Chapter 6

Geometric Design Manual – 2013 Cross Section Elements

6 CROSS SECTION ELEMENTS

6.1 Introduction

A cross-section will normally consist of the carriageway, shoulders or curbs, drainage

features, and earthwork profiles.
1. Carriageway- the part of the road constructed for use by moving traffic, including
traffic lanes, auxiliary lanes such as acceleration and deceleration lanes, climbing
lanes, and passing lanes, and bus bays and lay-bys.
2. Roadway- consists of the carriageway and the shoulders, parking lanes and viewing
areas
3. Earthwork profiles- includes side slopes and back slopes

For urban cross-sections, cross-section elements may also include facilities for pedestrians,
cyclists, or other specialist user groups. These include curbs, footpaths, and islands. It may
also provide for parking lanes. For dual carriageways, the cross-section will also include
medians. Typical Cross Sections are illustrated in Appendix E of this manual. Bus lay-bys,
parking lanes, passing lanes, and viewing areas are discussed in Chapter 14.

Lane and shoulder widths should be adjusted to traffic requirements and characteristics of
the terrain (Tables 2.1, 2.2, 2.3, 2.5). The cross-section may vary over a particular route
because these controlling factors vary. The basic requirements are, however, that changes
in cross-section standards shall be uniform within each sub-section of the route and that
any changes of the cross-section shall be effected gradually and logically over a transition
length. Abrupt or isolated changes in cross-section standards lead to increased hazards and
reduced traffic capacity and complicate construction operations.

In certain cases, however, it may be necessary to accept isolated reductions in cross-section

standards, for example, when an existing narrow structure has to be retained because it is
not economically feasible to replace it. In such cases a proper application of traffic signs
and road markings is required to warn motorists of the discontinuity in the road. However,
all narrow structures must be widened or replaced when the width across the structure is
actually less than the carriageway width for the road standard in question.

6.2 Lane Widths

The width of the running surface and shoulders of a road largely define its cost, other
things being equal, hence defining width standards that are acceptable both to the highway
authority and to the travelling public is vital.

The width of the carriageway and shoulders has a great influence on road safety; wide
roads generally being safer than narrower ones unless speeds on narrow roads are low.
However, quantifying this in economic terms is inherently very difficult.

Wider roads require less shoulder maintenance because fewer vehicles drive over the
vulnerable edge between the running surface and shoulder. Less maintenance is also
needed because wheelpaths are wider, wheel loads less concentrated and less damage is
caused to surfacings and road bases. Also, the weak area near to the edge of the running
surface receives less traffic and deteriorates less rapidly. On the other hand, the area of

Ethiopian Roads Authority Page 6-1

Chapter 6
Cross Section Elements Geometric Design Manual - 2013

surface that requires maintenance is larger on wide roads hence when re-surfacing is
required the cost is higher.

There is insufficient quantifiable evidence to justify carriageway and shoulder widths

purely on economic grounds using whole life cost principles. This requires knowledge
about the costs of accidents and their causes, knowledge about the road deterioration and
the effects of maintenance and much more. Knowledge about these issues has been
collected for many years and is contained in the publications related to the Highway
Development and Management Model HDM-4 and its predecessors. Such a model is very
useful for calculating the optimum standards for various aspects of the design of a road but
calculating optimum standards for widths remains a difficult problem. Because of this,
road widths are based on the long term international evolution of such standards modified
by local considerations. Judgement rather than precise calculations are required and the
final standards are based on consensus. A primary principle, however, is that higher
standards are required for higher levels of traffic. Standards for carriageway widths are
shown in Table 2.1 for all road design standards.

Auxiliary lanes at intersections often help to facilitate traffic movement. Such added lanes
are discussed in the Chapters 11 and 12.

6.3 Shoulders

A shoulder is the portion of the roadway contiguous to the carriageway for the
accommodation of stopped vehicles; traditional and intermediate non-motorised traffic,
animals, and pedestrians; emergency use; the recovery of errant vehicles; and lateral
support of the pavement layers. Shoulder widths for the different design standards, terrain
type, and urban/rural environment are shown in Table 2.2. Shoulders are not specifically
defined for gravel and earth roads because, in practice, the road material spreads across the
trafficable area with no distinct demarcation. On paved roads, shoulders vary from a
minimum of 0.5m up to 3.0m depending on the terrain and design classification.

The basic shoulder widths shown in Table 2.2 are increased on some of the lower road
standards if the number of Large Heavy Vehicles in the traffic stream is high (Table 2.3)
and if the number of motorcycles, non-motorised vehicles and pedestrians (measured in
terms of Passenger Car Units) exceeds 300 (Table 2.5). At the present time this limit is
based on average daily conditions. On market days the road within several kilometres of an
urban centre can contain a very high volume of non-motorised traffic and this can be
sufficient for the average over the week to exceed 300 and justify wider shoulders.
However many markets, although very busy, may not be quite large enough to exceed this
average. The design engineer should be aware of the local situation and, if he/she feels that
it is necessary (e.g. on the basis of safety and/or congestion) to widen the shoulders, he/she
should be encouraged to request a modification to the standards.

Where the carriageway is paved, the shoulder should also be sealed with a bituminous
surface treatment. This has several advantages;
i) prevents edge ravelling and the maintenance problems associated with parking on
an unpaved shoulder;
ii) controls ingress of moisture into the upper pavement layers;
iii) provides paved space for vehicular parking outside of the traffic flow;

Page 6-2 Ethiopian Roads Authority

 Chapter 6
Geometric Design Manual – 2013 Cross Section Elements

iv) provides a better surface for vehicles experiencing emergency repairs;

v) caters for the very heavy pedestrian traffic observed in the villages, traffic that
would otherwise use the roadway.
The road layout in terms of widths in urban and peri-urban areas is shown in Table 2.2.
The sealed shoulder width may increase to 3.5 metres where provision for a parking lane is
required and the shoulders should be paved rather than sealed. In fact, it is not strictly a
shoulder but an additional lane.

In cases where terrain is severe, the existing roadway width is narrow, and where the
shoulder width can only be maintained through an excessive volume of earthwork e.g. at
escarpment conditions, standards can be reduced through the Departure from Standard
process presented in Chapter 2.

6.4 Normal Crossfall

Normal crossfall (or camber, crown) should be sufficient to provide adequate surface
drainage whilst not being so great as to make steering difficult. The ability of a surface to
shed water varies with its smoothness and integrity. On unpaved roads, the minimum
acceptable value of crossfall should be related to the need to carry surface water away from
the pavement structure effectively, with a maximum value above which erosion of material
starts to become a problem.

The normal crossfall should be 3.0 percent on paved roads. Shoulders having the same
surface as the roadway should have the same normal crossfall. Unpaved shoulders on a
paved road should be 1.5 percent steeper than the crossfall of the roadway.

The crossfall or camber on unpaved roads does not remain the same for very long because
of movement of the gravel and its loss. Research has shown that the rate of deterioration of
a gravel road is highly dependent on the camber and the ability of the surface to shed water
effectively. The higher the initial camber, the longer it takes for the surface to deteriorate
to a shape that fails to shed water adequately, thereby causing accelerated deterioration. In
most circumstances, crossfalls/cambers as high as 6.0 percent should be used, although this
value may need to be modified for some types of gravel. .

For shoulder crossfall in super-elevated conditions, refer to Chapter 8.

6.5 Side Slopes and Back Slopes

Side slopes should be designed to ensure the stability of the roadway and to provide a
reasonable opportunity for recovery of an out-of-control vehicle.

Three regions of the roadside are important when evaluating the safety aspects:
i) top of the slope (hinge point),
ii) side slope, and
iii) toe of the slope (intersection of the fore slope with level ground or with a back
slope, forming a ditch).
Figure 6.1 illustrates these three regions.

Ethiopian Roads Authority Page 6-3

Chapter 6
Cross Section Elements Geometric Design Manual - 2013

Research has found that rounding at the hinge point can significantly reduce the hazard
potential. Similarly, rounding at the toe of the slope is also beneficial.

Carriageway Hinge point

Back slope

Ditch

Shoulder
Side slope
Toe of slope

Figure 6-1: Designation of Roadside Regions

Refer to Table 6.1 for details of side slopes and back slopes.

Embankment or fill slopes parallel to the flow of traffic may be defined as recoverable,
non-recoverable, or critical. Recoverable slopes include all embankment slopes 1:4 or
flatter. Motorists who encroach on recoverable slopes can generally stop their vehicles or
slow them enough to return to the roadway safely. Fixed obstacles such as culvert head
walls should not extend above the embankment within the clear zone distance.

A non-recoverable slope is defined as one which is traversable, but from which most
motorists will be unable to stop or to return to the roadway easily. Typically, vehicles on
such slopes can be expected to reach the bottom. Embankments between 1:3 and 1:4
generally fall into this category. Since a high percentage of encroaching vehicles will reach
the toe of these slopes, the clear zone distance extends beyond the slope, and a clear run-
out area at the base is desirable.

A critical slope is one on which a vehicle is likely to overturn. Slopes steeper than 1:3
generally fall into this category.

The selection of a side slope and back slope is dependent on safety considerations, height
of cut or fill, and economic considerations. Furthermore, the guidance in this chapter may
be most applicable to new construction or major reconstruction. On maintenance and
rehabilitation projects, the primary emphasis is placed on the roadway itself. It may not be
cost-effective or practical because of environmental impacts or limited right-of-way to
bring these projects into full compliance with the side slope recommendations provided
here.

Table 6.1 indicates the side slope ratios recommended for use in the design according to
the height of fill and cut, the material and practical experience in Ethiopia of the costs of
construction. It will be noted that with the single exception of roads in areas of black

Page 6-4 Ethiopian Roads Authority

 Chapter 6
Geometric Design Manual – 2013 Cross Section Elements

cotton soils, the recommended slopes are too steep to meet the recommendations for
adequate safety. Achieving a good safety design is clearly a function of overall cost and at
the present time is only likely to be viable for the highest classes of road.

Table 6-1: Slope Ratio Table – Vertical to Horizontal

Height of Slope Side Slope (V:H)
Material Back Slope
(m) Fill Cut
Earth Soil 0.0 – 1.0 1:3 1:3
1.0 – 2.0 1:2 1:2

>2.0 2:3 2:3

Strong Rock 0.0 – 2.0 4:5 2:1
>2.0 1:1 4:1
1:2
Weathered Rock 0.0 – 2.0 2:3 2:1
>2.0 1:1 3:1
Decomposed Rock 0.0 – 1.0 1:3 1:3
2.0 – 2.0 1:2 1:2

>2.0 2:3 2:3

Black Cotton Soil 0.0 – 2.0 1:6
(expansive clays)(1) - -
>2.0 1:4
Note 1 Move ditch away from fill as shown in Figure 6.2

This Table should be used as a guide only, particularly because applicable standards in
rock cuts are highly dependent on costs. Also certain soils that may be present at subgrade
level may be unstable at 1:2 side slopes and therefore a higher standard will need to be
applied for these soils. Slope configuration and treatments in areas with identified slope
stability problems should be addressed as a final design issue.

6.6 Roadside Ditches

For the detailed design of roadside ditches, the ERA Drainage Design Manual should be
used.

The choice of side drain cross-section depends on the required hydraulic capacity,
arrangements for maintenance, space restrictions, traffic safety and any requirements
relating to the height between the crown of the pavement and the drain invert

Under normal circumstances the adoption of a trapezoidal cross-section will facilitate

maintenance and will be acceptable from the point of view of traffic safety. It is much
easier and appropriate to dig and clean a trapezoidal drain with hand tools and the risk of
erosion is lower. The minimum recommended width of the side drain is 500mm. This
shape has high flow capacity and, by carefully selecting the gradients of its side slopes, it
will resist erosion.

The V-shape is the standard shape for a drainage ditch constructed by a motor-grader. It
can be easily maintained by heavy equipment but it has relatively low capacity
necessitating more frequent structures for emptying it. Furthermore the shape concentrates
flow at the invert and encourages erosion. Using a V-shaped ditch the minimum depth
should be 0.6m in mountainous and escarpment terrain, and 1.0m elsewhere.

Ethiopian Roads Authority Page 6-5

Chapter 6
Cross Section Elements Geometric Design Manual - 2013

The side slope and back slope of ditches should generally be no less than 1:2; however,
these slopes should conform to the slopes given in Table 6.1.

Side drains should be avoided in areas with expansive clay soils such as black cotton soils.
Where this is not possible, they should be kept at a minimum distance of 4-6 m from the
toe of the embankment, dependent on functional classification (6m for trunk roads), as
shown in Figure 6.2. The ditch in this instance should have a trapezoidal, flat-bottom
configuration.

Figure 6-2: Side Drain Ditch Location in Expansive Soils

6.7 Clear Zone

Once a vehicle has left the roadway, an accident may occur. The end result of an
encroachment depends upon the physical characteristics of the roadside environment. Flat,
traversable, stable slopes minimize overturning accidents, which are usually severe.
Elimination of roadside furniture or its relocation to less vulnerable areas is an option in
the development of safer roadsides. If a fixed object or other roadside hazard cannot be
eliminated, relocated, modified, or shielded, for whatever reason, consideration should be
given to delineating the feature so it is readily visible to a motorist.

For adequate safety, it is desirable to provide an unencumbered roadside recovery area that
is as wide as practical on a specific highway section. The cleared width should be a
minimum of 15 metres each side from the edge of the roadway for the higher road
standards.

For lower standard roads, the clear zone can be reduced. It should extend beyond the toe of
the slope. Lateral clearances between roadside objects and obstructions and the edge of the
carriageway should normally be not less than 1.5 metres. At existing pipe culverts, box
culverts and bridges, the clearance cannot be less than the carriageway width; if this
clearance is not met, the structure must be widened. New pipe and box culvert
installations, and extensions to them, must be designed with a 1.5-metre clearance from the
edge of the shoulder.

Horizontal clearance to road signs, marker posts, etc. must be a minimum of 1.0m from the
edge of the carriageway.

6.8 Right-of-Way

Right-of-ways, or road reserves, are provided in order to accommodate road width and to
enhance the safety, operation and appearance of the roads. The width of the right-of-way

Page 6-6 Ethiopian Roads Authority

 Chapter 6
Geometric Design Manual – 2013 Cross Section Elements

depends on the cross section elements of the highway, topography and other physical
controls together with economic considerations. Although it is desirable to acquire
sufficient right-of-way to accommodate all elements of the cross section and appropriate
border areas, right-of-way widths should be limited to a practical minimum in both rural
and developed areas affecting the economy of the inhabitants.

Right of ways will be equidistant from the centreline of the road to the left and to the right
of the carriageway. They should always be determined and shown on the final design plans
of road projects.

Road reserve widths applicable for the different road classes are shown in Tables 2-6
through to 2.16. In mountainous or escarpment terrain, a cut section may be of such depth
that the right-of-way width is exceeded from the top of cut on one side to the other top of
cut.

Additional areas required for outlets etc., should be provided in a manner that will not
endanger the future integrity of the drainage facility and will provide adjoining land
owners restricted use of this land after completion of the road.

Reduced widths should be adopted only when these are found necessary for economic,
financial or environmental reasons in order to preserve valuable land, resources or existing
development or when provision of the desirable width would incur unreasonably high costs
because of physical constraints. In such cases, it is recommended that the right-of-way
should extend a minimum of a nominal 3 metres from the edge of the road works.
However, where this occurs, it is advisable to restrict building activity along the road to
prevent overcrowding, to preserve space for future improvements, and to provide for sight
distances at curves. The distance across the carriageway from building line to building line
should be a minimum of 15m.

For dual carriageway roads it may be necessary to increase the road reserve width above
the given values.

6.9 Four-Lane and Divided Roads

For such roads a planning horizon exceeding the 15 or 20 years used for rural roads is
required. Four-lane and divided roads are necessary when the traffic volume is sufficient to
justify their use and, in urban/peri-urban areas, this may be anticipated in the foreseeable
future. Indeed, some cities and towns have assumed that they will eventually be inevitable
and have included four-lane roadways in their master plans.

A minimum median width of 5.0 metres is required to allow the provision of left-turning
lanes outside of the adjacent carriageway, and to avoid having a turning passenger vehicle
from the minor road protrude into the through lanes. Geometric standards for four lane
roads are given in Tables 2.1, 2.2, and 2.3.

6.10 Medians

The median is the total area between the inner edges of the inside traffic lanes of a divided
road, and includes the inner shoulders and central islands. The purpose of the median is to
separate opposing streams of traffic hence reducing the possibility of vehicles crossing into

Ethiopian Roads Authority Page 6-7

Chapter 6
Cross Section Elements Geometric Design Manual - 2013

the path of opposing traffic. This is accomplished by the selection of the width of the
median or by a physical barrier such as a guardrail.

Medians are also used to reduce the nuisance of headlight glare by the planting of shrubs
on the central island. The shrubs should not grow so tall that sunlight could fall into the
driver's eyes in bands - the stroboscopic effect encountered in avenues of trees in the early
morning or late afternoon. In addition, the stems of the shrubs should not grow so thick as
to become a further possible hazard to the motorist; a maximum stem thickness of 100 mm
is recommended. Medians should not, as far as possible, be obstructed by street furniture.

Median width depends not only on traffic volume but also on the function of the road and
on traffic composition. For example, a median functioning purely as a pedestrian refuge
could be much narrower than one protecting a turning vehicle (which could be semi-trailer
plus trailer).

A median width of 9.2 m eliminates most cross-median accidents, and this width is
recommended where no barriers are provided between opposing traffic flows. Where a
road is to be constructed in stages, the median should be wide enough to accommodate
future lanes, without falling below the recommended width in the final stage. Operational
difficulties may arise at intersections with very wide medians because of the duration of
turning movements. The median should, however, be wide enough to provide refuge to the
design vehicle.

With severe space limitations, it is possible to use medians that are as little as 1.5 m wide.
These would, however, serve only to accommodate back-to-back guardrails to ensure
vehicular separation. A median that is 5.0 m wide is able to accommodate a right turn lane
with provision for a pedestrian refuge but would also require guardrail protection to
separate the opposing flows of traffic.

It is suggested that the median island should be depressed rather than raised, because a
raised or kerbed median island, will automatically require the inner shoulder to be 3.0 m
wide to allow sufficient space for emergency manoeuvres, including stopping. A depressed
median also facilitates drainage.

The purpose of an outer separator is most frequently to separate streams of traffic flowing
in the same direction but at different speeds and also to modify weaving manoeuvres. In
general, the standards applied to medians are equally appropriate to outer separators.

Two different conditions dictate the steepness of the slope across the median namely
drainage and safety. The normal profile of a median is a negative camber, i.e. sloping
towards a central low point, to facilitate drainage. The flattest slope that is recommended is
10 per cent. Slopes flatter than this may lead to ponding and to water flowing from the
median to the carriageway. Slopes steeper than 1:4 make control of an errant vehicle
difficult, leading to a greater possibility of cross-median accidents. If surface drainage
requires a median slope steeper than 1:4, this aspect of road safety might justify replacing
surface drainage by an underground drainage system.

Differential, or split, grading requires the median to be sloped to absorb the height
difference between the carriageways. This is achieved, in the case of small height
differences, by locating the low point of the median eccentrically, retaining the maximum

Page 6-8 Ethiopian Roads Authority

 Chapter 6
Geometric Design Manual – 2013 Cross Section Elements

permissible slope. The limit is reached when the low point is adjacent to the lower
carriageway and functions as a side drain. If a steeper slope is required, the carriageways
have to be designed as completely independent roadways, with full-width shoulders,
guardrails if necessary, and sufficient distance between shoulder breakpoints, with the side
slope appropriate to the in-situ material, to accommodate the height difference between
carriageways.

The design of the ends of medians is described in Section 11.9.2.

6.11 Single Lane Roads

For low traffic volume roads (<75ADT) single lane operation is adequate because the
probability of vehicles meeting each other from opposite directions is small and the few
passing manoeuvres can be undertaken at very reduced speeds using either the shoulder or
passing bays. Provided sight distances are adequate for safe stopping, these manoeuvres
can be performed without hazard, and the overall loss in efficiency brought about by the
reduced speeds is small.

The lowest design standards (DC1 and DC2) are not sufficiently wide for passing and
overtaking to occur on the carriageway and passing bays must be provided. The increased
width at passing bays should be such as to allow two design vehicles to pass, i.e. a
minimum of 5.0 m width, and vehicles are expected to stop or slow to a very low speed.

Normally, passing bays should be located every 300 to 500 metres depending on the terrain
and geometric conditions. However, adjacent passing bays must be visible from each other.
Account should be taken of sight distances, the likelihood of vehicles meeting between
passing bays and the potential difficulty of reversing. In general, passing bays should be
constructed at the most economic locations as determined by terrain and ground conditions
such as transitions from cuttings to embankment, rather than at precise intervals.

The length of individual passing bays varies with local conditions and the type of design
vehicle but, generally, a length of 20 metres including tapers caters for most commercial
vehicles.

Significant cost savings may be realized in mountainous and escarpment terrain by

incorporating short lengths of DC1 standard within a DC2 road. This Departure from
Standard may be economically justified, especially in escarpment terrain, for design traffic
flows of less than 75 vehicles per day. However, appropriately placed, inter-visible passing
bays are essential to ensure the free flow of traffic.

6.12 Typical Cross Sections and Standard Cross Sections

Typical cross sections are illustrated in ERA Standard Detail Drawings, and cross sections
for the standard classes of roads are illustrated in Appendix E of this manual.

Ethiopian Roads Authority Page 6-9

 Chapter 8
Geometric Design Manual – 2013 Horizontal Alignment

7 DESIGN SPEED AND SIGHT DISTANCES

7.1 Design Speed

Design speed is defined as the maximum (actually the 85th percentile) safe speed that can
be maintained over a specified section of road when conditions are so favourable that the
design features of the road govern the speed. To ensure that a driver is presented with a
consistent speed environment, design speed is used as an index that essentially defines the
geometric standard of a road, linking many of the factors that determine the road’s service
level, namely traffic volume; terrain; pavement type; safety/population density; and road
function.

The concept of design speed is very useful because it allows the key elements of geometric
design to be selected for each standard of road in a consistent and logical way. For
example, design speed is relatively low in mountainous terrain to reflect the necessary
reductions in standards required to keep road costs to manageable proportions. The speed
is higher in rolling terrain and highest of all in flat terrain.

In practice the speed of motorised vehicles on many roads in flat and rolling terrain will
only be constrained by the road geometry over relatively short sections but it is important
that the level of constraint is consistent for each road class and set of conditions.

In view of the mixed traffic that occupies the rural roads of Ethiopia and the cost benefit of
selecting lower design speeds, it is prudent to select values of design speed towards the
lower end of the internationally acceptable ranges as shown in Table 7.1.

Table 7-1: Design Speeds

Design Design speed (km/h)

standard Flat Rolling Mountain Escarpment Urban
DC 8 120 100 85 70 50
DC 7 120 100 85 70 50
DC 6 100 85 70 60 50
DC 5 85 70 60 50 50
(1)
DC 4 70 60 50 (46) 25 50
DC 3 70 60 50 (46) (1) 25 50
(1)
DC 2 60 50 40(37) 20 50
DC 1 50 40 30 20 40
Note 1 The design speeds in mountainous terrain for unpaved roads has been adjusted slightly so that
the minimum radii of curvature are the same for both the paved and unpaved option. This
ensures that when a road is upgraded to paved standard, the existing curves are not already
‘over’ designed.

Changes in design speed, if required because of a change in terrain, should be made over
distances that enable drivers to change speed gradually. Thus changes should never be
more than one design step at a time and the length of the sections with intermediate

Ethiopian Roads Authority Page 7-1

Chapter 8
Horizontal Alignment Geometric Design Manual – 2013

standards (if there is more than one change) should be long enough for drivers to realise
there has been a change before another change in the same direction is encountered. In
general, a particular design speed should be used for a minimum distance of five
kilometres (i.e. considerably more than one single bend). Where this is not possible,
warning signs should be provided to alert drivers to the changes.

7.2 Stopping Sight Distance

In order to ensure that the design speed is safe, the geometric properties of the road must
meet certain minimum or maximum values to ensure that drivers can see far enough ahead
to carry out normal manoeuvres such as overtaking another vehicle or stopping if there is
an object in the road.

The distance a vehicle requires to stop safely is called the stopping sight distance. It mainly
affects the shape of the road on the crest of a hill (vertical alignment) but if there are
objects near the edge of the road that restrict a driver’s vision on approaching a bend, then
it also affects the horizontal curvature.

The driver must be able to see any obstacle in the road, hence on a crest curve, the stopping
sight distance depends on the size of the object and the height of the driver’s eye above the
road surface. The driver needs time to react and then the brakes of the vehicle need time to
slow the vehicle down. Hence stopping sight distance is dependent on the speed of the
vehicle and the efficiency of its brakes. The surface characteristics of the road also affect
the braking time so the values for unpaved roads differ from those of paved roads, although
the differences are small for design speeds below 60km/h.

The stopping distance also depends on the gradient of the road; it is harder to stop on a
downhill gradient than on a flat road because a component of the weight of the vehicle acts
down the gradient in the opposite direction to the frictional forces that are attempting to
stop the vehicle.

The stopping sight distance is given by the following formula;

V2
d = (0.278)(t )(V ) +
(254( f + g / 100))
where,
d = distance (metres)
t = driver reaction time, generally taken to be 2.5 seconds
V = initial speed (km/h)
f = coefficient of friction between tyres and roadway (see Table 7.2)
g = gradient of road as a percentage (downhill is negative)

On a flat road the value of g is zero. On a 5 percent downhill gradient the stopping distance
at 120 km/hr is typically 16 percent longer. At a 10 percent gradient it is nearly 40 percent
longer, as shown in Table 7.2. The Table also shows that for speeds above 50 km/hr, the
gradient of the road makes a significant difference and must be taken into account in
establishing safe sight distances.

Page 7-2 Ethiopian Roads Authority

 Chapter 8
Geometric Design Manual – 2013 Horizontal Alignment

Table 7-2: Stopping and Passing Sight Distances for Paved Roads
Minimum Passing Sight
Design Coefficient Stopping Sight Distance (m)
Passing Sight Distance to allow
Speed of Friction
Distance (m) manoeuvre to be
(km/h) (f) g=0 g = 5% g = 10% (from formulae) aborted (m)
20 .42 18 18 19 160 -
25 .41 23 24 25 190 50
30 .40 30 32 33 220 80
40 .37 45 47 50 285 135
50 .35 65 70 75 350 180
60 .33 85 90 105 415 230
70 .315 110 120 140 480 270
80 .305 140 155 180 545 310
85 .295 155 175 205 575 330
90 .29 170 195 230 610 345
100 .285 210 240 285 675 375
110 .28 245 285 340 740 405
120 .28 285 330 400 805 425

The coefficient of friction values shown in Table 7.2 have been determined from test
results such as those shown in Figure B.1 of Appendix B, using the lowest results of the
friction tests. The values shown in the main third column of Table 7.2 for minimum
stopping sight distance are calculated from the above formula.

Table 7.3 is similar to Table 7.2 but is for unpaved roads where the coefficients of friction
are lower and much more variable, depending on the properties of the gravel or soil.

It is important to note that the values in the Tables are for dry weather conditions. Stopping
sight distances are much longer in unfavourable wet conditions.

Full adherence to the required sight distances is essential for safety reasons. On the inside
of horizontal curves it may be necessary to remove trees, buildings or other obstacles to
obtain the necessary sight distances. If this cannot be done, the alignment must be changed.
In rare cases where it is not possible and a change in design speed is necessary, adequate
and permanent signage must be provided.

7.2.1 Stopping Sight Distances for Trucks

Trucks with conventional braking systems require longer stopping distances from a given
speed than do passenger cars. However, a truck driver is able to see the vertical features of
the obstruction from substantially further because of the higher driver eye height. This is
particularly important on crest curves (Chapter 9) where object height and the driver’s eye
height dictate the sight distance available. In addition, posted speed limits for trucks are
often considerably lower than for passenger vehicles.

Ethiopian Roads Authority Page 7-3

Chapter 8
Horizontal Alignment Geometric Design Manual – 2013

Separate stopping sight distances for trucks and passenger cars are, therefore, not generally
used in highway design. However, there is evidence that the sight distance advantage
provided by the higher driver eye level in trucks does not always compensate for their
inferior braking. Some reasons for the longer truck braking distances include:
• Poor braking characteristics of empty trucks. The problem relates to the suspension
and tyres that are designed for maximum efficiency under load;
• Uneven load between axles;
• Propensity of truck drivers not to obey posted speed limits;
• Inefficient brakes of articulated trucks, and
• Effect of curvature where some of the friction available at the road/tyre interface is
used to hold the vehicle in a circular path.

To balance between the costs and benefits in designing for trucks, truck stopping sight
distances should be checked at potentially hazardous locations. In general, the deceleration
rate for trucks is 1.5 m/s2 which is about half that of cars and is equivalent to a coefficient
of friction of half that shown in Table 7.2 and 7.3. This increases the stopping distance by
40% at lower speeds, increasing to 70% at 120 km/h. Where required (e.g. for vertical
alignment – see Chapter 9) the driver's eye height is taken as being at 1.8 m and the object
height is as defined in Tables 9.1 and 9.2.

Table 7-3: Stopping Sight Distances for Unpaved Roads

Design Speed Coefficient Stopping Sight Distance (m)

(km/h) of Friction (f)
g=0 g = 5% g = 10%
20 .34 19 19 21
25 .33 25 25 30
30 .32 30 35 37
40 .30 50 55 60
50 .28 70 80 90
60 .26 95 110 130
70 .25 125 145 175
80 .24 160 185 235
85 .24 180 210 270
90 .23 200 240 305
100 .23 240 290 375

7.3 Stopping Sight Distance for Single Lane Roads (Meeting Sight Distance)

For single lane roads, adequate sight distances must be provided to allow vehicles
travelling in the opposite direction to see each other and to stop safely if necessary. This
distance is normally set at twice the stopping sight distance for a vehicle that is stopping to
avoid a stationary object in the road. An extra safety margin of 20-30 metres is also
sometimes added. Although a vehicle is a much larger object than is usually considered
when calculating stopping distances, these added safety margins are used partly because of
the very severe consequences of a head-on collision and partly because it is difficult to

Page 7-4 Ethiopian Roads Authority

 Chapter 8
Geometric Design Manual – 2013 Horizontal Alignment

judge the speed of an approaching vehicle, which could be considerably greater than the
design speed. However, single lane roads have a relatively low design speed, hence
meeting sight distances should not be too difficult to achieve.

7.4 Intersection Sight Distance

Intersection sight distance is similar to stopping sight distance, Table 7.2, except that the
object being viewed is another vehicle that may be entering the road from a side road or
crossing the road at an intersection. On straight sections of road many vehicles will exceed
the road’s design speed but, being straight, sight distances should be adequate for vehicles
that are travelling straight through the junction on the major road. The situation is quite
different for vehicles that may need to slow down or stop at the junction. This is because
the time required to accelerate again and then to cross or turn at the junction is now much
greater hence longer sight distances are required. This topic is dealt with in Section 11.3
and summarised in Tables 11.1 and 11.2. Further details are also provided in Appendix C.

7.5 Decision Sight Distance

Stopping sight distances are usually sufficient to allow reasonably competent and alert
drivers to stop under ordinary circumstances. However, these distances are often
inadequate when:
• A situation arises that does not require an emergency stop but drivers need to make
complex decisions;
• Information is difficult to perceive; or
• Unexpected or unusual manoeuvres are required.

Limiting sight distances to those provided for stopping may also preclude drivers from
performing evasive manoeuvres, which are often less hazardous and otherwise preferable
to an emergency stop. Even with an appropriate complement of standard traffic control
devices, stopping sight distances may not provide sufficient visibility for drivers to
perceive and understand complex situations and to perform the necessary manoeuvres.

It is evident that there are many locations such as exits from freeways, or where lane shifts
or weaving manoeuvres are performed, where it would be prudent to provide longer sight
distances. In these circumstances, decision sight distance provides the greater length that
drivers need. If the driver can see what is unfolding far enough ahead, he or she should be
able to handle almost any situation. Decision sight distance, sometimes termed
‘anticipatory sight distance’, is the distance required for a driver to:
• detect an unexpected or otherwise ‘difficult-to-perceive’ information source or
hazard in a roadway environment that may be visually cluttered;
• recognize the hazard or its potential threat;
• select an appropriate speed and path; and
• initiate and complete the required safety manoeuvre safely and efficiently.

Because decision sight distance gives drivers additional margin for error and affords them
sufficient length to manoeuvre their vehicles at the same or reduced speed rather than to
just stop, it is substantially longer than stopping sight distance. Drivers need decision sight
distances whenever there is likelihood for error in information reception, decision-making,

Ethiopian Roads Authority Page 7-5

Chapter 8
Horizontal Alignment Geometric Design Manual – 2013

or control actions. Critical locations where these kinds of errors are likely to occur, and
where it is desirable to provide decision sight distance include:
• Approaches to interchanges and intersections;
• Changes in cross-section such as at toll plazas and lane drops;
• Design speed reductions; and
• Areas of concentrated demand where there is likely to be ‘visual noise’, e.g. where
sources of information, such as roadway elements, opposing traffic, traffic control
devices, advertising signs and construction zones, compete for attention.

The minimum decision sight distances that should be provided for specific situations are
shown in Table 7.4. If it is not feasible to provide these distances because of horizontal or
vertical curvature or if relocation is not possible, special attention should be given to the
use of suitable traffic control devices for advance warning.

Although a sight distance is suggested for the left side exit, the designer should bear in
mind that exiting to the left on a main road is in conflict with driver expectancy and is
highly undesirable. The only reason for providing this value is to allow for the possibility
that a left side exit has to be employed.

Table 7-4: Decision Sight Distances (metres)

Situations
Design Interchanges. Lane drop, Lane shift. Intersections.
Speed Sight distance to nose merge. Sight distance Sight
km/h Sight distance to to beginning distance to
Right exit Left exit taper area of shift turn lane
50 NA NA 150 85 150
60 200 275 200 100 200
80 250 340 250 150 250
100 350 430 350 200 350
120 400 500 400 250 400

7.6 Control of Sight Distance

Sight distances should be checked during design and adjustments made to meet the
minimum requirements. The following values should be used for the determination of sight
lines. Details of crest and sag curve design are to be found in Chapter 9.
a) Driver's eye height: 1.05 metres
b) Object height for stopping sight distance: 0.2 metres
c) Object height for passing sight distance: 1.30 metres
d) Object height for decision sight distance 0.00 metres

On the inside of horizontal curves, it may be necessary to remove buildings, trees or other
sight obstructions or widen cuts on the insides of curves to obtain the required sight
distance (see Figure 7.1).

Page 7-6 Ethiopian Roads Authority

 Chapter 8
Geometric Design Manual – 2013 Horizontal Alignment

Figure 7-1: Sight Distance for Horizontal Curves

Relevant formulae are as follows:

Length of Sight Line (S) = 2R sin(∆/2) where ∆ = Deflection angle (o)
Length of Middle Ordinate (M) = R(1-cos(∆/2)

Example:
Radius = 1000 metres, ∆ = 20o;
S = 2R sin(∆/2) M = R (1 – cos(∆/2)

= 2(1000)(sin(10o) = 1000(1- cos(10o)

= 347 metres = 15.2 metre

The available sight distance needs to be checked separately for both stopping and passing
sight distance, for each direction of travel.

7.7 Passing Sight Distance

The Passing Sight Distance is the minimum sight distance on a two-way road that must be
available to enable the driver of one vehicle to pass another vehicle safely without
interfering with the speed of an oncoming vehicle travelling at the design speed. Hence
factors affecting the safe sight distances required for overtaking are complicated because
they involve the capability of a vehicle to accelerate and the length and speed of the
vehicle being overtaken. Assumptions also need to be made about the speed differential
between the vehicle being overtaken and the overtaking vehicle. In view of all these
assumptions many road authorities have simply based their standards on empirical
evidence.

Ethiopian Roads Authority Page 7-7

Chapter 8
Horizontal Alignment Geometric Design Manual – 2013

Within the sight area, the terrain should be at the same level or a level lower than the
roadway, otherwise, for horizontal curves, it may be necessary to remove obstructions and
widen cuttings on the insides of curves to obtain the required sight distance. Care must be
exercised in specifying passing/no-passing zones in areas where the sight distance may be
obscured in the future due to vegetative growth.

The passing sight distance is generally determined by a formula with four components, as
follows:
d1 = initial manoeuvre distance, including a time for perception and reaction
d2 = distance during which passing vehicle is in the opposing lane
d3 = clearance distance between vehicles at the end of the manoeuvre
d4 = distance traversed by the opposing vehicle
The formulae for these components are as indicated below:

d1 = 0.278 t1 (v – m + a.t1/2)

Where
t1 = time of initial manoeuvre, s
a = average acceleration, km/h/s
v = average speed of passing vehicle, km/h
m = difference in speed of passed vehicle and passing vehicle, km/h

d2 = 0.278 v.t2.

Where
t2 = time passing vehicle occupies left lane, s
v = average speed of passing vehicle, km/h
d3 = safe clearance distance between vehicles at the end of the manoeuvre, and is
dependent on ambient speeds as per Table 7.5:
d4 = distance traversed by the opposing vehicle, which is approximately equal to d2
minus the portion of d2 whereby the passing vehicle is entering the left lane,
estimated as:
d4 = 2d2/3

Table 7-5: Clearance Distance (d3) for Different Ambient Speeds

Speed Group (km/h) 50-65 66-80 81-100 101-120
d3 (m) 30 55 80 100

The time t1 for the initial perception and manoeuvre is particularly variable and also
depends on the vehicle speed – more time is required for assessment when the overtaking
manoeuvre is considered to be dangerous.

The minimum Passing Sight Distance (PSD) for design is

PSD = d1 + d2 + d3 + d4

Page 7-8 Ethiopian Roads Authority

 Chapter 8
Geometric Design Manual – 2013 Horizontal Alignment

The resulting minimum sight distances for passing are as indicated in the sixth main
column of Table 7.2. For the reasons outlined herein, however, the preferable (or desirable)
PSD at 40km/h is 15% greater than the values quoted in the Table, rising to 40% greater at
120km/h.

A method of measuring and recording sight distances on plans is given in Appendix C.

Figure 7-2: Passing Sight Distance

7.8 Minimum Provision of Passing Sight Distance

An alternative design strategy is to base the passing sight distances on providing enough
sight distance for a vehicle to safely abort a passing manoeuvre if another vehicle is
approaching. The recommended values are shown in the seventh column of Table 7.2.

Sight distance records are also useful on two-lane highways for determining the percentage
of length of highway on which sight distance is restricted to less than the minimum needed
for passing. This is important in evaluating capacity. With recorded sight distance, as in the
lower part of Figure C-1 of Appendix C, it is a simple process to determine the percentage
of length of highway with a given sight distance or greater.
Passing Sight Distance is a desirable requirement for two-way single roadway roads.
Sufficient visibility for passing increases the capacity and efficiency of a road and should
be provided for as much of the road length as possible within financial limitations.
Table 7.6 gives guide values for the extent to which passing sight distance should be
provided.
Table 7-6: Guide Values for the Minimum Provision of Passing Sight Distance
Percent Passing Opportunity and Terrain
Design
Standard Urban/Peri-
Flat Rolling Mountainous Escarpment
Urban
DC8 50
50 25
DC7 33
DC6
25 25 0 20
DC5
15
DC4
20 20
DC3

Ethiopian Roads Authority Page 7-9

Chapter 8
Horizontal Alignment Geometric Design Manual – 2013

8 HORIZONTAL ALIGNMENT
8.1 General

The horizontal alignment consists of a series of straight sections (tangents), circular curves,
transition curves (spirals) and super-elevation. The horizontal curves are designed to
ensure that vehicles can negotiate them safely. The alignment design should be aimed at
avoiding sharp changes in curvature, thereby achieving a safe uniform driving speed.
Transition curves between straight sections of road and circular curves whose radius
changes continuously from infinity (tangent) to the radius of the circular curve (R) are used
to reduce the abrupt introduction of centripetal acceleration that occurs on entering the
circular curve. They are not required when the radius of the horizontal curve is large and
are normally not used on the lower classes of road. In Ethiopia their use is confined to
roads where the design speed is 80km/hr or greater.

In order for a vehicle to move in a circular path, an inward radial force is required to
provide the necessary centripetal acceleration or, in other words, to counteract the
centrifugal force. This radial force is provided by the sideways friction between the tyres
and the road surface assisted by the cross-fall or super-elevation.

In order to calculate the minimum horizontal radius of curvature, Rmin, for a particular
design speed, the equation is:

VD 2
Rmin =
127(e + f )
Where
VD = design speed (km/h)
e = maximum super-elevation (%/100)
f = side friction coefficient

The design speed is thus one of the main design parameters. Values for each class of road
under each of its operating conditions are shown in Table 2.1.

Values of the minimum radii of curvature for different design speeds and super-elevations
based on this formula are shown in Tables 8.1 for paved roads and Table 8.2 for unpaved
roads. For unpaved roads the super-elevation will not be constant but will vary as the
gravel is worn away. A value of 4 percent has been used in the calculations as a reasonable
compromise during the life of the gravel surface, assuming an initial maximum value of 6
percent.

Side friction coefficients are dependent on

i) vehicle speed;
ii) type, condition and texture of roadway surface;
iii) weather conditions; and
iv) type and condition of tyres.

The coefficient is considerably less than the longitudinal friction coefficient. Its value
decreases as speed increases but there is considerable disagreement about representative
values, especially at the lower speeds. Tables 8.1 and 8.2 were developed based on the

Page 8-10 Ethiopian Roads Authority

 Chapter 8
Geometric Design Manual – 2013 Horizontal Alignment

results of several studies (Appendix B). For paved roads the coefficient ranges from
between 0.18 and 0.25 at 20km/h down to between 0.09 and 0.16 at 120km/h.

Table 8-1: Minimum Radii for Horizontal Curves for Paved Roads
Design speed
20 25 30 40 50 60 70 80 85 100 120
(km/h)
Side Friction
0.23 0.22 0.21 0.19 0.17 0.16 0.14 0.13 0.12 0.11 0.10
Factor (f)
Super-elevation
15 19 30 55 95 145 215 300 350 515 780
= 4%
Super-elevation
15 18 27 50 85 135 195 270 310 455 685
= 6%
Super-elevation
15 17 25 50 80 120 175 240 280 410 610
= 8%
Super-elevation
15 16 25 45 75 110 160 220 255 375 555
= 10%

Table 8-2: Minimum Radii for Horizontal Curves for Unpaved Roads
Design speed
20 25 30 40 50 60 70 80 85 90 100
(km/h)
Side Friction 0.16
0.19 0.17 0.15 0.14 0.12 0.11 0.10 0.10 0.10 0.09
Factor 5
Super-elevation =
15 25 35 65 115 175 255 355 410 475 610
4%

For unpaved roads the friction is usually considerably less. In these calculations it has been
assumed that it is 80% of the value for paved roads but this is dependent on a tightly knit
and dry surface of good quality gravel with no loose stones; in other words a surface on
which the design speed could be maintained. A poorly bound surface with many loose
particles has a very low value of friction and it has to be assumed that vehicles will be
driven on such a surface at a speed that is much lower than the nominal design speed
dictated by the sight distances and radii of curvature.

The Tables above indicate the minimum radii of curvature for different design speeds and
road surfaces. In general, these radii should be used only under the most critical conditions.
The deviation angle of each curve should be as small as the physical conditions permit.
The deviation should be absorbed in the flattest possible curve so that passing
opportunities are not unduly restricted.

Changes in design speed (see Section 5.4), if required due to changes in terrain class,
should not be made abruptly but over sufficient distance to enable drivers to change speed
gradually. The change in design speed should not be greater than one design speed step
(usually 10 or 15 km/h) and the section with the lower geometric standards should be long
enough to be clearly recognizable by drivers and not, for example, just a single curve.

The physical terrain sometimes changes by two terrain classes, typically from mountainous
to flat. Where possible a transition section of road should be provided with limiting
parameters equivalent to the intermediate terrain type namely rolling terrain. Where this is

Ethiopian Roads Authority Page 8-11

Chapter 8
Horizontal Alignment Geometric Design Manual – 2013

not possible, adequate warning signs must be provided to alert drivers to the changes in
geometric standards.

Under normal circumstances sections of road will contain many curves that are larger than
the minimum radii specified in the design standards. For reasons of safety and driver
comfort it is inadvisable for consecutive curves to differ in radius by a large amount even
though they are both greater than the minimum. Indeed, all the various design elements
must be combined in a balanced way, avoiding the application of minimum values for one
or a few elements at a particular location when other elements are considerably above the
minimum requirements. Figure 8.1 shows the required ratio of radii for consecutive curves.
Consecutive horizontal curves are defined as curves where the distance between the end of
one and the beginning of the next is less than the radius of the larger curve. The best result
will be achieved when the two radii are similar (labelled ‘very good’ in the diagram). If the
ratio of radii falls into the ‘useable’ category some discomfort or inconvenience will be felt
because of the increase in centripetal force when entering the tighter curve.

Figure 8-1: Ratio of radii of consecutive horizontal curves

8.2 Cross-fall

For both paved and unpaved roads there are constraints on the maximum cross-fall, as
summarised in Tables 2.6 to 2.17. These constraints translate directly into minimum values
of horizontal radii of curvature.

Page 8-12 Ethiopian Roads Authority

 Chapter 8
Geometric Design Manual – 2013 Horizontal Alignment

8.3 Elements of a Circular Curve

The elements of circular curves are shown in Figure 8.2

Figure 8-2: Curve Elements

1) ∆ is the Deflection Angle (in degrees).

2) R is the Radius of the curve.

3) T is the Tangent Distance PC to PI; T = R.tan(∆/2)

4) E is the External Distance: E = R.[sec(∆/2) – 1]

5) L is the Curve Length. L = 2.π.R.∆/360

6) M is the Middle Ordinate M = R.[1 – cos(∆/2)]

7) C is the Chord Length from PC to PT C = 2.R.sin(∆/2)

8) Point PC is the Point-of-Curvature Station PC = PI – T

9) PT is the Point-of-Tangency Station PT = PC+ L

Example:
A curve has a deflection angle of ∆ = 23o 18’ 02”, and a radius of 1432.6m. The Point of
Intersection (PI) is 5+053.87. Calculate the tangent distance (T), external distance (E),
curve length (L), Point of Curvature (PC), and Point of Tangent (PT).

∆ 23°18'02'
T = R tan = (1432.6) tan = 1432.4(.2026) = 295m
2 2
Chapter 8
Horizontal Alignment Geometric Design Manual – 2013

R × 2π 1432.6 × 2 × 3.14
L = ∆× = 23.3 = 582m
360 360

 
∆  1 
E = R sec − 1 = 1432.6  − 1 = 1432.4(.02103) = 30m
2 ∆
 cos 
 2

PC = PI – T = 5+053.87 – 295.35 = 4 + 758

PT = PC + L = 4 + 758.49 + 582.51 = 5+341

8.4 Minimum Turning Radii

Buses, trucks, trucks with trailers and 4x4 utility vehicles require minimum design turning
radii of 12.8m, 13.7m and 7.3m respectively (Table 5.1). It is not possible to exclude any
of these vehicle categories from the lower standard roads and, as a certain amount of
tolerance is required for safe operations, the minimum horizontal curve radius of 15m is
specified in Tables 8.1 and 8.2 for all design standards.

For reasons of safety and ease of driving, curves near the minimum radius for the design
speed should not be used at the following locations;
i) On high fills, because the lack of surrounding features reduces a driver’s
perception of the alignment.
ii) At or near vertical curves (tops and bottoms of hills) because the unexpected
bend can be extremely dangerous, especially at night.
iii) At the end of long tangents or a series of gentle curves, because actual speeds
will exceed design speeds.
iv) At or near intersections and approaches to bridges or other water crossing
structures.

8.5 Isolated Curves

The horizontal curvature over a particular road section should be as consistent as possible.
Long tangent roadway segments joined by an isolated curve designed at or near the
minimum radius are unsafe. Long straight sections encourage drivers to drive at speeds in
excess of the design speed, hence sudden and unexpected sharp curves are dangerous.
Good design practice is to avoid the use of minimum standards in such conditions. For
isolated curves, the minimum horizontal curve radius as shown in Tables 2.6 through 2.16
should be increased by 50 percent. This will usually result in the ability to negotiate the
curve at a speed approximately 10 km/hr higher than the design speed.

8.6 Length of Tangent Sections and Curves

There are conflicting views about curve lengths. One school of thought maintains that the
horizontal alignment should maximise the length of road where adequate sight distances
are provided for safe overtaking. Overtaking is difficult on curves of any radius and hence

Page 8-14 Ethiopian Roads Authority

 Chapter 8
Geometric Design Manual – 2013 Horizontal Alignment

the length of curved road should be minimised. This requires curve radii to be relatively
close (but not too close) to the minimum for the design speed to maximise the length of
straight sections.

The alternative view is that very long straight sections should be avoided. Long tangent
sections increase the danger from headlight glare and usually lead to excessive speeding.
For example, a long tangent causes speeds to creep up to about 120 km/h or even higher
and the driver then has to reduce speed to negotiate the following curve, thereafter
accelerating again. Ideally, drivers should be encouraged to maintain a speed which is
close to the design speed to reduce the possibility of an error of judgment leading to an
accident. It has been found that a maximum tangent length, measured in metres, of 20
times the design speed in km/h, achieves this effect. For example, a design speed of 100
km/h suggests that tangents should, ideally, not be longer than about 2.0 km. Thus a safer
alternative is obtained by a winding alignment with tangents deflecting 5 to 10 degrees
alternately from right to left. Straight sections should have lengths (in metres) less than 20
x design speed in km/h. Such ‘flowing’ curves restrict the view of drivers on the inside
carriageway and reduce safe overtaking opportunities, therefore such a winding alignment
should only be adopted where the straight sections are very long. In practice this only
occurs in very flat terrain. The main aspect is to ensure that there are sufficient
opportunities for safe overtaking and therefore, provided the straight sections are long
enough, a semi-flowing alignment can be adopted at the same time. If overtaking
opportunities are infrequent, maximising the length of the straight sections is the best
option.

For small changes of direction it is often desirable to use a large radius of curvature. This
improves the appearance and reduces the tendency for drivers to cut corners. In addition, it
reduces the length of the road segment and therefore the cost of the road provided that no
extra cut or fill is required. A widely adopted guideline is that, on minor roads, curves
should have a minimum length of 150 metres for a deflection angle of 5o and that this
length should be increased by 30 metres for every 1o decrease in deflection angle. On
major roads and freeways, the minimum curve length in metres should be three times the
design speed in km/h. The increase in length for decreasing deflection angle also applies to
these roads. In the case of a circular curve without transitions, the length in question is the
total length of the arc and, where transitions are applied, the length is that of the circular
curve plus half the total length of the transitions.

The minimum length of tangent must allow for the run-off of the super-elevation of the
preceding curve followed by the development of that for the following curve. This distance
should be calculated during detailed design but, as a rule of thumb, a tangent length of less
than 200 m is likely to prove inadequate.

8.7 Reverse Curves, Broken-Back Curves, and Compound Curves

Curves are more frequent in rugged terrain. Tangent sections are shortened, and a stage
may be reached where successive curves can no longer be dealt with in isolation. Three
cases of successive curves are (see Figure 8.3):
i) Reverse Curve: a curve followed by another curve in the opposite direction.
ii) Broken-back Curve: a curve followed by another curve in the same direction but
with only a short tangent in-between.

Ethiopian Roads Authority Page 8-15

Chapter 8
Horizontal Alignment Geometric Design Manual – 2013

iii) Compound curve: curves in the same direction but of different radii, and without
any intervening tangent section.
The occurrence of abrupt reverse curves (having a short tangent between two curves in
opposite directions) should be avoided. Such geometrics make it difficult for the driver to
remain within his lane. It is also difficult to super-elevate both curves adequately, and this
may result in erratic operation.

The broken-back arrangement of curves (having a short tangent between two curves in the
same direction) should also be avoided except where very unusual topographical or right-
of way conditions dictate otherwise. Drivers do not generally anticipate successive curves
in the same direction hence safety is compromised. Problems also arise associated with
super-elevation and drainage.

The use of compound curves affords flexibility in fitting the road to the terrain and other
controls. Caution should, however, be exercised in the use of compound curves because the
driver does not expect to be confronted by a change in radius once he has entered a curve,
hence safety is compromised. Their use should also be avoided where curves are sharp.

Compound curves with large differences in curvature introduce the same problems as are
found at the transition from a tangent to a small-radius curve. Where the use of compound
curves cannot be avoided, the radius of the flatter circular arc should not be more than 50
percent greater than the radius of the sharper arc; ie. R1 should not exceed 1.5.R2. A
compound arc on this basis is suitable as a form of transition from either a flat curve or a
tangent to a sharper curve, although a spiral transition curve is preferred (see Section 8.10).

Figure 8-3: Reverse Curves, Broken-Back Curves, and Compound Curves

8.8 Widening on Curves and Embankments

The use of long curves of tight radii should be avoided where possible because drivers
following the design speed will find it difficult to remain in the traffic lane. Widening of

Page 8-16 Ethiopian Roads Authority

 Chapter 8
Geometric Design Manual – 2013 Horizontal Alignment

the carriageway where the horizontal curve is tight is usually necessary to ensure that the
rear wheels of the largest vehicles remain on the road when negotiating the curve; and, on
two lane roads, to ensure that the front overhang of the vehicle does not encroach on the
opposite lane. Widening is therefore also important for safety reasons.

Vehicles need to remain centred in their lane to reduce the likelihood of colliding with an
oncoming vehicle or driving on the shoulder. Sight distances should be maintained as
discussed above.

Widening on high embankments is recommended for design classes DC8 through to DC4.
The steep drops from high embankments unnerve some drivers and the widening is
primarily for psychological comfort although it also has a positive effect on safety.
Widening for curvature and for high embankments should be added where both cases
apply.

Curve widening is required on all standards of roads and should be sufficient to cater for
the design vehicle. Table 8.3 gives the values to be adopted in the design.

Table 8-3: Widening on Curves and High Fills

Radius of Curve Widening: Curve Widening: Fill Widening
Curve Single Lane Two Lanes Height of fill Amount
(m) (m) (m) (m) (m)
>250 0.0 0.0 0.0-3.0 0.0
120- 250 0.0 0.6 3.0- 6.0 0.3
60-120 0.0 0.9 6.0 - 9.0 0.6
40-60 0.6 1.2 Over 9.0 0.9
20-40 0.6 1.5 Over 9.0 0.9
<20 See Section 8.10: Switchbacks

Curve widening is generally not applied to curves with a radius greater than 250 metres
regardless of the design speed or the lane width. Widening should transition gradually on
the approaches to the curve so that the full additional width is available at the start of the
curve. Although a long transition is desirable to ensure that the whole of the travelled way
is fully usable, this results in narrow pavement slivers that are difficult, and
correspondingly expensive, to construct. In practice, curve widening is thus applied over
no more than the length of the super-elevation runoff preceding the curve. For ease of
construction, the widening is normally applied only on one side of the road. This is usually
on the inside of the curve to match the tendency for drivers to cut the inside edge of the
travelled way.

The height of fill is measured from the edge of the shoulder to the toe of the slope.

Widening is provided to make driving on a curve comparable with that on a tangent. On

older roads with narrow cross-sections and low design speeds and hence sharp curves,
there was a considerable need for widening on curves. Because of the inconvenience
attached to widening the surfacing of a lane, it follows that the required widening may not
always have been provided. Where a road has to be rehabilitated and it is not possible to
increase the radius of curvature, the designer should consider the need for curve widening.

Ethiopian Roads Authority Page 8-17

Chapter 8
Horizontal Alignment Geometric Design Manual – 2013

8.9 Switchback Curves

Switchback or hairpin curves are used where necessary in traversing mountainous and
escarpment terrain. Employing a radius of 20m or less, with a minimum of 10m, they are
generally outside of the standards for all road designs and are specified using the
guidelines listed in the Departure from Standards section in Chapter 2.

Switchback curves require careful design to ensure that all design vehicles can travel
through the curve. They must therefore provide for the tracking widths of the design
vehicles, as indicated in Figures 5.2 through to 5.4. These figures show that the minimum
outer radii for design vehicles DV2 through DV4 are 12.5m, 14.1m, and 12.5m,
respectively. Minimum inner radii are 8m, 7.4m, and 6m, respectively.

Switchback requirements can be determined which allow for:

• Passage of two opposing DV4 vehicles. This is recommended for Design
Standards DC8, DC7 and DC6
• Passage of a single DV4 and a DV1. This is recommended for Design Standards
DS4- DC5
• Passage of only a single DV4. This is recommended for Design Standards DC3,
DC3, DC2 and DC1.

Figure 8.4 illustrates a switchback curve.

Figure 8-4: Switchback Curve

Page 8-18 Ethiopian Roads Authority

 Chapter 8
Geometric Design Manual – 2013 Horizontal Alignment

For a design example, consider road standard DC3 which allows for only the passage of a
single DV4 vehicle. By superimposing Figure 3.11 for design vehicle DV4 over Figure 8.4
at the same scale, it can be shown that the requirements are (see Figure 8.5):

R = 10m Ri = 6m Rs = 14m

Thus although the normal carriageway width for a paved DC3 is 6.0m, at the switchback
curve 8m is required.

Figure 8-5: Switchback Curve for the Passage of Single DV4 Vehicle

Requirements vary depending on passage requirements, radius, deflection angle, and

design standard, and a template should be used based on the turning radii of the design
vehicle to ensure that the vehicles can negotiate each switchback.

It is important to provide relief from a severe gradient through the switchback. Gradient
parameters associated with a switchback curve are discussed in Section 9.5.

8.10 Transition Curves

The characteristic of a transition curve is that it has a constantly changing radius.

Transition curves may be inserted between tangents and circular curves to reduce the
abrupt introduction of lateral acceleration and therefore to enhance safety. They may also
be used between two circular curves.

For large radius curves, the rate of change of lateral acceleration is small and transition
curves are not normally required. It can also be argued that transition curves are not a
requirement for certain roads, particularly those of lower classification. Another possible

Ethiopian Roads Authority Page 8-19

Chapter 8
Horizontal Alignment Geometric Design Manual – 2013

strategy would be to consider transitions for roads where a significant portion of the curves
have a super-elevation in excess of 60 percent of the maximum super-elevation. For
Ethiopian roads, transition curves are a requirement for trunk and link road segments
having a design speed of equal to or greater than 80 km/hr. They are also required if the
radius of the circular curve is less than the values shown in Table 8.4.

Table 8-4: Transition Curve Requirements

Transition Required if
Design Speed (km/hr)
Radius of Curve is less than:
80 380
85 428
90 480
100 590
110 720
120 850

If a transition curve is required, the Euler spiral, which is also known as the clothoid,
should be used. The radius varies from infinity at that tangent end of the spiral to the radius
of the circular arc at the circular curve end. By definition, the radius at any point of the
spiral varies inversely with the distance measured along the spiral.

In the case of a combining spiral connecting two circular curves having different radii,
there is an initial radius rather than an infinite value.

8.11 Super-elevation

A tighter curve can be designed if higher values of super-elevation are used, but high
values of super-elevation are not recommended if the friction is low, such as in locations
where mud is likely to contaminate the road surface regularly. High values are also not
recommended where mixed traffic and/or roadside development severely limit the speed of
vehicles. In urban areas an upper limit of 4 percent should be used. Similarly, either a low
maximum rate of super-elevation or no super-elevation is employed within important
intersection areas or where there is a tendency to drive slowly because of turning and
crossing movements, warning devices, and signals.

Super-elevation is, however, a requirement for all standards of roads.

8.12 Super-elevation Runoff

In alignment design with spirals, the super-elevation runoff is provided over the whole of
the transition curve. The length of runoff is the spiral length, with the tangent to spiral (TS)
transition point at the beginning and the spiral to curve (SC) transition point at the end. The
change in cross slope begins by removing the adverse cross slope from the lane or lanes on
the outside of the curve on a length of tangent just ahead of TS (the tangent runout).
Between the TS and SC (the super-elevation runoff) the travelled way is rotated to reach
the full super-elevation at the SC. This procedure is reversed on leaving the curve. By this
design the whole of the circular curve has full super-elevation, as shown in Figure 8.6.

Page 8-20 Ethiopian Roads Authority

 Chapter 8
Geometric Design Manual – 2013 Horizontal Alignment

Figure 8-6: Spiral Curve Transition.

In the design of curves without spirals the super-elevation runoff is considered to be that
length beyond the tangent runout. Empirical methods are employed to locate the super-
elevation runoff length with respect to the point of curvature (PC).

Current design practice is to place approximately two-thirds of the runoff on the tangent
approach and one-third on the curve, as shown in Figure 8.7.

Figure 8-7: Circular Curve Transition

Table 8.5 gives super-elevation rates and length of runoff for horizontal curves at different
speeds for a maximum super-elevation of 8 percent. Table 8.6 gives super-elevation rates
and length of runoff for horizontal curves at different speeds for a maximum super-
elevation rate of 4 percent.

8.13 Shoulder Super-elevation

Figure 8.8 depicts shoulder super-elevation rates corresponding to carriageway super-

elevation rates. The figure shows that on the low side (inner shoulder) of super-elevated
curves, the shoulder super-elevation matches the roadway super-elevation. On the high
side (outer shoulder), the super-elevation is set such that the grade break between the

Ethiopian Roads Authority Page 8-21

Chapter 8
Horizontal Alignment Geometric Design Manual – 2013

roadway and the shoulder is 8 percent. An exception to this occurs at a maximum super-
elevation of 8 percent, where the resultant shoulder super-elevation would be an
undesirable flat configuration. Here the super-elevation is set at -1% to drain the shoulder.

Note: For design classes DC5 and lower the shoulder may be sloped with the carriageway, but the
shoulder should then be surfaced on the outside of the curve.

Figure 8-8: Shoulder Super-elevation (for Surfaced Roads)

Page 8-22 Ethiopian Roads Authority

 Chapter 8
Geometric Design Manual – 2013 Horizontal Alignment

Table 8-5: Super elevation Rates and Length of Run-Off: 8%

Vd=30km/h Vd=40 km/h Vd=50 km/h Vd =60 km/h Vd=70 km/h Vd=85 km/h Vd=100 km/h Vd= 120 km/h
R e e e e e e e e
(m) (%) L(m) (%) L (m) (%) L (m) (%) L(m) (%) L(m) (%) L(m) (%) L(m) (%) L(m)

7000 NC 0 NC 0 NC 0 NC 0 NC 0 NC 0 NC 0 NC 0
5000 NC 0 NC 0 NC 0 NC 0 NC 0 NC 0 NC 0 NC 0
3000 NC 0 NC 0 NC 0 NC 0 NC 0 NC 0 RC 56 2.4 101
2500 NC 0 NC 0 NC 0 NC 0 NC 0 RC 47 2.1 56 2.9 101
2000 NC 0 NC 0 NC 0 NC 0 NC 0 2.2 47 2.6 56 3.5 101
1500 NC 0 NC 0 NC 0 NC 0 RC 39 2.5 47 3.4 56 4.6 101
1400 NC 0 NC 0 NC 0 RC 33 2.1 39 2.6 47 3.6 56 4.9 101
1300 NC 0 NC 0 NC 0 RC 33 2.2 39 2.8 47 3.8 56 5.2 101
1200 NC 0 NC 0 NC 0 RC 33 2.4 39 3.0 47 4.1 56 5.6 101
1000 NC 0 NC 0 RC 28 2.2 33 2.8 39 3.5 47 4.8 56 6.5 101
900 NC 0 NC 0 RC 28 2.4 33 3.1 39 4.2 47 5.2 56 7.1 101
800 NC 0 NC 0 RC 28 2.7 33 3.4 39 4.6 47 5.7 56 7.6 103
700 NC 0 RC 22 2.2 28 3.0 33 3.8 39 5.1 47 6.3 56 8.0 108
600 NC 0 RC 22 2.6 28 3.4 33 4.3 39 6.5 47 6.9 56 Rmin = 665
500 NC 0 2.2 22 3.0 28 3.9 33 4.9 39 7.2 47 7.8 56
400 RC 17 2.7 22 3.6 28 4.7 33 5.7 39 7.8 51 8.0 64
300 2.1 17 3.4 22 4.5 28 5.6 34 6.7 44 8.0 55 Rmin = 395
250 2.5 17 4.0 22 5.1 28 6.2 37 7.3 48 Rmin = 270
200 3.0 17 4.6 24 5.8 31 7.0 42 7.9 52
175 3.4 17 5.0 26 6.2 33 7.4 44 8.0 52
150 3.8 18 5.4 28 6.7 36 7.8 47 Rmin = 175
140 4.0 19 5.6 29 6.9 37 7.9 47
130 4.2 20 5.8 30 7.1 38 8.0 48
120 4.4 21 6.0 31 7.3 39 Rmin = 125
110 4.7 23 6.3 32 7.6 41
100 4.9 23 6.5 33 7.8 42
90 5.2 25 6.9 36 7.9 43
80 5.5 26 7.2 37 8.0 43 emax = 8.0%
70 5.9 28 7.5 39 Rmin = 80
60 6.4 31 7.8 40 R = radius of curve
50 6.9 33 8.0 41 V = assumed design speed
40 7.5 36 Rmin = 50
30 8.0 38
e = rate of superelevation
Rmin=30 L = minimum length of runoff(does not include tangent runout)
NC = normal crown section
RC = remove adverse crown, super-elevation at normal crown slope

Note : Lengths rounded in multiples of 10m to permit simpler calculations.

Ethiopian Roads Authority Page 8-23

Chapter 8
Horizontal Alignment Geometric Design Manual – 2013

Table 8-6: Super elevation Rates and Length of Run-Off: 4%

Vd=30km/h Vd=40 km/h Vd=50 km/h Vd =60 km/h Vd=70 km/h Vd=85 km/h Vd=100 km/h

R e e e e e e e
(m) (%) L(m) (%) L (m) (%) L (m) (%) L(m) (%) L(m) (%) L(m) (%) L(m)

7000 NC 0 NC 0 NC 0 NC 0 NC 0 NC 0 NC 0
5000 NC 0 NC 0 NC 0 NC 0 NC 0 NC 0 NC 0
3000 NC 0 NC 0 NC 0 NC 0 NC 0 NC 0 RC 56
2500 NC 0 NC 0 NC 0 NC 0 NC 0 RC 47 2.2 56
2000 NC 0 NC 0 NC 0 NC 0 NC 0 2.1 47 2.6 56
1500 NC 0 NC 0 NC 0 NC 0 RC 39 2.2 47 2.7 56
1400 NC 0 NC 0 NC 0 NC 0 RC 39 2.3 47 2.8 56
1300 NC 0 NC 0 NC 0 RC 33 RC 39 2.5 47 2.9 56
1200 NC 0 NC 0 NC 0 RC 33 RC 39 2.7 47 3.2 56
1000 NC 0 NC 0 NC 0 RC 33 2.2 39 2.9 47 3.4 56
900 NC 0 NC 0 RC 28 2.1 33 2.4 39 3.2 47 3.5 56
800 NC 0 NC 0 RC 28 2.3 33 2.5 39 3.4 47 3.7 56
700 NC 0 NC 0 RC 28 2.5 33 2.7 39 3.5 47 3.9 56
600 NC 0 RC 22 2.1 28 2.7 33 2.9 39 3.7 47 4.0 56
500 NC 0 RC 22 2.3 28 2.9 33 3.1 39 3.9 47 Rmin = 490
400 NC 0 2.1 22 2.5 28 3.3 33 3.4 39 4.0 47
300 RC 17 2.3 22 2.8 28 3.6 33 3.8 39 Rmin = 285
250 RC 17 2.6 22 3.0 28 3.8 33 3.9 39
200 2.3 17 2.8 22 3.3 28 3.9 33 Rmin = 215
175 2.4 17 2.9 22 3.5 28 4.0 33
150 2.5 17 3.1 22 3.7 28 Rmin = 150
140 2.6 17 3.2 22 3.8 28
130 2.6 17 3.3 22 3.8 28
120 2.7 17 3.4 22 3.9 28
110 2.8 17 3.5 22 4.0 28 emax = 4.0%
100 2.9 17 3.6 22 4.0 28 R = radius of curve
90 3.0 17 3.7 22 Rmin = 100
80 3.2 17 3.8 22
V = assumed design speed
70 3.3 17 3.9 22 e = rate of superelevation
60 3.5 17 4.0 22 L = minimum length of runoff(does not include tangent runout)
50 3.7 18 Rmin = 60
40 3.9 19 NC = normal crown section
Rmin=35 RC = remove adverse crown, super-elevation at normal crown slope

Note : Lengths rounded in multiples of 10m to permit simpler calculations.

Use of emax = 4.0% should be limited to urban conditions.

Source: AASHTO

Page 8-24 Ethiopian Roads Authority

 Chapter 9
Geometric Design Manual – 2013 Vertical Alignment

9 VERTICAL ALIGNMENT
9.1 Introduction

Vertical alignment is the combination of parabolic vertical curves and tangent sections of a
particular slope. The selection of rates of grade and lengths of vertical curves is based on
assumptions about characteristics of the driver, the vehicle and the roadway. Vertical
curvature may impose limitations on sight distance, particularly when combined with
horizontal curvature.

Thus the two major aspects of vertical alignment are vertical curvature, which is governed
by sight distance criteria, and gradient, which is related to vehicle performance and level of
service. This chapter describes the mathematical concepts for defining the vertical
curvature of the road; defines the limiting characteristics for each road class; recommends
maximum and minimum gradients; indicates gradient requirements through villages;
develops the criteria for incorporation of a climbing lane; and provides vertical clearance
standards.

The vertical alignment should also be designed to be aesthetically pleasing. As a general

guide, a vertical curve that coincides with a horizontal curve should, if possible, be
contained within the horizontal curve, and should ideally have approximately the same
length.

A smooth grade line with gradual changes appropriate to the class of road and the character
of the topography is preferable to an alignment with numerous short lengths of grade and
vertical curves. The 'roller coaster' or 'hidden dip' type of profile should be avoided.

9.2 Vertical Curve Formula

Vertical curves are required to provide smooth transitions between consecutive gradients.
The simple parabola is specified for these because the parabola provides a constant rate of
change of curvature and, hence, acceleration and visibility, along its length. Equations
relating the various aspects of the vertical curve are as follows (Figure 9.1):

Y(x) = r.X2/200 + X.g1/100 + YBVC Equation 9.1

r = (g2 – g1)/L = G/L

Where
BVC = Beginning of the vertical curve. The coordinates are normally (0, Y(0)),
EVC = End of the vertical curve. The coordinates are normally (L, Y(L)),
Y(X) = Elevation of a point on the curve (metres)
X = Horizontal distance from the (BVC) (metres)
g1 = Starting gradient (%),
g2 = Ending gradient (%),
r = Rate of change of grade per section (% per metre),
L = Length of curve (horizontal distance) in metres,
G = g1 - g2 (%),

Ethiopian Roads Authority Page 9-1

Chapter 9
Vertical Alignment Geometric Design Manual – 2013

K = L/G = horizontal distance required to achieve a 1% change in grade

(metres),
Z = vertical distance from the tangent to the curve (metres)

Useful relationships are;

Equation of tangent g1 is Y(X) = Y(0) + g1.X/100

Equation of tangent g2 is Y(X) = Y(L) + g2.(X-L)/100

The y coordinate of the EVC is Y(L) = (g1+g2)L/200 + Y(0)

The Intersection Point always occurs at an x coordinate of 0.5L hence the elevation is
always;

Y(IP) = (g2+3.g1)L/800 + Y(0)

Examples of crest and sag vertical curves are shown in Figures 9.1
9 and 9.2,
2, respectively.

Figure 9-1: Crest Curve

 Chapter 9
Geometric Design Manual – 2013 Vertical Alignment

101

100

Elevation (metres)
99

98

97

96

95
0 50 100 150 200 250 300
Chainage (metres)

Sag curve Tangent g1 = -3% Tangent g2 = 2%

Figure 9-2: Sag Curve

Example:
For the crest curve shown in Figure 9.1 the two tangent grade lines are +6% and -3%. The
Beginning of the Vertical Curve is at chainage 0.000 and its elevation 100.0m. The length
of the vertical curve is 400m. Compute the End of Vertical Curve and the coordinates of
the Intersection Point.

The y coordinate of the EVC is Y(L) = (g1+g2)L/200 + Y(0)

= (6 - 3).400/200 + 100.0 = 106.0
The x coordinate is X(L) = 400.0
The coordinates of the IP are X(IP) = L/2 = 200.0 and
Y(IP) = (g2+3.g1)L/800 + Y(0)
= (-3 + 3x6).400/800 + Y(0) = 107.5m

9.3 Crest curves

Two conditions exist when considering the minimum sight distance criteria on vertical
curves. The first is where the sight distance (S) is less than the length of the vertical curve
(L), and the second is where sight distance extends beyond the vertical curve.
Consideration of the properties of the parabola results in the following relationships for
minimum curve length to achieve the required sight distances:

For S < L (the most common situation in practice):

Lm = (G.S2)/[200(h10.5 + h20.5)2] and therefore

Ethiopian Roads Authority Page 9-3

Chapter 9
Vertical Alignment Geometric Design Manual – 2013

Lm = K.G
where
Lm = minimum length of vertical crest curve (metres)
S = required sight distance (metres)
h1 = driver eye height (metres)
h2 = object height (metres)
K = is a constant for given values of h1 and h2 and stopping sight distance (S)
and therefore speed and surface friction.

For S > L

Lm = 2S - [200.(h10.5 + h20.5)2]/G

Eye height (h1) has been taken as 1.05 metres, and object heights h2 of 0.2 metres and 0.6
metres above the road surface.

In choosing which object height to adopt it is useful to know that in open country there is
very little evidence of vehicles being involved in accidents over crest curves because of
small objects in the carriageway. This may, of course, be because accident records are very
imperfect in many countries but evidence from the USA and Canada supports this
conjecture. In relatively flat terrain the most likely objects are another vehicle, a human
being crossing the road and a pothole. The first two are more than 0.6m high and the third
is at road surface level. An object height equal to the road surface itself is also applicable
for sag curves on the approach to a ford or drift where a driver may have to stop because of
the presence of surface water. For crest curves it has become more common to adopt an
object height of 0.6m to cater for night time conditions when the object is likely to be the
tail lights of a vehicle. However, in hilly and mountainous terrain the most likely object is
a fallen rock and in forested area, the branch of a tree. In such conditions an object height
of 0.2 metres is the best compromise.

The minimum lengths of crest curves have been designed to provide sufficient sight
distance during daylight conditions. Longer lengths would be needed to meet the same
visibility requirements at night on unlit roads but, even on a level road, low beam headlight
illumination may not show up small objects at the design stopping sight distances. There is
no point in providing a suitable sight distance if headlights are not bright enough to make
use of it, therefore these longer lengths of curve are not justified. From a safety point of
view, high objects such as vehicles and their tail lights will be adequately illuminated at the
required stopping sight distances. Approaching vehicles will be identified by the
approaching illumination and drivers should be more alert at night and/or be travelling at
reduced speed.

Similar calculations can be carried out based on passing sight distance. High values of K
result so that, in the situation where the crest of the curve is in cut, the increase in volume
of excavation will be significant. Although the designer should seek to provide as much
passing sight distance as possible along the length of the road, it may be useful to shorten
the crest curve in order to increase the lengths of the grades on either side rather than to
attempt achieving passing sight distance over the crest curve itself.

Page 9-4 Ethiopian Roads Authority

 Chapter 9
Geometric Design Manual – 2013 Vertical Alignment

Minimum values of K for crest curves are shown in Tables 9.1 and 9.2.

Table 9-1: Minimum Values for Crest Vertical Curves (Paved Roads)
K for Stopping
Design Speed Sight Distance K for Passing
(km/h) Sight Distance
h2 = 0m h2 = 0.2m h2 = 0.6m
20 2 1 1 10
25 3 1 1 30
30 4 2 1 50
40 10 5 3 90
50 20 10 7 130
60 35 17 11 180
70 60 30 20 245
80 95 45 30 315
85 115 55 35 350
90 140 67 45 390
100 205 100 67 480
110 285 140 95 580
120 385 185 125 680

Table 9-2: Minimum Values for Crest Vertical Curves (Unpaved Roads)
K for Stopping
Design Speed Sight Distance K for Passing
(km/h) Sight Distance
h2 = 0m h2 = 0.2m h2 = 0.6m
20 2 1 1 10
25 3 1 1 30
30 5 2 2 50
40 11 6 4 90
50 25 11 8 135
60 45 20 15 185
70 75 35 25 245
80 120 58 40 315
85 150 72 50 350
90 185 90 60 390
100 270 130 88 480

9.4 Sag Curves

It is assumed that adequate sight distance will be available on sag curves in daylight.
However, at night, visibility is limited by the distance illuminated by the headlamp beams.

Ethiopian Roads Authority Page 9-5

Chapter 9
Vertical Alignment Geometric Design Manual – 2013

Assumptions concerning the brightness of the headlights, their height above the road and
the divergence of the beams have been made and minimum sag curve lengths for this
condition have been computed. However the results lead to unrealistically long vertical
curves, especially at higher speeds, and the required sight distances may be in excess of the
effective range of the headlamp beam. Thus, the only likely situation when the calculations
are useful is on the approaches to fords and drifts and other similar locations where
flowing or standing water may be present on the road surface. Most of these structures
occur on low speed roads where headlamp illumination is more likely to reach the full
sight distances.

It is therefore recommended that, for most situations, sag curves are designed using a
driver comfort criterion of vertical acceleration. A maximum acceleration of 0.3m/sec2 is
often used. This translates into

K > V2/395

Where V is the velocity in km/hr.

The resulting curve length values are shown in Table 9.3.

Table 9-3: Minimum Values of K for Sag Curves

Design Speed (km/h) K for driver comfort
20 1.0
25 1.5
30 2.5
40 4.
50 6.5
60 9
70 12
80 16
85 18
90 20
100 25
110 30
120 36

9.5 Minimum Lengths of Vertical Curves

Especially for trunk and link roads, where the algebraic difference between successive
gradients is often small, the intervening minimum vertical curve, obtained by applying the
above formulae, becomes very short. This can create the impression of a kink in the grade
line. If the vertical alignment is allowed to contain many curves of short length, the result
can be a ‘hidden dip’ profile, and/or a ‘roller coaster’ type profile, as indicated in Figure
9.3. For this reason, where the algebraic difference in gradient is less than 0.5 percent, a
minimum curve length is recommended for purely aesthetic reasons. The minimum length

Page 9-6 Ethiopian Roads Authority

 Chapter 9
Geometric Design Manual – 2013 Vertical Alignment

should not be less than twice the design speed in km/h and, for preference, should be 400
metres or longer, except in mountainous or escarpment terrain.

Figure 9-3: Hidden Dip and Roller Coaster Profiles

Where a crest curve and a succeeding sag curve have a common beginning and end, the
visual effect created is that the road has suddenly dropped away. In the reverse case, the
illusion of a hump is created. Either effect is removed by inserting a short length of straight
grade between the two curves. Typically, 60 m to 100 m is adequate for this purpose

For lower standard roads (DC1, DC2 and DC3), no minimum curve length is specified. In
these cases, the curve lengths should be kept to a minimum to enhance drainage
capabilities, and the curve lengths should match the K values given in Tables 2.6 to 2.17
for stopping sight distance. Where the difference in grade is less than 0.5 percent, the
vertical curve is often omitted.

Ethiopian Roads Authority Page 9-7

Chapter 9
Vertical Alignment Geometric Design Manual – 2013

9.6 Maximum Gradients

Vehicle operations on gradients are complex and depend on a number of factors: severity
and length of gradient; volume and composition of traffic; and the number of overtaking
opportunities on the gradient and in its vicinity.

For very low levels of traffic of only a few four-wheel drive vehicles, various references
advocate a maximum traversable gradient of up to 18 percent. Small commercial vehicles
can usually negotiate an 18 percent gradient, whilst two-wheel drive trucks can
successfully manage gradients of 15-16 per cent except when heavily laden.

However, under normal operating conditions the level of service and considerations of
safety the geometric design should aim at achieving grades which will not reduce the speed
of heavy vehicles to such an extent as to cause intolerable conditions for following drivers.
It has been found that the frequency of truck accidents increases sharply when truck speed
is reduced by more than 15 km/h. A speed reduction of 20 km/h is recommended as
representing intolerable conditions. If gradients on which the truck speed reduction is less
than 20 km/h cannot be achieved economically, it may be necessary to provide auxiliary
lanes for the slower-moving vehicles

The vehicle fleet in Ethiopia is composed of a high percentage of vehicles that are under-
powered and poorly maintained. Some existing roads are avoided and under-utilised by
traffic because of an inability to ascend the existing grades. ERA has little choice but to
limit gradients based on the existing fleet, although this translates into an added cost to
develop the road infrastructure.

Recent research has quantified the costs of using earth or gravel surfaces on steep gradient.
It has shown that spot improvements whereby the steep sections are surfaced with one of a
variety of appropriate surfacings built by labour-based methods are very cost effective at
providing all year access and reducing maintenance costs and whole life costs. Thus the
limiting gradients on gravel and earth roads reflect this and it is expected that a spot
improvement strategy will become standard practice in Ethiopia.

Maximum ‘absolute’ gradient and maximum ‘desirable’ gradient are therefore extremely
important criteria that greatly affect both the serviceability and cost of the road.
Performance considerations have formed the basis for defining the limiting criteria as
summarised in Table 9.4 and shown in the design standards in Tables 2.6 to 2.17.

When gradients of 7 percent or greater are reached, consideration should be given to

paving the steep sections (spot improvements) to enable sufficient traction to be achieved
as well as to minimise maintenance requirements. As traffic increases, the economic dis-
benefits of severe gradients, measured as increased vehicle operating and travel time costs,
will justify reducing the severity and/or length of a gradient or paving the steep sections.
On the higher design classes of road, the lower maximum recommended gradients reflect
these economics. However, an economic assessment of alternatives to long or severe
gradients should be undertaken where possible.

Page 9-8 Ethiopian Roads Authority

 Chapter 9
Geometric Design Manual – 2013 Vertical Alignment

Table 9-4: Maximum Gradients for Paved Sections

Maximum Gradient (%), for Paved Sections
Topography DC8, DC7, DC6 DC5, DC4 DC3, DC2 DC1 Basic Access
D A D A D A D A D A
Flat 3 5 4 6 6 8 6 10
Rolling 4, 5 7 6 8 7 9 7 10
Mountainous 6, 7 9 8 10 10 12 10 12 NA NA
Escarpment 6, 7 9 8 10 10 12 10 12
Urban 6 8 7 9 7 9 7 9
Note: D is the desirable value, A is the absolute value.

Standards for desirable maximum gradients were set to assure user comfort and to avoid
severe reductions in the design speed. If the occasional terrain anomaly is encountered that
requires excessive earthworks to reduce the vertical alignment to the desirable standard an
absolute maximum gradient can be used. Employment of a gradient in excess of the
desirable maximum can only be authorized through a Departure from Standard (see
Section 2.2).

9.7 Maximum Gradients at Switchbacks

Where switchback curves are unavoidable in mountainous or escarpment terrain, there is a

need to reduce the maximum allowable gradient at any point through the curve. The
maximum allowable gradient through a switchback curve is 4 percent for road standards
DC8 to DC5, and 6 percent for DC4 to DC1. The minimum allowable gradient is 0.5%.

Corresponding crest and sag curves approaching the switchback curve must meet the
requirements of Sections 9.3 and 9.4, and the transitions must be completed outside of the
switchback curve. The sag curve above the switchback shall be made as long as possible to
allow ascending vehicles to accelerate at the flatter grade when leaving the switchback.

9.8 Minimum Gradients

The minimum gradient for the usual case is 0.5 percent. However, flat and level gradients
on uncurbed paved highways are acceptable when the cross slope and carriageway
elevation above the surrounding ground is adequate to drain the surface laterally. With
curbed highways or streets, longitudinal gradients should be provided to facilitate surface
drainage.

9.9 Gradients through Villages

In many instances the natural grade level is flat through villages. The adjacent roadside
ditches in such circumstances can readily become clogged and ineffective. Sometimes they
are deliberately blocked to provide access to adjacent property or to channel flow for
agricultural use. These practices lead to saturation of the subgrade and hence pavement
failure, and should be avoided.

Ethiopian Roads Authority Page 9-9

Chapter 9
Vertical Alignment Geometric Design Manual – 2013

9.10 Critical Length of Gradient

The critical length of gradient is the maximum length of a designated upgrade upon which
a loaded truck can operate without unreasonable reduction in speed. It is, to some extent,
dependent on the gradient of the approach because a downhill approach will allow vehicles
to gain momentum and thereby to increase the critical length. The critical length of
gradient also decreases, as gradient increases. This is shown in Table 9.5. Where it is
necessary to exceed the critical length of gradient on heavily trafficked roads, it is
desirable to provide either safe passing distances on the rise, or a climbing lane for heavy
vehicles.

9.11 Climbing Lanes

A climbing lane is an effective means of reducing the impact of a steep gradient. A

climbing lane is an auxiliary lane added outside the continuous lanes and has the effect of
reducing congestion in the through lanes by removing slower vehicles from the traffic
stream. It also enhances road safety by reducing the speed differential in the through lane.
The requirements for climbing lanes are therefore based on road standard, speed and traffic
volume.

Benefits from the provision of a climbing lane accrue because faster vehicles are able to
overtake more easily, resulting in shorter average journey times, reduced vehicle-operating
costs, and increased safety. Benefits increase with increases in gradient, length of gradient,
traffic flow, the proportion of trucks, and reductions in overtaking opportunities. The effect
of a climbing lane in breaking up queues of vehicles held up by a slow moving truck will
continue for some distance along the road.

Climbing lanes must be considered for roads when present traffic volumes are greater than
400 ADT. Thus the application of climbing lanes is limited particularly to trunk and link
roads. Table 9.5 was prepared according to the criteria that a 20 km/h speed reduction is
expected for a truck. It is used in the design to indicate locations where climbing lanes are
recommended.

A climbing lane layout is shown in Figure 9.4. Climbing lanes must be clearly marked and,
where possible, should end on level or downhill sections where speed differences between
different classes of vehicles are lowest to allow safe and efficient merging manoeuvres.
The introduction and termination of a climbing lane shall be effected by 100 metre long
tapers. The tapers shall not be considered as part of the climbing lanes.

The starting point of the grade can be approximated as a point halfway between the
preceding vertical point of intersection and the end of the vertical curve.

In escarpment terrain the carriageway and shoulder widths may have been reduced, hence a
climbing lane, which will increase the width considerably, may not be economically
justified. Consideration must be given to a balance between the benefits to traffic and the
initial construction cost. In sections requiring heavy side cut, the provision of climbing
lanes may be unreasonably costly in relation to the benefits. Reduced level of service over
such sections is an alternative.

Page 9-10 Ethiopian Roads Authority

 Chapter 9
Geometric Design Manual – 2013 Vertical Alignment

The climbing lane is sometimes not effectively utilised, especially when traffic flows are
heavy, because the drivers of slower vehicles fear that they will not be allowed to merge
with the faster vehicles where the climbing lane ends. The preferred layout forces faster
vehicles to merge with the slower, thus allaying this fear to some extent. This layout is
preferred based purely on that fact that a vehicle can merge more readily with a slower-
than with a faster-moving stream of traffic (see Figure 9.4).

A slow moving vehicle should be completely clear of the through lane by the time its speed
has dropped by 20 km/h, and remain clear of the through lane until it has accelerated again
to a speed which is 20 km/h less than its normal speed.

Table 9-5: Climbing Lanes

Critical length of gradient
Gradient Maximum desirable
Design Class above which a climbing
(%) length of gradient (m)
lane is required (m)
DC7 and DC6 4 300 900
5 240 800
6 200 700
7 170 600
DS7, DC6, DC5
8 150 500
9 130 400
10 Required 400
11 Required 400
DC4
12 Required 400

Ethiopian Roads Authority Page 9-11

Chapter 9
Vertical Alignment Geometric Design Manual – 2013

Figure 9-4: Layout for Climbing Lane

The performance characteristics of a heavy vehicle are such that, for a particular gradient,
the vehicle speed will decrease to a final ambient speed that can be maintained by that
vehicle on that grade. This limits, in most references, any discussion on the maximum
length allowable at a given grade even considering the employment of a climbing lane.
However, in the interests of factors such as vehicle operating costs and travel time losses,
the absolute recommended maximum lengths at any given grade are also indicated in the
last column of Table 9.5. When these distances are reached, it is necessary to design a
relief gradient of less than 6 percent between steep sections. The relief gradient must
extend a minimum of 100 metres.

Page 9-12 Ethiopian Roads Authority

 Chapter 9
Geometric Design Manual – 2013 Vertical Alignment

These values have also taken into consideration the safety factors associated with the
increase in speed resulting in the descent of steep grades. Although they may mitigate the
safety hazard, they do not eliminate it. For example, a non-braking heavy truck will
accelerate from 0 km/hr to 90 km/hr over a distance of about 500 metres at a descending
grade of 5 percent. This emphasizes the need to provide warning signs for such vehicles at
all long continuous grades. The use of ‘escape lanes’ is discussed in Section 13.9.

9.12 Vertical Clearances

Bridges over water shall normally have a minimum clearance height according to Table 9.6
unless a refined hydraulic analysis has been made. The standard minimum headroom or
clearance under bridges or tunnels shall be 5.1m for all classes of roads. This clearance
should be maintained over the roadway(s) and shoulders. Where future maintenance of the
roadway is likely to lead to raising of the road level, then an additional clearance of up to
0.1m may be provided. Light superstructures (e.g. timber, steel trusses, steel girders, etc)
over roadways shall have a clearance height of at least 5.3m. See ERA's Bridge Design
Manual for further reference.

Table 9-6: Vertical Clearance from Superstructure to Design Flood Level (DFL)
Design Flow at Bridge (m3/s) Vertical Clearance (m)
5 to 30 0.6
30 to 300 0.9
>300 1.2
Source: ERA Bridge Design Manual

Underpasses for pedestrians and bicycles shall not be less than 2.4m. For cattle and
wildlife, underpasses shall be designed as the normal height of the actual kind of animal
plus 0.5m, and for horse-riding the clear height shall be not less than 3.4m. Bridges above
railways shall have a clearance height of at least 6.1m- if not otherwise stated- to facilitate
possible future electrification.

Over existing pipe culverts and box culverts, the roadway elevation cannot be less than as
indicated in the ERA Drainage Design Manual.

Ethiopian Roads Authority Page 9-13

 Chapter 10
Geometric Design Manual – 2013 Phasing of Horizontal and Vertical Alignment

10 PHASING OF HORIZONTAL AND VERTICAL ALIGNMENT

10.1 Alignment Defects Due to Mis-phasing

Phasing of the vertical and horizontal curves of a road implies their coordination so that the
line of the road appears to a driver to flow smoothly, avoiding the creation of hazards and
visual defects. It is particularly important in the design of high-speed roads on which a
driver must be able to anticipate changes in both horizontal and vertical alignment well
within the safe stopping distance. It becomes more important with small radius curves than
with large.

Defects may arise if an alignment is mis-phased. Defects may be purely visual and do no
more than present the driver with an aesthetically displeasing impression of the road. Such
defects often occur on sag curves. When these defects are severe, they may create a
psychological obstacle and cause some drivers to reduce speed unnecessarily. In other
cases, the defects may endanger the safety of the user by concealing hazards on the road
ahead. A sharp bend hidden by a crest curve is an example of this kind of defect.

10.2 Types of Mis-phasing and Corresponding Corrective Action

Cases of mis-phasing fall into several types. These are described below together with the
necessary corrective action for each type.

When the horizontal and vertical curves are adequately separated or when they are
coincident, no phasing problem occurs and no corrective action is required. Where defects
occur, phasing may be achieved either by separating the curves or by adjusting their
lengths such that vertical and horizontal curves begin at a common station and end at a
common station. In some cases, depending on the curvature, it is sufficient if only one end
of each of the curves is at a common station.

10.3 Vertical Curve Overlaps One End of the Horizontal Curve

If a vertical curve overlaps either the beginning or the end of a horizontal curve, a driver’s
perception of the change of direction at the start of the horizontal curve may be delayed
because his sight distance is reduced by the vertical curve. This defect is hazardous. The
position of the crest is important because vehicles tend to increase speed on the down
gradient following the highest point of the crest curve, and the danger due to an unexpected
change of direction is consequently greater. If a vertical sag curve overlaps a horizontal
curve, an apparent kink may be produced, as indicated in Figures 10.1b and c.

The defect may be corrected in both cases by completely separating the curves. If this is
uneconomic, the curves must be adjusted so that if the horizontal curve is of short radius
they are coincident at both ends, or if the horizontal curve is of longer radius they need be
coincident at only one end.

10.4 Insufficient Separation between the Curves

If there is insufficient separation between the ends of the horizontal and vertical curves, a
false reverse curve may appear on the outside edge-line at the beginning of the horizontal
curve. This is a visual defect, illustrated in Figure 10.1d.

Ethiopian Roads Authority Page 10-1

Chapter 10
Phasing of Horizontal and Vertical Alignment Geometric Design Manual – 2013

Corrective action consists of increasing the separation between the curves, or making the
curves concurrent, as in Figure 10.1a.

10.5 Both Ends of the Vertical Curve Lie on the Horizontal Curve

If both ends of a crest curve lie on a sharp horizontal curve, the radius of the horizontal
curve may appear to the driver to decrease abruptly over the length of the crest curve. If the
vertical curve is a sag curve, the radius of the horizontal curve may appear to increase. An
example of such a visual defect is shown in Figure10.1e. The corrective action is to make
both ends of the curves coincident as in Figure 10.1a, or to separate them.

10.6 Vertical Curve Overlaps Both Ends of the Horizontal Curve

If a vertical crest curve overlaps both ends of a sharp horizontal curve, a hazard may be
created because a vehicle has to undergo a sudden change of direction during the passage
of the vertical curve while sight distance is reduced.

The corrective action is to make both ends of the curves coincident. If the horizontal curve
is less sharp, a hazard may still be created if the crest occurs off the horizontal curve. This
is because the change of direction at the beginning of the horizontal curve will then occur
on a downgrade (for traffic in one direction) where vehicles may be increasing speed.

The corrective action is to make the curves coincident at one end so as to bring the crest on
to the horizontal curve.

No action is necessary if a vertical curve that has no crest is combined with a gentle
horizontal curve.

If the vertical curve is a sag curve, an illusory crest or dip, depending on the “hand” of the
horizontal curve will appear in the road alignment.

The corrective action is to make both ends of the curves coincident or to separate them.

10.7 Other Mis-Phasing

Other types of mis-phasing are indicated in Figure 10.1:

• A sag curve occurs between two horizontal curves in the same direction in Figure 10.1g.
This illustrates the need to avoid broken back curves in design (see Chapter 8:
Horizontal Alignment).
• A double sag curve occurs at one horizontal curve in Figure 10.1h. This illustrates the
effect in this case of a broken back vertical alignment on design (see Chapter 9: Vertical
Alignment).
• Figure 10.1i shows a lack of phasing of horizontal and vertical curves. In this case, the
vertical alignment has been allowed to be more curvilinear than the horizontal
alignment.

Page 10-2 Ethiopian Roads Authority

 Chapter 10
Geometric Design Manual – 2013 Phasing of Horizontal and Vertical Alignment

10.8 The Economic Penalty Due to Phasing

The correct phasing of vertical curves restricts the designer in fitting the road to the
topography at the lowest cost. Therefore, phasing is usually bought at the cost of extra
earthworks and the designer must decide at what point it becomes uneconomic. He will
normally accept curves that have to be phased for reasons of safety. In cases when the
advantage due to phasing is aesthetic, the designer will have to balance the costs of trial
alignments against their elegance.

Ethiopian Roads Authority Page 10-3

Chapter 10
Phasing of Horizontal and Vertical Alignment Geometric Design Manual – 2013

Figure 10-1: Phasing of Horizontal and Vertical Curves

Page 10-4 Ethiopian Roads Authority

 Chapter 11
Geometric Design Manual – 2013 At-Grade Junctions

11 AT-GRADE JUNCTIONS
11.1 Introduction

A junction, or intersection, is the general area where two or more roads join. A
disproportionate amount of traffic accidents occur at junctions, and thus, from a traffic
safety aspect, junctions require attention and careful design. Good junction design should
allow transition from one route to another or through movement on the main route and
intersecting route with minimum delay and maximum safety. To accomplish this, the
layout and operation of the junction should be obvious to the driver, with good visibility
between conflicting movements.

Differing junction types will be appropriate under different circumstances depending on

traffic flows, speeds, and site limitations. Types of junctions are illustrated in Figure 11.1
and include:
i) Three-legged T and Y-Junctions.
ii) Four-legged Cross-Junctions.
iii) Multi-legged (>4) Junctions.
iv) Roundabouts.
v) Grade Separation (discussed in Chapter 12).
Intersections vary greatly in their design and complexity based primarily on the traffic
volume that they have to accommodate. The important factors to be considered in the
selection of an intersection type are:
i) Volumes of traffic to be accommodated.
ii) Proportion of traffic on each approach.
iii) Approach speeds.
iv) Land use and land availability.
v) Local environment.
vi) Cost of construction.
Safety at intersections is of key importance. For example, the National Safety Council in
the USA has estimated that 56 per cent of all urban accidents and 32 per cent of all rural
accidents occur at intersections. Poor design, e.g. inadequate sight distances, leads to
significantly higher injury and total accident rates, hence good design is vital if accident
rates are to be minimised. It should be noted that roundabouts have considerable safety
advantages over other types of at-grade intersections.
This chapter describes the design for all at-grade junctions. A checklist for junction design
is presented in Figure 11.19 at the end of the chapter.

11.2 Design Requirements

Intersections are of two basic types, those where traffic speed is uncontrolled and those
where some control is provided. Control may consist of a mandatory ‘stop line’ or
considerably more complex signal control or traffic management system.

Uncontrolled intersections have no control at all, or simply provide advice to drivers to

‘give-way’ or ‘yield’ on the basis of priority to the first to arrive at the junction or to traffic

Ethiopian Roads Authority Page 11-1

Chapter 11
At-Grade Junctions Geometric Design Manual – 2013

on the major road. Such intersections are used where traffic volume is low. When the
traffic is similar on all legs, the recommended capacity of such a junction is about 250
vehicles per hour in each direction on each leg. This increases to about 750 vehicles per
hour in each direction on the major road when traffic on the minor road is 10 per cent of
this. When traffic exceeds these values, additional features need to be included as
described in the following Sections.

Figure 11-1: Types of at-grade junctions

The design of junctions must take account of the following basic requirements:
i) Safety.
ii) Operational comfort.
iii) Capacity.

Page 11-2 Ethiopian Roads Authority

 Chapter 11
Geometric Design Manual – 2013 At-Grade Junctions

iv) Economy.
A junction is considered safe when it is visible, comprehensible, and manoeuvrable. The
right of way should follow naturally and logically from the junction layout and the types of
junctions used throughout the whole road network should be similar. The use of road signs
may be necessary and road markings and other road furniture will almost always be
required.

The basic objectives of good intersection design are:

• Reduction of the number of points of potential conflict to the minimum compatible
with efficient operation;
• Reduction of the complexity of conflict areas whenever possible;
• Limitation of the frequency of actual conflicts; and
• Limitation of the severity of those conflicts that do occur.

The tools available to apply these principles are interrelated:

i) Defining and arranging traffic lanes;
ii) Use of traffic islands of all sizes and types;
iii) Median islands;
iv) Horizontal and vertical approach geometry;
v) Pavement tapers and transitions; and
vi) Traffic control devices.

The first five elements are a range of physical features while traffic control devices are an
integral part of any intersection.

11.3 Selection of Junction Type

11.3.1 Junction Choice

The choice of a junction type requires knowledge of traffic demand, intersection

performance and accident prediction. It is often difficult to determine the best intersection
type for any particular situation, taking into account capacity, delay, safety and physical
layout factors. Several alternatives may be possible at a junction. A guide for the selection
of junction type based on different combinations of traffic flows is shown in Figure 11.2.
For lower volume junctions, the applicable junction type depending on traffic flow is
shown in further detail in Figure 11.3.

The basic advantages and disadvantages of different junction types, including grade
separation, are as follows:
Priority (T-Junction, Cross-Junction). For low flows. These can cause long delays. They
require sufficient stopping sight distance. Delays can be improved by signal installation.
Roundabouts. These are best for low to medium flows. They provide for minimum delays
at lower flows. They have been shown to be safer than priority junctions. They require
attention to pedestrian movements and the accommodation of slow-moving traffic.
Grade-Separation. This is only for high flows because they are expensive but they result
in minimum delays. Pedestrian movements also need special consideration.

Ethiopian Roads Authority Page 11-3

Chapter 11
At-Grade Junctions Geometric Design Manual – 2013

These options are discussed in more detail below.

Figure 11-2: Junction Selection Based on Traffic Flows

Figure 11-3: T-Junction Selection for Various Major and Minor Road Traffic Flows

Page 11-4 Ethiopian Roads Authority

 Chapter 11
Geometric Design Manual – 2013 At-Grade Junctions

11.3.2 T-Junctions

The basic junction layout for rural roads is the T-junction with the major road traffic
having priority over the minor road traffic. Applications of T-junctions include the
staggered T-junction, which caters to cross-traffic. Staggered T-junctions are often the
result of a realignment of the minor route to improve the angle of the skew of the crossing,
as shown in Figure 11.4. Where such staggered T-junctions are used, the left-right stagger
is preferred to the right-left stagger. The reason for this is that, in the latter case, a crossing
vehicle must re-enter the minor road by making a left turn on the major roadway. In other
words the number of turning manoeuvres that require traffic to cross a traffic stream
increases from 2 to 4. In such cases, the inclusion of a left-turning lane between the
staggers should be considered. The minimum distance between the T-junctions is shown in
Table 11.1.

Figure 11-4: Staggered X-Junctions

11.3.3 Cross Junctions

A cross junction is required where two highways cross each other (i.e. a crossroads). It
therefore has four legs. The overall principles of design, island arrangements, use of
turning lanes, and other parameters are similar to those used in T-junctions.

Experience in some countries has shown that converting crossroads into roundabouts can
reduce accident costs by more than 80 per cent.

Where more complex junction layouts involving the intersection of four or more roads are
encountered, these should be simplified by redesign to two junctions, or a roundabout
should be used.

Having selected the basic junction layout, it is necessary to adapt this basic layout in
accordance with the following principles to ensure that a safe, economic and geometrically
satisfactory design will be produced.

11.4 Angle of Intersection

The angle of intersection of two roadways influences the operation and safety of a
junction. Large skews increase the pavement area and thus the area of possible conflict.
Operationally they are undesirable because:
• Crossing vehicles and pedestrians are exposed for longer periods;
• The driver's sight angle is more constrained and gap perception becomes more
difficult;
• Vehicular movements are more difficult;

Ethiopian Roads Authority Page 11-5

Chapter 11
At-Grade Junctions Geometric Design Manual – 2013

• Large trucks require more pavement area; and

• Defining vehicle paths by channelization is more difficult.

For new intersections the crossing angle should preferably be in the range 75° to 120°. The
absolute minimum angle of skew is 60° because drivers, particularly of trucks with closed
cabs, have difficulty at this angle of skew in seeing vehicles approaching from one side.
The designer should be able to specifically justify using an angle of skew less than 75°. In
the remodelling of existing intersections, the accident rates and patterns will usually
indicate whether a problem exists and provide evidence on any problems related to the
angle of skew.

11.5 Distance between Adjoining Junctions

Level of service and driver perception is affected by the spacing of junctions. In certain
cases it may be necessary to limit the number of junctions for reasons of safety and
serviceability. Table 11.1 gives a guide to the minimum spacing for each road design
standard, and should be used for the design of new roads or when reviewing junction
layouts.

Table 11-1: Access Control

Minimum Spacing of
Design Standard
Junctions (m)
DC8 1000
DC7 500
DC6- DC5 300
DC4- DC3 100
DC2- DC1 20

11.6 Sight Distances and Visibility

The key to good design and maximum safety is to provide adequate sight distances for the
manoeuvres that are required. On a basic cross-road intersection these manoeuvres are left
turns and right turns from both the minor road and the major road and crossing manoeuvres
across the junction. The important factors are the time required to carry out the manoeuvre
and the time available to do so based on the sight distance and the speed of traffic. The
time required to carry out the manoeuvre depends upon:
i) Whether the vehicle is in motion and at what speed when it reaches the junction
(yield control – Figure 11.5) or begins from a stopped position (stop control –
Figure 11.6);
ii) The type and power of the vehicle;
iii) The length of the vehicle;
iv) The distance the vehicle needs to travel (number of lanes plus median, if present);
v) The gradient of the road which the vehicle has to negotiate;
vi) The perception and reaction time of the driver;

Page 11-6 Ethiopian Roads Authority

 Chapter 11
Geometric Design Manual – 2013 At-Grade Junctions

Thus calculating the time required is complex and varies considerably for different
conditions.

Figure 11-5: Visibility Splay for ‘Yield’ Conditions

Figure 11-6: Visibility Splay for ‘Stop’ Conditions

The time available to carry out the manoeuvre depends on the speed of traffic in the lanes
to be crossed. This speed is not the design speed of the road because drivers tend to slow
down when approaching junctions, even when on the major road. However the sight
distance needs to be at least as great as the product of the traffic speed and the time
required to carry out the manoeuvre.

Models have been developed for carrying out these calculations but require many
assumptions and are not reliable. The best information is obtained from empirical data, but
this is primarily based on research in western countries. As has been emphasised elsewhere
in this manual, the mix of traffic, its age spectrum, overloading practices etc. are entirely
different in Ethiopia. Furthermore, it is apparent from the discussion above that the
required sight distances also depend on driver behaviour. Nobody can doubt that driver
behaviour in western countries is significantly different to driver behaviour in Ethiopia. In
summary, it is not a simple task to calculate the optimum or minimum sight distances
applicable to different junction designs, different road classes and different mixes of traffic
in Ethiopia. A pragmatic approach is to utilise the available empirical data but to select
conservative options for safety. Appendix C summarises the empirical approach and shows

Ethiopian Roads Authority Page 11-7

Chapter 11
At-Grade Junctions Geometric Design Manual – 2013

how sight distances can be calculated for the majority of manoeuvres and for different
design vehicles.

The greatest sight distances are needed for the manoeuvres that take the longest to execute
(required time) and involve joining fast traffic (available time). This inevitably means that
heavy truck and trailer combinations require the greatest sight distances when joining a
main road. Catering for this situation is not always possible. The methods described in
Appendix C can be used to calculate sight distances for different design vehicles but it is
prudent to use DV4 for most designs.

Practical sight distances are shown in Tables 11.2 and Table 11.3 below and are generally
more conservative than the more precise values obtained using Appendix C. However,
when the gradients on any of the legs exceed 3 per cent and when multiple lanes are
involved, prudent use of Appendix C should be made, bearing in mind that driver
behaviour is likely to be different in Ethiopia and therefore conservative options should be
used.
Table 11-2: Minimum Sight Distances for ‘Yield’ Conditions
Main road design speed (km/h) 40 50 60 70 85 100 120
Sight distance, LA (m) 80 95 115 140 190 215 270

Table 11-3: Minimum Sight Distances for ‘Stop’ Conditions

Main road design speed (km/h) 40 50 60 70 85 100 120
Sight distance LS (m) 130 160 190 225 275 320 385

11.7 Lane Width and Manoeuvrability

All traffic lanes should be of adequate width and radius for the appropriate vehicle turning
characteristics. To accommodate truck traffic, turn radii shall be a minimum of 15 metres.

Where intersecting roadways have shoulders or sidewalks, the main road shoulder should
be continued through the intersection. Lane widths should normally be the same as that of
the major road so that approaches from the minor road are usually widened if necessary.

Where conditions are severely constrained, lane widths as low as 3.3 metres can be
considered, provided that approach speeds are below 80 km/h. In constricted urban
conditions on low speed-roadways, lane widths of 3.0 metres should be the minimum
adopted.

Offsets from the edge of a turning roadway to kerb lines should be 0.6 to 1.0 metres.
The edges of traffic lanes should be clearly indicated by road markings.

11.8 Horizontal and Vertical Alignment

The horizontal and vertical alignments through and approaching an intersection are critical
features. Simple alignment design allows for early recognition of the intersection and

Page 11-8 Ethiopian Roads Authority

 Chapter 11
Geometric Design Manual – 2013 At-Grade Junctions

timely focus on the intersecting traffic and manoeuvres that must be prepared. The
following are specific operational requirements at intersections:
• The alignments should not restrict the required sight distance;
• The alignments should allow for the frequent braking and turning associated with
intersections; and
• The alignments should not require a driver’s attention to be detracted from the
intersection manoeuvres and conflict avoidance.
As a general guide, horizontal curve radii at intersections should not be less than the
desirable radius for the design speed on the approach roads. For high-speed roads with
design speeds in excess of 80 km/h, approach gradients should not be greater than -3 per
cent. For low-speed roads in an urban environment this can be increased to -6 per cent.

11.9 Channelisation

11.9.1 Purpose of Channelisation

The operation of the junction depends principally upon the frequency of gaps that naturally
occur between vehicles in the main road flow. These gaps should be of sufficient duration
to permit vehicles from the minor road to merge with, or cross, the major road flow. In
consequence, junctions are limited in capacity, but this capacity may be optimised by, for
example, channelisation or the separation of manoeuvres.

At-grade intersections with large paved areas, such as those with large corner radii or with
angles of skew differing greatly from 90o, permit unpredictable vehicle movements,
require long pedestrian crossings and have unused pavement areas. Even at a simple
intersection there may be large areas on which vehicles can wander from natural or
expected paths. Under these circumstances it is usual to resort to channelisation.
The purpose of channelisation is to manage the conflicts that are inherent in any
intersection. There are eight principles of channelisation:
i) Undesirable or wrong-way movements should be discouraged or prohibited;
ii) Vehicle paths should be clearly defined;
iii) The design should encourage safe vehicle speeds;
iv) Points of conflict should be separated whenever possible;
v) Traffic streams should cross at close to right angles and merge at flat angles;
vi) High priority flows should have the greater degree of freedom;
vii) Decelerating, slow-moving or stopped vehicles should be separated from higher-
speed through lanes; and
viii) Refuge for pedestrians and the handicapped should be provided where
appropriate.

Thus channelisation is the process whereby a vehicle is guided safely through the
intersection area from an approach leg to the selected departure leg. Guidance is offered by
lane markings that clearly define the required vehicle path and also indicate auxiliary lanes
for turning movements. Various symbols are also used as road markings to indicate that
turns, either to the left or to the right, from selected lanes are mandatory. At intersections
that are complex or have high volumes of turning traffic, it is usually necessary to reinforce
the guidance offered by road markings by the application of:

Ethiopian Roads Authority Page 11-9

Chapter 11
At-Grade Junctions Geometric Design Manual – 2013

• Channelising islands;
• Medians and median end treatments;
• Corner radii;
• Approach and departure geometry;
• Pavement tapers and transitions;
• Traffic control devices including signs and signals; and
• Arrangement and position of lanes.

11.9.2 Channelising Islands

A traffic island is a defined area between traffic lanes for the control of vehicle movements
and which may also be used as a pedestrian refuge. Traffic islands may take the form of an
area delineated by barrier curbs or a pavement area marked by paint or a combination of
these. Traffic islands may be included in the design of junctions for one or more of the
following purposes:
i) Separation of conflicts.
ii) Control of angle of conflict.
iii) Reduction of excessive pavement areas.
iv) Regulation of traffic and indication of proper use of junction.
v) Arrangements to favour a predominant turning movement.
vi) Protection of pedestrians.
vii) Protection and storage of turning and crossing vehicles.
viii) Location of traffic signs.
Islands are either elongated or triangular in shape and are situated in areas not normally
used as vehicle paths, the dimensions depending upon the particular junction or bus stop
layout.

The layout of an island is determined by the edges of the through traffic lanes, turning
vehicles and the lateral clearance to the island sides. Island curbs should be offset a
minimum of 0.3 metres from the edge of through traffic lanes even if they are mountable.

Typical island shapes are illustrated in Figure 11.7. The designer should bear in mind that
islands are hazards and should be less hazardous than whatever they are replacing.

Islands may be kerbed, painted or simply non-paved. Kerbed islands provide the most
positive traffic delineation and are normally used in urban areas to provide some degree of
protection to pedestrians and traffic control devices. Painted islands are usually used in
suburban areas where speeds are low (in the range of 50 km/h to 70 km/h) and space
limited.

In rural areas, kerbs are not common and, at the speeds prevailing in these areas, typically
100 km/h or more, they are a potential hazard. If it is necessary to employ kerbing at a
rural intersection, the use of mountable kerbing should be considered.

As an additional safety measure, a kerbed island may be preceded by a painted island.

Page 11-10 Ethiopian Roads Authority

 Chapter 11
Geometric Design Manual – 2013 At-Grade Junctions

Non-paved islands are defined by the pavement edges and are usually used for large
islands at rural intersections. These islands may have delineators on posts and may be
landscaped.

Figure 11-7: Traffic Islands

Drivers tend to find an archipelago of small islands confusing and are liable to select an
incorrect path through the intersection area. As a general design principle, a few large
islands are preferred to several small islands.

Islands should not be less than 5 m2 in area to ensure that they are easily visible to
approaching drivers.

Islands are generally either long or triangular in shape, with the circular shape being
limited to application in roundabouts. They are situated in areas not intended for use in
vehicle paths. Directional islands are typically triangular with their dimensions and exact
shape being dictated by:
• The corner radii and associated tapers;
• The angle of skew of the intersection; and

Ethiopian Roads Authority Page 11-11

Chapter 11
At-Grade Junctions Geometric Design Manual – 2013

• The turning path of the design vehicle.

A typical triangular island is illustrated in Figure 11.8. The approach ends of the island
usually have a radius of about 0.6 metres and the offset between the island and the edge of
the travelled way is typically 0.6 to 1.0 metres to allow for the effect of kerbing on the
lateral placement of moving vehicles. Where the major road has shoulders, the nose of the
island is offset about one metre from the edge of the usable shoulder, the side adjacent to
the through lane being tapered back to terminate at the edge of the usable shoulder, thus
offering some guidance and redirection. A kerbed cross-section on the major road suggests
that the nose of the island should be offset by about 1.6 metres from the edge of the
travelled way, with the side adjacent to the through lane being tapered back to terminate
0.6 metres from the edge of the through lane.

Figure 11-8: Typical Triangular Island

Dividing, or splitter, islands usually have a teardrop shape (Figure 11.9). They are often
employed on the minor legs of an intersection where these legs have a two-lane, two-way
or four-lane undivided cross-section. They are also often employed on the minor legs of an
intersection where these legs have a two-lane, two-way or four-lane undivided cross-
section.
The principle function of a dividing island is to warn the driver of the presence of the
intersection. This can be achieved if, at the widest point of the island, its edge is in line
with the edge of the approach leg. To the approaching driver, it appears as though the
entire lane had been blocked off by the island. If space does not permit this width of island,
a lesser blocking width must be applied, but anything less than half of the approach lane
width is not effective.

Splitter islands are also used in the approach to roundabouts where there is a need to
redirect vehicles entering a roundabout through an angle of not more than 30o.

Page 11-12 Ethiopian Roads Authority

 Chapter 11
Geometric Design Manual – 2013 At-Grade Junctions

Dividing islands are usually kerbed to ensure that the island is visible within normal
stopping sight distance. However, it may be advisable to draw the driver's attention to the
island by highlighting the kerbs with paint or reflective markings. As in the case of the
triangular island, the nose of the dividing island should be offset by 0.6 m from the
centreline of the minor road. For the sake of consistency, the radius of the nose should be
of the order of 0.6 m.

The balance of the shape of the island is defined by the turning paths of vehicles turning
both from the minor road to the major road and from the major road to the minor.

Figure 11-9: Splitter Island

11.9.3 Medians

Median islands are discussed in detail in Section 6.10. The general layout of median
openings at intersections is normally dictated by wheel-track templates. However, median
openings should not be shorter than:
• The surfaced width of the crossing road plus its shoulders.
• The surfaced width of the crossing road plus 2.5 m (if kerbing is provided).
• 12.4 metres.

A further control on the layout of the median opening is the volume and distribution of
traffic passing through the intersection area. If the median is wide enough to accommodate
them, it may be advisable to make provision for speed-change and storage lanes. The

Ethiopian Roads Authority Page 11-13

Chapter 11
At-Grade Junctions Geometric Design Manual – 2013

additional lanes reduce the width of the median at the point where the opening is to be
provided and thus influence the median end treatment.

The median end treatment is determined by the width of the median. Where the median is
3 m wide or less, a simple semicircle is adequate. For wider medians, a bullet nose end
treatment is recommended. The bullet nose is formed by arcs dictated by the wheel paths
of turning vehicles and an assumed nose radius of 0.6 to 1.0 m. This results in less
intersection pavement area and a shorter length of opening than the semi-circular end.

Above a median width of 5 m, the width of the minor road controls the length of the
opening. A flattened bullet nose, using the arcs as for the conventional bullet nose but with
a flat end as dictated by the width of the crossing road and parallel to the centreline of the
minor road, is recommended. These end treatments are illustrated in Figure 11.10.

The bullet nose and the flattened bullet nose have the advantage over the semi-circular end
treatment that the driver of a turning vehicle has a better guide for the manoeuvre for most
of the turning path. Furthermore, these end treatments result in an elongated median which
is better placed to serve as a refuge for pedestrians crossing the dual carriageway road.

Figure 11-10: Median End Treatment

(Criteria for width of opening are described in the text)

Page 11-14 Ethiopian Roads Authority

 Chapter 11
Geometric Design Manual – 2013 At-Grade Junctions

11.10 Speed-Change Lanes

11.10.1 Purpose

Deceleration lanes for vehicles turning left or right from a major road are of particular
value on higher speed and higher volume roads when a vehicle slowing down to leave the
major road may impede the following vehicles and cause a hazardous situation. Similarly,
a vehicle joining a high speed road will also cause a hazardous situation unless it can
increase its speed to that of the traffic on the road before merging; hence an acceleration
lane is desirable. These are incorporated into the Standard Detail Drawing for all junctions
on trunk and link roads.

The length of such speed-change lanes are based on acceptable levels of discomfort for
decelerating and for accelerating. These lengths are greater than stopping sight distances
because the latter are concerned with emergency braking.

11.10.2 Right Turn Lane

Right turn lanes, comprising a taper section and deceleration lane, shall be provided for all
trunk and link road DC8, DC7, and DC6 junctions, and for other road standards meeting
any of the following conditions:
i) On four or more lane roads and divided highways.
ii) When the major road design speed is 100 km/hr or greater, and the present year
AADT on the major road is greater than 1500 AADT.
iii) When the present year AADT of the right-turning traffic is greater than 750
AADT.
A detail of the layout for the Right Turn Lane is given in Figure 11.11. The length of the
right turn lane including the taper, measured as shown in the Figure, is related to design
speed as indicated in Table 11.4. The width of the major approach lane shall be the same as
the width of the traffic lanes.

Figure 11-11: Layout for Right Turn Lane

Ethiopian Roads Authority Page 11-15

Chapter 11
At-Grade Junctions Geometric Design Manual – 2013

Table 11-4: Length of Right Turn Lane

Length of Length of deceleration section (LD)
Main road
diverging Exit control speed (km/hr)
design speed
section (LT)
(km/hr) 40 50 60 70 85 100
(m)
60 65 60 20 - - - -
70 75 85 45 25 - - -
85 80 130 90 70 45 - -
100 90 185 145 125 100 55 -
120 110 270 230 210 180 145 85

On up-hill gradients these distances are shorter and on down-hill grades they are longer.
The increase or decrease in length is linear and is 5 per cent for every 1 per cent change in
grade. Thus, for example, for a down-hill grade of 4 per cent the length should be
increased by 20 per cent.

11.10.3 Left Turn Lanes

Warrants are the same as for a right-turning lane. A separate lane for left turning traffic
(traffic turning left from the major road into the minor road) shall be provided for all trunk
and link road DC8, DC7, and DC6 junctions. Warrants for inclusion of left turn lanes for
other road standards are under any of the following conditions:
i) On four or more lane roads and divided highways.
ii) When the major road design speed is 100 km/hr or greater, and the present year
AADT on the major road is greater than 1500 AADT.
iii) When the present year AADT of the left-turning traffic is greater than 750 AADT.

A left turn lane consists of a taper section, a deceleration section and a storage section. The
minimum lengths for taper sections are as for right turn lanes (Table 11.4). A detail of the
layout for a Left Turn Lane for a single carriageway is given in Figure 11.12; the
configuration for dual carriageways is shown in Figure 11.13. The length of the storage
section is as indicated in Table 11.5.

Table 11-5: Lengths of Storage Sections for Left Turn Lanes

Left-Turning Traffic Length of Storage Section
(AADT) (LS) (m)
0-1500 20
1500-3000 40
>3000 60

Provision of left turn lanes can be made for both the major and minor road. On single
roadway roads where a left turn lane is to be provided, a painted central reserve shall
always be used.

In order to accommodate a left turn lane on a single roadway road the roadway has to be
widened to provide the required width. The widening shall be designed so that the through

Page 11-16 Ethiopian Roads Authority

 Chapter 11
Geometric Design Manual – 2013 At-Grade Junctions

lanes are given smooth and optically pleasing alignments. The width of the through lanes
at the junction shall be the same as the approach lanes.

Figure 11-12: Layout for Left Turn Lane, Single Carriageway

Figure 11-13: Layout for Left Turn Lane, Dual Carriageway

Ethiopian Roads Authority Page 11-17

Chapter 11
At-Grade Junctions Geometric Design Manual – 2013

The widening shall be provided by the deviation of both through lanes from the centreline.
This shall be achieved by introducing a taper of 100-metres length at the beginning and
end of the widening.

The total length of an acceleration lane (i.e. not including the merging taper) is shown in
Table 11.6.

Table 11-6: Length of Acceleration/Merging Lane

Main road Length of Acceleration/Merging Lane
Length of
design
merging Entrance control speed (km/hr)
speed
taper (m)
(km/hr) 40 50 60 70 85 100
60 65 110 60 - - - -
70 75 180 130 70 - - -
85 80 315 265 205 130 - -
100 90 460 415 350 280 150 -
120 110 700 655 595 525 400 240

11.11 Turning Roadways

Turning movements are accommodated either within the limitation of the crossing
roadway widths or through the application of turning roadways. Turning roadways can be
designed for three possible types of operation:
• Case 1 One-lane one-way with no provision for the passing of stalled vehicles.
• Case 2 One-lane one-way with provision for the passing of stalled vehicles.
• Case 3 Two-lane one-way operation.
Three traffic conditions must also be considered:
• Condition A Insufficient trucks in the traffic stream to influence design.
• Condition B Sufficient trucks to influence design.
• Condition C Sufficient semi-trailers in the traffic stream to influence design.

The lengths of turning roadways at intersections are normally short, so that design for Case
1 operation is sufficient. It is reasonable to assume, even in the absence of traffic counts,
that there will be enough trucks in the traffic stream to warrant consideration, and
Condition B is normally adopted for design purposes. Widths of turning roadway for the
various cases and conditions are given in Table 11.7. The radii in the Table refer to the
inner edge of the pavement.

The simplest design is a semicircle which is adequate for medians of up to 3.0 m wide. For
medians wider than 3.0 m, a bullet-nose end treatment is preferred. The bullet-nose is
formed by two portions of control radius arc and an assumed small radius, e.g. 0.6 m. The
bullet-nose closely follows the path of the inner rear wheel of the design vehicle and

Page 11-18 Ethiopian Roads Authority

 Chapter 11
Geometric Design Manual – 2013 At-Grade Junctions

results in less intersection pavement and a shorter length of opening than the semi-circular
end. For wider medians, a bullet-nose end requires shorter lengths of opening.
Above a width of 5.0 m the minimum lengths to provide for cross-traffic become the
controlling factor. At this stage the bullet nose end should be replaced by a flattened bullet
nose, the flat end being parallel to the centre line of the crossing road.

Table 11-7: Turning roadway widths

Inner Case 1 Case 2 Case 3
Radius
Condition Condition Condition
(m)
A B C A B C A B C
15 4.0 5.5 7.9 6.1 8.8 13.4 7.9 10.7 15.2
20 4.0 5.2 6.7 5.8 8.2 11.0 7.6 10.1 12.8
30 4.0 4.9 6.4 5.8 7.6 10.4 7.6 9.4 12.2
40 3.7 4.9 6.4 5.5 7.3 8.8 7.3 9.1 10.7
60 3.7 4.9 5.2 5.5 7 8.2 7.3 8.8 10.1
80 3.7 4.6 5.2 5.5 6.7 7.6 7.3 8.6 9.4
100 3.7 4.6 4.9 5.2 6.7 7.3 7.0 8.4 9.1
150 3.7 4.3 4.6 5.2 6.7 7.3 7.0 8.2 9.1

Figure 11.14 shows the design of a typical cross road junction illustrating turning sections,
channelization islands, deceleration lanes, tapers, medians and mountable kerbs.

11.12 Private Access

A private access is defined as the intersection of an unclassified road with a classified road.
An access shall have entry and exit radii of 6 metres or greater, depending upon the turning
characteristics of the expected traffic. The minimum width shall be 3m. A typical access is
shown in Figure 11.15; dotted lines show the possible edge of the corresponding shoulders.
The location of the access must satisfy the visibility requirement for ‘stop conditions’
described in Section 11.6. A drainage pipe shall be placed as required. The access shall be
constructed back to the right-of-way line, with a taper to match the existing access.

Ethiopian Roads Authority Page 11-19

Chapter 11
At-Grade Junctions Geometric Design Manual – 2013

Figure 11-14: Typical Cross-road Intersection

(showing most features described in the text)

Page 11-20 Ethiopian Roads Authority

 Chapter 11
Geometric Design Manual – 2013 At-Grade Junctions

Figure 11-15: Typical Access

11.13 Roundabouts

11.13.1 Typical Features

A roundabout is a one-way circulatory system around a central island, entry to which is

controlled by markings and signs. Priority is given to traffic already on the roundabout.
Roundabouts thus operate by deflecting the vehicle paths to slow the traffic and promote
yielding. Figure 11.16 illustrates the features of a typical roundabout.

Roundabouts provide high capacity and minimum delay. They also have a good safety
record largely because traffic speeds are low and the number of potential traffic conflicts is
greatly reduced, for example, from 32 at a cross-roads to just 8 at an equivalent four-
legged roundabout.

Despite their advantages, roundabouts are not appropriate in every situation. They may be
inappropriate:
a) Where spatial restraints (including cost of land), unfavourable topography or high
construction costs make it impossible to provide an acceptable geometric design;
b) Where traffic flows are unbalanced, with high flows on one or more approaches
leading to serious delays to traffic on the major road;
c) Where there are substantial pedestrian flows;
d) As an isolated intersection in a network of linked signalised intersections;
e) In the presence of reversible lanes;
f) Where semi-trailers and/or abnormal vehicles are a significant proportion of the
total traffic passing through the intersection and where there is insufficient space to
provide the required layout; and

Ethiopian Roads Authority Page 11-21

Chapter 11
At-Grade Junctions Geometric Design Manual – 2013

g) Where signalised traffic control situated down-stream of the roundabout could

cause a queue to back-up through the roundabout.

Roundabouts should be considered when:

a) Intersection volumes do not exceed 3,000 vph (vehicles per hour) at three-legged or
4,000 vph at four-legged intersections;
b) The proportional split between the volumes on the major and minor road does not
exceed 70/30 where, on three-legged intersections, the flow is less than 1,500 vph
or, on four-legged intersections, is less than 2,000 vph;
c) The proportional split between the volumes on the major and minor road does not
exceed 60/40 where; on three-legged intersections, the intersection flow is greater
than 1,500 vph or; on four-legged intersections, is greater than 2,000 vph.
A typical roundabout is shown schematically in Figure 11.16.

Figure 11-16: Roundabout Layout

11.13.2 Design Speed and Safety

The design speed within the roundabout should, ideally, be between 40 and 50 km/h.
Unfortunately this suggests a radius of between 60 and 90 metres hence requiring an
overall diameter of the roundabout of the order of 175 metres. Very often, the space for
this size of intersection is not available and a lower design speed must be accepted.

Roundabouts should not be introduced directly on rural roads where the design speeds of
adjacent sections are 90 km/h or greater. Where the design speeds on the approaches are
high, i.e. more than 15 km/h faster than the design speed within the facility, it may be

Page 11-22 Ethiopian Roads Authority

 Chapter 11
Geometric Design Manual – 2013 At-Grade Junctions

necessary to consider forcing a reduction in vehicle speed. This could be by means of

horizontal reverse curvature. The ratio between the radii of successive curves should be of
the order of 1.5:1.

Consideration should be given to the use of rumble strips and warning signs at the
approaches to warn drivers to anticipate the roundabout. Speed humps should not be
employed as speed-reducing devices on major roads or on bus routes and are usually
unacceptable on approaches to major roundabouts. They are, however, sometimes
acceptable in urban environments for lower flows and lower speeds.

Roundabouts are usually more difficult for pedestrians to cross than normal junctions
hence arrangements should be made to provide adequate directions.

11.13.3 Sight distance

Approaching drivers should have a clear view of the ‘nose’ of the separator (or splitter)
island. At the yield line and while traversing the roundabout, they should have an
uninterrupted view of the opposing legs of the intersection at all times. This requirement
suggests that the elaborate landscaping schemes sometimes placed on the central islands of
roundabouts are totally inappropriate to the intended function of the layout.

Sight distance for intersections as described in Section 11.6 should be provided on each
approach to the roundabout to ensure that drivers can see the nose of the splitter island. It
follows that roundabouts should not be located on crest curves.

11.13.4 The General Layout

The various components of a roundabout are illustrated in Figure 11.16. The general layout
of a roundabout should provide for the following:
i) Deflection of the traffic to the right on entry to promote movement and ensure low
traffic speeds.
ii) Adequate entry widths.
iii) Suitable visibility at any entry of each adjacent entry.
iv) Adequate circulation space compatible with entry widths.
v) Central islands of diameter sufficient only to give drivers guidance on the
manoeuvres expected.
vi) A simple and clear layout.
vii) Entry and exit deflection angles and central island radius should prevent through
speeds in excess of 50 km/h. This is accomplished by maximising the difference
between the shortest route a driver can take through the roundabout versus the
straight-line distance from an entry to the opposite exit. No vehicle path should
allow a vehicle to traverse the roundabout at a radius greater than 100 metres (see
Figure 11.17).

Deflection
An important component is the deflection forced on vehicles on the approach to the
roundabout. The intention is to reduce the speed of vehicles so that, within limits, the
greater the deflection the better. The limit is the minimum acceptable angle of skew at an

Ethiopian Roads Authority Page 11-23

Chapter 11
At-Grade Junctions Geometric Design Manual – 2013

intersection of 60o. This corresponds to a deflection on entry of 30o. The approach radius
should not exceed 100 metres, which corresponds to the recommended design speed of 40
to 50 km/h.

Figure 11-17: Vehicle Path through Roundabout

Entries and exits

The widths of single lane approaches to roundabouts are typically of the order of 3.4 to 3.7
metres. The entry width is one of the most important factors in increasing the capacity of
the roundabout and can be increased above the width of the approach by flaring, i.e. by
providing a passive taper with a taper rate of 1:12 to 1:15. The recommended minimum
width for a single-lane entry is 5 metres. If demanded by high approach volumes, the
flaring may add a full lane to the entry to increase capacity. The width of a two-lane entry
should be of the order of 8 metres. A variation on the two-lane entry is to have a single-
lane circulatory roadway with an auxiliary lane provided for the benefit of vehicles turning
right at the roundabout. The auxiliary lane is shielded from traffic approaching from the
left by moving the end of the splitter island forward to provide a circulatory road width
adequate only for single-lane travel. This approach could be adopted with advantage when
right-turning traffic represents 50 percent or more of the entry flow or more than 300 vph
during peak hours.

Circulatory roadway
The circulatory roadway width is a function of the swept path of the design vehicle and of
the layout of the exits and entries and generally should be either equal to, or slightly
greater (1.2 times) than the width of the entries. The width should be constant throughout
the circle. In the construction of the swept path of the design vehicle, it should be noted
that drivers tend to position their vehicles close to the outside kerbs on entry to and exit
from the roundabout and close to the central island between these two points. The vehicle
path, being the path of a point at the centre of the vehicle, should thus have an adequate
offset to the outside and inside kerbs. For a vehicle with an overall width of 2.6 metres, the
offset should be not less than 1.6 metres, with 2.0 metres being preferred.

Page 11-24 Ethiopian Roads Authority

 Chapter 11
Geometric Design Manual – 2013 At-Grade Junctions

To ensure that vehicles do not travel faster than the design speed, the maximum radius on
the vehicle path should be kept to 100 metres or less (Figure 11.17). As a general
guideline, the circulatory roadway should be sufficiently wide to allow a stalled vehicle to
be passed. The minimum roadway width for single-lane operation under these
circumstances is of the order of 6.5 metres between kerbs. Two-lane operation requires a
roadway width of about 9.0 metres. If trucks are present in the traffic stream in sufficient
numbers to influence design, the circulatory road width should be increased by 3 metres
both in the single-lane and in the two-lane situation. A significant proportion of semi-
trailers requires the width of the circulatory road width to be increased to 13 metres and 16
metres in the single-lane and the two-lane situation respectively. A circulatory road width
of 13 metres makes it possible for passenger cars to traverse the roundabout on relatively
large radius curves and at correspondingly high speeds. To avoid this possibility, the
central island should be modified as discussed below.

The cross-slope on the roadway should be away from the central island and equal to the
camber on the approaches to the intersection.

Central Island
The central island consists of a raised non-traversable area (except in the case of mini-
roundabouts) that is usually circular. The island is often landscaped but the landscaping
must not obscure the sight lines across the island. A mountable area or apron may be added
to the central island to accommodate occasional Large Heavy Vehicles and to allow the
circulatory width to be reduced to 9.5m. The apron should have crossfall steeper than that
of the circulatory road, principally to discourage passenger vehicles from driving on it and
a crossfall of 4 to 5 percent is recommended.

Splitter Islands
Splitter islands should be provided on the approaches to roundabouts to:
a) Allow drivers to perceive the upcoming roundabout and to reduce entry speed;
b) Provide space for a comfortable deceleration distance;
c) Physically separate entering and exiting traffic;
d) Prevent deliberate and highly dangerous wrong-way driving;
e) Control entry and exit deflections; and
f) Provide a refuge for pedestrians and cyclists and a place to mount traffic signs.
The sizes of splitter islands are dictated by the dimensions of the central island and
inscribed circle. As a general guideline they should have an area of at least 10 square
metres so as to ensure their visibility to the oncoming driver. The length of splitter islands
should be equal to the comfortable deceleration distance from the design speed of the
approach to that of the roundabout. Ideally, the nose of the splitter island should be offset
to the right of the approach road centreline by about 0.6 to 1 metre.

The roundabout depicted in Figure 11.18 and in the Standard Detail Drawings is acceptable
for traffic volumes of up to 15,000 (based on empirical evidence rather than gap-
acceptance theory).

The following steps may be followed in laying out trial geometry for a roundabout:
1. Select the general design criteria to be used.

Ethiopian Roads Authority Page 11-25

Chapter 11
At-Grade Junctions Geometric Design Manual – 2013

2. Select the appropriate design vehicle for the site. This will generally be the DV4
for all design standards.
3. Adopt a minimum design vehicle turning radius. This will generally be 15m radius.
Check the design using the template in Chapter 5.
4. Determine from traffic flows the number of lanes required on entry, exit and
circulation.
5. Identify the needs of pedestrians.
6. Identify the location of controls such as right-of-way boundaries, utilities, access
requirements, and establish the space available.
7. Select a trial central island diameter and determine the width needed of the
circulating carriageway.
8. Draw the roundabout.
9. Check that the size and shape is adequate to accommodate all intersecting legs with
sufficient separations for satisfactory traffic operations.
10. Lay out the entrance/exit islands.
11. Check the achievement of adequate deflection. Adjust as required.
12. Check site distances at approaches and exits.
13. Layout lane and pavement markings.
14. Layout lighting plan
15. Layout sign plan.

Figure 11-18: Roundabout Dimensions

Page 11-26 Ethiopian Roads Authority

 Chapter 11
Geometric Design Manual – 2013 At-Grade Junctions

11.14 Checklist for Junction Design

The following is a checklist of factors that need to be considered in the design of junctions.

Will the junction be able to carry the expected/future traffic levels without
becoming overloaded and congested?

Have the traffic and safety performance of alternative junction designs been
considered?

Is the route through the junction as simple and clear to all users as possible?

Is the presence of the junction clearly evident at a safe distance to approaching
vehicles for all directions?

Are warning and information signs placed sufficiently in advance of the
junction for a driver to take appropriate and safe action given the design speeds
on the road?

On the approach to the junction, is the driver clearly aware of the actions
necessary to negotiate the junction safely?

Are turning movements segregated as required for the design standard?

Are drainage features sufficient to avoid the presence of standing water?

Is the level of lighting adequate for the junction, location, pedestrians, and the
design standard?

Are the warning signs and markings sufficient, particularly at night?

Have the needs of pedestrian and noon-motorized vehicles been met?

Are sight lines sufficient and clear of obstructions including parked and stopped
vehicles?

Are accesses prohibited a safe distance away from the junction?

Have adequate facilities such as footpaths, refuges, and crossings, been
provided for pedestrians?

Do the design, road marking and signing clearly identify rights of way and
priorities?

Is the design of the junction consistent with road types and adjacent junctions?

Are the turning lanes and tapers where required of sufficient length for speeds
and storage?

Date: ................... Designer................................................

Date: ................... Responsible Engineer............................

Figure 11-19: Checklist for Junction Design

Ethiopian Roads Authority Page 11-27

 Chapter 12
Geometric Design Manual – 2013 Grade Separated Junctions

12 GRADE – SEPARATED JUNCTIONS AND INTERCHANGES

12.1 General

Grade-separated interchanges are usually associated with freeways. The fundamental

difference between a freeway and any other road is that it is subject to rigid control of
access. Entrance to and exit from a freeway may take place only at specified points,
typically remote from each other, and then only at very flat angles of merging and
diverging. As such, the freeway is characterised by the fact that all intersections along its
length are interchanges.

However, at any road junction, the flow of traffic, expressed in terms of a level of service,
can be enhanced by an interchange. Where two main roads intersect, it may be found that
traffic volumes are too high to be accommodated at an at-grade intersection, regardless of
the level of sophistication of the provision made for turning movements by means of
channelization, signalization and auxiliary lanes for through traffic. Generally, if an
intersection is likely to become a bottleneck, and all possibilities for improving its capacity
have been exhausted, an interchange is warranted. In the planning of a new road under
circumstances where close spacing of heavily trafficked intersections is anticipated, it may
be necessary to consider the provision of interchanges at points where the design level of
service cannot be achieved with at-grade intersections.

Interchanges are divided into two functional classes, referred to as Access (or Minor or
Service) interchanges, and Systems (or Major) interchanges. Access interchanges serve
local areas by providing access to freeways whereas systems interchanges are the nodes of
the freeway network, linking the individual freeways into a cohesive unit. These two
fundamentally different applications require different types of interchange layout.

An Access interchange provides free flow conditions for the main freeway with suitable
ramps to ensure that entering and exiting traffic do not affect traffic speeds on the freeway.
However the associated intersections with the minor road (or roads) are at-grade and
include the normal yield, stop and traffic controlled options described in Chapter 11.

In contrast, Systems (or Major) interchanges provide free flow conditions for more than
one freeway and ensure unhindered travel for all vehicles on them. Such interchanges are
very much more complex to design and build.

Thus the use of grade separation results in the separation of traffic movements between the
intersecting roads so that only merging or diverging movements remain. The extent to
which individual traffic movements should be separated from each other depends mainly
upon capacity requirements and traffic safety aspects; it also depends upon the extent to
which important traffic movements should be given free flow conditions.

The circumstances in which the use of a grade separated junction is warranted are usually
as follows:
i) An at-grade junction has insufficient capacity.
ii) The junction is justified economically from the savings in traffic delays and
accident costs.

Ethiopian Roads Authority Page 12-1

Chapter 12
Grade Separated Junctions Geometric Design Manual – 2013

iii) Grade separation is cheaper on account of topography or on the grounds that

expensive land appropriation can be avoided by its construction.
iv) For operational reasons.
v) Where roads cross freeways/motorways.

12.2 Choice of Scheme

In deciding on the location of a grade-separated junction, the following factors should be

taken into account:
i) Predicted traffic volumes.
ii) Cost of junction.
iii) Congestion control.
iv) Trip length (travel distance).
v) Size of urban areas.
From a study of conflicting traffic movements, it will generally be apparent which traffic
streams must be grade separated, leaving the other streams to be dealt with by junctions at
grade; the choice of these will depend upon the capacities needed. A study of the
characteristics of various types of grade-separated junctions is necessary, and a number of
alternative designs should be prepared. The final choice of scheme must satisfy capacity
requirements, geometric standards, and operational needs, and represent an economical
design. In some instances the choice of a particular design will be determined by the
adoption of two-stage construction, e.g. constructing an at-grade junction first and
providing grade separation later.

12.3 Types of Junction

12.3.1 Common Types

Figure 12.1 illustrates the common types of interchange. Grade-separated junctions

generally fall into four categories depending upon the number of roads involved and their
relative importance. These categories are as follows:
1) Three-way junctions.
2) Junctions of major/minor roads.
3) Junctions of two major roads.
4) Junctions of more than two major roads.
Each category is discussed briefly below with reference, where appropriate, to the basic
line diagram layouts shown in Figure 12-1.

12.3.2 Three-Way Junctions (Layouts A and B)

For some Y-junctions where grade separation of only one traffic stream is required, layout
A may be appropriate. The movements associated with the missing leg must be channelled
to another location and this is only appropriate if the traffic volumes on the missing leg are
low and capable of being served by an at-grade junction elsewhere.

Page 12-2 Ethiopian Roads Authority

 Chapter 12
Geometric Design Manual – 2013
201 Grade Separated Junctions

Layout B shows a typical three-leg

three leg junction. It is appropriate for traffic where the major
road is DC8 and the minor road is DC7 - DC3. This configuration is appropriate
propriate for traffic
volumes of up to 30,000 AADT on the four-lane
four lane major road (3,000 vehicles per hour).
With a single loop lane, it is appropriate for loop traffic of 1,000 vehicles per hour. Higher
loop traffic requires multiple loop lanes.

Figure 12-1: Typical Layouts for Grade-Separated

Grade Separated Junctions

12.3.3 Junctions of Major/Minor Roads (Layouts C and D)

Layouts C and D are the simplest for major/minor road junctions and both transfer the
major traffic conflicts
icts to the minor road. Layout C shows the ‘half clover leaf’ type of
junction, which has the advantage of being easily adapted to meet difficult site conditions.
Chapter 12
Grade Separated Junctions Geometric Design Manual – 2013

Layout D shows the normal ‘diamond’ junction, which requires the least land
appropriation. The choice between these options is generally dependent on land
requirements.

These configurations are appropriate for traffic volumes of up to 30,000 AADT on the
four-lane major road (3,000 vehicles per hour), with traffic of up to 10,000 ADT on the
minor road. They are appropriate for traffic where the major road is DC8 and the minor
road is DC7 - DC3. With a single loop lane, it is appropriate for loop traffic of 1,000
vehicles per hour. Higher loop traffic would require multiple loop lanes.

12.3.4 Junctions of Two Major Roads (Layouts E and F)

Layouts E and F show the two basic junction layouts for use where high traffic flows make
the simpler layouts unsatisfactory. They are appropriate for traffic volumes on both
crossing roads of between 10,000 and 30,000 AADT (3,000 vehicles per hour). Layout E
shows a ‘full clover leaf’ junction involving only one bridge but requiring a large land
appropriation. Layout F shows a typical roundabout interchange involving two bridges.
This layout is only suitable if the secondary road containing the roundabout is of a
relatively low design speed but carries a comparatively high volume of traffic.

12.3.5 Junctions of more than Two Major Roads

Junctions of more than two main roads are difficult to design, occupy large areas of land
and, requiring numerous bridges, are extremely expensive. This type of junction, although
unlikely to be required on rural roads in Ethiopia, can often be reduced by changes in the
major road alignments (which will simplify the traffic pattern) to a combination of the
simpler and economic layouts described above.

12.4 Geometric Standards

12.4.1 Design Speed

The geometric standards given in this manual for roads and at-grade junctions also apply to
grade separated junctions. However, the low design speeds of loops and other ancillary
roads necessitate further standards to be given.

The design speed for the through traffic movements is determined in accordance with
Chapter 5. Stopping sight distances appropriate for the design speed should always be
provided.

The design speed for loops and ramps is dependent on whether their terminations are free
flowing or a stop junction. The term ‘free flowing’ implies that the ramp terminals can be
negotiated at more or less the speed prevailing on the through road. Traffic on the
terminals thus diverges from or merges with traffic on the through road at very flat angles

For the ramps or loops of access-type interchanges, where the end of the exit loop
terminates at a road junction, the design speed should, ideally, be 40-50 km/hr. Higher
design speeds require higher radii of curvature and longer loops and therefore have a
significant cost implication. However the design speed should not be so low that it is
requires drivers who are leaving the freeway to reduce speed too quickly hence either

Page 12-4 Ethiopian Roads Authority

 Chapter 12
Geometric Design Manual – 2013 Grade Separated Junctions

compound curves are required suitable for an entry speed of 65% of the design speed of the
freeway or a deceleration lane must be provided on the freeway.

If a high volume of turning (exiting) traffic is expected, free flowing terminals at each end
of the loop or ramp will accommodate traffic entering and leaving at speeds close to the
operating speeds of the through and intersecting roads. A lower design speed in the middle
of the loop or ramp will have a restrictive effect on the capacity of the ramp and is
therefore unacceptable. Deceleration and acceleration lanes must also be provided on the
freeway (Section 11.10).
Where a dual carriageway intersects with another dual carriageway (Major Interchange),
the interchange between the facilities must be designed so that the loop roads do not entail
any significant reduction in the design speeds of the crossing carriageways.

12.4.2 Acceleration and Deceleration Lanes

The minimum standards to be applied for right turn deceleration lanes are the same as for
at-grade junctions (Section 11.10). The total length of the acceleration lane (i.e. not
including the merging taper) shall never be less than 150 metres or more than 400 metres.

12.4.3 Horizontal Curves and Super-elevation

The geometric principles described in this manual apply equally to the ramps for
interchanges. The maximum super-elevation for loops is e = 8% which, at a design speed
of 50 km/h, leads to a minimum radius of 80 metres. Where smaller radii are unavoidable,
warning signs are necessary.

Where transitions occur from high to low speeds the curves must be compound or
transitional, the radius at any point being appropriate for the vehicle speed at that point.

12.4.4 Vertical Curves and Gradients

To ensure reasonable standards of visibility, comfort and appearance, vertical curves

should be introduced at all changes in gradient. Vertical curve lengths should be
determined in accordance with Chapter 9 to provide safe stopping sight distances.

12.4.5 Widths and Gradients of Loops

The minimum carriageway width for loops on straight sections and horizontal radii greater
than 150m shall be 4.0m with shoulders of 1.5 metres on the near side and 1.0 metre on the
far side (widened by 0.5 metre where a guardrail is required). For loops on radii of 150
metres or less, the carriageway width shall be in accordance with Table 12.1.

The maximum up gradient should be 5% and the maximum down gradient should be 7%.

Table 12-1: Widths for Loops

Radius (metres) 25 30 40 50 75 100 150
Carriageway Width (m) 5.3 5.0 4.6 4.5 4.5 4.5 4.0

Ethiopian Roads Authority Page 12-5

Chapter 12
Grade Separated Junctions Geometric Design Manual – 2013

12.4.6 Clearances

The required vertical and horizontal clearances shall be in accordance with Chapters 6 & 9.

12.4.7 Capacity

Grade-separated junctions are generally designed using traffic volumes given in terms of
the Daily High Volume (DHV) rather than Annual Average Daily Traffic (AADTs). A
detailed traffic study and analysis can be made to determine these values. In the absence of
such a study, it can be assumed that the DHV in an urban area is 10% of AADT. The
capacity of each traffic lane, in DHV, is normally about 1000 vehicles per hour.

Thus, for example, Table 2.1 indicates a design traffic flow of 10,000 to 15,000 AADT for
Design Class DC8. The expected DHV is therefore 1000 to 1500. The capacity of this
facility would be exceeded at more than 1000 vehicles per hour per lane, which equates to
4,000 vehicles per hour for all four lanes, hence capacity will not be exceeded at 15,000
AADT.

These DHV values are necessary in choosing the number of lanes for the loops
corresponding to the junction.

12.4.8 Basic Lanes and Lane Balance

Basic lanes are those that are maintained over an extended length of a route, irrespective of
local changes in traffic volumes and requirements for lane balance. Alternatively stated,
the basic number of lanes is a constant number of lanes assigned to a route, exclusive of
auxiliary lanes.

The number of basic lanes changes only when there is a significant change in the general
level of traffic volumes on the route. Short sections of the route may thus have insufficient
capacity. This problem can be overcome by the use of auxiliary lanes. In the case of spare
capacity, reduction in the number of lanes is not recommended because this area could, at
some future time, become a bottleneck. Unusual traffic demands, created by accidents,
maintenance or special events, could also result in these areas becoming bottlenecks.

The basic number of lanes is derived from consideration of the design traffic volumes and
capacity analyses. To promote the smooth flow of traffic there should be a proper balance
of lanes at points where merging or diverging manoeuvres occur. In essence, there should
be one lane where the driver has the choice of a change of direction without the need to
change lanes.

At merges, the number of lanes downstream of the merge should be one less than the sum
of the number of lanes upstream of the merge plus the number of lanes in the merging
ramp. This is typified by a one-lane ramp merging with a two-lane carriageway that, after
the merge, continues as a two-lane carriageway as is the case on a typical Diamond
Interchange layout. This rule precludes a two-lane ramp immediately merging with the
carriageway without the addition of an auxiliary lane.

At diverges, the number of lanes downstream of the diverge should be one less than the
total number upstream of the diverge plus the number of lanes in the diverging ramp. The

Page 12-6 Ethiopian Roads Authority

 Chapter 12
Geometric Design Manual – 2013 Grade Separated Junctions

only exception to this rule is on short weaving sections, such as at Cloverleaf Interchanges,
where a condition of this exception is that there is an auxiliary lane through the weaving
section. When two lanes diverge from the freeway, the above rule indicates that the
number of freeway lanes beyond the diverge is reduced by one. This can be used to drop a
basic lane to match anticipated flows beyond the diverge. Alternatively, it can be an
auxiliary lane that is dropped.

Basic lanes and lane balance are brought into harmony with each other by building on the
basic lanes, adding or removing auxiliary lanes as required. The principle of lane balance
should always be applied in the use of auxiliary lanes. Operational problems on existing
roadways can be directly attributed to a lack of lane balance and failure to maintain route
continuity. The application of lane balance and coordination with basic number of lanes is
illustrated in Figure 12.2.

Figure 12-2: Principles of Lane Balance

Ethiopian Roads Authority Page 12-7

Chapter 12
Grade Separated Junctions Geometric Design Manual – 2013

12.4.9 Minimum Spacing of Successive Interchanges

The distance between two successive grade-separated junctions is an element of great

importance in ensuring the desired level of service. In suburban zones it is necessary to
establish a minimum distance between successive grade-separated junctions. The
recommended minimum distance is 2.0km.

12.5 Design Principles

Special design principles apply to grade-separated junctions and must be considered when
comparing the characteristics of alternative designs. The main principles and described
below:
1. The high speeds normally found on roads where grade separation is required and
the low design speeds of ancillary roads make it necessary to pay particular
attention to the transitions between high and low speed. This not only influences the
use of long speed-change lanes and compound curves but also the choice of types
of interchange which do not result in abrupt changes in vehicle speeds.
2. Weaving between lanes on the main roadway within the interchange is undesirable
and can be avoided by arranging for diverging points to precede merging points.
3. On a road with a large number of grade-separated junctions, a consistent design
speed is desirable for loops. This speed shall be not less than 65% of the speed of
the adjoining major road.
4. As a general rule, left-turning movements that are grade separated should be made
through a right-hand loop.
5. Unexpected prohibited traffic movements, especially where traffic is light, are
difficult to enforce and cause danger. If possible, the geometric layout should be
designed to make prohibited movements difficult, for example on one-way loops,
entry contrary to the one way movement can be restricted by the use of suitably
shaped traffic islands to supplement the traffic signs.

Page 12-8 Ethiopian Roads Authority

 Appendix I
Geometric Design Manual – 2013 Perspective in Geometric Design

Ethiopian Roads Authority Page I-1