01-Cam-Road Part 1 - Geometry
01-Cam-Road Part 1 - Geometry
bTdæan
sikSaKMerag
pøÚvfñl;
ROAD DESIGN
STANDARD
PART 1. GEOMETRY
CAM PW.03.101.99
2003
This document has been produced for the Kingdom of Cambodia as a joint Australia –
Cambodia project sponsored by the Australian Agency for International Development
(AusAID).
Valuable assistance and operational advice was provided by the staff of the Cambodian
Ministry of Public Works and Transport (MPWT) as follow:
I. Steering Committee (Appendix B)
1. Mr. Tan Hay Sien, Director of Infrastructure Department ............................................ Chairman
2. Dr. Yit Bunna, Director of Public Works Research Centre ........................Deputy Chairman
3. Mr. Tauch Chan Kosal, Director of Heavy Equipment Centre ............................................... Member
4. Mr. Lim Sidenine, Deputy Director of Bridge Construction Unit.................................... Member
5. Dr. Phung Katry, Director of Waterway Department ................................................... Member
6. Mr. Prum Sakun, Deputy Director of Cambodian Royal Railway................................. Member
7. Representative from Sihanouk Ville Port (Mr. Ma Sun Huot)................................................... Member
8. Representative from Public Works Laboratory (Mr. Keo Leap)................................................ Member
9. Representative from Phnom Penh Institute of Technology (Mr. Chhouk Chhay Horng).......... Member
10. Representative from Phnom Penh Public Works Department (Mr. Heng Nguon) ................... Member
11. Representative from Ministry of Water Resources and Meteorology ....................................... Member
Technical research and specialist input was provided by the Australian consulting firms of
McMillan Britton & Kell Pty Limited and Willing & Partners Pty Ltd.
Reproduction of extracts from this publication may be made subject to due acknowledgment
of the source.
Although this publication is believed to be correct at the time of printing, neither the MPWT nor
AusAID accept responsibility for any consequences arising from the use of the information
contained in it. People using the information should apply, and rely upon, their own skill and
judgement to the particular issue which they are considering.
SECOND PRINTING
FINANCED BY THE ASIAN DEVELOPMENT BANK LOAN NO. 1659 CAM (SF)
ROAD DESIGN STANDARD
FOREWORD
The Cambodia Road Design Standard is intended to be used for the design of all
new roads in the Kingdom of Cambodia. The Cambodian Road Design Standard
consists of the following complementary documents which shall be considered
together:
For the purpose of regulating and interpreting the provisions of this Standard, the
AUTHORITY shall be the Cambodian Ministry of Public Works and Transport.
TABLE OF CONTENTS
1.1 INTRODUCTION........................................................................................ 3
1.1.1 Design Standards....................................................................................... 3
1.2 DESIGN STANDARDS & ROAD CLASSIFICATION ............................... 3
1.2.1 Design Factors ........................................................................................... 3
1.2.2 Road Standards ......................................................................................... 5
1.2.3 Road Classification .................................................................................... 6
1.2.4 Access Control ........................................................................................... 9
1.2.5 Design Standards..................................................................................... 10
1.2.6 Utilities...................................................................................................... 10
1.3 DESIGN CRITERIA ................................................................................. 11
1.3.1 Topography .............................................................................................. 11
1.3.2 Design Vehicles and Characteristics ....................................................... 11
1.3.3 Speed ....................................................................................................... 15
1.3.4 Capacity ................................................................................................... 16
1.4 HORIZONTAL ALIGNMENT ................................................................... 20
1.4.1 General..................................................................................................... 20
1.4.2 Movement on a Circular Path .................................................................. 20
1.4.3 Superelevation ......................................................................................... 21
1.4.4 Maximum Side Friction............................................................................. 21
1.4.5 Horizontal Curvature ................................................................................ 23
1.4.6 Plan Transition Curves............................................................................. 24
1.4.7 Application of Superelevation .................................................................. 26
1.4.8 Superelevation Development Length....................................................... 27
1.4.9 Positioning of Superelevation Development ............................................ 29
1.4.10 Small Changes in Alignment.................................................................... 30
1.4.11 Widening on Horizontal Curves ............................................................... 31
1.4.12 Curves with Adverse Crossfall ................................................................. 32
1.4.13 Radius to Meet Sight Distance Requirements ......................................... 32
1.5 SIGHT DISTANCE................................................................................... 34
1.5.1 General..................................................................................................... 34
1.5.2 Driver Capability....................................................................................... 34
1.5.3 Constants Assumed for Calculation of Sight Distance ............................ 35
1.5.4 Height of Eye and Object Cut Off Height ................................................. 35
1.5.5 Stopping Sight Distance........................................................................... 36
1.5.6 Passing Sight Distance ............................................................................ 38
1.6 CROSS SECTIONS ................................................................................. 40
1.6.1 Elements of Cross Sections..................................................................... 40
1.6.2 Crossfalls ................................................................................................. 40
1.6.3 Lane Widths and Marginal Strips ............................................................. 41
1.1 INTRODUCTION
Geometric road design standards are used as an aid to achieving consistent and
operationally effective road designs. Rapid expansion and improvement to road networks
in developing countries precipitated the need for standards to:
(b) enable satisfactory designs to be produced, even where there is not a high degree of
expertise,
(c) ensure that road funds were not misspent, through inappropriate designs, or through
inadequate provision for future traffic growth or current operations.
The principal factors influencing the choice of design standard for a road are set out in the
following clauses.
The appropriate design standard for a particular road depends on both the overall
availability of finance and the state of development of the road network. When the overall
network is substantially adequate and finance is available, improvement projects will be
directed at operational safety and efficiency, and higher geometric standards are
appropriate. When the network is inadequate in terms of traffic demand, and funds are
limited, geometric standards may be lowered selectively on parts of the road system. The
state of the network and the funding position are partly dependent on population, and the
developed area over which the network must spread. In Cambodia, these vary between
geographical and administrative regions, and it is reasonable that the appropriate design
for individual roads should differ somewhat between regions.
1.2.1.2 Terrain
Research indicates that terrain is one of the factors that modify drivers’ expectation of
being able to maintain speed. Terrain is therefore included as one of the elements used in
selecting the design parameters in this standard. Terrain also has a significant effect on the
costs of achieving high geometric standards. In flat inland areas, high standard alignments
fit the terrain and can obviously be achieved at practically no additional cost, while in steep
mountainous country, costs escalate rapidly as standards are increased.
The volume of traffic to be carried by a road can be considered as the design loading that
the designer must satisfy. Traffic volume is an important design parameter and therefore a
complete methodology for considering traffic volume in the design of roads is provided in
this guide. Factors to calculate traffic volumes in terms of passenger car units and for
calculating the design capacity of a proposed road are provided.
The proportion of heavy commercial vehicles in the traffic stream influences not only the
structural design of the road pavement, but also some aspects of geometric design, be-
cause of the disparity of speeds on grades and also because of their greater width. The
effect of heavy vehicles lowers the level of service provided by the road due to two factors:
(a) The vehicle takes up more road than a car and thus a heavy vehicle is equivalent to
more than one car.
(b) The disparity of speeds leads to increased queuing and overtaking requirements.
1.2.1.5 Safety
Whatever design standard is adopted, safety is a major goal of road design. The theme of
enabling the driver to perceive hazards in time to take appropriate action, and of providing
geometric parameters appropriate to the likely speed of operation, runs throughout the
Standard. Further, vehicles can get out of control, and items like traversable batter slopes,
guard fencing, breakaway light standards and sign supports are desirable attributes of
what has been described as a ‘forgiving’ roadside.
Roads fall into a hierarchy of categories ranging from international expressway to District
road in rural areas and urban expressway to local road in urban areas. As road category is
one of the elements used to select design parameters a full definition is provided of each
category to assist designers.
1.2.1.7 Environment
1.2.1.8 Energy
The total road task makes considerable use of liquid fuels and other products derived from
crude oil. Grades exceeding about 5 % cause greater consumption of fuel by heavy
vehicles in the uphill direction than they save in the downhill direction. However, the
greatest changes in energy consumption related to transport spring from questions of
appropriate modes for long freight haulage, and from land use distributions in urban areas.
At the present stage, flattening of grades can rarely be justified on the basis of energy
saving alone.
In a situation of changing land use and growing traffic, no road can ever be regarded as
‘final’. There will always be requirements for future augmentation or modification. Where it
is obvious that medium term requirements would modify the best short term design for a
particular road, it is often possible to modify the design slightly to provide better options for
future action. While this ties up some funds and prevents their use on other current
projects, the effect can be much less than if the longer term design was adopted in the first
instance.
Urban areas are defined as areas having a population of at least 1,000 where buildings
and houses are gathered and business activity is prevalent. It covers all areas within the
gazetted Municipality limits and also includes areas expected to become urbanised within
the design period. Rural areas can be regarded as areas other than urban areas.
There is no fundamental difference in the principles of design for rural and urban roads.
Roads in urban areas, however, are characterised by busy pedestrian activities and
frequent stopping of vehicles owing to short intersection spacings and congested built-up
areas. Lower design speeds are usually adopted for urban roads and different
cross-sectional elements are applied to take into account the nature of traffic and adjoining
land use. It is for these reasons that variations in certain aspects of geometric design are
incorporated for these two broad groups of roads.
The design standard is classified into six groups (R6, R5, R4, R3, R2 and R1) for rural
areas and into six groups (U6, U5, U4, U3, U2 and Ul) for urban areas. These are in
descending order of hierarchy.
Roads which function to provide long distance travel, will require higher design speeds
whilst road which serve local traffic, where the effect of speed is less significant can have a
lower design speed. Also roads with heavier traffic will be provided with a higher standard.
(a) Standard R6/U6 : Provides the highest geometric design standard for rural or urban
areas. They usually serve long trips with high speed of travelling, comfort and safety.
It is always designed with divided carriageway and with full access control. The Rural
and International Expressways fall under this standard.
(b) Standard R5/U5: Provides also high geometric standard and usually serve long to
intermediate trip lengths with high to median travelling speeds. It is usually with partial
access control. The Rural Highway and Arterial fall under this standard.
(c) Standard R4/U4: Provides medium geometric Standard and serve intermediate trip
lengths with medium travelling speeds. It is also usually with partial access control.
The Rural Highway, Major Provincial, Minor Arterial and Major Collector fall under this
standard.
(d) Standard R3/U3: Provides low geometric standard and serves mainly local traffic.
There is partial or no access control. The Minor Provincial and Minor Collector fall
under this standard.
(e) Standard R2/U2: Provides the lowest geometric standard for two-way flow. It is
applied only to local traffic with low volumes of commercial traffic. District Roads and
Local Streets fall under this standard.
(f) Standard R1/U1: Provides very low geometric standard and is applied to very low
traffic where the chance of two-way flow is low.
The function of each road is defined according to its role in the National, Provincial or
Urban road networks. Roads are categorised considering their transportation
characteristics of trip length, design volume and design speed. The conflicting transport
requirements of mobility and accessibility are key design features and are used to
differentiate between the category design standards. Details of the road categories for
rural and urban roads, their location in the road network and their defining characteristics
of trip length, design volume and speed are provided in Table 1.2.1.
International National
Expressway National
RURAL
Access control is the condition where the right of owners or occupants of abutting land or
other persons to access, in connection with a road is fully or partially controlled by the
public authority.
Control of access is usually classified into three types for its degree of control, namely full
control, partial control and non-control of access.
Full Control of Access means that preference is given to through traffic by providing access
connecting with selected public roads only and by prohibiting crossings at grade or direct
private driveway connections. Partial Control of Access means that preference is given to
through traffic to a degree that in addition to access connection with selected public roads,
there may be some crossings trafficked roads, at grade intersections should be limited and
only allowed at selected locations.
To compensate for the limited access to roads with full or partial control of access, frontage
or service roads are sometimes attached to the sides of the main roads. In Non Control
Access there are basically no access limitations.
1.2.4.2 Application
The selection of the degree of access control required is important to preserve the as built
capacity of the road as well as improved safety to all road users. Two aspects pertaining
to the degree of access are:
(c) the selection of degree of access control depends on traffic volumes, function of the
road and the road network around the areas. Guidelines for selecting the degree of
access control are provided in Table 1.2.2 and Table 1.2.3.
Highway - R5 R4 - - -
Provincial - - R4 R3 - -
District - - - - R2 R1
Expressway U6 - - - - -
URBAN
Arterial - U5 U4 - - -
Collector - - U4 U3 - -
Local - - - - U2 U1
1.2.6 UTILITIES
Guidelines for the location of utilities within road reserves relative to property boundaries
are provided in Table 1.2.5.
1.3.1 TOPOGRAPHY
For design purposes topography is classified into flat, rolling or mountainous using the
following definitions:
FLAT terrain:
The topographical condition where highway sight distances, as governed by both
horizontal and vertical restrictions are generally long or could be made to be so without
construction difficulty or expense. The natural ground, cross slopes (ie. perpendicular to
natural ground contours) in a flat terrain are generally below 3 %
ROLLING terrain:
The topographical condition where the natural slopes consistently rise above and fall below
the road or street grade and where occasional steep slopes offer some restrictions to
normal horizontal and vertical roadway alignment. The natural ground cross slopes in
rolling terrain is generally between 3 – 25 %.
MOUNTAINOUS terrain:
The topographical condition where longitudinal and transverse changes in the elevation of
the ground with respect to the road or street are abrupt and where benching and side hill
excavation are frequently required to obtain acceptable horizontal and vertical alignment.
The naturals ground cross slopes in mountainous terrain are generally above 25 %.
The physical characteristics of vehicles and the proportions of various size vehicles using
the roads affect the geometric design of roads. The design vehicle to be used for geometric
design follows that used by AASHTO as in Chapter II of AASHTO “Design Vehicles “- A
Policy of Geometric Design of Highways and Stress (1994) with SI units . Information on
standard design vehicles are provided in Table 1.3.1, and Figure 1.3.1, Figure 1.3.2 and
Figure 1.3.3.
1.3.3 SPEED
Speed is a primary factor in all modes of transportation, and is an important factor in the
geometric design of’ roads. The speed of vehicles on a road depends, in addition to
capabilities of the drivers and their vehicles, upon general conditions such as the physical
characteristics of’ the highway, the weather, the presence of other vehicles and the legal
speed limitations.
The Speeds are selected to meet the needs of the road to fulfil its function. Thus roads
which are planned to provide long distance travel will be designed with a higher speed
while those which provides short distance travel can be given a lower design speed.
Operating speed is the highest overall speed at which a driver can travel on a given road
under favourable weather conditions and under prevailing traffic conditions without at any
time exceeding the design speed on a section by section basis.
Design speed is the maximum safe speed that can be maintained over a specified section
of the road when conditions are so favourable that the design features of the road governs.
The assumed design speed should be a logical one with respect to the topography, the
adjacent land use and the type of road. Every effort should be made to use as high a
design speed as practicable while maintaining the desired degree of safety, mobility and
efficiency. Table 1.3.2 and Table 1.3.3 indicate the selection of design speeds for rural and
urban roads respectively.
Note : Type I – Relatively free in road location with very little problem as regards land
acquisition, affected building or other socially sensitive areas.
Type II – Intermediate between I and III
Type III – Very restrictive in road location with problem as regards land acquisition,
affected building and other sensitive areas.
In the design of a substantial length of road, it is desirable, although not always feasible to
allow for a constant design speed. Changes in terrain and other physical controls may
dictate a change in the design speed on certain-sections. If a different design speed is
introduced, the change should not be abrupt, but there should be a transition section of at
least 1 km to permit drivers to change speed gradually before reaching the section of’ the
road with a different design speed. The transition sections are sections with intermediate
design speeds, and where the magnitude of the change in the design speeds are large,
more than one transition section will be required.
1.3.4 CAPACITY
The design of a road should be based on traffic data that defines the “loads” for geometric
design. Traffic data for a road or section of road should be based on recent traffic and axle
load surveys performed in accordance with procedures provided in CAM PW 03-102-99
Road Design Standard Part 2- Pavement.
ADT represents the total traffic for the year divided by 365, or the average volume per day.
Knowledge of the ADT is important for many purposes, such as determining annual usage
as justification for proposed expenditures or for design of structural elements of a road.
The projected ADT is also used to designate the standard of road as shown in Table 1.2.4,
Design Standards. However, the direct use of ADT in geometric design is not appropriate
because it does not indicate the significant variation in the traffic occurring during various
months of the year, days of the week and hours of the day. A more appropriate
measurement is by hourly volume that is used to determine the capacity requirement of the
road.
The traffic pattern on any road shows considerable variation in traffic volumes during the
different hours of the day and in hourly volumes throughout the year. It is difficult to
determine which of these hourly traffic volumes should be used for design. It would be
wasteful to base the design on the maximum peak hour traffic of the year, yet the use of
the average hourly traffic would result in an inadequate design.
To determine the hourly traffic best fitted for design, a curve showing the variation in hourly
traffic volumes, during the year is used. In the absence of the traffic survey data, the
hourly traffic used in design is the 30th highest hourly volume of the year, abbreviated as
30 HV. The design hourly volume, abbreviated DHV is the 30 HV of the future year chosen
for design.
The above criterion is applicable to most rural and urban roads. However, for roads on
which there is unusual or highly seasonal fluctuation in traffic flow such as holiday resort
roads, the 30 HV may not be applicable. It may be desirable, to choose an hourly volume
for design (about 50 percent of the hourly volumes) expected to occur during a very few
maximum hours (15 to 20) of the design year whether or not it is equal to 30 HV.
K is the ratio of DHV to the designed ADT. K’s value ranges from 7 % to 20 % and the
actual value should be obtained from traffic data. In the absence of traffic data k = 12 %
for urban roads and k = 15 % for rural roads should be used. For roads with highly distinct
fluctuations of traffic, whether seasonal, daily or hourly, it is recommended that traffic
surveys be carried out as the above k values may be unrealistic.
For 2-lane roads, the DHV is the total traffic in both directions of travel. On roads with
more, than two lanes, and on 2-lane roads where important intersections are encountered
or where additional lanes are to be provided later, knowledge of the hourly traffic load in
each direction of travel is essential for design.
The directional distribution of traffic during the design hour should be determined by field
measurements on the facilities. Generally in the absence of field data D value of 60 % can
be used in urban areas and 65 % in rural areas. Traffic distribution by directions is
generally consistent from year to year and from day to day except on some roads serving
holiday resort area.
Capacity is also usually stated in terms of passenger car units (p.c.u). Table 1.3.4 gives the
conversion factor to be used in converting the various classes of vehicles to passenger car
units.
Under ideal conditions, the possible capacity for uninterrupted flow are as follows:
(b) For multi-lane (per lane) = 2,800 pcu/hr. Ideal conditions consist of the following for
two-lane roads:
(i) No impediments to through traffic due to traffic control devices or turning vehicles
When one or more of these conditions are not met, the actual capacity will be reduced.
For additional information on the impact of these factors on volume capacity refer to
Highway Capacity Manual, Special Report No 209, Transportation Research Board,
Washington D.C., 1985.
Design volume is the volume of traffic estimated to use the road during the design year,
which is taken as 20 years after the completion of the road. The derivation of the design
hour volume (DHV) is as discussed in Section 1.3.4.2.
The service volume is the maximum volume of traffic that a designed road would be able to
serve without the degree of congestion falling below a preselected level. The level of
service defines the allowable level of congestion when traffic flow equals the design hourly
volume for each category of road. Levels of service for all road categories are provided in
Table 1.3.5.
Level of Remarks
Service
A Free Flow with low volumes, densities and high speeds.
Drivers can maintain their desired speeds with little or no delay.
B Stable Flow. Operating speeds beginning to be restricted some
what by traffic conditions. Some slight delay.
C Stable Flow. Speeds and manoeuvrability are more closely
controlled by higher volumes . Acceptable delay.
D Approaching Unstable Flow. Tolerable operating speeds which
are considerably affected by operating conditions. Tolerable
delay.
E Unstable Flow. Yet lower operating speeds and perhaps
stoppages of momentary duration. Volume is at or near capacity
congestion and intolerable delay.
F Forced Flow. Speeds and volume can drop to zero. Stoppages
can occur for long periods. Queues of vehicles backing up from, a
restriction downstream.
Volume capacity ratio is calculated by dividing the service volume for a given level of
service by the theoretical capacity of the road. Service volume is defined in Section 1.3.4.8.
Based on the design level of service volume capacity ratios can be provided for each
category of road. Design level service and related volume capacity ratios for rural and
urban roads are provided Table 1.3.6 and Table 1.3.7 respectively.
Table 1.3.6: Design Level of Service and Volume / Capacity Ratio (Rural )
Table 1.3.7: Design Level of Service and Volume / Capacity Ratio (Urban )
1.4.1 GENERAL
The horizontal alignment of a road is usually a series of straights (tangents) and circular
curves connected by transition curves. An alignment essentially without straight sections is
described as being curvilinear. A curvilinear alignment may consist of long, large radius
circular curves, with or without spiral transitions, or may consist of curves of variable
curvature conforming to a polynomial mathematical relationship.
Curvilinear alignment is suitable for divided roads. On two-lane roads in undulating country
it may be appropriate particularly if overtaking provision is not impaired. In flat inland areas,
it has also been used successfully. The curves, in keeping with the terrain are large and
hence do not restrict overtaking but do produce advantages of reduced headlight glare and
better perception of the speed of approach of opposing vehicles.
2
e+ f =V or V2 (1 - 1.)
gR 127 R
where:
e = pavement super elevation (m/m or tangent of angle). This is taken as positive if
the pavement falls towards the centre of the curve.
f = coefficient of side frictional force developed between vehicle types and road
pavement. This is taken as positive if the frictional force on the vehicle acts
towards the centre of the curve.
g = acceleration due to gravity = 9.8 m/s2
V = speed of vehicle (km/h)
R = curve radius (m)
Where f equals zero in the formula, the whole of the condition can occur on large radius
curves with positive superelevation or for slow moving vehicles on curves of any radius. At
low speeds, f can be negative, and the curve is then over-super elevated for that speed.
Curves are generally designed, however, so that a positive f is required for the range of
vehicle speeds likely to occur.
1.4.3 SUPERELEVATION
The super elevation to be adopted is chosen primarily on the basis of safety, but other
factors are comfort and appearance. The super elevation applied to a road should take
into account:
(a) design speed of the curve;
(b) tendency of very slow moving vehicles to track towards the centre;
(c) stability of high laden commercial vehicles difference between inner and outer
formation levels, especially in flat country;
(d) length available to introduce the necessary super elevation.
Use of maximum super elevation will need to be considered where the radius of a curve is
approaching the minimum for the design speed; ie towards the minimum radius provided in
Table 1.4.2. Normally, this will occur in steep terrain or where there are constraints on
increasing the radius of an individual curve in a group. In such cases, if super elevation
approaching the maximum is not applied, there may be a sharp increase in friction demand
by a vehicle travelling at the design speed as it enters the curve in question. While it can
be shown that the friction value may be within the normal range drivers use, a sudden
change in friction demand will lead to a sudden change in steering attitude of a vehicle
negotiating the road. Large fluctuations in friction demand are best avoided, as a driver
may not anticipate a sudden change in the response of his vehicle. The maximum super
elevation ranges from 0.10 in mountainous terrain to 0.06 to 0.07 in flat terrain.
At low and intermediate ranges of design speeds (below about 100 km/h), it will usually be
found desirable to super elevate all curves. The minimum superelevation should be at
least equal to the normal cross fall on straights, unless the radius of an individual curve is
so great that it can be regarded as a straight for the particular speed environment and
normal cross falls (or adverse super elevation) used.
The circular path formula given in Section 1.4.2 assumes that the path of a vehicle
traversing a curve assumes a constant radius equal to the curve radius. However, this
does not occur in practice. On short, small-radius curves, drivers tend to use the available
lane width so that vehicle path radius is increased and the actual f is below the value
implied by the circular path formula.
For longer, large-radius curves, however, drivers must make additional steering
adjustments to maintain the position of the vehicle within the lane. These tend to decrease
the radius of the vehicle path so that the actual f is greater than the value implied by the
formula.
The differences between the conditions assumed for the circular path formula and what
actually occurs in practice mean that f values appropriate for use in the formula cannot be
derived directly from known pavement skid resistance. However, drivers learn to assess
the speed appropriate to a given curve, and this can form the basis of a design criterion.
From measurements of this speed and curve geometry, the circular path formula can be
used to compute the equivalent f value that the driver regards as appropriate.
It should be noted that there are practical limits to the speed of vehicles around curves on
normal roads. Tests show that most cars can develop instability when the value of f
exceeds 0.6 g, where g is the acceleration due to gravity. Vehicles with a high centre of
gravity can develop roll instability at cornering rates as low as 0.28 g. Observations of
driver behaviour show that at speeds of 80 km/h, cornering rates above those considered
“comfortable” are regularly adopted, especially under dry conditions. At speeds less than
80 km/h even higher values of cornering rates are common, whereas at high speeds
(100 km/hr or more) the cornering rate observed is modest.
The f values given in Table 1.4.1 which apply to sealed and unsealed pavements, have
been derived from observations of driver speed behaviour on rural road curves, and are
appropriate for use with the circular path formula. The f values shown for speeds less than
90 km/h are those typically used by vehicles travelling at the design speed on the sharper
curves. The values shown for higher speeds are in excess of those likely to be required by
vehicles travelling at the design speed, in keeping with the concept that high speed
alignments should provide a high degree of comfort and safety for all road users (McLean
1983).
The values given in Table 1.4.1 are an upper limit and the f demand of a vehicle at the
design speed should not exceed them. Normal design will select an appropriate superele-
vation and the f demand can be checked from:
Vd2
fd = −e (1- 2.)
127 Ra
where:
fd = f demand of a vehicle at the design speed
Vd = design speed (km/h)
Ra = actual radius of curve (m)
e = proposed superelevation (m/m)
For normally accepted superelevation rates, f, will be somewhat below the values in Table
1.4.1 unless there is a particular curve where it is very difficult or costly to increase the
radius. In such cases, large increases in f demand between successive curves are best
avoided even if this means the use of superelevation towards the maximum.
It is important to note that the values for f given in Table 1.4.1 assume construction and
maintenance techniques which will ensure an adequate factor of safety against skidding.
The susceptibility of the wearing surface to polishing, the macro-texture of the surface and
the amount of bitumen used in the wearing surface are all important matters in the initial
construction of a pavement which contribute to skid resistance. Freedom from contamina-
tion by oil spillage or loose aggregate, and resealing when surface texture becomes too
smooth, are important aspects in maintenance of skid resistance. Normally, a pavement
that is properly maintained will retain adequate resistance to skidding under all but extreme
conditions of driver behaviour or weather.
The minimum radius of a horizontal curve for a given vehicle speed can be determined
from the formula given in Section 1.4.2:
V2
Rmin = (1 - 3.)
127 (emax + f min )
where:
Rmin is the minimum radius (m)
V is the vehicle speed (km/h)
emax. and fmax are the maximum values of superelevation and side friction respectively.
Using the values for fmax from Table 1.4.1, the approximate minimum radii for various
vehicle speeds for typical maximum superelevations are as shown in Table 1.4.2. When
both superelevation and the side friction values are at the maximum the vehicle speed is
equal to the limiting curve speed standard.
In practice it is usual to adopt radii greater than those in Table 1.4.2 and Table 1.4.3 and to
reduce superelevation and side friction below their maximum values.
A curve inserted to transition from the straight or tangent section of a road to a circular are,
or between adjacent curves, is known as a plan transition curve. Such a curve is used for
the following reasons:
(a) to provide a length over which superelevation and/or widening is applied.
(b) to facilitate positioning of the vehicle within narrower lane widths, particularly in lower
speed environments on lower radii curves.
(c) in some cases, appearance is improved particularly on a bridge where a rigid handrail
follows the exact geometry of the lane.
(d) appearance can also be improved on large-radius curves that are visible at the end of
a long straight.
(e) provides a length over which steering adjustment can be made (particularly between
reverse curves).
(f) provides a length over which speed adjustments may be made between curves of
different radii.
A wide range of curve forms (including even a large-radius circular curve) can be suitable
for a plan transition. However, a curve frequently used is the Clothoid where curvature
(reciprocal of radius) changes at a uniform rate along the curve. This would be the correct
form if vehicles travelled at uniform speed along the exact centre line of the roadway.
The plan transition length should be sufficient so that as superelevation is developed, the
rate of rotation of the pavement (see Section 1.4.8.1) should not exceed 0.025 m/m/sec
(angular rotation 0.025 rad/sec), of travel time at the design speed. In more difficult
situations, usually in the lower speed environment sections (below 80 km/h), this maybe
increased to 0.035 m/m/sec. The plan transition, if provided, will occupy that portion of the
superelevation development where the outer traffic lane goes from level to the full
superelevation (see Table 1.4.4 and Figure 1.4.1).
On wider pavements, the relative grade of the two pavement edges should be kept within
reasonable limits for the sake of appearance, this becomes a control on plan transition
length in these cases. Values using these criteria are discussed in Section 1.4.7 and 1.4.8.
Plan transitions involve a shift between the extended tangent line and the extended circular
arc. The value of the shift can be taken as L2/ 24R where L is the length of plan transition
curve, and R the radius of the circular arc. (This expression is correct for the Clothoid
where curvature increases linearly. It is an approximate shift value for other forms of plan
transition).
CL
CL
Where the shift is less than approximately 250 mm, the contribution by the transition to
positioning of vehicles or to appearance is negligible and it may be omitted.
Where the curve is to be widened (see Section 1.4.11), and the shift is not greater than
approximately half of the widening, a plan transition may also be omitted provided all the
widening is applied on the inside of the curve in the position normally occupied by the
transition.
Where a transition is desired for appearance purposes the length should be sufficient to
require a shift of 0.25 m to 0.5 m. In high speed environments in predominantly flat country
where curves are visible from long distances the appearance of the road can be improved
by increasing the length of the plan transition if necessary to produce a shift of about
0.5 m.
In lower speed environments, care should be taken when considering plan transition
lengths that are greater than those that are needed for the development of superelevation.
Drivers regulate their desired speed from the apparent curvature of the road ahead and, in
the low and medium speed environments, this results in some variation of curve approach
speeds. In these environments, therefore, longer plan transition lengths introduced merely
for appearance may affect the drivers’ perception of road curvature, and are probably best
avoided.
On straights, the pavement has normal cross fall to shed water. As discussed in Section
1.6.2, this cross fall is provided both ways from the centre on undivided rural roads. Each
carriage way on a divided rural road usually has one-way crossfall away from the median
on straight alignments.
A change from normal cross fall to superelevation occurs as the road changes from a
straight to a curved alignment, (except where adverse cross fall is adopted), or from a very
large curve with adverse cross fall to a curve with a lower radius.
The adopted position of the axis of rotation, ie the point about which the cross fall is rotated
to develop super elevation depends upon the type of road facility, total road cross section
adopted, terrain and the location of the road. On a two-lane, two-way road, the
superelevation is normally developed by rotating each half of the cross section (including
shoulders) about the carriageway centre line (axis of rotation). The application of
superelevation often results in the profiles at the edges of the carriage ways being in
opposite directions to each other, or to the resulting grade lines creating flat sections of
pavement (See Figure 1.4.1). Where drainage or appearance problems are thereby
created, the grading along the controlling edges of the carriageway and/or traffic lanes
must be determined and the relative grade or the rate adjusted accordingly.
On divided rural roads, where the median is relatively narrow, ie. less than 5m, the two
carriageways may be rotated about the centreline of the median. Where the median is
wide, the axis of rotation may also be the centreline of the median (provided that excess
cuts and fills do not result) or along the median edge of carriageway. In flat country, it may
be desirable to adopt the left (or outer) edge of carriageway as the axis of rotation or
introduce independently graded carriageways. Where the pavement is to be widened into
the median at a later stage, care must be exercised when placing the axis of rotation at the
first stage median shoulder or carriageway edge. Ultimate superelevated cross sections
must be checked to avoid steep median crossfalls or an unsightly saw tooth appearance.
Where carriageways are wide and it is necessary to reduce the depth of water on roadway
surfaces during the development of superelevation, either the rate of rotation must be
increased to values consistent with appearance, individual grade lines established for
shoulder and traffic lanes or the width of the contributing pavement reduced.
If the rate of change in level is abrupt, it may be unsightly. The change can be particularly
noticeable where a kerb, guard fence or bridge barrier defines the outer shoulder edge.
The rate of rotation of the pavement should not generally exceed 0.025 radians per second
of travel time at the design speed with a maximum rate of 0.035 (0.025 radians per second
rate of angular rotation is equivalent to a rate of change of superelevation of
0.025 m/m/sec or 2.5 % per second). As with other comfort and appearance criteria in this
publication, these and the following values should be regarded as reasonable values and
not inherently correct.
The relative grade is the difference between the grade of the edges of the carriageway and
the grade of the axis of rotation. This should be kept to acceptable limits to ensure a
smooth visual appearance.
When a change of grade occurs in the road in the vicinity of a superelevation development
length, the grade of the edge of the pavement or formation may be irregular. To ensure
that the edge presents a smooth appearance, the edge levels may be plotted to an
exaggerated scale and the profile smoothed by adjusting levels and superelevation
development length.
As the crossfall changes to superelevation on flat grades, a concavity may occur in the
grade line of the edge of the formation. Special drainage may be needed at the low spot
thus created, or a longer development length used to maintain a fall in the gutter line.
A vertical curve may be used to ease the grade changes from crossfall to superelevation at
the edges of the pavement and formation.
The minimum superelevation development length to satisfy the rate of rotation criterion can
be determined from the following expressions:
(e1 − e2 )Vd
Le = for a rate of 0.025 rad/sec
0.09
(1 - 4.)
(e1 − e2 )Vd
Le = for a rate of 0.035 rad/sec
0.126
Where:
Le = superelevation development length (m)
e1, e2 = crossfall or superelevation at ends of development length (m/m)
Vd = design speed (km/h)
Table 1.4.4 lists values (rounded to 5 m) satisfying these criteria (0.035 radians per second
for design speeds to 70 km/h and 0.025 radians per second for 80 km/h and above).
If the difference in grade between the edge of the carriageway and the axis of rotation are
kept below the values shown in Table1.4.5, a reasonably smooth visual appearance will
result. Once again, being subjective values, these should not be regarded as precise
criteria.
The expressions relating to rate of rotation and relative grade are as follows:
Where:
W = width from axis of rotation to edge of running lane (m)
Vd = design speed (km/h)
For a rate of rotation of 0.025 radians per second, the rate of rotation criterion results in
longer superelevation development length than relative grade criterion for widths of 3.5 m
and 7.0 m, ie. up to two lanes. Hence, for design speeds above 80 km/h, the lengths
given in Table1.4.5 satisfy both criteria in most cases.
The relative grade criterion will result in longer superelevation development lengths than
the rate of rotation criterion when the rate of rotation criterion is 0.035 radians per second
and the width from axis of rotation to edge of running lane comprises three or more lanes.
W (e1 − e2 )
Le = × 100 (1 - 5.)
Gr
Where:
Le = superelevation development length (m)
W = maximum width from axis of rotation to edge of running lane (m)
e1, e2 = pavement crossfall or superelevation at ends of the development length
(m/m)
Gr = allowable relative grade from Table 1.4.5.
may be continued to the curve of higher superelevation provided that the crossfall is
not excessive and the development point is positioned about the tangent point. For
longer distances between tangent points, the crossfall may be returned to normal or
carried through on one-way crossfall equal to the normal crossfall.
For the remainder of the transition the outer lane crossfall changes uniformly to the
maximum value of e required. The inner lane crossfall remains at -n until the outer lane
reaches a value of +n, from which point it continues to change at the same rate as the
outer lane.
The starting point for the superelevation development can either be established by
projecting the rate of change of the outer lane back from the tangent to spiral point (TS, in
Figure 1.4.1) until it reaches superelevation value of -n, or it can be fixed at a value
approximately 0.4 V, in advance of tangent to spiral point.
The length of arc of a curve is a function of the radius of the curve and the deflection angle.
At a small change in alignment the radius must be considerable if an arc of the minimum
length required by Table 1.4.6 is to be achieved. It is preferable (and usually more
economical) to adopt a circular curve of large radius with adverse crossfall rather than a
short curve with superelevation. The deflection angle below which such a treatment is
preferable is shown in column (a) of Table 1.4.6. The length of arc and the minimum radius
which should then be adopted are shown in column (b) of Table 1.4.6 and in Table 1.4.8
respectively.
At really small deflection angles the change in alignment is not apparent and the curve may
be omitted. The minimum angles at which curve are needed are shown in column (C) of
Table 1.4.6.
Table 1.4.6 Critical Factors for Horizontal Curve with Small Deflection Angles
Pavements may be widened on curves to maintain the lateral clearance between vehicles
equal to the clearance available on straight sections of road. Widening is required for two
reasons:
1. A vehicle travelling on a curve occupies a greater width of pavement than it does on a
straight as the rear wheels at low speeds track inside the front, and the front overhang
reduces the clearance between passing and overtaking vehicles. (At high speeds the
rear wheels track outside the front.).
2. Vehicles deviate more from the centreline of a lane on a curve than on a straight.
Other factors such as overhang of the front of the vehicle, wheel base and track width play
a part.
Approximate values for the amount of widening required by a single unit design truck on a
road pavement on a circular arc as shown in Table 1.4.7. Half the widening is applied on
each side of the centreline of curves with plan transitions. The whole widening may be
applied on the inside of plain circular curves, in which case the same effect as a plan
transition is achieved. For curves with a radius of 30 m or less, the actual paths of
articulated vehicles should be found from templates or vehicle path computer programs.
For roads wider than two lanes pavement widening is done by adding half the widening for
two-lane roads to each lane.
The transition of curve widening is uniform and coincides with the plan transition, if any.
Where there is no plan transition, about half the widening may be applied before the
tangent point and terminated at the end of the superelevation development length on the
arc.
Table 1.4.7 Recommended Values for Curve Widening for Two-lane Pavements
Curve Radius Total Amount of Widening (m)
(m) Normal width of two traffic lanes
6m 6.5m 7m 7.5m
30-50 2.0 1.5 1.5 1.0
50-100 1.5 1.0 1.0 0.5
100-250 1.0 1.0 0.5
250-750 1.0 0.5
Over 750 0.5
Adverse crossfall on curves should normally be avoided except on curves of large radius
which, for the speed environment of the section, can be regarded as straights (see Section
1.4.3.2). Such curves need not be superelevated, but the normal minimum crossfall used
on straights may be applied.
The radii in Table 1.4.8 can be regarded as approximate limiting values for adverse
crossfall equal to the normal straight crossfall value. It will be found that the (e +f) value
(see Section 1.4.2) of these curves all lie between 0.03 and 0.04 so that, with 0.03 adverse
crossfall the f demand will he between 0.06 and 0.07. This provides additional factors of
safety over the values given in Table 1.4.1.
70 900
80 1250
90 1500
100 2000
110 3000
120 4000
130 5000
Curves with radii near the minimum for the design speed (Table 1.4.2 and Table 1.4.3) do
not necessarily meet the sight distance requirements prescribed in Section 1.5. Where a
lateral obstruction off the pavement such as a bridge pier, cut slope or natural growth
restricts sight distance, the stopping sight distance appropriate to the design speed of the
curve determines the minimum desirable radius of curvature.
Figure 1.4.2 shows the relationship between horizontal sight distance, curve radius and
lateral clearance to the obstruction and is valid when the sight distance at the appropriate
design speed is not greater than the length of curve. This relationship assumes that the
driver’s eye and the sighted object are above the centre of the inside lane. When the
design sight distance is greater than the length of curve, a graphical solution is appropriate.
For alignments with lower design speeds, particularly in difficult terrain, it may not be
feasible to achieve the 2.0 and 2.5 seconds reaction times stopping sight distances
recommended in Section 1.5. In these situations, the designer should provide the
maximum sight distance practicable, and ensure that it is not less than the stopping sight
distance corresponding to a 1.5 second reaction time.
Where sight benches in side cuttings are required on horizontal curves, or a combination of
horizontal and vertical curves, the horizontal and vertical limits of the benching are
determined graphically.
Obs t r uc t i on
s
diu
28.65 S
O = R [ VERS ( )]
Ra
R
R [ - 1 R-O
S = COS ( )]
28.65 R
15
10
40
20
60
80
100
400
0
200
300
600
800
0
0
150
200
100
6
0
300
5
0
400
3
30 50 70 100 200 300 400 500
1.5.1 GENERAL
A principal aim in road design is to ensure that the driver is able to perceive any possible
road hazards in sufficient time to take action to avoid mishap. To provide a calculable
parameter which can be related to the geometry of the road, the concept of Sight Distance
is used. This concept is based on a number of somewhat stylised assumptions of particular
hazards and corresponding driver behaviour. The hazard is assumed to be an object, of
sufficient size to cause a driver to take evasive action, intruding into the driver’s field of
view (see Figure 1.5.1). Specific values are assumed for the driver’s reaction time (though
in practice there would be a distribution of values) and the dimensions determining the
geometry of the sight line. Normally, selection of extreme values for every parameter is not
appropriate, as the probability of all factors occurring together is extremely low, and the
resultant designs would become impractical. The assumed parameter values lead to sight
distances that produce a satisfactory design. Greater distances allow for less probable
hazard situations and thus produce greater margins of safety. Subject to their effect on
overtaking (see Section 1.5.6) and economics they may be an advantage.
An object in view may not always be perceived. There is evidence that when a driver is
travelling on sharp curves or when the vehicle is rapidly accelerating or decelerating and
the driver is subject to unusual forces, his ability to perceive an object is reduced. Fatigue
and drugs add to the time of perception and may increase an individual’s reaction time.
Peripheral vision declines as forward speed increases and there is some difficulty in
perceiving small hazards, eg. potholes, in time to take appropriate action when speeds
exceed 80 km/h to 100 km/h.
The values for determining the geometry of the sight line and for driver reaction time
adopted in this publication are as follows:
The height of a driver’s eye above the road when he is seated in a vehicle varies with the
type and condition of the vehicle, the build of the driver and his driving position. The height
of eye for drivers of heavy vehicles is above that for car drivers and 1.8 m may be adopted
for design purposes.
The object cut off height to be used in the calculation of stopping sight distance is a
compromise. Stopping is generally the result of the presence of another vehicle stopped
on the roadway, and visibility of the vehicle down to and including the rear lights (height
600 mm) would be necessary. The only case where the upper limits of eye height and
object cutoff height become relevant is the combination of a sag and an overhead
obstruction such as a bridge and 1.8 m and 0.6 m respectively are used in this case (see
Figure 1.5.1).
To perceive a very small hazard, eg a surface obstruction, a zero object cut off height
would be necessary. However, at high speeds, small pavement variations may not be
clearly visible. Most drivers travelling at 100 km/h find difficulty in avoiding and stopping
before such a small obstruction. At intersections where it is necessary to see road
markings, a zero height may be appropriate.
The length of vertical curve required at crests increases fairly significantly as the object cut
off value approaches zero and so the general figure adopted which produces satisfactory
designs is 200 mm. However, zero can be used in the case of intersections, or of
approaches to causeways and floodways where sand left by floodway water, or even
washouts, may occur.
In the case of small objects on the roadway, it is more likely that a driver will attempt to
take evasive action rather than to stop. The time available to manoeuvre, while not
recommended as a prime design parameter, is a useful measure when examining
variations of geometry in restricted situations.
1.5.5.1 Derivation
2
SD = RT V + V for V in m/s
2gd
(1 - 6.)
SD = RT V
2
+ V for V in km/h
3.6 254 d
where:
SD = stopping distance (m)
d = coefficient of longitudinal deceleration
g = acceleration due to gravity (9.8 m/s2)
RT = driver reaction time (sec)
V = initial speed of vehicle in the units shown.
Reaction time is defined as the time between the instant a hazard comes into view and the
instant the vehicle begins to slow after the brakes have been applied. There is no
conclusive evidence to support any particular reaction time value, but it is generally agreed
that values of 2.0 and 2.5 seconds are both reasonable and appropriate, and they are the
values which should be adopted as a general rule.
For low speed alignments in difficult terrain, it may not be practicable to design for, a 2.0
seconds reaction time, and lower values may have to be adopted. The minimum approach
should be to provide the best sight distance possible, and to check that it corresponds to a
reaction time of not less than 1.5 seconds. The lower reaction time values are acceptable
on consistently light alignments that will lead to an increase in driver alertness. Values
below 2.5 seconds should not be used for isolated alignment features where a driver is
unlikely to be alerted.
On long straight roads with low traffic volumes where sight distances are long and speeds
high, reaction times may be well in excess of 2.5 seconds. In such cases of unconstrained
sight distance, the stopping sight distance calculation involving reaction time is largely
irrelevant.
Recent tests at ARRB have found that, on good, dry pavements modern passenger cars
can consistently achieve deceleration rates in excess of 1.0 g (Samuels & Jarvis 1978).
However, the values used for design purposes should allow for the degradation of
pavement skid resistance when wet; and for a reasonable amount of surface polishing.
The coefficients of longitudinal deceleration values assumed are given in Table 1.5.1. The
lower values assumed for the higher speeds reflect the reduction in wet pavement skid
resistance with increasing speed and the need for lateral vehicle control over the longer
braking distances.
The assumed values of reaction time and coefficient of longitudinal deceleration enable a
notional stopping distance to be calculated from a given initial speed. If the design speed
is used as the initial speed, the stopping distance would be covered by a vehicle travelling
at the design speed for the road section whose driver elected to stop after perceiving a
hazard on the road ahead. This distance is considered to be the minimum sight distance
that should be available to a driver. The values calculated from the assumed RT and d
values, and the design speed are listed in Table 1.5.2.
It is not possible to specify design values of RT and d that are absolutely correct, or which
are likely to be correct for all situations. For design purposes, a minimum stopping sight
distance is used which is considered reasonable in terms of the limited empirical evidence
available. When combined with the assumed line of sight parameter values (see Section
1.5.4), these values lead to design standards that have been found to be satisfactory over
several decades of operational experience in a number of countries.
Table 1.5.2 lists the normal minimum values and lower limits for restricted situations using
shorter reaction time.
1.5.6.1 General
Most roads are two-lane two way on which vehicles frequently overtake slower moving
vehicles, the passing of which must be accomplished on a lane regularly used by the
opposing traffic. Passing sight distance for use in design should be determined on the
basis of the length needed to safely complete a normal passing manoeuvre.
The minimum passing sight distance for two-lane highways is determined as the sum of
four distances:-
a) distance traversed during the perception and reaction time and during the initial
acceleration to the point of encroachment on the passing lane.
b) distance travelled while the passing vehicle occupies the passing lane.
c) distance between the passing vehicle at the end of its manoeuvre and the opposing
vehicles.
d) distance traversed by an opposing vehicle for two-thirds of the time the passing
vehicle occupies the passing lane.
The total passing sight distance is determined by the sum of the above four elements.
Table 1.5.3 gives the minimum values to be used for each design speed.
In designing a road, these distances should be exceeded as much as practicable and such
sections provided as often as can be done with reasonable costs so as to present as many
passing opportunities as possible exceeded.
Specific adjustment for design use is not available. The effect of grade is not considered in
design as the effect is compensated either in upgrade or downgrade. However it should be
realised that greater distances are needed for passing on grade as compared to level
conditions. The designer should recognise the desirability of increasing the minimum as
shown in Table 1.5.3.
Sight distance adequate for passing should be provided frequently on 2-lane roads.
Designs with only minimum sight distance will not assure that safe passing can always be
made. Even on low volume roads a driver desiring to pass may, upon reaching the section,
find vehicles in the opposing lane and thus be unable to utilise the section or at least not be
able to begin to pass at once.
The percentage of length of road or section of road with sight distance greater than the
passing minimum should be computed. The adequacy of sight distance is determined by
analysis of capacity related to this percentage, and this would indicate whether or not
alignment and profile adjustments are necessary to accommodate the design hour
volumes. Generally this percentage should be about 60 % for flat terrain, 40 % for rolling
terrain and 20 % for mountainous terrain. The available passing sight distances should not
be concentrated in one section but be evenly distributed throughout the entire road. It is
important to note that in order to cater for high volume of traffic at high level of service, it
requires that frequent and nearly continuous passing sight distances be provided.
The elements of cross sections are shown in Figure 1.6.1. The width and slope of the
traffic lanes and shoulders are based on traffic needs and drainage requirements. The
form of the remainder of the cross section, namely the batter slopes of embankments and
cuttings, depends on the type of material to be excavated, environmental factors and the
importance of the road.
The widths and slopes of the various cross section elements may be varied within
acceptable limits to achieve a balanced, economical, functional and aesthetic result. This
section gives details of acceptable widths and slopes of the elements, together with
guidelines for selection of the appropriate values.
1.6.2 CROSSFALLS
1.6.2.1 General
Crossfall is the slope of the surface of a carriageway measured normal to the centre line.
The purpose of crossfall is to drain the carriageway on straights and curves and to provide
superelevation on horizontal curves (refer to Section 1.4).
Carriageway crossfalls on straights should be as flat as drainage needs permit, and these
in turn are conditioned by the type of pavement and nature of the surface. For a given
slope, the smoother the surface: the more efficient it is in shedding water, but this must not
override friction requirements for safety reasons.
On two-way carriageways, the traffic lane(s) is cambered to form an inverted ‘V’ that may
be rounded at its highest point, the crown. Recommended pavement crossfall on straights
for various pavement types are given in Table 1.6.1
On straights, the crossfall of road shoulders (even where they are sealed) can be up to 2 %
steeper than the crossfall of the traffic lanes they flank. Satisfactory shoulder crossfalls on
straights are given in Table 1.6.2. On curves with super-elevated pavements, both shoul-
ders can be super-elevated to crossfalls not less than the pavement crossfall.
Lane widths and the condition of the pavement surface are the most important features of
a road pertaining to the safety and comfort of driving. The capacity of a highway is
markedly affected by the lane width and in a capacity sense, the effective width of a
travelled way is further reduced when adjacent obstructions such as retaining walls, bridge
piers and parked cars restrict the lateral clearance.
The marginal strip is a narrow pavement strip attached to both edges of a carriageway. It
is paved to the same standard as the pavement structures. For divided roads, the marginal
strips are provided on both sides of the carriageway in both directions. The marginal strip
is included as part of the shoulder width and is demarcated from the through lane by lane
edge markings on the marginal strip.
Table 1.6.3 indicates the lane and marginal strip widths that are to be used for the various
road standards.
On 6-lane divided highways, the middle lane of the three lanes on each direction, should,
have a wider lane of 3.75 m when the commercial vehicle volume exceeds 20 % of the
total traffic.
A shoulder is the portion of the roadway continuous with the travelled way for
accommodation of stopped vehicle, for emergency use and for lateral support of the
pavement.
The normal useable shoulder width that should be provided along important roads is 3 m.
However, in difficult terrain and on low volume roads, useable shoulders of this width may
not be feasible. A minimum useable shoulder width of 0.6 m should be considered in such
cases. Table 1.6.4 and Table 1.6..5 give the widths of shoulders for the various road
standards in rural and urban areas respectively.
For shoulders to function effectively, they must be sufficiently stable to support occasional
vehicle loads, in all kinds weather without rutting. Paved or stabilised shoulders are
required to minimise rutting. Paved shoulders of the strength and standard as the
pavement should be considers for road standards R6, R5 and U6, U5 while stabilised
shoulders designed to cater for 50 % of the design traffic should considered be for road
standards R4, R3 and U4, U3. For road standards R2, R1, U2 and U1 the normal earth
type can be used. However, where gradients exceed 8 % the shoulder should also be
paved to prevent erosion.
1.6.5 MEDIANS
1.6.5.1 General
A median is a highly desirable element on all roads carrying four or more lanes and should
be provided wherever possible. The principal functions of a median are to provide the
desired freedom from the interference of opposing traffic, to provide a recovery area for
out-of-control vehicles, to provide for speed changes and storage of right-turning and U-
turning vehicles and to provide for future lanes.
For maximum efficiency, a median should be highly visible both night and day and in
definite contrast to the through traffic lanes.
Medians may be depressed, raised or flush with the pavement surface. They should be as
wide as feasible but of a dimension in balance with other components of the cross-section.
The general range of median width varies from a minimum of 1.0 m in a Type III urban
Figure 1.6.2 gives examples of kerbed and unkerbed medians while Table 1.6.6 and
Table 1.6.7 give the minimum and desirable widths and types of medians that are to be
applied to the various road standards. The median widths as expressed are as the
dimensions between the through lane edges and includes the right shoulders if any.
Table 1.6.8 gives the general guidelines for determining the type of crossing that is
required. Where the pedestrian and vehicle volumes does not fit into any of the category
shown, judgement is needed in the assessment of the type of crossing required.
For roads with dual 3-lanes or more, an overhead crossing or underpass should always be
considered if a pedestrian crossing is justified so as not to impede the smooth flow of traffic
unduly. Where justified, the location and design of the pedestrian crossing would require
an individual study. Where overhead pedestrian crossings are provided, side barriers must
be installed to prevent jay-walking. Such barriers shall be installed for a distance of 75 m
on both sides of the location of the crossing. The minimum spacing between crossings is
400 m.
1.7.1 GENERAL
The longitudinal profile of a road consists of a series of straight grades and vertical curves.
As well as smoothing the passage of a vehicle from one grade to another, the vertical
curves increase the sight distance over crests at the junction of the grades. They should be
simple in application and should result in a design that is safe, comfortable in operation,
pleasing in appearance and adequate for drainage. For simplicity, the parabolic curve with
an equivalent vertical axis centred on the vertical point of intersection is used.
Convex vertical curves are known as summit or crest curves, and concave vertical curves
as sag curves.
At crest curves, the minimum length may be fixed by stopping sight distance or
appearance requirements. However, lengths above the minimum may reduce the
overtaking sight distance available on the approaches. At sag curves, the length may be
fixed approximately by comfort related to vertical acceleration, appearance or, on a slightly
more deterministic basis, by drainage, headlight performance or overhead restrictions to
the line of sight.
The rate of change of grade to successive points on the curve is a constant amount for
equal increments of horizontal distance, and equals the algebraic difference between the
intersecting tangent grades divided by the length of curve or A/L in percent per metres.
The reciprocal L/A is the horizontal distance in metres required to effect a 1 percent
change in gradient and is a measure of curvature. This quantity (L/A), termed k, is used in
determining the horizontal distance from the beginning of the vertical curve to the apex or
low point of the curve. The k value is also useful in determining the minimum lengths of
vertical curves for the various design speeds.
The lengths of vertical curves used should be as long as possible and above the minimum
values for the design speeds where economically feasible.
Minimum lengths of crest vertical curves are determined by the sight distance
requirements. The stopping sight distance is the major control for the safe operation at the
design speed chosen. Passing sight distances are not used as it provides for an
uneconomical design. An exception may be at decision areas such as sight distance to
ramp exits where longer lengths are necessary.
The basic formulas for length of a parabolic vertical curve in terms of algebraic difference
in grade and sight distance (using an eye height of 0.92 m and object height of 0.15 m) are
as follows:
2
L= AS where S≤L
405
(1 - 7.)
405
L=2S− where S>L
A
Where:
L = length of vertical curve (m)
S = sight distance (m)
A = algebraic difference in grades (percent)
Table 1.7.1 indicates the minimum k values that are to be used in design for the
various design speeds.
Design Speed
(km/h) 120 100 80 60 50 40 30 20
Minimum K
Value 120 60 30 15 10 10 5 5
At least four different criteria for establishing lengths of sag vertical curves are recognised.
These are
(1) headlight sight distance,
(2) rider comfort,
(3) drainage control; and
(4) a rule of thumb for general appearance.
However, the headlight sight distance basis appears to be the most logical for general use
and this criterion is used to establish the, design values for a range of lengths of sag
vertical curves. It is again convenient to express the design control in terms of the k value.
Table 1.7.2 indicates the minimum k values that are to be used.
Longer curves are desired wherever feasible and should be used, but where k values in
excess of 55 are used, special attention to drainage must be exercised. Shorter sag
vertical curves may be justified for economic reasons, in cases where an existing element
such as a structure that is not ready for replacement controls the vertical profile.
Drainage of kerbed pavements are especially important on sag vertical curves where a
grade line of not less than 0.3 % within 15 m of the level point must be maintained.
1.7.5 GRADES
1.7.5.1 General
It is seldom practical to construct roads with grades sufficiently flat to permit all vehicles to
operate at the same speed. Therefore it is necessary to adopt a more modest design
standard, in which limits are placed on steepness and length of grade which may be used.
Suggested values have been formulated on the basis of a general maximum value that is
the limiting value to be used in ordinary design, with comments on situations where steeper
grades may be more appropriate.
There are three aspects to the design of grades that can be adopted in difficult terrain:
The poorer performing vehicles using the road (generally heavy vehicles in the lower
power ranges), must be able to climb the grade. This limits the maximum grade that can be
considered for roads open to the public. It only becomes an acceptable limit in low volume
situations, or for special purpose roads, eg to a specific tourist vantage point.
Grades cause the need for speed variations, gear changes and braking for all vehicles.
This is a quality of service consideration. Flatter grades, which enable a more consistent
travel speed, make fewer demands on both vehicle and driver and generally reduce vehicle
operating costs.
Grades cause speed disparities between vehicle types, leading to increased queuing and
overtaking requirements. This is a level of service problem. The increased overtaking
requirements and reduced service volumes can give rise to operational and safety
problems at higher traffic flows. The problem can arise from cars towing trailers as well as
from heavy commercial vehicles.
The maximum grade that can be negotiated by heavy vehicles is about 15 %. Motor cars
have, over recent years, developed in several ways. The increased performance of modern
light vehicles with higher efficiency engines and appropriate gear ratios and the increased
proportion of vehicles with automatic transmissions, has lessened the nuisance value to
the driver of occasional gear changes on steeper grades.
Uphill performance of motor cars has, therefore, ceased to be a major control on moderate
grades. In the higher volume ranges, interaction due to speed disparities between vehicle
classes is a more appropriate design criterion.
The safety aspect of uphill operation also relates to speed disparities. Rear-end accidents
can occur where the driver of a light vehicle comes unexpectedly on a slow moving large
vehicle.
Studies have shown that under favourable conditions of alignment and sight distance,
heavy commercial vehicles travelling downhill, even on very steep grades, appear to travel
as fast as their free speed on the flat. On sections with winding horizontal alignment and
grades in excess of 6 %, downhill speed seems to be restrained to about the same as
uphill speed. Car speed shows much less variation as would be expected since
performance is not a major control. Uphill and downhill speeds seem to differ by about
10-15 km/h spaced about evenly on either side of the flat gradient value. The efficiency of
car braking systems has improved over recent years and the downhill aspect of safety is
concerned primarily with the possibility of high speeds caused by equipment failure.
Grades used in design are, therefore, only controlled at the upper end by vehicle
performance. In most designs, the general maximum grade to be sought will be based on
The following situations may justify the adoption of grades steeper than the general
maxima:
comparatively short sections of steeper grade which can lead to significant cost
savings.
less important local roads where the costs of achieving higher standards are
In mountainous terrain where general maximum grades are not practical and generally on
less important local roads, maximum grades are fixed by the performance of heavy
vehicles. Grades as steep as 12 % to 15 % have been found satisfactory on low volume
roads. In extreme cases, for example access to an isolated vantage point where there are
no commercial vehicles, short lengths of grade as high as 33 % have been used.
On very steep grades, vehicles will travel relatively slowly. All vehicles ascending, and
most vehicles descending (particularly trucks), will use low gears. There is a possibility of
mechanical failure that could cause high descent speeds. On low volume roads, the
probability is relatively low and no special provision is necessarily made. On the more
important roads carrying higher traffic volumes where sections of adverse terrain may
require the use of maximum grades, it is desirable to increase the design speed of the
individual geometric elements progressively towards the foot of the steep grade. Appropri-
ate section design speeds are easily obtained and especially where percentages of heavy
vehicles are high, consideration should be given to construction of runaway vehicle
facilities.
Although speeds of cars may be reduced slightly on steep up grades, large differences
between speeds of light and heavy vehicles will occur and speeds of the latter will be quite
slow. It is quite important, therefore, to provide adequate horizontal sight distance to
enable faster vehicle operators to recognise when they are catching up to a slow vehicle
and to adjust their speed accordingly.
The term “critical grade length” indicates the maximum length of a designated upgrade
upon which a loaded truck can operate without an unreasonable reduction in speed. To
establish the design values for critical grade length, for which truck performance is the
determining factor , the following assumption are made:
The average running speed as related to design speed to approximate the speed of
vehicle beginning and uphill climb.
A maximum reduction in speed to half the design speed is allowed for design speeds of
80 km/hr or above except for design speeds of 50 and 40 km/hr where the allowable
minimum speeds are 30 and 25 km/hr respectively.
The critical grade lengths for the various speeds are as shown in Table 1.7.5. Where a
particular section is made up of a combination of upgrades, the length of critical grade
should take into consideration the entire section of the combination. A speed
reduction/recovery curve should be plotted to determine the critical grade length based on
the assumptions used above. Figure 1.7.1 gives the acceleration and deceleration curves
that can be used for this purpose.
1.8.1 GENERAL
The foregoing sections have concentrated primarily on the physical attributes of proper
road design to satisfy the requirements of safety and performance. It has been indicated
that, while these needs are of prime importance, some compromise may be necessary in
the interests of convenience and economy. Ultra-safe roads are of little value if access is
too inconvenient or if the additional cost reduces the number of projects that can be
undertaken.
Such issues are relatively tangible, and amenable to some form of assessment. There
remains, however, one area for consideration which is far from tangible or quantifiable, but
which is nevertheless of increasing importance; and that is the issue of amenity.
Considerations of amenity are those which concern the effect that a road and its traffic has
upon the environmental and aesthetic senses of users and of those others who are
affected by its construction and operation. The pleasing coordination of alignment and
grading, the fitting of the road to the natural contours of the land surface, and the
preservation or enhancement of the natural vegetation are all involved. This section
considers those less tangible factors which, while not subject to rigid analytical treatment,
are none-the-less important in the determination of the total efficacy of a road.
The traditional method of designing roads has been based on the limitations of manual
techniques wherein the problem is considered, separately, in three views; plan, longitudinal
profile and cross section. Such an approach has been born of the semi-graphical
techniques usually employed, and can clearly produce satisfactory results if carried out by
an experienced designer.
Equally, it can produce poor results if the designer only considers each view independently
of the other two, without proper consideration of the interrelation between the various
views.
Even conscientious adherence to the appropriate tabulations and charts included in the
previous sections of this publication will not guarantee a satisfactory result, if the three
traditional views are treated independently. The road user sees the road as a constantly
changing three-dimensional continuum, and unless designers take full cognisance of this
fact, they may not appreciate how the finished design will appear to the road user. It is the
appearance of the road to the driver that determines the driver's behaviour, and unless the
road appears to the driver, as the designer intended, then the design will have failed in one
of its most important attributes, that of satisfying the needs of the user.
The driver sees a foreshortened and, thus, distorted view of the road, and unfavourable
combinations of horizontal and vertical curves can result in apparent discontinuities in the
alignment, even though the horizontal and vertical designs each comply separately with the
provisions of previous sections. Such combinations can mask from the driver a change in
horizontal alignment or even a sag curve deep enough to conceal a significant hazard (the
hidden dip problem). Only the consideration of the road as a three dimensional entity can
reveal such deficiencies, and good design practice requires the elimination of all avoidable
hazards even though some additional expense may be incurred. The removal of the
hazard is not, however, the only benefit, as the improved safety and performance potential
is invariably accompanied by significantly enhanced amenity.
Not only is the driver's view constantly changing, but the duration of his view of successive
elements of a road is also varying. Features situated in a long, low sag remain in view for
a considerable length of time whereas other features at or near an abrupt crest or on a
tight curve are in view only fleetingly. It follows then that important features such as
intersections are most favourably located on long sag curves, whilst unavoidable
objectionable peripheral features are less offensive if located near crests or tight curves.
Visual clues to the driver from peripheral areas must be given adequate attention. While
the designer views the whole road layout at once, and is aware of all changes in alignment,
the driver sees much less at any one time. His inherently restricted view can be further
limited at night, or in other times of poor visibility. The designer must, therefore, provide the
driver with as many clues as possible as to what lies ahead, but must make sure that the
roadside conditions do not convey messages which are ambiguous or misleading.
The smooth curving face of a cutting can indicate the corresponding curvature in a road,
but a fill can give no sense of position if the view ahead is open to the sky, such as on a
crest where there may be no visual background of fixed objects. A driver will not expect a
curve, however, if his sole clues come from a line of service poles or trees which continue
straight on, or from a deceptive gap in a line of trees on the outside of the curve. Again, a
crest should not obscure a potential hazard such as a narrow bridge, a railway at-grade
crossing or a horizontal curve that requires a significant reduction of speed. Even small
crests can obscure important features. The visual clues presented to the driver by the view
of the road surface are especially important on sections of sharp curvature. This should
not be confused with the assumptions for sight distance calculations discussed in Section
1.5 related to stopping or avoiding a hazard.
It has been shown that the speed environment of a road is influenced partly by the nature
of the terrain, and partly by the horizontal alignment. It follows, therefore, that if the
indications of these two factors are similar, the road will provide the best level of
consistency in driver expectancy, and thus safety. Further, a road having both horizontal
and vertical curvature carefully designed to conform to the terrain will result in a structure
having the desirable aesthetic quality of being in harmony with the landform.
Perfect harmony of course is not always possible, and the designer must consider what
matters are beyond his control, and make full allowance for their influence on driver
behaviour. From Section 1.2 it will be clear that, while it is possible to build a road with a
high-speed environment in adverse terrain, it is unlikely that there will ever be sufficient
curvature in flat or gently rolling terrain to produce a low-speed environment. Design
speeds will be high in the latter cases because of the terrain. The grading needs to ensure
adequate sight distances to potential hazards on the road and, where such sections merge
into more constrained alignment sections, such transition must be accomplished gradually
rather than suddenly.
The suggestions in Section 1.3.3.3 cover transitioning from one speed environment to a
lower one. They are relevant in the context of fitting the road to terrain, where the terrain
changes. In flat open terrain, long straight road sections are common, but generally there
is advantage in avoiding excessive lengths of straight road. A gentle curvilinear design
always helps to keep the operating conditions under control and at the same time, affords
scope for far more sympathetic fitting of the road to the terrain. The increased flexibility of
this approach enables more pleasing designs to be produced at no extra cost: economies
in earthworks can often be achieved by fitting the road more closely to the terrain. In
addition, safety is enhanced by making the driver more aware of his speed, by allowing him
to make better assessments of the distances and speeds of other vehicles, by reducing
headlight or sun glare in appropriate circumstances, and by reducing boredom and fatigue.
Even in flat country curvilinear designs can be used. Radii must be very large, so that not
all the benefits of a curving alignment are achieved. Estimation of speed of oncoming
vehicles is not significantly improved over a straight alignment when radius exceeds about
5,000 metres to 10,000 metres. It is the opinion of experienced designers, however, that
sufficient benefit does still remain to make the exercise worthwhile.
Curvilinear design is most readily applicable to divided roads with their less stringent sight
distance requirements, but the principles are just as relevant to single carriageway roads
provided care is taken to ensure adequate overtaking opportunities are available. Very
large radius curves can provide overtaking opportunities and, as mentioned above, retain
at least some of the benefits of curving alignment. If the topography is such that 'natural'
curvature precludes the provision of overtaking sight distance, then artificially introduced
long straights could be lacking in harmony with the terrain, and may be subject to inhibiting
vertical curvature or require heavy earthworks. In such cases, a harmonious curving
alignment with the provision of overtaking zones may produce an economical as well as an
aesthetic solution.
Figure 1.8.1 illustrates basic examples of the method and benefits of proper fitting of the
road to the terrain and of proper coordination of horizontal and vertical elements. In
addition, there are some examples of poor design form, with indications of appropriate
remedial measures. These latter examples are typical of the results likely if the designer
does not consider the vertical and horizontal views simultaneously; particularly if a
'minimum' vertical standard is superimposed on a relatively unrestricted horizontal regime.
The diagrams are not intended to be comprehensive, but serve merely to demonstrate the
general concepts that should (or should not) be followed. In all cases, recognition of the
deficiency is sufficient to indicate the appropriate remedy, and the recognition of the
deficiency is dependent only on the designer taking a three-dimensional, rather than a
two-dimensional view of the problem.
Specific rules are not appropriate to good design, as each particular project has its own
peculiar problems and constraints. However, some benefit can be obtained from a
consideration of what combinations of horizontal and vertical elements are most likely to
produce satisfactory results, and visualising the schemes in these dimensions using
whatever aids are available.
PLAN
PLAN
The dynamic and safety aspects of horizontal alignment have already been discussed, but
further consideration must be given if the resulting standards are not to be used
mechanically as minima which can sometimes produce aesthetically poor results. While
Section 1.5 indicates the basic minimum curve and transition lengths required to produce
satisfactory results for performance and safety, appearance criteria may necessitate
considerably higher standards in some circumstances. The considerations set out in
Sections 1.8.5 to 1.8.7 assume greatest importance when the geometric elements of the
road are visible for some distance, ie generally in easy terrain. In mountainous areas, while
they will have some relevance, considerations of safety and economy may assume greater
importance.
Difficult terrain imposes sufficient obvious restraints that little problem is encountered in
providing road geometry in harmony with the terrain, and the driver can readily appreciate
the reasons for any restriction. Easy country, however, displays no such obvious restraints
and the designer must therefore ensure that the geometry does not appear forced or
unnatural. For this reason, curve radii considerably greater than the minimum specified in
Table 1.4.2 may be justified in flat terrain. Arc lengths of at least 500 metres may be
required to ensure that curves with small deflection angles do not appear as kinks.
Long straights may be acceptable in level country as the artificial introduction of curves
may result in the appearance of a forced alignment. Where possible, isolated curves
should be avoided in flat terrain, even if a fully curvilinear alignment is not employed.
Several curves in succession, from time to time, add welcome punctuation to a journey in
otherwise featureless terrain. Rarely is the terrain so flat that some legitimate curvature
cannot be introduced. The gradual introduction of successively tightening curves should be
used judiciously to alert a driver to a transition from easy terrain to more restricted
conditions.
Generally, compound curves should be avoided, unless the radii are very large, because of
the difficulty the driver has in detecting the change of curvature. However, compound
curves may be considered where:
the topography, or some other control, for example a bridge makes the use of a
single-radius curve impracticable; or
the compound curve replaces an otherwise 'broken back' horizontal curve.
The provisions of Section 1.4.5 regarding safety of compound curves are important. In
addition, the total curve length should be sufficient to allow a significant arc length for each
radius to be used. Where these conditions cannot be met, spiral transitions may be used
between successive circular arcs.
A free-flowing form that fits naturally to the terrain usually incorporates reversed horizontal
curvature. Such an alignment frequently looks better than any other, particularly in rolling
terrain. The designer, apart from satisfying the appropriate dynamic requirements, should
ensure that the curves are of correctly related radii and that they are sufficiently well
separated to allow pleasing warping of the superelevation. The guidelines of Section 1.4
indicate the requirements.
Horizontal curves in the same direction joined by short straights should be avoided, if
possible, as it is virtually impossible to provide the correct amount of superelevation
throughout, and it is equally difficult to produce a pleasing grading of pavement edges. In
practical terms, there is a maximum length of ‘short’ straight that can be tolerated without
visual or physical complication, just as there is a minimum length of 'long' straight which
provides effective separation of similar curves. While lengths of straight between these
two values may have to be used in constrained situations, avoidance of these will lead to a
more pleasing design.
A nominal length of straight corresponding to about 2 seconds travel time, that is a length
of 0.6 V metres, where V is the design speed in km/h is the maximum of the ‘short' range
that can be successfully accommodated. In most cases, a single curve with a slightly larger
radius could and should be used. Straights of this order do not permit reversion to two-way
crossfall between curves, but some improvement to appearance and safety can result if the
one-way crossfall is reduced somewhat on the straight.
The minimum length of 'long' straight that should be used between similar curves is such
that normal crossfall can be regained, and maintained for a distance corresponding to
about 4 seconds travel time. Based on the superelevation development lengths of
Table 1.4.4, this results in a straight of minimum length between 3 V and 4 V metres. Even
though such a straight may be visually satisfactory, there is advantage in exceeding this
minimum value. Where such a minimum cannot be obtained, the alignment should be
altered, either to increase the straight, or eliminate it entirely by the use of a compound
curve, a curve that is transitional between the main curves, or a long spiral alignment.
Much of the philosophy of horizontal alignment is equally valid to vertical alignment. Even
though the vertical scale is invariably much smaller than the horizontal, injudicious
selection of vertical components can be equally as hazardous, and as unsightly as poor
horizontal design. Indeed, some aspects of vertical design may warrant even closer
attention to detail than similar aspects of horizontal design. Research has indicated that a
driver's behaviour is influenced mainly by horizontal alignment and terrain; drivers do not
seem to respond to variations in vertical geometry. Vertical geometry cannot, therefore, be
determined in isolation, but must be related to the corresponding horizontal geometric
features.
Vertical grading composed of long tangents and generous vertical curves should be
adopted in preference to sections with numerous grade changes between short tangents.
Hidden dips should be avoided, and design checks of sight lines using a zero height
object, ie to the road surface, will reveal such deficiencies.
Road profiles should not include minor humps or hollows, which curtail sight lines, when
the horizontal alignment is adequate to provide over taking sight distance.
Broken-back profiles consisting of two or more vertical curves in the same direction
separated by short tangents should be avoided. Such profiles are disjointed and look
unsightly, especially in sag situations and should be replaced by single, longer vertical
curves. The difference in levels, and hence earthworks between the two situations is rarely
significant.
Reversed vertical curves can be designed to have common vertical tangent points and
such practice produces pleasing, flowing grade lines which are more likely to be in
harmony with the natural landform. The rate of change of vertical acceleration at the
tangent point must be considered, however, if either of the component curves is designed
for near limiting values of vertical acceleration. The algebraic sum of the adjacent rates of
vertical acceleration should remain within the acceptable limit, usually 0.1 g, if comfort
criteria are to be maintained.
The most pleasing three-dimensional result is achieved if the horizontal and vertical
curvature is kept in phase, as this relates most closely to naturally occurring forms. Where
possible, the vertical curves should be contained within the horizontal curves. This
enhances the appearance in sag curves, by reducing the three-dimensional rate of change
of direction, and improves the safety of crest curves by indicating the direction of curvature
before the road disappears over the crest. Thus, the best appearance requires the scale of
the vertical and horizontal movements to be comparable: a small movement in one
direction should not be combined with a large movement in the other.
Drainage structures in sag curves that are combined with horizontal curves require careful
design if a disjointed or kinky appearance is to be avoided. Culverts should introduce little
aesthetic difficulty if they are contained within embankments and are made sufficiently long
to accommodate full road formation widths.
Bridges built on combined horizontal and vertical curvature can present considerable
aesthetic problems, especially if reduced formation widths are used. Particular care should
be devoted to the design of the bridge kerbs and railings, as well as to the location and
transitioning of approach guard fences.
Horizontal curves combined with crests have less influence on the appearance of a road
than those combined with sags. Nevertheless, the effect on safety can be much greater, as
the crest can obscure the direction and severity of the horizontal curve. Minimum radius
horizontal curves, therefore, should not be combined with crest vertical curves.
APPENDIX A
Prakas No. 377, Dated 11th October, 2001
(Page 1)
APPENDIX B
Decision No. 328, Dated 13th November, 1998
END OF DOCUMENT