OUTLINE NOTES

ENGINEERING SURVEYING
 ENGINEERING SURVEYING

INTRODUCTION
Definition: Any survey carried out to support /facilitate the design,
construction and maintenance of engineering works.
Three components
- Setting out
- Earth works
- Deformation monitoring

1. ASPECTS OF ENGINEERING SURVEYING.

- Setting out for engineering constructions e.g. roads, railways, bridges,

huge buildings, dams, tunnels etc.
- Monitoring the deformation trend of engineering structures e.g. dams,
huge buildings, bridges etc.
- Industrial alignments e.g. installation of machines, mine field,
alignments, crane routes etc.
- Vertical alignments of high rise buildings etc.
- Prediction of deformation behaviour of engineering structures on the
basis of the already known deformation pattern of the natural ground.
This is a design requirement when, say, a structure is to be constructed
within, say, the Rift Valley.
- Mining surveying
- Any other surveying required in engineering works e.g. leveling for
drainage or railway or pipeline construction. etc.

2. ROLE OF ENGINEERING SURVEY.

To provide information that may be used to protect human life & property
and guarantee the safety of engineering structures. This main role may be
broken down into several parts.
i. Adequate monitoring of structural behaviour may lead to
accurate prediction of natural catastrophies such as earth
quakes, which in effect, would lead to the protection of
human life and property incase such natural disasters
occur.
ii. Risk to existing engineering structures due to earth
movements may be reduced through consistent
monitoring of the deformation behaviour of such
structures.
 iii. Aiding of economic planning and development through
provision of vital information required at the design stage
when structures are to be constructed.
iv. Aiding exploitation of natural resources. Major earth
resources are closely related to past movements of the
earth’s crust. The distribution of mineral deposits,
geothermal energy resources, oil wells etc. are intimately
tied to the tectonic movements. Hence, monitoring of
earth movements would help in the exploitation and
exploration of natural resources.
v. Corroborate/complement geological/geophysical data in
search of a better understanding of earth movements.
Through engineering surveying (deformation
monitoring), linear measurements may be obtained and
compared with geological/geophysical evidence
regarding earth movement.

PRINCIPLE OF ENGINEERING SURVEYING.

Largely same principles of general land surveying but guided or distinguished

by the following factors:-

i. Small scale operations requiring large scale maps or plans.

ii. High accuracy requirements.
iii. Higher need for statistical analysis.
iv. More emphasis on differences in measured values (horizontal
measurements and heights).
v. Repeat measurements from time to time.
vi. More frequent need for calibration of equipments.
vii. Results may be based on arbitrary (riot actual) controls.
viii. Mostly uses high precision EDM’s and precise leveling methods during
measurements..
ix. First-hand-information may be obtained through the analysis of raw-
data.
x. Because of high accuracy, values of the meteorological data are
essential.
 3. TECHNIQUES EMPLOYED IN ENGINEERING SURVEYING.

i. Terrestrial techniques to provide horizontal information.

- EDM measurements.
- Use of T3 theodiolites (accuracy 0‫ ״‬.1)
- Triangulation based on fixed or relative deformation
nets

ii. Precise leveling or any other appropriate leveling

method:
- Provide vertical information.
- Used to determine the rate of settlement or uplift of the
Structure.
- Used to determine differences in height which are
required in drainage, pipeline, railways constructions
etc.

iii. Global Positioning System (GPS)

- Provide both horizontal and vertical information.
- Appropriate when accuracies up to the centimeter level
are sought e.g. terrain movements, coastal deformations
etc.
iv. Other techniques.
- Photogrammetry.
- Remote sensing.
(1) SITE CONSTRUCTION

Control networks must be established in order to connect the site to the

national survey control grid. These horizontal and vertical controls will
eventually form the basis for the subsequent construction surveys.

❖ In construction surveys, precision is allowed to vary as follows:

- For excavation purposes: elevation given to 3cm

- For points on the structure: elevation given to three 3mm
- For alignment purposes: measurements to 3mm
- For prefabricated steel structures or members, high precision is required
e.g. <3mm.
- Ordinary measurements in construction to 3mm.
- Measurements for construction of buildings, bridges and dams given
1.5mm.

(2) HIGHWAYS AND STREETS.

Normally various horizontal and vertical control points are made available
along highway routes before the construction process is started. These
controls are established during the planning and design stage of the project.
These controls facilitate the necessary setting out before the construction
process is started. Highway construction surveys may be approached in two
ways:

- A layout from centerline controls and

- A layout from random control stations.

Choosing any one of the two approaches depends on terrain configuration,

vegetation cover, equipments available, construction requirements etc.

Setting out and construction of steel requires a procedure of surveying simiral

to that of highways. For streets, normally the curb is built first. The line and
grade for the top of each curb are indicated by hubs driven outside the curb
line at 15m intervals. Cut and fills are computed marked on the gravel stakes
next to each hub and are also tabulated in field notebook. The surveys for
street setting out and construction should identify the location of all surface
and underground utilities that may affect the project e.g. rail line, sewer line ,
underground water pipes etc. The planning at the design stage should address
and accommodate such utilities in the process of setting out and constructing a
street.
(3) RAILWAYS.

Setting out and construction surveys for railways are similar to those of
highways and streets. Before the construction process is started:-
- The located centre line is rerun.
- Missing stakes are replaced
- Control station hubs are referenced
- Borrow pits are staked out and slope stakes set.
- Lines and grades for structures are established on the ground

After grading the roadbed, alignment is established by precisely setting stakes

along the centerline at full stations on tangents and usually at intermediate
stations on horizontal and vertical curves. Spiral curves are also staked out at
this time. An additional line of hubs is set on one side of the track about 1m
from the proposed centre line of the rail, with the top of the hub usually at the
elevation of the rail. Track is usually laid on the sub grade and is lifted into
position offset the ballast has been placed and compacted around the ties.

4. SEWER AND PIPELINES.

(a) Sewer: The centerline is located on the ground using stakes (or any
other marks). The stakes are usually set at 15m intervals for uniform
grades and as close as 3m for vertical curves. At one side of the line,
safe distances from the excavation area, a parallel line of hubs is set
with the hubs at the same interval as those of the center line. In paved
streets or hard roads where it is not possible to put stakes or pegs, the
line and the grade are marked with spikes (driven flush), chisel marks
or paint marks.

When the stretch has been excavated, batter boards are set across the
stretch at the same interval used for station. The top of the board is set
at fixed whole number of decimeters above the sewer invert (inside
surface of bottom of sewer pipe). Then a measuring stick of the same
length is prepared. A nail is driven on the top of each batter board to
define the line. A cord is stretched tightly between these nails as the
sewer is being laid out. Then, the face end of each section of the pipe is
set at the proper distance below the cord a determined by measuring
with stick.
 (b) Pipelines: The procedure is similar to that of sewers except for the
following:
i. The interval between grade hubs or batter boards may be greater.
ii. Laying the pipes does not require as much care as laying the
sewer.
For both sewer and pipelines, the extent of excavation are computed to
provide basis of payment to the contractor. As usual, the survey data should
include the location of underground utilities close by or adjacent to the trench.
Laser alignment techniques may also be used to replace the method of batter
boards and string lines.

The simplest approach involves the use of laser-equipped theodolite to

establish a laser – beam parallel with flow line of the sewer or pipeline. The
alignment would first be used to provide rough grading to get the ditch
excavated and then for placing the pipe on line and at grade with the ditch.

5. TUNNELS
In tunnel surveys, a coordinate system based on field measurements and
computation of lengths, direction and the slope of line connecting given points
and the laying out of that line. For a short tunnel (e.g. highway tunnel through
a ridge) the following steps are needed:
i. Run a traverse and level between terminal points.
ii. Compute the length, direction and grade of the connecting line.
iii. Establish permanent monuments outside the tunnel on the
centerline at both ends.
iv. To fix the direction of the tunnel on each side of the ridge,
establish additional points at convenient surface location on the
centre line.
v. As constructions proceeds, the line at the end is given by setting
up at the permanent monument outside the tunnel, sighting the
fixed point on line and then setting points along the tunnel.
vi. Direct leveling to point on either the roof or the floor to provide
the grade. Distances are measured from the permanent
monuments to the station along the tunnels.

Rail or road tunnels or other tunnels in mountainous countries are often

several km long and are not uniform in either slope or direction. Since tunnels
are driven both from the ends and from intermediate points where shafts are
sunk to intersect the centre line of the tunnel. Surveys from such tunnels
include precise triangulation and/or triangulation systems tied to monuments
at the portals of the main tunnel and at the entrances of shafts. In addition a
precise system of differential levels connecting the same points is required.
From such surveys, the length, direction , and slope of each of the several
sections of the tunnel are calculated and the construction is controlled by
establishing these lines and grades as the work progresses.

Tunnel alignment may also be achieved through laser equipped instruments.

In addition, Electromagnetic Distance Measuring (EDM) instrument is also
quite useful in establishing distance to theodolite stations, points of curvature,
and chord points as the tunnel is set out. A gyroscopic theodolite is used to
establish absolute direction (azimuths), periodic determination of the azimuths
of the centre line of the tunnel, provides the necessary checks for maintenance
purposes.

6. BRIDGES AND CULVERTS.

Location survey provides sufficient information needed for the design of the
small bridges and culverts. However, a special topographical survey of the site
is required for long bridges and for grade-separation structures.

The site map should have the following features:-

- Scale: 1:1200 with5m contours for long bridges with rough terrain, and
scale 1:600 with contour interval of 1m urban regions and where terrain
is level.
- All data of location survey, line and grade of the roadway and the
marking and referencing of all survey points.
- Hydrographic survey which provides a continuous profile of the bottom
of the river at 3 to 8m intervals is required where the bridge crosses
rivers or other water bodies.

Planning for horizontal and vertical control for design mapping needs to
incorporate the following factors:

- Control points to be suitably located for the construction surveys.

- Long crossings over water require elaborate first-order triangulations
and/or trilateration to facilitate the setting out process before the
construction is started.
To establish the location of offshore structures simultaneous sights are taken
from two appropriate control points so as to locate the points by intersections.
Armed with EDM and theodolites, points may be located by distance and
direction and the survey accomplished in stages as deemed necessary. Laser-
equipped instruments may then be employed to provide alignments and
vertical positions before the construction process is started. The designed
construction specifications are then followed until the bridge is completed.

On completion of the structure permanent survey points are established to

assist in future monitoring of the deformation behaviour of the bridge
structure.

Setting out of a culvert requires the following steps:-

- At the intersection of the centre line of the culvert with the located
survey line of the highway or railway, turn off the angle of intersection.
- Project survey line, defining the direction of a culvert, for a short
distance beyond its ends and reference that line.
- If excavation is necessary for the channel to and from the culvert, stake
it out in the usual way it’s done for a roadway cut.
- Establish benchmarks near by and set hubs for convenient leveling to
the culvert. Establish lines and grades required for a particular type of
culvert.

7. DAMS

The design and setting out of a dam requires the following:

i. Topographic survey required to determine the feasibility of the project,
the approximate site of the reservoir and the optimum location and
height of the dam.
ii. Extensive soundings and borings are made and topography considered
in detail so as to fully define the dam, the crane routes roads and other
structure within the surrounding.
iii. A property line survey is made of the area to be covered by or directly
affected by the proposed reservoir.
iv. Before the construction is started adequate horizontal and vertical
control points are established at the appropriate locations preferably
using a rectangular coordinate system,
 v. The information gathered above are used to aid the design of the dam
and finally facilitate and guide the setting out process before the actual
construction is started.

Vertical controls are usually established through precise leveling, while,

horizontal controls are established through the use of theodolite and EDM’s.
Permanent controls for future deformation monitoring are also established.

8. BUILDINGS.

Appropriate horizontal and vertical control points are required in the design
and setting out of a building before the actual construction process
commences. Very accurate controls involving precise leveling and EDM
measurements are required; more so when storey building is to be constructed.
Precise angular measurements using theodolites need to be made to facilitate
building alignments and the subsequent excavations.

At the beginning of the excavation, the corners of buildings are marked by

stakes (pegs) which are eventually lost as the excavation progresses. The
setting out requires that sighting lines are established and referenced on each
outside building line and line of column.

Adequate permanent horizontal and vertical control points need to be

established as soon as the construction process is complete. This provides the
monitoring scheme to detect structural deformation behaviour of the building
in a periodic pattern.
Measurements of Distances
Distance used in plans relates to horizontal distance.

1. Chain
Old surveys were sometimes done with the surveyor’s (or Gunter’s) which was 60ft in
length and divided to 100 links each link 7-9 inches long.
Later the engineer’s chain which was 100ft long came into use. It had 100 links, each 1ft
long. All now obsolete.

2. Steel Tapes.
For accurate measurements, steel tapes are used most commonly. The common lengths are
50ft, 100ft, and 200ft although longer tapes are available. Tapes graduated in miter units
are available in lengths of 25, 30, 50, 100 metres.

3. Metallic Tapes.
Sometimes called woven tapes. Are cloth tapes with fine tinsel twisted around each thread
to prevent excessive stretching. They are made in lengths of 25, 50, 100ft.
Not recommended for precise results. They are used also for less accurate measurements.

4. Cloth Tapes.
Without woven wires, stretch so much as to be practically useless in surveying.

5. Invar Tapes.
Are made of nickel steel alloy, having a very small thermal coefficient of expansion. They
are used for surveys requiring a high degree of precision. For a given change in
temperature, the change in lengths of these tapes is about 1/30 to that of ordinary steel tape.

6. The adometer
Is an instrument attached to the wheel of a vehicle to measure it’s revolutions. Which
when multiplied by circumference, gives slope distance covered. This is rather inaccurate
instrument and is of little use in surveying.

7. The Stadia.
Distance ca be measured very rapidly by the stadia method where the distance is obtained
by sighting the telescope of a theodolite on a graduated rod held at the other end of the line,
and observing the interval or the rod included between two special horizontal cross-heirs
(stadia) set in the telescope. From the observed interval, the distance from the instrument to
the rod is readily computed.

This is a rather simple and fast approach but the drawbacks are the limited accuracy due to
the direct reading of the rod and the fact that the distances obtained are slope distances.

8. Substence bar
This device consists of two targets supported on each end of a metal bar or tube mounted
horizontally on a tripod. Distances are obtained by measuring angles sustended by the
targets.
Theodolite
Horizontal
 9. E.D.M.
These distance measurements are accomplished by transmitting beams of electromagnetic
waves between instruments located at the ends of the line to measured. THhe beams are
modulated precisely so that by comparing the transmitted and received beams, at one or
both ends of the lines, the distance can be determined.

Accurate distance determination depends upon accurate knowledge of the velocity of the
particular electromagnetic variation used, involving correction for temperature,
atmospheric pressure and relative humidity. These instruments generally require that the
ends of the lines can be intervisible and occupiable.

Advantages - Long distances

- High accuracy
- Less time of measurements
- Measurements are independent of roughness of terrain.

Distances may be measured easily when taping is difficult if not impossible.

i. Instrument utilizing radio waves.

These instrument system use the longer waves (several hundred meters of wavelengths)
over long ranges (hundreds of miles), yielding a limited absolute accuracy. Their main uses
are hydrographic and oceanographic surveying and in navigation where long ranges are
essential while accuracy equipments are rather relaxed. Cubic auto tape, Decca, Shoran,
Hiran, Loran and Omega.

ii. Microwave Instruments.

Utilizing wavelengths of a few centimeters. These instruments are capable of reaching
absolute accuracies of better the ±2cm over distance of up to 100 km.
The instrument system consists of two units, the master and the slave (or remote unit). The
master unit is located at one end of the line while remote unit is located at the other end.
Each unit transmits a modulated beam which is independent of the beam being received
from the other unit. This beam may be visualized as constructing a series of tape length
placed end to end traveling through space at the velocity of the microwaves.

The beam being received at each end of the line is compared to that being transmitted. The
results of the phase comparisons are then combined to provide the measurement data.
- Meteorological data to be obtained
- Successful measurement in haze, light rain, mist, light bush for there should be no
intervening terrain.
- Effects of the deflected water waves over water bodies and highways create
interference (noise).
 Microwave instruments

Instrument Range Reading Accuracy

Electro tape 10m - 50km 1cm
Distometer 50m – 100km 2cm
MRA2 100m - 150km 5cm
MRA3 30m – 80km 15cm
MRA4 50m – 50km 3cm
MRA101 100m – 50km 1.5cm
Distomat 50m – 150km 2cm

iii. Instrument Utilizing Visible Light

The use of electromagnetic waves in the visible light range has led to an additional
increase in accuracy, mainly because there extremely short wavelengths are less
affected by atmospheric conditions.

However, the range is considerably reduced especially for daylight operation. At night
and with a high powered light source, such as a laser, the range can be significantly
increased but will hardly be as long as for microwaves instruments.

For light waves, special reflectors are used. The reflectors consists of one or more
retroprisms which return light rays to the source, even if not placed directly
perpendicular to the direction of the rays.

The Geodimeter and Mekometer belong to this category. They use the same principle
of comparing modulated light returned from the reflectors with the transmitted beam.
Typical range 5km.

iv. Instruments utilizing infra-red radiations.

Although operating with approximately double the wavelength of the visible light type
instruments, the accuracy of infra-red equipment remains practically as good. However,
due to the lower power transmission of the infra-red diode, the range is limited to
approximately 2km.
Instruments of this type are usually compact, small, light weight and equipment with
direct numerical read out.

Instrument Manufacturer Range Accuracy

Hp 3800 HP, USA 3km 0.5cm

DM 500 Switzerland 0.5km 0.6cm
DM 2.5km 0.4cm
MA 100 Techometer, S.Africa 2km 1.5mm
Distomat D1-10 Wild Switzerland 1km 1cm
EOK 2000 Zeiss Jena, E. Germany 2km 1cm
SM 11 Zeiss Oberkoche, W. Germany 1km 1cm
v. Laser EDM Instrument
 There are several instruments which use laser light instead of ordinary light. This
means basically an extension of the range while maintaining all basic characteristics of
the instrument (L Geodimeter).

vi. Combined EDM Angular Measuring Instrument.

While many EDM instruments can be directly interchanged with Theodolites (forced
centering) or attached to theodolites (e.g. DI 10, Kern DM500), a few combined
instruments are on market. These instruments are sometimes called electronic
tacheometers or total station e.g. HP Station, Zeiss (Oberkoche) Regella and the
Geodimeter 700.

EDM PRINCIPLES.

Principles of distance measurements.

Although a wide variety of EDM instruments are currently available, most derive
distances by measuring either time delay between pulses of EM energy, normally
referred to as the pulse echo technique or by comparing transmitted and received
signals and deriving the phase difference between them. The former technique is
normally restricted to laser-based instruments.
The fundamental stages involved in the determination of a distance electronically
can be represented by a flow diagram. The basic EM signals generated by the
instrument is the Carrier Signal, and may be of visible , infrared, or microwave
wavelengths. The first two cases may be termed Passive EDM instrument in which the
EM signal is reflected from an unmanned reflector, whereas the latter forms the basis of
Active EDM instruments in which the returned signal is generated by a second manned
instrument at the terminal point.

Carrier Modulation.
Since the λ of the signal is very short (0.5mm – 30mm), in order to ensure that the
phase difference can be measured precisely, a measurement modulation signal is
superimposed on to the carrier signal, a process termed Carrier Modulation.
The transmitted signal is then either reflected or retransmitted along an identical
path back to the instrument. With electro-optical devices reflection of the signal is
achieved by means of precisely manufactured retro reflective prism, also referred to as
Corner Cube Reflectors. Generally speaking, the alignment of a reflector is not a
critical problem, although precise alignment of the measuring instrument is essential.
In the case of microwave instruments, the signal is received by a second remote
instrument and retransmitted back to the master instrument.

➢ The weak reflected signal is then amplified and compared in phase with the
original transmitted beam.
 ➢ The difference in phase between the two signals is derived either using a digital
pulse approach (electro-optical instruments) or a null approach (microwave
instrument) in which the phase difference between the signal is brought to zero
by introducing an additional phase shift, the required phase shift being
indicative of the original phase difference.

+
a b

The final distance is derived from the electronic and data processing facility in built in the
measuring system.
CARRIER SIGNAL CARRIER TRANSMITTER
Electro optical(EO) MODULATION Optical lens (EO)
λ=0.56-0.91 μ m Amplitude (EO) Dipole and
Microwave (μ) Frequency (M) Parabolic
λ=30mm reflection (M)

POINT B
Retro-reflector (EO)
PHASE Remoter transmitter
SLOPE MEASURENT MODULATION SIGNAL (M)
DIST. Digital (EO)

AMPLIFIER RECEIVER
Silicon photodiode Optical lens
CORRECTIONS Electronic circuit(M) (EO)
Refractive index Dipole(M)
Calibration
Slope
MSL
Scale factor

CORRECTED DIST DATA COLLECTOR

a – reference signal opens counting gate.

b – reflected signal closes gate.
∆t – timing pulses counted.

➢ The measurement of the phase difference between the signals is representative of

the fractional part of the total distance less than the integral value of the modulation
wavelength.

The distance AB is given by

D = Mλ + ∆λ…………….(1)

M-integral, ∆λ-fraction of wavelength

 The signal is normally reflected back to the instrument resulting in
2D = Nλ + ∆λ……………(2)

This is equivalent to the N revolutions of the vector OA plus the fractional excess equal
to φ. Angle φ is a measurement of the phase difference between outgoing and the
reflected signal and can be used to derive a value for
∆λ = φ λ/2π -----------------(3)
∆ϕ= ϕt – ϕr

D=Nλ/2 + ϕ/2π. ϕ/2 (4)

This is an idealized model, a more precise description of the situation

D= N.C/2fna + (ϕ/2π X C/2fna) + K3 + K2 + K1

Where,
C = velocity of the EM signal in a vacuum
na = refractive index of the atmosphere
f = modulation frequency
k2 = the zero error of the instrument
k3 = the cyclic error of the instrument
k1 = scale error

The integer N cannot be derived directly by a single measurement and is obtained

by introducing successive changes to λ which leads to a series of the equation of the
general form

D = Ni . λi/2 + ϕi/2π . λi/2

1st term is unknown, 2nd term is known, 3rd term is measured

ERRORS INFLUENCING EDM MEASUREMENTS.

Many different sources of systematic errors can affect EDM measurements

including those caused by instrument operation., the atmosphere and the instrument
maladjustment.
 1. Instrument operator errors.
➢ Precise centering of the instrument.
➢ Pointing to target
➢ Setting automatic correction values.

2. Atmospheric errors.
The effect of the prevailing atmospheric conditions (temperature, pressure, relative
humidity) on EDM measurements is analogious to the influence of temperature when
measuring distance with a steel tape. Variations in the temperature cause expansion or
contraction of the tape from it’s standardized length which if not corrected will lead to a
systematic error in the measured distance. Similarly, variation in the atmospheric condition
from those which are assumed by the instrument will cause an increase or reduction in the
measuring wavelength and lead to a systematic error if a correction is not applied.
Normally, the effect of the atmosphere is defined by the change in refractive index
(n) where
n = C/V

C = velocity of the electromagnetic waves in vacuum.

V = velocity in the medium (atmosphere)
n is assumed to be 1.000320
For the visible and infrared parts of the EM spectrum n is calculated using the Barell and
Sears’ formula.

(nt-1) = (ns-1) 273 . P - 15.002 E × 10-6

273 + t 760 273 + t
Where,
(ns -1) × 10-6 = 287.604 + 1.6288 0.0136
+
λ02 λ04

Where,
nt - prevailing refractive index
ns - standard refractive index of the air @0oC and 706mmHg pressure with a 0.03% CO2
content
t - temperature during observations
P – pressure during observations in mmHg
E – water vapour pressure in mmHg
λ0 – wavelength of signal I vacuum

t, P, E are normally observed at the terminal points of the line being measured
3. Instrument Errors
EDM device like other surveying instrument requires careful use and regular
calibration if they are to provide reliable and accurate results. Most instruments are subject
to relatively small but vitally important systematic errors caused by instruments
maladjustment. Modern EDM’s do not display results when maladjustment is severe. Scale
error (k1) occurs if the modulation frequency EDM instrument does not correspond exactly
with the design frequency value for the instrument. The error is proportional to the distance
measured and may often be expressed in parts per million (ppm) of the distance. Ideally it
should negligible however it may be as high as 20 to 30 ppm (20- 30mm/km) in extreme
cases.
Zero error (k2), also referred to as additional constant error, index error or reflector/
prism offset; represents the difference between the EDM measured distance between two
points (corrected for scale, cyclic and atmospheric error) and the distance between the
known points. It is caused if the internal measurement centre of the instrument (and
reflector) does not coincide with the physical centre of the instrument (and reflector) which
is plumbed over the measuring mark.
It also accounts for the variation in the path length through the retroflector caused
by refractive index variations. at the glass- air interface. It is a systematic error of constant
magnitude and therefore not proportional to the distance measured.
The value of zero error obtained during a calibration refers to a particular EDM
instrument and retroflector. If either the instrument or reflector is changed then a new zero
constant will exist.
Cyclic error (k3) also termed periodic error, revolver error or non-linearity error, is
analogous to circle graduation errors in a theodolite in the sense that they are periodic and
can have a systematic effect. They are generally the smaller of the three sources of
instrument errors and are caused by internal electronic contamination between the
transmitter and receiver circuitry, the effect being cyclic over the modulation wavelength.
The effect of this source of error is reduced by the instrument manufacturers by
electrical isolation and shielding of components within the instrument, although the effect
becomes more apparent with a weak return signal. It is eliminated by taking means of back
and forward distance readings in microwave instruments.

10.1 m(say)

● ● ● ● ● ●
A B1 B2………………………………………………... BN
100M
Accuracy of EDM.
Typically given as ± e mm + p mm/km. Overall standard error of measurement

D = distance measured.
e is constant
p is distance dependent.

e is important for short distances, while p is important for long distances.

Methods of Instrument calibration.

There two distance calibration approaches namely;

➢ Quick and calibration test which could be carried out on site.
➢ More sophisticated approach to (calibration on baseline) for more accurate
determination of the state of health of an EDM instrument.

Three distinct methods to instrument calibration can be identified.

• Laboratory methods
• Field methods using known baseline lengths
• Field methods using unknown baseline lengths

❖ Laboratory methods – Laboratory test can be carried out to assess both scale
and cyclic error components.

Advantages over field tests.

o They are quick and convenient to apply in practice.
o They enable the surveyor to precisely control the measurement conditions

Disadvantages
o They are limited to short range calibration measurements.

Scale error – determination in the laboratory by comparing the observed modulation

frequency of the EDM signal with a time reference frequency.

EDM Frequency counter Off – air frequency

Standard

Laboratory determination of scale error using standard frequency.

Cyclic error – can be determined in the laboratory by recording distance measurement to a
series of reflector positions, which extend over a distance equivalent to the half-
wavelength of the instrument. The simplest and cheapest arrangement involves the use of a
graduated metal bar.
The calibrated bar is constructed so that the prism can be moved in increment of say
5m from the EDM instrument to minimize pointing errors. A superior EDM e.g. laser may
also be used in place of calibrated bar.
If the effects of scale error are presumed to be negligible over such short distances
then each measurement to the bar Lm will be subject to only zero and cyclic errors, so that

LM = LT + K2 + K3……………..(1)

Where,
LM – Measured distance
LT - True distance (unknown)
K2 - zero error
K3 - cyclic error
For any other point i,

LiM = LiT + K2 + K3i………………..(2).

Subtracting equation (1) from (2) to eliminate the constant zero error.

∆LM = ∆LT + (K3 – K3i)

OR (∆LM - ∆LT) = K3 – K3i)

A graph may then be drawn to illustrate the difference between the distances (∆L) at
various values of L. The resulting graph will be as follows;

∆L 6

0
0m 5m 10m L

6
Unknown Versus Known baselines.
• Unknown more popular because stability of markers or pillars of baselines only
required for the duration of survey observation only.
• Known baselines require periodic monitoring of the interpillar distances
• Disadvantage of the unknown is inability to determine scale error in the field.

❖ Field method using unknown Baseline Lengths. – Measurements on a

calibration baseline can be used to determine all 3 error components (scale, zero
and cyclic). Either known baseline may be used. Two techniques which are based
on the use of an unknown length of baseline are:
• The three point method
• Schwendener method
The simplest procedure for a quick and appropriate determination of the zero error involves
the use of three baselines.
l13

∆ l12 ∆ ∆
1 2 l23 3

Zero error determination using 3-pt baseline method.

If the observed distances (1) are each in error by the same zero error (K2) and if the actual
distance 1 to 2 and 2 to 3 are d1 and d2, then ignoring all errors.
l12 = d1 + k2
l23 = d2 + k2
l13 = d1 + d2 + k2

Therefore
k2 = l12 + l23 – l13

If the baseline consists of more than two sections and each section 1 is measured, then if
the total length is L, a more precise value of K2 can be derived using

K2 = L - ∑ li

n-1
Where,
L – Total length of the baseline
1i – Length of each baseline section
n – Number of baseline sections
Schwendener method – Improves the determination of K2 by making use of a 7- point
baseline (six sections) which, taking all combination of distances, results in a total of 21
measurements. Since 7 are required for a unique solution, the solution is over determined,
with 14 redundant measurements.
01 12 23 34 45 56
02 13 24 35 46
03 14 25 36
04 15 26
05 16
06

Schwendener’s basic observation equation is of the form

( ∑li)1 + K2 = Mi + Vi
Where,
Li – unknown interpillar distances
K2 – zero error
Mi – observed values (21 distance)
Vi – residuals (21 values)

❖ Field Methods using known baseline lengths – The use of known values for the
length of baseline and baseline advantageous since it enables the values of the
scale error to be derived in addition to zero and cyclic errors. A variation on the
Schwendener’s methods which includes measured distances as given below;

( ∑li)i + K2 + K1 ( ∑li)i = Mi – (K3)I + Vi

Where,
1i – are the known or observed distance
K2 – zero error
K1 – scale error
K3 – known cyclic error (from laboratory measurements)
Vi - residuals
Mi – observed distances.

EDM Instrument and Applications.

➢ Geodetic control Networks.

EDM over long range (>10km) has been used since the original development of
EDM instruments to measure baselines and to strengthen the existing geodetic control
networks e.g. Tellurometer.

➢ Large scale surveys.

 Large scale (,1:1000 scale) route and site surveys form the central activities of most
surveying organizations. Other traditional techniques such as chain traversing and stadia
tacheometry are still used. However, they have been superseded by EDM such as
geodometer 122 (traversing) and polar detailing using tachometers and data recorders. The
use of such EDM instruments not only decreases the observation time but also enables
higher orders of precision to be achieved in comparison to more conventional methods.

➢ Setting out works.

The widespread adoption of EDM instruments in the construction industry has lead
to significant changes in the techniques of setting out civil works. Of particular importance
has been the move towards polar methods setting out using coordinates rather than the
more tradition technique based on grids and offsets.

➢ Tracking
Almost all modern electronic-Optical EDM instruments are capable of measuring
to a moving target. Generally the level of accuracy is of a lower order of accuracy than for
static mode measurements- typically 10-20mm.

➢ Integral speech communication.

One way communication between the instrument and target can be achieved by
modulating the infrared carrier signal with speech information.

➢ Visible guidance system.

A further facility offered by Geotronics is the track light system. In this case a
flashing light source is built into the instrument. The correct alignment for the target occurs
when the prism is located exactly along the direction being defined by the track light.
When the target is in this position, the survey assistant will view a white pulsing light. Any
deviation of line will be indicated by a change in the colour of the pulsing signal, which
turns red with movement to the right and green if the deviation is to the left of the correct
position.

➢ Deformation monitoring
The analysis of surveying measurements/observation of different epochs to determine
whether any structural deformation has occurred is becoming increasingly popular. Several
different techniques can be used for such operations including those based primarily on
angle, distance measurement or close range photogrammetry. The determination of precise
distance is of fundamental importance in all cases and is generally performed using
specialized instruments such as Kem Mekometer and Comrad Geomensor. Laser
interferometers may be used for very short range distance measurements. Dams, bridges,
waterways, manufacturing plants, landslides, tectonic movements, geothermal stations, can
be subject to successful monitoring using EDM measurements.

ELECTRONIC TACHEOMETER.
This refers to an instrument that is able to:-
• Electronically measure both angles and distances.
• Perform some limited computational tasks using an integral microprocessor.
• Reduce slope distance to horizontal
• Calculating chords from bearings and distances
• It may also store data in an internal memory unit or external solid-state data
recorder also known as total stations.

There are basically two designs.

• Integral
• Modular

➢ Integrated design.
The pri9mary feature is that the electronic theodolite and EDM instrument form a
single integrated unit.

Advantages.
- Need to transport only a single unit.
- Dispense with auxiliary cables that are often required to link separate units.
- Collimation of the telescope line of sight and the EDM signal properly aligned.
Examples include Shokisha Set2&Geodimeter 440.

➢ Modular design.
Electronic theodolite and EDM instrument are separate unit which can be independently
operated.

Advantages.
- More flexible
- Cost effective solution.

DIAGRAM…………………………..

DISTANCE MEASUREMENTS: EDM AND TOTAL STATIONS

 TABLE 5.2
Some Examples of Total Stations

Instrument Nikon Sokkia Zeiss

DTM-A5LG SET 3C Elta 5

Angle measurement
H accuracy ± 2‫״‬ ±3 ±5
V accuracy ± 2‫״‬ ±3‫״‬ ± 5‫״‬
Distance measurement
to one prism 2.3 km 2.2km 1.0km
to three prisms 3.1 km 2.9 km 1.5 km
accuracy ± (2 mm + 2 ppm) ± (3 mm+ 3ppm) ± (5 mm + 3
ppm)
Measurement time 3.0 seconds 3.2 seconds 3 – 4 seconds
Data displayed
H and V angles yes yes yes
SD, HD and VD yes yes yes
X, Y and Z cords yes yes yes
Setting out data yes yes yes
Data processing data recorder data recorder data recorder
field computer field computer field computer
memory card
Compensator single-axis dual-axis single-axis
Battery NiCad NiCad NiCad
7.2 V 6V 4.8 V

Distance Measurement

At present most total stations use a GaAs infra-red carrier source and phase comparison
techniques in order to measure distances. However, compared to theodolite-mounted
systems, nearly all total stations use coaxial optics in which the EDM transmitter are
combined with theodolite telescope. This makes the instrument much more easier to use on
site. Normally, a total station will measure a slope distance and the microprocessor uses the
vertical angle recorded by the theodolite along the line of sight (line of distance
measurement) to calculate the horizontal distance. In addition, the height difference
between the trunnion axis and prism centre is also calculated and displayed. All
instruments use some form of signal attenuation to protect the receiver.

Three modes are usually available for distance measurement.

Standard ( or coarse) mode which has a resolution of 1mm and a measurement time
of 1 – 2 seconds.
Precise (or fine) mode which again has a resolution of 1 mm and a measurement
time of 3 – 4 seconds. This is more accurate than the standard mode since the instrument
repeats the measurement and refines the arithmetic mean value.
 Tracking (or fast) mode in which the distance measurement is repeated
automatically at intervals of less than one second. Normally, this mode has a resolution of
10 mm and is used extensively when setting out since readings are updated very quickly
and vary in response to movements of the prism which is usually pole-mounted. Setting out
using total stations is discussed in chapter 14.
The range of a total station is typically 1- 3 km to a single prism assuming visibility
is good and up to a range of 500 m, which covers 90 per cent of the distances measured on
site, the precision of a typical total station is about 5mm. Most instruments allows for the
input of temperature and pressure which enables the distances readings to be automatically
corrected for atmospheric effects (see section 5.19). In addition, any value of prism
constant can be entered into the instrument to suit whatever prism is being used.

Control Panel

A total station is activated through its control panel which consists of a keyboard and
multiple line liquid crystal display (often abbreviated to LCD). The LCD of a total station
is moisture proof, it can be illuminated and some incorporate contrast controls to
accommodate different viewing angles. A number of instruments have two control panels
(one on each face of the theodolite) which makes them easier to use. The keyboard enables
the user to select and implement different measurements modes, enables instruments
parameters to be changed and allows special software functions to be used. Some
keyboards incorporate multi-function keys to carry out specific tasks whereas others use
keys to activate and display menu systems which enables the total station to be used as a
computer might be.
In addition to controlling the total station, the keyboard is used to code data
generated by the instrument. Angles and distances are usually recorded electronically by a
total station in digital form as raw data (slope distance, vertical angle and horizontal angle).
If a code is entered from the keyboard to define the feature being observed, the data can be
processed much more quickly by downloading it into appropriate software. On numeric
keyboards, codes are represented by numbers only whereas on alpha-numeric keyboards
codes can be represented by numbers and/or letters which gives greater versatility and
scope. The alphanumeric control panel of the Topcon GTS-6 is shown in figure 5.13.
Feature codes and their application to larger-scale surveys are discussed in section 9.11.

Figure 5.13…………………………………….

Power Supply

Rechargeable nickel-cadmium (NiCad) batteries are now standard for surveying

instruments and these are connected directly to the total station without using cables. For
angle and distance measurements, between two and ten hours’ use can be obtained from a
battery depending on the instrument. Most total stations are capable of giving a battery
power indication and some have an auto power save feature which switches the instrument
off or into some standby mode after it has not been used for a specified time.
It is good practice, no matter what assurances a manufacturer may give about the
life of a battery, to have a fully charged spare with the instrument at all times.

Useful Accessories

Geotronics and Nikon manufacture devices known as the Track light and Lumi-
Guide respectively. Both of these are similar and the Track light is a visible light which
enables a pole-mounted prism to be set directly on the line of sight without the need for
hand signals from the total station. The device consists of a flashing three colour light: if
the prism is to the left of centre of the line of sight, a green light is seen; and if the prism is
to the right, a red light is seen – as shown in figure 5.14. When the prism is on line, a
flashing white light is seen, the frequency of which doubles when it strikes the prism giving
confirmation that the prism is in the correct position.
The Geotronics Unicom is a communication system which allows speech to be
transmitted from a Geodimeter instrument to a prism. This consists of a small microphone
on the control panel which is activated by pressing a key and a receiver with small
loudspeaker mounted on the prism pole (see figure 5.15). The usual method of
communicating on site, however, is by use of short range VHF hand-held radios.

Figure 5.14……………………………

5.11 Onboard Software

As well as controlling the angle and distance functions of a total station, the microprocessor
is also programmed to perform coordinates and other calculations. Some of these are
described below although not all are available on every instrument.

Slope Corrections and Reduced Levels

From raw data (slope distance, vertical angle horizontal distance), a total station will
calculate and display horizontal distance and vertical distance and the height of the prism
are entered, the reduced level of the prism station can also be calculated and displayed (see
section 5.24). These are the most basic functions of a total station.

Figure 5.15……………………….
Horizontal Circle Orientation

The horizontal circle of a total station can be set o read a known bearing by entering the
easting (E) and northing (N) coordinates of the station occupied followed by the E and N of
a reference station is then sighted and the orientation program is activated to calculate the
bearing from the station occupied to the reference station and to set the horizontal circle to
display this bearing (figure 5.16). The instrument would now be ready for further
coordinate measurements or for setting out.

Coordinate Measurement

Having oriented the horizontal circle of a total station, the coordinates ……… can be
determined fairly easily. A new point is sighted,

N
N
reference
NR R station

A
S
NS instrument
station

E
ES ER

Figure 5.16 Horizontal circle orientation :Es Ns and ER NR are entered into total station which then calculate A and
orientates to this

and distance and circle readings taken: when using the coordinate measurement program
the instrument will now display the coordinates of the new point (figure 5.17). This can be
extended to three dimensions if the reduced level of the instrument station and appropriate
instrument and prism heights are also entered into the total station which will then display
the reduced level of the new point.

N
N
NR R new point

A
S
EP=ES+D sin B
NS instrument Np=Ns+D cos B
station
Figure 5.17 coordinate measurement: total station determines D and B and calculates Ep Np from instrument station

Traverse Measurements

Traversing consists of a series of distance and angle measurements taken between

successive points that enable the coordinates of those points to be calculated. Shown in
figure 5.18, this type of coordinate measurement is performed by a total station as a series
of horizontal circle orientations and coordinate measurements taken at each traverse
station. Traversing is discussed in much more detail in chapter 7 and the reader is strongly
advised to consult this before attempting to use a total station in control survey.
A4
reference
poit(E.N
known) A2

2
D2 etc
A1 D1 D3
A3

Resection (or free stationing)

At level two control points (1 and 2 in figure 5.19) are required for free stationing ,the
coordinates of which have to be entered into the total station .These points are sighted in
turn and distances and circle readings measured to each. Using this data, the instrument
calculates the coordinates of the instrument station. This can be a useful facility when
setting out as temporally control point can be established in any desired location to suit
need as. As with other coordinates functions, if reduced levels and prism heights at the
control points are also fed into the total station as well as the instrument height, the reduced
level of the instrument station can also be computed and displayed.
 Control point 1
(E1N1) Control point 2
(E2N2)

D1
D2

Instrument station
(E2N2)

Missing Line Measurements (MLM)

This software option allows a total station, from a single instrument position, to determine
the horizontal distance and height difference between a start point and series of
subsequently selected points. In figure 5.20, points 1 and 2 are sighted and the distance and
circle readings to them recorded from the instrument station .The MLM program then
computes the horizontal distance D12 and height difference ∆h12 between these points. If the
distance
 1
D12
∆h12 2

1
D12
D13
∆h12 2
∆h13

D13

∆h13
3
Instrument station

Instrument station

and circle reading to a third point are included in the sequence, the total station can display
D13 and ∆h13 (radial MLM) or it can display D23 and ∆h23 (continuous MLM). Any number
of points can be added to the sequence.

Remote Elevation Measuring (REM)

This function is used to determine heights at inaccessible points where it is not possible to
locate a prism. Since measurements are taken along an extended plumb line through the
prism, the prism must be positioned vertically above or below the point(s) to be surveyed.
The prism height p is entered into the instrument and the horizontal distance D to the prism
determined (figure 5.21). In REM mode, the total station will now display the height from
ground at the prism to any point along the vertical through the prism. In figure 5.21, the top
of a structure S could be sighted directly above the prism and its height h recorded. REM
can also be used to set profile boards at their correct heights.
 s

Point at S and display h

h

p
Point at prism and measure
D

Vertical through prism

Figure 5.21

Setting Out Functions

Total stations can be used for setting out with given horizontal angle and distance values or
with given coordinates.

When the horizontal angle and distance to be set out are known, these are entered into
the instrument which has already had its horizontal circle orientated to a reference station.
As soon as the appropriate key(s) are pressed to activate the setting out mode on the total
station, it displays the difference (dHA) between the entered and measured horizontal angle
values. In order to set the required direction for setting out, the telescope is rotated until a
difference of zero is displayed (that is, dHA = 0). Following this, a pole-mounted prism is
located on the line of sight as near to the required distance as possible: devices such as the
Tracklight or Unicom (see section 5.10) are useful for lining in the prism but VHF radios
with hand signals between instrument and prism are often used. Once aligned, the prism is
sighted and the distance to it measured by the total station. The difference between the
measured and entered distances is displayed and by moving the prism this difference is
reduced to zero to locate the point.
When the coordinates of the points to be set out are known, those of the instrument
and reference stations should also be known. Prior to measurement, the station coordinates
are entered into the instrument as well as those of a reference station and the horizontal
circle is orientated to the coordinate grid such that it displays bearing directly. Next, the
coordinate values of the point to be set out are entered into the total station. When the
setting out mode is selected the instrument displays the difference between calculated and
measured bearings. There is no need for the observer to calculate any bearings as the total
system does this automatically. As with the horizontal angle and distance mode, the
telescope is rotated until this difference is zero such that it is pointing in the required
direction for setting out. With pole-mounted prism located on this line of sight, the
horizontal distance to it is measured and the difference between this and the value
calculated by the total station is displayed. This is reduced to zero by moving the prism.
For further details of setting out by coordinates with total stations see chapters 10,
11 and 14.

5.12 Total Stations: What to do and what not to do

The previous section discussed the onboard software installed in total stations and a
number of different measurement functions were described. As far as site surveying is
concerned, some of these have created a situation where the emphasis on surveying is
apparently turning away from hand calculations and associated checking procedures to
good site practice and field checking procedures.
A note of caution is expressed here. Even though a total station can perform many
of the calculations often done manually on site. This does not mean the surveyor or
engineer should lose this ability. For this reason, surveying for engineers deals with and
strongly emphasizes coordinate calculations throughout. A knowledge of these may avoid
situations where too great a reliance is placed on the digital readout obtainable from a total
station and may avoid obvious mistakes when generati0ng data using setting-out functions.
In other words, the ability to process and manipulate coordinates by hand gives a surveyor
or engineer the ‘feel’ for a correct orientation and distance, especially when setting out.
It is accepted that the total station is a very sophisticated instrument but this in itself
can create problems. As an example, a traverse (see chapter 7) is a recognized surveying
procedure for obtaining the coordinates of control points. Using on board software, a total
station can produce a set of coordinates much more quickly than by traversing since
observations can be taken from a single instrument position. Although the methods used by
total stations for obtaining coordinates have their applications, they are very dangerous if
used in the long circumstances, especially when fixing the position of the control points.
There are many cases where time taken to arrive at the end result when using a total station
is reduced by such a large factor compared with an established field operations that the
temptations are often too great and mistakes occur. While every opportunity should be
made to make full use of a total station, any field procedure involving a total station that
does not include an independent check on field work must be treated as incomplete.
Through surveying for engineers, recommendations for good practice are given for all
procedures covered: there are still good reasons why these must be applied even when the
most up-to-date instruments are being used on site or elsewhere when surveying.

5.13 Specialised Total Stations

Geotronics AB of Sweden currently manufacture a number of total stations with unique

features.
The Geodimeter 464 (figure 5.22) has all of the features described in sections 5.10
and 5.11 but is in addition, a servo-driven total station. In use, all the operator needs to do
to align onto the prism is to point the instrument roughly at the prism and then press the
key to initiate a measurement. The servo motors the take over pointing the instrument at the
prism and readings are recorded automatically. This has some advantages over a manually
pointed system since a servo-driven instrument can ai9m and point much more quickly and
with a better precision. For setting out with the geodimeter 464, a point number is entered
and the instrument instantly computes setting-out data and then automatically positions
itself on the calculated bearing. If an elevation is stored, the instrument will also position
itself in the third dimension.
The Geodimeter system 500 is a range of total stations which allows users.

8
Curves

In the geometric design of motorways, railways, pipelines, etc., the design and setting out
of curves is an important aspect of the engineer’s work.
The initial design is usually based on a series of straight sections which positions
are defined largely by the topography of the area. The intersections of pairs of straights are
then connected by horizontal curves (see Section 8.2). In the vertical design, intersecting
gradients are connected by curves in the vertical plane.
Curves can be listed under three main headings, as follows:

(1) Circular curves of constant radius.

(2) Transition curves of varying radius (spirals)
(3) Vertical curves of parabolic form
8.1 CIRCULAR CURVES

Two straights, D1T1 and D2T2 in Figure 8.1, are connected by a circular curve of radius R:

(1) The straights when projected forward, meet at l: the intersection point.
(2) The angle ∆ and l is called the angle of intersection or the deflection angle, and
equals the angle T10T2 subtended at the centre of curve 0.
(3) The angle φ at l is called the apex angle , but is little used in curve computations.
(4) The curve commences from T1 and ends at T2; these points are called the tangent
points.
(5) Distances T1l and T2l are the tangent lengths and are equal to R tan ∆/2.
(6) The length of the curve T1AT2 is obtained from:

Curve length = R∆ where ∆ is expressed in radians, or

Curve length = ∆◦100/D◦ where degree of curve (D) is used (see Section 8.1.1)
(7) Distance T1T2 is called the main chord (C), and from Figure 8.1.

sin ∆/2 = T1B/T1O = ½ chord (C)/R C = 2Rsin∆/2

(8) lA is called the apex distance and equals

lO – R = R sec ∆/2 – R = R(sec ∆/2 – 1)

(9) AB is the rise and equals R – OB = R – R cos ∆/2

AB = R(l – cos ∆/2)

These equations should be deducted using a curve diagram (Figure 8.1).

1 ∆

A
∆ ∆
l2 l2
T1 B T2
∆
R l2 ∆
R

D2
D1
Figure 8.1

8.1.1 Curve designation

Curves are designated either by their radius (R) or their degree of curvature (D ◦). The
degree of curvature is defined as the angle subtended at the centre of a circle by an arc of
100m (Figure 8.2).
Thus R = 100m/D rad = 100 × 180◦/D◦ × Л

R = 5729.578/D◦ × m

Thus a 10◦ curve has a radius of 572.95978 m

8.1.2 Through chainage

Through chainage is the horizontal distance from the start of a scheme for route
construction.
Consider Figure 8.3. If the distance from the start of the route (Chn 0.00m) to the
tangent point T1 is 2115.50 m, then it is said that the chainage of T1 is 2115.50m, written as
(Chn 2115.50m)

If the route centre-line is being staked out at 20m chord intervals, then the peg
immediately prior to T1 must have a chainage of 2100m (an integer number of 20m
intervals). The next peg on the centre-line must therefore have a chainage of 2120m. It
follows that the length of the first sub-chord on the curve from T1 must be 92120 –
2115.50) = 4.50m.
100m

R R

Figure 8.2

●
l2

Origin of
scheme ● T●4
● T3

●
● T2
T1 l3
●
l1

Figure 8.3
 Similarly, if the chord interval had been 30m, the peg chainage prior to T1 must be
2100m and the next peg (on the curve) 2130m, thus the first sub-chord will be 2130 –
2115.50) = 14.50m.
A further point to note in regard to chainage is that if the chainage at l1 is known,
then the chainage at T1 = Chn l1 – distance l1 T1, the tangent length. However the chainage
at T2 = Chn T1 + curve length, as chainage is measured along the route under construction.

8.2 SETTING OUT CURVES

This is the process of establishing the centre-line of the curve on the ground by means of
pegs at 10m to 30m intervals. In order to do this the tangent and intersection points must be
fixed in the ground in their correct positions.
Consider Figure 8.3. The straights Ol1, l1l2, l2l3, etc., will have been designed on the
plan in the first instance. Using railway curves, appropriate curves will now be designed to
connect the straights. The tangent points of these curves will then be fixed, making sure
that the tangent lengths are equal, i.e. T1l1 = T2l1 and T3l2 = T4l2. The coordinates of the
origin, point 0, and all the intersection points only will now be carefully scaled from the
plan. Using these co-ordinates the bearing of the straights are computed and, using the
tangents lengths on these bearings, the co-ordinates of the tangent points are also
computed. The difference of the bearings of the straights provides the deflection angles (∆)
of the curves, which combined with the tangent length, enables the computation of the
curve radius, through chainage and all setting-out data. Now the tangent and intersection
points are set out from existing survey stations and the curves ranged between them using
the methods detailed below.

8.2.1 Setting out with theodolite and tape

The following method of setting out curves is the most popular and it is called Rankine’s
deflection or tangential angle method, the latter term being more definitive.
In Figure 8.4 the curve is established by a series of chords TX, XY, etc. Thus, peg 1
at X is fixed by sighting to l with the theodolite reading zero, turning off the angle δ1 and
measuring out the chord length T1X along this line. Setting the instrument to read the
second deflection angle gives the direction T1Y, and peg 2 is fixed by measuring the chord
length XY from X until it intersects at Y. The procedure is now continued the angles being
set out from T1l and the chords measured from the previous station.
It is thus necessary to be able to calculate the setting-out angles σ as follows:
Assume 0A bisects the chord T1X at right angles; then
 I

X Peg 1
●

δ R Y Peg 2
T1 ●

Figure 8.4

AT10 = 90◦ - δ1, but lT10 = 90◦

lT1A = δ1
By radians, the length T1X = R2δ1

δ1 rad = arc T1X/2R = chord T1X/2R

δ°1 = chord T1X × 180°/2R.Л = 28.6479 chord/R = 28.6479 C/R (8.2a)

or δ° = D° × chord/200 where degree of curve is used (8.2b)

(Using equation (8.2a) the angle is obtained in degree and decimals of a degree; a single
key operation converts it to degrees, minutes, seconds.)
An example will now be worked to illustrate these principles.
The centre-line of two straights is projected forward to meet at l, the deflection
angle being 30◦. If the straights are to be connected by a circular curve of radius 200m,
tabulate all the setting-out data, assuming 20m chords on a through chainage basis, the
chainage of l being 2259.59m.

Tangent length = R tan ∆/2 = 200 tan 15◦ = 53.59m

Chainage of T1 = 2259.59 – 53.59 = 2206m
1st sub-chord 14m

Length of circular are = R∆ = 200(30◦ .Л/180)m = 104.72m

From which the number of chords may now be deducted

i.e. 1st sub-chord = 14m

2nd, 3rd, 4th, 5th chords = 20m each.
 Final sub-chord = 10.72m.
Total = 104.72m (check)
Chainage of T2 = 2206m + 104.72m + 2310.72m

Deflection angles:

For 1st sub-chord = 28.6479 . 14/200 = 2◦ 00’19”

Standard chord = 28.6479 . 20/200 = 2◦51’53”

Final sub-chord = 28.6479 . 10.72/200 = 1◦32’08”

Check: The sum of the deflection angles = ∆/2 = 14◦59’59” = 15◦

Table 8.1

Chord Chord Chainage Deflection Setting-out Remarks

number length angle angle
(m) (m) ◦ ‘ “ ◦ ‘ “

1 14 2220.00 2 00 19 2 00 19 peg 1
2 20 2240.00 2 51 53 4 52 12 peg 2
3 20 2260.00 2 51 53 7 44 05 peg 3
4 20 2280.00 2 51 53 10 35 58 peg 4
5 20 2300.00 2 51 53 13 27 51 peg 5
6 10.72 2310.72 1 32 08 14 59 59 peg 6

The error of 1” is, in this case, due to the rounding-off of the nearest second and is
negligible.

8.2.2 Setting out with two theodolites

Where chord taping is impossible, the curve may be set out using two theodolites at T1 and
T2 respectively, the intersection of the lines of sight giving the position of the curve pegs.
The method is explained by reference to Figure 8.5. Set out the deflection angles
from T1l in the usual way. From T2, set out the same angles from the main chord T2T1. The
intersection of the corresponding angle gives the peg position.
If T1 cannot be seen from T2, sight to land turn off the corresponding angles ∆/2 –
σ1, ∆/2 – σ2, etc.

8.2.3 Setting-out using EDM

When setting out by EDM the total distance from T1 to the peg is set out, i.e. distances
T1A, T1B, and T1C etc. in Figure 8.6. However, the chord and sub-chord distances are
required in the usual way plus the setting-out angles for those chords. Thus all the data and
setting-out computation as shown in Table 8.1 must first be carried out prior to computing
the distances to the pegs direct from T1. These distances are computed using equation 7 in
8.1 i.e.
 T1A = 2 R sin σ1 = 2 R sin 2◦00’19” = 14.00m
T1B = 2 R sin σ2 = 2 R sin 4◦52’12” = 33.96m
T1C = 2 R sin σ1 = 2 R sin 7◦44’05” = 53.83m
T1T2 = 2 R sin (∆/2) = 2 R sin 15 ◦00’00” = 103.53m.

l
∆

Peg1 Peg2
● ●

T1 T2
L

B
A ●
●
C
●

T1 ●
Main cord T2

Figure 8.6

Setting – out EDM

In this way the curve is set out by measuring the distance direct from T1 and turning off the
necessary direction in the manner already described
8.2.4 Setting-out using co-ordinates

In this procedure the co-ordinates along the centre-line of the curve are computed relative
to the existing control points. Consider Figure 8.7:

(1) From the design process, the co-ordinate of the tangent and intersection points are
obtained.
(2) The chord intervals are decided in the usual way and the setting-out angles σ1,
σ2…σn computed in the usual way (Section 8.2.1).
(3) From the known co-ordinates of T1 and l, the bearing T1l is computed.

B Z
● C ○
A
● ●
X
○

● ○ ●D
T1 Y

T2
Figure 8.7 Setting –out coordinates

(4) Using the setting-out angles, the bearings of the rays T1A, T1B, T1C, etc. are
computed relative to T1l. The distances are obtained as in Section 8.2.3.
(5) Using the bearings and distances in (4) the co-ordinates of the curve points A, B, C,
etc. are obtained.
(6) These points can now be set out from the nearest control points either by ‘polars’ or
by ‘intersection’, as follows:
(7) Using the co-ordinates, compute the bearing and distance from, say, station Y to T1,
A and B.
(8) Set up theodolite at Y and backsight to X; set the horizontal circle to the known
bearing YX.
(9) Now turn the instrument until it reads the computed bearing YT1 and set out the
computed distance in that direction to fix the position of T1. Repeat the process for
A and B. The ideal instrument for this is a total station, many of which will have
onboard software to carry out the computation in real time. However, provided that
the ground conditions are suitable and the distances within, say, a 50m tape length,
a theodolite and steel tape would suffice.

Other points around the curve are set out in the same way from appropriate control
points.
Intersection may be used, thereby precluding distance measurement, by computing
the bearings to the curve points from two control stations. For instance, the theodolites
 are set up at Y and Z respectively. Instrument Y is orientated to Z and the bearing YZ
set on the horizontal circle. Repeat from Z to Y. The instruments are set to bearings YB
and ZB respectively, intersecting at peg B. The process is repeated around the curve.
Using co-ordinates eliminates many of the problems encountered in curve ranging
and does not require the initial establishment of tangent and intersection points.

8.2.5 Setting out with two tapes (method of offsets)

Theoretically, this method is exact, but in practice errors of measurement propagate round
the curve. It is therefore generally used for minor curves.
In Figure 8.8, line OE bisects chord T1A at right-angles, then ET1O = 90◦ - σ, CT1A
= σ1, and triangles CT1A and ET1O are similar, thus

CA/T1A = T1E/T1O … CA = T1E/T1O × T1A

i.e. offset CA = ½ chord × chord/radius = chord2/2R (8.3)

From Figure 8.8, assuming lengths T1A = AB = AD

D
F
A B
C
90
E
90 L2
R G
R
T1 H

R J

Figure 8.8

then angle DAB = 2σ, and so offset DB = 2CA = chord2/R (8.4)

 The remaining offset round the curve to T2 are all equal to DB whilst, if required,
the offset HJ to fix the line of the straight from T2, and equals CA.
The method of setting out is as follows:
It is sufficient to approximate distance T1C to the chord length T1A and measure
this distance along the tangent to fix C a right-angled offset CA fixed the first peg at A.
Extend T1A to D so that AD equals a chord length; peg B is then fixed by pulling out offset
length from D and chord length from A, and where they meet is the position B. This
process is continued to T2.
The above assumes equal chords. When the first and last chords are sub-chords, the
following (Section 8.2.6) should be noted.

8.2.6 Setting out by offsets with sub-chords

In figure 8.9 assume T1A is a sub-chord of length x; from equation (8.3) the offset CA = O1
= x2/2R.
As the normal chord AB differs in length from T1A, the angle subtended at the centre will
be 2θ not 2σ. Thus as shown in figure 8.8 the offset DB will not in this case equal 2CA.
Construct a tangent through point A, then from the figure it is obvious that angle
EAB = θ, and if chord AB = y, then offset EB = y2/2R.
Angle DAE = σ, therefore offset DE will be directly proportional to the chord
length, thus:

DE = O1/x . y = x2y/2Rx = xy/2R

Thus the total offset DB = DE + EB

y/2R(x + y) (8.5)

i.e. = chord/2R(sub-chord + chord)

Figure 8.9
Thus having fixed B, the remaining offset to T2 are calculated as y2/R and set out in the
usual way.
If the final chord is a sub-chord of length x1, however, then the offset will be

x1/2R(x1 +y) (8.6)

Students should note the difference between equations (8.5) and (8.6)
A more practical approach to this problem is actually establish the tangent through
A in the field. This is done by swinging an arc of radius equal to CA, i.e. x2/2R from T1. A
line tangential to the arc and passing through peg A will then be the required tangent from
which offset EB, i.e. y2/2R, may be set off.

8.2.7 Setting out with inaccessible intersection point.

In Figure 8.10 it is required to fix T1 and T2, and obtain the angle ∆, when l is inaccessible.
Project the straights forward as far as possible and establish two points A and B on
them. Measure distance AB and angles BAC and CBA then:

Building
1 ∆
C

B
A
T1 T2
R R

0
EC D

Figure 8.10

angle lAB = 180◦ - BAC and angle lBA = 180° - DBA, from which angle BlA is deduced
and angle ∆. The triangle AlB can now be solved for lengths lA and lB. These lengths
when subtracted from the computed tangent lengths (R tan ∆/2), give AT1 and BT2, which
are set off along the straight to give positions T1 and T2 respectively.

8.2.8 Setting out with theodolite at an intermediate point on the curve

Due to an obstruction on the line of sight (Figure 8.11) or difficult communications or

visibility on long curves. It may be may be necessary to continue the curve by ranging from
a point on the curve. Assume that the setting-out angle to fix peg 4 is obstructed. The
theodolite is moved to peg 3, backsighted to T1 with the instrument reading 180◦, and then
turned to read 0◦, thus giving the direction 3 – T. The setting-out angle for peg 4, σ4, is
turned off and the chord distance measured from 3. The remainder of the curve is now set
off in the usual way, that is , σ5 is set on the theodolite and the chord distance measured
from 4 to 5.
The proof of this method is easily seen by constructing a tangent through peg 3, the
angle A3T1 = AT13 = σ3 = T3B. If peg 4 was fixed by turning off σ from this tangent, then
the required angle from 3T would be σ3 + σ = σ4.

8.2.9 Setting out with an obstruction on the curve

In this case (Figure 8.12) an obstruction from the curve prevents the chaining of the chord
from 3 to 4. One may either:

(1) Set out the curve from T2 to the obstacle.

(2) Set out the chord length T14 = 2R sin σ4 (EDM).
(3) Set out using intersections from theodolites at T1 and T2.
(4) Use co-ordinate method.

8.2.10 Passing a curve through a given point

In Figure 8.13, it is required to find the radius of a circle which will pass through a point P,
the position of which is defined by the distance lP at an angle of φ to the tangent.

A 3
2 3 4 5
4 B
1 Tangent
3 4 5 Through
Peg3
T1 R

Figure 8.11 1 1

2 3

1 4

T1
T2
R

0
 Figure 8.12

I ∆

P
T2
ϴR
T1

R
R
0

Figure 8.13

Consider triangle lPO:

angle β = 90◦ - ∆/2 – φ (right-angled triangle LT2O )

by sin rule: sin α = lO/POsin β but Lo = R sec∆/2

sin α = sin β R sec∆/2/R = sin β sec∆/2

then θ = 180° - α – β, and by the sine rule: R = lP sin β/sin θ

8.3 COMPOUND AND REVERSE CURVES

Although equation are available which solve compound curves, (Figure 8.14) and reverse
curves (Figure 8.15), they are difficult to remember and students are advised to treat the
problem as two simple curves with a common tangent point t.
In the case of the compound curve, the total tangent lengths T1l and T2l are found as
follows:

R1 tan ∆1/2 = T1t1 = t1t and R2 tan ∆2/2 = T2t2 = t2t, as t1t2 = t1t + t2t

then t1lt2 maybe solved for lengths t1l and t2l which, if added to the known lengths T1t1 and
T2t2 respectively, give the total tangent lengths.
 I ∆=∆1+∆2

∆1 ∆2
t1 t t2
∆1/2
T1
∆1 T2
R1
01
R2
∆2
A B
02

Figure 8.14

I
∆1
02

∆2 R2

t
T2
T1

R1
∆1

01

∆2 I
2

Figure 8.15

In setting out this curve, the first curve R1 is set in the usual way to point t. The theodolite
is moved to t and backsighted to T1, with the horizontal circle reading (180° - ∆1/2). Set the
instrument to read zero and it will then be pointing to t2. Thus the instrument is now oriented
and reading zero, prior to setting out curve R2.
Incase of the reverse curve, both arcs can be set out from the common point t.
8.4 SHORT AND/OR SMALL RADIUS CURVES

Short and/or small radius curves such as for kerb lines, bay windows or for the construction
of large templates may be set out by the following methods.

8.4.1 Offsets from the tangent

The position of the curve (in Figure 8.16) is located by right-angled offsets Y set out from
distances X, measured along each tangent, thereby fixing half of the curve from each side.
The offsets may be calculated as follows for given distance X. Consider offset Y3, for
example.

In ∆ABO, AO2 = OB2 – AB2 (R – Y3)2 = R2 – X32 and Y3 = R- (R2 – X32)½

Thus for any offset Yi at distance Xi along the tangent

Yi = R – (R2 – Xi2)½ (8.7)

I
∆

D
Y3
Y2

B
PY1
X3

T1 T2
R
Y3

Figure 8.16

8.4.2 Offsets from the long chord

In this case (Figure 8.17) the right angled offsets Y are set off from the long chord C, at
distances X to each side of the centre offset Y0.
An examination of Figure 8.17, shows the central offset Y0 equivalent to the
distance T1A on Figure 8.16; thus:
 Y0 = R – [R2 – (C/2)2]½

Similarly, DB is equivalent to DB on Figure 8.16, thus: DB = R – (R2 – X12)1/2

and offset Y1 = Y0 – DB .. Y1 = Y0 – [R – (R2 –X12)1/2]
and for any offset Yi at distance Xi each side of the

mid point of T1T2: Yi = Y0 – [R – (R2 – Xi2)1/2] (8.8)

Therefore after computation of the central offset, further offsets at distances X i, each side
of Y0, can be found.

8.5 TRANSITION CURVES

The transition curve is a curve of constantly changing radius. If used to connect a straight
to a curve of radius R, then the commencing radius of the transition will be the same as the
straight , and the final radius will be that of the curve R (see Figure 8.28).
Consider a vehicle traveling at speed (V) along a straight. The forces acting on the vehicle
will be its weight W, acting vertically down, and an equal and opposite force acting
vertically up through the wheels. When the vehicle enters through the radius R at tangent
point T1, an additional centrifugal force (P) acts on the vehicle, as shown in Figures 8.24
and 8.25. If P is large the vehicle will be forced to the outside of the curve and may skid or
overturn. In Figure 8.25 the resultant of the two forces is shown as N, and if the road is
super-elevated normal to this force, there will be no tendency for the vehicle to skid. It
should be noted that as

P = WV2/Rg equ1 (8.9)

Super-elevation will be maximum at minimum radius R.

It therefore requires a length of spiral curve to permit the gradual introduction of super-
elevation, from zero at the start of the elevation of the transition to maximum at the end,
where the radius is the minimum safe radius R.
w

(a)
P
P R
O

w
Figure 8.25
w

(b)

Figure 8.24
8.5.1 Principle of the transition

The purpose then of a transition curve is to:

(1) Achieve a gradual change of direction from the straight (radius ∞) to the curve
(radius R).
(2) Permit the gradual application of super-elevation to counteract centrifugal force and
minimize passenger discomfort.
Since P cannot be eliminated, it is allowed for permitting it to increase uniformly along
the curve. From equation (8.9), as P is inversely proportional to R., The basic requirement
of the ideal transition curve is that its radius should decrease uniformly with distance along
it. This requirement also permits the uniform application of the super-elevation; thus the
distance l along the transition of radius is r and rl = c(constant):

l/c = 1/r eq2

From Figure 8.26, tt1 is an infinitely small portion of a transition δl of radius r; thus:

δl = rδφ eq3
1/r = δφ/ δl which on substitution above gives
l/c = δφ/ δl

integrating: φ = l²/2c .. l = (2cφ)1/2

putting a = (2c)1/2
l = a(φ)1/2 (8.10)
When c = RL, a = (2RL)1/2 and (8.10) may be written:
l = (2RLφ)1/2 eq4 (8.11)

The above expressions are for the clothiod curve, sometimes called the Euler spiral, which
is the one most used in road design.

t1

t
r

T1
0
Figure 8.26
8.5.2 Curve design

The basic requirements in the design of transition curves are:

(1) The value of the minimum safe radius (R), and

(2) The length (L) of the transition curve. Sections 8.5.5 or 8.5.6.

The value R may be found using either of the approaches Sections 8.5.3 or 8.5.4.

8.5.3 Centrifugal ratio

Centrifugal force is defined as P = WV²/Rg; however, this ‘overturning force’ is

counteracted by the mass (W) of the vehicle, and may be expressed as P/W, termed the
centrifugal ratio. Thus, centrifugal ratio:

P/W = V²/Rg (8.12)

where V is the design speed in m/s, g is acceleration due to gravity in m/s² and R minimum
safe radius in metres.
When V is expressed in km/h, the expression becomes

P/W = V²/127R (8.13)

Commonly used values for centrifugal ratio are

0.21 to 0.25 on roads, 0.125 on railways

Thus, if a value of P/W = 0.25 is adopted for a design speed of V = 50km/h, then

R = 50²/127 × 0.25 = 79m

The minimum safe radius R may be set either equal to or greater than this value.

8.5.4 Coefficient of friction

The alternative approach to find R is based on Road Research Laboratory (RRL) values for
the coefficient of friction between the car tyres and the road surface.
Figure 8.27(a) illustrates a vehicle passing around a super-elevated curve. The
resultant of the two forces is N. The force F acting towards the centre of the curve is the
friction applied by the car tyres to the road surface. These forces are shown in greater detail
in Figure 8.27(b) from which it can be seen that

F2 = WV²/Rg cos θ and F1 = W cos(90 – θ) = W sin θ

F = F2 – F1 = WV²/Rg cos θ - W sin θ

Similarly N2 = WV²/Rg sin θ, and N1 = W cos θ

 N = N2 + N1 = WV²/Rg sin θ + W cos θ

Then F/N = (WV²/Rg cos θ - W sin θ)/ (WV²/Rg sin θ + W cos θ) = V²/Rg - tan θ/
V²/Rg tan θ + 1

For department of transport (DTp) requirements, the maximum value for tan θ = 1 in 14.5 =
0.069, and as V²/Rg cannot exceed 0.25 the term in the denominator can be ignored and

F/N = V²/Rg - tan θ = V²/127R - tan θ (8.14)

F2
(A) (B)

A P 0

90○
2 F F1
P=WV 0
Rg 0
0
C B
W
N N1

W
N

Figure 8.27
prevent vehicles slipping sideways. F/N must be greater than the coefficient of friction u
between tyre and road. The RRL quote value for u of 0.15, whilst 0.18 may be used up to
50 km/h, ******

V²/127R >* tan θ (8.15)

*** example, if the design speed is to be 100km/h, super-elevation limited to 1 in 14.5

(0.069)***** u = 0.15, then

100²/127R = 0.069 + 0.15

… R = 360m

In the UK, the geometric parameters used in design are normally related to design speed.
Table 8.3 shows typical desirable and absolute minimum values for horizontal and vertical
curvature; there is also an additional lower level designated ‘limiting radius’, specific to
horizontal curvature.
Designs for new roads should aim to achieve desirable values for each design parameter.
However, absolute minimum values can be used wherever substantial saving in
construction or environmental costs can be achieved.
DTp Technical standard TD9/81 advises that in the design of new roads, the use of radii
*** than the limiting values is undesirable and not recommended.

8.5.5 Rate of application of super-elevation

It is recommended that on motorways super-elevation should be applied at a rate of 1 in

200, on full-purpose roads at 1 in 100, and on railways at 1 in 480. Thus, if on a motorway
the super-elevation is computed as 0.5m, the 100m of transition curve would be required to
accommodate 0.5 at the required rate of 1 in 200, i.e. 0.5 in 100 = 1 in 200. In this way the
length L of the transition is found.
The amount of super-elevation is obtained as follows:
From the triangle of forces in Figure 8.27(a)

tan θ = V²/Rg = 1/H = 1 in H

thus H = Rg/ V² = 127R/ V², where V is in km/h.

In the UK, the design speed V is replaced by the ‘average speed’ which is 63.6% of V,

and H = 127R/(0.636V) ² = 314R/ V² (8.16)

super-elevation = 1 in 314R/ V²

The percentage super-elevation (or crossfall), S may be found from:

S = V²/2.828R (8.17)

The DTp recommend the crossfall should never be greater than 1 in 14.5, or less than 1 in
48, to allow rain water to run off the road surface.
It is further recommended that adverse camber should be replaced by a favourable crossfall
of 2.5% when the value of V²/R is greater than 5 and less than 7 (see Table 8.3).
Driver studies have shown that whilst super-elevation is instrumental to driver comfort and
safety, it need not be applied too rigidly. Thus for sharp curves in urban areas with at-grade
junctions and side access, super-elevation should be limited to 5%.
The rate of crossfall combined with the road width, allows the amount of super-elevation to
be calculated. Its application at the given rate produces the length L of transition required.

8.5.6 Rate of change of radial acceleration

An alternative approach to finding the length of the transition is to use values for ‘rate of
change of radial acceleration’ which would be unnoticeable to passengers traveling by
train. The appropriate values were obtained empirically by W.H. Shortt, an engineer
working for the railways; hence it is usually referred to as Shortt’s Factor.

Radial acceleration = V²/R

Thus, as radial acceleration is inversely proportional R it will change at a rate

proportional to the rate of change in R. The transition curve must therefore be long enough
to ensure that the rate of change of radius, and hence radial acceleration, is unnoticeable to
passengers.
Acceptable values for rate of change for radial acceleration (q) are 0.3m/s3 and 0.6
m/s3.
Now, as radial acceleration is V²/R and the time taken to travel the length L of the
transition curve is L/V, then

Rate of change of radial acceleration = q = V²/R ÷ L/V = V3/RL

L = V3/Rq = V3/3.63.R.q (8.18)

where the design speed (V) is expressed in km/h.

Although this method was originally designed for railway practice, it is also applied
to road design. q should normally not be less than 0.3m/s3 for unrestricted design, although
in urban areas it may be necessary to increase it to 0.6m/s3or even higher, for sharp curves
in tight locations.

8.6 SETTING-OUT DATA

Figure 8.28 indicates the usual situation of two straights projected forward to intersect at l
with a clothoid transition curve commencing from tangent point T1 and joining the circular
arc at t1, The second equal transition commences at t2 and joins at T2. Thus the composite
curve from T1 to T2 consists of a circular arc with transitions at entry and exit.

(1) Fixing the tangent points T1 and T2

In order to fix T1 and T2 the tangents lengths T1l and T2l are measured from l back down
the straights or they are set out direct by co-ordinates.

T1l = T2l = (R + S) tan ∆/2 + C (8.19)

I
∆

B
X Y t1 t
2
circular arc
θ θ
A D
Transition
T1 T2
R R
∆-2

Figure 8.28
where S = shift = L2/24R – L4/3! × 7 × 8 × 23R3 + L6/5! × 11 ×12 × 25R5 – L8/7! × 15
× 16 × 27R7…..
and C = L/2 – L3/2! × 5 × 6 × 22R2 + L5/4! × 9 × 10 × 24R4 – L7/6! ×13 × 14 × 26R6…..

The values of S and C are abstracted from the Highway Transition Curve Tables (Metric)
(see Table 8.2).

(2) Setting out the transitions

Refering to Figure 8.29:

The theodolite is set at T1 and oriented to l with the horizontal circle reading zero.
The transition is then pegged out using deflection angles (θ) and chords (Rankine’s
method) in exactly the same way as for a simple curve.
The data are calculated as follows:

(a) The length of transition L is calculated (see design factors in sections 8.5.5
and 8.5.6), assume L = 100m.

Θ2 t1
Θ1 2 End of transition
T

Figure 8.29……………………………..

(b) It is then split into, say, 10 arcs, each 10 m length (ignoring through
chainage), the equivalent chord lengths obtained from:
A- A3/24R2 + A5/ 1920R4 , where A is the arc length

(c) The setting-out angles θ1,θ2,…… θn are obtained as follows:

Basic formula for clothoid: l = (2RLφ)1/2

Φ = l2/2RL = L/2R when l = L (8.20)

( l is any distance along the transition other than total distance L)

then the θ = Φ/3 - 8 Φ3/2835 - 32 Φ5/467775…... (8.21)

= Φ/3 – N, where N is taken from Tables and ranges in value from 0.1” when Φ = 3°, to
34’41.3” when Φ = 86° (see Table 8.1).
 Now φ1/ Φ = l12/L2 (8.22)

φ1 = Φ l12/L2 where l1 = chord length = 10m, say

Table 8.1 Interpolated deflection angles

For any point on a spiral where angle consumed = φ, the true deflection is φ/3 minus the correction Tabled below.
The back angle is 2 φ/3 plus the same correction.

Angle φ/3 Deduct Deflection Angle φ/3 Deduct Deflection

consumed N angle consumed N angle
− −
φ φ
° ° ‘ ‘ “ ° ‘ “ ° ° ‘ ‘ “ ° ‘ “

NIL
2 0 40 0.1 40 0 45 15 00 4 46.2 14 55 13.8
3 1 00 0.2 59 59.9 46 15 20 5 6.0 15 14 54.0
4 1 20 1 19 59.8 47 15 40 5 26.6 15 34 33.4

5 1 40 0.4 1 39 59.6 48 16 00 5 48.1 15 54 11.9

6 2 00 0.7 1 59 59.3 49 16 20 6 10.6 16 13 49.4
7 2 20 1.0 2 19 59.0 50 16 40 6 34.1 6 33 25.9

Continued at 1° intervals of φ

41 13 40 3 35.9 13 36 24.1 84 28 00 32 14.4 27 27 45.6

42 14 00 3 52.1 13 56 7.7 85 28 20 33 26.9 27 46 33.1
43 14 20 4 9.4 14 15 50.6 86 28 40 34 41.3 28 5 18.7

44 14 40 4 27.4 14 35 32.6

Reproduced with permission of the county surveyors’ Society

and θ1 = φ1/3 = N1 ( where N1 is the value relative to φ1 in Table 8.1)

Similarly φ2 = Φ l22/L2 where l2 = 20m

and θ2 = φ2/3 – N2 and so on

Students should note:

(a) The values for l1, l2, are accumulative.

(b) Thus the values obtained for θ1, θ2 etc. are the final setting-out angles and are
obviously not to be assumed.
 (c) Although the chord length used is accumulative, the method of setting out is still the
same as for the simple curve.

(3) Setting out circular arc t1t2

In order to set out the circular arc it is first necessary to establish the direction of the
tangent t1B (Figure 8.28). The theodolite is set at t1 and backsighted to T1 with the
horizontal circle reading [180° - (Φ – θ)]. Setting the instrument to zero will now orient it
in the direction t1B with the circle reading zero, prior to the setting-out the simple circular
arc. The angle (Φ – θ) is called the back-angle to the origin and may be expressed as
follows:

θ = Φ/3 – N
(Φ –θ) = Φ – (Φ/3 – N) = 2/3 Φ + N (8.23)

and is obtained direct from tables.

The remaining setting out data are obtained as follows:

(a) As each transition absorbs an angle Φ, then the angle subtending the circular
arc = (∆ - 2 Φ).
(b) Length of circular arc = R(∆ - 2 Φ), which is then split into the required chord lengths
C.
(c) The deflection angles δ° = 28.6479.C/R are then set out from the tangent t1B in the
usual way.

The second transition is best set out from T2 to t2, setting-out from t2 to T2 involves the
‘osculating circle’ technique (see Section 8.9).
The preceding formulae for clothoid transitions are specified in accordance with the latest
Highway Transition Curve Tables (Metric) compiled by the county Surveyors’ Society. As
the equations involved in the setting-out data are complex, the information is generally
taken straight from tables. However, approximation of the formulae produces two further
transition curves, the cubic spiral and cubic parabola (see Section 8.7).
In the case of the clothoid, Figure 8.28 indicates an offset Y at the end of the transition,
distance X along the straight, where

X = L – L3/5 × 4 × 2!R2 + Ls/9 × 42 × 4!R4

- L7/13 × 43 ×6!R6 + ……… (8.24)
Y = L2/3 × 2R – L4/7 × 3! 23R3 + L6/11 × 5! × 25R5
- L8/15 × 7! × 27R7 + … (8.25)

The clothoid is always set out by deflection angles, but the values for X and Y are useful in
the large-scale plotting of such curves, and are taken from tables.
Refer to the end of the chapter for derivation of clothoid formulae.

8.6.1 Highway transition curve tables (metric)

An examination of the complex equations defining the clothoid transition spiral indicates
the obvious need for tables of prepared data to facilitate the design and setting out of such
curves. These tables have been produced by the County Surveyors’ Society under the title
Highway Transition Curves Tables (Metric), and contain a great deal of valuable
information relating to the geometric design of highways. A very brief sample of the tables
is given here simply to convey some idea of the format and information contained therein
(Table 8.2)
As shown in Section 8.6(3), θ = φ/3 – N and the ‘back-angle’ is 2 φ/3 + N, all this
information for various values of φ is supplied in table 8.1 and clearly shows that for large
values of φ, N cannot be ignored.
Part only of Table 8.2 is shown and it is the many tables like this that provide the bulk of
the design data. Much of the information and its application to setting out should be easily
understood by the student, so only a brief description of its use will be given here.

Use of tables

(1) Check the angle of the intersection of straights (∆) by direct measurement in the
field.
(2) Compare ∆ with 2Φ, if ∆≤2Φ, then the curve is wholly transitional.
(3) Abstract (R+S) and C in order to calculate the tangent lengths = (R+S) tan ∆/2 + C.
(4) Take Φ from the tables and calculate length of circular arc using R(∆ - 2Φ), or, if
working in ‘degree of curvature’ D, use

100(∆ - 2Φ)
D

(5) Derive chainages at the beginning and end of both transitions.

(6) Compute the setting-out angles for the transition θ1 = θ from φ1/Φ = l12/L2 from
which θ1 = φ1/3 – N1, and so on, for accumulative values of l.
(7) As control for setting out, the end point of the transition can be fixed by turning off
from T1 (the start of the transition) the ‘deflection angle from the origin’ θ and
laying out the ‘long chord’ as given in the Tables. Alternatively, the right-angled
offset Y and distance X along the tangent may be used.
(8) When the first transition is set out, set up the theodolite at the end point and with
the theodolite reading (180° - (⅔Φ + N), backsight to T1. Turn the theodolite to
read 0° when it will be pointing in a direction tangential to the start of the circular
curve prior to its setting out. This process has already been described.
(9) As a check on the setting out of the circular curve take (R + S) and S from the
Tables to calculate the apex distance = (R + S) (sec ∆/2 – 1) + S, from the
intersection point l of the straights to the centre of the circular curve.
(10) The constants RL and D/L are given at the head of the tables and can be used as
follows:

(a) Radius at any point P on the transition = rp = RL/lp

(b) Degree of curve at P = Dp = (D/L) × lp, where lp is the distance to
P measured along the curve from T1.
 Similarly;
(c) Angle consumed at P = φp = lp2 lp2 D
or ×
2RL 200 L

8.10 VERTICAL CURVES

Vertical curves (VC) are used to connect intersecting gradients in the vertical plane. Thus,
in route design they are provided at all changes of gradient. They should be of sufficiently
large curvature to provide comfort to the driver, that is, they should have a low ‘rate of
change of grade’. In addition, they should afford adequate ‘sight distances ‘for safe
stopping at a given design speed.
The type of curve generally used to connect the intersecting gradients g1 and g2 is the
simple parabola. Its use as a sag or crest curve is illustrated in figure 8.37.

8.10.1 Gradients

In vertical curve designs the gradients are expressed as percentages, with a negative for a
downgrade and a positive for an upgrade.

e.g. A downgrade of 1 in 20 = 5 in 100 = -5% = -g1%

An upgrade of 1 in 25 = 4 in 100 = +4% = +g2%

The angle of deflection of the two intersecting gradients is called the grade angle and
equals A in Figure 8.37. The grade angle simply represents the change of grade through
which the vertical curve deflects and is the algebraic difference of the two gradients.

A% = (g1% - g2%)
I
T1 A%
T2
sag g2%
g1%
g1% g2%
Crest
T1 T2
A%
g1% I
T1
g1%
T1 I
T2 A
I A
g1% T2
g1%

g2% A T2 g2%
I
T2 T1

g1%
A
T1
I
g1%
 Figure 8.37

In the above example A% = (-5% - 4%) = -9% (negative indicates a sag curve).

8.10.2 Permissible approximations in vertical curve computation

In civil engineering, road design is carried out in accordance with the following documents:

(1) Layout of roads in rural areas.

(2) Roads in urban areas.
(3) Motorway design memorandum.

However, practically all the geometric design in the above documents has been replaced by
Department of Transport Standard TD 9/93, hereafter referred to simply as TD 9/93, with
Advice Note TD 43/84.
In TD 9/93 the desirable maximum gradients for vertical curve design are:

Motorways 3%
Dual carriageways 4%
Single carriageways 6%

Due to the shallowness of these gradients, the following VC approximations are

permissible, thereby resulting in simplified computation (Figure 8.38)

(1) Distance T1D = T1BT2 = T1CT2 = T1l + lT2), without sensible error. This is very
important and means that all distances may be regarded as horizontal in both the
computation and the setting out of vertical curves.
(2) The curve is of equal length each side of l. Thus T1C = CT2 = T1l = lT2 = L/2,
without sensible error.
(3) The curve bisects Bl at C, thus BC = Cl = Y (the mid-offset).
(4) From similar triangles T1Bl and T1T2l, if Bl =2Y, the T2l = 4Y.4Y represents the
vertical divergence of the two gradients over half the curve length (L/2) and
therefore equals AL/200.
(5) The basic equation for a simple parabola is

y = C.l2
 -g1%
Horizontal
T1 D

L1
Y1 B

Y
Y2 C
L1 C G
T2+g2%
Y F
L/2 H
I
M

L
J

Figure 8.38
where y is the vertical offset from gradient curve, distance l form the start of the curve, and
C is a constant. Thus, as the offsets are proportional to distance squared, the following
equation is used to compute them:

y1 l12
= (8.37)
Y (L/2)2

where Y = the mid-offset = AL/800 (see Section 8.10.7)

8.10.3 Vertical curve design

In order to set out a vertical curve in the field, one requires levels along the curve at given
chainage intervals. Before the levels can be computed, one must know the length L of the
curve. The value of L is obtained from the parameters supplied in Table 3 of TD 9/93
(reproduced below as Table 8.3) and the appropriate parameters are K-values for specific
design speeds and sight distances; then

L = KA (8.38)

where A = the difference between the two gradients (grade angle)

K = the design speed related coefficient (Table 8.3)
e.g. A + 4% gradient is linked to a – 3% gradient by a crest curve. What length of the
curve is required for a design speed of 100km/h?

A = (4% - (-3%)) = +7% (positive for crest)

From Table 8.3

C1 Desirable minimum crest K-value = 100

C2 One step below desirable minimum crest K-value = 55
from L = KA
Desirable minimum length L = 100 × 7 = 700m
One step below desirable minimum length = L = 55 × 7 = 385m

Wherever possible the vertical and horizontal curves in the design should be co-
coordinated so that the sight distances are correlated and a more efficient overtaking
provision is ensured.
The various design factors will now be dealt with in more detail.

8.10.3.1 K-value

Rate of change of gradient (r) is the rate at which the curve passes from one gradient (g1%)
to the next (g2%) and is similar in concept to rate of change of radial acceleration in
horizontal transitions. When linked to design speed it is termed rate of vertical acceleration
and should never exceed 0.3m/s2.
A typical design of badly designed vertical curve with a high rate of change of grade is a
hump-backed bridge where usually the two approaching are quite steep and connected by a
very short length of vertical curve. Thus one passes through a large grade angle A in a very
short time, with the result that often a vehicle will leave the ground and/or cause great
discomfort to its passengers.

Table 8.3

Design speed (kph) 120 100 85 70 60 50 V2/R

A Stopping sight distance, m

Desirable minimum 295 215 160 120 90 70
One step below desirable minimum 215 160 120 90 70 50
B Horizontal curvature, m
Minimum R* without elimination of adverse 2880 2040 1440 1020 720 510 5
camber and transitions
Minimum R* with superelevation of 2.5% 2040 1440 1020 720 510 360 7.07
Minimum R* with superelevation of 3.5% 1440 1020 720 510 360 255 10
Desirable minimum R with 1020 720 510 360 255 180 14.14
superelevation of 5%
One step below desirable minimum 720 510 360 255 180 127 20
R with superelevation of 7%
Two steps below desirable minimum 510 360 255 180 127 90 28.28
radius with superelevation of 7%

C Vertical curvature
C1 Desirable minimum* crest K-value 182 100 55 30 17 10
C2 One step below desirable minimum 100 55 30 17 10 6.5
crest K-value
C3 Absolute minimum sag K-value 37 26 20 20 13 9

Overtaking sight distances

C4 Full overtaking sight distances FOSD, m * 580 490 410 345 290
 C5 FOSD overtaking crest K-value * 400 285 200 142 100

* Not recommended for use in the design of single carriageways.

The V2/R values shown in Table 8.3 above simply represent a convenient means of
identifying the relative levels of design parameters, irrespective of design speed.
(Reproduced with the permission of the Controller of Her Majesty’s Stationery Office)

Commonly used design values for r are

3% on crest curves
1.5% on sag curves
thereby much larger curves to prevent rapid change of grade and provide adequate sight
distances.
Working from first principles if g1 = -2% and g2 = +4% (sag curve), then the change of
grade from -2% to +4% = 6%A, the grade angle. Thus to provide for a rate of change of
grade of 1.5%, one would require 400m (L) of curve. If the curve was a crest curve, then
using 3% gives 200m (L) of curve:

L = 100A/r (8.39)

Now expressing rate of change of grade as a single number we have

K = 100/r (8.40)

and as shown previously, L = KA.

8.10.3.2 Sight distances

Sight distance is a safety design factor which is intrinsically linked to rate of change of
grade, and hence to K-values.
Consider once again the hump-backed bridge. Drivers approaching from each side of this
particular vertical cannot see each other until they arrive, simultaneously, almost on the
crest: by which time it may be too late to prevent an accident. Had the curve been longer
and flatter, thus resulting in a low rate of change of grade, the drivers would have had a
longer sight distance and consequently and more time in which to take avoiding action.
Thu, sight distance, i.e. the length of road ahead that is visible to a driver, is a safety factor
and it is obvious that the sight distance must be greater than the stopping distance in which
the vehicle can be brought to rest.
Stopping distance is dependent upon:

(1) Speed of the vehicle

(2) Braking efficiency
(3) Gradient
(4) Coefficient of friction between tyre and road
(5) Road condition
 (6) Driver’s reaction time

In order to cater for all the above variables, the height of the driver’s eye above the road
surface is taken as being only 1.05m; a height applicable to sports cars whose braking
efficiency is usually very high. Thus, other vehicles such as lorries, with a much greater
eye height, would have a much longer sight distance in which to stop.

8.10.3.3 Sight distances on crests

Sight distances are defined as follows:

(1) Stopping sight distance (SSD) (Figure 8.39)

The SSD is the sight distance required by a driver to stop a vehicle faced with an
unexpected obstruction on the carriageway. It comprises two elements:

(a) The perception reaction distance, which is the distance traveled from the time the
driver sees the obstruction to the time it is realized that the vehicle must stop; and
(b) The braking distance, which is the distance traveled before the vehicle halts short of
the obstruction.

2m 2m

Envelope of visibility

1.05m

0.26m

Figure 8.39

The above are a function of a driver age and fatique, road conditions, etc. and thus the
design parameters are based on average driver behaviour in wet conditions. Table 8.3
provides values for desirable and absolute minimum SSD.
It has been shown that 95% of the driver’s eye height is 1.05m or above; the upper limit of
2m represents large vehicles.
The height of the obstruction is between 0.26m and 2.0m. Forward visibility should be
provided in both horizontal and vertical planes between points in the centre of the lane
nearest the inside of the curve.

(2) Full overtaking sight distance (FOSD) (Figure 8.40)

On single carriageways, overtaking in the lane of the opposing traffic occurs. To do so in
safety requires an adequate sight distance which will permit the driver to overtake in the
normal overtaking procedure.
The FOSD consists of four elements:

(a) The perception/reaction distance traveled by the vehicle whilst the decision to
overtake or not is made.
(b) The overtaking distance traveled by the vehicle to complete the overtaking
manoeuvre.
(c) The closing distance traveled by the oncoming vehicle whilst overtaking is
occurring.
(d) The safety distance required for clearance between the overtaking and oncoming
vehicles at the instant the overtaking vehicle has returned to its own lane.

It has been shown that 85% of overtaking takes place in 10 seconds and Table 8.3 gives
appropriate FOSD values relative to design speed.
It should be obvious from the concept of FOSD that it is used in the design of single
carriageways only, where safety when overtaking is the prime consideration.
For instance, consider the design of a crest curve on dual carriageway with a design speed
of 100km/h.
From Table 8.3:

C1 Desirable minimum K-value = 100

C2 One step below desirable minimum K-value = 55
C4 FOSD K-value = 400
As overtaking is not a safety hazard on a dual carriageway, FOSD is not necessary and one
would use:

2m 2m

Envelope of visibility

1.05m 1.05m

Figure 8.40
 L = 100 A (desirable minimum)
or L = 55 A (one step below desirable minimum)

Had the above road been a single carriageway then FOSD would be required and :

L = 400A

If this resulted in too long a curve, with excessive earthworks, then it might be decided to
prohibit overtaking entirely, in which case:

L = 55 A
would be used.
Although equations are unnecessary when using design tables, they can be developed to
calculate curve lengths L for given sight distances S, as follows:

(a) When S<L (Figure 8.41)

From basic equation y = Cl2

Y = C(L/2)2, h1 = C(l1)2 and h2 = C(l2)2
then h1 = l12 4l12 h2 4l22
= and =
Y (L/2)2 L2 Y L2

thus l12 = h1L2 AL

but since 4Y =
4Y 200

l12 = 200h1L

and l1 = (h1)1/2 (200L/A)1/2

Similarly l2 = (h2)1/2 (200L/A)1/2

S = (l1 + l2) = [(h1)1/2 + (h2)1/2] (200L/A)1/2 (8.41)

and L = S2A/200[(h1)1/2 + (h2)1/2]2 (8.42)

when h1 = h2 = h

L = S2A/800h (8.43)

(b) When S>L it can similarly be shown that

 L = 2S – 200/A[(h1)1/2 + (h2)1/2]2 (8.44)

and when h1 = h2 = h
L = 2S – 800h/A (8.45)

When S = L, substituting in either of equations (8.43) or (8.45) will give the correct
solution.

I
A

L1 L2

+g1%
-g2%
h2
h1

Y
T1 T2
L

Figure 8.41

e.g. (8.43) L = S2A/800h = L2A/800h = 800h/A

and (8.45) L = 2S – 800h/A = 2L - 800h/A = 800h/A

N.B. If the relationship of S to L is not known then both cases must be considered; one of
them will not fulfil the appropriate argument S<L or S>L and is therefore wrong.

8.10.3.4 Sight distances on sags

Visibility on sag curves is not obstructed as it is in the case of crests; thus sag curves are
designed for at least absolute minimum comfort criteria of 0.3m/sec2. However, for design
speeds of 70km/h and below in unit areas, sag curves are designed to ensure that headlamps
illuminate the road surface for at least absolute minimum SSD. The relevant K values are
given in C3, Table 8.3.
The headlight generally considered as being 0.6m above the road surface with its beam
titled up at 1° to the horizontal. As in the case of crests, equations can be developed if
required.
Consider Figure 8.42 where L is greater than S. From the equation for offsets:

BC/T2D = S2/L2 .. BC = S2(T2D)/L2

but T2D is the vertical divergence of the gradients and equals

A.L/100.2 .. BC = A.S2/200L (a)

also BC = h + S tan x° (b)

Equating (a) and (b): L = S2A(200h + 200Stan x°)-1 (8.46)

putting x° = 1° and h = 0.6m

L = S2A/120 + 3.5S (8.47)

Similarly; when S is greater than L (Figure 8.43)

BC = A/100(S – L/2) = h + S tan x°

T2

s
B

Headlight beam
0
X
A
h
I C D
T1

Figure 8.42

Figure 8.43…………………………………..

equating L = 2S – (200/l + 200S tan x°)/A (8.48)

when x° = 1° and h = 0.6m

L = 2S – (120 + 3.5S)/A (8.49)

8.10.4 Passing a curve through a point of known level

In order to ensure a specific clearance at a specific point along the curve it may be
necessary to pass the curve through a point of known level. For example , if a bridge
parapet or road furniture were likely to intrude into the envelop of visibility, it would be
necessary to design the curve to prevent this.
This technique will be illustrated by the following example. A downgrade of 4% meets a
rising grade of 5% in a sag curve. At the start of the curve, the level is 123.06m at chainage
3420m, whilst at chainage 3620m there is an overpass with an underside level of 127.06m.
If the designed curve is to afford a clearance of 5m at this point, calculate the required
length (Figure 8.44).

To find the offset distance CE:

From chainage horizontal distance T1E = 200m at -4%

Level at E = 123.06 – 8 = 115.06m

Level at C = 127.06 – 5 = 122.06m
Offset CE = 7m

Figure 8.44……………………………..

From offset equation CE/T2B = (T1E)2/(T1B)2

but T2B =the vertical divergence = AL/100.2, where A= 9

CE = AL2002/200L2 = 1800/L
L = 257m

8.10.5 To find the chainage of highest or lowest point on the curve

The position and level of the highest or lowest point on the curve is frequently required for
drainage design.
With reference to Figure 8.44, if one considers the as a series of straight lines, then at T1
the grade of the line is -4% gradually changing throughout the length of the curve until at
T2 it is +5%. There has thus been a change of grade of 9% in distance L. At the lowest
point the grade will be horizontal, having just passed through -4% from T1. Therefore, the
chainage of the lowest point from the start of curve is, by simple proportion,

D = L/9% × 4% = L/A × g1 (8.50)

which is the previous example is 257/9% × 4% = 114.24m from T1.

Knowing the chainage, the offset and the curve level at the point may be found.
This simple approach suffices as the rate of change of of grade is constant for a parabola,
i.e. y = Cl2, d2y/dl2 = 2C.

8.10.6 Vertical curve radius

Due to the very shallow gradients involved in VC design, the parabola may be
approximated to a circular curve. In this way vertical curves may be expediently draw on
longitudinal sections using railway curves of a given radius, and vertical accelerations
(V2R) easily assessed.
In circular curves (Section 8.1) the main chord from T1 to T2 = 2Rsin ∆/2, where ∆ is the
deflection angle of the two straights. In vertical curves, the main chord may be
approximated to the length (L) of the VC and the angle ∆, i.e.

∆ = A%
sin ∆/2 = ∆/2rads = A/200
L = 2RA/200 = AR/100 (8.51)

and as K = LlA = R/100, then:

R = 100L/A = 100K (8.52)

It is important to note that the reduced levels VC must always be computed. Scaling levels
from a longitudinal section, usually having vertical scale different from the horizontal, will
produce a curve that is neither parabolic not circular. The use of the railway curves is
simply to indicate the position and extent of the curve on the section.
Thus, having obtained the radius R of the VC, it is required to know the number of the
railway curve necessary to draw it on the longitudinal section.
If the horizontal scale of the section is 1 in H and the vertical scale is 1 in V, then the
number of the railway curve required to draw the VC is:

Number of railway curve in mm = R.V/ll2, with R in mm

e.g. If horizontal scale is 1 in 500, vertical scale 1 in 100 and curve radius is 200m,
the number of railway curve in mm = 200000 × 100/5002 = 80

8.10.7 Vertical curve computation

The computation of a vertical curve will now be demonstrated using an example.

A ‘2nd difference’ (δ2y/δl2) arithmetic check on the offset computation should automatically
be applied. The check works on the principle that the change of grade of a parabola (y –
C.l2) is constant, i.e. δ2y/δl2 = 2C. Thus, if the first and last chords are sub-chords of
lengths different from the remaining standard chords, then the change of grade will be
constant only for the equal-length chords.
For example, a 100m curve is to connect a downgrade of
0.75% to an upgrade of 0.25%. If the level of the intersection
point of the two grades is 150m, calculate:

(1) Curve levels at 20-m intervals, showing the second

difference (d2y/dl2); check on the computations.
(2) The position and level of the lowest point on the
curve.
Method

(a) Find the value of the central offset Y.

(b) Calculate offsets
(c) Calculate levels along the gradients
(d) Add/subtract (b) from (c) to get curve levels.

(a) Referring to Figure 8.38:

Grade angle A = (-0.75 – 0.25) = -1% (this is seen automatically).

L/2 = 50m, thus as the grades lT2, and lj are diverging at the rate of 1% (1m per 100m)in
50m, then

T2j = 0.5m = 4Y and Y = 0.125m

The computation can be quickly worked mentally by the student. Putting the above
thinking into equation form gives

4Y = A/100 . L/2 .. Y = AL/800 (8.53)

(b) Offsets from equation (8.37):

There are two methods of approach.

(1) The offsets may be calculated from one gradient throughout; i.e. y1,y2, EK , GM,
T2J, from the grade T1J.
(2) Calculate the offsets from one grade, say T1l, the offsets being equal on the other
side from the other grade lT2.

Method (1) is preferred due to the smaller risk of error when calculating curve levels at a
constant interval and grade down T1J.

From equation (8.37): y1 = Y × l12/(L/2)2

1st diff. 2nd diff.
T1 = 0m
0.02
2 2
y1 = 0.125 × 20 /50 = 0.02m − 0.04
0.06
y2 = 0.125 × 402/502 = 0.08m − 0.04
− 0.10
y3 = 0.125 × 602/502 = 0.18m −− 0.04
− 0.14
y4 = 0.125 × 802/502 = 0.32m − 0.04
−− 0.18
y2 = T2J = 4Y = 0,50m
The 2nd difference arithmetical check, which works only for equal chords, should be
applied before any further computation.
(c) First find level at T1 from known level at l:

Distance from l to T1 = 50m, grade = 0.75% (0.75m per 100m)

Rise in level from l to T1 = 0.75/2 = 0.375m

Level at T1 = 150.000 + 0.375 = 150.375m

Levels are now calculated at 20-m intervals along T1J, the fall being 0.15m in 20m. Thus
the following Table may be made.

Chainage Gradient Offsets Curve Remarks

(m) levels levels

0 150.375 0 150.375 Start of curve T1

20 150.225 0.02 150.245
40 150.075 0.08 150.155
60 149.925 0.18 150.105
80 149.775 0.32 150.095
100 149.625 0.50 150.125 End of curve T2

Position of lowest point on curve = 100m/1% × 0.75% = 75m from T1

Offset at this point = y2 = 0.125 × 752/502 = 0.281m

Tangent level 75m from T1 = 150.375 – 0.563 = 149.812m
Curve level = 149.812 + 0.281 = 150.093m

8.10.8 Drawing-office practice

(1) Design
(a) Obtain grade angle (algebraic difference of the gradients) A.
(b) Extract the appropriate K-value from Design Table in TD 9/81
(c) Length (L) of vertical curve = KA.
(d) Compute offsets and levels in the usual way.

(2) Drawing

To select the correct railway curve for the vertical curve on a longitudinal section.

(a) Find equivalent radius (R) of vertical curve from R = 100.L/A = 100.K
(b) Number of railway curve in mm = R mm × V/H2.

If horizontal scale of section is say 1/500, then H = 500.

If vertical scale of section is say 1/200, then V =200.
If the railway curves used are still in inches, simply express R in inches.

8.10.8 Computer-aided drawing and design (CADD)

The majority of survey packages now available contain a road design module, whilst highly
specialized programs such BIPS (British Intergrated Program System) and MOSS
(modeling systems) are standard tools in many road design offices.
Basically all the systems work from a digital ground model (DGM), established by ground
survey methods or aerial photogrammetry. Thus not only is the road designed, but
earthwork volumes, setting out data and costs are generated. The formation of DGMs and
illustrations are shown in chapter 1 (Sections 1.5.4 and 1.5.5 and Figure 1.19).

2019