatively high and C1 = 10 000 000 €; C2 = 700 €/m2
C1 = 20 000 000 €; C2 = 560 €/m2 number of weirs ==>
the water depth is large compared
to the total water level difference,
Figure 2-7 Influence of C1 and C2 on construction costs
then the costs function behaves
350 000 000
itself as a linear function. This is
demonstrated in Figure 2-7. € 300 000 000
The graph shows that the least
possible number of weirs always 250 000 000
results in the lowest construction
200 000 000
costs, which is not a very
surprising conclusion, however the 150 000 000
choice for the number of weirs is
usually not based on minimum 100 000 000
costs alone; the benefits or
50 000 000
completely different criteria have
to be considered.
1 2 3 4 5 6 7 8 9 10
C1 = 10 000 000 €; C2 = 700 €/m2 number of weirs ==>
C1 = 30 000 000 €; C2 = 300 €/m2

Figure 2-8 High fixed costs (C1) resulting in linear

construction costs

 Hydraulic Engineering
Faculty of Civil Engineering 30 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Costs of mitigating measures:

Side effects upstream from the weirs include a higher level of the ground water table, discharge problems
of local tributaries or discharge canals as a result of the higher water and groundwater levels, and of
course local levee (dike) enforcement. In case of a fixed weir, levee enforcement is necessary after
construction of the weir due to the achieved higher water levels in general and particularly during a high
runoff period. A movable variable weir allows for a temporary lapse in the weir-effect during this high
runoff period. However even for a movable weir, locally the water levels will be higher due to the smaller
cross section at the weir. Hence for both types of weir an enforcement of the levees is needed. For the
cross section of a river with forelands and summer levees, it is usually undesirable to have submerged
forelands during low runoff periods. In case of more permanent submerged forelands the summer levees
would have to be enforced.

Costs of mitigating measures can be approximated by a second order polynomial equation. The
2
construction costs of the levees and the costs of the draining are directly proportional to H . The costs for
the levee enforcement, more specific for bed or bank protection, are usually proportional to H.

A reasonably well expression for the costs Cmiti would be:

H tot 2
Cmiti = nC3 +
l 1
n6
( C 4 H 2
+ C5 H ) = C 4
n 2
H
+ C5 tot
n

Where: C4 , C5 = coefficients determined by extensive research into the possible

damages and measures

Besides the costs of additional civil engineering works, one must also take into consideration the full costs
of changes or damage to the ecologic system. By constructing the weir, the existing food cycle of the
surrounding waterway will be disturbed. Small animal species that can be counted as prey for larger fish
will be hampered in their descent of the river. This would lead to a disaster for the local fishing industry,
which can be translated fairly direct into financial costs. Alternatively the extra costs for provision and
maintaining fish migration facilities have to be taken into account.

The graph in Figure 2-9 illustrates the development of mitigation costs in time.

300 000 000

€
250 000 000

200 000 000

150 000 000

100 000 000

50 000 000

1 2 3 4 5 6 7 8 9 10

C3 = 2500000 €; C4 = 50000 €/m/km; C5 = 500000 €/m2/km number of weirs ==>

C3 = 2500000 €; C4 = 50000 €/m/km; C5 = 500000 €/m2/km

Figure 2-9 Mitigation costs in time

 Hydraulic Engineering
Faculty of Civil Engineering 31 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Operational and maintenance costs

The following costs will be included and described as operational costs, which is not completely correct; it
would be better to consider them as costs related to the operational stage of the project:
• Operational costs for every lock and weir complex
• Maintenance costs
• Delay costs ship navigation
Obviously the operational costs are not limited to the items previously mentioned.

Operational costs

The operational costs can be assumed to be proportional to the number of weirs. One should think about
the cost of operating of the weir and the locks (if required), electricity and other utilities for lighting, daily
maintenance etc.. In general manpower or labour costs are the major to be considered.

Maintenance costs

The maintenance costs per annum of a hydraulic structure are dependent on the nature of the structure
and the used materials. Usually the maintenance costs are a percentage of the construction costs, on
average 1% per year. For a first estimate of the number of weirs to be built, this percentage can be used
for the determination of the costs.

The total operational and maintenance costs can be summarised as follows:

CO & M = cwf ( γ n + δ Cconstr ,n )

Where: γ = annual operating costs of a weir

cwf = discount factor
δ = roughly 1%
Cconstr,n = total construction costs

The nearly linear cost function for operation and maintenance is demonstrated by means of a graph in the
figure below.
160 000 000
€ 140 000 000

120 000 000

100 000 000

80 000 000

60 000 000

40 000 000

20 000 000

1 2 3 4 5 6 7 8 9 10

C1 = 10000000 €; C2 = 700 €/m2 number of weirs ==>

C1 = 20000000 €; C2 = 560 €/m2

Figure 2-10 Operation and maintenance costs in time

 Hydraulic Engineering
Faculty of Civil Engineering 32 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Delay costs ship navigation

The cost of shipping delay is caused by the time it takes to pass the weirs or adjacent locks. Depending
on the intensity of ship traffic and the capacity of the weir or lock a queue may develop, resulting in ship
waiting time. Obviously, time is also consumed by the lock process; which includes the time needed to
open and shut the gates of the lock, the time required for adjusting the water level in the lock chamber, the
time needed for the ships to sail in and out of the lock and last not but not least, the number and size of
the ships per lock cycle. Simulation techniques are used to estimate the delay time for the (overall)
shipping process and in particular for the lock navigation process. Using the delay costs per unit of time,
the number of ships etc. one arrives at the (total) delay costs. A considerable number of simulations will
be made to vary the number of weirs with varying water level differences to estimate the delay the ships
will endure. Meaningful use of the simulation is a job on itself and, unfortunately, often simulation is not
possible in the early stages of a project because the necessary data is not available, or there may not be
enough budget.

For the purpose of rough cost calculation, needed to decide on the number of required weirs, a simple
method of approximation can be used. It is assumed that the delay time to for ships is divided in a waiting
period tw (the ship in the queue), and a lock period ts (the time to pass the lock).

t0 = t w + t s

Where: tw = queue time; dependent on shipping intensity and lock capacity

ts = lock cycle; the (average) time the average ship needs to pass the lock

If there is a stretch of river in which there is more than one sluice, then the ideal situation would be that;
for the first sluice there is a queue and lock cycle time and that for each sluice after that only lock cycle
time. In this scenario the shipping traffic is running optimally and the shipping speed is equal for all ships.
At the entrance of the first sluice the shipping traffic is optimised.

De lock cycle time can be approximated with:

ts = α + β H

Where:
α = the time for opening and closing of the lock gates and the time required
(by ships) for entering and leaving the lock chamber
β = coefficient, dependent on the dimensions of the lock chamber and the
capacity of the (water) filling and emptying system

The total delay for a ship:

H tot
t0 = tw + nts = tw + nα + nβ H = t w + nα + nβ = tw + nα + β nH tot
n

If per year there are N ships that pass through the weirs and the costs for the delay per ship on average is
equal to P Euros per hour, see Table 2-5 below for information on inland navigation costs, then the total
costs per annum:

C j = PN (t w + nα + β ( n H tot )

If a comparison is to be made between the construction costs and the total ship delay costs Ctot,o, the net
present value of both costs should be used and the lifetime of the weir has to be known. Usually weirs are
designed for a lifespan of 50 years.

 Hydraulic Engineering
Faculty of Civil Engineering 33 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

The costs over the full lifespan n years can be transferred to their net present with the formula below,
which takes inflation and interest rates into account:
T 100* C j
Ctot ,o = ∑
j =1 (100 + r ) j

Where: r = discount rate in % (financial interest rate on the market – inflation)

T = lifespan

In the previous formula the costs of ship delay (may) vary from year to year. If the costs per annum remain
constant then the net present value of the costs is:
T 100 * C j T
100
Ctot ,o = ∑ = Cj ∑ = C j cwf
j =1 (100 + r ) j
j =1 (100 + r )
j

Where: cwf = discount factor (Dutch: contante waarde factor)

With a lifespan of 50 years, 50 discount factors and their sum have to be computed, e.g. by means of a
spreadsheet. To avoid this cwf can be approximated by:
T
100 100  100 
cwf = ∑ = 1 −  for r < 8 this approximation is very inaccurate
j =1 (100 + r ) r  (100 + r )T 
j

The total costs for the delay can now be defined as:

(
Cdelay = cwf * P * N tw + nα + β nH tot )
The delay costs function increases monotonously with n. An example of this can be found in Figure 2-11.
1 000 000 000
€ 900 000 000
800 000 000
700 000 000
600 000 000
500 000 000
400 000 000
300 000 000
200 000 000
100 000 000

1 2 3 4 5 6 7 8 9 10
N = 21000 -; tw = 1 hour number of weirs ==>
N = 21000 -; tw = 2 hour

Figure 2-11 Delay costs linearly increasing with no. of locks

Stating the obvious: a minimum number of weirs results in minimum ship delay costs.

In Table 2-5 the cost and relative cost of inland waterway vessels can be found. Actually this information is
very hard to find if at all; so it should be used wisely.

 Hydraulic Engineering
Faculty of Civil Engineering 34 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Relative level of cost, based on price levels in 2005 in the Netherlands,

Rhine vessel sailing loaded taken as 100% (100%= 200€/hour in 2005)
(Ref: Rijkswaterstaat, Inland Navigation Model System, BVMS)
Tankers
CEMT Ship types Tonnage Loaded Empty Waiting
class (ton) Sailing Sailing at lock
0 Others 150 24% 24% 21%
I Barge 350 33% 32% 27%
II Campine Barge 550 40% 39% 32%
III Gustav Koenings / Dortmund-Eems 750 49% 46% 38%
Length <= 74m 950 57% 54% 44%
Length > 74m 1150 66% 62% 51%
IV Johan Welker / Rhine-Herne (L <= 86m) 1550 84% 79% 64%
Johan Welker / Rhine-Herne (L > 86m) 1950 113% 106% 89%
Va Rhine Vessels 2500 157% 148% 127%
Cargo vessels
0 Others 150 18% 18% 15%
I Barge 350 23% 22% 17%
II Campine Barge 550 29% 27% 21%
III Gustav Koenings / Dortmund-Eems 750 35% 33% 25%
Length <= 74m 950 42% 39% 30%
Length > 74m 1150 50% 46% 35%
IV Johan Welker / Rhine-Herne (L <= 86m) 1550 65% 60% 46%
Johan Welker / Rhine-Herne (L > 86m) 1950 79% 73% 56%
Va Rhine Vessels 2500 100% 92% 71%
Container vessels
0 Others 150 17% 17% 15%
I Barge 350 23% 22% 18%
II Campine Barge 550 29% 28% 22%
III Gustav Koenings / Dortmund-Eems 750 37% 34% 28%
Length <= 74m 950 45% 42% 34%
Length > 74m 1150 55% 51% 42%
IV Johan Welker / Rhine-Herne (L <= 86m) 1550 74% 69% 56%
Johan Welker / Rhine-Herne (L > 86m) 1950 92% 86% 71%
Va Rhine Vessels 2500 119% 111% 92%
Table 2-5 Relative level of cost for vessels, sailing and waiting

Reuse and/or removal costs:

Reuse and/or removal costs are by nature hard to predict when considering a lifespan of the structure in
the order of 50 to 100 years, which is not uncommon at all for lock and weir projects. The lifespan that
complicates the actual estimate, at the same time makes the cost estimating work easier because the
number of years inevitably results in a low discount factor cwf.
Reuse costs will be similar to costs that have to be made for any Greenfield or new built project, except for
having to consider costs for removal of parts of the existing structure; those costs come on top of the
required budget. Off course only the relevant ‘to be reused’ parts of the structure will be taken into
account.
Removal costs can be split in costs for demolishing the structure, or parts of the structure, and costs for
disposal of the construction and demolition waste or rubble. It is not uncommon to find there is a supply
and demand market for demolition material, especially (crushed) concrete and steel, resulting in negligible
disposal costs. On the other hand, take good notice of the fact that the costs for disposal of polluted
materials, e.g. polluted sediments or soil, can be considerable.

 Hydraulic Engineering
Faculty of Civil Engineering 35 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Total costs depending on number of weirs and water head

The total costs to be considered when one or more weirs have to be constructed and operated can be
estimated by taking the sum of all the above mentioned costs. The total costs Ctot are:

Ctot = Cconstr,n + Cdelay + Cmiti + CO&M



CH 2
( H 2
n
) H
=  nC1 + C2 Htot + 3 tot  (1+ δ cwf ) + cwf * P* N tw + nα + β nHtot + C4 tot2 + C5 tot + cwf γ n
n  n
Using this expression for the total costs, one can calculate the cost differences for various numbers of
weirs. The optimum number of weirs is the number for which the total costs are lowest.

In the following figures first a graph showing all the contributing cost items for one set of C-coefficients,
see Figure 2-12. Then 4 sets of C-coefficients have been used to produce the graph in Figure 2-13; for all
variations of C-coefficients an optimum can be seen for 2 weirs.

1 200 000 000

Constr
Delay
1 000 000 000
Miti
Opmai
800 000 000 Total
€
600 000 000

400 000 000

200 000 000

1 2 3 4 5 6 7 8 9 10
number of weirs ==>

Figure 2-12 Costs per item and total costs

1 600 000 000

€ 1 400 000 000

1 200 000 000

1 000 000 000

800 000 000

600 000 000

400 000 000

200 000 000

1 2 3 4 5 6 7 8 9 10
number of weirs ==>
Figure 2-13 Total costs for different sets of C-coefficients

 Hydraulic Engineering
Faculty of Civil Engineering 36 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

3. Overall lock design

3.1 Lock geometry and lock configuration

3.1.1 Typical inland waterway lock lay-out and geometry

After the analysis phase and preparing a comprehensive set of design criteria, the design work gets into
the phase of sketching or drawing the first overall concepts of the lock. Below a top view and longitudinal
section are presented of a typical inland navigation lock in the Netherlands.

1. Waiting- or lay-by berths 1. Wachtplaatsen

2. Guard- or guide wall or lead-in jetty 2. Geleidewerken
3. ‘Closing elements’ – lock gates 3. Afsluitmiddelen – sluisdeuren
4. Lock heads 4. Sluishoofden
5. Lock chamber 5. Schutkolk
6. Filling and emptying system 6. Vul- en ledigsysteem
7. Cut-off walls and screens (sheetpiles) 7. Voorzieningen tegen onderloopsheid en
to prevent piping achterloopsheid (kwelschermen)
8. Bottom protection 8. Bodembescherming
Figure 3-1 Lay-out of a navigation lock
The waiting berth [1] provides a safe mooring place for vessels that have to wait before being allowed to
enter the lock. A guard wall [2] is a navigational aid; facilitating the entrance to the lock and protecting the
structure when the ship loses rudder control. Entering locks may be difficult and potentially dangerous due
to crosswinds or local currents. For navigation locks in coastal areas a different layout may be more
appropriate.
When entering a lock the vessel will pass the (mitre) gates [3], which are built into the lock head [4]. The
closed gates and the lock head must be able to retain the head difference over the lock, thus they should
be designed to resist the forces caused by hydraulic pressure. Forces on the gates are transferred to the
supports, sometimes to the sill, in the lock head and result in a spalling force (spatkracht) on the lock head
walls and a (horizontal) line load on the sill respectively. During filling or emptying the ships remain in the
lock chamber [5].

 Hydraulic Engineering
Faculty of Civil Engineering 37 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

In general the lock process can be described as follows: The lock chamber is closed off from the outside
water by closing the gates [3]. Subsequently the water level inside the chamber must be adjusted to the
desired level; assuming the ship sails from left to right in the above figure, the water level must be raised.
When locking in the opposite direction the water level in the lock chamber will be lowered. In order to
achieve this, a filling and emptying system [6] is required, e.g. ducts in the lock heads [4]. After adjusting
the water level to the outside level the gates are opened, the ship sails out and continues its journey.
Usually the lock chamber is provided with berthing and mooring furniture, such as fenders, bollards,
mooring rings and ladders in recesses (inkassingen). All furniture has to be able to resist (predominantly)
horizontal forces due to berthing, mooring and deberthing.

Filling and emptying the lock chamber will cause local currents. These currents, as well as the propeller
wash caused by the navigating ships, may result in scour in front of the lock heads (on either side), which
in time may compromise the stability of the lock structure. To prevent this bottom protection [8] can be
provided.
Typically, water levels will be different at both ends of a lock. This will result in a ground water flow
underneath and on both sides of the lock structure. This seepage may lead to movement of soil particles
leading to cavities under and alongside the lock. To prevent this, locks must be provided with seepage or
cut-off walls and/or cut-off screens (kwelschermen), usually made of sheetpiles [7], acting as an
impermeable curtain perpendicular to the axis of the lock chamber.

Together with dikes (US: levees) and other structures the lock may be part of a water defense system.
The upper lock head (at the side of the highest water level) [4a], the upper head, must be able to retain
the highest water level. The lock head at the side with the lowest water level, the lower head [4], retains
the highest water level in the lock chamber, which is selected by the designer and not necessarily the
highest outer water level. In tidal areas lock heads are referred to as outer head (on the sea side) and
respectively inner head (on the land side).
As part of the water defense structure the navigation lock must be able to retain water under all
circumstances. If for any reason (e.g. a ship runs into a gate) the lock is damaged other measures must
allowed for. E.g. locks are often equipped with spare gates, bulkheads or stop logs (schotbalken) are
available to replace damaged gates and maintain the water retention function.

3.1.2 Special lock geometries and/or lock configurations

The lock chamber (schutkolk) of some historic locks is wider than the entrance to the chamber, whilst the
axes of the upstream and downstream reaches run parallel; are not in line, see Figure 3-2. Besides the
historic bayonet lock in France even a circular lock can be found in Lequetin, which is even worse from a
manoeuvring point of view; back then ships were smaller and shipping much slower.

Figure 3-2 Bayonet lock and circular lock

In modern times the ships must enter and leave the lock as soon as possible, time seems to be money,
and this has resulted in a more rational layout of the lock. The chamber width is the same as the entrance
width in the lock head. A cross section of the lock often has an U-shape, vertical retaining walls and of
course the horizontal floor. Together with well positioned guard walls this facilitates navigation because
the situation is easily observed and assessed by the master of the ship; a clear situation enables efficient

 Hydraulic Engineering
Faculty of Civil Engineering 38 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

navigation. A big help is to avoid assembling all sorts of furniture to the lock walls. If bollards, fendering
and ladders protrude from the wall they are observed as bigger obstacles than they are because of their
actual size; they cloud observation by the master of the ship who reduces speed accordingly. The furniture
should be assembled in recesses (uitsparingen) in the wall, maintaining a flat face, see Figure 3-3

Figure 3-3 Lock chamber (laid open) (drooggezet)

Depending on the capacity and intensity of a waterway one could consider constructing an extra set of
gates to subdivide the lock chamber in segments. In this way the lock cycle can be managed more
efficiently. The extra gates are often positioned at 1/3rd of the lock length, see Figure 3-4 (not to scale).
Advantages of such lock chamber partitions are:
• Time needed for filling and emptying the lock chamber is being reduced;
• In coastal areas the intrusion of salt water is being reduced;
• In case of damage due to collision the navigation lock is still partially operational.

Figure 3-4 Lock chamber partitions using different gate

configurations

 Hydraulic Engineering
Faculty of Civil Engineering 39 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

3.2 Lock Approaches

3.2.1 Lock approach layout

For positioning the lock not only the dimensions of the lock chamber(s) is of importance, but the size and
shape of the lock approach areas and the requirements to connect these navigation areas to the waterway as
well.

Outer port layouts - Inland navigation

General reference is made to the RVW (2005). The Figures below show the approach area or outer of a
lock. Often the adjectives upstream or downstream are added. Some times upper or lower pond is also
used as a name or description. Shown in the Figures are the length and widths, if applicable, and their
definitions. Subsequently Table 3-1 shows the dimensions of these lengths and widths depending on the
inland navigation vessel type (according to the CEMT class).

Figure 3-5 Approach area or outer port of a lock

Table 3-1 shows the dimensions of CEMT B Bch S Br Llb/Lch

characteristic lengths and widths for the outer Class
ports for different CEMT classes. [m] [m] [m] [m] [m]
I 5.1 6.0 3.0 5.0 1.0 – 1.2
II 6.6 7.5 3.5 6.0 1.0 – 1.2
III 8.2 9.0 4.5 7.5 1.0 – 1.2
IV 9.5 10.5 5.0 8.5 1.0 – 1.2
Va 11.4 12.5 6.0 10.5 1.0 – 1.2
Vb 11.4 12.5 7.0 11.5 1.0 – 1.2
Notes: - See figures for meaning of the symbols
- Lch is the length of the lock chamber
Table 3-1 Dimensions of outer ports with
berths on one side

 Hydraulic Engineering
Faculty of Civil Engineering 40 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Figure 3-6 shows the layout for

the outer port of a lock wit one
lock chamber. At the other side
the layout would be identical,
however, point mirrored.

Figure 3-6 Outer port with berths on one side

In waterways serving traffic with

an intensity warranting
construction of two or more lock
chambers, the outer port looks
as in Figure 3-7.

Figure 3-7 Outer port for twin lock and berths on both sides

When traffic intensities are too

low to construct dedicated lock
chambers, not only the lock
chamber but the outer port as
well, has to be shared by
different types of vessels. Figure
3-8 shows a layout for the outer
port where some of the berthing
structures can be used for both
professional and recreational
craft and still be kept separated
from each other. Figure 3-8 Outer port with waiting berths for recreational
craft

 Hydraulic Engineering
Faculty of Civil Engineering 41 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

When the waterway is a river where the discharge is controlled by a weir, often the lock and weir are
positioned adjacent to each other at one location. Multiple functions combined in one complex, resulting in
some extra demands to the layout of the lock outer port, or, vice versa, demands to the inlet and outlet of
the weir, as schematically shown in Figure 3-9. Care has to be taken that eddies or cross currents are
kept away far enough from the areas where manoeuvring is complicated, e.g. close to the lock entrance
areas. Approach walls can be used to mitigate the effects of adverse currents.

Figure 3-9 Lock approach area, discharge and injection

channels

3.2.2 Alignment requirements with respect to the inland waterway - inland navigation
General references are available in RVW (2005), PIANC InCom WG16 (1996b), PIANC InCom WG20
(1999), and PIANC Marcom WG49 “Horizontal and vertical dimensions of fairways”.

Numerous locks are or will be constructed in a curve of a waterway. Therefore the alignment of lock and
approaches has to suffice to requirements that apply curves in the waterway. The following applies to a
two lane waterway where the maximum current is 0.5 m/s.

The radii of curves in waterways should be chosen above a certain minimum:

• because rudder angle does not translate linearly in ship rotation
• to avoid speed reduction in the curve
• for required course corrections
• the loss of unobstructed view

Minimum radius: R ≥ 4L
Preferred radius: R = 6L

R = radius of the curve, relative to the axis of the waterway

L = length of the governing ship

Sailing through a curve the ship needs extra space, hence a width allowance ∆B should be added to the
lane width as follows:

2
∆B = 0.75 L /R

o
However, the width allowance can be reduced for β ≤ 30 , where β is the angle between the axes of the
straight stretches adjacent to the curve.

2 o
∆B = 0.75 L /R *β/30 for β ≤ 30

o
For R>10L, in that case β<20 , no width allowance is required, the curve is already sufficiently smooth or
wide enough.

 Hydraulic Engineering
Faculty of Civil Engineering 42 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Preferably the extra width in the curve will be provided in the inside curve. If this is not possible the outside
can be used or distribution over the in- and outside is possible.

Right before and after the curve a straight

channel stretch of 1.5L should be available. Care
has to be taken of a smooth transition from the
wide curve to the narrower channel or river
section. An angle α is proposed to define start
and end of the transition, see Figure 3-10.

Figure 3-10 Alignment and width corrections

Taking care of a clear unobstructed view in a curve, air draught of the vessel, prevailing water levels and
level of the bank, including buildings, come into play. Over a length of 5L the view has to be unobstructed,
see Figure 3-11. Visibility requirements for junctions of waterways are shown in Figure 3-12.

Figure 3-11 Required clearance for Figure 3-12 Required clearance for visibility at
visibility in curves junctions

3.2.3 Location of the lock and construction

Navigation locks are well used in the Netherlands since this is the most economic way to transfer ships
from one water level to the other for small differences in water level (< 10-15m). Since navigation locks
are used in case a ship has to pass a permanent water retention structure or a weir, logical locations to
find a navigation lock are:

• In a river next to a weir.

• At the start and/or end of a waterway with a controlled water level.
• In coastal areas, at a passage of a primary water retention structure.

 Hydraulic Engineering
Faculty of Civil Engineering 43 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Figure 3-13 Construction of lock-weir complex Hagestein, river Lek the Netherlands

3.3 Horizontal dimensions lock

Considering ship dimensions, there is a difference in the information available between inland and
maritime navigation. Generally the classification of ships in inland navigation seems to have progressed
more than classification of maritime vessels.

a) Inland navigation

The dimensions of a lock chamber in Europe are shown in Table 3-2 (RVW, 2005). The dimensions are
based on the CEMT table of 1992, for all waterways west of the river Elbe - Germany, and are applicable
for the so called minimum lock, which is a lock operating in a traffic situation where less than 10 000
inland navigation vessels, either in the up- or downstream direction, are locked every year.

CEMT Vessel size Lock chamber dimensions (*)

classes
1 2 3
length beam Draught length width sill depth
[m] [m] [m] [m] [m] [m]
I 38.5 5.05 1.80-2.20 43 6.0 2.8

II 50-55 6.60 2.50 60 7.5 3.1

III 67-80 8.20 2.50 90 9.0 3.1

IV 80-85 9.50 2.50-2.80 95 10.5 3.5

Va 95-110 11.40 2.50-4.50 125 12.5 4.2

Vb 172-185 11.40 2.50-4.50 210 12.5 4.7

1. The length between the lock heads

2. The clear width between the lock chamber walls or protrusion from the wall
3. The sill depth takes into account the draught of the vessel, squat and keel clearance. However,
no allowance has been included for translation waves
4. (*) The lock chamber dimensions do not take into account that the governing ship might be
entering the lock chamber with tug assistance.
Table 3-2 Dimensions lock chamber for inland navigation depending on CEMT classes

 Hydraulic Engineering
Faculty of Civil Engineering 44 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

b) Maritime navigation

Generally the variety in ships and ship sizes passing a maritime navigation lock is quite large. This may
explain why a convenient table with lock dimensions related to maritime vessels can not be found. Some
guidance on the dimensions of ships can be found in EAU 2004 (Recommendations on Waterfront
Structures Harbours and Waterways); PIANC MarCom WG48 Guidelines for Berthing Structures;
Thoresen C. (2003); and in ROM3.1-99, 2000 (Recommendations for the design of the maritime
configuration of ports, approach channels and harbour basins).

Usually the nautical conditions in coastal areas are more adverse than in inland regions and the shape of
maritime vessels is prone to a larger impact of those conditions on manoeuvrability. The faster the vessels
are able to manoeuvre in and out of the lock (chamber), the higher the traffic capacity. It all adds to the
recommendation to use simulation to determine the size of the lock chamber for maritime locks.

To start real or fast-time simulations to find the required size of the lock chamber a set of initial values for
the width and length has to be available. It is advised to do the following:
• Define the governing or design ship that should be able to pass the lock
• Assume a minimum lock chamber suffices from a traffic capacity point of view
• The length of the chamber is approximately 1.1 times the length of the design vessel.
• The width is 1.25 times the width of the design vessel.
The simulation work will be continued to check the assumption made. A variety of vessels with different
transit times has to be used to find out whether or not the capacity of the lock is sufficient.

In case the lock chamber is designed to accommodate 2 vessels, either in length or in width direction,
space for manoeuvring will be added only once, since the vessels enter the lock one by one. Hence, the
length is:
Lchamber= L+ (L x 1.1).

and similarly:
Bchamber= B+ (B x 1.25)

An upper boundary for the width of a lock chamber, assuming the ship is navigating on its own propeller
would be:
Bchamber= 2 x Bvessel.

The above initial lock chamber dimensions do not take into account that the governing ship might be
entering the lock chamber with tug assistance.

When tugs or other active means, other than the vessels own rudder and propellers, are used to control
the course of the vessel the required lateral clearance will be smaller than without assistance. The design
vessel for the new Panama lock complexes is a 12,000 TEU containership with a beam of 48.8 m. The
specified width between the lock chamber walls is 55 m, the minimum clearance on each side of the
chamber is 0.3 m. Hence, the usable width is specified at 54.4 m. If the useable width were to be used
completely the lock width over beam ratio would be 1.01. The useable lock width over design vessel’s
beam ratio is 1.15.

Obviously it is not recommended to use the ratio of 1.01 to start simulations for new lock design because
it applies to exceptional events. For the more frequent event of the design vessel navigating the lock, the
ratio of 1.15 can be considered as the minimum to start.

The depth of the sill should be determined using the maximum draught of the design ship with regard to
the selected minimum water level in the lock chamber, and including at least 1 meter nett keel clearance.
Selection of the minimum waterlevel in the lock chamber is governed by operational or availability criteria,
and depends on waterlevel variations, and related frequency of occurrence on both sides of the lock. Note

 Hydraulic Engineering
Faculty of Civil Engineering 45 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

that the stern of the vessels sinks deeper into the water when the engine is put to work again. Sometimes
it may be necessary to construct the sill at a lower level considering the volume of water that has to be
discharged and the ratio between the cross sectional area of the ship and the water area in the locks
cross section, when the ship enters the lock chamber. Fresh versus salt water phenomena may add to
this requirement.

c) Recreational navigation:

Depending on the traffic intensity recreational craft will have its own dedicated lock chamber or has to
make use of the spare capacity of the lock chamber dedicated to commercial navigation.

Considering the construction costs of a lock, a dedicated lock for recreational craft will not be a minimum
lock chamber, i.e. for use of one yacht only. To determine the size of the chamber it is advised to simulate
the traffic and lock situation.

Figure 3-14 shows some configurations for locking both commercial and recreational craft in one lock
chamber at the same time. Commercial vessels should get the highest priority, allowing these ships to sail
into and sail out of the lock before recreational craft. A complicating factor might be the difference in
allowable mooring forces on the different types of vessel due to emptying or filling the lock.

Figure 3-14 Lock chamber for simultaneous

use by two types of traffic

When two different types of traffic have to be locked in the same chamber it is advised to determine the
size of the chamber by means of simulation.

 Hydraulic Engineering
Faculty of Civil Engineering 46 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

3.4 Vertical dimensions of the lock

In order to deal with the vertical dimensions of the lock the following subjects have to be discussed:
1. Top of Structure and Bottom of Structure
2. Ship’s draught and keel clearance
3. Water levels
Although written in above sequence in this section the subjects are in fact completely cross connected and
both in reading this text and later during design in reality an iterative mind setting or approach will be best for
understanding and future use.

3.4.1 Top of structure (ToS) and Bottom of structure (BoS)

Top of Structure:

Since the lock often is a part of a flood defense system, but always part of a water retaining system, one
of the components along with dikes or levees and other ‘hard’ structures, Top of Structure (ToS) is an
important issue to be considered. Figure 3-15 shows contributors, definitions and some comments, to the
construction height of structures like dikes and breakwaters. Note this figure is the same as in the ‘Dutch’
CT2320 lecture notes, however, translated for convenience sake. For earth structures generally there will
be a significant difference between the construction height and Top of Structure level because the latter
has to be maintained throughout the years in spite of e.g. settlement.

Figure 3-15 Top of Structure – Earth structures

For structures like locks and quays etc., apart from agreed and accepted construction tolerances,
generally construction height is equal to ToS, see Figure 3-16 below. Indeed, contrary to earth structures,
for this type of structures ‘surplus height’ as a contributor to ToS generally could be neglected, although it
could be considered to take into account construction tolerances here.

 Hydraulic Engineering
Faculty of Civil Engineering 47 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Figure 3-16 Top of structure – locks, barriers, quays

Moveable lock elements – the gates:

The gates of the lock are moveable, contrary to e.g. other concrete lock structures or dikes. In closed
position the same ToS requirements apply to top of gate. Generally bottom of gate will be the same level
as top of the sill or top of the lock head floor, when closed; detailing issues related to seals taking care of
water tightness may result in some deviation to the previous. For some gate types, e.g. a lift gate or radial
gate, bottom of gate will have to be determined for the gate in opened position, considering the required
air draught and the design (high) water levels. Obviously, in opened position, top of gate and top of the
gate support structure are ‘practically’ related to bottom of gate in the opened position, not to flood
defense requirements.

Gates being frequently moved, and ships frequently manoeuvring in their vicinity, introduces the risk of
gate failure resulting in failure of the flood defense or water retaining system. Too increase safety of the
whole system, often a double gate will be provided, or at least provisions will be made for stop logs. The
double gate could be positioned in the same (outer) lock head, or the gate in the other (inner) lock head
could be considered as the back-up gate. In the latter situation the whole structure between inner and
outer gate has to at the same required ToS level.

There are other reasons to provide the lock with more than one gate in the lock heads, or to provide
intermediate gates, because a good intermediate gate arrangement has the following advantages:
• It takes less time to handle a small vessel
• Water will be saved in periods of draught
• The amount of salt water penetrating into fresh water channels is reduced
• It permits gates to be repaired or removed without shutting the lock down for traffic
See following Chapters and the Appendix for further elaboration.

In case the head difference between the upper head and the lower head exceeds 2 meters (A>2), see
Figure 3-17, it is usually considered better, from a construction costs point of view, to design the upper
lock head higher than the lock chamber and the lower lock head. In case the difference water between the
lower canal level (LCL) and the lower lock chamber level is larger than 2 meters (A>2) it is usually better

 Hydraulic Engineering
Faculty of Civil Engineering 48 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

to design a sill (drempel) at the lower lock head. In case the level difference is smaller than 2 meters
(A<meters) usually the most economic way is to construct the lock in the same plane by constructing the
lower lock head higher or by removing the sill.

For navigation lock in a conventional waterway the height of the chamber wall and the gate (4) is 1 meter
above the highest water level (HLL) in the chamber. The gates positioned at (3) and (5), see Figure 3-17,
have the top of the gate between 0.5 to 1 meter above Highest Canal Level (HCL), pending on the waves
to be expected. For reasons of practicality these doors (3&5) are at times designed and constructed to
match the height of doors at position (2) and (4). In this way only 2 different sizes of spare gates are
needed.

Figure 3-17 Characteristic water levels combined with direction mitre gates;
Construction level differences

Bottom of Structure:

In the above it was described how to determine ToS and it was mentioned that bottom of gate is generally
at the same level as top of sill or top of lock head floor. To determine top of sill or top of floor, the ship’s
draught, the keel clearance and water levels have to be taken into consideration, see section 3.4.2. Only
after knowing top of floor a start can be made to derive Bottom of Structure (BoS). If strength and stiffness
is the only requirement to the sill or floor structural calculations will result in the required floor thickness
and finally simply subtracting this thickness from top of floor will lead to BoS. Other functional
requirements, e.g. an emptying and filling system using culverts in the floor of the lock, may increase the
thickness of the floor, see Chapter 5.

Be aware that the deeper the level of BoS, generally, the higher the cost of lock construction.

3.4.2 Ship's draught and keel clearance

There is a minimum water depth required to enable ships to manoeuvre through the approach areas and
the lock itself. Table 3-2 showed the water depth to be provided for inland navigation vessels. For
maritime vessels and recreational craft a very approximate rule of thumb for the water depth Dwater would
be:
Dwater = D x 1.15 + 0.5 m where: D = ship’s loaded draught

 Hydraulic Engineering
Faculty of Civil Engineering 49 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

When designing for recreational (sailing) craft the above formula reduces to: Dwater = D + 0.5 m.

This because the shape of the keel of recreational craft, in cross section and longitudinal section is much
more curved than inland or maritime vessels, therefore, leaves more water area available for return lows
that would result in sinkage of the ship.

Figure 3-18 Minimum waterdepth, draught and keel clearance

The rule of thumb for the water depth is quite an approximation; often it will be necessary to produce a
better estimate for the required water depth. The minimum depth in a lock chamber is determined by the
following components, see Figure 3-18 as well:

1. Margin for transitional waves, seiches, etcetera (wave phenomena which can cause fluctuations)
2. Vessel draught
3. Squat (inzinken), which includes sinkage and trim of the vessel. A moving vessel increases its draught
due to return flow phenomena.
4. Hydraulic margin (influenced by the relation between the cross sectional area of the vessel versus the
lock chamber)
5. Safety margin (to compensate for margins in the previous components)

Please refer to e.g. the CT4330 lecture notes for further elaboration of these minimum water depth
components.

3.4.3 Water Levels

Due attention should be given to selection of the water levels in the lock. Extreme high and low water
levels, i.e. water levels occurring at low frequencies, generally are or should be related to the water
retention function of the lock. In design stages where strength and stability of the lock structures is
checked, the extreme water levels are translated in loads used for the Ultimate Limit State (ULS).
Frequently occurring water levels are important for the ship navigation function of the lock; these water
levels determine to a large extend the operational capacity of the lock. From strength and stiffness point of
view, frequent occurrence translates into loads and load combinations of the Serviceability Limit State.
See previous sections for descriptions of water retention function and lock operation (capacity) issues.

In Figure 3-19 typical water levels situation for a lock in coastal area are shown; it is left to the reader to
make a similar schematisation for a lock in an inland waterway.

 Hydraulic Engineering
Faculty of Civil Engineering 50 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Figure 3-19 Characteristic water levels of a lock in coastal areas

Water levels in the outer ports

Looking at a lock in coastal area, attention will be focused on sea levels first because (in general) the
variation in water levels is larger than those of the inland waterway, secondly there is an obvious water
retention aspect, the lock is part of the flood defense system.

Flood defense will require that the lock retains water up to a defined extreme high water level, here
referred to as the Storm Surge Level (SSL) but it could also be described as ‘Design Water Level’ (DWL).
Although it could be allowed that waves, generated by a storm surge, overtop the structure, obviously, the
structural integrity shall and will not be threatened. Often extra height of the structure will be added to
prevent wave overtopping. This will depend on the requirements for the amount of water allowed to
overtop the structure in relation to the quality of the water (salt or fresh water). Both the SSL and the wave
should be related to a probability of exceedance (for instance 1:10.000 for a storm surge).

Generally High High Water Spring (HHWS) and Low Low Water Spring (LLWS) are known all over the
world; often the values can be derived from tide tables for a given location. Using wind speed and fetch
(strijklengte), both wind set-up and wave height could be calculated to determine SSL or DWL, or at least
a fair approximation, and similarly the extreme lower water levels are estimated. Both the wind speed and
fetch depend on the wind direction and could be related to a probability of exceedance (for instance
1:10.000 for a storm surge). For many locations waterlevels and the related probability of exceedance are
known and e.g. presented in a graph, see Figure 3-20.

 Hydraulic Engineering
Faculty of Civil Engineering 51 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Figure 3-20 Water levels and frequency of exceedance for 4

locations along the Dutch coast

When designing the lower lock head the highest and lowest water level in the canal (HCL, LCL) are the
dominating factors.

In the absence of any wave data a minimum wave height of 1 meter should be taken into account for
coastal areas and 0.5 meter for inland navigation areas for the outer lock head. Note that incident waves
may be reflected, which increases the local wave height and the original Still Water Level (SWL), which
was equal to SSL.

Water levels in the lock chamber

Determining the extreme water levels to be maintained in the lock chamber, a balance has to be found,
between construction costs and safety, but it may be more important to consider the effect of water levels
on the emptying and filling time, hence on the capacity, of the lock.
Highest Lock Level (HLL) is chosen somewhere in the middle of the water level in the outer approach area
(SSL or DWL) and the water level in the inner approach area (HCL). The lowest Lock Level (LLL) is in
between the lowest water level in the outer approach area and lowest mean water level of the lower canal.
As a rule of thumb a navigation lock is designed to retain a water level that exceeds once a year the
highest lock level and the lowest lock level. Obviously the mentioned frequency depends on the
importance of the waterway and the traffic on the waterway.

 Hydraulic Engineering
Faculty of Civil Engineering 52 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

4. Structural elements of a navigation lock

In this chapter sections on the following subjects have been included:

1. Filling and emptying systems (F/E) 4. Lock chamber

2. Lock gates 5. Seepage cut-off screens
3. Lock head 6. Approach structures and berthing facilities
It might seem strange to begin a chapter on structural elements with a section on F/E systems, but filling
and emptying (F/E) the lock chamber is key to the efficient operation of the navigation lock. In various
ways a number of the structures to be designed and constructed, if not all, are influenced by the F/E
system of the lock.

4.1 Filling and Emptying Systems (F/E)

This section starts with a general description of F/E systems, including some terminology, fluid mechanics
or hydraulics, water saving basins, and selection criteria. In the next two sections two most important
selection criteria for F/E systems, the time required for F/E and hawser forces (tros krachten) will be
discussed quite thoroughly. Finally some specific subjects regarding F/E, viz. water saving basins, salt
water intrusion, ice control, and large water head difference phenomena, will be mentioned and provided
with some references for further study.

4.1.1 Typical F/E alternatives

There are two main systems for filling and emptying the lock, viz. the “through the heads system” and the
“longitudinal culverts system”, the first for low lifts, the other for high lifts. Some other main features and/or
characteristics of these F/E systems are mentioned in Table 4-1.
To distinguish the heads system from the longitudinal culvert system it is best to keep in mind that filling or
emptying through the heads results in a very turbulent water zone near the lock head; the disturbance will
spread through the whole lock chamber, see Figure 4-1. The intended use of a longitudinal culvert system
is to distribute the in or outflow over the whole length of the lock chamber, therefore substantially reducing
water turbulence and avoiding or reducing water level differences over the length of the lock chamber.
Quiet water in the lock chamber is important for the safe mooring of ships being locked.

Through the heads Through the heads / longitudinal culverts Through longitudinal culverts

Lift height H<10 m Lift height 10<H<15 m Lift height H>15 m

Short culverts, stilling chamber; Longitudinal culverts
Gate valves, short
or (possibly with stilling chamber or
culverts
Longitudinal culvert pressure chamber)

“Realtively” fast, simple, cheap <====> Relatively” slow, complicated, expensive

Table 4-1 Main characteristics of F/E systems

Figure 4-1 Turbulent water, due to a water jet, spreading through lock chamber

 Hydraulic Engineering
Faculty of Civil Engineering 53 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

In Figure 4-2, the pencil thin lines on the

Head filling Longitudinal culvert filling
left are transforming into a gate with
culvert
dimensions; adding more thickness or,
gate
extruding the gate and the culvert or
closed conduit on the right develops.

From a hydraulic point of view, head filling,

usually by means of valves in the gates, is
characterized as ‘open channel flow’
(stroming in open waterlopen), whilst the
flow of water through culverts is typically
known as ‘closed conduit flow’ (stroming in
Closed conduit flow in culvert:
Open channel Open Open gesloten waterlopen of leidingen).
• entrance loss
flow trough channel channel In case of flow through a gate, the
• wall friction
gate flow flow
• exit / Carnot loss available potential energy, which is a
Figure 4-2 Hydraulic differences between head and function of √mgh, is transformed in kinetic
2
culvert filling energy, a function of ½mv , without very
significant losses of energy. Regarding
closed conduit flow, the loss of energy, e.g. due to wall friction or due to elbows or bends in the culvert, will
be considerable and the effect on discharge has to be taken into account.

Translation waves not only run through the lock chamber during the filling process, but a negative
translation wave runs through the outer port of the lock and upstream reach as well. Vice versa, a
negative wave travels through the lock chamber when it is being emptied and a positive wave moves
through the downstream approach and further. In the lock chamber, generally with vertical concrete walls,
the wave will be reflected (more than once), whilst in the approach area, due to spread and e.g. bottom
friction the wave energy will usually dissipate quickly. Safe mooring of the vessels and safety of lock and
approach structures will be matter of concern. See section 4.1.3 for further discussion.

Head filling

In these lecture notes ‘head filling’ will be used for systems

where the lock is filled, or emptied near or at the head. Often
the closing element, the gate, has valves that are used to let the
water flow in or out, see Figure 4-3. The valves are positioned
at the bottom of the mitre gate to prevent the water jetting
through the valve from hitting the ships directly at the bow or
stern. The latter generally cannot be prevented if a sill is
constructed at the head. In that case short culvert systems, just
going around the gate and/or a part of the lock head are useful,
see Figure 4-4. Short culverts around or through the head will
be considered as a head filling system.

Head filling generally is used for locks with a lift of up to 6

meter. Using it for larger lifts results in a very turbulent water
near the gate and lock head, due to the large water head
difference, creating a zone where ships should not be moored.
This zone decreases the effective lock chamber length, or vice Figure 4-3 Head filling: gate
versa increases the total length of the lock. Of course the head with valves and breaker plates
filling system can be designed to minimize the turbulent water
zone, for instance by using short culverts and dissipation or stilling chambers (woelkelders), see Figure
4-4 and Figure 4-5. This may increase the use of the system even to a lift or water head of 15 m. In this

 Hydraulic Engineering
Faculty of Civil Engineering 54 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

range the combination of short culvert and stilling chambers has to compete with the longitudinal culvert
system.

Figure 4-4 Head filling: a pipe Figure 4-5 Stilling chamber below the lock gates
culvert around the lock head

If the requirements for F/E time and safe mooring are met, to make a choice between filling through the
gate, or by means of a short culvert system in or around the lock head with or without stilling chamber(s),
it will be best to look at the simplicity or complexity of providing valves in the gate versus culverts, possibly
with stilling chambers, in the lock head and the resulting costs.

A most simple head filling system has to Stilling chambers:

be mentioned: this would be lifting or
opening the whole gate. Nothing extra has The intended use of stilling chambers is to increase the lift
to be designed or constructed to the gate, of head or longitudinal culvert filling systems. Combined
which is good for costs, and the lock would with a longitudinal culvert the stilling chamber is often
be filled or emptied really fast by using the positioned under the lock chamber floor, generally covering
whole gate. The downside to this solution a large part of the horizontal area. In this low level position,
is obvious; all the water in the lock would water pressures are relatively high, therefore the name
be in turmoil and whatever ship being ‘pressure’ chamber is also used.
moored, it would be tossed around in the To enlarge the lift range the stilling chamber has to
lock and there would be a great risk it dissipate as much of turbulent water energy as possible,
would break its hawsers. and has to take care of a distributed inflow of the water in
the lock chamber. For that reason, often flow breaking
elements, such as concrete colums or beams, are
constructed into the stilling chamber.

Very turbulent water, and high water pressure, create the

conditions for water hammer and/or cavitation. Both these
phenomena are very sensitive to the dimensions and shape
of the lock culverts and the stilling chamber in particular. It
is advisable to involve hydraulic experts in the (preliminary)
design and, if possible, to base the design on either existing
stilling chambers or physical model investigation.

Longitudinal culverts:

For lock with lifts larger than 6 meter, going up to 25 meters or even over 25m, longitudinal culvert
systems are used. The principle is to distribute the inflow or discharge of water over (many) more
positions or locations in the lock chamber to avoid water zones too turbulent to moor ships. Many in or
outflow positions along the whole lock chamber length are used, hence the name longitudinal culverts.

 Hydraulic Engineering
Faculty of Civil Engineering 55 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Figure 4-6 ILCS, Marmet Lock (USA)

In-chamber Longitudinal Culvert System (ILCS), Marmet Lock (USA):
The ILCS consists of two parallel rectangular culverts; each culvert has two
sections with manifolds at each side of the culvert. A wall baffle is mounted on
the lock wall opposing these manifolds to dissipate the energy from the
current. The ILCS results in an asymmetric distribution of flow along the length
of the lock chamber.

The ILCS was developed to reduce the construction costs associated with
large concrete gravity walls. Because the longitudinal culverts are located
within the lock chamber, on top of the bottom floor, the lock walls can be
thinner, which results in a less costly wall design. Assuming the culverts have
to be ‘outside’ the navigable cross section, putting the culvert on top of the
bottom has to result in a lower level of the lock chamber floor. Hence, the cost
savings on the walls are reduced by deeper construction of the lock bottom.
Obviously there is the risk that the exits/inlets of the culverts are damaged,
e.g. by debris in the water or fallen anchors. Presumably it will take some extra
precautions to provide sufficient robustness of these exits/inlets; this will
further reduce the costs savings.

Lock dimensions: 360 ft long x 56 ft wide; 24 ft lift (100 m x 17 m; 7.3m)

Figure 4-7 Top view

on two culvert ILCS
Figure 4-8 Lifting a 6 barge push convoy

 Hydraulic Engineering
Faculty of Civil Engineering 56 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

In literature all sorts of similar or more specific names or descriptions for longitudinal culvert systems can
be found. Look for instance in Table 2-1 for the names and descriptions of simple and complicated
longitudinal culvert systems.

Complexity Hydraulic systems PIANC classification

[2009]
1 Simple longitudinal culvert - Wall culvert and side ports system Low lift height:
systems - In chamber culvert system up to 10 m

2 More complex longitudinal - Wall culvert bottom lateral system Intermediate lift height:
culvert systems - Wall culvert bottom longitudinal system between 10 and 15 m
- Longitudinal culverts under the lock floor
3 Very complex longitudinal - Dynamically balanced lock filling system High lift height:
culvert systems • Pressure chamber under the floor above 15 m

Table 4-2 Names and/or descriptions for longitudinal culvert systems PIANC (2009)

Obviously longitudinal culvert systems generally will be substantially longer than ‘short’ ones used for head
filling, thus, the literal length of the culvert could be a distinctive criterion. Remember this: rather than
looking at the length of the culverts used, it is better to look at the position and/or concentration of the
inflow and/or outflow ports in the lock chamber to characterize the F/E system.

The water may enter or leave the lock chamber in horizontal direction along or perpendicular to the lock
axis, vertical or in any angle or orientation that is deemed necessary. The culverts can be positioned in or
adjacent to the lock walls, in, below or above the lock chamber floor. For al these situations the culverts
may be separated or integrated into the structure, see Figure 4-9 trough Figure 4-8.

Figure 4-9 F/E system with culverts in the lock chamber wall; culvert
exits near or on the lock chamber bottom

Separate culvert systems have to constructed in any way in case the intake or discharge points of the
water are at some distance from the lock chamber, see for instance the plan view in Figure 4-10.
Positioning the intake or outlet a bit further away could be necessary due to the effects of the negative
(inlet) or positive (outlet) translation waves in the approach areas of the lock.

Separation or integration of the culverts and either the wall or bottom of the lock chamber depends, on
one hand, on the required size of culvert, on the other hand, on the available area or space in the lock wall
or bottom. The structural integrity of these elements should not be compromised. Figure 4-6 and substory,
see previous page, describe a culvert solution for an ‘out of’ the wall and ‘out of’ the bottom culvert
structure.

 Hydraulic Engineering
Faculty of Civil Engineering 57 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Longitudinal section

Plan / Top view

Figure 4-10 F/E system with external culverts and pressure chamber
in the lock chamber floor

rd
A 3 F/E system:

In the above description an underlying assumption was the use of natural water head. In some situations
management of either the quantity, or the quality of water at the upstream side of the lock, results in
alternative F/E systems. Gates and culverts may do part of the work, but basically pumps are used to
overrule the natural water head or water level difference.

Water saving basins or side ponds

Quite frequently the design and

construction of water saving basins
is required. In Panama the new
locks are 61 m wide and 457 m
long; they have a lift of say 13 to
17 m. Water of lake Gatun, which
is at an higher altitude, gets lost
due to the locking process. In a
lock chamber being filled or
3
emptied, about 400 000 m of
water flows in or out (160 Olympic
swimming pools of 2m depth). It is
important to use the lake’s water Courtesy Panama Canal
sustainable and economically, Authority (2008)
hence, to save as much water Figure 4-11 Water saving basins and large (external)
during locking. culvert systems
Figure 4-11 shows a worked open

 Hydraulic Engineering
Faculty of Civil Engineering 58 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

illustration of the water saving basins and their emptying and filling system. Safety of mooring is not an
issue in the basins but it is in the lock chamber. Thus the lock chamber has more and better distributed
inlet/exit ports than the water basins. Look at the size, shape and position of the culverts between basin
and chamber; it all may be described by the phrase large underground tunneling work. It is not hard to
imagine that the water saving basins and underground culverts significantly increase the cost of a lock
complex (in the order of 2-4 times the costs).

Selection of a F/E system

Given the importance of the F/E system for the design and construction of the whole lock it is best to
carefully select the system; if need be, use a multi criteria analysis or evaluation. Looking at the
description of the F/E systems in the above the following criteria for emptying and filling a lock can be
defined:
1. time/duration: the time required for emptying or filling the lock linearly adds to the lock cycle, thus
to the capacity of the lock, see Chapter 2. In the next section formulas for computation of the F/E
time will be derived. In general F/E has to take as less time as possible, under the condition of the
next criterion on hawser forces (tros krachten);
2. hawser forces of the vessels moored in the lock chamber have to remain well below preset limits,
depending on the traffic type and/or chosen design vessels. Safety for the ship and the lock
(structure) is the background for this criterion; ships should not break loose from their mooring
lines. See section 4.1.3 for calculation of the forces and limits to hawser forces. Do mind that the
water discharging from the lock chamber should not create mooring problems for vessels laying
by or moored in the approach area of the lock;
3. water management usually requires very specific measures determined by the situation at hand
and it will result in special criteria. It would either be not very meaningful, or simply impossible to
provide ‘general’ criteria in these lecture notes;
4. operation and maintenance robustness;
5. costs
6. etc.

The issues of point 1 and 2 in the above will be dealt with in Sections 4.1.2 and 4.1.3 respectively; they set
forth the basic requirements a filling system has to suffice to in a given situation. If more than one
alternative suffices and if budget permits, the issues under point 3 to 6 come into play and a multi criteria
analysis, to find more selection criteria, followed by the multi criteria evaluation, will result in selection of
the F/E system to be further elaborated.

In PIANC report, Innovations in navigation lock design (2009), the M factor method, from China, for selection
of the type of filling system, is introduced to the greater global public. M is defined as follows:
T
M= where T is the time to fill the chamber in [min] and H is the lift height in [m]
H

The values for M and choices for the F/E For Chinese inland waterways:
system on the right have been defined for a
certain set of maximum hawser forces 3.5<M through heads system
2.5<M<3.5 through heads system or simple
used in China.
longitudinal culvert system
2.4<M simple longitudinal culvert system
(Only for conceptual design it may be 1.8<M<2.4 a more complex longitudinal culvert
assumed, there is not a significant system
difference between the maxima in China or M<1.8 a rather/very complex (advanced)
elsewhere). longitudinal culvert system

 Hydraulic Engineering
Faculty of Civil Engineering 59 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

4.1.2 Time required for F/E

Filling or emptying the lock chamber takes time, in fact it is the biggest contributor to the lock cycle time,
both for inland and maritime navigation, see Section 2.2 and Table 2-4 in specific. Therefore the time
required for F/E is of direct influence on the capacity of the lock, however, and vice versa the
requirements set for the lock capacity will determine the time for F/E. Hence, it is necessary to be able to
compute the time a certain F/E system needs to do the job.

Head filling

To analyse the process of filling or emptying the lock chamber and finally determining the total time Ttotal
required for F/E, it is useful to obtain formulas describing, for instance the discharge and the changing
water level. For that purpose several assumptions have to be made:
• Both the upstream and downstream canal are assumed to be infinitely large basins, therefore the
average water level remains the same in both canals
• The lock chamber will be filled or emptied through valves in the gate or using a (short) culvert
system through or around the lock head; see next assumption
• Inertia of the water mass in the lock chamber, in the (short) culvert system or stilling chamber will
be neglected in the following. Friction of the water flow is neglected
• The translation wave resulting from the filling process has a relatively small wave height compared
to the considered water depth

Valves in the gate completely open(ed):

From a fluid mechanics point of view, head filling through valves in the gate could be considered as a
case of ‘submerged’ flow (verdronken uitstroming), see Figure 4-12 and lecture notes ‘Introduction to
Hydraulic Engineering’ CT2320. Formula (1) may be used to determine the discharge Q:

2 3
1

Q = ms ⋅ b ⋅ a ⋅ 2g(h1 − h3 )
2
U 1

2g
where:
ms = discharge coefficient for
2 2
U2 U3 submerged flow (≈0,8)
2g 2g
b = width of the valve(s) [m]
H1 h1
a = height of the valve [m]
H3
(µ = contraction coefficient [-])
h3 In the following:
a 2
µa f = a * b = total valve area [m ]
z = h1 – h3 = water level difference

Q = ms ⋅ f ⋅ 2gz (1)
Figure 4-12 Submerged flow under a gate. Discharge formula

Formula (1) describes a static situation, which disagrees with the ‘dynamic’ reality of a water level
difference z that is changing in time due to filling or emptying the lock chamber. Q and z have to be
considered as time dependent variables. The principle of mass conservation can be used to find a relation
between the discharge Q(t) and the water level change dz. Multiplying the time dependent discharge Q(t)
with sufficiently small time periods dt, mathematically spoken integrating Q(t) over time, has to equal the
volume of water flowing in or out the lock chamber. The same volume is also equal to the horizontal area
A of the lock chamber multiplied with the change in water level dz.

 Hydraulic Engineering
Faculty of Civil Engineering 60 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Use this simple physical reality to write down Formula (2):

dz
Q(t )* dt = − A * dz ⇒ Q(t ) = − A * (2)
dt
Rewriting Formula (1) as Q(t ) = ms f 2 gz (t ) , substitution in (2) and rearranging results in:

dz m f ms f
=− s 2 g dt after integration: z(t ) = − 2g * t + Const. (3)
z(t ) A 2A

First the valves in the gate are closed and there is no water flow, then, at t = 0, the valves are opened and
discharge starts due to the initial water level difference z = ∆H. (Suppose, for the time being, opening or
closing of the valves takes no time at all). Hence, the boundary condition z=∆H at t=0 can be defined and
used to solve the constant in Formula (3):
ms f
For t = 0; z = ∆H ⇒ z =− 2 g * t + ∆H ⇒
2A
2
m f  mf
z =  s  2g * 1 t 2 − s 2g ∆H * t + ∆H (4)
 A  4 A

Substitute (4) in (2) and find after derivation:

ms 2 f 2 g
Q=− t + ms f 2g ∆H (5)
A

The total filling time is easily determined, by using Formula (5) and the boundary condition t = 0; z = ∆H
and Q = 0:

ms 2 f 2 g 2 A ∆H
Q=− t + ms f 2g ∆H = 0 ⇒ Ttotal = (6)
A ms f 2g

With Formulas (5) & (6), and the data below, the
45.00 5.00
graph(s) on the right, for z(t) and Q(t) respectively, have
4.50
been drawn. 40.00
2
A = 1080 [m ] 35.00
4.00

ms = 0.85 [-] 30.00

3.50
2
f = 5 [m ] 3.00
Q [m3/s]

2 25.00
z [m]

g = 9.81 [m /s] 2.50

∆H = 4.5 [m] 20.00
2.00
15.00
1.50
2 A ∆H
Ttotal = = 243 s 10.00
1.00
ms f 2 g 5.00 0.50

0.00 0.00
0 50 100 150 200 250 300
Observe that the discharge Q decreases linearly in time,
Discharge Q
whilst the water level difference z shows a parabolic ==> t [s]
Water level diff erence z
decrease.
Since the water level difference does not change a lot in Figure 4-13 Graph for the decrease in Q
the later stage of the process, z slowly approaches zero, and z over time
it is more or less common practice to open the whole gate before t = Ttotal. The criterion for z being
acceptably low or not is determined by the sudden water movement in the lock chamber or in the
approach area of the lock, hence, by the order of magnitude of ship displacements and hawser forces.

 Hydraulic Engineering
Faculty of Civil Engineering 61 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Including the time required to open the valves in the gate:

In reality gates or valves can not be opened or closed without the passing of some time. The time required
for opening the valves will be denoted by th. In the previous paragraphs the formulas for Q(t) and z(t) have
been derived under the condition of fully opened valves, hence, t ≥ th. Now the development of the
discharge and water level difference for 0 ≤ t ≤ th, will be determined to find an expression for Ttotal again.

Q and z during the period 0< t ≤ th:

Assume a linear lifting process of the valves. Therefore a term t/th will be added to Formula (1) for the
discharge, hence:
t
Q(t ) = ms f 2 gz (7)
th
Equating the Formulas (2) and (7), followed by some rearrangement makes the left and right hand side of
equation (8) ready for integration:

t dz ms f 2g dz
ms f
th
2 g z = −A
dt
⇒ ∫ Ath
t dt = −∫
z
(8)

Use the start or boundary condition t = 0, then the water level difference z = ∆H, and find:

ms f 2g 1 2 ms f 2 g 2
2
t = −2 z + 2 ∆H ⇒ z = ∆H − t ( 9)
Ath 4 Ath
Take the square of z to find z for t ≤ th:
2
m f m f 
z = ∆H − s 2 g∆H 1 t 2 +  s  g 1 t 4 (10)
Ath 2 8
 Ath 

Substitution of (9) in (7) results in Q for t ≤ th:

t  m f 2g 2  ms f 2 g∆H m2 f 2 g
Q = ms f 2 g  ∆H − s t  ⇒ Q= t − s 2 t3 (11)
th  4 Ath  th 2 Ath
 

In case filling or emptying continues after t = th there will be a remaining water level difference ∆Hrest and it
will take some time to complete the process with the valve(s) completely opened. The total time will be
equal to the addition of th to twide open. Formula (6) can be used to compute twide open if the remaining water
level difference ∆Hrest is known. Use Formula (10) for t = th to find the remaining water level difference
∆Hrest.
2
mf m f 
Formula (10) for t = th : z(th ) = ∆H rest = ∆H − s 2g∆H 1 th2 +  s  g 1 th4
Ath 2 8
 Ath 
2
 m f  2 A ∆H rest 2A
∆H rest =  ∆H − s 2g 1 t 2
subst. in (6) : twide open = = ∆H rest
Ath 4 h
  ms f 2g ms f 2g

2A  ms f  2 A ∆H
twide open =  ∆H − 2g 1 th2  = − 1 th ⇒
ms f 2g  Ath 4 2
 ms f 2g

 Hydraulic Engineering
Faculty of Civil Engineering 62 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

2 A ∆H 1
Ttotal = th + twide open = + th (12)
ms f 2g 2

Do note the effect of more accurately including the lifting period th. The total time Ttotal, increases with
(only) 50% of the lifting period th compared to the period with an immediately wide open valve.

In the above it has been assumed that the lock chamber was not completely filled or emptied when the
process of lifting the valve came to a stop. In one of the following paragraphs Ttotal will be determined
when filling and emptying does come to a stop before the valve is completely opened, hence Ttotal < th.

Some observations with regard to the relation between valve opening time th and discharge Q(t)

Consider the following situations:

1. Relatively fast opening of the valves; i.e. when

the valves are fully opened there is still enough
water level difference to result in an increasing
discharge. See upper graph.
2. Relatively slow opening of the valves; i.e. when
the valves are finally fully opened the discharge
is already decreasing because the remaining
water level difference is small. See lower
graph.

The two illustrative graphs in Figure 4-14 are drawn

with the same data for A, ms, f, g and ∆H as in the
above.
Note:
The discharge curves should have been cut off
beyond th larger than 90 and 180 s
respectively, since Formula (4) is only valid for
t≤th.

Maximum discharge occurring during the lifting

process:
In situation 1 the maximum discharge Qmax is easily
found by substituting th into (11), which results in:

ms f 2 g Figure 4-14 Graphs showing discharge

Qmax = ms f 2g ∆H − th (13) Q for different lifting periods
2A

In situation 2 the maximum discharge has to be determined by time derivation of Formula (11) and finding
the time t for which the derivative equals 0. At that moment, for that value of t, Q=Qmax. Substitution of the
obtained expression for t into (11) results in Qmax for situation 2.

dQ ms f 2 g ∆H m2 f 2 g 2 A 2 g ∆H
= − 3 s 2 t2 = 0 ⇒ t2 = th
dt th 2 Ath 3ms fg

4 A ∆H
substitute t= th in (11) and find :
3ms f 2 g

 Hydraulic Engineering
Faculty of Civil Engineering 63 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

3
m f 2 g ∆H 4 A ∆H m 2 f 2 g  4 A ∆H 
Qmax = s . th − s 2  th  (14)
th 3ms f 2 g 2 Ath  3ms f 2 g 
 

There is a third situation to be considered besides the two already mentioned in the above:

3. (Very) Slow opening of the valves;

discharge starts and stops before the valves
are fully opened. See the graph at the right.

Notes:
- The negative values for Q have no physical
meaning. The water stops flowing at t≈ 540
s before t=th=600 s.
- The maximum discharge for this situation
can be computed using Formula (14).

For situation 3 the length of the filling or

emptying period, Ttotal, can be determined using
Formula (11) and realizing the discharge Q is Figure 4-15 Graph showing discharge for
equal to zero when the lock chamber has been very slow lifting
filled or emptied:

ms f 2 g ∆H m f 2g
Q= t − s 2 t3 = 0 0 < Ttotal ≤ th (11)
th 2 Ath

ms f 2 g ∆H m f 2g 4 A ∆H th
t = s 2 t3 ⇒ t= (14)
th 2 Ath ms f 2 g

4*1080 4.5 *600

Situation 3 : t = Ttotal = = 540.4 [ s ]
0.85*5 2*9.81

Alternatively, for situation 3, it is possible to find Ttotal using the condition that the volume of water V being
discharged over this period is equal to A∆H.
Ttotal

t =0
∫ Q(t ) dt = V t = 0, V = 0 t = Ttotal , V = A∆H

The boundary condition for t=0 does not result in extra information (Why?); the one for t=Ttotal does. For
convenience sake notation t will be written instead of Ttotal..

t
 ms f 2 g ∆H ms2 f 2 g 3  mf 2 g ∆H 1 2 ms2 f 2 g 1 4
∫0 Q (t ) dt = ∫

th
t −
2 Ath2
t dt

⇒
th 2
t −
2 Ath2 4
t = A∆H
 
 8 A ∆H t h  2  16 A2 ∆H th2 
Finally this results in : t4 −   t +  =0 use the
 m f 2g 2 2
 s   ms f 2 g 

4 A ∆H t h 4 A ∆H th
ABC - formula to find : t2 = ⇒ Ttotal = (15)
ms f 2 g ms f 2 g

 Hydraulic Engineering
Faculty of Civil Engineering 64 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Notes:
- For the situations 1&2, as well as for situation 3, it is possible to use the integration below to find the
volume of discharged water for any t as long as t ≤ th , for situation 1&2, respectively as long
as t ≤ Ttotal for situation 3.
t
 m f 2 g ∆H m f 2g 
∫0 Q (t ) dt = V (t ) ⇒ V (t ) = ∫  s
 th
t − s 2 t 3  dt
2 At h 
 
- It is important to observe that the slower the valves in the gate are lifted the smaller the resulting
maximum discharge, see the graphs in the above.

The development of the discharge for 0 < t < Ttotal

Figure 4-16 Graphs showing discharge during the whole lifting period
Note:
In section 4.1.3 it will be shown that hawser forces vary in agreement with the derivative (afgeleide) of Q
with respect to time.
- Look at the graphs in Figure 4-16 for the discharge Q, and observe that going from the first branch
(t<th) to the second branch (t>th) of the graph, there is not a smooth transition in Q for t=th:
o In the right-hand graph for th=180 s, the tangents (raaklijnen) on either side of th do not have
the same direction. Mathematically spoken, the derivative on the left of th has a different value
than the derivative on the right of th.
o In the left-hand graph for th=90 s, the derivative of Q not only has a different (absolute) value
but changes sign as well from plus to minus. From a mathematical point of view, when
increase changes into decrease, the derivative changes sign from + to -.
- Look at the graph in Figure 4-15 and note that, when comparing the three graphs, this one has the
smoothest development of discharge Q and of its derivative.
Manipulating the lifting of the gate(s), selecting a smaller or larger th, in other words, faster or slower
opening, is an effective way to influence the hawser forces in the lock.

The development of the water level difference z. 5,0

for 0 < t < Ttotal 4,5 th = 180 s

4,0 Ttotal = 333 s
3,5
Figure 4-17 shows the variation of water level 3,0
z [m]

difference z in time. The Formulas (10) and (4) and 2,5

2,0
previously mentioned data for A, ms, f, g and ∆H, 1,5
have been used to produce the graph. 1,0
The change in z is small in the later stage of the 0,5
0,0
process. The whole gate could be opened, provided 0 100 200 300 400
that the resulting longitudinal wave is small enough ==> t

with regard to ship movements and hawser forces. Figure 4-17 Variation of z in time

 Hydraulic Engineering
Faculty of Civil Engineering 65 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Longitudinal culvert filling

In contrast with the head filling system, described in the previous section, energy losses in the culvert
system have to be considered, viz.:
1. energy loss due to:
• wall friction in the culverts
• local energy dissipation while entering and exiting the culvert
• losses in bends and at bifurcation points (splitsingspunten),
• etcetera (trash racks, valve shafts, transitions in culvert shapes and/or size).
2. Besides this, rectangular shaped conduits perform worse than smooth and/or rounded (pipe)
conduits having the same cross sectional area. The rectangular shape increases the order of
magnitude of several of the losses previously mentioned under point 1.

Looking at all the loss phenomena involved it seems hard to imagine that similar formulas for the
calculation of Q, development and maximum, and the F/E time Ttotal can be used culvert filling, as the
ones for heaf filling, see e.g. Formula (11) and (5) for Q, and Formula (12) for Ttotal. However, it would be
very convenient to use similar expressions and be able to compare the two types of F/E systems. The
solution is to use the discharge coefficient ms as the dust bin or garbage can (vuilnisbak), and translate all
those extra losses into a lower value for ms. Before determining the ms for a culvert system some
comparisons will be made between “through the heads” and “through longitudinal culverts” systems.

Figure 4-18shows the differences between head filling and culvert filling of the lock chamber with regard to
the discharge Q and the period required for filing Ttotal. The Q(t) line for head filling was already shown
before in Figure 4-16, three culvert alternatives are added to the graph. The alternative culverts have a
shape or configuration similar to the one drawn in Figure 4-21. The single culvert 100% has a cross
sectional area just as big as the area used for the openings in the gate in the head filling alternative. A
single wall culvert would fill the lock chamber from one side, which has big disadvantages regarding
turbulence of water in the lock. To fill the lock chamber from both sides, two wall culverts have to be
constructed. A 50% discharge on either side requires half the cross sectional area, which means that the
width and height dimensions have to be scaled down with √2 for the two smaller culverts.

Figure 4-18 Head filling versus culvert filling with regard to discharge Q and period Ttotal

The graphs in Figure 4-18 were made for a lock chamber servicing CEMT class IV or class Va vessels;
inland waterway navigation. With 5½ to 8 minutes, the systems fill or empty fast enough from a lock
operation point of view, see Table 1-4. Hawser forces need to be checked for this situation, see section
4.1.3 on hawser forces.
In Figure 4-19 the discharge curve for a bigger lock and a higher lift are presented. (The horizontal area of
2
the chamber is 7500 instead of 1080 m ; the lift 20 m i.o. 4.5 m). The lock should service a push convoy,
consisting of 3*2 barges and a 40 m long push barge, which is beyond the largest CEMT class. In spite of

 Hydraulic Engineering
Faculty of Civil Engineering 66 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

the (much) larger openings or cross sections of the culverts the time required for F/E increases
substantially to about 12 or 18 minutes. For new lock construction the latter would have to be considered
as too long. The discharge curves are smoother because the valves are opened slowly, in 10 minutes.
Hawser forces need to be checked for this situation as well, however, it is easy to predict that head filling
will fail the test, because in the worst case the water is jetting through the valves with a velocity of about 8
m/s (simply divide Qmax by f). Head filling using short culverts around or through the head will do better
provided quite a large stilling chamber is being used. In case of longitudinal culverts, the distributed inflow
of water should prevent potential hawser problems.

Figure 4-19 Discharge curves for head and culvert filling, for a large lift lock (∆H=20)

Remember that, for the same water head ∆H and the same cross sectional area f, the discharge Q will
develop slower and the maximum discharge will always be lower for a culvert system than for a head
filling system. Longitudinal culverts may be designed and constructed with a larger cross sectional area to
obtain the same capacity. There is quite a big cost difference between filling through the gate valves or
through longitudinal culverts. Construction of even larger culverts has less bearing on costs than the
switch from head to culvert filling.

And again there is another hydraulic 30

phenomenon to be considered: overtravel of
===>

the water in the culvert. At the end of the F/E head filling

period there will be no water head ∆H left; 25 culvert

Q [m3/s]

culvert - overtravel
based on the formulas the water in the culverts
should stop flowing, but it won’t. The kinetic 20
energy of the mass of water in the culvert will
act as the new driving force. The water will
15
overtravel up to the moment that driving force
will be compensated by the newly created
water level difference. This water head ‘on the 10

other side’ will reverse the direction of flow and

the same will happen again. When the water 5
head has been reduced to nothing, the water
will overtravel again, obviously in the other
direction. An oscillatory motion in the culvert, 0
0 100 200 300 400 500 600 700
lock chamber and part of the outer pool will be
the result, see Figure 4-20. Finally, the -5 t [s] ===>
oscillations will come to an end due to friction
and other energy losses. It will take too much Figure 4-20 Overtravel in culvert

 Hydraulic Engineering
Faculty of Civil Engineering 67 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

time to wait for the overtravel effect to dampen out. A possibility to shorten the process is to close the
valve(s) in the culvert. The closing of the valve from a certain moment close to the end of the F/E period
would have to be introduced into the formulas, as was the opening of the valve. It will definitely change the
discharge graph negatively. Discharge will start to decrease earlier and faster, and the time required for
F/E will increase. There are some other downsides to be taken care of when a valve in a culvert is being
closed, viz. cavitation and water hammer, see section 4.1.4.

Determination of discharge coefficient ms for a longitudinal culvert system:

To explain how the discharge coefficient for a longitudinal culvert has to be determined it is best to use an
example. A more or less realistic example of a longitudinal culvert, constructed into the wall of a lock, is
shown in Figure 4-21. Although the sketch is schematic, it does include the main energy loss contributors
like entrance and exit losses, bends, and bifurcations and obviously wall friction. The flow of water is as
follows:
1. from the upstream or upper pool more or less horizontal, depending on the shape of the entrance,
but in any way sideways out of the waterway into the culvert entrance.
0
2. through a bend; here vertically down in 90 . The bend may be less sharp and the culvert axis may
be inclined instead vertical, all depending on the actual situation. The vertical distance bridged will
be in the order of the difference in bottom level between the up and downstream water pool
respectively.
0
3. through a bend, here 90 again, and further in horizontal direction into the (bottom) culvert that
runs along the biggest part of the lock chamber and close to the lock bottom. The water passes
the shaft with the lifted upstream valve, which opened up the culvert for the flow of water.
4. every so much meters a part of the water takes an exit turn, here in the horizontal plane, and
flows into the lock chamber. The exit ports are close to the lock chamber bottom.
5. the closed downstream valve prevents the water from flowing through the remaining culvert into
the lower pool.

uppe r poo l

o pe n ed lo w e r p o o l
va lv e cl os ed
m a in c ulv e rt v alv e
w a ll/ bo tt om

lo c k
cha
m be
r

Figure 4-21 3D sketch of a longitudinal culvert system

With a bit of imagination the longitudinal culvert can be considered as a part of a pipeline network; the lock
chamber is the other part of the network. Regarding the culvert formulas based on closed conduit flow
theory will be used, however, for flow in the lock chamber open channel flow formulations would be more
appropriate. In spite of this, the lock chamber will be treated as if it was a closed conduit, a special type of
course, to be able to use formulas, engineering experience and/or results for closed conduit flow and the
pipeline systems world.

Below, for convenience sake, once more the basic formula to describe the discharge Q, depending on a
water level difference (z or ∆H) and discharge coefficient ms:

Q = ms ⋅ f ⋅ 2gz or ms ⋅ f ⋅ 2g∆H (1)

 Hydraulic Engineering
Faculty of Civil Engineering 68 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

A formula commonly in use to calculate losses in pipeline systems is the following:

 L  U2
∆Htot =  λ + ∑ξi  (16)
 Deq i  2 g
 
Where: ∆Htot = total energy loss [m]
 L  U 2 ∆Hf = energy loss due to wall [m]
∆H f =  λ
 D  2 g
(17) friction
 eq 

is the part of the losses due to wall friction, ∆Hi = any other energy loss [m]
and: λ = friction coefficient [-]
  U2 L = length of the culvert [m]
∆H i =  ∑ ξi  (18)
 i  2g Deq = equivalent diameter for the [m]
rectangular profile
accounts for all the other losses, such as ξi = loss/ dissipation coefficient i [-]
2
entrance, exit losses, elbows or bends, g = gravitational acceleration [m/s ]
bifurcations (splitsingspunt), etc.. U = flow velocity [m/s]

Rewrite Formula (16) to make it look like Formula (1) and introduce the cross sectional culvert area f in
order to replace U by Q; then find an expression for discharge coefficient ms for closed conduit flow:
 L  U2 2 g ∆Htot 
∆Htot =  λ + ∑ ξi  ⇒ U2 = 
   L  
 Deq i  2 g + ∑ ξi 
 λ   ⇒
 Deq i  
Q = fU 

f2 f
Q2 = 2 g ∆Htot ⇒ Q= 2 g ∆Htot ⇒
 L   L 
λ + ∑ ξi   λ + ∑ ξi 
 D  
 eq i   Deq i 
− 12
 L 
Combine with Formula (1) and find : ms =  λ + ∑ ξi  (19)
 D 
 eq i 
Having found a formula linking ms to all sorts of losses that occur in close conduit flow the, remaining task,
in following paragraphs, is to find or determine the values to be used for the appropriate loss coefficients.
The work may be laboriously, depending on the number of losses to be taken into account, but is not too
complicated.
The previous statement, ‘not too complicated but laboriously’, is correct from a more general point of view,
even for a single pipeline, however, a pipeline network is subject of investigation here. Looking a bit more
detailed, a first complication is the fact that the discharge in each pipeline, or part of the pipeline of the
network, is different, see Figure 4-35. There is a main pipeline, the main longitudinal culvert, and then
multiple exits into the lock chamber, which are schematised as smaller branching off pipelines. The whole
discharge Q flows through the first or main branch(es), then at every exit branch or chamber outlet, Q is
getting smaller and smaller in the main branch. Even without the losses it needs an iteration to find Q in
every part of the system. On top of it friction and other losses depend on Q, vice versa Q depends on
those losses; an iterative process involving more than 2 variables is the result.

It is practical to start with determination of the losses ξi, depending on a varying Q. After that step the
losses in the individual pipes, and possibly derived individual discharge coefficients ms,i, have to combined
to an overall discharge coefficient for the whole pipe network. However, first some remarks on the cross
sectional shape.

General reference is made to the lecture notes of the course CT2140 or CT2320 or any other reference
work in the field of hydraulic engineering or fluid mechanics. Look specifically for the sections on closed
conduit flow in pipelines,

 Hydraulic Engineering
Faculty of Civil Engineering 69 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Circular versus rectangular cross section; steel pipe versus concrete culvert:

A culvert system will be used for larger

volumes of water. Generally the larger the
volume, the larger the required cross
sectional area of the culvert will be; the
dimensions and shape will be beyond what
is readily available as standard prefab
construction material, whether in steel or
concrete. The largest or main parts of the
culvert system will be constructed in-situ.
Since the fast majority of lock walls is
constructed using concrete, it is safe to
assume that the larger parts of the culvert
will be rectangular shaped.
Smaller parts, like nozzles (spuitstukken) Figure 4-22 Nozzles used as culvert exits, lock
may be constructed from standard pipe Sülfeld - Germany
material. Look e.g. at the lock near Sülfeld,
Germany, see Figure 4-22; pipes/nozzles are being used in a specially designed configuration to break the
water jets and distribute the inflow of water into the lock chamber as much as possible.

In following paragraphs, losses will be discussed for closed conduit flow in pipeline networks. Figure 4-23
shows secondary flows in a rectangular and a triangular shaped cross section, which typically do not occur
in circular shaped cross sections. Travelling along the wall of a circular pipe, a water particle would not
encounter discontinuities, whilst the perimeters of
e.g. rectangular and triangular shaped pipes
contain pronounced discontinuities in every corner.
The energy consumed by the resulting secondary
flows or eddies reduces the energy available for
the primary flow parallel to the axis of the pipe. The
available area, and the width-height ratio of the
cross section, are factors determining the number
and the development into smaller or larger eddies.

The wetted part of the perimeter P and area A of a

cross section of arbitrary shape are related to each
Figure 4-23 Losses due to the shape of the
other by the hydraulic radius R. Formulas (20) and
cross section (Idelchik)
(21) show how to compute R for a circular and a
rectangular cross section.
For circular cross sections :
Traditionally friction formulas have been developed A 1 πD 1 2
R= = 4 = 4D (20)
for pipes, having circular cross sections; far less P πD
for other types of cross sections. Looking e.g. at For rectangular cross sections :
the Jain formula, calculating the wall friction (loss),
A b.h
two parameters, the pipe diameter D and the R= = (21)
Reynolds number Re, take the shape of the pipe P 2(b + h)
into account implicitly. On second thought, it is the
pipe diameter alone bringing the shape into the
2.b.h
equation, because the Reynolds number includes Deq = 4 R = (22)
D. Substituting the diameter D with 4R does not (b + h)
change the results for pipes, however, provides the
opportunity to introduce a range of differently shaped cross sections by means of four times the
appropriate hydraulic radius, see the development of Formulas (20) trough (22).

 Hydraulic Engineering
Faculty of Civil Engineering 70 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Wall friction – loss coefficient ξf :

Colebrook (1939), combining the work of e.g. Chézy, Von Karman, Darcy and Weisbach, proposed one
formula for calculation of the friction coefficient λ for both smooth and rough pipes of circular cross
sections:
1  k / D 2.51 
= 2log10  s +  (23)
λ  3.7 Re λ 

The Colebrook formula is an implicit equation for λ, which requires quite some iteration to find a result.
Alternatively the Moody diagram could be used to find the friction factor, see Figure 4-24.

Figure 4-24 Moody diagram

The drawback of iteration using the Colebrook formula, or reading the Moody diagram numerous times,
was circumvented by Jain (1976), who suggested the following explicit equation for the friction factor:

1  k / D 5.74  0.25
= −2log10  s + 0.9  or f = 2
(24)
f  3.7 Re    ks / D 5.74  
log10  3.7 + Re0.9  
  
ks
for : 10−6 ≤ ≤ 10−2 and 5000 ≤ Re ≤ 108
D
The Jain formula saves quite some computational time, especially when a spreadsheet will be used to
calculate through all the individual pipelines and determine the overall ms of the pipeline network as well.

2
Note: Don’t get confused in the following where the symbol f is used for either the area [m ] of the valve,
or for the friction coefficient [-].

Next, the information to determine loss coefficients for entrance, bend elbow, bifurcation, end bend, and
exit losses will be presented and discussed.

 Hydraulic Engineering
Faculty of Civil Engineering 71 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Intermezzo: Idelchik’s handbook of hydraulic resistance

The russian researcher I.E. Idelchik and fellow researchers
did extensive research into loss coefficients for pipes, single
pipes, in networks, and as flowpipe in machinery. Losses for
gas and water flow, for circular and rectangular shapes, and
a lot of other things water can flow through were included.
Idelchik gathered all the results together and in the 1960’ies
the work was first translated into French, later it became Figure 4-25 Illustration of flow and
available in English. Today it is still being reprinted. eddies in a bend or elbow

The handbook of hydraulic resistance contains many

drawings explaining how water flows through pipes, bends
elbows, distributors, trash racks, etc.. New editions still use
the illustrations made years ago, presumably because they
provide such clarity to the actual behaviour of flow in pipes,
closed conduit flow.
Figure 4-26 Distributor of constant
Test results are presented in graphs and tables. For many cross section and bifurcations
situations, formulas to calculate the loss coefficient ξ are
included. For those cases the formula has not been
presented yet, nowadays availability of (e.g.) spreadsheets
make it easy to find the equation describing a line or curve in
a graph. Since, a set of tabular values can be plotted in a
graph, it is also possible to find a formula to compute the loss Figure 4-27 Type of pipe and wall
coefficients based on the data presented in tables. friction

Figure 4-28 Loss coefficients for reunion of flows (lock emptying)

λ, f = friction coefficient [-] Rc = hydraulic radius - circular profile [m]

ks = equivalent sand roughness [m] Rr = hydraulic radius - rectangular profile [m]
ν= kinematic viscosity
2
[m /s] b= width of the culvert [m]
Re = Reynolds number [-] a,h = height of the culvert [m]
2 3
A, F = cross sectional area of flow [m ] Qp = supply discharge [m /s]
3
P= wetted perimeter [m] Ql = lateral discharge – branching off [m /s]
3
D= diameter of the pipe [m] Qr = rectilinear discharge – continues [m /s]
Table 4-3 Symbols, nomenclature

 Hydraulic Engineering
Faculty of Civil Engineering 72 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Entrance loss – loss coefficient ξentrance :

Entering the conduit or culvert, the water flow generally separates

from the wall and an eddy zone develops, see Figure 4-29. Due to
the flow contraction µ the cross sectional area A reduces, which
has to result in an acceleration of the flow at that point.
Immediately after the contraction the available section is larger
and the flow will decelerate, which results in energy loss. The loss
coefficient is calculated as follows:
2
1 
ξentrance =  − 1 (25)
µ 
Numerous tests showed that the contraction coefficient µ for a Figure 4-29 Contraction of the
sudden change of the cross section shall be taken equal to 0.85. flow at the entrance

Bend or elbow loss – loss coefficient ξbend :

Figure 4-30 shows the track a

water particle follows when
travelling through an elbow or
bend. Observe:
• the track is quite curly and far
from an averagely straight
flow line
• eddies develop on either side
of the elbow or bend The
larger the angle α between the
pipes, the larger the eddies Figure 4-30 Bend or elbow loss
that will develop.
• whether the pipe is circular or rectangular, it is split in half and in each part a whirl spins around (in
opposite direction).
Obviously losses will occur in this turbulent flow; for some angles α and for circular pipes, the loss
coefficients are shown in Figure 4-30.

In the culvert considered, the angle of the two pipe may not be the only
property that changes, but the size may change as well; look for
instance at the connection main culvert to exit port in Figure 4-21. For
situations where only the width changes, from bo to b1 as shown at the
top of Figure 4-31, the loss coefficient is calculated as follows:
  a 
−0.09  −0.4 −0.727*ln  o  
a   b    bo  
ξ = 1.15  o  * 1  (26)
 bo   b0 
Looking at the exit ports, not only the width b changes but the height a
changes as well. In that case, it is probably better to use Formula (23)
and consider the loss as a bifurcation loss instead of a bend loss.

The niche in the bottom illustration of Figure 4-31, is a sort of dead-end

street and adds to the loss in the bend. This would be taken into
o
account by a multiplication factor 1.2 over ξ. A 90 bend with a niche Figure 4-31 End bend (top),
would result in a total loss coefficient of 1.44. niche (bottom)

 Hydraulic Engineering
Faculty of Civil Engineering 73 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Bifurcation loss – loss coefficient ξbif :

Remain
At every exit pipe from the main culvert, at every bifurcation, there
Supply
is a supply flow, a lateral flow branching off, and there is the
remaining flow in the main culvert that continues, see Figure 4-32.
As expected there is contraction with related losses in the exit or
lateral pipe, however, the same occurs behind the bifurcation in
the on-going pipe or main culvert. Eddy areas, reducing the Lateral /
effective cross section for the flows past the bifurcation, develop. Branch-off
The loss coefficients are calculated as follows:
Figure 4-32 Bifurcation, flow
  Q  lines and eddies
Q   −1−0.25*ln  l  
 l 
 Qp  F  

 Qp  
 
ξ pl = ζ pl = 0.85e  
* l  (27)
 Fp 
3 2
 Qr   Qr   Qr 
ξ pr = ζ pr =  -0.7*  -0.3*  +0.7 (28)
 Qp   Qp   Qp 

Fp = cross sectional area supply pipe

Fr = cross sectional area on-going pipe; generally Fr=Fp
Fl = cross sectional area lateral pipe
ζpl = loss coefficient lateral or branch-off pipe
ζpr = loss coefficient on-going pipe

Fp Fr
Qprimaire Qremain

Fl

Figure 4-33 Idelchik graph

Qlateral bifurcation loss coefficients

Exit losses – loss coefficient ξexit :

When exiting from the culvert and entering the lock chamber
the flow decelerates because of being distributed over a (much) D2
D1
larger space. The loss coefficient is calculated as follows: U1
2
U2
 U 
ξexit = 1 − 2  since : Q = AU
1 1 = A2U 2
 U1 
2
 A 
ξexit = 1 − 1 
Deceleration
(29) area
 A2 
Figure 4-34 Exit loss
Here:
coefficient
A2 = bexit hexit
bexit = the lock chamber length divided by the number of exits in the wall
hexit = the water depth in the lock chamber.

Since the water depth varies during the F/E process, hexit would be dependent on z(t); this complicates
computation of ξexit and the whole calculation of the overall discharge coefficient ms. Generally A2 is
substantially larger than A1, the cross sectional area of the exit port, and the resulting ξexit will be small.
Therefore it will be justified to use the average water depth in the lock chamber during F/E.

 Hydraulic Engineering
Faculty of Civil Engineering 74 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Example: upper pool

Calculation of the loss in the supply culvert / pipeline up to, but not including, the
opened valve. opened
valve
2
ks = 0.3 [mm] Amain = 5 [m ] P = 9 [m]
-6 2
ν = 1*10 [m /s] b = 2 [m] Rr = 5/9 [m]
2
g = 9.81 [m/s ] h = 2.5 [m] Dequi = 20/9 [m]

Loss Qi A Dequi U Re f ξ
3 2
contributions [m /s] [m ] [m] [m/s] [-] [-] [-]
µentrance = 0,85 [-] 25,0 5 5,0 11111111 0,031 ξentrance
L1 = 2 [m] 25,0 5 2,222 5,0 11111111 0,010282 0,002 ξfriction
Bend 1 = 90 [degrees] 1,2 ξbend
L 2 = 4,5 [m] 25,0 5 2,222 5,0 11111111 0,010282 0,005 ξfriction
Bend 2 = 90 [degrees] 1,2 ξbend
L3= 8 [m] 25,0 5 2,222 5,0 11111111 0,010282 0,009 ξfriction
+ =====
U2 25 2,448 Σξ
∆Hloss = Σξ = 2.448 = 3.12 [m]
2g 2*9.81 Hence, in the supply culvert the loss in water head
Q 25 amounts to about 3 m.
ms = = = 0.64
f 2 g ∆H 5* 2*9.81*3.12 The discharge coefficient ms, computed in two different
− 12 ways, see Formulas (1) and (19), is equal to 0.64.
 L 
= ( 2.448)
− 12
ms =  λ + ∑ξi  = 0.64
 Deq i 
  Note: the biggest loss contributors are the two bends.

Calculating through the pipeline network / the calculation algorithm:

Some basic rules and/or start conditions to be able to calculate through a pipe network, see Figure 4-35:
• The network consists of pipes (e.g. pipe AD or pipe IJ) and nodes (here A through J)
• Each node in the network can only have one pressure
• The pressure change from A to D by a clockwise path, pipe AC then CD, must equal to the
pressure change using the anti-clockwise path, pipe AB then BD
• For this specific pipe network for culvert and lock chamber:
– There are four loops, respectively ACDB, CEFD, EGHF and GIJH
– The water head between the upper pool and the lock chamber, ∆H, is known, or has to be
assumed.
– The pressure in the nodes B, D, F, H and J is assumed to be the same because the average
water level of the lock chamber will be the same in all these nodes. Hence, the pressure drop
over AB, is equal to, for instance, the pressure drop over ACEGIJ, irrespective of the
increasing number of pipe pieces. (AB only one pipe; from A to J 5 pipe pieces)
• Flow into each node or junction must be equal to flow out of the node or junction
• For this specific pipe network for culvert and lock chamber:
– The inflow into the pipe network will be in node A, the outflow out of the network in the nodes
B, D, F, H and J
– The sum of the outflow or discharge in the nodes B, D, F, H and J is equal to the inflow in
node A
• Given a certain flow in the pipes, losses can be calculated for every pipe, hence, the pressure
drop or head loss is known for each pipe.

To start the actual calculation process:

• An initial discharge has to be assigned to each pipe in the network. For instance 20% of Qinflow in
the pipes AB, AD, AF, AH and AJ; 80, 60, 40 and 20% in the pipes AC, CE, EG and GI
respectively; whilst the discharge in pipes BD, DF, FH and HJ is zero.

 Hydraulic Engineering
Faculty of Civil Engineering 75 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

• Calculate the head loss in each pipe and calculate

the summed up head loss going clockwise and Background – Cross-method:
anti-clockwise through the first loop on the left In 1936 Hardy Cross worked around his
side of the network (ACDB). Generally the clock moment distribution method, for structural
and anti-clockwise head loss around the loop analysis of large buildings, into a similar
won’t be the same. method to determine the discharge of
• In case the clockwise head loss is the largest than pipelines in complex water supply networks.
the flow through the involved pipes has to be Until recent decades, it was the most
reduced. Vice versa the flow through the anti-
common method for solving such problems.
clockwise pipes has to be increased. Use e.g. the
Hardy Cross correction factor.
For each loop:
If necessary the discharges in the pipes have to
be adjusted over and over again until the head ΣQ = 0
loss left around the loop is equal to the loss going Σ∆Hloss ,clock = Σ∆H loss ,anti −clock
right around the loop.
• Now move to the next loop (CEFD) in the network Correction factor:
and use the previously computed results for pipe
CD, which is the common pipe of these two loops. Σ∆H loss,clock − Σ∆Hloss,anti −clock
Again check the clock and anti-clockwise head ∆Q =
loss. For this loop there will be an iteration as well;  ∆Hloss ,clock ∆H loss,anti −clock 
2  Σ +Σ 
flows or discharges will have to be adjusted, head
 Qclock Qloss,anti −clock 
losses recalculated and checked again, until the
results left and right around the loop are balanced.
Then for the remaining loops EGHF and GIJH the Q = ms ⋅ f ⋅ 2gz (1)
same has to be done (EF and GH are common
pipes).
• Having calculated through the network from left to right, now the calculations have to be done in
the other direction,(from loop GIJH to loop ACDB). The whole iterative calculation process
continues until head loss differences in every node and loop, for the entire network are negligable.
• Now the water level difference z or water head related to the input discharge Q, is known for the
longitudinal culvert system. Knowing Q, z and f, the cross sectional area of the main culvert, the
discharge coefficient ms can be calculated with Formula (1).
• Compute other combinations of Q, z and ms. It is recommended to calculate Q, z and ms, say for
every 0.5 m of water level difference. A slight variation in the figures for ms may be observed,
which is not strange given all the numerics. Compute and use the average ms; this will yield
sufficiently accurate results.
• With ms, the discharge curve (Formulas (11) (5)) can be drawn and the total filling time (Formula
(12)), can be determined for this longitudinal culvert system.

Clearly the computational process described in the above is best done by means of a spreadsheet or
using some other network analysis software. (In Excel the iteration option and the Solver add-in are very
convenient).

Open Longitudinal culvert Closed

valve Main branch valve

Qinflow, H1 A C E G I Q=0

Outer Culvert Outer

port ∆H exits port

Upper Lower
pool gate B D F H J pool
gate
closed closed
Q=0 Q=0
H2 Qoutflow
Lock chamber
Figure 4-35 Culvert and lock chamber schematized to pipe network

 Hydraulic Engineering
Faculty of Civil Engineering 76 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Notes:

- An alternative for iterative calculation trough a pipe network can be found by comparing the pipe
network with an electricity network. There is an analogy between the voltage (Volt), current (Ampere)
& resistance (Ohm), and the water head (∆H), discharge (Q) & losses or the flow resistance K.
U2 Σξ
∆H = Σξ ⇒ ∆H = U 2
or ∆H = KU 2
≙ V = RI
2g 2g
Observe the square sign above U, which is not there for I, compare the expressions for (total)
resistance, and see the extra term 2/√(K1K2) in the denominator for parallel resistance.

Expressions for resistances

Serial resistance in pipe networks Serial resistance in Serial scheme

electricity networks
U1 = U 2 = U
Vtot = V1 + V2
U or I U or I
∆H tot = ∆H1 + ∆H 2 = K1U12 + K 2U 22 K1 or R1 K2 or R2
I =I
∆H tot ∆H1 or ∆V1 ∆H1 or ∆V1
K tot = = K1 + K 2 Rtot = R1 + R2
U2
Parallel resistance in pipe Parallel resistance in Parallel scheme
networks electricity networks
∆H = K1U12 = K 2U 22 = K totU tot
2
 V =V
⇒
U tot = U1 + U 2  I tot = I1 + I 2 U1 or I1
K1 or R1
U or I U or I
1 1
K tot = Rtot = U2 or I2
1 1 2 1 1
+ + + K2 or R2
K1 K 2 K1 K 2 R1 R2
In a spreadsheet the total resistance for two parallel exit pipes is easily extended to the total
resistance for n exit pipes. Then the parallel exit resistance has to be entered into the formula for a
serial scheme to combine the supply line part of the culvert (up to the valve) with the exit part.

- In terms of hydraulics, the system should be balanced so that the hydraulic loss for the filling process
is governed by the nozzles in the floor and not by the hydraulic losses in inlet, culverts or pressure
chamber. As a rough estimate the cross-sectional area of the pressure chamber should be at least
the same as the sum of the cross-sectional areas of the feeding culverts. As a rule of thumb, the ratio
between the sums of cross-sectional areas for the nozzles and the culverts can be used for an initial
design:

α= ∑
nozzles
Anozzle ∑
culverts
Aculvert

If the parameter α is small (<1), the filling process will be slower but smoother. If it is bigger (>1.5), the
filling will be faster but rougher. The indicative figures for α give some guidance for the ratio between
the cross sectional areas of the nozzles and culvert or the pressure chamber. Later the ratio should be
optimized on the basis of a laboratory or numerical model.

- How to incorporate the closing of the valve at the end of the filling period is a subject left to the reader.
It makes sense to assume a linear closing process; linear with regard tot time.

- In the previous all the explanations and (example) calculations were in fact dedicated to the filling
proces. When the emptying process has to be calculated through, a first obvious difference is that
entrance and exit, and the related losses, switch; pipe lengths and bends are different, no bifurcations
but joints where flows come together. Setting up the calculation (process) for emptying is left to the
(student) readeer and considered good practice to get to an understanding of all the previous.

 Hydraulic Engineering
Faculty of Civil Engineering 77 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

4.1.3 Hawser forces

Wind, waves or currents in the lock chamber or in the approach areas of the navigation lock will set a ship
in motion or result in hawser forces (tros krachten) in case the ship has been tied to the structure by
mooring lines. Uncontrolled motion of the ship will be prohibited, as a consequence, breaking of mooring
lines must be prevented, hence hawser forces must be limited. Limiting the hawser forces imposes quite
some demands on F/E.
In this section hawser forces will be considered assuming the lock has a “through the heads” system for
water leveling. The F/E of a lock is accompanied by (translation) wave and flow phenomena in the lock
chamber and these phenomena are generally more pronounced for a “through the head” than for a
“longitudinal culvert” system. Head F/E results in a rather concentrated disturbance near the head, which
irrevocable spreads throughout the whole lock chamber, or through the outer port. The reason of being of
a longitudinal culvert system, which is much more expensive, is to reduce the order of magnitude of water
turbulence and flow by means of distributed in or outflow in the lock chamber.
Inside the lock chamber the flow in or out at a chamber end (upper head during filling, lower head during
emptying) occurs mainly in the longitudinal direction of the chamber. Generally the water area outside the
lock chamber is less confined and the wave and flow phenomena will spread more and/or will be damped.
Although this section will focus on ships being moored in the lock chamber it should not be forgotten to
check the hawser forces of ships in the lock approach area or at the waiting berths, even though the
problems to be expected are of a lesser order than in the lock chamber.

Longitudinal force components during F/E:

Figure 4-1 shows how turbulent the water mass

may behave in the lock chamber. Even tough
the water movements may be directed in all
directions, most of the resulting forces will be in
the longitudinal direction, parallel to the lock
chamber axis. The forces in longitudinal are
split in a number of components. Not all these
components are relevant for both filling, see
Figure 4-36, and emptying, see Figure 4-37.
Below the ‘bullet-letters’ correspond to the
letters in the figures:

a. Translation waves:
Because of the non-permanent character of
the discharge through the openings, series
of translation waves are generated at these
openings, which propagate in the chamber.
The waves fully reflect against the gates of
both chamber ends, only partial against the
bow and the stern of the vessel.
The translation waves create an oscillating
movement corresponding to the own
frequency of oscillation of the chamber.
(The oscillation period is the time needed to Figure 4-36 Forces in longitudinal direction
travel up and down through the chamber). during filling

b. Difference in impulse over the vessel length; see the frame ‘Analogy’ on the next page:
Concentrated water jets with high flow velocity may occur behind the gate openings. These high flow
velocities decrease in the longitudinal direction of the chamber due to turbulent exchange with the
surrounding water. Besides and under the vessel, the flow pattern changes substantially due to

 Hydraulic Engineering
Faculty of Civil Engineering 78 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Analogy: impuls force on a broad crested weir (lange overlaat) and a ship in a lock chamber during
F/E. Flip the weir over the horizontal axis in the Figure below and recognize the floating
ship. (Reference is made to the courses CT2320 or CT3310).

= 1 2 ρ gh1
F1 = Force on water body in 2

section 1 [N]
= 1 2 ρ gh2
F2 = Force on water body in 2

section 2 [N]
q = Discharge per running = U1.h1
3
meter [m /s/m’] = U 2 .h2
Impuls balance: U1 = ‘Average’ flow velocity in = q / .h1
dU section 1 [m/s]
F = m.a = m. ⇒ F .dt = d (mU )
dt U2 = ‘Average’ flow velocity in = q / .h2
F1 − F2 − F3 = ρ q (u1 − u2 ) section 2 [m/s]

partial blocking of the cross section by the vessel. At the bow of the vessel the concentrated filling flow
is less able to break down rapidly because of the limited surrounding water available, at the stern of
the vessel the flow detaches and only starts spreading and breaking down further on. From the filling
point at the upper head to the lower head, the average discharge decreases because a shorter part of
the chamber behind the preceding cross section has to be filled. The product of ρqu (impuls) is not the
same in every section, and varies from moment to moment. On top of this, the water levels, h1 and h2,
vary in time and from section to section as well. Use of a stilling chamber results in a substantially
lesser concentrated flow near the head, hence less turbulent exchange will occur in the lock chamber,
nonetheless, an impuls force will develop.
When emptying a lock the energy is not broken down in the chamber but outside the lock, thus the
turbulence exchange in the lock is considerably less, which reduces impuls differences.

c. Friction (vessel, lock chamber floor and

wall):
Friction between the water and the
chamber floor, the chamber walls and the
skin of the vessel result in a water level
difference in the longitudinal direction of the
chamber.

d. Jet of water against the bow:

Could be considered as a first order impuls
force; consequently the ‘difference in
impuls over the vessel length’ (bullet b
above) would have to be considered as a
second order impuls force.

e. Difference in water density:

If there are density differences between the
water of the lock approach and the
chamber, internal (density) waves are
generated during F/E. These waves Figure 4-37 Forces in longitudinal direction
propagate at low speed in the chamber and during emptying
reflect against the vessel and the gates.
These waves are accompanied by water level differences in the longitudinal direction of the chamber
as well.

The F/E system should be designed such that a water jet directly hitting the ship is impossible. Of the
other longitudinal force contributors, only the translation waves and the differences in impuls over the
ships length are significantly contributing to hawser forces.

 Hydraulic Engineering
Faculty of Civil Engineering 79 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Analytical method to determine the mooring load

Since translation waves and the differences in impuls over the ships length were identified as significant
contributors to hawser forces, further analysis to develop a formula for calculation is meaningful:
• For impuls differences the formulas are shown in the frame ‘Analogy’ on the previous page. The
difficulty is to estimate the different water levels and flow velocities in two different sections at the
same time. The subject will not be further elaborated in these lecture notes.
• For translation waves a formula will be derived below.

Mooring load due to a translation wave

Assume the water surface in the lock remains a straight plane, on average, in spite of al local turbulences
or a running wave front. However, on one end of the lock the water level is higher than on the other. With
a bit of imagination it is clear that the average water surface rotates due to the translation wave, see
Figure 4-38. The rotation or slope of the water surface is equal to dz/dx. A force, the hawser force, is
required to prevent the ship from sliding down, which would be initiated by the dead weight G of the ship.

The hawser force F has to

be equal to: dz
dz dx
F= G = iG (30) F
dx Q
G, the dead weight is
G
equal to the vessel’s
water displacement. An L
expression has to found
relating dz/dx or i to the Figure 4-38 Average slope dz/dx of the water surface due to a
incoming discharge Q. translation wave

Examine the translation wave, which is between Q and i and start with the translation wave velocity:
c = wave velocity [m/s]
c = g ⋅h (31) g = gravitation constant
2
[m/s ]
h = water depth (average) [m]
The wave velocity is influenced by vessel presence; the vessel blocks the available water area in the cross
section. This is compensated for in the following formula:
2
g ( Av - n) Av = area of the cross section of the lock; bh [m ]
c= (32) n = cross section area of the ship below water level 2
[m ]
b b = width of the lock chamber [m]
The time needed for a translation wave to pass a vessel with a length L can be calculated by:
L
T= (33)
c
Discharge Q into the lock and the change of water level dz in the lock should be equal to each other. Use
this to link Q, dz and the translation wave velocity c. First assume a constant discharge Q, then consider a
constant increase of the discharge in time:
Q dz dQ 1
dz = ⇒ = ⋅ (34)
b⋅c dt dt b ⋅ c
Now Formula (35) for the water surface slope i is derived using the above Formulas (30) through (34):
dz z dt ⋅ T dQ 1
dz
dQ 1
i= = = = ⋅ ⇒ i= ⋅ (35)
dx L L dt bc 2 dt g ( Av − n)
The water slope i is proportional to the change in the discharge Q with regard to time, in other words,
proportional to the time derivative of Q.

 Hydraulic Engineering
Faculty of Civil Engineering 80 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

In section 4.1.2 formulas for the discharge and its time derivative were found. The second derivative with
respect to time yields the maximum dQ/dt. For the period the valves are being lifted, t ≤ th:

ms f 2 g ∆H m2 f 2 g dQ ms f 2 g ∆H m2 f 2 g
Q= t − s 2 t3 (11) = − 3 s 2 t2 (36)
th 2 Ath dt th 2 Ath
d 2 (Q) 3ms2 f 2 g dQ ms f 2 g ∆H
= − t =0 ⇒ t =0 ⇒ Maximum = (37)
dt 2 Ath2 dt th
For t ≥ th:
ms 2 f 2 g dQ m 2 f 2g
Q=− t + ms f 2g ∆H (5) =− s (38)
A dt A

Depending on the data for ms, f, ∆H, th, and A, the maximum for dQ/dt will be calculated either by Formula
(37) or Formula (38). Equating the modulus or absolute value of Formulas (37) and (38) results in:

ms f 2 g ∆H m 2 f 2g 2 A ∆H
=− s ⇒ th = (39)
th A ms f 2 g
Note:
- Select a th smaller than calculated by Formula (39) and the dQ/dt will be the largest right at the start of
the F/E process; calculate dQ/dt with Formula (37).
- Using a th larger than calculated by Formula (39) has the effect that the maximum of |dQ/dt| occurs
when the valves are completely opened, at the end of the F/E process. The maximum |dQ/dt| has to be
calculated with Formula (38).
- th equal to the value calculated by Formula (39): Recall Formula (6) and (12) from section 4.1.2,
which compute the total F/E time Ttotal for a process with valves being lifted linearly in time and for a
process with instantaneously lifting (th = 0) of the valves. Substitute Formula (39) in Formula (12) to
find:
2 A ∆H 1
Ttotal = + th (12) ⇒ Ttotal = 1 1 th (40)
ms f 2 g 2 2

The frame below shows a short hawser force calculation to illustrate the different results for head filling
and longitudinal culvert filling, but, more important, to get a quantitative feel for the matter.
Hawser force calculation: Head filling: ms = 0.85
2 2
dQ m f g 0.85 *52 *9.81 2

Using data from section 4.1.2: =− s = = 0.164 m3 / s 2

2 dt A 1080
A = 1080 [m ]
dQ ms f 2 g ∆H 0.85*5* 2 *9.81* 4.5
ms = 0.85 [-]
2
= = = 0.222 m3 / s 2
f = 5 [m ] dt th 180
2
g = 9.81 [m /s]
Governing dQ is the 0.222 :
∆H = 4.5 [m] dt
th = 180 [s] dQ 1 0.222
i= ⋅ = = 0.8*10−3 (= 0.8 0 00)
Some additional data required:
2
dt g ( Av − n) 9.81* (54 − 25)
Av = 54 [m ]
n
**
= 25 [m ]
2 F = iG = 0.8*10−3 * 20 200 = 16 kN
**
G = 20200 [kN] / 2020 [ton]
Longitudinal culvert: ms = 0.54 (Figure 4-18)
** 3
Based on a 2.5 x 9.5 x 85 m ms ,culvert 0.54
F= iG = *0.8*10 −3 * 20 200 = 10 kN (= 0.5 0 00)
CEMT class IV ship ms ,head 0.85
Note:
- Use of a longitudinal culvert system results in smaller hawser forces than use of a head filling system.

 Hydraulic Engineering
Faculty of Civil Engineering 81 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

- The permillages are slightly smaller than the guideline figures mentioned in the next section. This
agrees with the fact that only the hawser force due to translation waves has been computed. Other
force contributions, e.g. impuls differences or friction, were not included.

Practical method to determine the mooring load

Often a vessel in a navigation lock will be stopped in its manoeuvers by putting the forward hawser around
a lock bollard and the ship’s bollards. By easing the hawser of one of the ship’s bollards carefully, the
hawser force has to be controlled by tightening or slipping of the free hawser end, the friction will slow the
vessel down. It is a matter of definition, whether the hawser forces in the previously described process,
should be considered as mooring load or as a berthing load because it slows down the ship.
At rest, in the lock chamber or in the outer ports, emptying and filling the lock chamber results in loads on
the vessel that have to be resisted by the mooring lines. Moored vessels with a mooring or hawser line
configuration as shown in Figure 4-39 are not able to resist much transverse forces. The resistance to
transverse forces is especially low when fixed lock bollards are used and the lines are continuously
slipping to allow the ship to rise and fall with the changing water level. The preferred direction of
transverse mooring forces is towards the wall (contact pressure between ship and wall, no hawser force).

Inland vessel Recreational craft Maritime vessel

Figure 4-39 Hawser configuration for different type of vessels in the lock chamber

Generally the smaller the ship (recreational craft) the smaller the number of mooring lines, and the larger
the probability that only one hawser at the time is transferring the whole tensile force, or not, in case of
slipping or breaking. For the lock the resulting mishap will be fairly negligable.

Inland navigation vessels are generally moored using 2 hawsers. The following is an indication for the
mooring force to be expected in one hawser:
• Ships up to
0
600 tonnes: 1,5 /00 of the water displacement
• Ships up to
0
2 000 tonnes: 1,0 /00 of the water displacement
• Push barge convoy up to
0 0
10 800 tonnes: 0,7 /00 - 1,0 /00 of the water displacement

Mooring of seagoing vessels is different because generally four hawsers will be used, which will be held
under constant tension by the winches of the ship. In conditions of transverse wind or water flow
sometimes one or more extra hawsers will be used in the transverse direction (imagine the amount of time
required to handle all the hawsers). An indication for mooring loads of seagoing vessels, mainly
determined by the tension winch:
• Vessels of over 20.000 DWT:
0
0,25 /00
• Vessels of over 100.000 DWT:
0
0,12 /00
During design the above indications for the mooring forces, related to the water displacement or DWT of
the vessel, can be used. Obviously the governing ship has to be selected wisely, and this might prove to
be difficult in case of large variations in the types of vessels using the lock.

Another approach would be to determine the governing mooring load based on the type of hawser most
probably used. Instead of selecting a governing ship a hawser type has to be selected, which still requires
the designer to consider one or more design ships, however, generally there is considerably less variation
in the break loads of hawsers. The heaviest cables which can still be operated manually are steel cables
with a diameter of 22 mm and a fracture load of circa 105 kN. If hand winches are present, like on push
barges, heavier cables can be used. Nylon hawsers absorb more energy due to their larger elasticity.

 Hydraulic Engineering
Faculty of Civil Engineering 82 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Software used to determine the mooring load

The above calculations or methods are used to obtain a first estimate of the water level gradients caused
by translation waves and the hawser forces. The maximum hawser force is either the result of reducing
the vessel’s speed when slowing down or mooring, or forces induced by filling and emptying the lock
chamber. Seemingly more accurate calculation of the hawser force could be possible e.g. using the
program named LOCKFILL. Figure 4-40 shows the graphical results of a calculation with the program for
the forces in longitudinal direction.

Figure 4-40 Forces in longitudinal directions, calculated using LOCKFILL;

during filling (upper graph), during emptying (lower graph)

 Hydraulic Engineering
Faculty of Civil Engineering 83 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Reduction of hawser forces

Different methods exist to reduce the hawser force:

a. Use of a longitudinal lock culvert:

A culvert with valves positioned in longitudinal direction will encounter the following problem. At the
start the first valve will process most of the water due to the slow acceleration of the water mass in
the culvert. This problem can be diminished by adjusting the valve opening. In this way the
discharge will be divided more evenly thus reducing the longitudinal forces. For instance by
positioning 2 valves at ¼ and ¾ of the chamber length, the chamber will be divided into four parts
with a head filling of ¼ Q each. This is an optimal distribution to reduce the longitudinal forces.
Additionally, the draw-off spots have to be placed as deep as possible to prevent water entering the
lock to interfere with the moored vessels.
b. Use of energy dissipating chamber:
When large differences in the water level are present one can use an energy dissipating chamber to
reduce the entrance speed of the water. By lifting the gate partially water will flow underneath the
gate and enter a chamber. This chamber will reduce the flow velocity.
c. Adjustment of the valve lift program:
By adjusting the valve lift program the hawser force can be reduced. For instance one can start by
slowly opening the valves, opening the valves serially or by temporarily interrupting the lifting. When
the translation wave reflects at the opposing gate the lifting velocity can be increased.
d. Use of many smaller fill openings instead of a few larger openings:
Smaller water jets dissipate their energy easier than large.
e. Breaking the water jet:
The energy of the incoming water (jet stream) can be reduced by:
• The valve shape;
• Positioning breaker plates in front of the valve opening;
• Using a Venetian blind so one valve will open later in time.
f. Use of a bed protection:
If water jet velocities permit a very effective solution to reduce the hawser forces is the use of a bed
protection inside the lock chamber. The bed protection creates a lot of friction, which dissipates a
lot of energy. However, it is quite an expensive solution as well.

4.1.4 Specific subjects regarding F/E

As announced in the beginning of this section F/E, water saving basins, salt water intrusion, ice control,
and large water head difference phenomena, will either be discussed in this paragraph or provided with
some references for further study.

Water saving basins / Side ponds

As explained in Section 2.1.1 every lock cycle water is lost in the upstream reach of the waterway and
added to the downstream reach. The amount of water lost per leveling cycle, Vloss is:

Vloss = A ⋅ ∆H + Wup − Wdown (41)

where:
A = the horizontal water area of the lock chamber
W up = is the water displacement of the ship navigating in the upstream direction
W down = is the water displacement of the ship navigating in downstream direction

There are several options to reduce the water loss due to locking ships. For instance instead of
constructing one large navigation lock, two smaller locks could be constructed. One large lock, leveling
over ∆H, uses twice as much water as two smaller locks, leveling over ½∆H, placed serially.

 Hydraulic Engineering
Faculty of Civil Engineering 84 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Another option would be to construct one or more water saving basins or side ponds (spaarbekken). Part
of the water will be stored in the basin(s) when the lock chamber is emptied. During filling the water from
the basins will be discharged into the lock chamber. One side pond with a horizontal water surface equal
to the lock chamber will save 33% water. During emptying of the chamber ⅓ will flow into the side pond
and ⅔ will be lost. The construction of two side ponds will result in a reduction of 50 %, see Figure 4-41.

Figure 4-41 Principle of water saving basin / Side pond

Assume the horizontal water area A of the lock chamber and the water basin(s) are the same. Divide the
water head ∆H, the water level difference between the upstream and downstream waterway, into an
integer number of N steps. Therefore the lock chamber contains N water volumes with water height ∆H/N,
‘above’ the downstream water level. Let n be the number of water saving basins and let the water basin
height be ∆H/N, hence equal to one of the N water volumes in the lock chamber. Given the condition that
free flow is being used, N has to be larger than 3 because the upper water volume in the lock flows in the
first basin, which is ∆H/N lower, and the last and lowest volume of water in the lock is discharged into the
downstream reach, not in a basin. If this last volume was stored in a basin a pump would be needed to get
the water back into the lock. The amount of water saved or lost per lock cycle is
calculated as follows:
90.0
n 
Vsaved = A.∆H . .100%  %
N  80.0

 N ≥3 (42)
N −n 70.0

Vlost = A.∆H . .100% 

N  60.0

The graph in Figure 4-42 shows that there is little extra 50.0

water saved using 4 or even more water saving basins. 40.0

30.0
Disadvantages of the use of water saving basins are the 0 1 2 3 4 5 6 7 8
n
costs and the extended passage time.
Figure 4-42 Water saved per lock cycle
versus number of basins n

Salt water intrusion

See Appendix 2

Ice control
See Appendix 3

(Both Appendices more or less correspond with text from the report PIANC, Innovations in navigation lock
design, InCom-106, Brussels, 2009, (ISBN 978-2-87223-175-1).

 Hydraulic Engineering
Faculty of Civil Engineering 85 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Large water head difference phenomena

Water hammer and/or cavitation are phenomena that are closely related to larger water head differences.

Water hammer

To prevent overtravel in longitudinal culverts, the valve in the supply line will be shut when the discharge Q
drops below a certain (low) limit. If a valve was immediately closed in a culvert the force, using a/the
momentum equation, necessary to stop the water movement would be equal to:
du
F = m.a = m (43)
dt
Immediate closure would imply dt = 0 resulting in an infinitely large force. Fortunately the equipment used
for closing the valve is not that fast. What does develop is a shock, however, the compressibility of the
water and the elasticity of the culvert will reduce the impact. The shock wave (over pressure) travels back
through the culvert to the upstream reach, where it is reflected by water surface of the infinitely large water
volume and returns as an under pressure wave to the valve. The reflection process will continue as long
as the wave energy is not dissipated by the friction along the culvert walls. The wave celerity c can be
expressed as:
1 If:
c= (44) ρ = 1000 kg/m
3
 1 D  6 2 o
ρ  +  Ebulk = 2.18 * 10 kN/m (for 20
 Ebulk twall .E young  Celcius)
D =5m
Where:
twall = 1 m
c = wave celerity [m/s] 6 2
3 Eyoung = 10 * 10 kN/m (cracked
ρ = density of the water [kg/m ]
concrete)
Ebulk = bulk modulus of water; a measure for the
2 then:
compressibility of water [kN/m ]
c = 1021 m/s
D = pipe or equivalent culvert diameter [m]
twall = wall thickness of the pipe or culvert [m]
Travelling back and forth in a 300 m long
Eyoung = Young’s modulus of pipe or culvert walls
2 culvert, takes 2*300/1021≈ 0.6 s
[kN/m ]
Suppose:
The pressure increase due to sudden interruption of
U0 = Q0/A = 25/20 = 1.25 m/s
the flow is; Rankine (1870), Joukowski (1898):
then for sudden closure:
∆p = ρ cU 0 (45) ∆p = ρ cU 0 = 1000*1021*1.25 ⇒
∆p ≈ 1.25 . 106 kg / ms 2 = 1250 kN / m 2
The pressure increase for slow closure of the valve,
formula by Allievi (1929): If:
L = 300 m
N N2 
∆p = p0  + +N (46) tvalve = 180 s
2 4  p0
2
= 200 kN/m (∆H*γ= 20*10)
 
ρ LU 0
N= (47) then for slow closure:
p0tvalve ρ LU 0 1000*300*5
N= = = 0.042
Where: p0 tvalve 200 000*180
2
∆p = pressure increase [kN/m ]  0.042 
0.0422
U0 = water velocity when valve closes [m/s] ∆p = 200  + + 0.042  ⇒
2  2 4 
p0 = original (hydrostatic) water pressure [kN/m ]  
∆p = 45 kN / m 2

 Hydraulic Engineering
Faculty of Civil Engineering 86 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Note:
- Given a wave shock celerity in the order of magnitude of hundreds meters per second, it is safe to
assume that the return shock wave arrives back at the valve when the closing process is still
continuing. As a result Formula (46) for ‘slow’ closure of the valve has to be used for computation of
the extra pressure on the valve due to water hammer.
- In the exceptional case that the time needed for closing the gate is smaller than the time needed for
the shock wave to travel back and forth, then Formula (45) for ‘sudden’ closure has to be used to find
the extra pressure on the valve due to water hammer.

- An over pressure shock wave travels back in the opposite discharge direction, however, at the other
side of the valve an under pressure wave travels in the same direction as the discharge. Once
reflected by the downstream water mass, this wave returns as over pressure to the valve.

Cavitation

Some times the flow velocity can become so high that the local pressure decreases to below the vapour
pressure, or the negative shock wave due to valve closure results in under pressure. Below the vapour
pressure bubbles filled with vapour will develop in the fluid. When these bubbles reach places where the
pressure is higher again, they implode and cause damage to structures and equipment (concrete or steel
lining, valves, etc). This phenomenon is known as cavitation; some characteristics of cavitation:
• unstable flow pattern;
• less efficient discharge through the culverts (loss coefficient cavitation to be introduced);
• vibration of the valves and their lifting equipment;
• considerable wear of the valves etc., or the concrete culvert walls (cavitation erosion);
• a lot of noise.

It is possible to reduce cavitation at the inlet/outlet valves by adding air to the water. Damage can be
reduced by linings made of steel or the use of high strength concrete with steel fibers. An additional
measure which has a positive effect is by widening the culvert down stream. The implosion of the bubbles
will occur further away from the culvert or lock wall.

References on water hammer and cavitation:

- Fluid-structure interaction in case of waterhammer with cavitation, A.S. Tijsseling, Doctoral thesis
Delft University of Technology, 1993
- Elementary Hydraulics, J.F. Cruise, M.M. Sheriff, V.P. Singh, Thomas Nelson, Student edition,
2007
- Water-resources engineering, D.A. Chin, Pearson Prentice Hall, Upper Saddle River, New Jersey,
nd
2 Edition, 2006
- Pipeline design for water engineers, David Stephenson, Elsevier scientific publishing company,
Amsterdam, 1976

 Hydraulic Engineering
Faculty of Civil Engineering 87 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

4.2 Lock Gates

In history, when locks were constructed with masonry (metselwerk) walls, gates were constructed from
wood and kept as small as possible. Good reasons to use a small gate were the limited strength of the
wood, lack of the (mechanical) power that is needed to move a large gate, and costs. Frequently locks
were constructed with gates having a span smaller than the width of the lock chamber, see the mentioned
bayonet lock in Section 3.1.2. The historical technical limitations do not exist anymore, moreover, given
nowadays economic importance of waterway transport, those limitations would not be accepted anymore.
Steel, even concrete or synthetic gates, are simply designed and constructed up to the size required. The
time needed to pass a navigation lock has become a dominating factor and as a result reduction of the
time to operate the lock gate as well. Mechanical and electrical devices, taking care of the movement of
the gate, nowadays have been developed to such an efficiency level, that it takes 3 minutes on average,
to open or close gates for both inland and coastal locks, see Table 2-4.

4.2.1 Gate function

A gate (sluisdeur) in a navigation lock will function as a means or an element, able to separate as well as
connect different bodies of water. The following functions and design requirements can be defined:

Water retention function:

The separation of bodies of water most frequently having a different water level requires:
• Prevention of water flowing from a higher water level to a lower water level. Although the gate
closes off by far the biggest part of the wet cross section of the lock head, at the wet perimeter of
the gate, leakage and seepage will occur.
Depending on project specific requirements limited leakage and seepage will or won’t be allowed. In
coastal areas strict requirements are often imposed to maintain a separation between salt and fresh
water.
• Resist the resulting hydrostatic pressure caused by the difference in water level on either side of the
gate and transfer this load to the supports at the lock head. The direction of the resulting hydrostatic
load is always from the higher water level side to the lower water level side; hence, the direction
changes when the water level on the other side gets the highest. Difference in direction can be
caused e.g. by tide. If needed the gate has to be able to resist the force in both directions.

Separation or connection function:

• In view of navigation, water bodies must be separated on one side of the lock, see retention in the
above, as well as connected on the other side of the lock. In general simply opening or lifting the
gate results in too large and too turbulent water movements either in the lock chamber or in the lock
approach areas, which means an existing water level difference has to be leveled out. Water
movements, more specific the resulting mooring forces are not allowed to exceed a certain
threshold. A means to take care of acceptably quiet water movements is to provide openings that
can be gradually adjusted in size.
• It has to be possible to move the gate completely out of the cross sectional area required for
navigation to provide safe passage for navigation.

4.2.2 Gate types

Using the most important functions and resulting requirements, first to generate, then to select gate
alternatives, generally speeds up the overall design process. For navigation locks, the way to move the
gate out of the required cross sectional area for navigation is quite a distinguishing criterion. Gate
movement often is a translation or a rotation, hardly ever a combination of the two. Given the three
dimensions there are 6 theoretical movement possibilities. Looking at the most commonly used gates,
descriptions in further detail will follow, in real life only 4 of these possibilities are frequently in use, see
Table 4-4.

 Hydraulic Engineering
Faculty of Civil Engineering 88 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Table 4-4 Gate types and gate movements

Translation Rotation axis Used
direction
Mitre gates (puntdeuren) vertical frequently
Single leaf gate (enkele draaideur) vertical frequently
Lift gate (hefdeur) vertical frequently
Submersible lift gate (zakdeur) vertical
Rolling-, sliding or caisson gate horizontal frequently
(roldeur)
Radial gate or sector gate vertical
Radial gate or tainter gate horizontal

Mitre Gates (puntdeuren)

Mitre gates (puntdeuren) consist of two leaves, symmetrical with regard to the centre line of the lock,
rotating around the vertical axis. The force resisting principle is very simple; in closed position both leaves
point in the direction of the upper water level. On a global force level the water pressure results in normal
forces in the gates and spalling forces acting on the lock head. On a local force level bending moments
and shear are introduced in the doors as well.

Mitre gates generally should not be exposed to significant current or wave actions when partially open. For
this reason mitre gates are unsuitable for use as a weir (stuw) or flood gate (vloed deur) but are frequently
used as lock gates. In some cases, when for instance hydraulic cylinders are used to open en close the
gates, the gates can be designed to resist a small reverse water head difference.

In open position the gates are turned in recesses in the walls of the lock head. In order to prevent damage
when vessels navigate through the lock head, there should be enough distance between the gate and the
face of the recess wall, or fenders should be attached either to the gate or the face of the recess wall.

Figure 4-43 Water pressure (W), contact force (H) and spalling force (S) on a mitre gate

Bits and parts of the mitre gate:

Stresses due to the weight of the gates are transmitted to the heel post (achterhar) and then via the pivots
to the collar strap (halsbeugel) and socket (taats). All the items must be designed to resist the horizontal
forces; the bottom pivot and socket also carry the whole (underwater) weight of the gate. The axis of
rotation of the gate should be in such a position that, on one hand, separation of the gate and the concrete

 Hydraulic Engineering
Faculty of Civil Engineering 89 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

of the recess walls is ensured during the whole move from open to closed gate position and vice versa.
On the other hand, to retain water in closed position the space between the gate, hence the axis, and the
recess wall should be as small as possible to limit water flow around the gate. To accomplish both, a
solution would be to choose an eccentric position for the gate in the gate, i.e. at some distance from the
centre line.

A Elevation (Aanzicht)
B Section (Doorsnede)

1 heel post (achterhar)

2 mitre post (voorhar)
3 girder (regel)
4 top girder (bovenregel)
5 bottom girder (onderregel)
6 diagonal strut (schrankschoor)
7 planking or sheeting (beplanking)
8 diagonal strap (trekstang)
9 top pivot / gudgeon pin (halsprop)
10 metal straps (ijzeren beugels)
11 T strap (T – kruk ijzer)
12 L strap (L – kruk ijzer)
13 valve opening in gate (schuifopening)
14 pivot and socket (taats en
taatsplaat)
15 shoe for the pivot (schoen met
taatskom)
16 through bolt (doorgaande
bout)

Figure 4-44 Parts of (old) wooden mitre gates *

* Courtesy of PIANC: http://www.waterdictionary.info/

Moving the gate:

Generally speaking, the equipment for operating the gates is located above the highest upstream water
level and as near as possible to the lock wall (so there is a danger of being damaged by downstream
bound ships). Operation of a mitre gate, results in torsion stress which should be taken into account when
designing and dimensioning a gate. Some basic information on the equipment will be presented below;
the civil engineer has to keep in mind that design of the equipment, including all the necessary mechanical
and electrical items, the installation and control systems, is a job on its own, best done by engineers with
relevant experience.

Electro mechanical:
The so called "panama wheel" is most widely used in locks to operate the gate. The wheel is operated
through a set of gears put into motion by an asynchronous motor. In spite of the constant speed of the
gear wheel the rate of angular displacement of the leaf is reduced at the beginning and at the end of the
operation, when stresses to be resisted may be the largest.

Hydraulically (oil):
Today almost exclusively hydraulic cylinders are used in new design. A hydraulic jack fed by a volumetric
pump and attached to a frame in its centre, acts on the piston rod, the end of which is connected to the
gate. The jack is positioned at 1/4 to 1/2 of the gate length from the axis of rotation.

 Hydraulic Engineering
Faculty of Civil Engineering 90 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Figure 4-45 Position of hydraulic jack moving the mitre gate

Advantages of mitre gates are:

• No air draught limitations. High superstructures such as lifting towers or pillars are not needed.
• Low cost, effective solution for smaller locks. The three-hinged mechanics scheme results in a
relatively light gate structure.
• The equipment for mitre gates does not have to carry the weight of the gate; the centre of gravity
remains in a horizontal plane during opening and closing. Resulting stresses during operation are
reduced to those due to frictional forces at the pivots, socket and collar strap and the dynamic effect
of moving the gate through the water.
• Opening and closing of the gates takes a short time.

Disadvantages are:
• The length of the lock chamber needs lengthening as well as the upper head by the recesses of the
gates.
• To retain water in both directions a double set of doors is required.
• Very accurate mounting and frequent checking of the contact at points of load transmission (thrust
blocks and quoin blocks) is necessary.
• Mitre gates can not be opened or closed under a water head difference
• Operation may be disturbed by debris and ice.

For increasing widths of the sluice, the type of material and the design of gates changes. Mitre gates
(puntdeuren) which are often constructed in wood become less suitable and are replaced e.g. by steel
rolling gates. The width of mitre gates is limited to approximately 20-25 meter (10-12.5 m per door). What
is especially important for this is the height width ratio. When the ratio is smaller than 1 the forces on the
pivot points are relatively large and may become a problem, especially the (horizontal) force on the upper
one.

Advantages of mitre gates often outweigh their disadvantages as is demonstrated by their utilization
throughout the world. However, since vessel size is increasing this type is less frequently applied. For
reasons of construction and maintenance costs (and sometimes the status as monument) the gates of
small and medium size of navigation locks are made of either steel or wood. A new development is the
use of glass fiber reinforced plastics (GFRP), which have little need for maintenance.

Single leaf gate (enkele draaideur)

The single leaf gate (enkele draaideur) could be considered as half a mitre gate, their appearance, shape
and materials used are similar, however, there are some significant differences. Where the mitre gate
transfers the hydrostatic load for a considerable part by normal forces, the single leaf gate brings it all
away by bending moments. The recess in the wall for the free end of the leaf gate has a different shape,
see Figure 4-46, which is a bit better for streamlining water flow.

 Hydraulic Engineering
Faculty of Civil Engineering 91 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

If selected at all, the single leaf gate is most frequently used for locks having a small width, e.g. for the
locking of recreational craft.

Figure 4-46 Single leaf gate compared with a mitre gate

Advantages
• Very suitable for locks of smaller width because of simple construction and operation.
• No air draught limitations.
• It is possible to lock the door on the free end which will make it possible to retain water in two
directions.
• Forces on the gates are transferred parallel to the lock wall (when the closed leaf gate is
perpendicular to the lock axis).

Disadvantages
• Recesses for single leaf gates are longer than those for equivalent mitre gates, thus single leaf gates
need longer structures.
• Opening and closing of the gate results in a lot of water displacement.
• More rigid and thus heavy construction needed for the pivots, socket and collar strap.
• Vulnerability for ice ad debris.
• Not suitable for wide locks because the supports will be very heavy.

Radial gates with vertical axis of motion (sectordeur)

Radial gates with a vertical axis of rotation are often referred to as sector gates. The skin plate of the gate
has the curved shape of a (part of) a cylindrical circle, and has to be stiffened to resist the hydrostatic
water pressures, see the Figures in this and the following paragraph. The lower and upper gate arms,
triangular shaped trusses in the horizontal
plane, support the stiffened skin plate
horizontally and vertically. Between the upper
and lower arm there will be diagonal braces
or struts, see Figure 4-47, to provide the
required vertical stiffness. At their rear ends
the gate arms transfer the loads to pivot
points, which are attached to, better said
casted into, the lock head wall. Due to the
shape of the skin plate the resulting force of
the hydrostatic pressure has its working line
through the pivot.

Sector gates have the advantage of being

statistically determinate, can resist reverse
heads and can be operated whether there is
Figure 4-47 Sector gate during construction
a water head or not. The weight of the gates
may exceed that of the equivalent mitre gates, making them more expensive. Another cost setback are

 Hydraulic Engineering
Faculty of Civil Engineering 92 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

the large and deep recesses in the lock heads. Sector gates are more often used for guard locks or storm
surge barriers where closure under free flow and unlimited air draught is needed.

Moving the gate:

Similar to mitre gates, sector gates can be pushed into movement using mechanical equipment, e.g.
machinery or hydraulic jacks. The driving force can be applied to the upper gate arm. Contrary to mitre
gates, those have to push away the water body in front of the gate, the sector gate cuts through the water
with its edge. Therefore the installed power of the equipment used is relatively small, even though the gate
itself is heavier. Again, the design of the equipment to be used, including all the necessary mechanical
and electrical items, the installation and control systems, is best done by engineers with relevant
experience. Note: the space for the recesses, the equipment and the ‘wiring’ that has to be provided,
generally requires quite a lot of detailing work to be done during concrete structure design and results in
quite some complications during construction due to all the required cast-in items.

An alternative for the use of mechanical

devices, to move the door, is using the
hydrostatic water pressure difference due to
differing water levels on either side of the
gate. Extra skin plate would be required at
the straight face of the sector gate closest to
the lock head wall, see Figure 4-48. With
appropriate sealing along the edges this
plate closes off the water body in the recess.
If the water level in the recess is the highest,
the sector gate moves out, vice versa if the
water level in the recess is lower than in the
lock chamber or lock head the gate will move
into the recess, i.e. open. To change the Figure 4-48 Horizontal cross section sector gates
water levels in the recess, water will be let in
or out through the culverts either by means of a natural water head or using pumps.

Advantages of sector gates are:

• The possibility to close the gate in flow conditions.
• Relatively easy to move because it cuts through the water with a small surface.
• No air draught limitations, hence no lifting towers or pillars.
• The equipment for operating the gates does not have to carry the weight of the gate; the centre of
gravity remains in a horizontal plane during opening and closing. Frictional forces at the pivots and
the sealing result in horizontal forces.
• Opening and closing of the gates takes a short time.
• It is possible to retain water in two directions.

Disadvantages are:
• Relatively large amount of material because of the developed length of skin plate and the gate arm
trusses that have to be braced.
• High support loads at the pivots resulting in heavy cast-in items in the lock head wall due to weight
of the gate. To mitigate this disadvantage float elements assembled to or integrated in the gate may
be used.
• The recesses required to move the gates out of the way are large and substantially increase the
size of the lock head structure.
• Operation may be disturbed by debris and ice.

 Hydraulic Engineering
Faculty of Civil Engineering 93 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Note:
If the sector gate is moved using hydrostatic water pressure some of the advantages and
disadvantages mentioned for sector gated in the hereinabove, become bigger, others may even
be reversed.

Radial gates with horizontal axis of rotation

Radial gates with a horizontal
axis of rotation are frequently
known as tainter gates.

Reference is made to the

descriptions of the sector gate in
the above. Compared to sector
gates the tainter gate:
• A single tainter gate spans
the whole width of the lock
head opening. Generally two
sector gates are used.
Lifting to open for navigation:
• In lifted position there will be
Figure 4-49 Stiffener principles tainter or sector gates
a limitation to the air draught
of passing vessels.
• The equipment has to be
able to carry the weight of the
gate. Counter weights may
be used to mitigate this
disadvantage.

Lowering to open for navigation:

• Lowered there is no air
draught limitation
Figure 4-50 Hydrostatic forces (left); Forces during lifting and
• To lower the gate a deep
lowering (right)
recess in the lock head is
required, significantly
increasing the lock head structure, complicating construction because of the much deeper foundation,
hence considerably increasing costs.
• Again, the equipment has to carry the weight of the gate. Besides counter weights, floating elements
assembled to the gate may be used as well to mitigate this disadvantage.

It is possible to use this type of gate for guard gates, flood defense or storm surge barriers, provided the
gate body is designed high and strong enough to resist underflow or/and overflow.

Figure 4-51 Radial or tainter gate during construction (left & middle); hydraulic jack (right)

 Hydraulic Engineering
Faculty of Civil Engineering 94 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Lift gate in upward direction (hefdeur)

Lift gates spanning the entire width of the lock chamber and operating in vertical recesses enable
hydrostatic loads to be transmitted to the lock head. Lifting the gate opens the lock. The deadweight of the
gate is balanced by counterweights in order to reduce the operating forces of the equipment. High gantries
or guide towers (heftorens) are required in most cases to guide the gate when moved into open position.
This is illustrated in Figure 4-52.

Figure 4-52 Lock head for a lift gate

The advantages of lift gates are:

• Little space is needed for the lock head, thus enabling the total lock length to be reduced to a
minimum.
• Considering the driving gear, it is possible to open the gate under a water head because the force
directions are in different planes.
• Not seriously affected by debris and ice in closed or opened position.
• Easy to control and repair.
• The support system of the gate is statically determinate and therefore largely insensitive to differential
settlement.
• It is possible to retain water in two directions.

Operational disadvantages are:

• Air draught limitation of vessels.
• The superstructure, lift towers or columns, is large and heavy.
• Lifting the whole gate results in very strong underflow.
• Spillage of water on vessels passing the gate.
• Difficulties with the roller tracks because of the deflection of the bottom of the gates when they are
under full head.
• Balancing the gate may be complicated and expensive. For opening, floating ice and debris, as well
as ice frozen to the gate may increase the weight of the gate and negatively influence the balance with
the counterweights.

Submersible lift gate (zakdeur)

The submersible lift gate (zakdeur) is lowered for the passage of ships. The first phase of lowering can be
used for filling and emptying the chamber. For purposes of repair the gate is lifted above the water level.
This type of gates takes little horizontal space and has the advantage that vessel height is not limited.

 Hydraulic Engineering
Faculty of Civil Engineering 95 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

However, it requires a large recess deep into the ground to accommodate the gate and its construction
and maintenance costs are high.

Figure 4-53 Cross-section of a submersible gate

Most of the horizontal loads on the gate are diverted to the side recesses; a small amount is transferred to
the bottom support (onderaanslag). The submersible gate is able to retain water in 2 directions.

Caisson, rolling or sliding lock gate (roldeur)

The caisson, rolling or sliding lock gate (roldeuren) is preferred where large spans have to be bridged
without limiting the air draught. This type of gate can only be used when sufficient space exists besides
the lock, because the width of the lock head structure is about two times the doors length, see illustrations
in Figure 4-54.

The rolling gate moves on underwater rails or sliding tracks, which bears the risk of the gate getting
jammed due to accumulation of sediments or debris on those tracks. Especially if the gate frequently
remains in the open position it should be considered to wipe the rail or track clean before or during gate
operation. Moving the gate, reducing weight and frictional forces, thus reducing wear and tear as well, is
easier when buoyancy tanks are used. When the gate is in closed position the tanks can be filled, now
they serve as ballast tanks, to ensure the necessary contact pressure.

Figure 4-54 Rolling lock gate and its lock head; cross sections and 3D impression

A combination of factors results in the generally considerable thickness of rolling gates. Obviously one
factor is the large width of the lock to be spanned by the gate. The need for stability results in wheels or
slides on both sides of the gate and two rails or sliding tracks under one door, not just one in the center of

 Hydraulic Engineering
Faculty of Civil Engineering 96 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

the door, which adds to the thickness. Ever so often the thickness of rolling gates can be used to provide
passage to light traffic.
In closed position the rolling gate can be considered as a simply supported beam; in fully opened position
there are hardly any horizontal forces on the door. In those positions stability of the gate is not a problem.
During the opening or closing process the ‘upper’ corner of the gate is completely unsupported. A force on
this door area, e.g. due to a remaining water level difference or large flow, may result in significant stability
problems. Keeping the resulting vertical force within a distance of 1/6 of the gate’s thickness to the centre
line of the gate is a good measure for providing stability, but requires a certain thickness of the gate.

The advantages of rolling lock gates are:

• Suitable for locks of large width.
• No air draught limitation.
• It is possible to retain water in two directions.
• Easy to maintain.

Disadvantages:
• Requires a lot of space besides the lock (head).
• Requires a large recess or lock head.
• Can not be opened or closed under water head conditions.

Innovative gate types

The following has been included to show some developments and/or possible gate innovations; it is not
proven technology what is shown, in fact there is a good chance the idea will never leave the drawing board or
design stage. And of course gate innovations are not limited to what is shown here.

suspended gate
Figure 4-55 Innovations of mitre gates

Figure 4-56 Innovative gate types

4.2.3 Transfer of forces

As an example the transfer of forces through a mitre gate will be discussed. The following items have to
be analysed:
a. Identify possible (hydraulic) loads and load combinations and calculate their magnitude
b. Schematisation and selection of supports (opleggingen)
c. Determine the load transfer through the gate or door

 Hydraulic Engineering
Faculty of Civil Engineering 97 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

a. Identify possible loads and load combinations and calculate their magnitude

To identify loads and the combinations it helps to look at the gate in opened and closed position, or
consider the situation when the gate is being moved.
1. Load on a mitre gate in closed position: e.g. hydrostatic water pressure
2. Load on a mitre gate in opened position: e.g. dead weight
3. Loads on the doors due to debris when trying to close
4. Loads on the mitre gate subjected to ice
5. Loads on mitre gates due to ship impact
6. Loads on a mitre gate under wave attack
7. etc.

These situations and loads combine to a number of load combinations. As example combinations 1 and 2
will be elaborated in the hereinafter.

b. Schematisation and selection of supports (opleggingen).

When schematising and selecting the type of supports for the gate, not only the transfer of forces has to
be taken into consideration, but preventing leakages or flow under or around the closed gate as well.

In opened position the gate is simply supported, see left-hand side illustration in Figure 4-57. The bottom
pivot provides both the vertical reaction and a horizontal reaction force. The upper pivot provides the
horizontal reaction in the opposite direction. Rotation in the horizontal plane around the axis through the
supports is possible. Looking at deformations globally, the gate will get out of square (schranken) if it is
not stiffened sufficiently in its plane. This is a problem if there is too little clearance (speling) between the
bottom of the gate and top of the lock head floor. In the act of opening or closing the gate, the bottom of
the mitre post (voorhar) would touch and scratch the floor.
Obviously leakage of water or sealing is not an issue at al when the gate is open (or being opened).

Figure 4-57 Supports, loads and resulting deformations

When closed and retaining water, there are two (main) alternative systems to support a mitre gate. In the
first alternative, in the middle in Figure 4-57, the mitre gate has a vertical support at the bottom pivot point
(taats). Horizontally there will be a line support at the mitre post (voorhar), and either one or two simple
supports at the heelpost (achterhar), the last depending on the clearance in the bottom pivot shoe.

To be able to install the gates, there has to be a certain clearance or construction tolerance available in
the bottom pivot point. After installation of the door into the lock head, little can be done about the bottom
pivot or the available clearance. Often the upper pivot is readjustable to facilitate installation, which is also
convenient for later operational stages.
It is possible to design and construct the pivot shoe and socket (bottom) with such a large clearance that
the heel post is being pushed against the concrete wall of the recess by the water pressure. Suppose
enough (horizontal) clearance is available in upper pivot as well, than the heel post touches the wall over
its entire length. This is the second support alternative in closed position. In that case the gate has line

 Hydraulic Engineering
Faculty of Civil Engineering 98 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

supports on three sides for horizontal loads, the fourth side, the top girder, is free or unsupported, see
Figure 4-57 on the right.
For both alternatives supports and loads have been drawn at the right-hand side door, the deformations
have been indicated at the left-hand side door. The deformations of the gate are important for analysing
and preventing the leakage of water, usually done by providing sealing between the gate’s edges and the
‘surrounding’ concrete. Hand in hand with larger shear and torsion stresses, that are bigger for simply
supported plates than for line supported plates, go larger deformations of the gate that result in more
leakage. Larger deformations and leakages require larger seals, see Section 4.2.4, which increases initial
and maintenance costs.

c. Determine the load transfer through the gate or door

Sketch how forces acting on a structure are being transferred. In case of a mitre gate draw up how the
loads are transferred by beams, girders, the axis and hinges to the lock head.

This time consider a closed mitre gate first: the governing load to be considered is the maximum water
level difference.

Global force transfer

From construction mechanics point of view, the V-shape of the two leaves is a statistically determinate
three-hinged truss, which transfers the water pressures by normal forces to the supports in the gate
recesses; this is the global or overall force action. Locally the water pressure results in bending moments
in the truss.

Figure 4-58 shows a top view of forces acting on a closed mitre gate, only one of the doors has been
included, as well as the force vector diagrams for decomposition of the main forces. The resulting
hydrostatic load W is resisted by the force H, the force between the 2 doors of the mitre gate, and spalling
force S (spatkracht), the force acting on the lock head and vice versa. For reasons of symmetry the forces
H between the 2 doors, in the point(s) of contact of the mitre posts (voorharren) are in equilibrium. They
are of the same magnitude but their direction is exactly opposite. Both H and S can be split up in a
direction parallel (N) and a direction perpendicular (F) to the door.

S W │S│=│H│
½W = -F
H

N N F
S
F H

Figure 4-58 Forces and reactions on a closed mitre gate

Since the reaction force S can be divided in a force parallel and a force perpendicular to the gate these
forces are a function of the hydrostatic pressure W and the angle α. Expressed in formulas:
W W
S=H = F =½ W N=
2sin α 2 tan α
From the above formulas it can be derived that for small angles α the spalling force is high. This would
suggest using a larger deflection angle α. However, when increasing this angle the gates will become
larger, therefore requiring a heavier structure, but the soliciting load will be larger as well, hence the gate
will be heavier. For reasons of economics usually an angle of tan α = 1:3 has been determined as the
most optimum one.

 Hydraulic Engineering
Faculty of Civil Engineering 99 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

As shown in the previous section, the spalling force S at the heelpost, or F and N, may work as a
concentrated load (puntlast) or as a distributed load on the lock head structure and gate, action is
reaction, depending on the support conditions. The same goes for the mitre post; depending on the
detailing of the gate the force H will either be more concentrated or distributed.

Local or internal force transfer

The flow of forces through the gate, the local or internal force transfer, is strongly influenced by the type,
the arrangement or composition the girders and posts of the gate. The ‘local’ arrangement has to be in
agreement with the ‘global’ situation, ie. the support conditions of the gate, see previous paragraphs. For
the girder and post arrangement the following two main options exist:

1. Load transfer equally distributed along the vertical posts, illustrated in Figure 4-59. The gate will be
constructed using a relatively large number of horizontal girders (regeldeur), more or less equally
spaced, which results in an equally distributed load on the heel and mitre post. The horizontal water
pressure is predominantly carried away by horizontal internal forces. In agreement with the equal
distribution the pivots should have quite some clearance to allow contact, either direct or via sealing,
of the heel post along the whole height with the lock head wall. As such, load transfer along the post
of the lock head is via distributed line loads N and F.

Figure 4-59 Flow of forces in a gate with (many) horizontal girders

2. Load transfer concentrated in a view points, see Figure 4-60. Instead of a predominant horizontal
force flow, a larger part of the load is flowing to the supports via vertical routes. The gate will be
constructed using two heavy girders (regels) between the heel and mitre post. The water pressure is
transferred via plates and subsequently posts (stijlen) to the main girders. Positioning of the 2 heavy
girders should be such that they carry away about the same load and such that there is an optimum
between the bending moments in the heelpost and the other posts.

Both the heel and mitre post are simply supported, the heel post by the upper and bottom pivot, the
mitre post will be provided with two contact points or surfaces that can be less or more pronounced.
No, or only very little clearance will be allowed in the pivots at the heel post. In view of nowadays
abundant availability of high precision (with very little clearance) ball bearings (kogellagers), being able
to resist large loads, this is not a problem anymore.

Figure 4-60 Concentrated flow of forces in a gate – simply supported

Given the availability of appropriate ball bearing supports, it is more straightforward and cost effective to
use the principle of global and local force transfer through concentrated flow of forces and simple
supports.

 Hydraulic Engineering
Faculty of Civil Engineering 100 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Now consider the open gate again.

The governing load in, when the gate is in an open position, will be the dead weight of the gate. Unless a
very special upper pivot point is designed and constructed (this was done for the gates of the
Oranjesluizen in Amsterdam), the downward dead weight will be resisted by an upward reaction force in
the lower pivot point. The destabilizing moment to be resisted will result in 2 equally large horizontal forces
in opposite directions; tension on the upper support and compression on the lower support, if the lock
head is taken as reference. This is illustrated in Figure 4-61.

ΣV = 0 → FV = W
ΣH = 0 → FHbot = FHtop
ΣM = 0 → FHbot * h = W * a
a
→ FHbot = W *
h
a = 21 b

Figure 4-61 Balance of forces for an open gate

When the size of the gate increases, the gate will tend to get out
of square (schranken) due to its dead weight. This is a problem
especially for gates made of wood, because wood elements are
usually limited to a certain size and need a lot of connections.
From a stiffness or deformation point of view the welding of
steel is much more convenient. The shear deflection of wooden
mitre gates can be limited by using beams or girders in a
diagonal position, i.e. by using (wooden) struts. A steel tension
bar is a better solution because it adds considerably less
weight. This is illustrated in Figure 4-62.
Steel gates with appropriate stiffener arrangements are much Figure 4-62 Strut or tension bar
stiffer than wooden gates, and getting out of square is rather
easily solved, if it would be a problem at all.

4.2.4 Sealing
Retention of water is one of the main functions of the lock and the gate is the means to this end. To retain
all the water the gates must be connected completely watertight to the surrounding lock head structure,
and the doors of e.g. sector or mitre gates should not be left ajar (op een kier). However, to move the gate
there must be some clearance between the gate and the lock head and to leave a small gap between two
doors could be better than running the risk of damaging then during closure. To prevent the flow of water
through these clearances and to allow movement of the gate as well, sealing (waterafdichting), generally
(soft) wood or rubber profiles attached to either the gate or the lock head, is being used.

For mitre gates the positions to use a seal are:

• Between the heel post (achterhar) of the mitre gate and the wall of the lock head; this seal runs in
vertical direction.
• Between the lower girder of the gate (onderregel) and the lock head floor, or lock bottom; this is a
horizontal seal.
• Between the mitre posts (voorharren) of the 2 doors of the mitre gate; another vertical seal.

 Hydraulic Engineering
Faculty of Civil Engineering 101 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Sealing between the lower gate girder (steel) and the

chamber floor:
On the bottom girder of the gate two steel strips are fillet
welded (hoeklas), see Figure 4-63. Between the strips a
hollow rubber profile reinforced with a steel U-profile is
assembled to the gate by means of a long bolt running
through the welded strips, the U-profile and the rubber of
the seal. A nut on the other side, fixes the whole seal to
the gate. The U-profile keeps the rubber material in shape
and position in case of gate displacements perpendicular
to this cross section. When the gate is closed the rubber
profile makes contact with a metal strip casted into the Figure 4-63 Sealing between lock floor
mitre sill, and only needs a little compression to prevent and the bottom girder of a steel gate
water seeping through.

Sealing between the heel post of a wooden mitre gate and the lock head:
In Figure 4-64 on the left-hand side, the heelpost of a wooden mitre gate door and the posts on the wall of
the lock head are shown. If there is sufficient clearance in the pivots, the door is firmly pushed against the
posts on the wall. The usually hard wood heel post is not equally stiff in both directions, as a result sealing
is not perfect at the spot where line support N works. Using a compressible seal is in conflict with transfer
of the line force N. Use of a soft wood seal, pine or oak (grenen of eik), at the spot where line force F
works solves the leakage problem. On the right-hand side of the Figure a sketch to illustrate what is the
result, considering the displacement of points of the heel post, of the eccentric position of the axis;
eccentric to the heel post centre. Especially when there is little or no clearance in the pivots such a sketch
is better used to determine the best position for one or two seals. In this situation the advise is to use
rubber sealing which is more easily compressed.

Figure 4-64 Top view on the heel post of a wooden mitre gate and the post on the lock head

In both figures in the hereinabove the steel strips (aanslagstrippen) on/at the surface of the concrete, have
steel anchors at the back to cast them into the wall. The usually stainless steel strips have to be
positioned with more than the usual accuracy. It is not uncommon to cast the biggest part of the concrete
walls first using block outs, which keeps hollow spaces free in the wall. In a later stage anchors and stips
are positioned in the block outs and on the wall and carefully casted into the wall.

4.2.5 Gate selection

When selecting a gate not only the gate has to be taken into consideration, but the surrounding lock head
structure as well. Although costs will be used as a criterion in a gate selection procedure, the first to be
considered are the criteria derived from functional requirements. The following could be taken into
account when selecting the most suitable gate:

a. Type of navigation lock and lock width:

For a first ‘rough’ selection of gate alternatives, i.e. on a general level without considering any specific
local conditions or a special ship traffic situation, Table 4-5 could be used.

 Hydraulic Engineering
Faculty of Civil Engineering 102 3/3/2011
Delft University of Technology
 CT 3330 Hydraulic Structures Locks

Table 4-5 Gate selection per type of navigation lock and depending on lock width
Type of Lock Rolling
Mitre gate Single leaf gate Vertical lift gate
lock width* gate
Single Double Double Single Double Double Single Double
sided sided sided sided sided sided sided sided
water water water water water water water water
retention retention retention retention retention retention retention retention

Very
X
Single small
sided
Small X X
water
retaining
inland
Medium X X
lock
Large X X
Very
X
Double small
sided
Small X X X
water
retaining
inland
Medium X X X
lock
Large X X X X

Small X X X
Double
sided Medium X X X
water
retaining Large X X X
sea lock
Very
X
large
* Lock width explanation:
• Very small 4-6 m • Medium 10-16 m • Very large >24 m
• Small 6-10 m • Large 16-24 m

b. Retention of water in one or both directions:

Mitre gates are only able to retain water in one direction. When water retention is required in two
directions a double set of mitre gates must be installed, or special measures must be taken to equip a
single set for this task. A double set of mitre gates results in an increased length of the lock heads.
Rolling or sliding gates, or lift gates are possibly a more suitable solution.

c. Opening and closing of gates under head difference:

Unless special precautions are made, mitre gates and rolling gates should not be opened or closed
under a water head difference for reasons of strength and stability.

d. The area available for lock heads:

Often the lock head for a lift gate requires the least of area and is relatively easy to construct, however,
the big disadvantage of the lift gate will always be the limitation of the air draught of the passing

 Hydraulic Engineering
Faculty of Civil Engineering 103 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

vessels. Due to their height lift gates often are a landmark, which is difficult to fit in harmonically with
the surrounding environment. Except the submersible gate, all the other gate types have a larger
horizontal footprint, relatively spoken, than the lift gate. The construction and maintenance cost
disadvantage of submersible gates generally overrides the horizontal space advantage.

e. Width of the lock chamber:

Lift and rolling gates are able to span the full width of a lock chamber. However, the bigger lock width
and larger gate go hand in hand with a larger and heavier lock head structure. Since the 2 doors of
mitre gates have about half the size of the chamber width this will results in lighter lock head
structures.

f. Equipment to operate or move the gate:

The centre of gravity of lift gates is moved vertically, this requires a large amount of power, thus
considerable machinery. Although this can be reduced significantly by the use of counter weights, a
certain price has to be paid with regard to machine power on one hand and the required strength and
stiffness of the concrete structure on the other hand. The same will apply for a submersible gates and
sector gates with a horizontal axis.
Mitre, rolling gates and sector gates with a vertical rotation axis move in horizontal direction, which is
less demanding for equipment. The installed power and the equipment will be smaller.

g. Vulnerability for ship impact

Supports of mitre gates are vulnerable for ship impact. Usually both the gates as well as the supports
are damaged severely. Even in opened condition the door might suffer from ship impact. By their more
sturdy structure, rolling or sliding gates are better equipped to resist vessel impact. As an additional
advantage a rolling or sliding gate cannot be damaged when it is opened because at that time the gate
is completely protected in its recess.

h. Inspection, maintenance and repair:

Moving or sliding parts of a gate are subjected to quite some wear and tear, therefore regular
inspection, maintenance work and repair are necessary. Lift gates are lifted out of the water frequently,
which offers good opportunities for inspection, small maintenance and minor repairs. Rolling or sliding
gates can temporarily be set in dry in their recesses; the recess functions as a temporary dry-dock in
that case. However, to keep the navigation lock operational it will be necessary to provide two gates
per lock head. For inspection and repair on the rails or sliding track the navigation lock is locally
dewatered by using a steel watertight structure which is open on the bottom and functions as a kind of
diving bell. Mitre gates are usually lifted out of the water for maintenance and repair; this requires a
double set of doors if the lock has to remain operational. When inspecting or repairing the lock head a
maintenance limpet (inspectie kolk) is used to inspect the pivot, pindle, quoins, pivot shoe or
supporting frame. A limpet can have a box shape and is put in position using a mobile crane and is
ready to operate after the water is pumped out. Before positioning the limpet the canal bed must be
cleared of objects that might obstruct the sealing of the limpet.

4.3 Lock Head

4.3.1 Functions and requirements
Design of a lock head goes hand in hand with design of the gate. Full reference is made to section 4.2,
especially to the illustrations showing lock heads, for a better understanding of this section.

The lock head has three main functions:

1. Accommodate the gate and whatever is necessary for opening and closing the gate:
The lock head needs recesses for the gate in open position. The recess must be designed in such a
way that collision with the gate is prevented as much as possible when ships are passing. Furthermore
all the equipment and other items to operate the gate need space or a place in the lock head structure.

 Hydraulic Engineering
Faculty of Civil Engineering 104 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

2. Retention of water, sealing, prevention of seepage (groundwater):

On a general level, being part of the lock implies the lock heads have to retain water. On a detailed
level this results in design of sealing ‘between’ the gate and the lock head structure, see section 4.2.4.
Besides the previous, ground water flow underneath and alongside the lock head structure must be
avoided, not only for quantitative or qualitative water management reasons, but to prevent erosion and
piping of the structures as well. Therefore, lock head and lock chamber have to be connected
watertight and under and besides the lock head cut-off screens may have to be constructed.
3. Load transfer:
When the lock gates are closed the lock heads have to transmit the resulting hydrostatic load via the
walls and floor to the foundation, without too much of deformation.

Below, some of the principal design requirements for the lock heads are derived from the functions, others
are the result of operational considerations or boundary conditions.

Choice of filling and emptying system

The lock will be filled and emptied either by a head filling system, or by a longitudinal filling system. The
choice of a filling and emptying system may pose strict constraints on the lock head design and has been
described in section 4.1.

Safety of the flood or storm defense system

The lock may be part of the coastal defense or the flood protection system. This will definitely determine
the Top of Structure level of the lock head(s) and will also contribute to a positive decision on providing the
lock head with a second or back-up gate. The word contribute has been chosen deliberately for two
reasons. First, there is the much more economical alternative of using stop-logs, see next paragraph,
second, other reasons, e.g. operational security or the maintenance of gates, will have to be used as well
to justify the costs of a spare gate.

Need of Stop-log recesses

These are vertical slots in the lock head walls, positioned in front and behind the gate. In these recesses
horizontal beams, so-called stop-logs (schotbalken), can be lowered; piled on top of each other a retaining
wall is the result. Traditionally stop-logs are wooden beams, nowadays steel stop-logs are often used for
larger spans and for larger water head differences. Stop-logs serve the following purposes:
1. Back-up of the primary gate:
See previous paragraph.
2. Emptying the lock chamber:
When the whole lock chamber has to be emptied and the gates can not be used for this purpose,
because they are in maintenance as well, the stop-logs can be used to create a temporary retaining
wall. The most common solution is to apply two rows of stop-logs at a distance of a few decimeters
and fill up the gap with clay in order to create a watertight structure.
3. Protection of the closing structures:
This may be needed, for instance, to protect the closing structure against floating ice.

Equipment to move the gate:

Obviously the equipment, and installations, for gates will be completely different depending on the gate
type, but it will be installed in or onto the lock heads. This will govern the design of the lock heads in
varying extent, since the forces in a heavy lift gate and the provisions to be made; will be completely
different from those needed for a wooden mitre gate.

 Hydraulic Engineering
Faculty of Civil Engineering 105 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

4.3.2 Type of lock heads

Due to the large structural stiffness that is required and the large forces, e.g. hydrostatic water pressure,
horizontal soil pressure and concentrated loads caused by operating the gate and the equipment or
machinery, the design is in most cases based on an open concrete box or U-profile (bakprofiel). Usually a
monolithic structure is chosen in which the floor is rigidly connected with the walls as can be seen in
Figure 4-65. By far the most lock heads have a U-shaped structure to enclose the required cross sectional
water area with as little as concrete as possible.

Figure 4-65 Cross section of a monolithic U-shaped structure for a lock head
As mentioned before the standard U-shape will transform in more complicated 3D shapes because
recesses for the gates have to be provided and enclosed spaces for machinery and equipment.

Depending on the gate type, the construction method and local conditions variations to this theme are
designed and constructed. In Table 4-6 a number of alternatives are shown with description of the
situations where or when to apply them, for a lock head with mitre gates. It may help to look at the Figures
in the previous section on gates to develop similar ideas for lock heads for other gate types.

Table 4-6 Lock head alternatives

Description Application area
1 Standard U- This type of lock head is still used frequently, because
shaped lock it is simple to construct and it is a very rigid structure.
head build in-situ All the other alternatives are based on this U-shaped
structure.
2 Pneumatically The lock head is constructed at ground level and
immersed lock pneumatically immersed to its operational depth. The
head construction alternative is used when only little
horizontal area is available (no building pit needed).
3 Stripped U- The stripped lock head is built in a dry dock, then
shaped head, transported and immersed on a prepared gravel bed at
floated in its final destination. After ballasting with concrete the
head will be finished. This alternative will be applied
when building activity at the lock location is not desired
and a short construction period is required.
4 U-shape lock This lock head is only applied for very large sea locks,
head with a thus only when the width of the floor is relatively large
construction joint compared to the wall height. Apart from lower internal
in the middle forces the hinge in the floor (only) has disadvantages
regarding construction, different rotations at the joint,
leakages etc..
5 U-shape lock This alternative is a variation of alternative 4, the
head with two difference is the use of two joints instead of one. The
construction same (dis)advantages apply.
joints

 Hydraulic Engineering
Faculty of Civil Engineering 106 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Besides the normal lock head alternatives in Table 4-6, Table 4-7 represents two flexible lock head
alternatives. The purpose of a flexible lock head is to be relative easy extendable. This way, with minor
structural adjustments, the lock will be capable of serving the shipping traffic for its whole structural
lifespan. In general locks become already too small in 25-50 years, while a normal lock structure as a
lifespan of about 100 years. Thus, this type of locks is from a Life Cycle Management (LCM) point of view
a good alternative for lock (re)construction.

Table 4-7 Flexible lock head alternatives

Description Application area
1 Floatable lock The stripped lock head is built in a dry
head that is dock, then transported and immersed on a
floated in and can prepared gravel bed at its final destination.
be replaced quite After ballasting with ‘sand’ the head will be
easy finished. When the head becomes too
small it is floated out and replaced by a
See 3 in larger lock head of any width and/or depth
that is required.
2 Extendable lock For this flexible head a standard U-shaped
head; to be lock head is constructed. However in this
widened by head a demolishable wall is placed, which
demolishing the initially reduces the width of the head.
temporary wall When the width of the original U-shaped
structure is required the wall can be
demolished and the gates replaced.

4.3.3 Structural engineering aspects

The most important structural engineering aspect in the design of the lock head is the check on the
stability strength and stiffness of the structure. For stability the lock head, or any other structure, is
considered as a monolythical thing on a foundation. Checking the stability of the structure translates into
checking the strength of the soil under the structure. For the check on the strength and stiffness of the
structure itself a more detailed look is required and the elements of the structure, not the structure as a
whole, are taken into consideration. Of course the properties of the construction materials, often concrete
and steel, come into play as well.
Note that stability and strength of the structure have to be checked for ULS conditions, using load factors
to obtain the required safety, whilst in SLS deformations are checked without the use of load factors. In
the remainder of this Section, stability, strength and stiffness will be further elaborated without making
distinction of ULS and SLS, without the use of load factors.

Stability
For overall stability the following failure mechanisms must be checked:

Macro stability: Micro stability:

1. Bearing capacity (vertical) ΣV = 0 5. Piping (onder- en achterloopsheid)
2. Lateral bearing capacity ΣH = 0
3. Overturning - No tension ΣM = 0
4. Overturning – Redistribution & σ max ΣM = 0

In the listing above ‘Overturning’ has been mentioned twice. The check on overturning using the criterion
of no tensile stresses under bottom of structure is more conservative, i.e. results in bigger foundations,
than the criterion for overturning where redistribution of stresses is allowed, provided the maximum
bearing capacity of the subsoil is not exceeded.

 Hydraulic Engineering
Faculty of Civil Engineering 107 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

1. Bearing capacity (vertical)

When a lock is constructed besides the waterway, on average more weight is removed, excavating the
soil above the anticipated level of bottom of Structure, than added due to the weight of the concrete
structure. The exception to this general rule is the lock head structure, where bigger volumes of concrete,
for instance for lift towers or the walls around the recess for a rolling gate may increase the downward
load on foundation level. Constructed in the waterway, usually there is a surcharge on the bottom
foundation level all over the structure, but this may be quite limited. Most frequently even the soil
underneath a lock head has enough vertical bearing capacity to support the head with a shallow
foundation. A first, rough estimate to be used for the bearing capacity of a shallow foundation would be
2
about 300 kN/m . Of course a Brinch Hansen calculation to determine a more accurate figure for the
bearing capacity has to be made.

If a pile foundation is necessary an initial pile bearing capacity of 1000 kN per □ 450 pile (square 450 x
2
450 mm pile), compression, is a safe figure to work with, provided the pile toe is in a layer where the cone
2
value is around 10 N/mm (CPT-value). Generally the use of the Koppenjan formulas for pile bearing
capacity results in higher figures.

The upward groundwater pressure is a very important parameter because it reduces Σ Vs, the resulting
vertical downward force. In a first, strongly simplified, approximation the decrease in piezometric level can
be assumed to vary linearly with the position along the longitudinal axis of the construction. When vertical
seepage cut-off screens (schermen tegen onderloopsheid) are applied, the seepage length can be
approximated as sketched in Figure 4-66; a 2D-approach is used here. With total seepage length the
formula, shown in the Figure, for the upward pressure at a given x-coordinate can be found.

Figure 4-66 Upward ground water pressure

The true pressures will have to be determined by drawing a square net or by using a finite element flow
model; see the Manual 'Ground water flow' for an example. From the above it will be clear that the
locations of the seepage cut-off screens are an important factor in the determination of the upward
pressure under the lock floor, hence, in the stability of the lock head.
Keep in mind that short time intervals of high or low water levels will hardly have any noticeable effect on
the upward ground water pressure.

Note:
− The soil pressure under the structure remains the same from left (h1) to right (h2) whether the
piezometric level, read the water pressure, varies or not, assuming the dead weight of the structure is
nicely distributed. The latter also depends on the stiffness of the whole structure, lock heads and lock
chamber. Since the soil pressure is equal to the sum of the effective stress plus the water pressure
(σsoil = σ’eff + σw), an increase in water pressure reduces the effective stress.
− A reduced effective stress requires less vertical bearing capacity of the soil. Water carries a part of
the load. However, less effective stress reduces the lateral or horizontal bearing capacity, see next
paragraph.

 Hydraulic Engineering
Faculty of Civil Engineering 108 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

2. Horizontal or lateral bearing capacity

In Figure 4-67 the main loads on a lock head are sketched. Two
resultant forces can be distinguished:
Σ H: The resultant force of the horizontal water pressure on
the gates and the lock head. In the direction perpendicular
to this Figure e.g. soil pressures have to be added.
Σ V: The resultant force of the vertical forces, being: the dead
weight of the lock head and gate(s), the weight machinery
or equipment, the weight of the water above the lock floor; Figure 4-67 Main (hydrostatic)
reduced by the upward (groundwater) pressure. loads on a lock head

The stability requirement against sliding is: Σ H < f.Σ V

When the forces ΣH and ΣV are divided by the surface area A
and f is replaced by tan(δ ) the following formula results: τ < tan(δ ).σ n' with: δ = 2 φ
3
In this expression, σ’n is the effective soil stress (korrelspanning), φ is the angle of internal friction of the
subsoil. In Ultimate Limit State appropriate load factors and material factors have to be taken into account.

Overturning in general
After checking the vertical and horizontal equilibrium, the structure has to be checked on overturning.
Considering shallow foundations, soil pressures, generally linearly varying, have to provide the resisting or
stabilizing moment. Similar to the checks on vertical and horizontal stability, and using the already
computed results of those checks, the resultants of the vertical and horizontal forces are calculated. The
overturning moment will be computed using the arm between Bottom of Structure (BoS) and the working
line of the horizontal resultant, and the arm or eccentricities of the vertical force(s). Instead of continuing
the analysis with an overturning moment M, the calculated moment is translated into the resultant vertical
force and an eccentricity e of this force. The multiplication ΣV.e has to equal ΣM.

Figure 4-68 Resulting ΣM, ΣH, ΣV and etot

Note:
In construction mechanics courses the point of origin used to do calculations is often chosen at
the utmost right hand side and in the centre line of the structure or centre line of a part of it. From
a mathematical or systematical point of view this is a wise decision because it avoids most of the
confusion about plus-minus signs to be used for forces and their contributions to the moments,
which depends on direction and position.
In hydraulic engineering it is more convenient to select the centre and bottom of the structure as
the point of origin to check overturning stability, see Figure 4-68 and further explanations
hereafter. Often the use of the centre bottom point of the structure has the advantage that the
large dead weight force does not contribute to the moment.

In Figure 4-69 the effect of an increasing eccentricity, from left to right, on the soil pressures under the
bottom slab is shown, assuming a homogeneous soil body of constant stiffness under the structure (the
stiffness may vary from layer to layer though). Alternatively the force Vs could be larger, or both the

 Hydraulic Engineering
Faculty of Civil Engineering 109 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

eccentricity and force could increase. As explained a larger force Vs is the result of a larger overturning
moment due to either larger forces or larger eccentricities (in order not to cloud the Figure, M and H have
not been drawn).

In the middle and right-hand side illustration of Figure 4-69, the largest effective stress σeff under the
foundation is calculated as follows:
1. the stabilizing or resisting force Vr is working opposite and in the same working line as the
soliciting force Vs
2. knowing the eccentricity e of the force Vs, the distance a can be determined
3. Vr is the resultant force of the soil pressures, which are assumed to vary linearly. Given the linear
variation of soil pressures, there has to be a triangular pressure distribution. The work line of Vr
has to run through the centre of gravity of the triangle, hence, a is equal to ⅓ of the base of
triangle. As a result Vr equals ½ * σeff * 3a; with some rearranging σeff is ⅔Vr/a.

Figure 4-69 Effect of increasing eccentricity on soil pressure distribution

3. Overturning – No tension
When Vr is the equal opposite in the same working line as Vs the structure is stable; there is no danger for
overturning. However, often the requirement that the whole base has to remain in contact with the soil
beneath it will be put forward. This demand can be presented in formulas as follows for foundations with a
rectangular footprint:
ΣV ΣMs,tot ΣV ΣV
σ eff
'
= − s,tot − 2
= − s,tot − s,tot ≥0 hence : e ≤ 16 b
b b 1 b2
6b
1
6

The above mentioned requirement could be popularly phrased as: ‘No tension allowed’ under the base
slab of the structure’. To suffice to this requirement the vertical component of the resulting force has to
stay within the core of the foundation, e has to smaller than 1/6 of the base.
Obviously it has to checked that the effective stress σ’eff is not exceeding the maximum bearing capacity
of the (sub)soil, which can be determined using the Brinch Hansen.

4. Overturning – Redistribution
As illustrated in Figure 4-68 the overturning moment, thus the eccentricity and/or vertical force, may be
very large. In steel material this would result in compression ‘and’ tension stresses, but the only possible
force interaction between the bottom of the structure and the soil is a compression pressure, tension is not
possible due to the nature of soil. The tension pressures that would be needed to resist the overturning
moment are redistributed. As a matter of fact there are no ‘tension pressures’, but there will be increased
compressive stresses under the part of the base that remains in contact with the soil.

Again a check has to be made on the effective stress σ’eff, which should not exceed the maximum bearing
capacity of the (sub)soil. Compared to the similar check when the ‘No tension’ requirement is used, there

 Hydraulic Engineering
Faculty of Civil Engineering 110 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

is a higher chance on failure, i.e. the bearing capacity being too small for the required effective stress,
because the stresses under the effectively working part of the base are definitely higher.

Some observations on the above stability checks:

• The higher the resulting effective stress under the structure in ULS checks, the higher the
stresses that have to be used doing the SLS checks on (differential) displacements, although
without load factors. Especially the situation with redistribution of stresses for overturning may
become critical in SLS as well.
• The lock head length or width, do not forget to check the other direction, required for stability can
be found by iteration through the above stability checks. By the way, the final length of the lock
heads will be the result of either the stability requirements, or the required length for gate
recesses, or other functional or construction requirements.
• It is important to make a distinction between the upper lock head and the lower lock head, taking
the generally higher water levels ‘outside’ as a starting point. Most probably, horizontal water
pressure on the upper lock head is partly absorbed by the lock chamber behind the head. The
lower lock head has no other structure behind its back, when the water is high in the lock
chamber, so this situation is less favourable.
For sea locks the situation is a little different, because in this case the design level for the upper
head is storm surge level, and the design level for the lower head is only HLL. So in this case the
loads on the lower head are less. Of course this situation changes when the lower head also
functions as a secondary sea defence.

5. Piping (onder- en achterloopsheid)

Piping is the process of (ground)water flow, in the soil under and/or around the structure, having enough
transport capacity to erode soil particles. As soon as the first particles are carried away, more water will be
able to flow along the same route because of the reduced flow resistance and more soil particles will be
washed away. The process strengthens itself and a pipe will develop. This will undermine the structure
and, in the end, result in instability of the whole structure. To evaluate piping, the formulas of Bligh and
Lane (ref Manual § 20.3 "piping") can be used.

Piping is influenced by the type of lock chamber that is applied: is it a closed basin or has a permeable
chamber floor, for instance a filter and rubble stone protection, been used? In the first case the piping
length (kwellengte) is related to the full length of the lock, in the second case the piping length is related to
the length of the lock head only. The second case results in a larger hydraulic gradient and a greater risk
of piping.

Strength and stiffness

For the check on the strength and stiffness of the structure a more detailed look at the structure itself, and
the constituting elements is required. Nevertheless, the work starts with the same forces used for the
stability analysis or checks; now the flow of forces within the structure, not the force transfer into the
subsoil has to be determined and checked. To do so, the properties of the construction materials, often
concrete and steel, have to be considered.

From a structural engineering point of view lock heads are structures that should be designed using 3D
models. 2D schematisation is less meaningful because of the truly 3D shape of the lock head, see e.g.
Figure 4-54. Also forces are acting along all 3 axes, for instance ‘free’ water pressures acting along the
lock axis, soil and ground water pressures perpendicular to the same axis. The forces and the resulting
load combinations to be analysed for the head and the spreading of these forces through the 3D structure
is quite laborious if not complex. Nonetheless, starting from scratch 2D approximations can and will be
used to arrive at a first conceptual design for the 3D lock head, but in following design loops the 3D FEM
models will be used to do the final checks on strength and stiffness.

 Hydraulic Engineering
Faculty of Civil Engineering 111 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

4.3.4 Construction aspects

There are several construction methods for lock heads; this paragraph will deal with some of them, taking
in-situ construction of a standard U-shaped lock head as a starting point.

A large part of the lock head has to be constructed below the existing ground water level. Consequently a
construction pit or braced excavation is needed, to construct the lock head. Design of the construction pit
or braced excavation depends on several conditions at the lock location:
• Available space for construction (horizontal)
• Strength of the soil
• Permeability of the soil in combination with (differences in) piezometric levels or groundwater
head(s)
Based on these conditions some construction methods and the temporary structures required are
mentioned in Table 4-8. All the options need dewatering to a certain extend to keep the building pit dry.

Table 4-8 Characteristics of some construction methods for a U-shaped lock head
Slope combined with a seepage screen Sheet pile wall
Deep
impermeable
layer

This option is selected when there is This option is selected, when there is an
enough building space and an impermeable layer, but not so much space
impermeable layer is available to build the lock head
Underwater
concrete with
tensions piles

Not possible

This type of building pit is constructed

when not much space is available and a
thin layer of underwater concrete is
desired.
Under water
concrete
without
tensions piles

Not possible

The same as above, but no tension piles

are required, because the underwater
concrete is made thick enough to resist the
upward water pressure.

 Hydraulic Engineering
Faculty of Civil Engineering 112 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

4.4 Lock Chamber

Although a 3D structure, lock chamber design generally is based on analysis of the 2D cross section.
Therefore, designing a lock chamber is less complex than designing the lock heads. The two main
components are the lock chamber walls and the lock chamber floor. In the next two sections the different
types of walls and floors are described and in the third section the combination will be discussed. Finally,
lock chamber types and selection criteria are listed.

4.4.1 Lock walls

The lock chamber wall has several functions:
- soil and water retention;
- guiding the vessel through the navigation lock;
- mooring the vessel during the F/E process.

Important for the preliminary design are the forces due to horizontal earth pressure, generally the
governing load, water pressure, deadweight, surcharge load, hawser forces and ship's impacts. The wall
has to be designed such that overall stability is guaranteed (∑M=∑V=∑H=0) and that it is able to resist all
the forces (material strength > load). Stability and strength are generally considered in the Ultimate Limit
State (ULS).

In addition, in Serviceability Limit State (SLS), to these forces the lock walls are subjected to (tensile)
stresses induced by temperature. These stresses result in cracks. A way to prevent this problem is the
use of expansion joints (dilatatievoeg). Depending on the wall thickness these joints are positioned at a
distances of approximately 25 meter.

The horizontal forces can be resisted in several ways, e.g. by means of:
- a gravity structure
- a retaining wall structure
- a wall combined with a deep foundation
The vertical forces will be taken up by the soil, either more uniformly distributed by a shallow foundation
or more concentrated by using a deep or pile foundation. In following paragraphs several wall structures
will be discussed in more detail.

Gravity structure (uniformly distributed bearing):

The worklines of the resulting horizontal

and vertical force usually do not go
through the centre of gravity of the
structure, hence, there is an overturning
moment. Stability is obtained by a linearly
varying distributed soil pressure at the foot
of the structure, see Figure 4-70 for the
resulting soil pressures. A safe,
conservative condition for stability would
be to demand that the workline of the
resulting external force lies within the core
of the footing. In that case the whole
footing remains in contact with the soil and
Figure 4-70 Forces on a gravity structure
no major redistribution of soil pressure is
necessary to find equilibrium.

The vertical dead weight forces G and V are reduced by the upward water pressure (not shown in Figure
4-70). The resulting vertical downward force has to be resisted by the soil pressure O beneath the wall.

 Hydraulic Engineering
Faculty of Civil Engineering 113 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

The horizontal force H must be resisted by soil friction W at the bottom of the structure. The following must
be valid:
W >H Where: f = friction factor
W = f (G + V ) G = dead weight of the structure
V = weight of the soil
To provide a large enough soil friction force W, care has to be taken of:

a. Sufficient friction between structure and soil. Make sure that the bottom surface of the wall is rougher
than the soil. The formula f = tan(δ) can be applied. For neutral soil pressures δ≈½φ, whilst for sheet
pile walls δ≈⅔φ, where φ is the angle of internal friction. For a smooth wall f is considerably smaller;
use a δ << /3φ.
2

b. Enough deadweight G. This may lead to larger dimensions of the structure; alternatively heavier
materials may be used, e.g. natural stone. (Rock obtained during excavation can be used in the
concrete mix).
c. A large external force V, if the above measures were not sufficient. The reinforced concrete wall
structure may be designed and constructed as an L-shaped retaining wall; the bottom slab large
enough to be loaded with sufficient deadweight force V.

Note:
- Horizontal support for the wall may be partially or fully produced by the lock floor structure, see
section 4.4.3;
- Typically, gravity structures are structures on or with a shallow foundation.

Retaining wall structure:

For this type of structure, sheet pile

walls, combi-walls and diaphragm
walls can be used. Sheet pile walls
and combi-wall are made of steel H
profiles, U-shaped or Z-shaped sheet
piles or steel piles (circular cross
section), whilst diaphragm walls are
made of reinforced concrete. Figure
4-71 presents an anchored steel sheet
pile retaining wall.

The wall is schematised as a simply

Figure 4-71 Sheet pile structure
supported beam; a girder on two
supports. The horizontal force H, the resultant of active soil pressure, will be resisted by the anchor force
A and the passive soil pressures B. Most of the vertical force will be carried away by soil, part of it
increases the horizontal earth pressure and ‘creates’ a vertical friction force along the wall. Toe resistance
of the sheet pile will make up the difference between active (down) and passive friction (up).

To reduce the bending moments in the wall, several measures are possible:

a. Put the anchor at an efficient position. By moving the anchor down, the span reduces and the span
moment decreases as well. An optimum level of the anchor results in a span moment has the same
order as the support moments.
b. Use more than one anchor.
c. If B is not the result of passive soil pressure, but the reaction force of another structure, B should be
positioned as high as possible, thus reducing the span. B is obviously often supplied by the bottom of

 Hydraulic Engineering
Faculty of Civil Engineering 114 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

the lock chamber. The lock bottom slab and opposite lock wall form a system, which may be
optimised, see section 4.4.3.
d. Drive the sheet pile deeper into the soil; instead of a simply supported beam, a beam with a simple
and a fixed support, is a better schematisation for this situation. The anchor is still a simple support,
however, deeper down there is a fixed support. The sheet pile will rotate around point S. Due to the
rotation a passive soil force C is being introduced. Combined with force B the sheet pile is fixed in the
soil; the wall structure is statically indeterminate. The anchor force could be considered obsolete, not
necessary for the stability anymore, however, the anchor generally is not omitted in order to limit the
bending moments and to limit the deformations of the wall.

Note:
- When steel sheet piles are used, they will be covered with wooden posts, or concrete slabs, and
fender beams (regels), to prevent corrosion and damage to wall and ships.
- For steel structures, considering maintenance, preventing corrosion is the key issue. Usually steel
sheet piles have a shorter life span than concrete diaphragm walls.

Wall combined with a deep foundation

If the bearing layer in the subsoil is

located at a very deep level and the
lock chamber has a considerable
depth, which results in a large retaining
height, the solution shown in Figure
4-72 can be used. This usually applies
to sea or coastal locks. In the past the
solution was used to by-pass the lessr
strength of the materials of that time.
Nowadays, this type of structure is
mainly used for quay-walls. The
horizontal and vertical forces of the top Figure 4-72 Retaining wall with a deep foundation
structure are transferred by the piles into the subsoil. The sheetpile wall may contribute to vertical
resistance, but its main function is to retain soil and water, hence it resists horizontal loads as well

4.4.2 Lock chamber floor

For design of the lock chamber floor (sluisvloer) the following functions have to be considered:

1. Water retention function – or not:

Retaining water implies resisting water pressure. A concrete floor is impermeable and will be loaded
by hydrostatic water loads. If the floor of the lock chamber is permeable, for instance because of the
natural soil bottom or use of a (rock) filter, there is a lesser chance on water pressure built up. The
type of floor influences the stability of the lock heads and the chamber walls. The permeable type of
floor has to be checked on:
- Ground-water flow towards the lock chamber, so during a high ground-water level and LSP inside
the chamber;
- Currents and turbulences caused by the filling and emptying system;
- Currents and turbulences caused by ship movement's en ship propellers (wash load).

2. Load transfer function:

A concrete floor, due to its stiffness, generally has considerable influence on the stability, hence the
flow of forces in the structures next to it. Either it has to be specifically designed not to attract these
forces, e.g. by using expansion joints in a concrete floor or use of a filter, or it has to be designed and
constructed to resist the forces.

 Hydraulic Engineering
Faculty of Civil Engineering 115 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

A distinction can be made in a water-tight floor, a permeable floor without support function and a
permeable floor that supports the lock walls. In general impermeable structures are rigid. Therefore scour
is not a threat. The three types mentioned will be discussed a bit further, focusing on loads and load
combinations.

Impermeable floor:

The impermeable floor has to be able to resist the two extreme load situations listed below. Usually the
second situation, and the resulting load combination, is governing structural design.

1. Low ground-water level outside and HLD (high locking datum, HSP) inside the lock chamber:
This results in a force in a vertical downward direction which, together with the dead weight of the
floor, has to be resisted by the soil underneath the structure. In case the soil has insufficient bearing
capacity for a shallow foundation piles have to be used.
2. High ground-water level outside and LLD (low locking datum, LSP) inside the lock chamber, or even
no water in the lock for maintenance reasons:
The resulting vertical load is equal to G times the difference between the piezometric level of the
ground-water under the floor and the height of the column of water above the floor, reduced with the
dead weight of the floor.

If the resulting vertical load on the floor is positive, upwards, then the following structural solutions could
be considered:
a. Increase the dead weight of the floor structure by applying a thicker floor or using heavier
material.
b. Supply the floor with anchors or piles. Duly consider the fact that the load on the piles may change
from tension to compression, and vice versa, caused by varying ground-water levels or the
filling/emptying of the lock chamber.
c. Connect the floor with the chamber walls. The walls provide support to the floor if there is a
resulting upward force on the floor; in the other case they transfer vertical downward to the floor.
d. A combination of the solutions a, b and c.

With a high load in upward direction (risk of floatation!) solution b is financially more attractive than
solution a, especially when tension piles are used. Solution c generally is a good option. Nowadays the
requirement that a lock has to be set dry is demanded less frequent than in the past.

Permeable floor, without supporting the lock chamber walls:

In a permeable floor the relation between the permeability of the floor kfl and of the underground kgr is of
importance. Consider the following situations:
1. kfl ≥ kgr
By using a flow net or a finite element program the ground-water flow can be calculated. The hydraulic
load due to ground water flow has to be combined with flows and turbulences, caused by ships or the
filling and emptying system. Next step is to investigate the stability of the upper layer of the floor
structure; the soil or the top stones of the filter have to be cohesive, heavy or large enough to resist
the hydraulic loads.
2. kfl < kgr
Especially when kfl << kgr (in that case the floor is almost impermeable) the total upward pressure has
to be resisted by the floor structure, the bottom soil or filter layers. In fact the situation agrees with a
clay layer above permeable sand layers: there is great risk of hydraulic burst-up. To prevent this, the
clay layer has to be thick enough (to provide a counter load) and has to be of homogeneous
constitution and thickness.

 Hydraulic Engineering
Faculty of Civil Engineering 116 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

If a filter is used as lock bottom, groundwater will flow into the lock chamber. If the filter is not working
properly water pressure will built up in the filter and in combination with other hydraulic loads scour will
occur, which might endanger the stability. To reduce groundwater flow or piping, short sheet pile walls
may be used to increase the seepage length, which results in a smaller exit velocity of the groundwater
flow.

Permeable floor, which supports the lock chamber walls:

This type of lock chamber floor is almost the same as the impermeable floor option, however, in this
option, holes are made in the floor and filled with gravel, so the water pressures under and above the floor
are equal. Consequently, the lock chamber cannot be dewatered, but it is protected against scour; piping
is an issue. If the holes in the floor are large, it might be better to describe the floor structure as a number
of beams, acting as struts stempels) between the walls, with a gravel filling besides or between them.
In any way, this floor type maintains the wall support function, so the lock chamber wall can be more
slender compared to a normal permeable floor.

4.4.3 Combination of lock chamber walls and floor

The natural soil or filter bottom is not considered here, because there is little to be combined from a
structural point of view; the walls will be designed as stand alone structures. What remains to be
discussed in this paragraph is the concrete lock chamber floor, either permeable or impermeable, in
combination with the lock walls.

A stand alone concrete floor of certain minimum thickness is impermeable, but may not be heavy enough
to resist upward water pressures. Increasing the thickness of the floor will solve this problem, usually at
the cost of a large amount of extra concrete and extra excavation with all the related problems. Very thick
floors are generally made of mass concrete (stamp beton) and structural reinforcement is ‘not’ used. (The
thinner the concrete floor, the more need there is to use reinforcement). As a result of the horizontal earth
pressures behind the walls a normal force (compression) will is introduced in the floor, vice versa the floor
provides a support force to the walls. Although stand alone, there is structural interaction.

Quite some (design) effort would be necessary to provide a watertight joint between the floor and the wall
structures. Even in this stand alone version, due to the normal load and floor-wall interaction, bending
moments may be introduced, which result in vertical forces on the wall. To prevent this, a watertight joint
could be designed, which transfers the normal forces but not the shear forces and moments. In a following
alternative the thick gravity floor is replaced by a reinforced concrete floor able to resist bending moments.
However, the bending moment is not transferred into the walls, normal and shear forces are, see Figure
4-73. Water tightness can be obtained by positioning rubber slabs at the contact surfaces between floor
and wall.

Figure 4-73 Lock floor as a reinforced concrete floor

Since the bending moment in the floor increases quadratically with increasing lock width, a possibility to
reduce this is the introduction of support moments on the floor. Hence, the joint connection develops into
a full connection stiff concrete joint between the walls and the floor. The continuous concrete cross
section, fully integrated or monolithic structure is the result, see Figure 4-74.

 Hydraulic Engineering
Faculty of Civil Engineering 117 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Figure 4-74 Upward forces with a reduced field moment

An additional advantage of this option is that the large weight of the wall is transferred via the floor to the
soil resulting in a more equal distributed load on the subsoil. There will be less problems with peak soil
pressures and settlements and displacements in the soil will be less.

In navigation locks in coastal areas the lock is not

only very wide but also very deep, due to the
vessel size.. A continuous floor with such a wide
lock is not very sensible due to the size of the
bending moment. Therefore, joints are often
applied which work as hinges and still are capable
to transfer normal and shear forces. This is
illustrated in Figure 4-75. Figure 4-75 Cross section of a navigation
lock with a hinge halfway

In the above the focus was on the cross section of the lock chamber; now some attention will be paid to the
longitudinal section.

Usually locks are built with expansion joints which separate the chamber into sections with a typical length
of 25-30 meter, see Figure 4-76. The expansion joints allow each block to move independently and
reduce stresses that otherwise might occur due to differential settlement or expansions due to concrete
hardening or other changes in temperature. The expansion joint needs to be water tight, to keep the water
in the lock and to prevent groundwater and soil from coming into the lock.

Figure 4-76 Standard lock with 25 m long Figure 4-77 Monolithic lock
elements and water stops

If the calculated differential settlements are above the limit, the only solution may be to construct the lock
chamber without expansion joints. This joint less monolithic structure results in a very robust structure that
can resist not only differential settlement in general but also local unexpected settlements (e.g. due to
scour), see Figure 4-77. It is not necessary to build the whole structure at once. Construction joints can be
used to separate different segments. Reinforcement can be carried through the construction joints.

A monolithic lock design results in increased internal stresses. Joints normally reduce these stresses by a
small translation or rotation, compressing the (rubber) expansion joint material. The extra internal stresses

 Hydraulic Engineering
Faculty of Civil Engineering 118 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

in monolithic design have to be combined with the other loads (water pressure, earth pressure,
deadweight, etc.) and the structure has to be designed for the Ultimate Limit State and the Serviceability
Limit State.

For the future, a properly designed and constructed monolith structure without joints can reduce
maintenance and eliminate major problems with leaking expansion joints that appeared on a regular basis
in the past. [PIANC 2009, Innovations in Navigation lock design]

4.4.4 Chamber selection

By making a choice of chamber several factors play a part. The most important ones are:
a. Initial costs, life cycle costs;
b. Width of the lock chamber, height of the walls and their mutual relation;
c. Water levels, including groundwater level;
d. Condition of the soil; such as bearing capacity and the permeability of the bottom;
e. Construction requirements: such as the dimensions of the available construction site and the
permissibility of groundwater lowering during construction;
f. Availability of material and equipment.

Table 4-9 Wall-floor combinations in a lock chamber

Floor A. Impermeable floor, A. Impermeable floor, with B. Permeable floor
Wall gravity floor (C*) tension piles (C*)
Gravity
structure
No additional value

Retaining wall
structure

Retaining wall
with a deep
foundation

Continuous
concrete cross
section No additional value No additional value

One joint
concrete cross
section No additional value No additional value

C* : Floor option C has the same layout as floor option A, but the lock floor contains holes, so no large
water pressures differences will occur.

 Hydraulic Engineering
Faculty of Civil Engineering 119 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

4.5 Seepage cut-off screens

The hydraulic gradient over the lock will initiate a groundwater flow underneath and alongside the lock
structure (onder- en achterloopsheid). If the groundwater flow or seepage becomes strong enough to
transport soil particles, erosion endangers the stability of the lock. This failure mechanism, referred to as
piping, because the eroded voids are pipe shaped, is illustrated in Figure 4-78.

Figure 4-78 Failure mechanism due to piping

To prevent piping the following could be done:

• Construction of a geometrically closed filter on the bottom and on the slopes at the downstream side
of the structure.
• Reducing the permeability of the soil. This would imply the use of injection techniques, which is usually
expensive.
• Extending the seepage length. This can be done by:
− Enlarging the lock itself, which would be quite expensive;
− Applying vertical seepage screens under and beside the lock, perpendicular to the longitudinal
axis. The screens cut off the groundwater flow, thus, the name cut-off screens. This is the most
common solution.

The cut-off screens under and besides the lock must be positioned in the same vertical plane; in this way
they form a continuous flange or collar around the structure, in the ground. The top of the screens besides
the lock will b slightly above the local groundwater level. Cut-off screens are constructed from sheet piles
and their length is determined by a hydraulic study. The minimum width of the screens must be equal to
the width of the construction pit used for construction of the lock, because backfilled soil is generally more
loosely packed than the undisturbed soil. For this reason the permeability will be higher and piping a
greater risk.

Leakage through the sheet pile wall must be avoided, so during the pile driving attention must be paid to
the correct interlocking of the sheets.

The connection between the sheet pile wall and the lock floor must be and must remain impermeable.
This may be problematic under the lock bottom. When the lock design is based on a pile foundation, the
soil underneath the floor may settle a little with respect to the lock itself. The embedded cut–off screen will
settle along with it, which results in a gap between the screen and the lock floor, the pipe that has to be
prevented. The solution is to embed or anchor the top of the sheet piling into the floor.
An additional number of small cut-off screens in between the main screens may be necessary to close the
gap under a pile foundation lock. These screens can be relatively short, about one meter, since their only
function is to seal the gap due to settlement.

 Hydraulic Engineering
Faculty of Civil Engineering 120 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

When the lock has a shallow foundation a different problem occurs. In this case, settlement of the subsoil
may cause the lock floor to settle more than the sheet piles, so the lock floor will effectively be resting on
the sheet pile wall, introducing undesirable bending moments in the lock floor. This situation can be
avoided by installing a compressible material, e.g. cork or foam plastic, at the top of the sheet piling, see
the sketch for the detail in Figure 4-79

Seepage under the downstream lockhead

Seepage under the upstream lockhead

Figure 4-79 Connection cut-off Figure 4-80 Position of the seepage screens in case of an
screen to the lock floor permeable lock chamber floor

The connection between the side screens and the lock wall is usually obtained by casting part of a single
sheet pile into the concrete wall itself. The lock of this sheet pile then sticks out from the wall and forms
the starting point for the rest of the sheet pile screen.

It must be noted that all the above considerations only apply to locks with impermeable floors. When floors
with filter structures are applied, the seepage screens will have to be positioned under the lock heads as
can be seen in Figure 4-80.

The position of cut-off screens influences the upward groundwater pressure. The best location for a cut-
off screen is at the upstream side of the upper lock head. When the hydraulic gradients are large the
seepage length must be extended further and a second screen must be applied. The best position for this
second screen is at the upstream side of the lower lock head. Especially the lower head benefits from a
reduction in upward pressure because it is not supported by an adjacent lock chamber. When the lock is
constructed in a tidal area and the hydraulic gradient changes direction with the tide it is common to use
two seepage screens, one on each side of the lock, even when the gradients are small.

4.6 Lock approach structures

Berthing facilities have to be provided for ships having to wait for the next locking opportunity. Approach or
guidance works are structures that guide the ships quickly and safely into the lock chamber upon entering;
safe for the ship and the lock. One possible layout of berthing facilities and guidance works, typical for
North-Western Europe, is depicted in Figure 4-81.

For proper guidance to entering ships, the angle α must be small. However, a small α means a large distance
between the berthing facilities and the lock itself, which leads to a longer lock cycle and smaller lock
capacities. A good compromise can only be found from practical experience and model studies. In the
Netherlands this has lead to a value of tan(α) = 1:6 for inland waterway locks that are mainly used by self-
propelled cargo ships, not by push convoys combinations.

Ships that leave the lock must not be hindered by waiting ships, so the berthing facilities must be
constructed away from the line through the lock walls (see dashed line in Figure 4-81). It is recommended
to allow for some clearance between this line and the sides of the waiting ships. This will increase the
sailing velocities of the outgoing ships, reduce the locking times and improve the capacity of the lock. The
dimensions of the berthing area ( W x L ) are usually equal to the dimensions of the lock chamber ( w x l ).

 Hydraulic Engineering
Faculty of Civil Engineering 121 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

In this way all ships that will enter for the next locking are at the shortest possible distance from the lock.
When a large supply of ships is expected the berthing area can be extended.

Lst = Slowing down length (2,5 times the length of Lc = Length of the lock chamber
the design ship) S = Safety margin in between the waiting areas
W c= Width of the lock chamber and the fairway
Figure 4-81 Design of the lock approach

Locks that are also used by push convoys are usually only marginally wider than the convoy (e.g. lock
chamber width 24 m against a convoy width of 22.80 m). In this case a slightly bent and slowly narrowing
guidance wall is applied to improve safety. It must be understood that a push convoy is difficult to
manoeuvre because of its large mass (a four-unit push convoy weighs 8800 tons, compared to 1350 tons
for a common Europe class vessel).

Berthing facilities near sea locks are usually not constructed as a single continuous area, but as a series
of detached berths. In this way the tugboats, that may be needed to assist large sea-going vessels with
entering or leaving the lock, have more manoeuvring space. In this situation the guidance wall will be a
continuous structure because of its function.

The main design requirements can be

deducted from the definitions that were
given in the introduction:
• The structure must be able to absorb
the impact of a ship without damage to
the ship or the structure. (both berthing
facilities and guidance works)
• The structure must provide a sheltered
waiting place for ships, hence, must
resist the mooring forces caused by
waiting ships (berthing facilities)
• The structure must be designed in such
a way that ships can sail alongside
smoothly. (both berthing facilities and
guidance works)
• When needed, there must be a
connection between the ship and the
river bank to enable the ship's crew to
go ashore (berthing facilities)

The first requirement determines the

magnitude of the forces that must be
absorbed. The third requirement almost
inevitably leads to the design of a
Figure 4-82 Bird's eye view of a lock approach
 Hydraulic Engineering
Faculty of Civil Engineering 122 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

continuous structure. The last requirement shows that additional facilities (e.g. gangways, ladders etc.)
may be needed.

Loads:

Ships can cause two types of loads on the structure:

• Berthing or collision forces (Aanvaarbelasting)
• Mooring forces (Afmeerbelasting)

For berthing or collision forces a flexible structure would be best because larger displacements dissipate
more kinetic energy. For mooring forces a stiff structure would be the best solution.

So, in a way, the impact forces and the mooring forces require contrary solutions. For large ships this has
lead to the development of separated structures: berthing dolphins to absorb impact forces and mooring
dolphins.

Structural alternatives

The above analysis leads to three observations:

• the main design problem is the absorption of a horizontal force
• flexible structures are attractive
• continuous structures are preferred

A common solution in Europe that satisfies these three observations is depicted in Figure 4-83 below.

Figure 4-83 Steel guard wall Figure 4-84 Guidance walls with floating
girders

This solution consists of a number of vertical steel piles that have been driven into the subsoil to provide
the desired degree of resistance. These piles are connected by one or more horizontal steel girders. The
number of girders depends on the possible water level variations. A hardwood fender is constructed in
front of this girder because wood provides a better material to guide ships than steel, because it is more
flexible. When the water level variations are large the fixed steel girders can be replaced by floating
structures, see Figure 4-84

Sometimes sheet pile walls are used as guidance works. These are very stiff constructions, so to avoid
damage to ships a more flexible structure must be placed in front of the sheet piles, for instance a
framework made of azobé (type of wood).

In the USA approach walls are used quite differently compared to Europe. The vessels, in the USA
generally large push convoys, keep in touch with the approach wall, ‘slide or slip’ along the wall, while they

 Hydraulic Engineering
Faculty of Civil Engineering 123 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

are manoeuvring into the lock chamber. In Europe the vessels avoid every contact. The different
procedure is reflected in the layout of lock and approach structures. In Europe the guard wall has an angle
with lock axis, see Figure 4-81, in the USA there is no such angle and the face of the approach wall is
flush with the face of the lock chamber wall.

There is an explanation for the typical manoeuvring procedure in the USA. Often the navigation lock is
part of a weir complex in the river and constructed close to one of the river banks. In the approach zone of
the lock the flow pattern of the water is in the direction of the weir, where water discharges, adjacent to the
lock, see Figure 4-85.
The vessels are ‘pulled’ into this flow and
to maintain course, into the navigation
lock, quite some rudder and propeller
action is required. However, the vessels
can not use their engine power because
the push convoy has to be slowed down to
avoid collision with the lock. The approach
wall is a real guidance structure; it pushes
the convoys in the right direction and
prevents them from ending up in front of
the weir.

Figure 4-86 and Figure 4-87 show some

pictures of the floating approach wall for
the navigation lock in Olmsted – USA.
Figure 4-85 Birds-eye view Olmsted weir & lock - USA

Figure 4-86 Floating approach wall; Olmsted Figure 4-87 Nose pier of approach wall during
lock USA construction

 Hydraulic Engineering
Faculty of Civil Engineering 124 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

5. References

Most important literature in the field of navigation lock design:

- Ontwerp van Schutsluizen deel 1, Bouwdienst Rijkswaterstaat, (in Dutch; publication in English
available
- Ontwerp van Schutsluizen deel 2, Bouwdienst Rijkswaterstaat, (in Dutch)
- PIANC, Final Report of the International Commission for the study of Locks, Supplement to
Bulletin. 55, 1986
- PIANC, Innovations in navigation lock design, InCom-106, Brussels, 2009, (ISBN 978-2-87223-175-1)

Other literature:
th
- Handbook of hydraulic resistance, I.E. Idelchik, Begel House Inc., New York, 4 Revised and
augmented edition, 2007
- Fluid-structure interaction in case of waterhammer with cavitation, A.S. Tijsseling, Doctoral thesis
Delft University of Technology, 1993
- Elementary Hydraulics, J.F. Cruise, M.M. Sheriff, V.P. Singh, Thomas Nelson, Student edition,
2007
- Water-resources engineering, D.A. Chin, Pearson Prentice Hall, Upper Saddle River, New Jersey,
nd
2 Edition, 2006
- Pipeline design for water engineers, David Stephenson, Elsevier scientific publishing company,
Amsterdam, 1976
- EAU 2004, Recommendations of the committee for Waterfront Structures Harbours and
th
Waterways, Ernst&Sohn Verlag, Berlin, 8 Edition, 2006 (ISBN 3-433-01666-6)

Other Lecture Notes:

-
- Inleiding Waterbouwkunde, CT2320, K.G. Bezuyen, M.J.F. Stive, G.J.C. Vaes, J.K. Vrijling, T.J.
Zitman, maart 2009, VSSD (in Dutch)
- Ports and Terminals, CT4330, H. Ligteringen

Other PIANC reports:

- Standardization of ships and inland waterways for river/sea navigation, InCom-WG16, Brussels 1996
- Advanced and automated operation of locks and bridges, InCom-WG18, Brussels 1996
- Factors involved in standardising the dimensions of class Vb inland waterways (canals), InCom-
WG20, Brussels 1999
- Technical and economic problems of channel icing, InCom-WG23, Brussels 2004
- Developments in the automation and remote operation of locks and bridges, InCom-WG96,
Brussels 2008
- Guidelines for Berthing Structures, MarCom-WG48, Brussels (To be published 2009)
- Horizontal and vertical dimensions of fairways, MarCom-WG49, Brussels (To be published 2009)
- Life Cycle Management of Port Structures, Recommended Practice for Implementation, MarCom-
WG-103, Brussels 2008

Internet sites:
www.haringvlietsluizen.nl (October 2005)
www.dubo-centrum.nl/projecten/project.php?recID=103 (Oktober 2005)
www.keringhuis.nl/engels (January 2006)
www.canaux-historiques.com/d2m/ouvrage/montech/fiche_ouvrage/fo_00.html#
www.falkirk-wheel.com
http://en.wikipedia.org/wiki/Falkirk_Wheel
www.pancanal.com/esp/plan/estudios/0260-exec.pdf
www.pancanal.com/esp/plan/estudios/0276-01.pdf
www.pianc.org

 Hydraulic Engineering
Faculty of Civil Engineering 125 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

 Hydraulic Engineering
Faculty of Civil Engineering 126 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Appendices

A1 Inland waterway vessels, CEMT classes & transport capacity ............................................................129

A2 Water Management..............................................................................................................................130
A3 Ice control in locks................................................................................................................................136
A4 Case study:Naviduct in Enkhuizen .......................................................................................................139

 Hydraulic Engineering
Faculty of Civil Engineering 127 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

 Hydraulic Engineering
Faculty of Civil Engineering 128 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

A1 Inland waterway vessels, CEMT classes & transport capacity

CEMT Class I

CEMT Class II

CEMT Class III

CEMT Class VIb

≈ CEMT Class II

CEMT Class Va

CEMT Class Vb

CEMT Class Va

CEMT Class Va

≈ CEMT Class IV

 Hydraulic Engineering
Faculty of Civil Engineering 129 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

A2 Water Management
Water shortage and salt water intrusion are two locks, if no further constraints are applied. In special
important issues concerning water ‘resource’ situations, where the impact of the filling/emptying
management. These issues will interfere with lock process on the upstream/downstream waterways is
design to an extend that depends on local conditions. relevant (i.e. waves, surges), locks with water saving
basins can be operated faster than standard locks,
Impact of climate changes has to be considered as an as the amount of water which is exchanged with the
important source of uncertainty for the long term water waterway is smaller.
resource planning.
From water resource management point of view, the
A2.1 Water shortage - Quantity optimisation goal is to find a good balance between
In arid regions, areas with a small water supply or in water consumption, pumping costs and the possible
canals, the water consumption of locks can be an transport of contaminants through the lock.
important optimisation goal. It should be considered
that obviously a climate change is taking place, which A2.2 Salt water intrusion - Quality
might be a relevant factor during the lifetime of the Salt water intrusion in inland waterways by locking
lock. ships in a navigation lock induces large problems
when this water is used for drinking water or to
In canals the water consumption of a lock must be sprinkle agriculture areas. Another problem is that
compensated by pumping stations if no natural water the salt water may penetrate in the soil and replace
supplies for the canal are available. This results in the fresh ground water, which would induce problems
high costs for building the pumping stations, for the vegetation.
maintenance and energy supply which depend on the
amount of consumed water. Salt water intrusion is always accompanied by loss of
fresh water. Because salt water has larger density
If the water on the upper side of the lock is also then fresh water, it naturally enters in the lock
needed for other purposes (e.g. drinking water, chamber and replaces partially the fresh water. The
irrigation), water saving can be very economic, as less loss of fresh water is mainly a problem during the
water has to be transported / cleaned / desalinated. It summer, when the need of fresh water is the highest.
must be considered on which side of the lock the
water quality is better and if exchange must be Salt water intrusion problems not only play a role in
controlled. This must be carefully judged, as the the design of navigation locks between the sea and
implications are not obvious. an inland waterway provides salt water intrusion, but
also for locks between a salted inland navigation way
Especially the water resource management of large and a fresh water storage basin.
locks (lengths > 200 m and/or lift height > 15 m) is not
an easy question to solve. European directives (EEC,
1997) require that the friendliest solution from an The amount of salt water intruding on a fresh water
environmental point of view must always be studied, storage basin by a navigation lock, when no
even without considering economical questions. So preventive measures are taken to separate salt and
when we have to consider a large water head, even if fresh water, depends on the number of navigation
the most feasible technical and economical solution is cycles, the size of the lock and the difference in water
to build a standard lock (without saving system), it is levels (lift height).
mandatory to study other solutions (several locks with
reduced water head or the use of water saving Due to the difference in density of salt and fresh
basins), which might increase significantly the building water, currents in the lock chamber will occur during
and operational costs. the navigation cycle. These currents cause forces on
the ships, which have to be considered.
It must be considered that locks with water saving
basins, lock ladders or twin locks (with water To reduce the salt water intrusion and the loss of
exchange between the locks) show a slower fresh water few solutions can be mentioned:
operational speed than the equivalent standard type of

 Hydraulic Engineering
Faculty of Civil Engineering 130 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

- Air-bubble-barrier in the lock head to avoid water for navigation locks without separation of salt
exchange of salt and fresh water as the gates are and fresh water.
opened.
- Movable sill on the locks floor to reduce the water 1. Lock Navigation cycle
depth to a minimum for the ships to be locked. A navigation lock between the sea or a tidal river with
- Systems to remove the salt layer during locking by: salt water and a canal with fresh water will be
i) Leveling and transversal exchange considered in the following. The amount of chlorine in
(Duinkerken, NL), the salt water on the seaside of the navigation lock
ii) Leveling and, transversal and longitudinally will be much higher than in the canal-side (fresh
exchanges, waterside).
- Use of multiple lifts,
- Salt water sump (see P R Cardiff [P.R. 10-02]) Consider the navigation cycle at the moment that the
For the description about these systems see gate at the salt-water side of the lock has been
PIANC’1986 1986 on Locks (Section 12), see opened for quit a long time, so the lock-chamber will
attached CD-ROM Directory A3. be completely filled with salt water having the same
density as the water at the seaside.
Besides technical solutions also management
solutions can be mentioned, such as: The gate at the seaside will close and the emptying
- reducing the opening time of the gates, of the lock-chamber starts, discharging salt water to
- reducing the number of lock cycles, the canal when the water-level of the canal is lower,
- in case a lock is equipped with two lock or filling with fresh water when the water-level of the
chambers, a bigger one and a smaller one, use canal is higher.
the lock chamber which fits the best to the ships
to be locked. Empting the lock chamber will bring an amount of
salt water in the canal, which is mixed with the fresh
The above mentioned technical solutions are very water. The density of the salt water in the lock-
expensive and technically complicated. The chamber will not change.
management solutions are not complicated and do
not increase the building costs or maintenance costs. Filling the lock chamber with fresh water, in case of
higher water level on the canal, the average density
One of the best references about Water quality control of the salt water in the lock chamber will decrease.
at ship locks, Prevention of salt and water exchange
is Kerstma et al (1994) in which a number of projects Opening the gate at the canal-side, the salt water in
are described. the lock-chamber will exchange with the fresh water
in the canal by internal flow, with the final result that
Obviously, for the Panama Canal lock project salt the total lock-chamber-content enters the canal and
water intrusion studies, including the required lands as a rather thin layer on the bottom of the
modelling work, were conducted. These studies were canal. The water in the lock-chamber has turned
critical in discarding the use recycling pumped water, fresh.
of one and two lift locks, and the adoption of three lift
locks with water saving basins. Executive summaries The exchange of salt water will be stopped when the
study can be found in/at: gate at the canal-side closes.
http://www.pancanal.com/esp/plan/estudios/0260-
exec.pdf After closing the gate at the canal-side the leveling of
http://www.pancanal.com/esp/plan/estudios/0276- the lock-chamber starts again, empting fresh water
01.pdf on the sea or tidal river when the water level on the
sea is lower, or filling with salt water when the water
level on the sea is higher.
A2.2.1 Salt water intrusion and loss of fresh
water during the lock navigation cycle Empting the lock chamber will bring an amount of
This section presents the lock navigation cycle and fresh water to the sea. This is a large loss of fresh
the calculation of salt water intrusion and loss of fresh water. The density of the fresh water in the lock-
chamber will not change.

 Hydraulic Engineering
Faculty of Civil Engineering 131 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

The factor 1.4 (conversion from Vsalt to C ) depends

Filling the lock chamber with salt water, in case of on the temperature of the water. For a water
0
higher water level on the sea or tidal river, the average temperature of 10 °C the factor is 1.4. For 20 C the
density of the fresh water in the lock chamber will factor is approximately 1.3.
increase.

After filling the lock-chamber the gate at the sea-side A2.2.2 Induced flow in the lock chamber
opens and the total amount of fresh water in the lock- A flow is created in the lock chamber due to
chamber will be pushed away by salt water. The difference in density of salt and fresh water during
exchange of fresh water will be stopped of course the navigation cycle.
when the gate at the sea-side closes.
1. Filling and emptying the lock chamber
Ships in the lock-chamber will decrease the amount of When a fresh-water-lock chamber is filled up with
exchanged salt or fresh water with their water- salt water, the salt water sinks to the bottom of the
displacement tonnage. chamber and the fresh water layer stays above the
salt water (Fig. A1)
2. Calculation of salt water intrusion and loss of
fresh water Filling up a salt-water-lock chamber with fresh water,
The formula to determine the amount of salt water the fresh water flows into as a layer above the salt
entering in the canal at each lock cycle is: water layer.

Vsalt = (hs − z d ).blc .llc − Vs (maximum amount) The water flowing into the lock chamber reflects
against the gates and particularly against the bow
- and stern of the moored ships. Due to this flow and
and the mass of CL - ions is:
its reflection, an internal wave is generated with a
( ρ1 − ρ 2 ) large wavelength.
C = V salt (maximum amount)
1 .4 When emptying the lock chamber there is no wave
generated inside the lock chamber, but only in the
The formula for the amount of fresh water entering in approach channel.
the sea at each navigation cycle is:

V fresh = (h fr − z lc )blc .l lc − Vs (maximum amount)

with:

3
Vsalt = volume salt water m
3
Vfresh = volume fresh water m
3
Vs = volume displacement of m
ships
-
C = mass of CL - ions kg
hs = water-level of salt water side m
hfr = water-level of fresh water m
side
zd = threshold level of fresh m Fig. A1: Lock filling/emptying - Density flow during
water side filling the fresh-water-lock chamber with salt water
zlc = bottom-level of lock- m (above) or the salt-water-lock chamber with fresh
chamber water (below)
blc = width of lock-chamber m
llc = length of lock-chamber m 2. Exchange of salt and fresh water by opening the
ρ1 = density fresh water kg/m
3 gates
ρ2 = density salt water kg/m
3 As soon as the gates are opened the exchange of
salt and fresh water starts due to the density

 Hydraulic Engineering
Faculty of Civil Engineering 132 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

difference (Fig. A2). In case of a fresh-water-lock

chamber and a salt-water-approach channel a salt A2.2.3 Exchange of salt and fresh water as
layer will enter over the bottom of the lock chamber function of time
and the fresh water will flow at the surface of the Opening the gates between a salt-water-lock
approach channel. When the gates of a salt-water- chamber and a fresh-water-approach channel, the
lock chamber are opened the fresh water of the fresh water will flow into the lock chamber at the
approach channel will flow into the lock chamber at water surface with a speed, calculated with the
the water surface and the salt water will flow into the formula:
approach channel along the bottom.
ci = 0.53 ε .g .(hbi − z k )
As long as the ships are not sailing there will be a
minimum of mix of the salt and fresh water.
with ci the velocity of propagation of fresh water
The incoming water layer reflects against the gates (m/sec).
and particularly against the bow and stern of the
moored ships, and generates an internal wave with a For a lock with a fresh-water-lock chamber and a
large wavelength. salt-water-approach channel the factor 0.53 changes
to 0.43.

The exchange of fresh and salt water at the location

of the opened gates is stable during a long time and
depends of the dimensions of the open gates. The
discharge is calculated by:

Qu = bd .(hbi − z d ).C u ε .g .(hbi − z k )

with:

3
Qu = discharge of exchanged water m /s
bd = width of lock m
hbi = water-level of fresh water side m
zd = threshold level of fresh water m
side
zk = bottom level of lock chamber m
Cu = coefficient of discharge of -
exchanged water (= 0.2)
ε = (ρ1- ρ2)/ ρ2 = relative difference -
in density
3
ρ1 = density fresh water kg/m
3
ρ2 = density salt water kg/m
2
g = gravitational acceleration (about m/sec
9.81)

After a certain period, the fresh water front reaches

the closed gates, reflects against these gates and
travels back in the direction of the opened gates. The
fresh water front reaches the open gates after a
period t1:

2l k
t1 =
ci
Fig A2: Gate Opening - Density flow during exchange
of salt-water-lock chamber with fresh water

 Hydraulic Engineering
Faculty of Civil Engineering 133 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

On this moment (t1) the volume of salt water in the

lock chamber replaced by fresh water is: The lock is used for commercial navigation (approx.
4700 ships/year) and a lot for pleasure navigation
Vu = Qu .t1 (approx. 20.000 pleasure boats/year).

with Vu , the exchanged volume of water (m )

3 The number of navigation cycles is approximately
5000 a year from the Spaarne to the Zijkanaal C
(from fresh to salty water) and approximately 5400 a
In these calculations the volume of the ships in the
year the other way (from salty water to fresh water).
lock has also to be taken into account. When the
ships leave the lock their volume will be replaced by -
The average mass of CL - ions on the side of
fresh or salt water water.
Zijkanaal C is 1565 mg/ltr and the maximum mass is
A2.2.4 Example 3500 mg/ltr. In the near future this average mass of
-
CL ions will increase to 2250 mg/ltr, due to the
As example for the calculations of salt water intrusion
proposed construction of a new large lock (70 x 500
on a fresh water storage basin we look at the design
m) at IJmuiden.
of the renovation of the “Grote Sluis” (Fig. A3),
Spaarndam, The Netherlands (2007).
With the above mentioned data, calculations are
-
made for the total mass of CL - ions each navigation
This lock is situated in the western part of the country,
cycle, taken in account that the gates stay open after
between Zijkanaal C, a side branch of the North Sea
the ships have left the lock chamber, which is up to
canal, and the Spaarne, a large fresh water basin with
now the usual lock operation mode at the “Grote
a lot of flower-plantations (tulips) in it's surroundings.
Sluis”.
Zijkanaal C (side branch of the Noordzeekanaal) is
At each navigation cycle a mass of approximately
salted through the large locks at IJmuiden, the -
18.000 kg CL - ions are added to the fresh-water-
entrance to the port of Amsterdam.
storage basin, the “Spaarne”. This means in one year
an amount of approximately 100.000.000 kg
The renovation of the “Grote Sluis” is necessary -
(100.000 ton) CL - ions. It is worth to take some
because of the bad condition of the lock, the gates,
actions to reduce this amount of salt!
the mooring places and so on, and also to reduce the
large amount of salt water, which enters in the fresh- -
To reduce this amount of CL - ions the lock chamber
water storage basin at each navigation cycle and
will be divided in two parts, a small and a larger lock
which is very harmful for the flower-plantations.
chamber to have a better link with the number of
ships to be locked. Also the lock method will change,
The lock chamber of the “Grote Sluis” has a length of
witch means, among other things, that the gates will
80 m, a width of 25 m and a depth of 5.1 m. The gate-
be closed immediately after the ships have left the
openings have a width of 12 m.
lock chamber and the pleasure boats have to wait
until the total lock chamber is filled.
The water level difference is normally 0.40 m, at witch
the salt water side (Zijkanaal C) is the highest side.
70a

100a
101

Fig. A3: Overview of the “Grote Sluis”te Spaarndam in The Netherlands

 Hydraulic Engineering
Faculty of Civil Engineering 134 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

A2.2.5 Additional Forces on ships A2.2.6 Situation on the approach channel

Filling and exchanging the lock chamber causes By exchanging of water of the lock with water in
additional forces on the moored ships, due to the the approach channels, the average density of the
difference in density between salt and fresh water. water in the approach channels will change. The
The largest forces are generated when the gates salt-water-approach channel will become little less
are open and the lock chamber exchanges salt to salty and the fresh-water-approach channel will
fresh water, or opposite. become more salty.

The maximum force on a ship caused by the The fresh water on the salt-water-approach
difference in density is approximately: channel will be as a thin fresh water layer upon the
salt water. This thin fresh water layer flows to the
ε (d .b − d s .bs )
2 2 sea and will be mixed with the salt seawater by
F 'r = .d s − k k wind, ship propellers and so on.
2.l s .C b d k .bk − d s .bs The salt water on the fresh-water-approach
with: channel will take the shape of a thin salt water
layer positioned on the bottom of the approach
F’r = relative longitudinal force - channel. This salt water layer extends over a long
(relative to ship tonnage or distance from the lock into the approach channel.
displacement)
ls = length of the ship m The salt water on the fresh-water-approach
Cb = block coefficient (*) - channel is harmful for the agriculture, environment,
bk = width of the lock chamber m fishery, and so on, and should be removed.
bs = width of the ship m
ds = depth of the ship m To remove this salt water layer the best way is to
dk = water level depth m collect it in a deep sink in the bottom of the
ε = (ρ1- ρ2)/ ρ2 = relative - channel and to flush it away with a discharge
difference in density sluice or pumping station.
Another way is to avoid salt water on the fresh-
(*) The block coefficient (Cb) is the ratio between water-approach channel, as much as possible.
the real ship volume under water and the volume This requires special lock systems (see
of a block equals to length x width x depth of the PIANC’1986 Report, section 12)
ship.

 Hydraulic Engineering
Faculty of Civil Engineering 135 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

A3 Ice control in locks

The following focuses on methods to avoid ice largely dependent on the water temperature in the
formation in locks and in the locks entrance lock. Experience has shown that the water
o
channels (emptying and filling). For more temperature should be approx. 0.3 – 0.5 C before
information on ice formation in channels see the use of bubbling systems or the continuous
PIANC InCom Working Group 23 “Technical and use of current inducers in ice control is possible.
Economic Problems of Channel Icing” (published Otherwise the use of these methods leads to an
in 2004). Some of the results of this report, the increased ice formation, which worsens the ice
ones most relevant for ice control in locks, are situation. In this case it is advisable to stop the
presented in Table 5.1. The report points out the traffic for the cold period instead of trying to
fact that the methods for ice control at locks are continue it.

Methods Means Results

Thermal Conducting additional heat in Efficient and considerably cheap if waste heat is
engineering to the water mass available. Otherwise expensive.
Steam Quite inefficient
Hot water Efficient
Bubbling systems Efficient if there is heat capacity in the water mass,
otherwise may increase ice formation
Solar heat Inefficient
Wind power Inefficient
Heat of the ground water Could help in locks
Dusting of ice sheet (mixing Works if performed at the right moment, otherwise
with ashes) inefficient
Electrical methods Resistors in the lock wall Works, but the changing of resistors is difficult
Heating panels and mats on Efficient, but damage resistance is poor. Quite
the lock walls expensive and reduce the usable width of the lock.
Resistor panels on gates Efficient
Radiant heaters Inefficient
Chemical methods Low-adhesive coatings on the Poor duration and require regular renewal. Reduce
lock walls the ice adhesion strength of ice, but do not eliminate
the need for manual ice removal.
Lowering of the freezing point Calcium chloride and sodium tested with poor results.
of water
Mechanical Bubbling systems, high flow Efficient in controlling ice movements in the locks,
methods air screening and surface especially in the gate recesses and the upper funnel
current inducers area of the lock.
Use of manpower Requires a lot of labour, reliable, used together with
other methods
Chipping Widely used in removal of ice from structures,
reliable, but time consuming
Excavator Used during the most severe winters to relieve the ice
situation in the lower funnel and waiting areas of the
locks, may damage the canal structures
Chainsaw Relatively inefficient and vulnerable. May cause
damage to the canal structures.
High pressure water jet Expensive
Table 5.1 : Methods for ice control in locks (InCom-WG23, PIANC 2004)

To estimate the ice formation in locks and canals in the locks and canal, a series of field test were
with ongoing traffic and the need for additional heat carried out in Finland. The tests included full scale

 Hydraulic Engineering
Faculty of Civil Engineering 136 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

tests on the Saimaa Canal with warm water

conduction to the canal. The purpose of the tests Vs = (E1+E2)/(L*(1-e))
was to find out the needed heat input to the canal in
order to control the ice situation so that the brash In order to get a stabilised ice situation in the lock
ice layer thickness would not exceed 1.0 m. the heat input must be at the same level as the
heat loss from the lock. The results indicated that
The results from the canal sections can be the heat loss from the canal is between 1 – 4
generalized to the locks. The theoretical kW/m, when the brash ice layer is approx 1.0 m
calculations were checked and calibrated with the thick. Most of the heat loss takes place through the
field study measurements. brash ice layer. When the trough in the canal is
approx. 20 m wide, the heat loss is approx. 50 –
2
The amount of brash ice forming during the ice 200 W/m . In lock the heat loss is somewhat
breaking is approximately: bigger, if the brash ice layer is kept thinner and if
the bubbling system and current inducer are in use.
3 2
VL = (q*t*Br)/(L*(1-e)) (m /m) where Anyhow, the heat loss of 200 W/m corresponds to
o
the heat loss from open water at -10 C
Br = width of the trough (lock) temperature, which means that the required heat
t = time between the ice breakings input very seldom exceeds this despite the use of
3
L = fusion heat of ice (307 MJ/m ) bubbling and current inducers except in severe
e = porosity of brash ice (approx. 0.33) winters. These results indicate that for instance the
0.9 0.3
q = the heat loss in ice formation, 12*T *n , maximum heat input to control the ice situation in a
where 110 m long and 12 m wide lock would be approx.
T = mean temperature between the ice breakings 66 – 264 kW. This heat input can theoretically be
o o
( C) done for instance by conducting 3 litres/s of +30 C
n = frequency of ice breaking (times a day) water to the lock.

The thermal energy needed to melt the ice is With regards to ice formation, the Finnish Maritime
approximately: Administration carried out a major study on the
Saimaa Canal during 2004 - 2005. As a result of
1.5
E1 = Tw*v *k1*A*b (W/m) where this study this administration created an ice
formation simulation model for the Saimaa Canal.
Tw = temperature difference between the flowing
o
water and ice melting point ( C) The study consisted of field measurements in
v = flow velocity (m/s) different canal ice conditions with different canal
k1 = trough roughness factor heat capacity. These measurements were the basis
1.5 3.5 o
A = constant 4,800 (W*s )/(m * C) in creation of an ice formation simulation model,
b = width of the bottom of the brash ice layer (m) which was used to verify the previous theoretical
k1 = (1+4*h1)*C=1 calculations of the required heat input needed in
h1 = mean thickness of the brash ice layer (m) controlling the ice situation in the canal.
C = equalizing factor (h1/h1max)
h1max = mean maximum thickness or local The simulation consisted of three traffic situations
maximum thickness of the brash ice layer (one, two or four ship passages per day). The
results pointed out that the designed 120 MW heat
The vessel passages also consume the thermal outlet (discharge of warm water with 1 km intervals)
energy when it rotates the warm water melting the is sufficient in controlling the ice situation. During
ice. This energy can be estimated with formula: mild and normal winter vessels with 1A-ice class
could operate almost without icebreaking
E2 = k2*Tw*Q (W/m) where assistance. During severe winters icebreaking
assistance is needed and extreme winters the use
k2 = vessel factor (0.5 – 2.0) of convoys is needed. The simulation model also
o
Q = constant (50 MJ/ C*m a vessel passage) pointed out that short traffic stops effectively reduce
the formation of ice. The model also pointed out
The amount of ice melted between the ice that the melting effect in between the warm water
breakings is approximately:

 Hydraulic Engineering
Faculty of Civil Engineering 137 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

discharge points varies more than predicted in the

theoretical calculation.

New vessel and propulsion systems (such as

double acting ships and rudder propulsion systems)
ease the ice navigation problems as they become
more common.

 Hydraulic Engineering
Faculty of Civil Engineering 138 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

A4 Case study: Naviduct in Enkhuizen

A4.1 Introduction
The navigation lock in Enkhuizen (the
Netherlands) has been constructed
between 1969 and 1971; it is the Northern
connection between the IJsselmeer and
the Markermeer. The navigation lock is
suitable for vessels of class CEMT Va.
Adjacent to the navigation lock there is a
stop lock (spuisluis). Across the
navigation lock a road is situated. The
bridge can be opened when sailing ships
have to pass.

The existing navigation lock is causing

delays for vessels and road traffic during
the (busy) summer season. The capacity Figure App3: 1 Naviduct Enkhuizen
of the navigation lock is insufficient for
recreational ships. It is expected that the number of ships will increase within the next decade causing
more and more delays for the vessels. In addition, the bridge crossing the lock is causing traffic delays
when opened.

To resolve this problem a Naviduct will be built on the intersection between ship- and road traffic. These
functions have to be combined. The Naviduct is an aquaduct which is able to function as navigation lock.
In this way the capacity of the road and waterway will be increased.

Figure App3: 2 Artist impression of the Naviduct

 Hydraulic Engineering
Faculty of Civil Engineering 139 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Figure App3: 3 Situation near Enkhuizen and the Naviduct

A4.2 Terms of Reference (Programma van Eisen)

A4.2.1 General
The registered water levels between 1976 and 1990 have been analysed. Based on this data the high
water levels have been determined. In addition to this wave action caused by wind has been determined.
The increase in water level in the IJsselmeer over the operational life span is estimated at 0.20 meters.

IJsselmeer Markermeer

Design level based on NAP+1,90 m

1/10.000 yrs NAP -1,05
Significant wave height 0,69 m
Wave period 2,2 s
Wave length 7,70 m

Design level based on NAP +1,70

1/1.000 yrs NAP -1,50 0,90 m
Significant wave height 2,5 s
Wave period 10,0 m
Wave length

 Hydraulic Engineering
Faculty of Civil Engineering 140 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

The soil conditions can be summarized as:

- Weak soil layers with low permeability consisting of clay and peat till NAP-13.0 m
- Sand layer, till NAP -25.0 m
- Clay layer, till -31.0 m
- Capillary height (stijghoogte) water in weak soil layer NAP -0.40 / -0.20 m
- Capillary height Sand Layer NAP -2.00 m

Based on prognoses the number of vessels passing the locks is

Vessel type/number of passages 1989 2010

Inland navigation 4.500 4.650

Other commercial traffic 5.800 5.000
recreational vessels > 20 m 5.000 10.000
recreational vessels < 20 m 55.000 80.000

A4.2.2 Functional requirements

The global dimensions are determined by 2 parallel navigation locks which have an efficient length of 125
m and width of 12.5 m. The sill depth is determined at NAP -4.50 m. The maximum water level in the lock
is NAP + 0.45, the minimum is NAP -0.5 m. Locking is possible at water level differences less than 1.00
m.

A risk analysis has been performed which states that a protection of the lock gates against collision is not
necessary. During periods of frost and ice, it should be able to operate the locks. The design life of the
construction is 100 yrs, the gates and other replaceable elements must be designed cost efficiently.

The navigation lock has to be designed to retain:

• High water level on the IJsselmeer (probaility of failure is 1/10.000)
• High water on the Markermeer (probability of failure is 1/1000)

The level of the gates is at NAP +2.00, the level of lock at NAP +2.65 m. The height of the existing
Houtribdijk is NAP + 6.40 m. A new dike will be constructed at NAP + 4.00 m. At a water level exceeding
NAP+ 0.70 m or at water level differences larger than 1.00 meter all gates will be closed.

A4.3 Design

Excavation works

Costs associated with excavation work (grondverzet) account for a large factor in the budget. For this
reason the lock will be located as much to the North as possible, against the Houtribdijk. The orientation in
East-West direction depends on the chosen steepness of the road (5%). The orientation in North-South
direction is chosen in a way as to limit the length of the breakwater (leidammen) and ensure sufficient
manoeuvring space for the vessels in the waiting and berthing area. This resulted in an orientation which
0
makes an angle of 10 to the north as shown in Figure App3: 4.

 Hydraulic Engineering
Faculty of Civil Engineering 141 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Figure App3: 4 Sheetpiles (damwand) around the artificial polder to construct the Naviduct

The design is made in such a way that the road and navigation lock is located in an artificially created
polder. The lowest level, consists of the road, is at an elevation of NAP -10.50 m.

The weak soil layer is excavated to a depth of NAP -11.50 and replaced by sand. The remaining soil layer
with a thickness of 1.50 meter remains to function as a vertical barrier between the different water levels
(capillary height). The artificial polder has sheet piles which act as a barrier in horizontal direction. The
sheet piles are founded in the clay layer. The weak layers near the Houtribdijk are dredged away.

Lock gates and lock head

The gates should operate independently. To minimize the width the island between the 2 lock chambers
should be as small as possible. Hence, the hydraulic gate driving machinery has been positioned behind
each other, parallel to the lock axis, as is illustrated in Figure App3: 5.

 Hydraulic Engineering
Faculty of Civil Engineering 142 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Figure App3: 5 Hydraulic system of the lock gate

The lock head has been designed around the selected gates. A cross section is shown in Figure App3: 6.

Figure App3: 6 Cross section of the lock head

 Hydraulic Engineering
Faculty of Civil Engineering 143 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Lock chamber

The lock chamber differs from conventional navigation locks because the lock chamber is constructed
above a road. Demands on the water retention of chamber floor and bottom are therefore high. In addition
to these demands the lock chamber retains soil in the artificial polder.

Figure App3: 7 Cross sections of the 2 lock chamber

The slice in the middle of the lock chamber is one big monolithic construction consisting of the lock
chamber, pump room, and road intersection. The pumps are installed for the drainage of the artificial
polder.

 Hydraulic Engineering
Faculty of Civil Engineering 144 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Figure App3: 8 Cross section Naviduct at waterway-roadway intersection

A4.4 Construction
The following construction methods were taken into consideration:

a) construction pit with dewatering (open bouwput en bemaling).

b) cofferdam with sheet piles and underwater concrete (bouwkuip)
c) cofferdam with sheet piles and underwater concrete and tension piles (bouwkuip met trekpalen)
d) pneumatic caisson (pneumatisch afzinken)

The use of a construction pit was selected since this proved to be the most cost efficient.

Figure App3: 9 Construction methods

A4.5 Waiting- and berthing place

The lock is enclosed by a waiting and berthing place in the North as well as the South. The function is to:
• Provide an area to wait
• Protect the navigation lock against wave attack
• Create a calm area which facilitates manoeuvring

 Hydraulic Engineering
Faculty of Civil Engineering 145 3/3/2011
Delft University of Technology
CT 3330 Hydraulic Structures Locks

Figure App3: 10 Waiting and berthing locations Naviduct

A4.6 As-built
The construction of the Naviduct started in 1999 and was finished in 2003.

Figure App3: 11 Naviduct Enkhuizen

 Hydraulic Engineering
Faculty of Civil Engineering 146 3/3/2011
Delft University of Technology