GRAVITY DAM STRUCTURES

by

JACOB KROON

SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS

FOR THE MASTER’S DEGREE IN

CIVIL ENGINEERING

FACULTY OF ENGINEERING
UNIVERSITY OF PRETORIA
PRETORIA

MAY 1984
 FOREWORD

This translation of the original treatise was completed in 2022. It is not a revised edition, merely an
attempt at a complete English translation (with the Author’s approval) of the treatise which was
published in Afrikaans in 1984. The Author has provided a list of known errata in the original
document, and corrections have been incorporated in this translation.
Although most of the theory and methodology in this document is still valid and it still serves the
dam design section as well as always, certain known changes have occurred in the last 38 years
which may bear mentioning:
When this treatise was written, what is now the Department of Water and Sanitation was a
Directorate in the Department of Environmental Affairs. The Author often refers to the
‘Directorate’, this can now be interpreted to mean the ‘Department’. The section names such as
Hydrology, Mechanical and Electrical, Planning, Operations, etc. are still broadly the same, with
most of these now called ‘Directorates’.
More recent innovations in gravity dam construction such as the now common use of roller
compacted concrete and the subtle change these innovations brought were obviously not all known
at the time of writing. Rollcrete (as it was then known) is briefly mentioned in the treatise in the
conclusions and recommendations on page 16-3. The chapters covering river diversion,
construction joints, and concrete specifications do not cover this more recent construction method.
(Roller compacted concrete dams are not normally constructed in separate blocks but may have
‘induced’ joints separating blocks). The concrete mixture used for RCC has also evolved.
Asbestos cement pipes are no longer used.
Many of the dams referenced in this treatise have undergone name changes.
Where known changes have occurred, footnotes have been provided on the relevant pages.
Mr. Kroon retired from the Department in May 2015 after a long and successful career. He remains
as interested as always in the field of dam design.
DECLARATION

I declare that the treatise that I herewith submit for the Master’s degree in Civil Engineering to the
University of Pretoria, has not been submitted by me for a degree at any other university.

...............................
J. KROON
 SUMMARY
GRAVITY DAM STRUCTURES
BY
JACOB KROON

LEADER: Prof. Dr. A. Rooseboom

DEPARTMENT: Civil Engineering

DEGREE FOR WHICH SUBMITTED: M. Eng. in Civil Engineering

This treatise is a manual which could be used in design offices during the planning and design
of gravity dams. Particular attention has been paid to local circumstances, and experience in the
Directorate of Water Affairs has as far as possible been documented.
An additional aim of the treatise was to obtain more uniform standards, designs, and
specifications for the purpose of bringing about cost savings.
Particular attention was given to the following aspects of dam design:
(a) Site selection
(b) Geological investigations
(c) Design standards
(d) Dam spillways
(e) Non-overspill section of the dam
(f) Outlet systems
(g) Energy dissipation
(h) Foundation grouting and drainage
(i) River diversion
(j) Joints and water sealing
(k) Galleries and tunnels
(l) Concrete for gravity dams
(m) Gauge plates and water level recorders

Sufficient information has been collected to enable the complete planning and design of a
gravity dam to be done. A new design aid has been developed to ease the design of splitters.
 ACKNOWLEDGMENTS

The author acknowledges with thanks the leadership of Professor A. Rooseboom in the preparation
of this treatise. Without his good advice and encouragement this treatise would not have seen the
light of day.

The Director-General of the Department of Environmental Affairs is thanked for permission to

publish this treatise.

A heartfelt word of thanks to Mr. F. Hollingworth (Deputy Chief Engineer) for his cooperation and
good advice. Thanks also go to the draftsmen who assisted with the preparation of the drawings
and to the typists of the Department who typed the treatise.

Finally, a word of heartfelt thanks to my wife and parents who encouraged and supported me to
tackle and complete this task.
 (i)

CONTENTS

CHAPTER SUBJECT PAGE

1.0 PROBLEM STATEMENT 1-1

2.0 DEFINITION SKETCH 2-1

3.0 SITE SELECTION 3-1

3.1 Alternative dam sites 3-1

3.2 Factors which could influence choice of site 3-1
3.2.1 Topography 3-1
3.2.2 Geology 3-1
3.2.3 Ancillary structures 3-2
3.2.4 Local circumstances 3-2
3.2.5 Site access 3-2
3.2.6 Availability of construction material 3-2
3.2.7 Method of construction 3-2
3.2.8 River diversion 3-3
3.2.9 Siltation 3-3

4.0 GEOLOGICAL INVESTIGATIONS 4-1

4.1.0 The planning phase 4-1

4.1.1 Reconnaissance study 4-2
4.1.2 The borehole survey 4-3
4.1.3 Results, conclusions, and recommendations 4-4
4.2 The final design phase 4-7
4.3 The construction phase 4-8

5.0 DESIGN STANDARDS 5-1

5.1 Introduction 5-1

5.2 Loads 5-1
5.2.1 Hydrostatic loads 5-1
[Link] Full supply level 5-1
[Link] High flood level 5-1
[Link] Probable maximum flood level 5-3
 (ii)

CHAPTER SUBJECT PAGE

[Link] Toe water loads 5-3

5.2.2 Own weight 5-3
5.2.3 Silt load 5-4
5.2.4 Uplift forces 5-4
5.2.5 Earthquake loads 5-6
5.2.6 Other loads 5-8
5.3 Load combinations 5-8
5.3.1 Normal category 5-9
5.3.2 Abnormal category 5-11
5.4 Analysis of the structure 5-15
5.5.0 Design criteria 5-16
5.5.1 Allowable stresses in the concrete 5-16
5.5.2 Safety against sliding 5-16
5.5.3 Remarks 5-17
5.6 Sample calculation 5-17

6.0 DAM SPILLWAYS (FLOOD OUTLETS) 6-1

6.1 Purpose of spillways 6-1

6.2.0 Typical spillways 6-1
6.2.1 Ogee spillways 6-1
6.2.2 Free overfall spillways 6-3
6.2.3 Morning Glory spillways 6-4
6.2.4 Side channel spillways 6-5
6.2.5 Siphon spillways 6-6
6.2.6 Auxiliary spillways 6-6
6.2.7 Controlled spillways (Sluice gates) 6-7
6.3.0 Layout and choice of spillway 6-10
6.4.0 Components of spillways 6-11
6.5.0 Model studies 6-11
6.6.0 Controlled versus uncontrolled spillways 6-12
6.7.0 Spillway design standards 6-12
6.8.0 Design of ogee spillway 6-14
6.8.1 Design information 6-14
6.8.2 Determination of effective spillway length 6-14
 (iii)

CHAPTER SUBJECT PAGE

6.8.3 Determination of the spillway design head (Ho) 6-17

6.8.4 Determination of discharge at levels other than Ho 6-17
6.8.5 Discharge head due to PMF 6-22
6.8.6 Determination of spillway profile 6-23
[Link] Simplified spillway profile 6-23
[Link] More accurate spillway profile 6-24
6.8.7 Transition to horizontal level 6-27
6.8.8 Cavitation 6-28
6.9.0 Control gates 6-29
6.9.1 Radial gates 6-29
6.9.2 Vertical gates 6-30
6.9.3 Fishbelly flaps 6-31
6.10 Spillway profile of controlled spillways 6-32
6.11 Discharge through a radial gate 6-33
6.12 Discharge through a vertical gate 6-34
6.13 Emergency gates 6-37
6.14 Loads on gates 6-37

7.0 THE NON-OVERSPILL SECTION OF THE DAM WALL 7-1

7.1 Section selection 7-1

7.2 Freeboard 7-3
7.3 Upper trajectory of spillway nappe profile and training wall design 7-5

8.0 OUTLET WORKS 8-1

8.1 Introduction 8-1

8.2 General guidelines and principles for the design of outlet works 8-2
8.2.1 Capacity of the outlet works 8-2
8.2.2 Location of the outlet works 8-3
8.2.3 Abstraction systems 8-5
8.2.4 Control system 8-5
 (iv)

CHAPTER SUBJECT PAGE

8.3.0 Components of the outlet works 8-6

8.4.0 Intake section 8-6
8.4.1 Approach channel 8-6
8.4.2 Trash rack 8-6
[Link] Hydraulic energy losses through trash racks 8-7
8.4.3 Intake section of a multi-level abstraction system 8-9
8.4.4 Intake section of a single-level abstraction system 8-11
8.5.0 Pipeline section 8-14
8.5.1 Bellmouths 8-14
8.5.2 Bend- and bifurcation pipe fittings 8-16
8.5.3 Tapered pieces 8-18
[Link] Pipes 8-21
[Link] Pipeline losses 8-21
8.5.5 Flow gauging 8-23
8.5.6 Isolating valves 8-24
[Link] Butterfly valve 8-24
[Link] Sluice valve (Gate valve) 8-28
8.6.0 Outlet section 8-31
8.6.1 Outlet house 8-31
8.6.2 Control valves 8-32
[Link] Sleeve valve 8-32
[Link] Needle valve 8-36
8.7 Sample calculation 8-37

9.0 ENERGY DISSIPATION 9-1

9.1 Introduction 9-1

9.2 Hydraulic jump stilling basin 9-2
9.3 Ski-jump/Flip bucket 9-4
9.4 Slotted- and solid buckets 9-5
9.5.0 Splitters 9-6
9.5.1 Introduction 9-6
9.5.2 Operating principle of splitters 9-9
[Link] Design of splitter layout 9-13
[Link] Definition of symbols 9-13
[Link] Design criteria 9-15
[Link] Loads on splitters/step 9-25
9.6 Sample calculation 9-26
 (v)

CHAPTER SUBJECT PAGE

10.0 FOUNDATION GROUTING AND DRAINAGE 10-1

10.1 Introduction 10-1

10.2 Grouting investigation 10-2
10.2.1 Geological factors influencing grouting 10-2
10.2.2 Permeability investigation 10-3
10.3 Grouting design 10-4
10.3.1 The necessity to grout 10-4
10.3.2 Curtain grouting 10-5
[Link] Position of grout curtain 10-5
[Link] Orientation of grout curtain 10-6
[Link] Depth of grout curtain and drainage holes 10-7
[Link] Grouting intensity 10-8
[Link] Stage lengths 10-10
10.4 Grouting methods 10-10
10.5 Grout specifications 10-12
10.6 Monitoring of grouting 10-12

11.0 RIVER DIVERSION 11-1

11.1 Purpose 11-1

11.2 Factors influencing the river diversion system 11-1
11.3 Standards 11-2
11.4 River diversion methods 11-3
11.5 Multistage diversion 11-5

12.0 JOINTS AND WATER SEALING 12-1

12.1 Purpose 12-1

12.2 Contraction joints 12-1
12.3 Expansion joints 12-2
12.4 Construction joints 12-2
12.5 Shear keys (Shear grooves) 12-3
12.6 Waterstops 12-5
 (vi)

CHAPTER SUBJECT PAGE

13.0 GALLERIES AND TUNNELS 13-1

13.1 Purpose 13-1

13.2 Location 13-1
13.3 Dimensions 13-2
13.4 Ventilation 13-2
13.5 Access 13-3
13.6 Construction 13-3
13.7 Analysis 13-3
13.8 Lighting 13-4
13.9 Drainage 13-4

14.0 CONCRETE FOR GRAVITY DAMS 14-1

15.0 GAUGE PLATES AND WATER LEVEL RECORDERS 15-1

15.1 Purpose and uses 15-1

15.2 Description 15-2
15.2.1 Gauge plates 15-2
15.2.2 Water level recorder 15-2
15.3 Installation 15-2

16.0 CONCLUSIONS AND RECOMMENDATIONS 16-1

16.1 Conclusions 16-1

16.2 Recommendations 16-1

APPENDIX A A-1
REFERENCES B-1
 1-1

1.0 PROBLEM STATEMENT

The purpose of this treatise is to create a manual which can be used in design offices in
South Africa for the design of “Gravity Dam Structures”, and to document local
experience. It is compiled to specifically meet local requirements. An additional aim is to
obtain more uniform standards, designs, and specifications, based on practical experience
obtained in the Directorate of Water Affairs for the purpose of bringing about cost savings.

Gravity dam structures under consideration are solid concrete structures which are
designed and shaped so that their own weight will enable them to remain stable against the
loads acting on them. The assumption throughout this treatise will be that the structure is
built on rock. Gravity dam structures can however also be built on weaker foundations,
designs must then specifically make provision for this.

Mainly structural aspects will be dealt with and not related aspects such as the
determination of flood sizes, backwater curves, geological surveys, etc. The subject of
arch gravity dams falls outside the scope of this treatise and only straight gravity walls will
be treated.
 2-1

2.0 DEFINITION SKETCH

LEGEND:

1. Grout curtain 14. Training wall 27. Access road

2. Drainage holes 15. Foundation (Contact level) 28. Rip-rap
3. Shear keys – alternative to 4 16. Drainage / Inspection gallery 29. Wet well
4. Shear groove – alternative to 3 17. Apron 30. Dry well (sometimes)
5. Waterstops 18. Outlet block 31. Trash rack
6. Dam heel 19. Valve control room 32. Gantry crane
7. Dam toe 20. Windows 33. Crane rail
8. Contraction joints 21. Control valves 34. Automatic water level recorder
9. Upstream face 22. Handrails 35. Non-overspill crest
10. Downstream face 23. Retaining wall 36. Entrance to dam crest
11. Spillway crest 24. Access to valve control room 37. Dam blocks
12. Splitters 25. Parking area 38. Intake tower
13. Step 26. Gallery entrance 39. Operating platform

FIGURE 2.1: Isometric projection of a typical gravity dam layout

 3-1

3.0 SITE SELECTION

3.1 Alternative dam sites

After the need is identified to supply water to a certain area the alternative dam sites must
be compared to find the best solution (most economical) for the problem. The two most
important factors which play a role here are:

(a) To find a suitable foundation to resist the loads that will be caused by the proposed
structure, and to

(b) Find a suitable dam basin. A basin which is deep in comparison to the surface area it
covers is desirable to limit evaporation.

Cost estimates must finally be done to find the most economical site together with the least
possible disturbance to the local environment.

3.2 Factors which could influence choice of site

3.2.1 Topography

A dam wall in a deep narrow valley will use less concrete and will consequently have a
lower cost. The layout of the final structure also plays a role, that is to say the position of
spillway, outlet works, access, etc.

3.2.2 Geology

A foundation which is relatively free of large faults and shear zones is preferable because
repair work to overcome the above-mentioned shortcomings is a time consuming and
expensive process and the success achieved is not always consistent.
 3-2

It is however possible to construct low gravity dam walls on weak material where leakage
under the dam is not important, but this usually requires flatter downstream slopes to
obtain stability. The depth and bearing capacity of the foundation also influences dam site
selection as a result of cost and stability considerations.

3.2.3 Ancillary structures

While weighing up the alternative sites consideration must already be given to the possible
layout and placement of ancillary structures such as the outlet system, spillway, etc. An
acceptable design cannot be done for a site which is not suitable.

3.2.4 Local circumstances

The choice of a site must minimise the requirement for realignment of roads, rails,
telephone lines, power lines, etc. Loss of good agricultural land must also be a serious
consideration when selecting the final site, as well as disturbing the environment.

3.2.5 Site access

If the site is located near the already existing infrastructure the cost of the structure should
be much lower because it is expensive to supply new access roads, rails, and power lines.

3.2.6 Availability of construction material

Construction material which is available close at hand, such as concrete aggregate, sand,
etc. is a factor which cannot be ignored and should be considered from the outset.

3.2.7 Method of construction

Construction methods also play a role but are usually less important than other factors.
 3-3

3.2.8 River diversion

Provision is usually made to allow the average annual peak flow to safely pass though the
river diversion during the construction period. It is therefore necessary during dam site
selection to look at possible river diversion systems which can be used. With concrete
dams it is usually not as important as at earthfill dams where a 1 in 20 year flood must be
passed safely.

3.2.9 Siltation

Where siltation is a problem the catchment area should be as small as possible.

In the author’s experience site geology plays the main role when comparing alternative
dam sites. Unfortunately, the available good dam sites in South Africa are decreasing and
the designer is often forced to use a given site. It is for this reason that gravity dam
structures are becoming more important because arch, double curvature arch, and buttress
dams require better foundations.
 4-1

4.0 GEOLOGICAL INVESTIGATIONS

As the safety of the structure is largely dependent on the foundation on which it is

constructed, it is very important that a capable Engineering Geologist is included in the
professional team as soon as possible to evaluate the foundations. It is then also his
responsibility to identify the geological conditions which could possibly lead to the failure
of the dam wall which may result in loss of life and serious damage. Designers are to a
large extent dependant on the information and advice supplied by the Geologist, and it is
unthinkable that any dam project should be attempted without the assistance of a
Geologist.

During the creation of a gravity dam three phases are distinguished as far as Geological
investigations are concerned in the Directorate of Water Affairs, namely:

(i) The planning- and preliminary design phase

(ii) The final design phase
(iii) The construction phase

4.1.0 The planning phase

After the different alternative dam sites have been evaluated by the Planning Section, the
geological investigation is carried out. The purpose of the investigation is to evaluate the
foundations geologically and to provide information to the Design Engineer so that he can
proceed to do a preliminary design. Additionally, he must give professional advice
regarding the feasibility of building a dam structure on the foundation.

The importance of the investigation cannot be overemphasized because it largely determines

the final position of the structure. If the geology is incorrectly evaluated during this phase, it
could mean that either the choice of site must be changed, or design changes must be made
which could bring about serious cost increases and which could even render the scheme
uneconomical.
 4-2

It is not far-fetched to say that the more information that can be gathered during this phase,
the better it will be for the successful completion of the scheme.

The report which is the product of this investigation must report on the following aspects of
the geology.

4.1.1 Reconnaissance study

Before the drills and other equipment arrive on site the Engineering Geologist must carry
out a reconnaissance study. This study must include the following aspects:

(a) He must do a complete literature study regarding the geology of the area where the
proposed site is located. This includes inter-alia the stratigraphy, slope and reach,
bearing capacity, strength, thickness, and faults of the various rock types. Aerial
photos may also be used to plan his investigation. An additional important step at
this stage is to carry out an on-site investigation and based on the findings, to do the
planning for the drilling of boreholes and the digging of test pits and/or trenches.

(b) During the reconnaissance survey the geologist should already consult the records
and literature to determine the possibility of the alternative sites lying in seismically
active areas. If they do, he must carry out further research and add these findings to
his report.

(c) During his reconnaissance visit he must also look at the availability, quality,
quantity, and proximity of the different construction materials which will be needed.
He can only do this properly if he has already researched what to expect at the
various sites. The geologist must also have the required knowledge to evaluate the
different types of materials to determine if they will be suitable for their ultimate
purpose. Building materials are mainly concrete aggregate and sand in the case of a
concrete dam.
 4-3

(d) During the reconnaissance phase it may not yet be known what type of dam will
ultimately be built and the Geologist must take this into consideration when he
reports on the available materials, borrow areas, quarries, etc.

4.1.2 The borehole survey

The next step in the geological survey is to get drill rigs on site and to drill according to the
predefined plan. The extent of the survey depends largely on the importance and cost of the
scheme. It is self-evident that more drilling will be required for a high arch dam than for a
low gravity dam. It is also evident however, that the more information the geologist can
gather at this stage, the better the preliminary design can be done and the more accurate the
final cost estimate will be. The expenditure at this stage will usually not exceed 1% of the
final cost of the scheme and no stone must be left unturned to complete the full geological
picture.

Drilling is usually carried out with a diamond drill bit and drill cores are obtained. Drill
cores are placed in boxes and are safely stored. The lengths of cores are determined and are
also written on the boxes. Hole depths are usually up to 20 m in solid rock when the rock
type is consistent. Depths may vary based on the type of dam, height of dam, and geological
characteristics of the site. Water pressure tests are also carried out to determine the
permeability and grout-ability of the rock types. Holes are usually drilled vertically, but
holes can also be drilled at an angle if it is felt that extra information can be obtained. There
is no rule determining the number of holes to be drilled, but the author believes that a
minimum of one hole every 30 m is required. Holes are usually drilled along the proposed
centreline of the dam to obtain a longitudinal section.
 4-4

Extra holes must also be drilled perpendicular to the centreline to enable the geologist to
prepare cross sections of the geology. More holes will provide better data and a grid of
holes is preferable.

Based on an initial interpretation of the borehole results it is advisable to now dig several
inspection trenches in a parallel and perpendicular fashion to confirm the results of the
drilled cores. The trenches will also enable the geologist to confirm whether the borehole
results are representative of the site or are local features. This is important because it is not
practical to drill holes at spacings of one metre. It must however be clearly stated that the
trenches are not a guarantee that other conditions will not be found when full excavations
are carried out. It is even possible that large shear zones will only be found when the real
excavation is carried out.

While the drills are on site a number of cores can also be obtained from the planned quarry
and/or borrow pit areas to allow the geologist to evaluate the quality and quantity of
construction material available.

4.1.3 Results, conclusions, and recommendations

The end result of this phase’s investigation is the report in which the results are documented,
conclusions are drawn, and recommendations are made.

The practice in the Directorate is that the results of each borehole are given on a geological
datasheet. On the datasheet the following information, inter alia, is indicated for each
borehole:

Type of drill rig, drill operator, coordinates of borehole, height and depth of borehole,
orientation of borehole, datum, percentage of core recovered, water test results, rock type,
particle size, discontinuity spacing, degree of weathering, strength, etc. Figure 4.1 shows an
example of such a datasheet. Photos of the cores are also included in the report.
 4-5
FIGURE 4.1: Example of borehole datasheet
 4-6

With the aid of all the borehole results, the Geologist then draws sections (longitudinal- and
cross sections) of the geology and indicates all required detail such as proposed excavation
levels, water table, weathering, shear zones, rock types, etc. on these.

In his report the Geologist also comments on the following aspects:

(a) He discusses the geological formations and comments on whether it is desirable as a

foundation. He gives professional advice regarding the strength and bearing capacity
of the foundation. He also gives his opinion about the grout-ability of the foundation
and whether this is desired based on the water tests.

(b) He recommends which of the alternative sites is most suitable.

(c) He broadly comments on the possibility of seismic activity in the area.

(d) The Geologist recommends an excavation depth.

(e) He comments on the potential erosion of the riverbed to enable the designer to
decide whether an apron is required or not.

(f) Finally, the Geologist gives his opinion about the available construction material at
the alternative sites with regard to the type, weathering, hardness, grading, shape,
organic content, quantity, quality, and proximity to the site.

(g) As soon as his report is issued, the Geologist’s task is temporarily complete although
he must remain available to answer any queries from the design engineer.
 4-7

4.2 The final design phase

After a site is selected the next phase of the geological investigation is carried out where
every aspect of the geology is examined in detail and reported on.

The first step in this phase is to determine whether additional drilling needs to be carried out
to clarify any uncertainty which may result from the provisional survey or the specific
structure which is proposed. Since the site selection has been finalised at this stage, it is
desirable to have a prepared excavation drawing so that portions of the excavation can be
done to determine if the geology corresponds to the predictions or if additional drilling
needs to be done. If additional drilling is required, it is usually done to determine specific
information regarding the foundation.

During this phase it may also be necessary to dig tunnels into the steep flanks to further
examine the foundations and to obtain in-situ material for testing. (For example, shear tests
and to determine certain design moduli).

Geophysical investigations (for example micro seismic and magnetic seismic) can be
invaluable to complete the geological picture of the site. These must however be read in
conjunction with the borehole results and inspection trench results.

By sending samples of the foundation to a laboratory, the shear strength, internal angle of
friction, and permeability of the foundation material can be determined, these will be used
during stability analysis.

If the structure lies within a seismically active area this issue must now be thoroughly
examined and reported on in detail.
 4-8

From the results of the initial and possibly also later water tests the Geologist should make
recommendations regarding test sections to determine the grout-ability of the foundation.
To do this, he must study the pattern, orientation, and continuity of the seams and cracks and
possibly also shear zones.

In this phase the borrow areas must also be examined in detail to obtain finality regarding
the various construction materials. Problems which may arise must be pointed out and
reported. Three or four times the required volume of material should be available because it
often happens that material is rejected during construction.

The report should also contain detailed sections and a final excavation depth must be
proposed for the structure. The report must contain all the information required by the
designer to complete the final design. It is important during this phase for there to be
continuous professional cooperation and communication between the Engineer and
Geologist.

4.3 The construction phase

The Geologist’s main function during this period is to inspect the foundations and to report
promptly about any observed weakness. It is important that he monitors the excavations and
compares results with borehole results and observations in test pits and trenches, he must
report on any differences which may endanger the safety of the structure. It is also the
Geologist’s responsibility to map the geology of each block and to update the geological
drawings which must be included in his completion report when the scheme is completed.

Prior to placing of concrete in each block’s foundation the Geologist must provide his
written permission to do so. Photos are taken of each completed excavation portion and are
held for archival purposes.
 4-9

The Geologist must subsequently supervise the grouting of the foundation on behalf of the
Design Engineer, and he must provide the Engineer with advice regarding this activity to
obtain the best results. The preliminary grouting specification is no more than a guide and it
must be adapted in the light of the information obtained from test grouting.

Effective communication between the Designer, Contractor, and Geologist is of the utmost
importance during construction to facilitate the effective and economical completion of the
work.

During the first impoundment of the dam the Safety Engineer should visit the dam
periodically to detect possible leaks and to report on whether it is safe to continue with
impoundment. It may also be necessary to carry out further grouting of the foundation, with
which the Geologist can assist.

The Geologist’s final task is to compile a completion report in which the complete
geological picture must be painted.
 5-1

5.0 DESIGN STANDARDS

5.1 Introduction

This portion of the treatise is compiled based on design standards which are already used
in the Directorate (501, 502).

The design criteria is that stresses in the dam section under different load conditions may
not exceed certain safe allowable stresses.

5.2 Loads

5.2.1 Hydrostatic loads

Four hydrostatic loads are specified which correspond to the water depth in the dam during
the full supply-, high flood-, and probable maximum flood conditions as well as the toe
water depth.

[Link] Full supply level

In this case the water depth is the difference in height between the full supply level
(usually the spillway crest height) and the lowest foundation level.

[Link] High flood level

The flood with a return period of 1 in 200 years is arbitrarily chosen as the high flood, also
called the design flood. The water depth in this case is the difference in height between the
water height in the dam during high flood conditions and the lowest foundation level.
Flood absorption calculations must be carried out to determine the flood height at the dam.
Computer program 110/005 is available to determine the flood height. If the actual inflow
hydrograph is not available, the Directorate suggests the following shape for the inflow
hydrograph (503):
 5-2

FIGURE 5.1: Inflow hydrograph

0,87
,
with

where Tc is the concentration time in hours (h),

L is the length of the longest water course (km), and
S is the dimensionless slope of the longest watercourse.

Usually, the flood characteristics for a specific dam are supplied by the Hydrological Section but if
they are not available the designer must determine them using the latest hydrological reports. Detail
of these methods is not discussed as it falls outside the scope of this treatise. These reports are
available from the Hydrological Section and the library.
 5-3

[Link] Probable maximum flood level

The progression of the probable maximum flood (PMF) through the dam is analysed to
determine the maximum flood height. In contrast to the design flood hydrograph, it is
usually found that lowering of the peak flow rate is insignificant because the volume of the
dam is small compared to the volume of the PMF flood hydrograph. The water depth is
equal to the difference in height between the flood height at the dam wall and the lowest
foundation level.

[Link] Toe water loads

Loads due to the water pressure against the downstream toe under the different flow
conditions as discussed in the preceding three paragraphs must also be considered when
determining the stability of the dam wall. The water depth against the toe is determined by
calculating the tailwater levels for flow downstream of the dam wall. Program 110/177 is
recommended for use*. The water depth used in the stability calculation is the difference
between the tailwater height and the lowest foundation level at the toe of the wall. The
downstream face of the dam is usually not vertical, and the load therefore also has a
vertical component which must be considered.

5.2.2 Own weight

The weight of the wall is calculated by assuming that the specific weight of the concrete is
24 kN/m3. This is higher than that of reinforced concrete but is justified because the wall
is constructed using mass concrete which could have a maximum aggregate size of 120
mm or more (depending on the type of mixer used) resulting in a higher unit density. The
masses of sluice gates, piers, bridge decks, etc. may be added to the own weight, but their
influence is generally negligible, and they can usually be excluded.

* This mainframe software no longer exists.

 5-4

5.2.3 Silt load

Horizontal pressure because of silt increases linearly with depth like a fluid with a density
of 4 kN/m3. The vertical load as a result of silt is determined as for soil with a weight of
9 kN/m3 (501). Both these loads exclude the load as a result of water, and this must be
added if the water loads are not regarded separately.

For design purposes, the 100-year silt level is used. The Hydrological Section is
responsible for providing the silt load of the river. In this regard ‘Technical report number
61’ which covers sediment production in South Africa can be useful. ‘Technical report
number 63’ deals with the settlement of silt in the dam basin and can be used to determine
the silt depth at the dam wall. These reports are available from the Hydrological section.
If the above method is not used, the Design office method can be followed. In this method,
the 100-year silt volume is calculated and assumed to settle horizontally in the basin.
Three quarters of the height against the wall is then used as the silt design level. For
design purposes this will be acceptable because it is based on a conservative assumption.
When this silt level is reached, the safety of the structure must be re-evaluated.

5.2.4 Uplift forces

As a result of seams, cracks, and joints in the foundation and concrete, internal pressure
exists from the water on the foundation and in the concrete. Uplift forces act throughout
the entire body of the structure and under the base. For the design it must be assumed that
the uplift pressure acts over the entire area and not just a portion of the level under
consideration. Uplift forces are considered at any horizontal section at any level. Two
distinct cases are considered depending on whether the dam has a drainage system or not.
 5-5

(a) Partial uplift pressure

When a dam has a good cut-off curtain, foundation, and internal drainage system the
uplift pressure can be reduced. The Directorate assumes that the full water pressure at
the heel decreases linearly to a third of the difference between the upstream- and
downstream pressure at the drainage hole positions, and decreases linearly from there
to the water pressure at the toe of the dam.

FIGURE 5.2: Uplift pressure

 5-6

(b) Full uplift pressure

When there is no drainage system or if it is assumed that the drainage system is

blocked, full uplift pressure is assumed. In this case it is assumed that the uplift
pressure decreases linearly from the water pressure at the heel to the water pressure at
the toe of the wall. See the dashed line in Figure 5.2 for the full uplift pressure
distribution on the base of the dam.

5.2.5 Earthquake loads

Earthquakes cause accelerations in the earth’s crust which result in increased water- and
silt pressures which can be detrimental to the stability of the structure. Earthquakes must
therefore also be considered during the design of dam walls.

Horizontal and vertical accelerations take place, but in this Directorate only the effect of
the horizontal acceleration is considered. In the absence of specific seismological data, it
is assumed that earthquakes cause an acceleration of 0,1 g with a period of 1 second. The
effect of the earthquake on the hydrostatic load on the structure is calculated based on the
Westergaard theory. For the stability calculation it is assumed that uplift- and silt
pressures remain unchanged. Paragraph 163 of the first edition of “Design of Small Dams
(504)” (United States Department of the Interior) gives more information regarding the
topic. In the Directorate the Westergaard theory is applied as follows:
 5-7

0,82
"
0,1 usually
#
2
*
3

FIGURE 5.3: Earthquake loads

(a) A horizontal force with magnitude 0,1 W is applied at the centre of gravity of the
structure, where W is the total dead weight of the section (gates, piers, bridge deck,
etc. included).

(b) An additional horizontal load Pe is applied to the upstream face of the wall due to the
earthquake, at a height of 0,4 H above the base of the section. The symbols in Figure
5.3 are defined as follows:

Ce – dimensionless constant
α – acceleration constant due to earthquake
pe – additional hydrostatic pressure due to earthquake at depth y below water level
H – total water depth at wall
γw – specific weight of water (10 kN/m3)
a – acceleration due to earthquake
g – gravitational acceleration (10 m/s2)
Pe – total force due to additional pressure

Other loads act normally.

 5-8

5.2.6 Other loads

Other loads such as wind- and temperature loads are also present but can be ignored during
the stability analysis due to their negligible influence.

5.3 Load combinations

In the Directorate four different load combinations are considered for gravity dam
structures. These combinations are divided into two categories for normal and abnormal
conditions. It is evident that the designer would be willing to take a greater risk with the
abnormal cases than with the normal cases. Safety factors and allowable stresses are
therefore adjusted to make provision for this. The combinations are also selected so that two
loads with a low probability of occurrence are not combined in one load condition because
the chance of them occurring simultaneously is very small. The load combinations are as
follows:
 5-9

5.3.1 Normal category

A. Water height at the design flood level (1 in 200 year flood)

Own weight of structure
Toe water load
Silt load (Expected after 100 years)
Partial uplift force (If a drainage system is present)

FIGURE 5.4: Load combination A without a drainage system

 5-10

FIGURE 5.5: Load combination A with a drainage system

 5-11

5.3.2 Abnormal category

B. Water height at full supply level

Own weight of structure
Silt load (Expected after 100 years)
Partial uplift force (If a drainage system is present)
Earthquake loads

FIGURE 5.6: Load combination B without a drainage system

 5-12

FIGURE 5.7: Load combination B with a drainage system

 5-13

C. Water height at the design flood level (1 in 200 year flood)

Own weight of structure
Toe water load
Silt load (Expected after 100 years)
Full uplift force resulting from blocked drainage system

FIGURE 5.8: Load combination C (Blocked drainage system)

 5-14

D. Water height at probable maximum flood level (PMF)

Own weight of structure
Toe water load
Silt load (Expected after 100 years)
Partial uplift force (If a drainage system is present)

FIGURE 5.9: Load combination D (Probable maximum flood)

If a dam has crest gates, the full hydrostatic load in cases A, B, and C must be calculated
for closed gates. This is justified because the possibility exists that the gates are not all
opened during a flood or that the gates cannot be opened due to a mechanical or electrical
problem.
 5-15

5.4 Analysis of the structure

In the Directorate, the above combinations (A to D) are analysed for the following two
cases:

(i) The stress distribution is calculated at the base of the dam using the beam theory. Uplift
forces are applied as external forces. Calculated stresses must then fall within the
allowable limits (See Chapter 5.5.1).

(ii) Safety factors against sliding are calculated using the following equation:

- + / tan 1
,

with
C – cohesive strength (kPa)
A – area or the base under consideration
N – vector sum of all vertical forces including uplift forces

ϕ
H – vector sum of all horizontal forces
– angle of internal friction

The following values may be used if no laboratory results are available:

C for concrete 3 000 kPa

tan ϕ 0,8 for rock and concrete individually as well as for interface between
concrete and rock

C for rock depends on site (Use 1 000 kPa as a conservative low value for rock)
 5-16

Analysis over the past few years has shown that load combination B is mostly the critical
case and designers are advised to analyse this case first. (See the Sample calculation at the
end of this chapter). In the analysis of the dam wall, the following values may be used:
Weight of concrete 24 kN/m3
Weight of water 10 kN/m3

5.5.0 Design criteria

5.5.1 Allowable stresses in the concrete

Normal (A) Abnormal (B, C, and D)

Tension in the concrete None * 0,2 MPa maximum
Compression in the concrete 0,25 U 0,25 U

U – Compressive strength of concrete after 1 year

* This tension can only be allowed if the foundation is good and does not have many
horizontal seams.

5.5.2 Safety against sliding

The following minimum values for the safety factor F against sliding must be obtained:

F ≥ 4 for combination A
F ≥ 2 for combinations B, C, and D
 5-17

5.5.3 Remark

The above safety factors and allowable stresses only apply for dams which are constructed
on good rock foundations. They do not apply for structures built on alluvium or sand.

5.6 Sample calculation

Test the structure in the following diagram for stability:

C = 1 000 kPa
U = 25 MPa
tan ϕ 0,8
 5-18

Experience has shown that load combination B yields the critical stresses. This case is
therefore examined first to save time. If the chosen section does not meet the standards set,
the required adjustments are made to the section, and it is re-tested. Other load
combinations are subsequently analysed.

The first step is to draw the free body diagram for the structure indicating all the loads
(Figure 5.10).

FIGURE 5.10: Free body diagram

 5-19

Lever arm
H V MB
around B
(kN) (kN) (kNm)
(m)
31
*3 4 10
2
4 805,00 - 10,33 -49 651,67

1
*5 4 4 4 20
2
800,00 - 6,67 -5 333,33

78 4 4 34 4 24 - 3 264,00 23,00 75 072,00

0,1 78 326,40 - 17,00 -5 548,80

1
7 4 21 4 30 4 24
2
- 7 560,00 14,00 105 840,00

0,1 7 756,00 - 10,00 -7 560,00

1 31
98 4 21 4 4 10
2 3
- -1 085,00 14,00 -15 190,00

31
9 44 4 10
3
- -413,33 23,00 -9 506,67

2 31
9 4 4 10 4 4
3 2
- -413,33 23,67 -9 782,22

2
*: 4 0,82 4 0,1 4 31 4 10
3
525,35 - 12,40 -6 514,34

7 212,75 8 912,33 71 824,97

 5-20

The next step is to calculate the eccentricity e.

; ; >?@
: <= <
2 2 >A
12,50 < 8,06
4,44 m

From beam theory it follows that tension will occur in the heel if the resultant of all forces
lies outside the middle third of the base.

; 25
4,25 E 4,44 ⇒ tension
6 6

The stress distribution over the base must now be calculated. From beam theory it follows
that:
>A 6:
I J<1 K L
; ;
8 912,33 6 4 4,44
I J<1 K L 4 10N MPa
25,00 25,00
I 0,02 MPa tension and 0,74 MPa compression
 5-21

The tension is acceptable because 0,02 < 0,2 MPa (5.5.1)

The compression is acceptable because 0,74 < 0,25% × U = 6,25 (5.5.1)

The safety factor against sliding can now be calculated.

- + A tan 1
,

1 000 4 25 + 8 912,33 4 0,8

7 212,75
∴F 4,45 > 2 ⇒ safe against sliding 5.5.2

The section is therefore safe for combination B. Similarly, the stability of the section must
now be tested for combinations A, C, and D.
 6-1

6.0 DAM SPILLWAYS (FLOOD OUTLETS)

6.1 Purpose of spillways

The purpose of a dam spillway is to return surplus water, which cannot be handled by the
outlet system, safely and in a controlled manner to the river channel or drainage system
without endangering the stability of the dam or it’s foundation.

6.2.0 Types of spillways

The following types have been used in the Directorate although some will not be used at a
gravity dam because they cannot be adapted to its specific shape.

6.2.1 Ogee spillway

FIGURE 6.1: Ogee spillway

 6-2

This type of spillway is used most often in the Directorate of Water Affairs. The shape of
the crest corresponds to the trajectory of the bottom face of the free-falling water nappe
over a sharp crested weir.

Methods of energy dissipation which can be used with an Ogee spillway are splitters,
trajectory buckets, slotted buckets, or hydraulic jump stilling basins. Methods of energy
dissipation are described in detail in Chapter 9. In the Directorate the most popular method
is to use splitters, on condition that the dam is high enough for their use. Ogee spillways
are cheap although the upper concrete lifts require curved shuttering and compaction is
often difficult because the shape it not easy to construct.

Ogee spillways are usually positioned in the river portion of the dam with concrete training
walls on either side to channel the spilled water into the river course. Examples of dams
where this type of spillway was used includes, inter alia the following: Fanie Botha dam*,
Blyderivierpoort dam, Erfenis dam, Hendrik Verwoerd dam†, and P.K. le Roux dam‡. The
spillway is usually made as long as possible to reduce the flow depth. It may be necessary
to construct return channels parallel to the dam centreline to convey flood water from the
flanks back to the river course. This principle is used at the Vaal dam spillway. The
heights of training walls are determined from model studies. The design of an Ogee
spillway will be examined later in this Chapter (6.8.6).

* Fanie Botha dam is now called Tzaneen dam.

† Hendrik Verwoerd dam is now called Gariep dam.

‡ P.K. le Roux dam is now called Vanderkloof dam.

 6-3

6.2.2 Free overfall spillway

This is a spillway where the flood water falls freely from the crest of the wall and plunges
into the river. This type of spillway is not used with gravity dams, being prevented by the
sloped downstream face of the wall. It is normally used with arch- and buttress dams.

FIGURE 6.2: Free overfall spillway

 6-4

6.2.3 Morning Glory spillway

This is a spillway which is used primarily at earth- and rockfill dams.

FIGURE 6.3: Morning Glory spillway

The practical maximum flow rate for this type of spillway is about 1 500 m3/s which can be
a disadvantage. Three different flow modes can occur depending on where the control is, at
the spillway, in the mouth, or at the outlet. Model studies are needed to refine the design of
this type of spillway.
 6-5

6.2.4 Side channel spillway

FIGURE 6.4: Side channel spillway

This type of spillway is positioned on one of the flanks of the dam, more- or less
perpendicular to the centreline of the dam. It is normally used with earth- and rockfill
dams. Design theory for this type of spillway is well established and is based on the
conservation of continuity and linear momentum as the flow increases along the channel.
It is assumed that all energy is dissipated over the crest and that only axial forces cause
flow. Readers are advised to refer to “Design of Gravity Dams (601)” for more
information regarding this type of spillway.
 6-6

6.2.5 Siphon spillway

FIGURE 6.5: Siphon spillway

This type of spillway is used where it is desirable to keep the water level constant during
floods. Discharge increases rapidly with an increase in water level, but the spillway cannot
discharge more than its design flow. Construction of this type of spillway is difficult and it
is seldom used.

6.2.6 Auxiliary spillways

Auxiliary spillways may sometimes be constructed on the flanks, or in saddles occurring on

the circumference of the dam basin which can discharge excess water back to the river along
a different water course or ravine. The crest height of these spillways can be higher than the
high flood level but lower than the non-overspill crest height to allow it to fail when
overtopped. It is important for the foundation of the auxiliary spillway not to erode deeper
than the full supply level of the dam, to prevent water loss. This type of spillway is usually
built where the probable maximum flood (PMF) is large compared to the design flood.
Economy therefore plays a large role.
 6-7

6.2.7 Controlled spillways (Sluice gates)

The most commonly used sluice gates in the Directorate are vertical- and radial gates but
Fishbelly Flaps are also sometimes used. Radial gates are preferred due to the smaller
lifting forces, better flow characteristics, and the fact that no side grooves are required which
need to be waterproofed. To prevent negative pressures on the crest when sluice gates are
partially open, the spillway crest is made flatter downstream of the sluice gate than that of a
normal spillway. Accurate control and operating systems are required, and they are often
determined from model studies. The sequence in which sluice gates are opened to obtain the
best downstream flow pattern is important. During floods it is possible to release water pre-
emptively and thus reduce the flood peak. Unfortunately, this method can also increase the
flood peak if the incoming flood peak is calculated incorrectly. Sluice gate vibrations can
occur for smaller openings and the gates should then be opened slightly more. It is also
better to open a few sluice gates completely rather than to open all gates partially. Sluice
gates can be used in three ways at spillways:

(a) Crest gates

Crest gates are constructed on top of the spillway and by opening them, surplus water
can be released from the dam over the normal spillway. This method has, for example,
been used at Vaal dam, Hartebeespoort dam, and Doorndraai dam.
 6-8

FIGURE 6.6: Crest gate

Installing crest gates on an existing dam is a popular method for raising a dam (as with
Vaal dam and Doorndraai dam).
 6-9

(b) Bottom outlets

FIGURE 6.7: Bottom outlets

Large openings are formed in the bottom of the dam wall which are controlled by gates.
Examples of such outlets are at Floriskraal dam and Gamkapoort dam. Such openings
can be used for river diversion during construction. These sluice gates should be opened
regularly to avoid possible silt build-up. An emergency gate is essential to allow
maintenance work to be carried out on the control gate. This type of sluice gate is often
used together with uncontrolled spillways, and this gives greater control of the system.
As a result of the large pressure head on the opening, the discharge per unit width is
considerably larger than that of a normal spillway. It is therefore useful where the width
of the valley is too narrow to accommodate a normal spillway.
 6-10

(c) Chute spillway

These are bottom outlets which are constructed on the flanks of the dam through which
flood water is returned to the river channel using chutes. The most well-known
examples are at Hendrik Verwoerd dam† and P.K. le Roux dam‡. Large gates are
normally used resulting in large discharge capacity. The slope of the flank must be
gradual to facilitate construction and restrict flow velocity, thus preventing cavitation
damage. At the end of the chute a ski-jump (above the tailwater level) or a roller bucket
is built to dissipate the energy before the water is returned to the river. A ski-jump will
cause the water to erode a stilling basin in the riverbed until equilibrium is achieved.
The position of this stilling basin must be such that it will not undercut and destabilize
the dam. The roller bucket can also only be used if the foundation downstream is not
prone to erosion.

6.3.0 Layout and choice of spillway

The layout and choice of spillway depends largely on the size of spillway required, which
depends in turn on the size of flood which must be passed. Topography and valley shape
also influence the choice. A wall in a narrow valley will probably require bottom outlets,
while a wide valley may accommodate an ogee spillway. Foundation quality can have a
large influence on the energy dissipation method which may in turn influence the type of
spillway, although the foundation can also be protected with concrete. The slopes of the
flanks are important when choosing a chute spillway because high water velocities can
cause cavitation. The river diversion method may also influence the type of spillway
chosen. The channel of a bottom- or morning glory outlet can easily be used to divert the
river during construction. Tailwater depths must also be considered when making a choice.
A high existing tailwater level can form a natural stilling basin while a low tailwater level
will require the construction of an end wall to form a basin.

† Hendrik Verwoerd dam is now called Gariep dam.

‡ P.K. le Roux dam is now called Vanderkloof dam.

 6-11

6.4.0 Components of spillways

Five components, which are not equally important, can be defined namely:

(a) Approach channel: This occurs with bottom outlets and chute outlets. Water
velocities must be low enough to prevent erosion of the channel sides.

(b) Control structure: This is the crest of the spillway or the control gate itself.

(c) Flood channel: This is the chute or downstream surface of the dam which transports
the water to the energy dissipator.

(d) Energy dissipators: Structures which are shaped and placed to reduce the water
energy. This includes splitters, stilling basins, hydraulic jumps, etc.

(e) Discharge channel: The channel which transports the water with reduced energy,
back to the river channel.

6.5.0 Model studies

Where the possibility exists that the riverbed or banks can be eroded or where a complex
flow pattern is expected, it is desirable to carry out a model study to examine the effects of a
design. Model studies can also be used to determine the order in which gates should be
opened to optimise flow and reduce spray on the outlet system. Model studies are also
extremely useful to determine if the energy dissipators are operating as planned. It is often
necessary to refine the design using the model in order to improve the system.
 6-12

6.6.0 Controlled* versus uncontrolled spillways

A controlled spillway is one where some form of mechanism may be used to alter the crest
position. An uncontrolled spillway’s crest is at a fixed level. The following table shows a
comparison between the two types which may influence choice:

Uncontrolled spillways Controlled spillways

No operator is required. Operator is required.

Debris is not a problem. Debris can be a problem.
No electrical power is required. Electrical power is required.
Simple system. Often complex.
Discharge < Inflow, so no legal problems. Could cause larger floods resulting in legal
No control over water level during floods. claims.
No pre-emptive discharge possible. Control over water level until capacity of
Cheap spillway. gates is exceeded.
Pre-emptive discharge is possible.
Expensive system.

It is often possible to combine the two systems at a dam which ensures better control.

6.7.0 Spillway design standards

The following design standards are applied in the Directorate.

(a) The spillway (controlled, uncontrolled, or combined) must be able to discharge the
1:200 year flood safely for a concrete dam. Spillways for earth- and rockfill dams
must be able to discharge the 1:500 year flood safely. This is not necessarily
economically justifiable, but experience has shown that this forms a good design
standard. In the current system there are too many unknown factors to carry out a
meaningful hydro-economic analysis. The effect of flood absorption is not
considered because the volume of the incoming flood (1:200 year) is usually so large
compared to the capacity of the dam that the absorption which occurs can rather be
regarded as an additional factor of safety. The level to which the water rises during
the 1:200 year flood (design flood) behind the spillway is known as the high flood
level (HFL). The spillway profile is also determined using the design flood.

* The current practice in the Department is to avoid the use of controlled spillways completely. Many existing dams with
controlled spillways are having these replaced with alternatives, such as labyrinth or piano key spillways.
 6-13

(b) The freeboard height to provide for waves resulting from wind
(Usually 1,0 to 2,0 m) determines the non-overspill crest height.

Chapter 7 contains information pertaining to the wave height.

(c) Finally, the passage of the probable maximum flood though the dam is analysed.
The rule here determines that for an earth- or rockfill dam the full flood must be
handled by the spillway without any discharge over the non-overspill crests unless
these are auxiliary spillways. For gravity- and concrete dams flood water may
discharge over the non-overspill crests during the probable maximum flood, but this
may not lead to failure of the dam.

FIGURE 6.8: Free board sketches

To determine the flow depth over the concrete non-overspill crest, the total length of this
portion is regarded as a broad crested weir for discharge purposes.

A complete design of an Ogee spillway is now shown as an example.

 6-14

6.8.0 Design of Ogee spillway

The symbols and methodology of “Design of Gravity Dams (601)” will be used.

6.8.1 Design information

(i) 1:200 year flood peak Q = 1 315 m3/s

(ii) Probable maximum flood peak Q = 5 400 m3/s
(iii) Height of crest above riverbed 25 m
(iv) Nett length of spillway 120 m (Eight blocks of 15 m each)
(v) The gravity dam has a vertical upstream face.
(vi) No provision needs to be made for raising of the dam
(i.e., No piers or spillway profile adjustment for gates)
(vii) Splitters will be used for energy dissipation.
(viii) Length of non-overspill crest is 800 m.

6.8.2 Determination of effective spillway length

Piers and side wall contractions on the spillway reduce the effective length. The effect can
be calculated using the following equation:

L = L’ – 2 (N Kp + Ka) He

L = Effective length of spillway crest

L’ = Nett length of spillway crest
N = Number of piers on spillway
Kp = Pier contraction coefficient
Ka = Abutment (side wall) contraction coefficient
He = Actual head on crest
 6-15

The pier contraction coefficient, Kp, is affected by the shape and position of the pier nose, the width
of the pier, the energy head compared to the design energy head and the approach velocity. For
conditions at the design head, Ho, the following pier contraction coefficients may be assumed: -

FIGURE 6.9: Pier contraction coefficient, Kp

The abutment contraction coefficient, Ka is affected by the shape of the abutment, the angle
between the abutment wall and the flow direction, the head in relation to the design head, and the
approach velocity. For conditions at the spillway design head, Ho, the following abutment
contraction coefficients may be assumed: -
 6-16

FIGURE 6.10: Abutment contraction coefficient, Ka

For the example we assume that the abutment radius is 0,3 Ho.
Then:
L’ = 120 m
N=0
Ka = 0,10

i.e. L = 120 – 2 (0,10) He = 120 – 0,2 He

 6-17

6.8.3 Determination of the spillway design head (Ho)

Discharge over an ogee spillway is given by the following equation:

\
[ :]

where Q = discharge in m3/s

C = variable discharge coefficient
L = effective spillway crest length in m
He = actual head being considered on the crest, including the kinetic energy height,
Ha, due to the approach velocity

At the design head this equation is as follows:

\
[ ^ ^]

where Co = 2,18 when He = Ho

and L = 120 – 0,2 Ho
\
i.e. 1 315 2,18 120 < 0,2 ^ ^]
and Ho = 2,944 m satisfies the above equation.

This means that the high flood level (design flood level) is 2,944 m above the full supply
level if we assume that the approach velocity is low. This is acceptable if the height of the
wall (P) is large relative to the spillway design head (Ho). If this is not the case, then the
approach velocity must be calculated to determine its influence on the spillway head.

6.8.4 Determination of discharge at levels other than Ho

Due to the effective spillway crest length (L) and the discharge coefficient (C) being
functions of Ho, the rest of the discharge curve must also be determined.
 6-18

For spillway heads lower than Ho, C is smaller than Co and for spillway heads larger than
Ho, C is larger than Co. The ratio of C/Co versus He/Ho is shown in Figure 6.11 which is
reproduced from “Design of Gravity Dams (601)”. The ratio can also be calculated
mathematically from the following equation:

:
^ _0,728 + 0,272` a
^

where C = Discharge coefficient at actual head on crest (He)

Ho = Spillway design head
Co = 2,18 = Discharge coefficient at design flood
He = Actual head on crest

The above equation may be simplified by inserting Co = 2,18 then:

:
1,587 + 0,593`
^

For the example, the following then applies: -

[ : 8,

:
[ _1,587 + 0,593` a b120 < 0,2 :c : 8,
2,944

By substituting the various values for He in this equation, the discharge table can be
calculated e.g.:
 6-19

FIGURE 6.11: Discharge coefficients for other than the design head

FIGURE 6.12: Factors for definition of nappe-shaped crest profiles

 6-20

He (m) Q (m3/s) ΔQ m3/s

0,2 18,7
0,4 54,8
0,6 103,3
0,8 162,6
1,0 231,5 231,5
1,5 442,1
2,0 702,2 470,7
2,5 1 007,8
3,0 1 356,0 653,8
3,5 1 744,8
4,0 2 172,5 816,5
5,0 3 139,6 967,1

From the table and Figure 6.13 it is evident that the discharge at low spillway head is small
but that it increases rapidly as determined by the equation.
 6-21
FIGURE 6.13: Spillway head discharge curve (Example)
 6-22

6.8.5 Discharge head due to PMF

The total length of the non-overspill crest (NOC) is regarded as a broad-crested weir.
Calculation of the head is required in order to test the stability of the chosen section and to
determine the level at which water will flow over the NOC. If the flow depth over the
NOC is less than 1 m, the possibility of adding a concrete balustrade on the upstream side
should be investigated. If not, an aluminium balustrade should suffice.

The discharge equation which applies for the NOC is:

Q = 1,70 L H 1,5

where Q = Discharge in m3/s

L = Total length of NOC crest
H = The head behind the NOC

For the example, we assume that the NOC is 5 m above the full supply level. The method
which must be followed is to find the sum of the discharges over the spillway and the
NOC, and when this total is equal to the PMF the desired spillway length is found. The
following table indicates how this is done:

He (m) Q1 (m3/s) H (m) Q2 (m3/s) Q (m3/s)

5,0 3 139,6 0,0 0,0 3 139,6

5,5 3 677,0 0,5 480,8 4 157,8
6,0 4 249,0 1,0 1 360,0 5 609,0

From Figure 6.13 the spillway head is now read off. In this case the total spillway head is
5,95 m, or 0,95 m higher than the NOC.

With the heads determined for the design flood and the probable maximum flood, the
stability of the spillway and non-overspill sections can now be tested for the different load
cases.
 6-23

6.8.6 Determination of spillway profile

The spillway profile is determined according to the book “Design of Gravity Dams (601)”
from page 164 onward.

To find the optimum discharge, the lower nappe surface over a sharp-crested weir must be
simulated for the ogee profile. The shape of the profile depends on the head, the slope of
the downstream face of the spillway section, and the crest height above the floor of the
approach channel (which influences the approach velocity).

[Link] Simplified spillway profile

The following cross-section for the spillway profile is recommended for dams with a
vertical upstream face (601).

FIGURE 6.14: Simplified spillway profile

The above profile should be adequate for the provisional designs and for dams where the
refined spillway profile described below is not required.
 6-24

[Link] More accurate spillway profile

Based on an extensive study and using model studies, “Design of Gravity Dams (601)”
proposes the following profile:

= j
^ g h i
^
k
A"
*+ ^

k ^ ^

A"
"
2#

FIGURE 6.15: More accurate spillway profile

The symbols have the following definitions:

P – Height of crest above approach channel

Ho – Design head above crest (1:200 year flood)
Ha – Kinetic energy height due to approach velocity of water
K – Constant
n – Constant
Va – Approach velocity of spillway water

K and n depend on the downstream slope of the spillway and the approach velocity, Va of
the discharged water.
 6-25

The values of K and n can be read from Figure 6.12. Values for R1, R2, Xc, and Yc can be
calculated with the aid of Figure 6.16.

Figures 6.12 and 6.16 are only applicable to spillways with vertical upstream faces.
Sloping upstream faces are only used in exceptional cases and are therefore not included in
this treatise. The reader is referred to “Design of Gravity Dams (601)” if more information
about this type of spillway is required.

For the example the following applies:

k ^ ^
= 2,18 x 2,9441,5
i.e. q = 11,01 m3/s/m

k
Then,
A"
*+ ^

11,01
A"
25,0 + 2,944

i.e. Va = 0,39 m/s

A"
Then,

"
2#
i.e. Ha = 0,008 m ≃ 0
And,
"
0,03
^

From Figure 6.12 and 6.16 it then follows that:

K = 0,50
n = 1,87
 6-26

FIGURE 6.16: Determination of spillway profile constants Xc, Yc, R1, and R2
 6-27

m
0,283
^
therefore Xc = 0,850 m

n
0,127
^
therefore Yc = 0,375 m

o8
0,530
^
therefore R1 = 1,600 m

o
0,235
^
therefore R2 = 0,700 m

Substituting the values for K and n in the profile equation, we obtain the following:

= 8, p
0,50 J L
2,944 2,944

or 0,195 =8, p

By substituting values for x in the above equation, the profile can be plotted.

6.8.7 Transition to horizontal level

The curved spillway’s slope changes constantly up to the point where its slope is equal to
the downstream spillway slope. This point is calculated as follows:

Suppose the downstream slope is l vertical to m horizontal. For gravity dams, m usually
varies from 0,7 to 1,0 depending on foundation conditions.
 6-28

=
g j
We know that ^ ^

1 j
q ^ g r = jN8 q= s t
^
Differentiate y with respect to x:

q 1 j
^ g r = jN8 s t
q= ^
And

q u
q= v
But

u 1 j
^gr= jN8
s t
v ^
Therefore

This can be solved for x and then for y from the first equation which will yield the point of
transition from the parabolic curve to a constant slope.

For the example:

Assume m = 0,75

1 1 8, p
2,944 0,5 1,87 = , p
s t
0,75 2,944
Therefore

Yielding x = 4,427 m
and y = 3,149 m

6.8.8 Cavitation

Cavitation occurs when the pressure of the water is lowered to its vapour pressure. Water
vapour molecules are formed which then collapse in the high-pressure areas. This happens
very rapidly and shock waves are formed in the water which can damage the adjacent
concrete spillway.
 6-29

Model tests have shown that the design spillway head can be safely exceeded by 50%
before any possibility exists that cavitation can occur on the spillway (601). The designer
must however still consider the possibility that cavitation can occur. Fortunately, the
probability of a flood large enough to cause cavitation occurring is small and it’s duration
short.

6.9.0 Control gates

6.9.1 Radial gates (601, 602)

Radial gates are usually manufactured from steel in a cylindrical segment which is attached
to support bushes using radial arms. Bushes carry the full compressive force resulting
from the water. Due to the resultant of the compressive forces acting through the bushes,
the gate experiences no moments which need to be overcome when the gate is opened.
This is the fundamental principle of the radial gate. The lifting force to open the gate only
needs to overcome the weight of the gate and friction.

A radial gate is the ideal type of control gate to use when a spillway with a high discharge
capacity is required. Additional advantages of this type of gate are that they close under
their own weight and no side slots are required. They are also simple, reliable, low cost,
and require simple sealing details. Their main disadvantage is that they take up a lot of
space.

Vibration problems can be experienced at bottom outlets when the sluices are flooded from
the downstream side.
 6-30

FIGURE 6.17: Typical layout for a radial gate

6.9.2 Vertical gates

Vertical gates include all flat rectangular structural steel sluices which move vertically.
Such gates are usually supported by wheels which roll inside slots on either side of the
gate. As with the radial gate, it requires seals to waterproof the sides and bottom. The gate
is usually raised or lowered using cables or rods. Sometimes the gate is heavy enough to
close under its own weight.

A large space is required above the gate to lift it out of the water, which is a disadvantage.
The size of these gates should be limited to 20 m2 due to the large lifting forces required,
although the ones at Vaal dam are 54 m2.
 6-31

Vertical gates are sometimes preferred to radial gates because they require narrower piers
and therefore do not reduce the required spillway length.

FIGURE 6.18: Typical layout of a vertical gate

6.9.3 Fishbelly flaps (602)

This curved sluice gate hinges around its bottom edge. The position of the gate is adjusted
on one- or both sides using control mechanisms.

The greatest advantage of this type of gate are the long lengths which it can have, and the
fact that the load is transferred directly to the wall and not via a pier. It is not desirable to
use this type of gate on a dam with a low design flood because it would require a narrow
crest on which the gate will not fit well. Such gates should not be higher than 4 m. The
gate must be designed for the dynamic effect of the discharged water and vibrations may
present a problem. The bottom of the water nappe must be very well aerated.
 6-32

FIGURE 6.19: Typical layout of a fishbelly flap

6.10 Spillway profile of controlled spillways

FIGURE 6.20: Spillway profile of controlled spillway

 6-33

As with uncontrolled spillways, the high flood level is used as design flood to calculate the
spillway profile. The profile used is the trajectory of the lower water nappe face which is
formed when the gate is just partially opened. Such a profile has a parabolic shape which
is defined as follows:

=
= tan w +
4 ^5 w

where H is the total head, and

θ is the angle between the gate and the centreline

This profile is flatter than the normal ogee spillway and the discharge properties are
therefore less favourable. It is important to shape the profile like this to prevent negative
pressures from occurring at partial opening of the gates which can lead to cavitation and
gate vibration. To prevent this a slightly higher gate can be used at the downstream end of
the crest. This forces the flow lines downward and creates a positive pressure on the crest.

6.11 Discharge through a Radial gate

The discharge is given by the following orifice equation:

2
[ 2# s < t
3 8
 6-34

FIGURE 6.21: Discharge through a radial gate

The value of C can be found from Figure 6.22. For large schemes it is important to carry
out model studies to calibrate the discharge curve.

6.12 Discharge through a vertical gate

Figure 6.23 can be used to calculate the discharge (603).

 6-35
FIGURE 6.22: Discharge through a radial crest gate
 6-36

NOTES:

Q = FREE FLOW DISCHARGE AT HEAD H

QG = DISCHARGE AT HEAD H AND GATE OPENING Go

FIGURE 6.23: Discharge through a vertical sluice gate

 6-37

6.13 Emergency gates

To do maintenance on control gates, it is necessary to place emergency gates upstream of

the control gates. These are usually vertical gates which can consist of one- or more parts
and which are lowered in slots using a gantry crane. After the emergency gates are lowered,
maintenance or repair work can be started on the control gates. A dam should have at least
two emergency gates for security purposes and to hasten maintenance work. The emergency
gates can be stored in an opening in the wall or in position in one of the channels. This
second method is usually used if the emergency gates consist of several parts.

6.14 Loads on gates

The following loads must be considered:

(a) Normal loads resulting from the water must be determined using hydraulic
theory.

(b) Additional wave forces can act on a gate if the water surface of a dam has a
long enough straight-line wind fetch and strong winds occur.

(c) The effect of floating objects, for example tree trunks which can collide with
the gate must be considered.

(d) Torsion at piers must be considered where one gate is shut, and the adjacent
gate is fully open.

(e) Wind loads on an open gate must not be ignored.

 7-1

7.0 THE NON-OVERSPILL SECTION OF THE DAM WALL

The non-overspill section is the part of the dam wall which is usually constructed on either
side of the spillway on the sides of the valley. One or more blocks of this portion usually
houses the river outlet system of the dam. Access to the drainage gallery is also usually
obtained via this section. The stability of the non-overspill section is calculated using the
standards laid out in Chapter 5 of this treatise. Access to the upstream face of the outlet
system is gained from the crest of the non-overspill section (NOC).

7.1 Section selection

For aesthetic reasons and to facilitate construction, the downstream face of the non-
overspill section is chosen to coincide with the downstream face of the spillway section.
The following sketch clarifies this:
 7-2

FIGURE 7.1: Downstream face of spillway and NOC section

The crest width should be about 5 m to enable vehicle access to the crest. Handrails made
of aluminium or reinforced concrete must be erected on either side of the crest, with a
minimum height of 1,1 m. Aluminium handrails are easier to erect than concrete
balustrades which are usually precast, but costs must be determined before a final decision
is made.
 7-3

7.2 Freeboard*

FIGURE 7.2: Freeboard

The freeboard of a dam is the vertical distance between the high flood level (HFL) and the
crest height of the non-overspill section (NOC). Freeboard is necessary to prevent
unnecessary overtopping of the non-overspill section due to waves. Such untimely
overtopping can cause damage downstream of the non-overspill section which is not
desirable. The height of the waves depends on the wind fetch distance, wind speed, depth
of water, and duration of “storm” on the dam basin. Figure 7.3 can be used to calculate
wave heights for broad, open water surfaces. Smaller surfaces will lead to lower waves.
The wind fetch distance is the longest straight distance which exists on the dam surface
between the dam wall and dam bank. The wind speed must be the maximum which can
occur at the dam site. This information can be obtained from the nearest Weather office.
It is recommended that a minimum freeboard of 1,5 m is used for gravity structures.

* Freeboard calculations are now done according to the SANCOLD Guidelines on Freeboard for Dams Volume II (2011)
 7-4
FIGURE 7.3: Graph for determination of Wave Height and Wave Period
 7-5

7.3 Upper trajectory of spillway nappe profile and training wall design

Training walls are usually constructed between the spillway and non-overspill sections to
channel flood water over the spillway. These walls prevent water from unnecessarily
flowing over the boundaries of the spillway section. For the training walls to be effective,
they must be able to channel at least the 1 in 200 year flood over the spillway area. To
design the wall, it is necessary to determine the profile of the top trajectory of the flood
water nappe over the spillway. This can be done in two ways:

(a) The theoretical method

FIGURE 7.4: Theoretical method to determine upper nappe trajectory

 7-6

Using the energy equation, it follows that:

A
^+ + q cos
2#
If the approach velocity is ignored. The symbol definitions are as in Figure 7.4

From the continuity principle it follows that:

[
k Aq

Where Q is the 1 in 200 year flood and L is the nett length of the spillway.

q
arctan s t
q=
q = jN8
g/ h i
q= ^
But

= jN8
arctan shg r i t
^
Therefore

k
And A
q

By substituting this into the first equation it follows that:

k = jN8
^+ + q cos yarctan shg r i tz
2#q ^

By substituting values for y, we can solve for d from this equation to plot the profile.

Profiles for floods other than the design flood can also be determined by substituting the
appropriate He for Ho.
 7-7

(b) Model studies:

=8, 2 : ,
American engineers (701) have carried out model tests on spillways with the shape
which is often used and is quite common. The approach velocity
was also ignored in these studies and the results are reflected in Table 7.1.

FIGURE 7.5: Model study method to determine upper nappe trajectory

 7-8

Centreline of Gate
Without Bridge Piers Adjacent to Bridge Piers
Openings

H/He 0,50 1,00 1,33 0,50 1,00 1,33 0,50 1,00 1,33

X/He Y/He

-1,0 -0,490 -0,933 -1,210 -0,482 -0,941 -1,230 -0,495 -0,950 -1,235
-0,8 -0,484 -0,915 -1,185 -0,480 -0,932 -1,215 -0,492 -0,940 -1,221
-0,6 -0,475 -0,893 -1,151 -0,472 -0,913 -1,194 -0,490 -0,929 -1,209
-0,4 -0,460 -0,865 -1,110 -0,457 -0,890 -1,165 -0,482 -0,930 -1,218
-0,2 -0,425 -0,821 -1,060 -0,431 -0,855 -1,122 -0,440 -0,925 -1,244
0,0 -0,371 -0,755 -1,000 -0,384 -0,805 -1,071 -0,383 -0,779 -1,103
0,2 -0,300 -0,681 -0,919 -0,313 -0,735 -1,015 -0,265 -0,651 -0,950
0,4 -0,200 -0,586 -0,821 -0,220 -0,647 -0,944 -0,185 -0,545 -0,821
0,6 -0,075 -0,465 -0,705 -0,088 -0,539 -0,847 -0,076 -0,425 -0,689
0,8 0,075 -0,320 -0,569 0,075 -0,389 -0,725 0,060 -0,285 -0,549
1,0 0,258 -0,145 -0,411 0,257 -0,202 -0,564 0,240 -0,121 -0,389
1,2 0,470 0,055 -0,220 0,462 0,015 -0,356 0,445 0,067 -0,215
1,4 0,705 0,294 -0,002 0,705 0,266 -0,102 0,675 0,286 0,011
1,6 0,972 0,563 0,243 0,997 0,521 0,172 0,925 0,521 0,208
1,8 1,269 0,857 0,531 1,278 0,860 0,465 1,177 0,779 0,438

TABLE 7.1: Results of model study for top nappe trajectory

The author has compared the two methods for the section without bridge piers and they
differ by less than 5%. With the profile known, the training walls can now be designed as
indicated in the following sketch.
 7-9

FIGURE 7.6: Training wall layout

 8-1

8.0 OUTLET WORKS

8.1 Introduction

A dam is a man-made obstruction which is built in a water course to make use of the
impounded water for some or other purpose. To release the water safely through the dam
for its intended use, an outlet works is required. An outlet works is a combination of
structures and mechanical equipment which is provided for the safe control of the
impounded water.

The impounded water can be used for one or more of the following purposes:

(a) To supply water to the lower reaches of the river if the dam is used for flow
regulation or pollution control.

(b) To supply irrigation water to canals, pipelines, or the river.

(c) To release water to water purification plants for water supply schemes.

(d) To release water under pressure to turbines if hydro-electric power generation

is the objective.

An outlet works can of course also fulfil a combination of two or more of the above-
mentioned functions and may even be used for river diversion during construction.

The outlet works must naturally also be able to empty the dam within a reasonable
timeframe in case of an emergency or for urgent repair work to the upstream side of the
wall. This is especially important during the first critical filling of the dam when possible
flaws may surface.
 8-2

From the aforementioned, it is evident that the outlet works is essential for the useful
application of the impounded water. A lot of time and attention must be spent on this
system during the planning- and design phases to ensure a functional outlet works. It is the
part of the dam which will be in continuous operation and the success of the scheme will
depend largely on the proper functioning of the outlet works.

Since no two outlet works can be identical due to different needs and sites, the author will
only attempt to lay down the broad principles of the design of an outlet works. Certain
consistent aspects, such as properties of different types of valves, trash racks, etc. will
however be examined in detail.

8.2 General guidelines and principles for the design of outlet works

8.2.1 Capacity of the outlet works

In the Directorate, it usually a function of the Planning section to determine the required
outlet capacity of a dam. The minimum and maximum discharge rates are usually also
coupled to dam water levels. According to “Van Robbroeck (801)”, the following should
be considered when determining the capacity of the outlet works:

(a) Possible future raising of the dam.

(b) The possibility that an additional dam may later be constructed upstream, where the
water discharged from that dam will also have to be discharged by the older dam.

Allowance must naturally be made for possible future extension. When a dam is
completed, new water uses are stimulated from the dam for which no provision was made
in the planning phase and adequate provision should be made for this.
 8-3

During the design, provision is made to satisfy the entire requirement even if maintenance
is needed on the system. This usually means that the system is duplicated. This may
appear at first to be an uneconomical and unnecessary step, but in practice this has proved
to be justified. An organization responsible for providing water can be badly embarrassed
if it is unable to do this due to inadequate outlets. To provide additional outlets at a later
stage is extremely difficult and expensive.

When the capacity of the system is determined, river and canal losses, etc. must also be
determined in order for the outlet works to compensate for these.

8.2.2 Position of the outlet works

Various factors play a role to determine the final position of the outlet works. The
importance of these factors obviously varies from site to site and depend on the specific
requirement. These factors are as follows:

(a) If the outlet works needs to supply water to a canal, water purification plant, power
station, etc. it is desirable that the facility is constructed on the same side of the river
to avoid a river crossing. This may not always be possible and may result in
alternative positions needing to be considered.

(b) Access to the outlet works is of cardinal importance. The water control officer visits
this part of the dam regularly and it must be easily accessible. For maintenance
purposes it must also be easy to reach the system. Access is especially dependant on
topography, but also on where the main access to the dam will be located. It may
prove to be necessary to build a bridge over the river downstream of the dam, but
attempts must be made to avoid this. If a bridge is however required, it may prove
useful during construction of the dam.
 8-4

(c) It is most convenient to construct the outlet works next to the spillway because
the same stilling basin or apron can be used for energy dissipation where needed.
Experience has however shown that at least one dam block is required between the
spillway and the outlet works because water spray caused by the spillway creates
operational problems in the outlet house. In the past, outlet works were sometimes
placed in the middle of the river with the spillway on either side of it. This required
a tunnel through the wall to provide access to the outlet house, and a bridge over the
spillway to provide access to the intake tower which is expensive.

(d) The topography of the site plays an important role. The system must be positioned to
allow discharged water to be easily returned to the river channel, so the system must
be close to the river. The outlet works must also be placed in a position where dead
storage is limited to a maximum of 10% of the storage capacity. Storage space is
required for aquatic life, recreation, and siltation. Excavations may be allowed to
allow the outlet works to fit, but this should be avoided due to cost considerations.

(e) Where the outlet works forms part of the river diversion system, the outlets must be
placed as low and as close to the river as possible. It is however usually cheaper to
keep blocks low or to leave openings in the concrete which can later be filled for the
river diversion.

(f) A further aspect which must be borne in mind is the geology in the area of the outlet
works. If the rock is unable to absorb the eroding forces of the outlets, expensive
stilling basins and aprons must often be constructed to absorb the energy. The outlet
block must also be founded on good rock.

(g) An outlet works must preferably also be positioned where control valves will be
above tailwater levels. If this is not done, the valves must be specially designed to
function in submerged conditions.
 8-5

8.2.3 Abstraction systems

Sometimes the best quality water available in the dam must be provided to the consumer.
Storage levels in the dam can rise or fall depending on the season and the water quality can
vary with depth. It is therefore necessary to provide equally spaced inlets at different
depths in order to be able to provide the best quality water. This is called a multi-level
abstraction system.

Each of the inlets is supplied with an isolation valve to allow for selection of water.
Downstream of the isolation valves the inlets are joined to a common outlet pipe which is
also supplied with a control valve. One emergency gate is also provided to enable the
closure of the inlets at the different inlet levels to carry out repair- and maintenance work
to the isolation valves. The author is of the opinion that this system will become more
popular in future as a result of the decreasing quality of water due to pollution.

An alternative method is to have a single-level abstraction system. With this system the
inlets are also provided at various levels, but only the lowest inlet is used due to the others
being sealed with blank flanges. As silt levels build up, inlets higher up are sequentially
brought into use by removing the blank flanges and sealing the lower inlets.

Examples of various outlet works will be examined in greater detail later in this chapter.

8.2.4 Control systems

The control system which will be used depends on the water delivery system. When water
is provided to a purification works or power station, the control valve is situated at the end
of the pipeline or turbine. When water is delivered to a canal, the control mechanism is
usually at the transition of the pipeline to the canal. For river outlets the Directorate
usually places the control valve at the end of the pipeline.
 8-6

8.3.0 Components of the outlet works

For simplicity an outlet works can be divided into three components, namely the intake-,
pipeline-, and outlet sections which in turn have their own sub-components. Each
component will subsequently be examined, and the detail of each sub-component will be
discussed.

8.4.0 Intake section

The layout of the intake section depends on the type of abstraction system which is chosen,
either a single-level or multi-level abstraction system. The most notable sub-components
of the intake section are the approach channel, trash racks, emergency gates, gantry cranes,
etc. Other aspects which will require attention are the siltation, water proofing,
accessibility, ventilation, and drainage of the intake section.

8.4.1 Approach channel

This is the area immediately upstream of the intake structure. The bottom inlet pipe must
be placed at least 1 m above the access channel to limit the abstraction of silt as much as
possible. If steep slopes with unstable material exist near the intake section, these must be
removed before first impoundment because they can slip and block the inlets with
catastrophic results. The approach channel must be wide enough to allow a steady stream
of water to flow towards the inlets with a reasonably low velocity to prevent sediment from
being transported.

8.4.2 Trash racks

To keep larger logs and debris from entering the outlets works, a pre-cast reinforced
concrete trash rack is usually provided upstream of the inlets. To prevent debris from
gathering on the trash rack and being drawn in, the velocity downstream of the trash rack
must be limited to about 0,5 to 1 metre per second. Low velocities are also desired to keep
hydraulic losses through the trash rack low.
 8-7

Drawing 84 807/82 on page 8-39 shows an example of a coarse trash rack. For the
structural design of the trash rack, “Van Robbroeck (801)” indicates that the hydraulic loss
due to blockage can be up to 2 m and that the trash rack must be designed for this
differential pressure. Coarse trash racks are usually not removable. These trash racks are
lowered into slots which are cast into the concrete sides of the wet well.

Where the outlet works supplies water to a pumping station, turbine, or water purification
works a fine scren is usually provided between the coarse trash rack and the inlets. These
are usually made from aluminium to make them as light as possible and corrosion resistant.
The openings in these screens depend on the smallest opening in the valve, turbine, or
pump through which the water must pass. In sensitive cases an extra set of fine screens are
provided so that the downstream screen will catch any debris which may drop off when the
upstream screen is lifted by the gantry crane. Sometimes “trays” are provided on the
upstream sides of the screens to catch debris when they are raised. The openings in the
fine screens used by the Directorate are usually about 25 mm squares.

Channels for the fine screens are formed by attaching angle profiles, treated to prevent
corrosion (or made of corrosion resistant material) to the sides of the wet well using
corrosion resistant anchors. These channels can also be cast into the concrete itself.

[Link] Hydraulic energy losses through trash racks

These hydraulic loss calculations are based on model studies conducted by Kirschmer and
described by “Zowski (802)”. The loss equation is as follows:

•
} A^
∆ℎ gs t sin
~ 2#
 8-8

where: Δh = Energy loss through trash rack in m

t = Thickness of bars in mm
b = Clear space between bars in mm
Vo = Approach velocity in m/s

α
g = Gravitational acceleration in m/s2
= Angle trash rack makes with the horizontal
K = Shape factor constant

A definition sketch follows:

FIGURE 8.1: Trash rack shape factor constants

 8-9

The author has carried out calculations, and the energy losses determined using the above
equation differ very slightly from those in “Design of Gravity Dams (803)” and the
“Engineering Manual (804)”. The above equation is however the only one which takes the
slope of the trash rack into consideration, although it is usually installed vertically.

In closure it must be clearly stated that the success of trash racks depends largely on how
often and thoroughly they are cleaned.

8.4.3 Intake section of a multi-level abstraction system

Figure 8.2 on the following page shows a typical example of the intake section of a multi-
level abstraction system.

As Figure 8.2 clearly shows, the inlet pipes at different levels are joined to a common
outlet pipe. The inlet pipes are each controlled by a selection valve (isolating valve) which
allows the operator to choose the abstraction depth and thereby ensure that the best quality
water is consistently delivered. At the upstream end of the inlet pipes bellmouths are
provided to obtain the best flow conditions and to limit the hydraulic energy losses. More
about bellmouths in the pipeline section of this treatise.

Butterfly valves are normally used as isolating valves at the different intake heights. These
valves take up less space. Valves can be controlled locally or remotely. At present the
Operations section prefers the local control system because it requires less maintenance
and ensures that the water control officer visits this part of the system regularly.
 LEGEND:

1. Gantry crane
2. Grating
3. Handrail
4. Emergency gate
5. Operating platform
6. Ladder
7. Bellmouth
8. Access gallery
9. Trash rack
10. Fine screens
11. Dry well
12. Wet well
13. Common abstraction pipe
14. Isolating valve
15. Drainage gallery
16. Drainage pipe
17. Outlet structure

8-10
18. Outlet valve

FIGURE 8.2: Cross section through multi-level abstraction system

 8-11

Pipes are preferably placed in a “dry well” where they can be regularly treated against
corrosion and the pipes can even be replaced later if needed. In all systems, provision must
be made for the inside of the pipes to be inspected and even treated if required. It is of
course important that the well in which the pipes are installed is well ventilated to limit
corrosion. This can be done mechanically with a fan or by installing ventilation pipes.

Drainage of the well is important. Leakage can occur and standing water can hasten the
corrosion of mechanical components.

Access to the various valves is important and must be provided in a manner which will
make it easy to reach the components. Access is usually obtained via galleries in the wall.
Platforms and convenient ladders are then supplied in the wells. The material used for
manufacturing the platforms, ladders, etc. should be corrosion resistant and require the
least possible maintenance. If the dam is very high, an elevator can be considered in the
core of the dam with access at convenient intervals to the well.

The gantry crane over the well must have the capacity to handle all the components in the
well. This crane must also be designed to be capable of handling the emergency gate on
the upstream side of the various inlets.

8.4.4 Intake section of a single-level abstraction system

Figure 8.3 shows a typical example of the layout of the intake section of a single-level
abstraction system. As the drawing clearly shows, inlets are provided at various levels. In
contrast to the multi-level abstraction system the upper inlets are initially sealed off for
future use when the lower inlets have silted up.
 LEGEND:

1. Gantry crane
2. Grating
3. Handrail
4. Emergency gate
5. Provision for temporary platforms
6. Sealed pipes for future use
7. Trash rack
8. Fine screens
9. Wet wells
10. Bellmouth
11. Drainage gallery
12. Isolating valve
13. Outlet structure
14. Outlet valve

8-12
FIGURE 8.3: Cross section through single-level abstraction system
 8-13

This type of outlet system is usually used at river outlets where the quality of water that is
released is not very important. When outlets are provided for stabilisation of river flow
this type is also normally used.

Rough permanent concrete trash racks and removable fine screens also usually form part of
this type of outlet to protect the mechanical elements of the system.

The inlet pipes are provided with bellmouths to ensure better flow patterns. A gantry crane
capable of handling the different mechanical components such as the emergency gate,
flanges, floors, etc. must be provided. The emergency gate must be able to function at full
flow conditions to close the openings for maintenance and safety purposes.

Provision must be made to fit temporary floors in the well. The floors are necessary for the
maintenance work which will be done on the pipes and when required, to bring higher
inlets into operation. An alternative method is to lower a platform down the well to the
correct level using the gantry crane.

The walls of the well should be as least 1 m thick, but 1,5 m walls are preferred. Thicker
walls are easier to construct and there is a greater chance that the leakage through the wall
will be negligible due to the longer flow path. During construction of the walls precautions
should be taken to ensure a waterproof wall by vibrating the concrete thoroughly.
 8-14

8.5.0 Pipeline section

The pipeline section is the portion of the outlet system which connects the intake section,
and the outlet section of the outlet works using pipes through the core of the dam wall.
This pipeline usually starts with a bellmouth on the upstream end and ends with the control
valve on the downstream end. Between the ends of this portion we find isolating valves,
flow meters (sometimes), bends, and bifurcations.

8.5.1 Bellmouths

Bellmouths are provided at the upstream end of the pipeline for at least two reasons.
Firstly, to ensure better flow patterns with associated low energy losses. It is important to
have uniformly transitioning flow lines into the pipeline, especially at high velocities. If
release were to occur at the inlet end, serious cavitation could occur inside the pipe.

Secondly, the upstream face of the bellmouth is machined which makes it possible for the
emergency gate to seal tightly against it when required.
 8-15

The energy losses through the bellmouth are calculated as indicated in Figure 8.4 (803,
805, 806)

FIGURE 8.4: Bellmouth energy loss constants

Approximate values for k are substituted into the equation:

‚
∆ℎ •
2#
The symbols have the normal meaning.
 8-16

8.5.2 Bend- and bifurcation pipe fittings

Sometimes pipes are required to change direction or split. Bend- and bifurcation pipe
fittings are used for this purpose. Where pipes split or change direction inside shafts, the
forces acting on the pipes must be determined from first principles to ensure that the pipes
can be properly anchored.

The energy losses which occur through bends can be calculated with the aid of Figure 8.5
(807) by substituting the applicable loss constant k into the equation:
‚
∆ℎ •
2#
The symbols have the normal meaning.
 8-17

FIGURE 8.5: Pipe bend energy loss constants

 8-18

Figure 8.6 shows the energy loss constants k for the typical ways pipes bifurcate.

FIGURE 8.6: Pipe bifurcation energy loss constants

8.5.3 Tapered pieces (Reducers and expanders)

Reducers and expanders are regularly used in the pipeline section. They are often installed
after the bifurcation of pipes when the larger diameter is no longer needed, and at the end
of the pipeline upstream of the control valve to allow installation of a smaller valve.
Smaller valves cause higher velocities which result in better energy dissipation.
 8-19

The energy losses through reducers are calculated as follows (803):

FIGURE 8.7: Energy losses through reducers

•
∆ℎ ƒA < A8 „
2#

#;
• 0,1 if tan … `
A

1
Where A A + A , and
2 8

k = 0,5 when α = 90° at a sudden contraction

 8-20

The energy losses through expanders are calculated as follows (803):

FIGURE 8.8: Energy losses through expanders

•
∆ℎ ƒA < A „
2# 8

α 2° 5° 10° 15° 20° 25° 30° 40° 50°

k 0,02 0,12 0,16 0,27 0,40 0,55 0,66 0,90 1,00

 8-21

It follows from the principles of hydraulics that expanders must be more gradual than
contractors because the velocity downstream decreases and therefore the static energy head
increases which could cause cavitation. “Design of gravity dams (803)” recommends that:

#; 1
tan … ` where A bA + A c
2A 2 8

Design of Gravity dams (803) also recommends that α should not be larger than 10°.

[Link] Pipes

Different materials are suitable for the pipe(s) which join the in- and outlet sections. The
type of pipe used depends on the water quality and so the corrosion resistance plays a
major role.

In the past, concrete pipes were commonly used in the core of the dam wall but the current
tendency is to use treated mild steel pipes, stainless steel pipes, or cast-iron pipes because
they can be treated against corrosion more easily. The choice of type of pipe in this
Directorate is usually left to the Mechanical Engineer because it is regarded as a
mechanical issue.

[Link] Pipeline losses

The Manning pipe flow equation:

8
o
A
r
can be used to calculate the friction losses.

To calculate pipeline losses, the Manning equation must be written in the appropriate form:
8
o
A
r
 8-22

Thence:

8 Ar A r ℎ
and •
o o

Since the system capacity is the determining factor, the discharge Q must be inserted into
the equation as follows:

r [
ℎ • Valid for all types of pipe
o -

; ˆ;
o and -
4 4
For a circular pipe,

r [
∆ℎ •
; ˆ;
‰ Š J
4 L
therefore
4

10,294 r [
∆ℎ 8‹
;
and so

The symbols have the following meaning:

V= velocity of fluid through the pipe in m/s

- Area
hydraulic radius o in m
* Wetted Perimeter
R=

S= energy gradient
8
n= Manning roughness in 5/v

Δh = energy head loss in m

L= length of pipe in m

Q= flow in pipe in m3/s

A= cross sectional area of conduit in m2

D= diameter of pipe in m
 8-23

In the literature the following n-values are given for different materials which can be used
in these calculations (803, 808, 809).

Type Minimum Maximum

Cast iron 0,012 0,015

Steel: Galvanized 0,013 0,017

Riveted joints 0,012 0,016

Welded joints 0,010 0,014

Concrete 0,010 0,016

It is recommended that the maximum n-value is used to determine the conduit size, and the
minimum value to calculate the highest flow velocities. This means that a conservative
pipe size will be obtained and that the energy dissipation measures will not be under-
designed. Since a pipe will seldom be less than 600 mm in diameter no provision needs to
be made for aging.

8.5.5 Flow measurement

Currently (1984) the policy in the Directorate is not to install flow meters in the outlet
works, but to rather build a gauging weir downstream of the dam in the river channel. The
reason for this is that the Hydrology- and Operations Sections are of the opinion that the
gauging weirs are more reliable and require less maintenance than flow meters.
 8-24

In the past, venturi meters or orifice plate meters were installed in the pipeline to measure
the discharge. There are now other types of flow meter on the market, such as
electromagnetic and doppler flow meters, etc.

Flow over the gauging weir is registered using an electronic process inside the outlet
house. Instant flow rate can be registered, as well as the total volume after a certain time
interval.

8.5.6 Isolating valves

Isolating valves are installed in the pipeline upstream of the control valve so that repair-
and maintenance work can be carried out on the pipeline itself or on the control valve.
Where a flow meter forms part of the outlet system the isolation valve is installed as close
as possible to the inlet section, otherwise it is placed in the outlet house. This is done
because there are strict rules for each flow meter type regarding unaffected flow upstream
and downstream of the flow meter. The isolating valves which are used most often in the
Directorate are butterfly- and sluice valves. The latter takes up a lot of room and is
preferably used only in pipes smaller than 600 mm in diameter.

[Link] Butterfly valves

Butterfly valves are very popular due to their ease of operation, low cost, compact size,
lack of moving parts and low energy losses. The energy loss due to a butterfly valve is
however higher than the energy loss caused by a sluice valve (sliding valve).

A butterfly valve consists of a cylindrical or conical body with a circular disk mounted on
a transverse rotating axle which is carried in two bushes diagonally opposite each other in
the valve body. An external operating mechanism rotates the disk through 90° from the
fully closed to the fully open position. Seals are provided to shut tightly when it is closed.
The disk and valve body are shaped to avoid sudden velocity changes. For large valves
and heads the disk is operated using hydraulic cylinders or rotors.
 8-25

Design detailing and specification of valves in the Directorate is always done by the
Mechanical Section.

Water velocities through the butterfly valve must be restricted where rubber seals are used
(which is usually the case) to about 7 m/s. A first indication of the size of pipe required is
obtained by conciliating the peak flow rate and the maximum flow velocity of 7 m/s
through the valve.

Energy losses caused by a fully open valve may be calculated as follows:

0,4 A
∆ℎ
2#
where V is the velocity through the valve and the other symbols have their normal
meaning. Energy losses resulting from other openings can be calculated using Figure 8.9.
Figure 8.10 shows a typical example of a butterfly valve.
 8-26

Basic equation:

1,234 A
∆ℎ .
• 2#

where Δh = Energy head loss

k = Discharge coefficient
v = velocity in m/s
g = gravitational acceleration

FIGURE 8.9: Energy loss through a butterfly valve

 8-27
FIGURE 8.10: Typical butterfly valve
 8-28

[Link] Sluice valves (sliding valve)

Sluice valves are generally used as isolating valves in an outlet system. This kind of valve
can, like the butterfly valve, not be used as a control valve because the possibility exists
that erosion and cavitation of the seals and valve body can occur. Advantages of this type
of valve are; relatively simple design, easy control, low energy losses because water flows
straight through, and the fact that the water can flow through the valve in both directions.

Disadvantages associated with a sluice valve are that large valves take up a lot of space and
that the valve opens and closes slowly.

Energy losses are calculated as before by using the equation:

•A
∆ℎ
2#

where k = 0,1 to 0,15 for a valve with a guide pipe, and

k = 0,2 to 0,3 without a guide pipe.

The first mentioned is normally used in the Directorate.

Figures 8.11 and 8.12 show the two commonest types of sluice valve namely the wedge-
and parallel types. The wedge type has better sealing characteristics and is subject to less
wear on the blade due to its shape.
 8-29
FIGURE 8.11: Wedge type sluice valve
 8-30
FIGURE 8.12: Parallel type sluice valve
 8-31

8.6.0 OUTLET SECTION

The outlet section consists of the outlet house and control valves at the downstream end of
the wall if it is a river outlet system. Where the outlet system provides water to a pumping
station, power station, or purification plant the system is adapted for the specific use.

8.6.1 Outlet house

The outlet house is probably the part of the dam which is most often visited to control the
outlets and during the design stage attempts must be made to provide a practical design
which will meet all the requirements.

During the design specific attention must be given to the drainage, access, lighting, and
ventilation of the outlet house. If the runoff water level is higher than the lowest point
inside the house a pump must be provided together with a one-way valve, otherwise
drainage can occur automatically. Access to the outlet house must be comfortable and
unhindered. Lighting is provided using waterproof windows and fluorescent lights.
Material used to manufacture the window frames must be corrosion resistant and to
facilitate maintenance, windows must be replaceable and cleanable from the inside. An
ideal solution to this problem has not yet been found and experimentation is still taking
place. Insect- and burglar proofing must also form part of the window. Lights must be
placed where they will not easily be damaged by activity inside the outlet house and where
they can also be easily replaced if required.

Ventilation may be done mechanically but usually the ventilation obtained through the
windows and entrances is adequate.

Outlet houses must also be designed so that the maintenance required on mechanical
components is easily carried out. Depending on what needs to be removable hatches,
trollies, etc. must be provided to do maintenance work. Hatches are normally used when
the mechanical equipment is removed using a mobile crane. Enough lifting hooks must be
provided at strategic positions in the roof of the outlet house to allow hoisting equipment to
be attached. Obviously, the roof must be structurally designed for this.

To allow pipes or valves to be removed later for repair work, connection flanges must be
provided at appropriate positions.
 8-32

8.6.2 Control valves

At the end of the pipeline the control valve is installed which regulates discharge through
the outlet system. The control mechanism for the valves is placed inside the outlet house.

Valves should never be open less than 15% because unnecessary wear occurs at smaller
valve openings. Sizes of the valves must therefore be chosen to allow for overlap in the
capacity of the different valves, so a large valve opened 15% should release less water than
a smaller valve which is fully open at the same head.

The control valve which is used most often in the Directorate is the sleeve valve although
the needle valve is also sometimes used. The latter is more expensive and is more
complicated. Experience shows that sleeve valves deliver very good service and there is
no reason not to recommend them.

[Link] Sleeve valves

For various reasons, sleeve valves are used most often as a control valve in the Directorate,
and they are very popular with Engineers who have to design outlet works. Some of the
reasons why the valve is so popular will now be briefly discussed.

Sleeve valves provide an easy and effective method to regulate and control the water which
is discharged. The cost of sleeve valves is lower than that of other discharge valves. The
advanced design of the valve enables it to discharge water effectively at both low- and
high-pressure heads. Sleeve valves vibrate very little during operation and require very
little maintenance.
 8-33

Since the sleeve valve has a high discharge coefficient, the rest of the system does not have
to be very large, and this saves costs.

As a result of the balanced hydraulic forces on the only moving part, the sleeve, little effort
is required to open or close the valve. The valve helps to dissipate the energy of the water
by spraying it in a hollow diverging cone shape which turns into a fine spray and in so
doing restricts erosion of the riverbed or apron. If the wide spray pattern is unacceptable a
downstream hood is used to constrict and direct the stream. Due to this effective energy
dissipation, a stilling basin is not usually required with these valves.

The sleeve valve consists of the body, the cylindrical sleeve gate, the seals, and the
operating mechanism. The internal diameter of the body is the same as that of the pipe to
which it is bolted. Radial ribs which are attached to the body protrude above the body and
support the cone at the end of the valve. The tip of this cone points upstream and the cone
angle is equal to 45°. Discharge from the radial openings formed by the cylindrical body,
ribs, and cone is controlled by the cylindrical sleeve gate on the outside of the valve body.
Contact surfaces of the body and sleeve gate are made of corrosion resistant material. An
operating mechanism moves the sleeve gate downstream to close the radial openings,
thereby closing the valve. The upstream end of the sleeve gate is provided with a seal
which slides over the outside of the body. Figure 8.13 shows a typical sleeve valve.
 8-34

FIGURE 8.13: Typical Sleeve valve

FIGURE 8.14: Typical Needle valve

 8-35

Discharge through a sleeve valve may be calculated using the following equation:

[ g; √

where: Q = discharge in m3/s

K = valve opening constant (see table below)
D = diameter of the valve in m
H = nett pressure head in m

Nett pressure head is the difference in height between the dam level and the centreline of
the valve (or the tailwater level if the valve is submerged) minus the sum of the energy
losses between the inlet and the sleeve valve.

FIGURE 8.15: Nett pressure head

 8-36

% Valve opening 10 20 30 40 50 60 70 80 90 100

K 0,45 0,82 1,22 1,57 1,86 2,16 2,38 2,61 2,80 2,96

Energy losses caused by a sleeve valve may be calculated with the following table and
equation:

% Valve opening 10 20 30 40 50 60 70 80 90 100

k 58,17 17,11 7,16 3,94 2,49 1,60 1,13 0,78 0,54 0,38

A
∆ℎ •
2#

where: Δh = energy loss caused by sleeve valve in m

k = valve loss constant
V = velocity through valve (= Q/A) in m/s
g = gravitational acceleration in m/s2

The constants given in the above tables are those for the “Ainsworth” valves which are
most often used in the Directorate (810).

[Link] Needle valves

Needle valves are also sometimes used as control valves in the Directorate. Needle valves
usually have a cylindrical aperture which is sealed using a conical piston. Figure 8.14
shows a typical example of a needle valve.
 8-37

Needle valves are used from the fully closed to fully open positions to regulate water
through the outlet works. As with the sleeve valve, the valve is usually installed at the end
of the pipeline to allow the valve to discharge water to the atmosphere. It can also be used
to release water when submerged but then the valve design must provide for this.

Hydraulic forces acting on the piston are more- or less balanced so the force required to
open the valve and keep it open at a specific position is not large. To obtain the minimum
leakage in the closed position, however, requires larger forces.

Discharge through a fully open needle valve may be calculated with the following
equation:

[ - 2#

where: Q = discharge in m3/s

C = discharge coefficient
A = cross sectional area of inlet pipe just upstream of valve in m2
g = gravitational acceleration in m/s2
H = nett pressure head in m

The discharge coefficient C varies from 0,45 to 0,60 depending on the valve
manufacturer’s specific design. Energy losses caused by a needle valve are calculated
using the equation:
A
∆ℎ •
2#
where k varies between 1,8 and 3,9 depending on the manufacturer.

8.7 Sample calculation

Question: Determine the discharge curves for one set of outlets as shown on
drawings 84 804/82 and 84 807/82, included herewith.

Solution: Note that this problem is approached backwards. Normally an outlet

is chosen, analysed, and adapted until the needs are satisfied.
8-38
8-39
 8-40

8.7.1 Step 1: Energy loss calculation

The first step is to calculate the energy losses caused by the trash racks, the pipe in- and
outlets, the bellmouth, the pipeline, bifurcations, bends, butterfly valve, reducers, sleeve
valves, and outlet losses.

Let the different discharges in the respective pipes be as shown in the layout sketch below:

[Link] Trash rack losses: 84 807/82 and [Link]

•
} A^
∆ℎ1 gs t sin
~ 2#

t = 120 mm
b = 120 mm

α = 90°
g = 9,81 m/s2

K = 3,7

[ [
A^ where is the water level above the centreline of the pipe.
- 3

Therefore:
•
120 [ sin 90°
∆ℎ1 3,7 s t
120 9 4 2 4 9,81

20,95 4 10N [
∆ℎ1
 8-41

[Link] Pipe in- and outlet losses: 8.5.1

These losses are calculated as for bellmouths and from first principles.

A
∆ℎ2 •1 + •2
2#

[ 0,5 + 1,0
‰ Š 4 2 4 9,81
4̂ 4 1,8

∴ ∆ℎ2 11,81 4 10N [

[Link] Bellmouth: 8.5.1

The bellmouth used is countersunk, therefore k = 0,2

A
∆ℎ3 0,2
2#

0,2 [
‰ Š 4 2 4 9,81
4̂ 4 1,2

∴ ∆ℎ3 7,97 4 10N [

 8-42

[Link] Pipeline losses: [Link]

10,294 r [
∆ℎ 8‹
;

Section 1:
L = 13 m; n = 0,017 (Assume galvanised pipe); D = 1,2 m ∅

10,294 4 13 4 0,017 4 [
∆ℎ40 8‹
1,2

∴ ∆ℎ40 14,63 4 10N [

Section 2:
L = 8 m; n = 0,017 (Assume galvanised pipe); D = 1,2 m ∅

10,294 4 8 4 0,017 4 [8
∆ℎ41 8‹
1,2

∴ ∆ℎ41 9,00 4 10N [8

Section 3:
L = 8 m; n = 0,017 (Assume galvanised pipe); D = 0,7 m ∅

10,294 4 8 4 0,017 4 [
∆ℎ42 8‹
0,7

∴ ∆ℎ42 159,5 4 10N [

 8-43

[Link] Bifurcations: 8.5.2 and 84 804/82

A
∆ℎ •
2#

Section 1:
k = 0,35 for flow straight past the bifurcation

A
∆ℎ51 0,35
2#

0,35 [8
∆ℎ51
2 4 9,81 4 ‰ 4 1,2 Š
4̂

∴ ∆ℎ51 13,95 4 10N [8

Section 2:
k = 0,4 for α = 30°

A
∆ℎ52 0,4
2#

0,4 [
∆ℎ52
2 4 9,81 4 ‰ 4̂ 4 0,7 Š

∴ ∆ℎ52 137,7 4 10N [

 8-44

[Link] Bends: 8.5.2

Assume R/D = 2 (safe assumption)

1 ˆ o
Ju^# s t + ‘L
• 2‘ ;
1 ˆ
‰u^# 2 + Š
• 24 6̂
6̂
∴ k = 0,09 say k = 0,1

Then
A
∆ℎ6 0,1
2#

0,1 [
∆ℎ6
2 4 9,81 4 ‰4̂ 4 0,7 Š

∴ ∆ℎ6 34,41 4 10N [

[Link] Butterfly valve losses: [Link]

0,4 A
∆ℎ
2#
 8-45

Pipeline 1:

0,4 [8
∆ℎ71
2 4 9,81 4 ‰ 4 1,2 Š
4̂
∴ ∆ℎ71 15,94 4 10N [8

Pipeline 2:

0,4 [
∆ℎ72
2 4 9,81 4 ‰ 4̂ 4 0,7 Š

∴ ∆ℎ72 137,7 4 10N [

[Link] Reducers: 8.5.3

Pipeline 1:

600 < 305

tan 0,164
1 800
#; 9,81 4 0,905
Value of
A A

Assume V1 = 7 m/s due to limit on flow velocity through the butterfly valve.

7 4 1,2
Then A 12,31 m/s
0,905

#;
and 0,242 ≮ tan
A
 8-46

Assume k = 0,2
0,2
Then ∆ℎ81 ƒA < A8 „
2 4 9,81

0,2 [8 1 1
_ < a
2 4 9,81
‰ 4 0,610 Š ‰ 4 1,2 Š
4̂ 4̂
∴ ∆ℎ81 111,4 4 10N [8

Pipeline 2:

350 < 190

tan 0,089
1 800

7 4 0,7
A 11,76 m⁄s based on same assumption as before.
0,54

#; 9,81 4 0,54
Then 0,196 ≮ tan
A 11,76
Assume k = 0,2
0,2
Then ∆ℎ82 ƒA < A8 „
2 4 9,81

0,2 [ 1 1
_ < a
2 4 9,81
‰ Š ‰ Š
4̂ 4 0,38 4̂ 4 0,7
∴ ∆ℎ82 723,7 4 10N [
 8-47

[Link] Sleeve valves: [Link]

Case 1: Sleeve valves 100% open, therefore k = 0,38

Pipeline 1:

A 0,38 [8
∆ℎ911 0,38
2#
2 4 9,81 4 ‰4̂ 4 0,610 Š

∴ ∆ℎ911 226,8 4 10N [8

Pipeline 2:

A 0,38 [
∆ℎ921 0,38
2#
2 4 9,81 4 ‰4̂ 4 0,38 Š

∴ ∆ℎ921 1,506 [

Case 2: Sleeve valves 20% open, therefore k = 17,11

Pipeline 1:

A 17,11 [8
∆ℎ912 17,11
2#
2 4 9,81 4 ‰4̂ 4 0,610 Š

∴ ∆ℎ912 10,21 [8

Pipeline 2:

A 17,11 [
∆ℎ922 17,11
2#
2 4 9,81 4 ‰4̂ 4 0,38 Š

∴ ∆ℎ922 67,80 [
 8-48

[Link] Outlet losses:

Pipeline 1:

A [8
∆ℎ101
2#
2 4 9,81 4 ‰ 4 0,610 Š
4̂
∴ ∆ℎ101 596,8 4 10N [8

Pipeline 2:

A [
∆ℎ102
2#
2 4 9,81 4 ‰ 4 0,38 Š
4̂
∴ ∆ℎ102 3,963 [

8.7.2 Step 2: Determining discharges

All the energy losses at the various components are now known in terms of Q, Q1, Q2, and
y. The equations can now be combined to calculate the different discharges as follows:

[Link] Case 1:

610 mm ∅ Sleeve valve 100% open and 380 mm ∅ sleeve valve closed, therefore,
Q Q1 and Q2 0.

The energy balance equation is now compiled.

y = Δh1 + Δh2+ Δh3+ Δh40+ Δh41+ Δh51+ Δh71+ Δh81+ Δh911+ Δh101

20,95 4 10N [8
Then + 1,008 [8
 8-49

By substituting different y values, Q1 can be calculated, as follows:

RL (m) y (m) Q1 (m3/s)

250 30 5,45
245 25 4,98
240 20 4,45
235 15 3,86
230 10 3,15
225 5 2,23

[Link] Case 2:

610 mm ∅ Sleeve valve 20% open and 380 mm ∅ sleeve valve closed, therefore,
Q Q1 and Q2 0.

The energy balance equation is:

y = Δh1 + Δh2+ Δh3+ Δh40+ Δh41+ Δh51+ Δh71+ Δh81+ Δh912+ Δh101

20,95 4 10N [8
Then + 10,992 [8

By substituting different y values as before, Q1 can be calculated.

RL (m) y (m) Q1 (m3/s)

250 30 1,65
245 25 1,51
240 20 1,35
235 15 1,17
230 10 0,95
225 5 0,67
 8-50

[Link] Case 3:

610 mm ∅ Sleeve valve closed and 380 mm ∅ sleeve valve 100% open, therefore,
Q Q2 and Q1 0.

The energy balance equation is:

y = Δh1 + Δh2+ Δh3+ Δh40+ Δh42+ Δh52+ Δh6+Δh72+ Δh82+ Δh921+ Δh102

20,95 4 10N [
Then + 6,696 [

By substituting different y values as before, Q2 can be calculated.

RL (m) y (m) Q2 (m3/s)

250 30 2,12
245 25 1,93
240 20 1,73
235 15 1,50
230 10 1,22
225 5 0,86

[Link] Case 4:

610 mm ∅ Sleeve valve closed and 380 mm ∅ sleeve valve 20% open, therefore,
Q Q2 and Q1 0.

The energy balance equation is:

y = Δh1 + Δh2+ Δh3+ Δh40+ Δh42+ Δh52+ Δh6+Δh72+ Δh82+ Δh922+ Δh102

20,95 4 10N [
Then + 72,991 [
 8-51

By substituting different y values as before, Q2 can be calculated.

RL (m) y (m) Q2 (m3/s)

250 30 0,64
245 25 0,59
240 20 0,52
235 15 0,45
230 10 0,37
225 5 0,26

[Link] Case 5:

Both valves 100% open, therefore, Q Q1 + Q2.

There are now 3 unknowns and to solve this, 3 equations are required. This case is a
combination of cases 1 and 3.
From case 1:
20,95 4 10N [
+ 34,41 4 10N [ + 973,9 4 10N [8 <<<<< A

From case 3:
20,95 4 10N [
+ 34,41 4 10N [ + 6,662 [ <<<<< B

and [ [8 + [ <<<<< C

A – B : 973,9 4 10N [8 6,662 [ <<<<< D

From which: Q1 = 2,615 Q2

Substitute this into (C):
Yields Q = 3,615 Q2
 8-52

By substituting the above into (B) it follows that:

273,8 4 10N [
+ 7,111 [

By substituting different y values as before:

RL (m) y (m) Q2 (m3/s) Q (m3/s) Q1 (m3/s)

250 30 2,05 7,41 5,36

245 25 1,87 6,76 4,89
240 20 1,68 6,07 4,39
235 15 1,45 5,24 3,79
230 10 1,18 4,27 3,09
225 5 0,83 3,00 2,17

8.7.3 Step 3: Velocity control

The maximum flow velocity through the butterfly valve must be controlled.

[Link] Pipeline 1:

[ 5,44 4 4
A 4,81 m⁄s E 7 m⁄s therefore within limits.
- ˆ 4 1,2
 8-53

[Link] Pipeline 2:

[ 2,12 4 4
A 5,51 m⁄s E 7 m⁄s therefore within limits.
- ˆ 4 0,7

8.7.4 Step 4: Overlap checking

The last step is to compile the discharge curves (Figure 8.16) using the compiled tables and
to check the overlap between the different valves. From the curves we can conclude that
the overlap obtained is adequate.

Similarly, the discharge for different outlet combinations may be calculated.

If the chosen layout is not satisfactory, the necessary changes are made to component sizes
and the calculations are repeated.
 8-54
FIGURE 8.16: Discharge curves for an outlet system
 9-1

9.0 ENERGY DISSIPATION

9.1 Introduction

Water discharged over the crest of a dam during a flood possesses an enormous amount of
potential energy which is converted to kinetic energy. If this energy is not dissipated there
is a danger that the structure could fail due to possible erosion and undercutting of the
foundation of the wall on the downstream side. The purpose of the energy dissipation is
therefore to reduce the kinetic energy before it reaches the natural channel downstream of
the dam wall.

Methods used apply one or more of the following principles, namely:

(a) Hydraulic jumps.

(b) Breaking up the main water stream to form a spray.
(c) Distributing the flood water over a larger area.
(d) Stimulating turbulent friction losses.

The following specific methods are normally applied at gravity dam structures and make
use of the above principles, namely:

(i) The hydraulic jump stilling basin.

(ii) The ski-jump/flip bucket.
(iii) Slotted and solid bucket.
(iv) Splitters.

The operation and design of the first 3 types are well described in the literature (901, 902,
903). In this chapter, splitters are described in detail according to Roberts’ description
(904, 905) and the other types are just briefly treated.
 9-2

9.2 Hydraulic jump stilling basin

Super-critical flow in the approach chute is converted to sub-critical flow with the aid of a
hydraulic jump inside the stilling basin or at the downstream end of the spillway. Due to
the intense turbulence generated in the jump, large amounts of energy can be dissipated.

It is extremely important during the choice and design of this kind of energy dissipator that
the hydraulic jump remains in the stilling basin for the entire spectrum of expected floods
to prevent serious erosion. The stability of the jump is largely dependent on the tailwater
level. End sills and baffle blocks can also be provided in the stilling basin to stabilize and
shorten the jump. The best jumps are obtained when the Froude number is between 4,5
and 9 in the water conduit upstream of the jump.

The most notable disadvantages of this kind of dissipator are the large excavations which
are sometimes required for the stilling basin, and waves which are propagated downstream
of the stilling basin. A model study should be conducted to test the design for all the
discharges which can occur.

A typical example of what the stilling basin could look like is shown in Figure 9.1.
Readers wanting to learn more about the subject are advised to study references 901, 902,
and 903.
 9-3
FIGURE 9.1: Typical layout of a hydraulic jump stilling basin
 9-4

9.3 Ski jump/flip bucket

A ski jump is often used at low dams or where the tailwater level is too low to form a
stable hydraulic jump, so where the excavation required for the stilling basin would be too
deep. The method basically consists of a flip bucket being built at the end of the flood
channel. While the water travels through the air, air is absorbed (±20%) before it plunges
back into the river over a wider area where it erodes a plunge pool.

The distance from the downstream toe of the wall to where the plunge pool will form is
given by the following equation (Symbols are defined in the sketch):

= ℎ
1,9 s t sin 2ž

(The constant 1,9 makes provision for the energy loss which occurs in the air)

10 … R …20 m
20° … φ … 40°
 9-5

A model study may be carried out to determine the depth of the pool. During this study it
can also be determined if the plunge pool will expand upstream and possibly undercut the
foundation.

A determining factor in the design of the dissipator is that the flip bucket may not be
submerged. This means that the lip must still be above the tailwater level.

The most notable disadvantages are that a mist can be formed which may affect nearby
structures such as power stations and due to the broader base of the dam, greater uplift
forces will be experienced. Where the water velocity becomes high the approach channel
becomes expensive.

F.W. Becker (910) has developed a method with which the flow distribution in the
approach chute may be analysed. Spreading is achieved using a curved bottom.

Revised forms of the flip bucket are often used to obtain optimal operation over the full
spectrum of expected floods.

References 901 and 903 contain more information regarding the design of the ski jump
energy dissipator.

9.4 Slotted- and solid buckets/Submerged buckets

Slotted- and solid buckets are suitable where a high tailwater level exists or where a free
hydraulic jump will not form. A surface roller and a ground roller are formed in the
process and the energy is reduced by turbulent friction.
 9-6

Unfortunately, the operation of this type is sometimes not effective at low flows because
the surface roller does not form due to a too low tailwater level. It must be noted that the
ground roller does carry material towards the toe and not away from it because this may
undercut the foundation. The slotted- and solid buckets work best for Froude numbers
between 3 and 10. Model studies are also required here to test operation over the total
spectrum.

References 901 and 902 provide more information regarding the design and operation of
this kind of energy dissipator.

9.5.0 Splitters

9.5.1 Introduction

The method of using splitters as energy dissipators was developed during 1936 by Lt. Col.
D.F. Roberts during the construction of Loskop dam and since then the method has been
used with good results on more than 25 dams in South Africa.
 9-7

Lt. Col. Roberts carried out an extensive model study to perfect the splitters and his results
are published in reference 906 which may also be used as a design aid.

The method is based on breaking up the solid stream of water passing over the crest into a
system of streams with different trajectories. This is done by constructing the splitters and
a continuous step under them against the downstream slope of the wall under the spillway.
Energy losses occur when the different streams collide with each other, and the water is
also aerated in the process.

Flood water then hits the riverbed (apron) over a wider area with less intensity due to the
energy loss, so much so that an apron is often not required if the quality of the rock in the
riverbed is good (e.g., Braam Raubenheimer* dam) or a reduced thickness apron. A typical
layout of splitters is shown in Figure 9.2.

Despite the fact that Roberts developed the splitters for a maximum spillway head of 3 m,
splitters are in use at Hendrik Verwoerd† and P.K. le Roux‡ dams which can handle a
spillway head of up to 9 m using aeration which was introduced after further model studies
(907, 908). If such high spillway heads are used model studies should be carried out to test
the operation of the splitters.

* Braam Raubenheimer dam is now called Kwena dam.

† Hendrik Verwoerd dam is now called Gariep dam.

‡ P.K. le Roux dam is now called Vanderkloof dam.

 9-8
FIGURE 9.2: Typical layout of splitters
 9-9

The main advantages of this method of energy dissipation are that it is simple to apply,
cheap, and also highly effective. At about half of the dams where splitters were used, no
apron was required resulting in significant cost savings. (909).

The largest disadvantage of splitters is that an enormous amount of spray is created which
hinders nearby power stations or outlet works and for which provision must be made in the
design, by for example separating the outlet block and spillway by one dam block width. It
must be added that South African dams mostly spill in the winter or summer for short
periods and also not every year.

In the following section the operation and design will be described in accordance with
references 904, 905, and 906. A sample calculation will also be done. (Designers are
advised that model studies must be carried out if there is any deviation from the
requirements as recommended by Roberts (906)).

9.5.2 Operation of splitters

The operation of splitters will be described with reference to Figures 9.3 and 9.4. It is also
necessary to define a number of terms, namely:

Main stream - the undisturbed water stream flowing over the crest before being divided by
the splitters.

Top stream - the stream flowing straight over a splitter in a downstream direction.
 9-10
FIGURE 9.3: Diagrammatic representation of hydraulic operation of splitters
 9-11
FIGURE 9.4: Diagrammatic representation of hydraulic operation of splitters
 9-12

Side streams - those portions of the water stream flowing over the sides of a splitter and
hitting the step.

Space stream - the stream flowing between two splitters.

Vortex stream - the stream formed under the splitter when the two side streams on either
side converge.

The dome - except at very low flow rates the two side streams and top stream form a
dome shaped curtain of water which flows over the front and sides of a
splitter.

The operation of splitters can then be described as follows:

(i) When water reaches the splitters, it is divided to flow over the top of the splitter to
form the side- and top streams and the rest flows between the splitters to form the
space stream.

(ii) Space streams converge (narrow) on the step and because the velocity remains
constant the water depth increases. If the splitters are close enough together the
space stream leaves the step in a thin unstable vertical stream.

(iii) Side streams hit the step along a line near the side of the space stream and together
with the top stream a dome is formed. The base of the dome is narrower at the origin
than at the side of the step. The two side streams flow towards each other after
hitting the step to meet at the centreline of the splitter. The two streams then
combine to form the vortex stream which moves outwards and upwards to collide
with the top stream and thereby cause a vortex. It must be noted that the side streams
do not merge with the space stream.
 9-13

(iv) With increasing flow rate these vortices expand laterally and with the correct spacing
of the splitters the adjacent vortices mix. The thin vertical space stream is very
unstable and breaks up completely.

(v) At very low head (Ha) a vortex is not formed because the side streams do not collide
with the top stream and also the space stream does not converge fast enough, but
because the flow rate is low the streams break up naturally.

(vi) At a specific head (Hc) a critical stage is reached depending on the dimensions and
position of the splitter. At heads higher than Hc the breakup is less complete, there is
a lot of vibration, the vortices are replaced by a solid stream of water, and the side-
and space streams form a single stream. This is commonly called submergence or
drowning of the splitters. This condition is undesirable due to the fact that the
splitters are then no longer effective.

[Link] Design of splitter layout

[Link] Definition of symbols

The symbols which will subsequently be used to completely describe the splitter have the
following meanings and must be referenced with Figure 9.5.
 9-14

FIGURE 9.5: Definition sketch for splitter layout

 9-15

He - Spillway head at any given time

Ho - Design spillway head
Hc - Spillway head at which the splitter starts to submerge
Ha - Spillway head at which effective operation of the splitter starts
C - Difference in height between top of step and spillway crest
P - Difference in height between top of splitter and spillway crest
L - Length of top of splitter
W - Width of top of splitter
T - Difference in height between the top of the splitter and the top of the step
S - Horizontal distance between two adjacent splitters
Rmin - Distance on apron (foundation) level between the toe and the nearest water
stream
Rmax - Distance on apron (foundation) level between the toe and the furthest water
stream
x - Difference in height between the top of the water flowing over the step and the
water level in the dam
y - Difference in height between the water level on the step and the apron
(foundation)
θ - Angle between the horizontal and the downstream face of the dam wall
d - Water depth on the step
Ls - Step width
f - Correction factor to provide for the reduced velocity on the step due to friction
and air resistance in the prototype

[Link] Design requirements

The formulae which follow were determined empirically and must not be deviated from.

(a) Ho as previously determined (6.7.0) is assumed as the design spillway head, that is
the 1 in 200 year flood head.
 9-16

(b) Splitters are not effective if the sum of x and y is not at least four times the design
spillway head, Ho, that is:
=+ ≥4 ^ (Equation A)
This requirement means that splitters cannot be used at relatively low dams.

(c) Roberts (906) recommends that Ho be increased by 20% to ensure that the
submerged condition is not reached, therefore Hc = 1,2 Ho.

In addition, Roberts (906) found that the following ratio exists between the
submerged condition and where the effective operation of the splitters begins:

[
4,4 sand thus 9,2t Equation B
" ["

From the above it follows that:

[
1,2 and thus 1,2 8,
1,3
^ [^

^ 4,4 [^
and and thus 3,67 8,
7,0
" 1,2 ["

Since the ratio of the 200 year flood (Qo the design flood) to the 5 year flood is
usually about 3,5 this means that the splitters will already begin to operate
effectively for floods smaller than the 5 year flood because Qo/Qa = 7,0 which is
exactly what is desired. It is much more important that the splitters are effective
for floods with high frequencies than those with low frequencies because the risk
that floods larger than the 200 year flood will occur is very small and in any case
the duration of such large floods will be short.
 9-17

The splitter is therefore designed for a spillway head of Hc where:

Hc = 1,2 Ho (Equation C)

(d) Distribution of the spilled water at riverbed level (apron level) depends on the
position of the step relative to the spillway crest. For design purposes it is preferable
that the spilled water lands as close to the downstream toe as possible during the
design flood. It may even be allowable for a small quantity of water to land on the
downstream face of the wall in which case Rmin will have a negative value. If Rmax
becomes too large, a wide and expensive apron is required if the rock is not solid.

Distribution is the difference between Rmax and Rmin and it must be an acceptable
distance, usually at least 10 m but it again depends on the height of the dam and
the amount of water being discharged. Rmax and Rmin may be determined with the
following equations:

o¤¥¦ 2§ = – cot w Equation D

and o¤¨j § = – cot w Equation E

where f can be obtained from the dimensionless graph, Figure 9.6 (904).
 9-18
FIGURE 9.6: Relationship between f and C/Ho
 9-19

The design procedure is to substitute values for x and y into equations D and E until an
acceptable distribution of the streams is obtained. The author has compiled figure 9.7
based on already constructed systems to simplify the initial choice of x and y. It follows
from the geometry that:
x = (He + P) + (T - d) (Equation F)

Because (T – d) <<<< (Hc + P) equation F may be simplified to:

x ≃ He + P (Equation G)

For the design flood He = Ho and thus x = Ho + P

 9-20
FIGURE 9.7: Curve to simplify the choice of P
 9-21

(e) For optimum results, Roberts (906) recommends that the following must also
apply:

S = T = L = 1,33W (Equation H)

The absolute value for W must be determined with the aid of Figure 9.8. The
value of W however also depends on the geometry of the dam block width and the
ratio of W/P to Hc/P may therefore deviate slightly from the prescribed curve.
 9-22
FIGURE 9.8: Relationship between Hc/P and W/P
 9-23

(f) The step width must lie within the following limits:

1,25L … Ls … 1,5L

The author has compiled a flow diagram (Figure 9.9) for the design of splitters
which considers all the preceding requirements. At the end of this chapter is an
example of a splitter design which was done using this flow diagram.
 9-24

FIGURE 9.9: Flow diagram for designing splitters

 9-25

[Link] Loads on splitters/step

Model studies carried out by Sogreah (908) have shown that the splitters and step must be
designed for the hydraulic loads as indicated in the following sketch:

FIGURE 9.10: Loads on splitter and step

The point load in the above sketch makes provision for impulse loads such as a floating
falling log or something substantial. Normal structural calculations are carried out to
determine the reinforcement required in the step and splitter.
 9-26

9.6 SAMPLE CALCULATION

9.6.1 Given details

H = 40 m
cot θ = 2/3
Block width = 15 m
Ho = 3 m
1:200 year flood: Q = 1 720 m3/s
1:5 year flood: Q = 460 m3/s

The sample calculation will be carried out according to the flow diagram, Figure 9.9.

9.6.2 Step 1: Test the head requirement (Requirement [Link] (b))

(x + y) ≥ 4Ho (or H ≥ 3Ho)

x + y = 43 m
4Ho = 12 m
∴ (x + y) > 4Ho ⇒ Correct

9.6.3 Step 2: Test flood ratio (Requirement [Link] (c))

[200 1 720
246 m /s
7 7
But Q5 = 460 m3/s

[200
∴ E [5 ⇒ Correct
7

9.6.4 Step 3: Calculation of Hc

Hc = 1,2 Ho = 1,2 × 3,0 = 3,6 m

 9-27

9.6.5 Step 4: Choice of P using Figure 9.7

^ 3 * *
* * 40
Satisfies
P
curve?
10 0,30 0,25 No
11 0,27 0,28 No
12 0,25 0,30 Yes
13 0,23 0,33 No

9.6.6 Step 5: Calculation of x, y, Rmin, Rmax, and Distribution (Equation G)

x = Ho + P = 3 + 12 = 15 m
y = 43 – 15 = 28 m

* 12
≃ 4
^ ^ 3
From Figure 9.6 it follows that f = 0,94

o¤¨j § = – cot w
2
0,94 √15 4 28 – 28 4
3
0,60 m

o¤¥¦ 2§ = – cot w
2
2 4 0,94 √15 4 28 – 28 4
3
19,86 m
∴ Rmax is within reasonable limits.

“Distribution” = Rmax - Rmin

= 19,86 – 0,60
= 19,26 m
∴ The “Distribution” is good enough.
 9-28

9.6.7 Step 6: Calculation of Hc/P and W

3,6
0,30
* 12

7
∴ 0,075 from Figure 9.8
*
and W = 0,075 × 12
∴ W = 0,90 m

9.6.8 Step 7: Calculate S, L, and T

S = L = T = 4/3 W = 1,2 m (provisionally)

9.6.9 Step 8: Calculate the number of splitters per block

15 15
r 7,1
+7 1,2 + 0,9
Say n = 7 splitters per block
So, 7 (S + W) = 15
7 (4/3 + 1) W = 15

∴ W = 0,918 m
„ Unpractical dimensions
and S = 1,224 m

Let W = 0,900 m

15 < 7 0,9
then 1,243 m
7
Let L = T = 1,25 m

9.6.10 Step 9: Recalculate W/P

Š This point plots on Figure 9.8 ∴ Correct

W/P = 0,9/12 = 0,075
Hc/P = 0,3
 9-29

9.6.11 Step 10: Calculate Ls

1,25L … Ls … 1,5L
∴ 1,563 … Ls … 1,875
Let Ls 1,700 m

9.6.12 Step 11: Check distribution at smaller floods

5 year flood: Q = 460 m3/s

He = 1,3 m
∴ x = 1,3 + 12 = 13,3 m
and y = 28 m
*+ 12 + 1,250
10,2
: : 1,3
From Figure 9.6, f = 0,83

o¤¨j § = – y cot θ
2
0,83 13,3 4 28 – 28 4
3
<2,65 m
∴ Not too much water on the toe.

o¤¥¦ 2§ = – y cot θ
2
2 4 0,83 13,3 4 28 – 28 4
3
13,37 m
∴ Rmax is good enough.
“Distribution” = Rmax - Rmin
= 13,37 – (-2,65)
= 16,02 m
∴ Very good “Distribution”, even for small floods.
The designer can test for other floods but in this specific example it should not be
necessary.
∴ The design is OK.
 10-1

10.0 FOUNDATION GROUTING AND DRAINAGE

10.1 Introduction

Proper foundation treatment of big dams is one of the most important problems in modern
construction. If the dam is to function as designed, it is important that the foundations
must be capable of safely resisting the loads which the dam exerts on them. If the
foundations in their natural condition have shortcomings it is sometimes possible to
strengthen them by pumping grout in under high pressure.

In the Directorate grout is used to reduce seepage, to reduce uplift forces, and to increase
the bearing capacity of the rock. Seepage and uplift forces are normally reduced by
forming a grout curtain. This consists of one or more rows of holes parallel to the dam
centreline which are grouted. Together with the grout curtain there is a system of drainage
holes directly downstream and parallel to the grout curtain. The purpose of these holes is
to relieve the hydrostatic pressure caused by water seepage through or around the grout
curtain.

Grouting is a technique involving the injection of a liquid solidifying mixture (consisting

of a suspension, a solution, or an emulsion) under high pressure into a rock mass. This is
done in such a manner as to displace as much air and water in seams and cracks or contact
zones with the purpose of reducing water seepage in the rock, and to provide adequate
strength to the rock mass after the mixture has cured and bonded with the rock (1001,
1002).

Before a grouting plan is compiled, the dam foundation must be thoroughly examined by
the Engineering Geologist. Based on his observations the characteristics of the rock and
the position and orientation of faults, seams, and voids is determined. Using this
information, a meaningful grouting- and drainage plan is then compiled. This plan is then
adjusted based on circumstances during the construction stage to obtain the best end
product.
 10-2

10.2 Grouting investigation

During the different phases of foundation investigation, the Engineering Geologist looks at
the groutability of the dam foundation and makes relevant recommendations in his reports.

According to Houlsby (1003) there are two characteristics which should be examined
during the foundation investigations of a concrete dam pertaining to possible grouting,
these are the geology and the permeability of the foundation. Houlsby regards seamed
rock which is relatively free of weathering as groutable.

10.2.1 Geological factors influencing grouting

Houlsby (1003) indicates that the following geological factors influence grouting:

(a) Spacing of open seams: More grout holes are needed if the seams are closer
together.

(b) Width of the open seams: Seam widths of about 2 mm will be easiest to grout.
Seams narrower than 0,5 mm are difficult to grout and seams wider than 5 mm
require special sealing techniques.

(c) Slope and reach of seams: The orientation of the holes must be such that as many
seams as possible are crossed.

(d) Rock strength: With strong rock, rock movement during grouting will be lessened.

(e) Quality of the rock: If the rock is of good quality the grout holes will not easily
collapse.

(f) Internal rock stresses: High internal tectonic rock stresses cause possible rock
movement which requires special grouting techniques.
 10-3

(g) Uniformity of the rock mass: Irregular seams, varied rock types, faults, etc.
require complicated grouting patterns.

(h) Tendency to begin piping: When the seams are filled with material which promotes
piping, special grouting techniques must be applied.

10.2.2 Permeability investigation

The permeability determination currently used in the Directorate is the Lugeon water test
method as described by Houlsby (1003). The Lugeon-unit is defined as the use of 1 litre of
water per metre of the test hole per minute at a pressure of 1 MPa. (This is approximately
equivalent to a k value of 1,2 × 10-5 cm/sec according to Darcy’s law).

Houlsby gives an indication of the scale of the Lugeon unit as follows:

1 Lugeon: This is the degree of permeability which will be found in dense formations
where (almost) no grouting will be required.

3 Lugeon: Grouting will be required for a concrete dam especially if seepage may be a
problem.

5 Lugeon: Extensive grouting will be required for a concrete dam.

10 Lugeon: Grouting is required for any type of dam.

The Lugeon test is carried out by pumping water into the test holes in a prescribed
sequence. Standard pressures are maintained for specific time periods in stages to form a
complete picture with depth of the permeability of a hole. It is important that the test is
carried out carefully because the final design and specification of the grouting depends
largely on the test results.
 10-4

Because the Engineering Geologist is responsible for the execution, monitoring, and
reporting with regard to the Lugeon test, the detail of the test is not discussed further.
Readers wanting to learn more are advised to study Houlsby’s (1003) paper.

10.3 Grouting design

According to Houlsby (1004) the question to be answered is whether grouting is required,

and if so, the intensity required. When these questions are answered, the layout is designed
to meet the foundation conditions and the specifications are compiled to obtain the
required intensity.

10.3.1 The necessity to grout

If the answer to one of the following questions is positive the foundation should be filled
with a cementitious grout mixture:

(i) Will the water loss be so large that waterproofing will be economically viable?

(ii) Is there a danger that piping due to seepage will occur in the foundation of the
dam?

(iii) Is grouting required together with drainage holes to control hydraulic uplift
pressures?

(iv) Is grouting required to increase the strength of the foundation?

The answer to the third question is usually positive based on experience in the Directorate
and it is only in exceptional cases where grouting is not done. The intensity required by
the Directorate for concrete dams is 1 Lugeon.
 10-5

10.3.2 Curtain grouting

[Link] Position of grout curtain

At gravity dams the curtain is positioned as close to the upstream heel as possible to obtain
the maximum benefit together with the drainage system of reducing the pressure on the
base. It must be noted that the curtain itself does not reduce the uplift pressure and that the
drainage holes form an inherent part of the system. An additional purpose of the curtain is
to reduce the seepage to such an extent that the drainage holes are able to handle the
possible flow more easily. Figure 10.1 indicates the typical position of a grout curtain of a
gravity dam.

FIGURE 10.1: Position of grout curtain

 10-6

[Link] Orientation of grout curtain

The curtain must preferably be orientated so that the most important and maximum
possible quantity of seams are crossed by the grout curtain. The curtain must preferably
slope upstream as indicated in Figure 10.2.

FIGURE 10.2: Orientation of grout curtain

To obtain the best cut-off of seams it is usually necessary that the holes not only slope
upstream but also to the left or right. It is unfortunately difficult to indicate these hole
directions for drilling. Such a system is also prone to undetected faults which lead to gaps
in the grout curtain. It is therefore safer for practical reasons to only slope the holes
upstream.
 10-7

“Houlsby (1004)” recommends the pattern indicated in Figure 10.3 when the grouting
holes in the flanks of the dam have different orientations. Overlap is preferable to avoid
gaps in the formed curtain.

FIGURE 10.3: Grout pattern

[Link] Depth of grout curtain and drainage holes

The required curtain depth depends on the height of the dam and the geological formation
under it. There are differences of opinion in the literature regarding this aspect. “Design
of Gravity Dams (1006)” recommends a depth of 30% to 40% of the hydrostatic head in
hard, dense rock and up to 70% in weaker foundations. The “American Institute for Civil
Engineers (1007)” recommends a depth of one third of the hydrostatic head plus a constant
which varies between 8 and 20 metres. “Houlsby (1004)” advises that the curtain must
have the same depth as the hydrostatic head or up to an adequately dense base in the
foundations.
 10-8

In the Directorate a minimum of two thirds of the hydrostatic head is accepted for the
depth of the curtain, but a depth which ensures enough resistance against the hydrostatic
head is the principle which should apply. Engineering- and Geological judgement play a
significant role in the ultimate choice of the depth.

At the top ends of the flanks (abutments) of the wall the curtain should not be less than 12
to 15 metres deep and the curtain must seal tightly up to the HFL as indicated in Figure
10.4.

FIGURE 10.4: Recommended depth of grout curtain

[Link] Grouting intensity

The grouting intensity obtained in a curtain depends on various factors, namely:

(i) The spacing of the holes: In the Directorate the holes are drilled at 2,5 m centres as
indicated in Figure 10.5.
 10-9

FIGURE 10.5: Spacing of grout curtain holes

Primary holes are drilled and grouted first, then the secondary holes and finally
the tertiary holes. If two adjacent primary holes do not require grouting, then no
holes are drilled between them. (So, 1 secondary and 2 tertiary holes fall away). If a
secondary hole does not require grouting, then the adjacent tertiary holes are not
drilled. Ideally, the holes should require less grout as they are drilled closer together,
so tertiary holes should take less grout than primary holes. Unfortunately, this does
not always happen in practice and each hole’s results should be evaluated
independently.

(ii) Desirability to drill angled holes in order to cross the most important seams. This
aspect has been discussed under [Link].
 10-10

(iii) Depth of holes and the associated higher allowable pressure. Under [Link] we
have already looked at the depth aspect of grout holes. The allowable applicable
pressure increases as the hole depth increases. The pressure is measured at the
top of the hole.

In the Directorate it is recommended that the grout pressure must be as high as possible
without causing a displacement of the rock/structure. As a guide, a pressure of 15 × d kPa
is used to start at the top of the grout hole where d is the depth of the hole in metres. This
corresponds with the recommendation of “Houlsby (1005)” for average rock conditions.

It must be noted that if any rock displacement occurs the pressure must be reduced
immediately, and the cause of the movement investigated. The first indication of a
possible rock displacement is that the pressure falls, and the mixture take-up increases
dramatically.

[Link] Stage lengths

Stage lengths depend on the minimum length worth drilling and the allowable pressure
increment per stage (short stages lead to minimum pressures at the upper ends of the
stages). The following stages are currently prescribed in the Directorate:

Stage 1: 0 – 6 m
Stage 2: 6 – 15 m
Stage 3: 15 – 30 m
Stage 4: 30 – bottom of hole

The above-mentioned stage lengths take the available drill pipe lengths into account.

10.4 Grouting methods

Houlsby’s (1004) schematic representation of the different methods is included for

convenience as Figure 10.6 and explains how the different systems work.
 10-11

FIGURE 10.6: Schematic representation of the different grouting methods

 10-12

Experience has shown that the downstage method with packers delivers the best results but
unfortunately it is also the most expensive method due to the multiple drill setups and the
loss of packers. The cheapest method is the upstage grouting method. This method may
give problems if the packer does not seal against the rough sides of the hole or if the hole
collapses due to movement. The downstage method without packers is preferred because
the higher lying rock which is normally of poorer quality can usually be strengthened
under relatively lower pressure to allow higher pressure to be applied to lower lying rock
without concern about what is happening higher up. Readers wanting to know more about
the grouting process are advised to read Houlsby’s paper (1005) regarding construction.

10.5 Grout specifications

Drawing 84 628/82 which is a typical grout specification of the Directorate is included

herewith as Figure 10.7.

10.6 Monitoring of grouting

During the construction of the curtain the Engineering Geologist must carefully monitor
the work together with the construction personnel and must report regularly to the design
engineer. If adjustments are required to improve the quality of the grouting, they must be
discussed with the design engineer to allow the issuing of a revised specification.
 10-13
FIGURE 10.7: Grout specifications
 11-1

11.0 RIVER DIVERSION

11.1 Purpose

The purpose of a river diversion system at a dam is to minimise the damage due to
untimely flooding during construction and to allow construction to proceed unhindered or
with the least possible delays.

11.2 Factors influencing the river diversion system

The size and flood potential of the river in which the dam is built is one of the first aspects
which must be examined. It is obvious that it is usually easier to divert a river with a low
runoff and flood potential.

Another aspect which largely determines the type of diversion is the kind of dam which
will be constructed. If a concrete dam is built, the risk of large damage is smaller than for
an earthfill dam because the completed concrete sections provide greater resistance to
damage during a flood compared to an earthfill dam which will in all likelihood breach and
wash away.

Terrain layout is a further aspect which can influence the diversion system. The shape of
the valley in which the dam is built also plays a role. A wide valley with low flanks, for
example, does not lend itself to the use of a bypass tunnel due to the low cover on the
tunnel crest.

The quality of the system will also depend on the duration of construction. The longer the
construction period, the more durable the system should be because it is exposed to greater
risks.
 11-2

A further aspect and probably the most important in South Africa is the flow characteristics
of the river. The majority of our rivers become a small stream in the dry season. This
allows us to programme the construction work at concrete dams to concentrate on the river
portion during this period. An extensive river diversion system is then required.

Sometimes there are existing dams upstream of the new one which can assist with
absorbing the larger floods, thereby limiting the size of diversion system required.
Because our country is water scarce, it is often not possible to reduce the water level in the
upstream dam merely to create flood absorption capacity.

11.3 Standards

In the Directorate of Water Affairs, it is the Construction Engineer’s responsibility to

design the system because he must ultimately determine the risk he is willing to take. This
does not however prevent the designer from making recommendations regarding this
aspect, in fact the designer is often asked by the Contractor (Construction section in the
case of Water Affairs) to prescribe the detail of the system. To do this, certain standards
are required.

“Design of Gravity Dams (1101)” recommends that the 1 in 25 year flood be used to
design the system if the construction period is longer than 2 years, and the 5 year flood if it
is shorter. The author feels that the standards are unnecessarily conservative for concrete
dams because South African river flow is mostly seasonal in nature. By programming
construction so that the flanks are constructed during the wet season and the river section
during the dry season the river diversion system can be reduced to the minimum. The
author does not believe that the risk should be unnecessarily increased by making the
diversion system too small. If the work cannot be programmed in this way and
construction needs to continue in the river section during the wet season the 2 year flood
should be used to design the system. Another recommendation is that construction work
which needs to proceed must not be inaccessible for longer than 2 days. Since the
Resident (Construction) Engineer still carries the responsibility the above should only
serve as a recommendation and not as a guideline. It is emphasized that the above
recommendation only applies for concrete dams and not for rock- and earthfill dams.
 11-3

11.4 River diversion methods

The method of diversion depends on the following factors:

(i) The size of the design flood or river flow rate will, to a large extent, determine the
method and detail of the diversion.

(ii) Another factor influencing the choice is the possible damage caused and the
subsequent delay if the excavation is flooded due to a system which is too small.

(iii) The diversion depends on the type of dam to be constructed.

(iv) The diversion method must naturally maintain the river flow downstream. Fish
deaths and water shortages can occur if this is not done.

(v) The diversion must also not create unnecessary impoundment because unbought
ground, roads, borrow areas, etc. can be flooded.

One or a combination of methods may be followed to divert the river. This includes
tunnels, canals, flood channels, flood conduits, and coffer dams. The reader is advised to
refer to chapter 5 of “Design of Gravity Dams (1101)” for more information regarding the
above methods. The multistage diversion method which is normally used in the
Directorate will be schematically explained using Figure 11.1.
 11-4

DESCRIPTION LAYOUT
STAGE 1: (First dry season)

1. At the start of the first dry season a coffer dam is built of

soil/rock/concrete, high enough so that the peak winter flow
does not overtop the coffer dam. The coffer dam must
provide enough workspace.
2. Do the excavation, place concrete in all the blocks and build
the flood conduit large enough to handle the peak winter flow.
The flood conduit must have a closure mechanism on the
upstream side (See “Design of Gravity Dams” for detail).
3. Get the blocks high enough so that they will not be
overtopped by normal summer flow.
4. If desired, at least parts of the coffer dam should be removed
at the end of the dry season to allow flow in the summer.

STAGE 2: (First wet season)

1. Work is concentrated on flanks.

2. During floods, water flows over the coffer dam and over the
low blocks. Shuttering, equipment, etc. must be removed
from the flood area.

STAGE 3: (Second dry season)

1. Build the new coffer dam and concentrate work on the river
section, so complete excavation and place concrete up to a
safe level.
2. Enough blocks must be left low to allow summer flood water
to pass without unnecessary impoundment.
3. Remove shuttering and equipment at the end of the winter to
prevent damage.

STAGE 4: (Second wet season)

1. Concentrate on the flanks and on the high blocks of the river

section if possible.
2. Try to get the outlet works operational.
3. Complete the flanks if it is the last wet season.

STAGE 5: (Last dry season)

1. Complete the work in the river section.

2. Seal the flood conduit as soon as the outlet works is able to
maintain the river flow and keep the water level low enough
to allow completion of all the blocks.
3. Impoundment can begin as soon as the work is completed.

FIGURE 11.1: Multi-stage river diversion

 11-5

11.5 Multi-stage diversion

For the purpose of this explanation, it is assumed that the construction period is longer than
2 years. This means that the system will be in operation for at least 2 wet seasons. This
does not mean that the system cannot also be used for shorter periods, but certain
adjustments will need to be made. For the purpose of this description, it is also assumed
that this is a summer rainfall area.
 12-1

12.0 JOINTS AND WATER SEALING

12.1 Purpose

Temperature and stress cracks which can form in the concrete of the wall are undesirable
because they can endanger the serviceability of the dam. Joints which are placed in the
dam are therefore actually planned cracks. Contraction- and expansion joints as well as
construction joints are found in gravity dam structures.

According to “Design of Gravity Dams (1201)” contraction and expansion joints are used
in the structure to allow for the volumetric changes in the concrete which occur after the
concrete is placed. Contraction joints are provided to prevent tensile stress fractures when
the concrete shrinks due to temperature drops. Expansion joints are provided to prevent
displacement of adjacent concrete blocks or build-up of stresses due to deflection.
Construction joints are provided to facilitate construction, to limit temperature stresses due
to hydration of the concrete, and to simplify the installation of mechanical equipment by
placing of secondary concrete. Construction joints are also needed when the concrete
placement is interrupted for a long period for any reason.

12.2 Contraction joints

To prevent the formation of cracks due to temperature stresses gravity dams are currently
still built using transverse vertical contraction joints at constant spacing over the length of
the dam wall. These contraction joints stretch over the full cross section of the wall from
the foundation to the crest of the dam*.

Contraction joints are so formed that there is no connection between two adjacent dam
blocks and no reinforcement passes through the joint. Spacing of contraction joints
depends on the terrain, the placement method, shrinkage characteristics of the concrete,
and the capacity of the batch plant. The author recommends that a minimum width of
7,5 m and a maximum width of 15 m be used. For aesthetic and practical reasons, the
block width for a specific dam should remain constant.

* With roller compacted concrete dams there are no physical contraction joints, but cracks are induced to form at
specific locations where waterstops are provided.
 12-2

In the Directorate of Water Affairs, the forming of contraction joints in the longitudinal
direction of the dam wall is not allowed for gravity dam structures unless the joints are
later grouted so that each dam block acts monolithically. This is a time consuming and
expensive process.

At gravity dam structures the transverse contraction joints are not grouted because each
block is capable of resisting its own loads safely.

Contraction joint grouting is however done at structures where stability is obtained by

collective arch- and gravity action, so where the full structure must act monolithically
under load conditions.

12.3 Expansion joints

Expansion joints are mainly provided to provide for expansion due to temperature
increases in structures. Expansion joints at gravity dam structures are usually found
downstream of the dam when another structure (canal, pump house, power station, etc.) is
erected directly against the dam wall. No connection between the two structures is made.
On the contrary, the joint is usually filled with an elastic material. The width of the joint
depends on the deflection which can be expected.

12.4 Construction joints

“Design of Gravity Dams (1201)” defines a construction joint as follows:

“A construction joint in concrete is defined as the surface of previously placed concrete

upon or against which new concrete is to be placed and to which the new concrete is to
adhere when the previously placed concrete has attained its initial set and hardened to such
an extent that the new concrete cannot be incorporated integrally with the earlier placed
concrete by vibration.”
 12-3

Construction joints are mainly needed to facilitate the work during construction but also
serve to provide cooling time so that temperature build-up due to hydration of the concrete
can be limited. Appendix A gives full details about the preparation of the various kinds of
construction joints before the placing of concrete may proceed.

12.5 Shear keys (shear grooves)

Shear keys are provided in the face of the contraction joint for the following reasons:

(a) To prevent differential settlement of adjacent dam blocks. This is done because
the bearing capacity of the foundation may vary from one place to another and to
bridge possible faults in the foundation.

(b) The shear keys (-grooves) increase the shear resistance between the blocks and
therefore the dam as a whole against shear.

“Design of Gravity Dams (1201)” states that shear keys are not always required at gravity
dam structures, but the author recommends that they are used because this will ensure a
safer design.

The advantage of the shear groove compared to the shear key is that the groove can be
attached permanently to the shuttering and can therefore be used repeatedly. Shear keys
are however better at preventing differential settlement.

A typical section of a contraction joint is indicated on drawing 85 119/82 which is included

as Figure 12.1. The contraction joint on this drawing is provided with shear keys.
 12.4
FIGURE 12.1: Typical contraction joint detail
 12-5

12.6 Waterstops

A dam structure consists of separate dam blocks which are separated by transverse
contraction joints*. At gravity dam structures it is not necessary to grout these joints and
the entire dam is therefore not monolithic. Water will therefore leak through these joints
unless waterstops are provided in the joints.

The layout of the waterstops is also indicated in Figure 12.1.

Previously metal seals made of copper or stainless steel were commonly used but these
days rubber- and/or polyvinyl chloride (PVC) waterstops are used. The PVC waterstops
have the advantage that they can be easily joined and that they are cheaper than those made
of rubber. A rubber waterstop can however handle larger movement without damage than
a PVC waterstop.

When the designer specifies a waterstop he must ensure that the chemicals in the water will
not erode the waterstop. The type which is prescribed is normally a centre bulb waterstop
which is pictured below.

FIGURE 12.2: Centre bulb waterstop

* With roller compacted concrete dams there are no physical contraction joints, but cracks are induced to form at specific
locations where waterstops are provided.
 12-6

The bulb body is in the plane of the joint and it is shaped to absorb relative movement.

Just a single waterstop is used in the spillway section on the downstream side to near the
foundation. This is done to drain the joint and thereby avoid the build-up of hydrostatic
pressure in the joint. The two waterstops on the upstream side are provided to doubly
ensure that positive waterproofing is obtained because the installation of waterstops is a
difficult task. In fact, the installation of the waterstop is more critical than the material
from which it is made.

Anchorage of the waterstop on the upstream side of the wall in the foundation is also an
important detail. A recommended layout (Detail A) is also indicated on Figure 12.1 and
should be used.

The width (W) of the waterstop depends on the maximum aggregate size (A) of the
concrete and the width (J) of the joint so that:

7 ≥ 6- + ¬
7<¬
n ≥
2

FIGURE 12.3: Waterstop width

 13-1

13.0 GALLERIES AND TUNNELS

13.1 Purpose

The first objective with galleries and tunnels in a gravity dam structure is to facilitate the
crucial inspections of the wall after its completion. This includes the inspection of the
foundations, the concrete, the drainage system of the structure, the reading and
maintenance of instrumentation, and the maintenance of drainage pumps. An additional
objective of the galleries in the structure is to allow access to carry out repairs which may
be required. By including drainage galleries and tunnels the construction of drainage- and
grouting holes is simplified. Secondary galleries may also be required for access, grouting
of construction joints, and cooling of concrete.

13.2 Location

FIGURE 13.1: Position of drainage gallery

 13-2

The recommended position of the drainage gallery is shown in Figure 13.1. The gallery
must be as close as possible to the foundation so that the drainage system operates as
efficiently as possible due to the low water level in the boreholes. In the longitudinal
direction the alignment must obviously allow the drainage water to flow to the deepest
point into the sump from which the water must then be pumped out.

Steep slopes against the flanks must be avoided because they will make inspections
difficult. The author recommends that slopes not be steeper than 30°. Such slopes must be
provided with convenient stairs and handrails on the sides. Secondary galleries which may
be required for grouting, access, and cooling of the concrete may be smaller and can be
closer to the upstream side to fit into the narrower portions higher up in the dam wall.

13.3 Dimensions

Experience has shown that a gallery with a height of 3,0 m, and a width of 2,5 m works
well. This size creates enough workspace for drill rigs, grout pumps, etc. to operate in
after the gallery is completed. Tunnels should not be smaller than 3,0 × 2,5 m because this
will make blasting of the tunnel difficult. With galleries, it is recommended that the roof is
arched to allow shuttering to be stripped more easily after concrete placement, and to
prevent unnecessary stress concentration. A drainage channel of 300 × 300 mm must be
provided in the gallery at a minimum slope of 1:200.

If the rock formations allow it, the tunnels may be left un-lined although a concrete floor is
essential to facilitate access.

13.4 Ventilation

It is essential that all galleries and tunnels have adequate ventilation. The minimum
requirement is the equivalent of a 300 mm diameter pipe at the end of each reach of the
gallery or tunnel.
 13-3

The pipes must furthermore be adequately protected against ingress of insects or birds
which may hinder the ventilation.

13.5 Access

Access must preferably be provided on both flanks. This not only improves ventilation,
but also provides more flexibility to the system. The entrances must be above the 200 year
flood line but preferably also above the PMF level to prevent flooding of the galleries and
tunnels which could endanger the stability of the dam. The position of the entrance/s
normally depends on the site layout and no fixed rule can be laid down. Appropriate
measures must be taken at the entrances to prevent undesired access by people, animals,
insects, and birds.

13.6 Construction

Galleries may be constructed in two ways, namely:

(i) In-situ concrete is placed around erected shuttering.

(ii) The gallery is formed by precast reinforced concrete elements which serve as
shuttering for the mass concrete. The advantage of this method is that it is faster,
ensures a better finish, and is easier to align.

13.7 Analysis

The gallery may be structurally analysed using the beam theory or a two-dimensional
analysis may be carried out. The latter method is more accurate but will take longer.
Loads must provide for the silt load and for the fact that the dam may later be raised.

Tunnels are analysed using available two-dimension computer software in the Directorate
such as “Planestress” *.

* This mainframe software no longer exists.

 13-4

13.8 Lighting

Adequate lighting must be provided in all galleries and tunnels. Waterproof fittings are
required, and all mounting materials must be corrosion resistant due to the corrosive nature
of the atmosphere in the galleries and tunnels.

13.9 Drainage

At the deepest point in the gallery or tunnel a sump must be formed in the concrete. A
submersible pump is normally placed in the sump to empty it automatically when required.
It is also preferable to create a facility on either side of the sump to measure the flow in the
respective reaches of the gallery into the sump. Figure 13.2 (Drawing 81 410/81) indicates
the detail of a prefabricated fibreglass Parshall flume* which may be used.

* Prefabricated stainless-steel flumes are now used.

 13-5
FIGURE 13.2: Detail of prefabricated Parshall flume
 14-1

14.0 CONCRETE FOR GRAVITY DAMS

Concrete is the most important material used in the construction of a concrete wall because
the wall consists exclusively of it.

To ensure stability, the concrete mast possess adequate strength, it must be durable for the
lifetime of the structure, and it must also be impermeable to allow the dam to impound
water. To comply with the above the following requirements apply:

(a) Ingenious engineers and technicians, who can design concrete mixes in a well-
equipped laboratory that meet the requirements of strength, durability, and
impermeability.

(b) Adequate quantities of suitable fine and coarse aggregate, cement,

slagment/pulverised fuel ash, additives, and water to continuously manufacture
good concrete.

(c) Enough trained personnel and reliable equipment to continuously produce and
place good concrete.

(d) Meticulous control over the mixing, placing, and post-placement treatment of the
concrete by the contractor on the site.

(e) Continuous quality control by means of the production and testing of concrete
cubes, and the regular reporting of results to the design engineer.

Because “Concrete for Gravity Dams” could be the topic for a dissertation of its own, we
finish by including the concrete specifications compiled by the author as Appendix A.
“Design of Gravity Dams (1401)” may be consulted if there are any specific topics about
which the reader needs to learn more.
 15-1

15.0 GAUGE PLATES AND WATER LEVEL RECORDERS

15.1 Purpose and uses

Gauge plates and water level recorders are indispensable components of any large dam
when it comes to operation and record keeping. These aids are used to monitor the water
level in the dam. During various operational tasks it is necessary to know the water level
in the dam, these are:

(a) When there is a calibrated uncontrolled spillway the discharge over the dam can be
calculated if the water level is known. Calibration of the spillway may be done with
a model study or by using the method described in chapter 6.

(b) When sluices are present, it is important during floods to know whether the water
level is rising or falling. It is especially important because opening too many
gates may cause a higher flood peak downstream in the river than that entering
the dam.

(c) By reading the gauge plates regularly (for example daily) a complete water level
record can be built up. Because the water level recorder results are plotted
continuously, an unbroken record is created. It is however necessary to read both
because mechanical problems may be experienced with the recorder. In fact, a
recorder must be regularly checked to determine if its reading is the same as that of
the gauge plates. The gauge plate reading is used to calibrate the recorder.

(d) With the water level known, it is easier for the Water Control Officer to determine
how much to open the river outlet valves to obtain a specific discharge, especially
since flow meters are no longer in common use.
 15-2

(e) The gauge plate reading is also used to determine the impounded water volume at a
specific time. By comparing the volume with the inflow and discharge the
evaporation losses, amongst others, can also be determined.

Readings obtained from recorders vary less because waves are supressed in the gauge well.
When the water level is low, the gauge plates are often difficult to read due to the long
distance between the reader and the plate.

15.2 Description

15.2.1 Gauge plates

Gauge plates are rectangular steel plates with a metric graduation applied to one side. The
plates are treated to provide resistance to corrosion and ensure durability.

15.2.2 Water level recorders

A float is placed in a gauging well near the upstream face of the wall. Small pipes at
regular intervals join the dam to the well so that the water level is the same in both. The
float is connected to a mechanism which records the water level automatically. An ink pen
plots the water level on special graph paper which fits around a rotating drum. The drum
usually makes one complete rotation per week but drums with longer rotation periods are
also available*.

15.3 Installation

Plates are installed, and the recorder so calibrated that the datum level (zero level)
corresponds with the lowest level where water can still be freely released from the dam
(Lowest drawdown level). Figure 15.1 indicates what is meant:

* These mechanical recorders have largely been replaced by electronic recorders which can be read remotely.
 15-3

FIGURE 15.1: Lowest drawdown level

The positions of gauge plates and recorders must be such that the water level in their
vicinity is not disturbed or drawn down by the spillway or outlets.

Gauge plates are normally attached to the dam wall itself with corrosion resistant mounting
materials. They must be as easy to read as possible at all times.

It is even necessary to attach single plates above the non-overspill section because gravity
dams can be completely inundated if the structure is designed for this.

Requirements for the layout of an automatic water level recorder are shown in Figures 15.2
and 15.3.
 15-4

SPECIFICATIONS:

1. 300 … ; … 700
j
-
2. - "¨ … where,
400
¨®8

j
Total cross sectional area
- "¨ <
of all the connection pipes
¨®8

- < Cross sectional area of recorder well

FIGURE 15.2: Typical layout of water level recorder

* Reinforced concrete pipes are now used for the recorder well rather than asbestos cement pipes, and 50 mm diameter
HDPE pipes are typically used for connection pipes.
 15-5

FIGURE 15.3: Relationship between depth and diameter of water level recorder well for specific
connection pipe spacing
 16-1

16.0 CONCLUSIONS AND RECOMMENDATIONS

16.1 Conclusions

(a) The main objective of making a design manual available which can be used in
design offices to design gravity dams has been achieved. The manual specifically
focuses on local conditions and the many years of experience of the Directorate of
Water Affairs has been chronicled as far as possible.

(b) If the standards, designs, and specifications contained in this manual are uniformly
applied costs may be saved.

(c) Sufficient information is available to comprehensively plan and design a gravity

dam. Knowledge regarding the topic keeps increasing. Industry journals and
symposiums contribute to the sharing of knowledge.

(d) Design of splitters may be accelerated now that an aid has been developed which
facilitates the initial choice of position.

(e) The manual meets the need to provide guidance to young engineers about the
planning and design of gravity dams.

16.2 Recommendations

Certain gaps in the existing knowledge have become obvious and the following
recommendations regarding future examinations are made.
 16-2

In the first place, it is recommended that a study be carried out to determine what amount
should be spent on geological surveys, during the reconnaissance- and design phases of
dam design. It still happens far too frequently that foundation conditions differ completely
after excavation is started, from what the geological reports predict. Cost increases which
appear out of proportion to additional geological investigation often occur. It may even
happen that the wrong dam site was selected due to inadequate geological surveys.

Corrosion of mechanical and electrical components still creates large problems in dams
due to the corrosive atmosphere which exists inside and outside the dam despite
ventilation. Research is still being done on this topic to find the most suitable treatment for
the respective components. Treatment is expensive but it still sometimes happens that it
does not meet the required objective. Research in this area should be accelerated to find
satisfactory solutions for the problem as quickly as possible.

Technical information regarding the flow characteristics of isolation– and control valves is
hard to obtain. This may be attributed to the manufacturers not wanting their product
information to become available to their competitors. In the author’s opinion the
manufacturers should be approached collectively to find a solution to the problem.

Grouting is currently still more of an art than a science. At design level a lot has been done
to place grouting on a more scientific footing. Much more needs to be done to test the
success of these measures by obtaining drill cores, digging inspection trenches etc. Only
then can experience be gained and the success of a method be determined with more
certainty.
 16-3

To be able to build cheaper dams remains a challenge in the modern age of rising costs and
inflation. At a concrete dam the concrete used is the largest expense. Current research
should be aimed at finding methods to reduce concrete unit costs. This research should be
focused on investigating cheaper mixes and well as potentially cheaper methods of
placement such as roller compacted concrete. Roller compacted concrete is concrete with
a low cement content (about 140 kg/m3) which has zero slump, and which is compacted
with vibrating rollers. At the same time, construction methods should be investigated
which will allow the best use to be made of roller compacted concrete, such as the
associated shuttering systems.

Flood control in dams with controlled spillways is of great importance because larger
floods may be caused downstream than incoming floods if too many sluice gates are
opened*. Better monitoring of runoff from the catchment area will make more accurate
operation possible. The telemetric instrumentation of the catchment area is a field which is
still fairly barren, and in which further work should be done.

While compiling this treatise the author has come to realise that a great deal of knowledge
exists regarding the topic, but that it is not currently presented in a usable format. He
therefore recommends that further design manuals are compiled to satisfy this need.
Possible topics for such treatises include the following:

(a) Raising of gravity dams

(b) Reinforced concrete design in gravity dams
(c) Computer software
(d) Instrumentation of gravity dams
(e) Concrete mixture design for mass concrete

Finally, the need also exists that similar manuals should be compiled which cover the
design of other types of dams, such as arch- buttress, earthfill dams, etc.

* The current practice in the Department is to avoid the use of controlled spillways completely. Many existing dams
with controlled spillways are having these replaced with alternatives, such as labyrinth or piano key spillways.
 A-1

APPENDIX A – CONCRETE SPECIFICATION

CONCRETE SPECIFICATION:

1.0 GENERAL

All concrete used in the construction shall comply with the requirements specified
hereafter.

2.0 MIX

2.1 Description

The mixes shall be designed for mass batching by the Materials Engineer. Batching by
nominal proportions or by volume shall not be permitted. The mix shall be designed so as
to produce concrete of optimum density, impermeability, durability and strength. For this
purpose aggregates of several nominal sizes shall be used as required.

2.2 Determination of Mix Proportions

Aggregate samples shall be forwarded to the Materials Laboratory so that suitable mixes
can be determined to conform to the specifications laid down in the table below.

3.0 BATCHING AND MIXING OF CONSTITUENTS

3.1 General

Concrete may be batched or mixed on the site, at a central plant, in a truck mixer or by a
combination of both. The operations of batching and mixing shall at all times be under the
control and supervision of a competent operator. The Resident Engineer (RE) shall afford
the Design Engineer (DE) reasonable access to all phases of concrete batching and mixing
operations. The RE shall arrange for the testing and calibration of all scales and measuring
devices as frequently as the DE may require.
 A-2

CONCRETE SPECIFICATION:
*S = Standard deviation Exterior Exterior Exterior
mass mass mass
Pumped
concrete concrete concrete
concrete
(Defined (Defined Reinforced (Defined
Interior Interior Interior (Also to be
as 1,25 m as 1,25 m Reinforced Reinforced Reinforced Prestressed concrete as 1,25 m No-fines
mass mass mass used for
minimum minimum concrete concrete concrete concrete (Aggressive minimum concrete
concrete concrete concrete float and
thickness thickness water) thickness
screeding
from from from
mixes)
exposed exposed exposed
faces) faces) faces)
Concrete class (28 day
strength in MPa/maximum 15/125 15/100 15/75 20/100 20/75 25/38 25/19 25/75 35/38 25/19 20/125 27/19 -
aggregate size in mm)
10% minimum strength at
28 days as measured and
150 mm cubes in MPa (wet 15 15 15 20 20 - - - - - 20 - -
screened to pass 38 mm
mesh)
5% minimum strength at 28
- - - - - 25 25 25 35 25 - 27 -
days in MPa

Maximum allowable
4 4 4 4,5 4,5 5 5 5 6 5 4,5 5 -
standard deviation in MPa
Target mean 28 day
strength for mix design
20 20 20 26 26 33 33 33 45 33 26 35 -
until degree of control is
known in MPa
Target mean 28 day
strength for mix design
15+1,3 S* 15+1,3 S* 15+1,3 S* 20+1,3 S* 20+1,3 S* 25+1,6 S* 25+1,6 S* 25+1,6 S* 35+1,6S* 25+1,6 S* 20+1,3 S* 27+1,6 S* -
after degree of control is
known in MPa
Portland cement/Slagment
30/70 30/70 30/70 50/50 50/50 50/50 50/50 50/50 100/0 50/50 50/50 50/50 50/50
ratio

Minimum cementing
150 150 150 200 200 250 250 250 300 400 200 300 -
materials content in kg/m3

Maximum cementing
- - - - - - - - 500 - - - -
materials content in kg/m3

Maximum water/cementing
0,70 0,70 0,70 0,55 0,55 0,55 0,55 0,55 0,55 0,55 0,55 0,55 0,40
materials ratio

Maximum size of aggregate

125 100 75 100 75 38 19 75 38 19 100 19 19
in mm

Minimum size of aggregate

- - - - - - - - - - - - 9,5
in mm

Entrained air (proposed) % 3 – 3,5 3 – 3,5 3 – 3,5 3 – 3,5 3 – 3,5 4 – 4,5 4 – 4,5 4 – 4,5 2,5 – 4,5 4 – 4,5 3 – 3,5 3,5 – 4,0 -

Minimum compacting
0,80 0,80 0,80 0,80 0,80 0,90 0,90 0,90 0,90 0,93 0,8 0,93 -
factor

Slump in mm 6 - 50 6 - 50 6 - 50 6 - 50 6 - 50 50 - 100 50 - 100 50 - 100 50 - 100 75 - 125 6 - 50 75 - 125 -

Aggregate/cementing
- - - - - - - - - - - - 5/1
materials ratio
 A-3

3.2 Batching of Constituents

The cement, slagment, aggregates, water, admixtures, and other constituents, as applicable,
shall be batched in the proportions as determined by the mix design. Each batch shall be
dumped cleanly into the mixer without loss of cement or spilling of the aggregates. The
constituents shall be batched as follows:

(a) Cement and Slagment

All cements shall be obtained from approved sources, and they shall comply with the
following specifications:

Portland Cement SABS 471*

Slagment SABS 626†

The RE may elect to use cement bagged during manufacture or delivered in bulk supply.
Batching of cement shall be by mass. Measurement of the cement by volume will not be
permitted under any circumstances. Bulk storage of cement shall be in approved
containers and batching shall be by mass measuring device capable of a delivery accuracy
of one percent of the required mass. Cement may be mass measured on a separate mass
meter. Loose cement shall be transported to the mixer either separately in a waterproof
compartment containing the total amount of cement required for the batch or placed in
contact with the aggregates comprising the batch provided that, in the case of the latter
method of transport, the concrete is mixed within ninety (90) minutes of such contact.
When cement is stored in sacks, they must be so arranged that they can be used in the order
of which they were delivered to site.

(b) Aggregates

Aggregates shall be transported or moved from the stockpiles or other sources of

storage or supply to the batching location or plant in such manner that grading
uniformity is maintained, and that contamination is avoided.

* SABS 471 has been replaced by SANS 50197-1.

† SABS 626 has been replaced by SANS 55167-1.

 A-4

Should separate supplies of aggregate material of the same size group, but of
different moisture content, specific gravity or surface characteristic affecting
workability be available at the batching location, withdrawals shall be made from
one supply exclusively and the material therein completely exhausted before starting
upon another. At the time of batching all aggregates shall have been dried or drained
sufficiently to result in a stable moisture content such that no visible separation of
water from aggregate will take place. In no event shall the free moisture content of
the fine aggregate at the time of batching exceed eight percent (8%) of its saturated,
surface dry mass. The batch-to-batch uniformity of all aggregates shall be such that
variations of moisture content within one (1) hour do not exceed seven kilograms (7
kg) in the batch quantity for one cubic metre (1 m3) of concrete or that a gradual
change does not exceed twice this variation in a period of four (4) hours. Batching
shall be as follows:

Batching shall be by mass. The RE shall provide such equipment and mass
measuring devices to enable the aggregates to be accurately batched by mass.

Bins should be so constructed that mixing of coarse aggregate in different sizes does
not occur. Flow of aggregate through gates in sidewalls of bins must be prevented.

Bin gradings on the different class sizes must be carried out daily until fairly
constant values are obtained then at least once weekly. Corrections should be made
regularly according to bin gradings to ensure that concrete is manufactured
according to proportions given in the mix design.
 A-5

(c) Water

Water must be clean and free of acids, alkalis, and organic matter, that may impair
the strength and durability of the concrete. Water, measured by volume or by mass,
shall be added at the mixer. Water containers and measuring equipment capable of
automatically discharging the desired amount of water directly into the mixer for
each batch of concrete shall be provided. The accuracy of measurement shall be
such as to deliver quantities not varying by more than two percent (2%) from the
indicated amounts. Calibration of the measuring equipment shall be checked on
each day on which concrete is mixed. Addition of water from uncalibrated
containers will not be permitted.

The total water content of the batch shall be considered to consist of the water added
in the mixer plus the water content of the aggregates. Aggregate water content shall
be predetermined and appropriately allowed for. Very rigid control of the quantity of
water added shall be exercised. Under no circumstances shall the water content of
any batch exceed that determined and specified.

(d) Admixtures

Admixtures shall be dispensed in liquid form. Dispensers shall have sufficient

capacity to measure at one time the full quantity required for once batch. Unless
liquid admixtures are added to pre-measured water for the batch, their discharge into
the batch shall be arranged to flow uniformly into the stream of water from
beginning to end of its flow into the mixer. Dosages of admixtures shall not vary
from the specified dosage by more than five percent (5%). If more than one
admixture is used each shall be dispensed by separate equipment. Chloride
containing admixtures are prohibited.
 A-6

3.3 Mixing

For the purpose of these specifications mixing shall be interpreted to imply wet mixing
where the required water content of the batch is added during the mixing operations. Dry
mixing of the cement and aggregates where added water is not so introduced shall be held
to mean bringing these constituents into contact for which the requirement is that mixing
shall be effected within ninety (90) minutes of such contact.

In the case of an automatic plant the weighing scales shall be so interlocked that a new
batch of materials cannot be delivered until the weighing hoppers have been completely
emptied of the previous batch.

For manual discharge from the hoppers a method of signalling shall be employed to ensure
that materials are not omitted or added more than once for each batch of concrete.

Mixing shall be by one or more of the following methods as approved:

(a) Hand mixing on site

The RE shall resort to hand mixing only when a small quantity of concrete for minor
work is required or in an emergency subject to the approval of the DE.

(b) Machine mixing on site

Concrete shall be mixed on site in mechanical batch mixers of the revolving drum or
pan type and of individual batch capacity not less than nought comma two cubic
metre (0,2 m3) nor less than a single cement sack batch. Each mixer shall be
equipped with suitable pickup and throw-over blades/paddle in the drum/pan
arranged in such a manner that concrete of the required
 A-7

quality and consistency may be produced within the specified mixing time. Blades
worn down more than twenty millimetres (20 mm) shall be replaced. In addition, the
mixer shall be equipped with an approved timing device to ensure mixing for the
minimum time specified.

At all times a spare mixer shall be held in readiness to run on fifteen (15) minutes
notice in the event of a breakdown. The volume of a batch shall not exceed the
manufacturer's rated capacity of the mixer. The charging of water into the mixer
shall begin before the cement and aggregates shall be operated at enter the drum.
During mixing, the drum speeds recommended by the manufacturer. The mixing of
the concrete in the machine shall continue for at least one comma five (1,5) minutes
and for not longer than five (5) minutes after the last of the constituents comprising
the batch has been added.

The concrete shall be mixed in such quantities as are required for immediate use.
Remixing of concrete will not be permitted.

On cessation of mixing for a period in excess of forty-five (45) minutes, the mixer
shall be thoroughly washed out before reuse.

The inner surfaces of the mixer shall be clean and free from hardened concrete.

(c) Central Plant mixing

Concrete shall be batched and mixed at a central mixing plant at the site of the works
or remote from the site. The provisions for machine mixing on site, specified
hereinbefore, shall apply equally.
 A-8

(d) Transit mixing

Cement and aggregates shall be batched at a central batching plant and the concrete
mixed in a mixer truck in transit to the site or immediately before discharge. The
truck mixer shall be of approved design and shall be operated according to the
manufacturer's recommendations. The volume of concrete in the drum shall not
exceed the manufacturer's rating.

4.0 TRANSPORTING, PLACING AND CURING

4.1 Transporting

Concrete shall, after mixing, be discharged into watertight barrows, buckets, hoppers,
trucks, or other approved containers for transportation to the point of placement.
Alternatively, for truck mixing, the concrete may be transported in the mixer prior to
discharge. No water may be added after mixing. The interval between the delivery of
batches shall in no case exceed thirty (30) minutes. Transportation may be effected by one
or more of the following methods:

(a) in non-agitator containers. These containers shall permit delivery of the concrete to
the location of placement in a thoroughly mixed and uniform mass. Uniformity shall
be deemed satisfactory if samples from the one-quarter and three-quarter points of
the load do not differ by more than five millimetres (5 mm) in slump. Covers shall
be provided where required for protection the elements. Concrete so transported
shall be placed within thirty (30) minutes of mixing.

(b) in agitator trucks. These trucks shall, when loaded to the Manufacturer's guaranteed
capacity, be capable of maintaining and discharging the concrete in a thoroughly
mixed and uniform mass. Concrete so transported shall be placed within ninety (90)
minutes of mixing.
 A-9

(c) Pumping of concrete shall not be permitted unless approved by the Engineer for a
specific purpose. Approved type of equipment adequate in capacity shall be used.
The equipment shall be so arranged that damage to freshly placed concrete due to
vibration is avoided. The operation of the pump shall be such that a continuous
stream of concrete without intervening air pockets is delivered. Subsequent to pump
the entire equipment shall be thoroughly cleaned.

4.2 General Placing

Concrete shall be placed in any part of the structure only after the forms, staging,
reinforcing steel, bearing assemblies and/or cable sheaths, mandrels and anchorages and
preparations for casting have been inspected and approved in writing by the RE. It shall be
placed promptly, with a minimum of handling to avoid the segregation of aggregates or the
displacement of reinforcement.

Each portion of the structure, between construction or expansion joints, as shown on the
drawings, shall be completed in a continuous operation with no interruption. The concrete
shall be deposited in the forms in horizontal layers as near to the final position as possible;
it shall not be deposited in a heap in one part of the mould and worked into position in the
process of compacting. The concrete shall be laid continuously in layers and each layer
shall be thoroughly compacted before being covered by the next layer. Where compaction
is by rodding and spading, the layers shall not exceed two hundred millimetres (200 mm)
in depth; where compaction is by vibration, the layers shall be as specified under Clause
4.4(b).

Where placing would involve dropping the concrete more than two metres (2m), it shall
not be freely dropped vertically, but must be lowered in barrows or buckets or dropped
through hoses, etc. Chutes, which cause the segregation of the aggregates, shall not be
used, nor shall water be added to prevent it sticking to the sides and bottom of the chutes.
Where the Engineer considers that segregation has occurred, the concrete shall be remixed
by hand before being placed in the forms.
 A-10

The concrete shall be well worked at all faces with properly constructed tools so as to
obtain a fair and clean surface. All concrete is to be worked into intimate contact with the
reinforcing steel without displacing it. Care is to be taken, as far as possible, to prevent the
formation of laitance. No plastering of surfaces will be allowed.

After the initial set of the concrete, the forms shall not be jarred, and no stress shall be
placed on the ends of projecting reinforcement until the concrete has taken its final set.

Mass concrete shall generally be placed in 1, 5 or 2 m lifts and spread in layers 500 to 550
mm thick. All concrete shall be thoroughly compacted by vibration immediately after it
has been placed in position. Before concrete is placed into the foundation, the foundation
shall be thoroughly cleaned and properly chipped, drained and dewatered so that no water
runs over or stand on the surface on which the concrete is being placed.

Immediately prior to concrete placement a layer of mortar approximately 15 mm thick and

of the same composition as the mortar in the concrete shall be evenly spread and
thoroughly brushed into all crevices so as to ensure complete contact between the
foundation and the concrete. The concrete shall be placed immediately on the fresh
mortar. Concrete shall not be placed during periods of prolonged or heavy rainfall.

4.3 Construction joints

4.3.1 Bonding Fresh Concrete to Older Concrete

(i) When the older concrete is not more than two (2) hours old: In the case of a
horizontal joint, the concrete of the previous lift shall be struck off before the
concrete stiffens up. The fresh concrete shall be placed without further preparation.
 A-11

In the case of vertical joints, the surface of the older concrete shall be wire brushed
to roughen the surface and all loose material shall be removed. The fresh concrete
shall be placed without further preparation.

(ii) When the older concrete is more than two (2) hours but less than three (3) days old:

In the case of a horizontal joint the surface of the older concrete shall be struck off as
described in (i) above. In the case of both vertical and horizontal joints the surface
of the old concrete shall be thoroughly hacked, and wire brushed or if possible sand
blasted to expose the coarse aggregate. This surface shall then be washed with clean
water to remove all laitance, dirt, and loose particles. Immediately before placing
the new concrete a thin layer of plastic mortar shall be applied to the surface of the
older concrete. The mortar shall consist of cement and sand mixed in the proportion
contained in the concrete mix, i.e., omitting the coarse aggregate. The fresh concrete
shall be placed against the layer of the mortar while it is still plastic.

(iii) When the older concrete is more than three (3) days old:

Preparation of surface before resumption of concrete – hardened concrete surfaces

on which fresh concrete is to be placed shall be clean, rough, wet, and free from all
laitance, and friable and greasy matter, in order to obtain the best possible bond.
Surfaces which have been cleaned and scoured while the concrete was green, will
generally only require cleaning with a high-pressure air/water jet. Surfaces where
curing has ceased and which have dried out, shall be kept continuously wet for 7
days prior to placing of fresh concrete. Surfaces which through long exposure have
become contaminated with oil grease, soil, or dirt; surfaces which were not cleaned
before the final set and shuttered construction joints, shall be scoured and roughened
until all unsatisfactory concrete and all laitance coatings and stains are removed.
 A-12

Immediately before concreting, all standing water shall be blown off the concrete
surfaces. A layer of mortar approximately 15 mm thick and of the same composition
as the mortar in the concrete shall, wherever practical, be spread uniformly over the
receiving surface and worked thoroughly into all irregularities by brushing. Concrete
shall then be placed immediately upon the fresh mortar.

(iv) Horizontal joints - concrete surfaces which form the faces of construction joints
shall, soon after initial set but before final set, be cleaned with high pressure air
water jets until all laitance is removed and clean sound aggregate is exposed. The
surface then shall be continuously moist cured for a minimum period of 14 days or
until covered with fresh concrete. At least 72 hours shall elapse between the vertical
placement of the 1,5 m or 2 m lifts.

(v) Vertical longitudinal joints - not normally allowed.

4.3.2 Temporary breaks

In the course of placing concrete, it may become necessary by reason of an unavoidable

breakdown of plant, or for other reasons acceptable to the Resident Engineer, to interrupt
concreting. On interruption of concreting, the exposed surface of the deposited concrete
shall be protected fully from the sun by covering with damp hessian. On resumption of
work, all laitance shall be cleaned from the surface of the deposited concrete, and freshly
mixed concrete shall be added immediately and thoroughly united with the old concrete.

If the period of interruption exceeds thirty (30) minutes, concrete shall be treated as
described under Clause 4.3.1 (i) above.

4.4 Vibration of concrete

(a) General

All concrete shall be vibrated to obtain the maximum density.

 A-13

Immersion Vibrators

The diameter of the vibrating element shall be as large as possible, consistent with
easy working between the reinforcing bars. The frequency shall be between 6 000
and 12 000 cycles per minute measured while the vibrator is immersed in concrete,
and the apparatus shall have sufficient power to enable it to vibrate the appropriate
mass of concrete without any appreciable retardation or labouring. There shall be at
least one vibrator at each point where concrete is being deposited, and one vibrator
shall be kept on hand ready for instant use in case of breakdown. The total number
of vibrators shall be such as to enable each batch of concrete to be compacted
without causing delay to the next batch.

Only men skilled and experienced in the use of vibrators shall be employed for this
type of work. To obviate displacement of the shuttering by vibration, additional
strength shall be provided in the formwork.

(b) For vibration, the concrete shall be laid continuously in layers, the depth of which
does not exceed the length of the vibrating element. Each such layer of concrete
shall be thoroughly compacted before being covered by the next layer. At corners
and ends, however, the concrete should be raised slightly above the surrounding
level to obviate accumulation of mortar at these points. The vibrator shall be
inserted vertically into the concrete and shall be withdrawn slowly. The spacing of
the insertions shall not exceed one metre (l) or a distance equal to the radius over
which the vibration is visibly effective. Internal vibration shall be applied directly to
the concrete and not through the forms or reinforcing steel. The vibrating element
should be inserted to the bottom of freshly laid concrete but should not disturb
partially hardened layers. Under no circumstances shall the vibrator be used to push
concrete or cause it to flow horizontally into position.
 A-14

(c) Period of Vibration

The period of vibration cannot be stated with certainty but must be determined
carefully by trial. In general, the period shall not be less than one hundred (100) or
more than two hundred (200) seconds per square metre of surface area over which
the vibration is effective. The intensity and period of vibration, however, shall be
sufficient to cause complete consolidation of the concrete, as indicated by the
cessation of contraction and the appearance of a thin layer or mortar on the surface of
the concrete. Vibration shall be discontinued immediately after the appearance of
these symptoms. After vibration, the concrete shall be spaded in corners, angles and
against forms to release air bubbles which may have been trapped in these positions.

4.5 Protection of Concrete whilst curing

During the first stages of hardening, concrete shall be protected from the harmful effects of
sunshine, drying winds and cold, and also from running or surface water and shocks.

No concrete shall be placed when the temperature is at or below 2°C. During cold
weather, when the temperature remains below zero for extended periods, the concrete shall
be protected by means of tarpaulins, straw covering or similar means, to the satisfaction of
the Engineer, for at least five (5) days after placing. Care shall be exercised to avoid
laying the covering materials too soon after placing the concrete and thus causing foreign
matter to adhere to the fresh concrete.

All horizontal or other un-shuttered faces shall be protected against damage and markings
from persons or vehicular traffic during the curing period.
 A-15

4.6 Curing

Particular attention shall be paid to curing. All exposed surfaces shall be kept continuously
moist or otherwise effectively sealed for a minimum period of 14 days. Tops of horizontal
construction lifts shall be ponded for the minimum period or until the next lift is placed.
Other surfaces can be either: ponded, continuously sprayed, covered with a moisture
retaining material and kept continuously wet, or membrane cured by the application of an
approved liquid curing compound.

5.0 CONCRETE FINISH

5.1 Screeded and floated finishes to spillway crest - The concrete must be well vibrated and
finished with a heavy screed. If bleed water does not dry from the surface within 1 hour, it
shall be mopped up with blankets or clean hessian. When the concrete has taken its initial
set, as determined by firm pressure with the finger leaving an indentation 3 to 4 mm deep,
the surface shall be steel floated with a few and as heavy strokes as possible. If the float
leaves ridges in the finish, the surface shall be re-floated after 30 minutes.

5. 2 Shuttered finishes - The shutters shall be dressed on the interior surface with a suitable
shutter preparation agent prior to any concreting. The shutters must be cleaned, freed from
any concrete particles, and redressed after every usage. Panels should be discarded as soon
as the surface deteriorates in any way. Fillets shall be placed in the corners of the shutter
surfaces. Care must be taken that construction joints are in a straight line. This is purely
for aesthetic reasons.

6.0 CONCRETE QUALITY CONTROL

The following procedure for sampling and testing of concrete shall be adhered to:

6 .1.0 Initial concrete (Defined as that concrete placed at the time when final aggregates and/or
final mix design are not yet available)
 A-16

6.1.1 For testing of 28 day strength

3 x 150 mm cubes from one mix for every 50 m3 concrete placed with a maximum of 12
cubes per day.

6.1.2 For testing of 7 day strength

2 x 150 mm cubes from the same mixes as from which the 28 day strength cubes are taken
for every 50 m3 of concrete placed.

6.2.0 Production concrete (Defined as the concrete placed when the final mix design is used and
the batching plant is in full operation).

[Link] At start of production

[Link] For testing of 28 day strength

120 x 150 mm [Link] from 40 different mixes at 3 cubes per mix plus
4 x 300 mm cubes taken at 1 cube from every 10th mix.

[Link] For testing of 7 day strength

2 x 150 mm cubes from each mix which is sampled for 28 day strengths.
The cubes at the start of the production stage shall be taken within the shortest possible
period.

[Link] First three months after start of production

[Link] For testing 28 day strength

6 x 150 mm cubes from 6 different mixes per day taken at near equal time intervals over
that day plus 1 x 300 mm cube every second day.

[Link] Quality control for remainder of concrete placing programme.

[Link] For testing 28 day strengths

3 x 150 mm cubes per day taken from 3 different mixes at near equal intervals for a given
day plus 1 x 300 mm cube every week.

6.3 The 150 mm cubes shall be tested on site. The 300 mm cubes shall be sent to the Materials
Laboratory for testing.

6.4 The above procedure shall be repeated for each type of concrete.
 A-17

6. 5 A register of numbering shall be kept for concrete, sand, and aggregate samples.

6.6.0 A daily register shall be kept in which the following properties and data concerning the
concrete cubes shall be noted.

6.6.1 Mix proportions

6.6.2 Type of concrete

6.6.3 Workability: compaction factor and slump

6.6.4 Where sampled (mixer number)

6.6.5 Where placed (block number, reduced level, and chainage)

6.6.6 Weather condition at day of placing (cloudy sky, windy, not or cool, etc.)

6.7 Weekly reports shall be made in triplicate on form DW 614 for each type of concrete and
sent to:

(a) Chief Engineer: Civil Design

(b) Control Technician, Materials Laboratory
(c) Resident Engineer's file

6.8 Moulds for test cubes

Moulds for the making of test cubes shall be supplied by the Materials Laboratory and
shall be of approved construction. The sides shall be of steel or cast iron at least twelve
millimetres (12 mm) thick and shall be so fabricated that all faces of the cube are plane and
smooth, opposite faces being parallel and adjacent faces truly square to one another. The
mould shall be provided with a base plate attached to the mould by screws or bolts, so that
no leakage of water during filling occurs. All faces of the mould shall be lightly oiled to
prevent adherence of the concrete.
 A-18

The number of moulds shall be sufficient to make the quota of cubes specified.

6.9 Compacting of test cubes

Test cubes are to be compacted either by hand or by mechanical vibration.

(a) Compaction of hand

Where compaction of the concrete in the mould is done by hand the test cube shall
be compacted with a steel bar four hundred millimetres (400 mm) long and having a
ramming face of six hundred square millimetres (600 m2) and a mass of two
kilograms (2 kg). The concrete shall be placed in the mould in layers of
approximately fifty millimetres (50 mm) deep and each layer tamped with not less
than thirty-five (35) strokes of the bar before the succeeding layer is placed. The
strokes of the bar shall be distributed in a uniform manner over the cross-section of
the mould.

After the tamping of each layer, a mason's trowel shall be worked lightly along the
sides of the mould to ensure a smooth finish on the faces of the cube. After the top
layer has been compacted, excess concrete shall be struck off and the surface finally
finished off level with the top of the mould using a trowel.

(b) Compaction by Mechanical Vibration

Where the concrete in the structure is being compacted by mechanical vibration, the
test mould shall be completely filled with the concrete proud of the surface of the
mould. The cube shall be compacted by vibrating platform or by placing a
mechanical vibrator against two opposite outer sides of the mould in turn, so as to
produce full compaction of the concrete with neither segregation nor excessive
laitance. If necessary, the vibrator shall be suitably cushioned to prevent concrete
from being flung out of the mould due to violent vibration.
 A-19

On completion of compaction any excess concrete shall be struck off and the surface
finally finished off level with the top of the mould using a trowel.

6.10 Curing of Test Cubes

Where site curing of the concrete is by the method specified in Clause 4.6 the cubes in the
moulds shall be stored indoors in a place free from vibration and draughts, under damp
sacks or wet sand. After twenty-four (24) hours the cubes shall be removed from the
moulds, clear distinguishing marks painted on, and then placed in a water bath. The marks
shall be painted on the top or bottom surface of the cube as moulded, so that any pair of
sides, as cast, may be used for testing. The cubes shall be cured in the water bath until they
are ready for testing or for despatch to the laboratory. The water bath, which shall be
indoors, shall be protected against extremes of temperature and the water kept at a
reasonably constant temperature. The cubes shall be completely immersed in the water
and shall be protected against the rays of the sun and against wind and frost.

6.11 Cube testing equipment

All concrete cubes shall be tested in a laboratory approved by the DE. The cube testing
equipment shall be provided by the Materials laboratory. The testing machine shall be
maintained in good working order and tested, calibrated, and serviced at regular intervals
to ensure true and accurate readings.

6.12 Packing and Transport of test cubes

All test cubes not being tested on site shall be despatched to the Materials Laboratory so as
to ensure their arrival before the time when testing is due. Subject to this requirement the
cubes shall remain at the site of works as long as possible. All test cubes must be
submitted with a completed form DW 36 (in triplicate) giving full details.
 A-20

The cubes shall be well packed in wet sand or other suitable damp material in such a way
that the cubes shall remain damp during transit. Sturdy packing boxes or containers shall
be used so that the risk of damage during transit is kept to the minimum.

6.13 Testing of Concrete Cubes

Testing of the cubes shall be in accordance with British Standard Specification B.S.l881.

6.14 Test results

The compressive strength of each test cube, tested under standard conditions, shall be
calculated in megapascals (MPa) from the maximum load sustained by the cube before
failure. The compressive strength of the concrete shall be defined as the average
compressive strength of the test cubes of the same mix. The compressive strength of the
concrete shall not be less than the strength required for works cubes as laid down in the
table and in addition no single cube shall have a strength less than eighty-five percent
(85%) of the minimum strength specified for works cubes at twenty-eight (28) days. The
compressive strength of prestressed concrete, in addition to satisfying the twenty-eight (28)
day requirements, shall be such that the required strength as specified and/or indicated on
the drawings at prestressing age or at an age of twenty-eight (28) days, whichever is the
lesser, is achieved.

6.15 Rejection of Concrete

Concrete failing to satisfy the strength requirements, containing areas of excessive

honeycombing, or displaying excessive shrinkage cracking or other defects which, in the
opinion of the Engineer, is likely to prove deleterious to the strength of the structure as
designed or to its long-term durability will be considered as sufficient cause for rejection.
 A-21

7.0 CONCRETE REPAIR

7.1 Immediately after shutter stripping the concrete shall be inspected for defects by the
responsible Quality Control Engineer or his representative and areas of honeycombing and
other defects shall be marked by him.

7. 2 The defective areas shall be chipped or cut so that solid coarse aggregate is exposed
ensuring that all defects are removed to the satisfaction of the Quality Control Engineer.

7. 3 Deep cavities can be repaired by the dry pack method, or concrete replacement, depending
on their shape and shallow cavities by mortar replacement.

7.4 Particular attention shall be paid to proper keying and bonding, and a suitable epoxy
bonding agent used where required.

7.5 Reference

CONCRETE MANUAL
Chapter 7 – Repair and Maintenance – U.S. Bureau of Reclamation.

8.0 DEGREE OF CONTROL

The standard deviation of the cube test results from their mean value is regarded as an
index of the scatter and consequently of the degree of site control.

The table below indicates the degree of control for general construction testing.
 A-22

Degree of site control Standard deviation MPa.

Excellent Below 3,0

Very good 3,0 - 4,0
Good 4,0 - 5,0
Poor 5,0- 7,0
None Above 7, 0

9.0 PERMISSIBLE DEVIATIONS

The Contractor shall construct each of the various parts of the Works within the limits of
the permissible deviation set out below appropriate to the degree of accuracy specified in
the project specification or shown on the drawings.

If no degree of accuracy is specified, Degree of Accuracy II shall apply.

Permissible deviation, mm
Degree of accuracy
III II I
(a) Reinforcement

(1) Spacing between two adjacent bars ±25 ±20 ±15

(2) Longitudinal location of bends and ends

of bars ±40 ±30 ±20

(3) Cover to reinforcement +30-0 +20-0 +10-0

 A-23

Permissible deviation, mm
Degree of accuracy
III II I
(b) Formwork

Formwork shall be so constructed as to

ensure that the position of the finished
work will be as specified, subject to the
relevant permissible deviation given in
(c) or (d) below, as applicable.

(c) Foundations: mass and reinforced

concrete

(1) Position on plan of any edge or surface

measured from the nearest grid line or
agreed centre line ±50 ±35 ±20

(2) Linear dimension on plan cast against

excavation sides ±60 ±40 ±20

(3) Linear dimension on plan cast against

formwork ±30 ±20 ±10

(4) Formation level -40+20 -30+15 -20+10

(5) Surface level -30+15 -20+10 -10+5

(d) Elements or components above

foundations

(1) Position on plan of any edge or surface

measured from the nearest grid line or
agreed centre line ±25 ±15 ±5
 A-24

Permissible deviation, mm
Degree of accuracy
III II I
(2) Linear (other than cross-section)
dimensions ±30 ±20 ±10

(3) Cross-section dimensions -10+20 -5+15 ±5

(4) Level (deviation from designed level

with reference to the nearest transferred
datum (TD) of the upper or lower
surface, as may be specified, of any slab
or other element or component) -20+10 -15+5 -10+0

(5) Vertically 5 per 5 per 2 per

metre metre metre
height height height
with with with
maximum maximum maximum
of 70 of 50 of 30

(6) Abrupt changes in a

continuous surface 10 5 2

(7) Flatness of plane surface (maximum

deviation of the surface from any
straight line of length 3 m joining two
points on the surface, determined by
means of a straight edge the ends of
which are supported on identical blocks
of suitable thickness placed one over
each of the two points). 10 5 3

The tolerances for bow or camber, twist, and squareness, and slip form concrete,
prestressed concrete, and precast concrete will be stated in the project specification where
applicable.
 A-25

(e) Cover to reinforcement

No deviation from the minimum cover of concrete over reinforcement specified will
be permitted.

10.0 GROUTING/CONCRETING OF RECESSES OR MECHANICAL PARTS

For the grouting/concreting of recesses or mechanical parts an expanding additive shall be

used in the following proportions of cement:

grout: 0,50% water/cement ratio: 0,33 to 0,37

concrete: 0,25% maximum water/cement radio: 0,50

The expanding additive shall be Injex powder* or an approved equal.

Injex powder is available from:

KLS-Sternson
Roosevelt Park
2129

* This product no longer exists.

 B-1

REFERENCES

501 Office notes by Mr. F. Hollingworth (Deputy Chief Engineer) Directorate of

Water Affairs, November 1970

502 Conversations with Mr. F. Druyts (Deputy Chief Engineer) Directorate of

Water Affairs.

503 Conversations with Mr. Kovacs (Principal Engineer) Directorate of Water

Affairs.

504 United States Department of the Interior Bureau of Reclamation:

Design of Small Dams. Third Printing, 1965.

601 Design of Gravity Darns. United States Department of The Interior.

Bureau of Reclamation. United States Government Printing Office,
Denver: 1976.

602 F. Hartung, Gates in Spillways of Large dams. Question 41, Eleventh

Congress on Large Dams (ICOLD - 1973).

603 Hydraulic Design of Spillways. Headquarters, Department of The Army.

Office of the Chief of Engineers, D.C., 20315. EM 1110-2-1603.
31 March 1965.

701 Hydraulic Design of Spillways. Headquarters, Department of The Army.

Office of the Chief of Engineers, D.C., 20315. EM 1110-2-1603.
31 March 1965.

801 Class notes, Applied Hydraulics 5SD, University of Pretoria. 1974.

Hydraulics of Dams. T.P.C. van Robbroeck.

802 Trashracks and Raking Equipment. Thaddens Zowski. Water Power,

September 1960, p342-348.
 B-2

803 Design of Gravity Darns. United States Department of The Interior.

Bureau of Reclamation. United States Government Printing Office,
Denver: 1976.

804 Hydraulic Design of Reservoir Outlet Structures: Engineering and

Design. USA Army Corps Engineers. United States of America,
Department of the Army, Engineer Manual EM 1110-2-1602. Year 1963.

805 Centrifugal Pump Texican. KSB Frankenthal, December 1975.

806 Manual of British Water Engineer Practice Volume II, Fourth Edition.
Engineering Practice. Institution of Water Engineers 1969.
W.O. Skeat and B. Dangerfield.

807 Hydraulic Design Criteria, Sheet 228-1, Bend Loss coefficients,

Waterways Experiment Station, U.S. Army Engineers, Vicksburg, Miss.
USA.

808 Streeter, Victor L. Fluid Mechanics. Fifth edition. McGraw-Hill

book company, 1971.

809 Ven te Chow. Open-channel Hydraulics. McGraw-Hill book company,

1959.

810 Conversations with met Mr. Keinhofer of Ainsworth Engineering

Johannesburg, Telephone (011) 6157190.

901 "Hydraulic design of stilling basins and energy dissipaters" USBR,

Engineering Monograph, No. 25, 1964.

902 Design of Gravity Dams. United States Department of the Interior.

Bureau of Reclamation. United States Government Printing Office
Denver, 1976.
 B-3

903 Elevatorski, E .A., "Trajectory bucket - type energy dissipators"

Journal of the Power Division, ASCE, PO 1, February 1958.

904 Department of Water Affairs, Forestry and Environmental Affairs.

Division of Special Tasks. "Hydraulic Design of Dams". Report V/H001
July 1980. Dr C.P.R. Roberts.

905 Class notes. Applied Hydraulics. University of Pretoria, May 1972.

Hydraulic Design Aspects of Controlled and Uncontrolled Spillways by
Dr C.P.R. Roberts, Department of Water Affairs.

906 Roberts, D.F. The dissipation of energy of a flood passing over a

high dam. Transactions, South African Institution of Civil
Engineers, Vol. XLl, 41st Session, 1943.

907 Orange River Project, Hendrik Verwoerd and Van der Kloof Dams.
“Overspill Crest and receiving apron”. Sogreah, R8 939, February 1965.
(Consultants to International Orange River Consultants (Pty) Co.).

908 Orange River Project, P.K. le Roux Dam: Tests on Flood Spillways and
Power Station Intake. Sogreah, R360 063, September 1977.

909 Roberts, C.P.R. "Energy dissipation by dam crest splitters". The Civil
Engineer in South Africa, November, 1977.

910 Bekker, F.W. Divergent canal transitions with curved bottoms for
supercritical flow. The Civil Engineer in South Africa, January 1981,
Volume 23, November 1, Page 11.

1001 Technical Memorandum 646. Pressure Grouting. United States

Department of the Interior Bureau of Reclamation. Design and
Construction Division. Denver, Colorado June 1957.
 B-4

1002 Grouting and its use with water retaining structures. D.B. Badenhorst. Office
notes Department of Environmental Affairs.

1003 "Foundation Grouting for dams" - Part I: Investigation. Bulletin 47,

April 1977 of the Australian National Committee on Large Dams.

1004 "Foundation Grouting for dams" - Part II: Design of the Australian
National Committee on Large Dams. Bulletin 48, August 1977 of the
Australian National Committee on Large Dams.

1005 "Foundation Grouting for dams" - Part III: Construction Bulletin 50,
August 1978 of the Australian National Committee on Large Dams.

1006 Design of Gravity Dams. United States Department of the Interior.

Bureau of Reclamation. United States Government Printing Office
Denver, 1976.

1007 Proceeding of the A.S.C.E. Journal of the Soil Mechanics and

Foundation Division. April 1962.

1008 Grouting Manual, Fourth Edition 1981, Water Resources Commission,

N.S.W., Australia.

1101 Design of Gravity Dams. United States Department of the Interior.

Bureau of Reclamation. United States Government Printing Office
Denver, 1976.

1201 Design of Gravity Dams. United States Department of the Interior.

Bureau of Reclamation. United States Government Printing Office
Denver, 1976.

1401 Design of Gravity Dams. United States Department of the Interior.

Bureau of Reclamation. United States Government Printing Office
Denver, 1976.

1501 Conversations with Mr. J.J. van Heerden, (Principal Engineer),

Department of Environmental Affairs.