Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

Surveying II

Download as pdf or txt
Download as pdf or txt
You are on page 1of 92

Vtusolution.

in
SURVEYING-II(10CV44)

PART-A

UNIT-1

THEODOLITE SURVEY

1.1Theodolite and types

Theodolite is the most precise survey instrument used commonly by engineers for

in
measuring horizontal and vertical angles accurately

Theodolites are broadly classified into two as

n.
1.Transit

2.Non-transit
tio
1.Transit theodolite: A theodolite in which if the telescope can be revolved through a
complete resolute about its horizontal axis in the vertical plane is called as a transit theodolite.

2.Non transit theodolite: This kind of theodolites are plain or „Y‟theodolites,in which the
telescope cannot be transited.
u
Theodolites are also classified into two as
ol

1.Vernier theodolites

2.Micrometer theodolites, based on the system used to observe the reading.


us

1.Vernier theodolite: verniers are used to measure accurately the horizontal and vertical
angles.A 20” verinier theodolite is usually used.

2.Micrometer theodolite : An optical system or a micrometer is used to read the angles in this
case.The precision can be as high as 1”
vt

* Fundamental Axis and part of transit theodolite

DEPARTMENT OF CIVIL ENGINEERING, Page 1

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

in
n.
1.2 Parts of theodolite

1.Telescope: The telescope of the theodolite is mounted on a spindle known as “Trunnion


axis”. In most of the transit theodolite an internal focusing telescope is used. It consists of
tio
object glass, a diaphragm and an eye-piece. The main functions of the telescope is to provide
line of sight.

2.The vertical circle: The vertical circle is rigidly connected to the transverse axis of the
telescope and moves as the telescope is raised or depressed. It is graduated in degrees with
u
graduations at 20‟. Graduation in each quadrant is numbered from 0‟ to 90‟ in the opposite
directions from the two zeros placed at the horizontal diameter of the circle.
ol

3.The index frame or T-frame or Vernier frame: It consists of a vertical portion called
dipping arm and a horizontal portion called an index arm. The 2 verniers of the vertical circle
are fixed to the two ends of the index arm. The index arm can be rotated slightly for
us

adjustment purpose, with the help of clip screw.

4. The standard or A-Frame: Two standards resembling the letter A are mounted on the upper
plates. The trunnion axis of the telescope is supported on these. The T-Frame and the arm of
vertical circle clamp are also attached to A-Frame.
vt

5.Levelling head: It consists of 2 parts namely

a)Tribrach- It is the upper triangular plate which carries 3 levelling screws at the

three ends of the triangle.

b)Trivet or the lower plate (foot plate) used three grooves to accommodate the 3

DEPARTMENT OF CIVIL ENGINEERING, Page 2

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

levelling screws.

The leveling head has 3 main functions namely

1.To support the main part of the instrument

2.To attach the theodolite to the tripod

3.To provide a mean for leveling.

6.The two spindles : Inner spindle is conical and fits into the outer spindle which is hollow.

in
Inner spindle is also called upper axis and outer spindle is called lower axis.

7.The lower plate (scale plate): It carries the circular scale which is graduated from 0-360‟.It

n.
is attached to the outer spindle which turns in a bearing within the tribrach of the leveling
head.It is fixed using lower clamping screws lower tangent screws enable slow motion of the
outer spindle.
tio
8.Upper plate(vernier plate): It is attached to the inner axis and carries 2 verniers with
magnifiers at two extremities diametrically opposite.Upper damping screw and a
corresponding tangent screw are used for moving upper plate.

9.The plate levels : The upper plate carries one or 2 plate levels which can be centred with
u
the help of foot screws.

10.Accessories:
ol

a)Tripod : with 3solid legs

b)Plumb bob : for centering


us

c)Compass : tubular or trough

d)Striding level : for yesting the horizontality of the transit axis or trunnion axis.

1.3Fundamental lines
vt

These are basically 2 planes and 5 lines in a theodolite .The planes are horizontal plane with
the horizontal circle and vernier; and vertical plane with vertical circle and vernier.

The fundamental lines are

1.Vertical axis

2.Horizontal axis

DEPARTMENT OF CIVIL ENGINEERING, Page 3

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

3.Line of collimation (line of sight)

4.Axis of plate level

5.Axis of altitude level

6.Axis of striding level,if provided

1.4 Definitions and Terms

in
1. centering: Setting the theodolite exactly over an instrument station so that its vertical axis
lies immediately above the station point is called centering

2. The vertical axis : It is the axis about which the instrument can be rotated in a horizontal

n.
plane.

3.The horizontal axis: It is the trunnion axis about which the telescope

4.Line of sight or line of collimation: It is the imaginary line passing through the intersection
tio
of the cross hairs (vertical and horizontal) and the optical center of the object glass and its
continuation

5.Axis of level tube : It is also called as bubble line,it is the straight tangential line to the
u
longitudinal curve of the level tube at its centre

6.Axis of the altitude level tube: It is the axis of the level tube in altitude spirit level
ol

7.Transiting: It is the process of turning the telescope vertical plane through 180‟ about the
trunnion axis. This process is also known as plunging or reversing.
us

8.Swinging the telescope: It is the process of turning the telescope in horizontal plane.If the
telescope is rotated in clock wise direction , it is known as right swing and other wise left
swing.

9.Face right observation: If the vertical circle is to the left of the observer, then the
vt

observation is ca;;ed as face left

10.Face right observation: If vertical circle is to the right of the observer,then the observation
called as face right.

10.Telescope normal and telescope inverted: If the telescope is in such a way that the face is
left and bubble is up,then it is said to be in normal position or direct.If the face is right and
bubble is down then the telescope is said to bein inverted position or reversed position.vertical
circle to the right of the observer,if originally to the left and vice versa.it is done by first

DEPARTMENT OF CIVIL ENGINEERING, Page 4

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

revolving the telescope through 180‟ in a vertical plane and then rotating it through 180‟ in
the horizontal plane,ie first transiting and then swinging the telescope.

1.5 Temporary adjustments of a transit theodolite.

The temporary adjustments of atransit theodolite is done by 3 important operations.

1. Setting up: The instrument have to be setted up properly on the station point.the tripod
stand should be approximately leveled before fixing the instrument.this is achieved with the

in
help of moving the legs of the tripod.there is a small spirit level on the tripod head for the
leveling of tripod.centering of the instrument over the station mark is achieved by a plumb
bob or by using optical plummet.

n.
2. Levelling up: After centering and approximate leveling ,accurate leveling is to be carried
out with the help of the foot screws and using the plate level tube.in this step the vertical axis
of the instrument is made truly vertical.Levelling the instrument depends on the number of
foot screws available.
tio
For a screw head,the procedure for leveling is as fallows:

a)Turn the upper plate until the longitudinal axis of the plate level is paralle to the line joining
any two foot screws(let it be A and B)
u
ol
us
vt

b)hold the 2 foot screws A and B between the thumb and the fore fingers of each hand and
turn them uniformly so that the thumb move either towards each other until the bubble is
central.Bubble moves in the direction of the left foot screw.

DEPARTMENT OF CIVIL ENGINEERING, Page 5

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

c)Turn the upper plate through 90‟ until the axis of the level passes over the position of the
third leveling screw C

d)Turn this leveling screw until the bubble is central

e)Return the upper plate to original position (fig1) and repeat step(b)

f)Turn back and repeat step (c)

g)Repeat steps (e) and (f) for 2-3 times until the bubble is central.

in
h)Now rotate the instrument through 180‟ and check whether the bubble is in the centre.

3. Ellimination Of Parallax: Parallax is a condition in which the image is formed will not lie

n.
on the plane of the cross hair,this can be eliminated by focusing the eye-piece and the
objective.

For focusing the eye-piece ,hold a white paper infront of the objective and move eye-piece in
tio
or out, until the cross-hairs are distinctly visible.objective lense focused by rotating the
focusing screw,until the image appears clear and sharp.

1.6Measurement Of Horizontal Angles


u
Theodolites are majorly used to measure horizontal and vertical angles.Horizontal angles are
usually ,easured by using any of these methods.
ol

1.Ordinary method

2.Method of repetition
us

3.Method of reiteration

1.Ordinary Method

FIG
vt

To measure an angle POQ,THE FOLLOWING PROCEDURE IS USED.

DEPARTMENT OF CIVIL ENGINEERING, Page 6

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

1.Set up the instrument at 0 , Set it up,level it accurately and perform the temporary
adjustments

2.Release the upper clamp screw and lower clamp screw.Turn the upper and lower plates such
that the vernier A reads „zero‟ (0) and the vernier circle is to the left of the observer.Clamp
both the plates and bring the vernier A to zero to coincide with the main scale zero using the
upper tangent screw.Check the reading on vernier A,it should read 180‟

3.Loosen the lower clamp and rotate the telescope to view point P.Clamp lower plate and

in
using lower slow motiomn screw sight P exactly.Check the readings on both th vernier to see
that it had not changed.

4.unclamp the upper clamp and rotate the instrument clock-wise until point Q is bisected

n.
tighten the clamp and using tangent screw bisect Q accurately.

5.Reading is observed from verners A and B .Reading of A vernier gives angle POQ and B
vernier gives 180‟+POQ
tio
Read degres,minutes and seconds from the vernier scale by observing which line on
the vernier scale is having correct coincidence with the reading in the main scale.

In a 20‟ transit theodolite ,the least count is 20” or the minimum reading which can be
u
measured from the scale is 20”.The reading coinciding with the vernier-zero is considered to
be the main scale reading .If there is no exact coincidence for the vernier zero line ,then the
ol

reading to the immediate left of the vernier scale,on the main scale should be considered.This
reading should be added with the vernier reading for the total value.

Reading onmain scale=128‟ 40‟


us

Reading on vernier scale=3‟ 00”

Therefore total reading =128‟40‟+3‟00”

=128‟43‟00”
vt

In B scale ,the degree reading is not required ,where as the minutes reading from the main
scale is noted and add with vernier reading and this will give the B scale reading.

6.Enter the readings in a field book of tabular format

DEPARTMENT OF CIVIL ENGINEERING, Page 7

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

1.7 Tabular Column

in
n.
7.Change the face by transiting and repeat the same process.

8.The mean of the 2 vernier reading gives the angle on face right
tio
10.Average horizontal angle is calculated from the mean horizontal angle of face left and face
right values.

Repetition Method
u
This method is used for very accurate work.In this method,the same angle is added
several times mechanically and the total angle is divided by no of repetitions to obtain the
ol

correct value of angle.there are 2 methods by which this method can be conducted

To measure an angle POQ by the method of repetition,the following procedure is adopted


us

1.Obtain the first reading of the angle following the procedure outlined in the previous
method.Read and record the value.

2.Loosen lower clamp,and turn the the telescope clockwise to sight P again and bisect
properly using lower tangent scew.check the vernier and see that the readings are not
vt

changed.

3.Unclamp the upper clamp and turn the instrument clockwise and sight Q again

4.Repeat the process for 3 times

5.consider the average horizontal angle for face left by dividing the final reading by three

6.change face and make 3 more repetitions find the average angle.

DEPARTMENT OF CIVIL ENGINEERING, Page 8

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

7. Total average angle is obtained by adding up the results of 2 faces and then dividing by 2

For high precision surveys, repetition method can be conducted in two ways

a)the angle is measured respectively for six times, keeping the telescope normal (face
left) and then calculating the average.

b)In another way ,angle is measured clockwise by first 3 with clockwise with face left
and last 3 with telescope inverted.Then in anticlockwise also 3 face left and face right
observations are taken.

in
1.8 Elimination of errors by method of repetition

The following errors are eliminated by adopting method of repetition

n.
a)Errors due to eccentricity of verniers and centres by measuring both vernier readings.

b)Errors due to line of collimation not being perpendicular to the horizontal axis of the
tio
telescope.

c)Errors due to horizontal axis of telescope not being perpendicular to the vertical axis.

d)Erroe due to the line of collimation not coinciding with the axis of the telescope
u
These 3 errors can be eliminated by changing their face of the theodolite.

e)Errors due to inaccurate graduations this can be eliminated by taking 2 vernier readings
ol

f)Error due to inaccurate bisection of the object this eliminated by taking repeated readings.

Reiteration Method
us

This method is also known as direction method or method of series several angles are
measured successivelu and finally the horizon is closed.

To measure a series of angles AOB,BOC,COD etc by reiteration,this procedure is fallowed


vt

DEPARTMENT OF CIVIL ENGINEERING, Page 9

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

1.Set the instrument at O, level it and centre it.

2.Measure the angle AOB in the same way as already explained.

3.similarly bisect the successive ranging rods C,D etc and keep onobserving the
readings.Each included angle is obtained by taking the difference of 2 consecutive readings.

Angle BOC=angleAOC – angle AOB

4.Finally close the horizon by sighting A.The reading in the vernier should be zero (360).If

in
not ,note down the reading and distribute it evenly to all angles.

Repeat the same steps in other face

n.
The sets of reading are usually taken first in clockwise direction and then after changing the
face in anticlockwise direction.

1.9 MEASUREMENT OF VERTICAL ANGLES tio


A vertical angle is an angle between the included line of sight and horizontal.the instrument
has to be leveled with respect to the altitude bubble for measuring vertical angles

1.Level the instrument with reference to plate level


u
2.keep the altitude bubble tube parallel to 2 foot screws and bring the bubble central.rotate
telescope 90‟ and adjust the bubble using the 3rd foot screw.repeat the procedure till the
ol

bubble is central.

3.loose the vertical clamp screw,rotate the telescope in vertical plane .to sight the object use
tangent screw for correct bisections.
us

4.read vernier C and D.mean gives correct vertical angle.

5.change the face and continue the procedure .

If the vertical angle is measured above the horizontal line,it is called angle of elevation or in
vt

other case as angle of depression.

1.10 Uses Of Theodolite

Theodolite is not only used for measuring horizontal angles and vertical angles.but it is also
used for the following:

1.To measure a magnetic bearing of a line

DEPARTMENT OF CIVIL ENGINEERING, Page 10

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

2.To measure direct angles

3.To measure deflection angles

4.To prolong a straight line

5.To run a straight line between 2 points

6.To locate the intersection points of 2 straight line

7.To lay off a horizontal angle etc.

in
1.11 PROLONGING A STRAIGHT LINE

1.When The Instrument Is In Adjustment

n.
I. Method A: Set the instrument at A and sight B accurately.Establish point C in the line
of sight shift the instrument to B, sight C and establish point D.The process is
continued till the last point.
II.
tio
Method B :Set the instrument at B and take a back sight on A.Clamp all the screws
and then plunge the telescope,if the instrument is in good adjustment point C will be
established.Similarly shift the instrument to C,back sight B ,plunge the telescope and
establish D,continue the procedure till the end .
u
2. When instrument is in poor adjustment (not in adjustment)
ol

If the instrumrnt is not in adjustment ,then instead of B,C,D some other points
B‟,C‟,D‟ etc will be established.

In such a case,set the instrument at B,take a back sight to A.plunge the telescope and
us

establish point C1,change the face and take back sight on A.Plunge the telescope to establish
C2 at the same distance. „C‟ wil be in midway between C1 andC2.shift the instrument to „c‟
and repeat the process. The process is repeated till the end point.This method is also called as
Double sighting.
vt

DEPARTMENT OF CIVIL ENGINEERING, Page 11

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

UNIT – 2

PERMANENT ADJUSTMENT OF DUMPY LEVEL AND


TRANSIT THEODOLITE

Permanent adjustments of a level are made to establish the fixed relationship


between its fundaments lines. The fundamental lines of a dumpy level are

in
1) The line of collection

2) Axis of the bubble tube

n.
3) Vertical axis

The desired relationship between the fundamental lines are

1) Axis of the bubble tube should be perpendicular to the vertical axis.


tio
2) Horizontal cross-hair should lie in a plans perpendicular to the vertical
axis.

3) Line of collination of the telescope should be parallel to the axis of


u
the bubble tube.
ol

2,1Principle of Reversal

This principle states that if any errors exist in a certain part of the
instrument, it is doubled on reversing the position of that part. The apparent
us

error on reversal becomes twice the actual error.


vt

DEPARTMENT OF CIVIL ENGINEERING, Page 12

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

If AD is perpendicular to AC. Let AB make an angle „α‟ with AD. If


the triangle ABC is revolved about A thorough 180°, such that B and C
occupy B‟ and C‟ . The angle between the original position of the line
AB and new position on reversal AB‟ is 2α. This principle is used in
testing of levels and theodolites.

Adjustment # 1

Desired relation :- The axis of the bubble tube should be perpendicular to the

in
vertical axis when the bubble is central.

Object :- To make the vertical axis truly vertical to that the bubble remains

n.
central after doing temporary adjustments.

Test:-
tio
(i) Set up the level on firm ground and level it carefully in the usual way. The
bubble will now be central in 2 positions at right angles to each other, one
bring parallel to pair of foot screws and the other over the 3rd foot screw

(ii) Bring the telescope over third foot screw and turn it to 180°. If
u
the bubble remains central, the adjustments is correct.
ol

Adjustments: - Means of adjustment are the capstan headed suits at one


end of the tube, connecting the bubble tube to the telescope.

i) If the bubble does not remain in the central positions, note down the
us

deviation (Say „2r‟ divisions). Bring the bubble half way back (n
divisions) by raising or lowering the end of the bubble tube by means
of capstan headed nuts and remaining half with the foot screw beneath
the telescope.
vt

ii) Turn the telescope through 90° so that it lies parallel to the other
foot screws and bring the bubble to the centre of its run by means of
these screws.

iii) Rotate the telescope and see if the bubble traverse. If not, repeat the
whole process until the adjustments is correct.

DEPARTMENT OF CIVIL ENGINEERING, Page 13

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

When the level tube is brought to the centre of its run when the
vertical axis is not truly vertical, the line looks like this

in
n.
When the instrument is now rotated through 180° „cd‟ remains as it is but the
new position of „ab‟ becomes a‟ b‟ as the angle 90°- e remains fixed and „ac‟
tio
becomes cb‟. Hence the error from the horizontal is 2e which is double
the error from the level tube axis and the vertical axis. Hence half the
errors is adjusted by the capstan headed screws.
u
Adjustment # 2

Desired relation :- The horizontal cross hair should lie in a perpendicular to the
ol

vertical axis.

Object :- To make the horizontal cross hair truly horizontal when the instrument
us

is levelled and the bubble is at the centre.

Test :- (i) Sight a well-defined object A (object 60m away) at one of the
horizontal hair.
vt

(ii) Rotate the end level slowly about its spindle until the point A is traced
from one end of the hair to the other hair.

(iii) If the point does not deviate from the hair, the adjustments is
correct.

DEPARTMENT OF CIVIL ENGINEERING, Page 14

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

Adjustments: - Loose the capstan screws of the diaphragm and turn it


slightly until by further trial the point appears to travel along the
horizontal hair.

Adjustments # 3

Desired relation: - The line of collimation of the telescope should be


parallel to the axis of the bubble tube.

in
Object: - To make the line of sight parallel to bubble axis. So that line
of sight is truly horizontal when the bubble is central. This adjustments
is very necessary.

n.
Test:-The test is called as „Two – Peg Test‟.

i) Drive 2 pegs A and B at a known distance „D‟ (Say 60-100m apart)


tio
on a fairly level ground and drive a 3 rd peg at „0‟, exactly midway
between A and B.

ii) Set up the level at 0 and level it accurately. Take staff readings on A
and B, Let the readings be „a‟ and „b‟ respectively. The bubble must
u
traverse while taking the readings.
ol

iii) Shift the level and set it up at „0‟ which is at „d‟ m away from A
(or B) and on the line BA produced. The level can also be placed
between A and B. Take the readings on A and B as a1 and b1 ,
us

with bubble central.

iv) Find difference between „a‟ and „b‟ and between a₁ and b₁. The first
one is a true difference when as second one is apparent level
difference. If both difference are same, their line of collimation is in
vt

adjustment.

True difference = a – b

Apparent difference = a₁ – b₁

DEPARTMENT OF CIVIL ENGINEERING, Page 15

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

in
n.
Adjustment:-

the true difference


tio
i) Find the true difference is a rise or a fall.

ii) Add to the reading on the peg A, near the


instrument (a₁), if it is a fall or dednet it from a₁ if it is a rise. The
will give the reading on the far peg (b₁) in level with a₁. Let the
u
reading be e₁;
ol

therefore; e₁= a₁ ± true difference [+ = fall]

[- = rise]
us

iii) Compare the readings e₁ and b₁; if b₁ is greater than e₁, the line of
collimation is included upwards, or vice versa.

iv) Find the corrections to be applied to the readings on both pegs,


using the fall formula.
vt

Correction to the reading on the far peg

= c₁ = (D+d)/D (b₁-e₁)

Corrections to the reading on the near peg

= c2 = d/D (b₁ – e₁)

DEPARTMENT OF CIVIL ENGINEERING, Page 16

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

Correct reading on the peg, B = b₁ ±c₁

Correct reading on the peg, B = a₁ ±c₂

v) Adjustments is made on far peg and checked on the near peg.

Look through the telescope and observe if the correct reading on the
correct reading on the far peg is seen above or below the horizontal
hair. Bring the cross-hair to the correct reading on the far peg by

in
moving the diaphragm by means of the diaphragm screws, loosening one
and tightening the other. If the correct reading is greater than the
observed one, cross-hair has to be lowered and vice versa.

n.
vi) Check the adjustments by reading the staff on the near peg A. The
observed reading should now agree with the calculated correct reading on
peg A. Repeat the adjustment until perfect.
tio
Test: - (Same as Reciprocal levelling)

i) Fix 2 points A and B on a fairly level ground at a distance of


100m. Set the instrument very close to A. Let it be „0‟.
u
ii) With staff kept at A, take the reading throughly the objective. The
ol

cross-hair will not be visible but the reading can be measured by


keeping a pencil point on the staff. This reading is called true rod
reading. Take staff reading at B.
us

The difference is apparent difference,

Let it be equal to h′ = a₁ - b₁

iii) Move the instrument close to B, let it be 0, and take readings to A


vt

and B.

New apparent difference, h′′ = a₂ - b₂

iv) If h′ = h′′, then the instruments is in adjustments.

DEPARTMENT OF CIVIL ENGINEERING, Page 17

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

Adjustments

i) Calculate the true level difference in elevation, as in reciprocal


levelling,

H = (a₁ - b₁) + (a₂ - b₂)

If H is positive, then B is higher than A and ricevers.

in
ii) Calculate the correct staff readings at B

n.
= a₁ ± H [+ = for fall]

[- = for rise]

iii) Keep the staff at A and sight at through the instrument set up at
D.
tio
Loose the capstan screws of diaphragm and raise or lower diaphragm
so as to get the same staff reading calculated.

Permanent Adjustments Of Theodolite Fundamentals Lines Of A


u
Theodolite And The Desired Relates

The fundamentals lines of a theodolite are


ol

i) Vertical axis

ii) Horizontal axis


us

iii) Line of collimation or Line of sight

iv) Axis of plate level


vt

v) Axis of altitude level

The desired relation between the fundamental lines are

a) The axis of the plate level must be in a plane perpendicular to the


vertical axis.

[Aim:- Vertical axis will be truly vertical when the bubble is at the
centre of its run]

DEPARTMENT OF CIVIL ENGINEERING, Page 18

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

b) The line of collimation must be perpendicular to horizontal axis at its


intersection with the vertical axis. Also, if the telescope is external
focussing type, the optical axis, the axis of the objective slide and the
line of collimation must coincide

[Aim:- Line of sight will generate a vertical plane when the telescope is
rotated about the horizontal axis]

c) The horizontal axis must be perpendicular to the vertical axis.

in
[Aim:- The line of sight will generate a vertical plane when the
telescope is plunged]

n.
d) The axis of the altitude level must be parallel to the line collimation.

[Aim:- vertical angle will be


tio free from index error due to lack of
parallelism]

e) The vertical circle vernier must read zero when the line of collimation
is horizontal.
u
[Aim:- vertical angle will be free from index error due to displacements
of the vernier.]
ol

Permanent Adjustments

The permanent adjustments of a theodolite are as follows:-


us

Adjustments # 1 :- Adjustments of Plate Level

Desired Relation:- The axis of the plate bubble should be perpendicular


to the vertical axis when the bubble is central.
vt

Object:- To make the vertical axis truly vertical. Once the relation is
maintained, the horizontal circle and the horizontal axis of the telescope
will be truly horizontal.

Test :-

i) Set up the instrument on a firm ground level the instrument using


temporary adjustments.

DEPARTMENT OF CIVIL ENGINEERING, Page 19

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

ii) Wen the telescope is on the third foot screws, swing it telescope is
on third foot screw, swing it through 180°. If the bubble is at centre,
then adjustments is correct.

Adjustment

i) Level the instruments with repeat to altitude bubble till it remains


central in two positions at right angles to each other.

in
ii) Swing the telescope through 180°. If the bubble moves from its centre,
bring it back halfway with the levelling screws and half with clip
screws.

n.
iii) Repeat till the altitude bubble remains central in all positions. The
vertical axis is now truly vertical.

iv) Centralize
screw.
plate level
tio
of the horizontal plate with capstan headed

Adjustments # 2 :- Adjustment of line of sight


u
Desired Relations :- The line of sight should coincide with the optical
axis of the telescopes.
ol

Object :- The objects is to place the intersection of the cross – hair in the
optical axis. Thus, both horizontal as well as vertical hair are to be
adjusted.
us

Test:-

i) Set up the instrument on a firm ground level the instrument using


temporary adjustments.
vt

ii) When the telescope is on the third foot screw, swing it through
180°. If the bubble is at the centre, then adjustments is correct.

Adjustments :-

i) Level the instruments with respect to altitude bubble till it remains


central in two positions at right angles to each other.

DEPARTMENT OF CIVIL ENGINEERING, Page 20

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

ii) Swing the telescope through 180°. If the bubble moves from its
centre, bring it back halfway with the levelling screw and half with cap
screw.

iii) Repeat till the altitude bubbles remains central in all positions. The
vertical axis is now truly vertical.

iv) Centralize plate level of the horizontal plate with capstan headed
screw.

in
Adjustment # 2 :- Adjustments of Line of Sight

Desired Relations :- The line of sight should coincide with optical axis of

n.
the telescope.

Object :- The object is to place the intersection of the cross-hair in the


tio
optical axis. Thus, both horizontal as well as vertical hair are to be
adjusted.

The adjustments of horizontal hair is important only in case of external


focussing telescope.
u
Vertical hair adjusted will ensure the line of collimations to be
ol

perpendicular to the horizontal aixs.

Test :- Level the instruments carefully, suspend a plumb bob at same


distance and sight it through the telescope . If the image of plumb bob
us

string is parallel to vertical hair, the vertical cross-hair is truly vertical.


The horizontal hair will then be horizontal.

Adjustment of horizontal hair :-


vt

Test:-

i) Level the instruments carefully with all clamps fixed.

ii) Take a staff reading placed at some distance a part. Note the reading
on the reading on the vertical circle also.

DEPARTMENT OF CIVIL ENGINEERING, Page 21

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

iii) unclamp the lower clamp, transit the telescope and swing it through
180°. Set the same reading. If the staff reading is same, the horizontal
hair is in adjustment

Adjustment:-

Use the top and bottom capstan screws of the diaphragm until the staff
reading is the mean of the two readings. Repeat the test and check.

in
Adjustment of Vertical hair:-

Test:-

n.
i) Set the instrument on a level ground, so that a length of about 100m
is available on either side of the instrument. Level it.

ii) Sight a point A about 100m away. Clamp the horizontal movements.
tio
iii) Transit the telescope and establish a point B to the other side at the
same level as A, such that OA = OB.
u
ol
us
vt

iv) Unclamp the horizontal movements and turn the telescope to sight A.

v) Transit the telescope. If it intersects B, the line of sight is


perpendicular to the horizontal axis.

DEPARTMENT OF CIVIL ENGINEERING, Page 22

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

Adjustment:-

i) Mark point C in the line of sight at same distance as that of B.

ii) Join C and B and establish D point such that CD = ¼ CB.

iii) Use side capstan screws of diaphragm, using the vertical hair to the
image of D.

iv) Change the face and repeat the procedure, til error free.

in
[Principle involved is double applications of principle of reversal].

Adjustment # 3 :- Adjustments of Horizontal Axis

n.
Desired Relations:- Horizontal axis should be perpendicular to the vertical
axis. tio
Object:- To make the horizontal axis truly horizontal when the instrument
is levelled. This adjustment ensure that line of sight revolves in a
vertical plane.
u
Test (Spire Test):-

i) Set up the instruments near a high building level it.


ol

ii) Sight a well defined point „A‟ on the top of the building or spire.
Clamp the horizontal screws.
us

iii) Depress the telescope and sight a point B on the ground as close to
the instruments as possible.

iv) Change the face and sight B. Clamp the horizontal plates.
vt

v) Raise the telescope and sight A. If it is sighted then the instruments


is in adjustment.

DEPARTMENT OF CIVIL ENGINEERING, Page 23

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

in
Adjustment

n.
Adjustment # 4 :- Adjustment of Altitude Level and Vertical Index Frame

a) Clip and tangent screw on separate arms


tio
b) Clip and tangent screw on same arm.

a) Clip and tangent screws on separate arms

Most of the modern day theodolite have clip and tangent screws on
u
separate arms and have altitude level on the index arm.

Desired Relations :- To make the line of sight horizontal when the bubble
ol

is central and vertical circle reading is zero.

Object :- Vertical circle reading will not be zero otherwise then the line
of sight is horizontal. This error is know as index error.
us

Test :-

i) Level the instrument with respect to the plate level, set the vertical
circle to read zero using vertical clamp screw and tangent screw.
vt

ii) Bring altitude bubble to its centre using clip screws.

iii) Observe a staff reading held at 75-100m away.

iv) Release the vertical circle clamp, transit the telescope and swing by
180°. Relevel the bubble by clip screw, if necessary. Set vertical circle to
zero again.

DEPARTMENT OF CIVIL ENGINEERING, Page 24

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

v) Read the staff held on the same point. If the reading is unchanged ,
adjustment is correct.

Adjustment:-

i) Bring the line of collimation on to the means reading by turning the


vertical circle tangent screw.

ii) Return the vernier index to zero by means of clip screw.

in
iii) Bring the altitude bubble to centre using capstan screw.

If the altitude bubble is on the telescope, then test for adjustments is

n.
the same.

Adjustments is proceeded by bring the bubble to the centre using


adjusting screw attached to telescope instead of capstan screws.
tio
If the clip and tangent screws are on the same arm then . The test is
done by 2- peg method used in dumpy level. Adjustment is then done to
the vertical index frame.
u
Numericals from Unit – 2
ol

1) A dumpy level was set up at C exactly midway between two pegs


A and B 100m apart. The readings on the staff when held on the pegs
A and B were 2.250 and 2.025 respectively. The respective staff reading
us

on A and B were 1.875 and 1.670. Calculate the staff readings on A and
B to give a horizontal line of sight.

Solution :-
vt

True level difference = hɑ -hb = 2.250-2.025

(when the instrument at C) H = 0.225m

Apparent level difference = H′ = ha′ - hb′

( when instrument at D) = 1.875-1.670 = 0.205m

DEPARTMENT OF CIVIL ENGINEERING, Page 25

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

H ≠ H′ , therefore; Line of sight in the net distance AB will be tan α = (H


– H′)/AB = (0.225 – 0.205)/ 100 = 0.020/100

H′ < H , line of sight is inclined upwards

. : Correct staff reading at A = 1.875 – AD tanα

= 1.875 – 20 * (0.02/100) = 1.871m

Correct staff reading at B = 1.670 – DB tanα

in
= 1.670 – (120 * 0.020)/100 = 1.646m

Check :- True level diff = 1.871 – 1.646 = 0.225m

n.
AN/AD = tan α

AN = AD tan α
tio
Correct staff reading at A = ha′ - MN

Similarly correct staff reading at B = hb′ - PQ .


u
2) The following observations were made during the testing of a dumpy
ol

level

Instrument at Staff reading on


us

A B

A a₁=1.702 b₁=2.244

B a₂=2.146 b₂=3.044
vt

Distance AB = 150m

Is the instrument in adjustment ? To what reading should ∟oc adjusted


when instrument was at B?

If RL of A = 432.052m , what should be the RL of B

Solution :-

DEPARTMENT OF CIVIL ENGINEERING, Page 26

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

Level difference between A and B, when instrument was at A = 2.244 –


1.702 = 0.542m (Rise)

Level difference when instrument was at B,

= 3.044 – 2.146 = 0.898m (Rise)

True level diff = (a₂ - b₂) + (a₁ - b₁)

in
= (2.146 -3.044) – (1.702 – 2.244)

n.
T.L.D = -0.898m + -(0.542) = -0.720m (fall)

E = -0.178m (inclined downwards)


tio
Staff reading at A for collimation adjustment
u
a₂ + e = 2.146 + 0.178 = 2.324m

check
ol

T.L.D = 3.044 – 2.324 = 0.720

RL of A = 432.052m (given)
us

Therefore RL of B = RL of A - T.L.D = 432.052 – 0.720 = 431.332m.


vt

3) In a 2 peg test of a dumpy level, the following readings were


taken :

i) The instrument at C ( midway between A and B, 100m apart, staff


reading on A = 1.628m

staff reading on B = 1.320m

DEPARTMENT OF CIVIL ENGINEERING, Page 27

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

ii) Is the line of collimation included upwards or downwards and how


much ? with the instrument at ∆, what should be the staff reading on
B in order to place the line of collimation truly horizontal.

Solution :-

True level difference, H = ha – hb = 1.682 – 1.320 = 0.362m

Apparent level difference, H′ = ha′ - hb′ = 1.528 – 1.178 = 0.350m

in
H > H′ , Therefore line of sight is inclined upwards.

Amount of inclination = H - H′ = 0.362 – 0.350

n.
e = 0.012 in 100m

Correct staff reading at B for collimation to be truly horizontal.


tio
= 1.178 – e = 1.178 – 0.012 = 1.166m

Check TLD = 1.528 – 1.166 = 0.362m

4) To test the line of colliation of a dumpy level, the instrument was


u
set up exactly midway between 2 pegs A and B, 80m apart. The
readings on the staff held on A and B were 1.620 and 1.565 reply. The
ol

instrument was then moved and set up at B, then the staff readings
were 1.385 and 1.325. Is the LOC inclined upwards / downwards and by
how much ? with the instrument at B, What should be the staff reading
us

on A in order to place the LOC truly horizontal ?

Solution :- True level difference , H = ha – hb = 1.620 -1.565 = 0.055m

Apparent level difference , H′ = ha′ - hb′


vt

1.385 – 1.325 = 0.06

H′ > H, Therefore line of sight is inclined downwards.

R e, Inclination = 0.06 – 0.055 = 0.005m in 80m(downward)

With the instrument at B,

DEPARTMENT OF CIVIL ENGINEERING, Page 28

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

Correct staff reading on A = 1.385 ± 0.005 = 1.380m

Check T.L.D = 1.380 - 1.325 = 0.055m

5) In a 2 peg test following reading were taken

instrument at readings on

O A B

in
1.655 1.350

A 1.425 1.112 (A and B = 100m)

n.
Find staff reading on B when instrument is at A?

Solution :- True level difference, H = 1.655 – 1.350 = 0.305m


tio
Apparent level difference, H′ = 1.425 – 1.112 = 0.313m

H′ > H , Therefore Line of collimation inclined downwards

Amount of inclination, e = 0.313 – 0.305 = 0.008m (downward)


u
When instrument at A,

Staff reading on B = 1.112 + 0.008 = 1.12m


ol

Check T.L.D = 1.425 – 1.120 = 0.305m

6) In testing a dumpy level, reciprocal levels were taken,


us

Instrument at Reading on

A B
vt

A 1.825 1.470

B 1.460 1.135

Is LOC IS adjustments? What should be staff reading on A during


second set up of instrument?

Solution :- T.L.D = (1.460 – 1.135) + (1.825 – 1.470) = 0.325 + 0.355

DEPARTMENT OF CIVIL ENGINEERING, Page 29

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

2 2

= 0.340m (rise)

Collimation error, e = (a₂ - b₂) – (a₁ - b₁)

= (1.460 – 1.135) – (1.825 – 1.470) = (0.325 – 0.355)

in
2 2

= -0.015m

n.
Inclination of LOC is downwards staff reading at A for second set up
is = 1.460 + 0.015 = 1.475m

Check T.L.D = 1.475 – 1.135 = 0.340m


u tio
ol
us
vt

DEPARTMENT OF CIVIL ENGINEERING, Page 30

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

UNIT-3

TRIGNOMETRIC LEVELLING

Trignometric Leveling

It is the process of determining the differences of elevations of stations from


observed vertical angles and known distances,which are assumed to be either
horizontal or geodetic lengths at mean sea level.

in
Heights And Distances

Trignometrical leveling is an indirect method of leveling in which the difference

n.
in elevation are measured using vertical angles,which are measured with the help of a
transit theodolite.Distances are either measured or compacted using trignometrical
calculations.Thus this method is also called as a method of heights and
tio
distances.There are various cases in this they are

CASE 1

Base of the object is accessible


u
ol
us
vt

Consider an object Q,which is at distance of Dm from the instrument station P and is


accessible.

Let A- Centre of the instrument

DEPARTMENT OF CIVIL ENGINEERING, Page 31

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

Q‟- Projection of Q on horizontal plane through A

h‟-Height of instrument at P

h-Height QQ‟ from line of collimation to Q

S-Staff reading on the BM

Alpha-Vertical angle from line of sight to Q

in
Consider triangle A at Q

Tan alpha =h/d

n.
h=D tan α

RL of Q=RL of instrument axis (IA)+h

If RL of P is known ,then


tio
RL of Q=RL of P+S+D tan alpha
If reading on staff on bm is given then
 RL of Q =RL of BM+S+D tan alpha
u
This method is applicable only when D is small .if D IS Large then the
combined correction for curvature and refraction should be applied.
ol

Then
RL of Q=RL of BM+S+D tan alpha + or – Cc
Where Cc=0.0673D2 meter (D is in kilometer)
us

+ -for angle of elevation


- -for angle of depression

CASE 2
vt

Base of the object inaccessible – Instrument stations in the same vertical plane
as the elevated object

(SINGLE PLANE METHOD)

INSTRUMENT A NES AT SAME LEVEL

DEPARTMENT OF CIVIL ENGINEERING, Page 32

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

in
If the base of the object and the instrument station are intervisible but not inaccessible

n.
, two instrument stations are need so tha they are in the same vertical plane as the
elevated object.the procedure is as fallows
tio
1.Set up the theodolite at P,level the instrument and centre it with respect to
altitude bubble.

2.Observe Q and clamp the plates.read the vertical angle alpha 1


u
3.Transit the telescope so that LOS is reversed.mark another point R on
ground. Measure distance P and R
ol

4.Repeat the steps in face right

5.Take the staff reading S on the BM


us

6.Shift the instrument to R and do the temporary adjustments.measure the


vertical angle alpha 2 on both faces

CASE 3
vt

Base of the object inaccessible – instrument stations not in the same vertical plane

(DOUBLE PLANE METHOD)

DEPARTMENT OF CIVIL ENGINEERING, Page 33

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

in
n.
Let P and R be the 2 instrument stations in the different planes with the object Q
tio
1.Set the instrument at P.make the temporary adjustment and level respect to altitude
bubble measure alpha 1

2.Sight point R with reading on horizontal circle as zero and measure θ1


u
3.Take back sight on bm (s)
ol

4.Shift instrument to R ,measure vertical angle α 2 and horizontal angle θ2

TOTAL STATION
us

A total station is an optical instrument used as a primary equipments for modern


surveying.it is the combination of an electronic theodolite and an electronic distance
meter (edm).it also needs a software running on an external computer as a data
collectors
vt

SALIENT FEATURES OF TOTAL STATION

1.A total station consists of an electronic theodolite,an electronic distance meter


which makes possible to determine the coordinates of a reflector by aligning the
instruments cross hair on the reflector and simultaneously measuring the vertical
angle,horizontal angle,slopes etc.

DEPARTMENT OF CIVIL ENGINEERING, Page 34

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

2.Micro processor takes care of reading ,reading and computations

3.Key board with function keys for data input.

4.Removable data storage.large battery capacity and on board application programs in


one unit.

5.Exchanging data and configurations between instruments and computers or


transferring data didirect to data location

in
6.Horizontal and vertical distances,angles etc help calculating co ordinates of survey

7.Some total station are enabled with GRS interface

n.
8.Distance measurement is accompalished with a modulated microwave or infra red
carrier signal geerated by an emitter and bounced off of the object to be measured.
tio
9.The modulation pattern in the returning signal read and interpreted by the on-board
computer

10.porro prism is usually used as reflector


u
11.Modern total stations are robotic which can be operated by remote control
ol

Advantages Of Total Station Over Conventional Instruments

1.Most total stations can measure angle to a least count of 5”


us

2.It is very easy to level the total station than the conventional instruments

3.All the computations can also be made with a total station other than taking the
readings and recording it for further use.
vt

4.The microprocessor of the total station perform variety of tasks like averaging
multiple angle and average distance measurement,calculation of co-ordinate slope
corrections,remote object elevations,atmosphere and instrumental corrections

5.GPS facility of total station enhance the capacity for high precision measurements
and reduces the consequences of poor accuracy of line of sight observation

DEPARTMENT OF CIVIL ENGINEERING, Page 35

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

6.Distance measurement upto accuracy of 1/100th a foot can be obtained

7.User friendly in all aspects

8.While taraditional survey is a data gathering activity,total station survey,seting


out,data gathering,data storage,computations etc are all performed side by side

Aplication Of Total Station

in
1.It mainly measures vertical angle,horizontal angle,slope distances etc.

2.It can compute horizontal and vertical,and coordinates of actual positions of


surveyed points,from known points.

n.
3.Many mathematical operations like averaging multiple angle measurements distance
measurement,calculation of rectangular coordinates,slope corrections,distances
tio
between remote points,atmospheric and instrumental corrections etc.

4.It can store the data for a long term

5.The data can be transferred to the computer


u
6.The traverse closure can be calculated for tapographic data for closed traverse
ol

7.ploting can be conducted any desired scale with the data stored in coordinated files.
us
vt

DEPARTMENT OF CIVIL ENGINEERING, Page 36

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

UNIT 4

TACHEOMETRY

Basic principle

Tacheometry is a branch of angular surveying in which the horizontal and vertical


distances of points are obtained by Instrumental observations this method is rapid and

in
accurate.

The common principle in all tacheometric survey is that the horizontal distance b/n an
instrumental station and a point as well as the elevation point,relatively to the instrument can

n.
be determined from the angle subtended at the instrument by a known distance at point and
vertical angle from instrument to the point.

Uses of tacheometric survey tio


1.it is rapid in rough and difficult terrain where ordinary leveling is tedious,chaining is
inaccurate,diffcult and slow.

2.used when obstacles such as steep and broken ground,deep ravines and streches of water are
u
met with.

3.used to prepare contor maps requiring both the horizontal as well as vertical control.
ol

4.used in hydrographic survey,location surveys,road surveys, railway and reservoir surveys.

5.used for checking more precise instruments.


us

Types of tacheometric survey

There are 3 types-

1. Stadia method
vt

2. Tangential method

3. Measurment by means of special instruments.

Stadia method is further classified into two:

A. Fixed hair method b. Movable hair method

DEPARTMENT OF CIVIL ENGINEERING, Page 37

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

Stadia method:

Instruments employed in stadia method are

1.tacheometer:

It is a transit theodalite having a stadia telescope with 2 horizontal hairscalled stadia hairsin
addition to regular cross hairs.

2.stadia rod:

in
It is a rod with 5cm to 15cm width and 3 to 4 cm long. A leveling staff also can be used as a
stadia rod.

n.
Fixed hair method:

In this method stadia hair interval is fixed when a staff is sigthed through the telescope, a
certain length of staff(staff intercept) is intercepted by the stadia lines and from this values the
tio
distance from the instrument to the staff station may be determined.

Priciple of stadia method(tacheometric eqn for horizontal line of sight)

Let o be the optical centre of the object glass


u
A,b&c –the bottom, top and the central axis at diaphram

A,b&c –the points on the staff cut by the three lines


ol

Ab=i=interval b/n stadia lines

Ab=s=staff intercept
us

F=focal length of the object glass

U=the horizontal distance from the optical centre to the staff

V=the horizontal distance from the optical centre to the image of the staff.
vt

U&v are the conjugate focal distance

D=the horizontal distance from o to the vertical axis of the tacheometer.

D= the horizontal distance from o to the vertical axis of the instrument to the staff.

DEPARTMENT OF CIVIL ENGINEERING, Page 38

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

From similar triangles aob and aob

I/s=v/u

V=iu/s……1

From the formulae of lenses 1/f=1/u+1/v……..2

in
1/f=1/u+1/(iu/s)=1/u+s/iu

1/f=1/u+s/iu=(1+s)/iu

1/f=(i+s)/iu

n.
Iu=(i+s)*f

U=(i+s)*f/i=(i/i+s/i)*f
tio
U=( 1+s/i)*f=f+f(s/i)

But d=u+d

D=f+f(s/i)+d
u
D=(f/i)*s+(f+d)
ol

or

D=ks+c
us

This eqn is known as the distance eqn or the tacheometric eqn

The quantites (f/i)&(f+d) are the tacheometric constants


vt

(f/i)=k it is called as multiplying constant

(f+d)=c, adittion constant

The value of (f/i) or k actually 100.

DEPARTMENT OF CIVIL ENGINEERING, Page 39

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

Determeination of tacheometric constants(field measurments)

First method

 Sight any far object and focus it properly


 Measure the distance along the top of the telescope b/n the object glass and the plane
of the cross hairs with a rule.
 Measure the distance d
 Measure several lengths d1,d2…along ab from instrument position a and obtain the

in
staff intercept s1,s2,s3…….at each of the length
 Add f & d to find c=f+d
 Knowing c determine the several radius of f/i or k from eqn d=ks+c

n.
 Mean of the several values give the required values of the multiple constants(f/i)

Second method:


tio
Measure a line accurately oa about 300 long on a farely level ground and fix pegs
at the interval 30m
 Set up the instrument at o and obtain the staff in tercept by taking the stadia
u
readings on the staff held vertically on each of the pegs
 Substitute the values of d and s in eqn d=ks+c from the member of eqn formed by
ol

the substituting values d and s


D1=ks1+c d2=ks+c
us

D1*s2-d2*s1/(s2-s1)

Distance and elevation formulae when staff held vertical :

Tacheometric eqn for horizontal line of sight:


vt

DEPARTMENT OF CIVIL ENGINEERING, Page 40

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

D=ks+c, k=100&c=0

Then d=100*s

Rl of p=rl of bm +s1-h

Tacheometric eqn for inclined line of sight

in
n.
tio L=ks‟+c

Cos(alpha)=d/l

D=lcos(alpha)
u
Sin (alpha)=v/l

V=l sin(alpha)
ol

Let d&c are the 3 points on the staff cut by the upper middle and the lower cross hairs ,db is
stadia reading=s
us

From fig2 bc=cb=s/2

D=lcos(alpha)…….1
vt

V=lsin(alpha)………2

L=ks‟+c……3

Form fig4 cos (alpha)=s‟/s

or

s‟=scos(alpha)

DEPARTMENT OF CIVIL ENGINEERING, Page 41

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

For angle of elevation,

Rl of p=rl of bm +s1+v-h

For angle of depression

Rlof p=rlof bm+s1-v-h

H is the middle hair reading or actual hair reading

Tacheometric eqn for the line of sight inclined and the staff held normally in line of

in
sight:

n.
tio
In this case the line of sight is perpendicular to the staff
u
Axial hair reading h is inclined from triangle cfb
ol

cf=h cos(alpha)…..1

D=delta‟ *g+gh……….2

A‟g=lcos (alpha)……..3\
us

Cg=lsin(alpha)……..4

Fb=hsin(alpha)……5
vt

D=lsin(alpha)+hsin(alpha)……..6

Here l=ks+c

Tacheometric eqn for inclined line of sight:

D=(ks+c)cos(alpha)+hsin(alpha)

D=ks cos (alpha)+cos(alpha)+hsin(alpha)…….7

DEPARTMENT OF CIVIL ENGINEERING, Page 42

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

If c=0

D=kscos(alpha)+hsin(alpha)…..8

V=ks sin(alpha)+csin(alpha)…………9

If c=0

V=kssin(alpha)….10

For the angle of elevation

in
Rlof b=rlof bm +s1+v-hcos(alpha)…..11

For angle of depression

n.
Rlofb=rlof bm+s1-v-hcos(alpha)…….12

Moving hair method tio


In this the instruments used are a theodalite equipped with a diapharm which has
stadia hairs which can be moved by a separate sliding frame by micrometer screw with a large
graduated head.
u
Distance through which the stadia wires are moved is given by the sum of the
readings.eventhough stadia interval is variable,staff intercept remains constant.
ol

The horizontal distance, d is given by the formula,d=ks/n+(f+d)

Where, n=sum of micrometer readings.


us

Tangential method

When telescope is not fitted with stadia diapharm,this method is used. The horizontal and
vertical distances from the staff stations from the instruments may be computed from
observations taken to 2 vanes or targets on the staff at known distance (s) apart usually,
vt

1. When both are angles of elevation

DEPARTMENT OF CIVIL ENGINEERING, Page 43

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

in
D= s/(tan alpha2-tan alpha1)

V=dtan(alpha2)

n.
Rlof q=rlof bm+s1+v-h

2. When both are angles of depression


D= s/(tan alpha1-tan alpha2)
tio V=dtan(alpha2)

Rlof q=rlof bm+s1-(v+h)

Problems
u
1. Two distances of 20&100 are accurately measured and the intercepts on the staff b/n
ol

the enter stadia meter were 0.196m @ the former distance and 0.996 @ the lateral.
Calculate the tacheometric constants.

Solun:
us

D1=20m d2=100m

S=0.196m s2=0.996m

d=ks+c
vt

d1=ks1+c

d2=ks2+c

20=k*0.196+c…….1

-100=-k*0.996+c…… 2

-80=-0.8k

DEPARTMENT OF CIVIL ENGINEERING, Page 44

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

k=80/0.8 = 100.

k*0.196+c=20.

c=20-k*0.196 = 20-19.6 = 0.4

c=0.4

in
n.
u tio
ol
us
vt

DEPARTMENT OF CIVIL ENGINEERING, Page 45

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

in
n.
u tio
ol
us
vt

DEPARTMENT OF CIVIL ENGINEERING, Page 46

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

in
n.
u tio
ol
us
vt

DEPARTMENT OF CIVIL ENGINEERING, Page 47

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

in
n.
u tio
ol
us
vt

DEPARTMENT OF CIVIL ENGINEERING, Page 48

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

in
n.
u tio
ol
us
vt

DEPARTMENT OF CIVIL ENGINEERING, Page 49

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

in
n.
u tio
ol
us
vt

DEPARTMENT OF CIVIL ENGINEERING, Page 50

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

in
n.
u tio
ol
us
vt

DEPARTMENT OF CIVIL ENGINEERING, Page 51

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

in
n.
u tio
ol
us
vt

DEPARTMENT OF CIVIL ENGINEERING, Page 52

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

in
n.
u tio
ol
us
vt

DEPARTMENT OF CIVIL ENGINEERING, Page 53

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

in
n.
u tio
ol
us
vt

DEPARTMENT OF CIVIL ENGINEERING, Page 54

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

in
n.
u tio
ol
us
vt

DEPARTMENT OF CIVIL ENGINEERING, Page 55

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

in
n.
u tio
ol
us
vt

DEPARTMENT OF CIVIL ENGINEERING, Page 56

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

in
n.
u tio
ol
us
vt

DEPARTMENT OF CIVIL ENGINEERING, Page 57

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

in
n.
u tio
ol
us
vt

DEPARTMENT OF CIVIL ENGINEERING, Page 58

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

in
n.
u tio
ol
us
vt

DEPARTMENT OF CIVIL ENGINEERING, Page 59

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

in
n.
u tio
ol
us
vt

DEPARTMENT OF CIVIL ENGINEERING, Page 60

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

in
n.
u tio
ol
us
vt

DEPARTMENT OF CIVIL ENGINEERING, Page 61

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

in
n.
u tio
ol
us
vt

DEPARTMENT OF CIVIL ENGINEERING, Page 62

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

in
n.
u tio
ol
us
vt

DEPARTMENT OF CIVIL ENGINEERING, Page 63

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

in
n.
u tio
ol
us
vt

DEPARTMENT OF CIVIL ENGINEERING, Page 64

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

UNIT 6

CURVE SETTING (COMPOUND CURVE AND REVERSE


CURVES)

Compound curves

in
n.
u tio
ol
us

A compound curve consists of 2 or more circular curves arcs of


different radii with their centres of curvatures on the same side of the
vt

common tangent. Compound curves are required when space restrictions


preclude a single circular curve and there are some interversing
obstacles.

Elements of a Compound Curve

DEPARTMENT OF CIVIL ENGINEERING, Page 65

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

The figure shows a compound curve T₁, T₃T₂ having 2 simple curves, T₁
T₃ and T₃ T₂; meeting at a common point T₃ known as point of
compound curvatures (P.C.C)
Rs = Radius of smaller curve

RL = Radius of larger curve

∆s , ∆L = Deflection angles

in
Ls, Ll = length of curves

ts , tL = tangent lengths

n.
Ts , TL = tangent lengths on the sides of smaller and larger curves.

1) Tangent lengths for the circular curves


tio
ts = T₁M = MT₃ = Rs tan ∆s/2

tl = T₃N = NT₂ = RL tan ∆L/2

2) Total deflection angle


u
∆ = ∆s+∆L
ol

3) Tangent length of the compound curve

In ∆MBN ; we have ;
us

MN = MT₃ + T₃N = ts + tL

BM = MN Sin ∆L = (ts + tL) Sin ∆L

Sin ∆ Sin ∆
vt

[by sine formula BM . = MN . chainages]

Sin ∆L Sin (180 – (∆s +∆L)

Chainages of T₁ = chainages of B - T₃

Chainages of T₃ = chainages of T₁ + ls

DEPARTMENT OF CIVIL ENGINEERING, Page 66

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

Chainages of T₂ = chainages of T₃ + lL = chainages of T₁ + ls + lL

Relationship between elements of a Compound Curve

Case-1 :- Given :- ∆, Rs, RL and ∆s (or ∆L)

Required :- ∆L (or ∆s) , Ts and TL

[Note :- This is the most common case]

in
Solution

We have , ∆L = ∆ - ∆s

n.
Ts = (ts + tL) Sin ∆L + ts

Sin ∆
tio
TL = (ts + tL) Sin ∆s + tL

Sin∆

Ts = Rs tan ∆s
u
2
ol

Case – 2 :- Given :- ∆ , Rs , RL and Ts

Required :- ∆s , ∆L and TL
us

Solution :- The shorter curve T₁T₃ is prologged to a point E until the


central angle at Os is equal to ∆.

BM . = MN .
vt

Sin ∆L Sin(∆s + ∆L)

But ∆s + ∆L = ∆

BM . = MN .

Sin ∆L Sin ∆

But MN . Sin ∆L

DEPARTMENT OF CIVIL ENGINEERING, Page 67

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

Sin ∆

But MN = ts + tL .: BM = (ts + tL) Sin ∆L . . ①

Sin ∆

similarly, BN = MN

Sin ∆s Sin∆

in
and BN = MN . Sin ∆s

Sin ∆

n.
= (ts + tL) . Sin ∆s →②

Sin ∆

T₃ = T₁M +MB

= ts + (ts +tL) Sin ∆s →③


tio
Sin ∆
u
= TL = BN +NT₂
ol

= (ts + tL) Sin ∆s + tL →④

Sin ∆
us

4) Length of the compound curve l = ls + lL

we have ls = ∏ * Rs *∆s and lL = ∏ * RL * ∆L

180 180
vt

l = ∏ * Rs *∆s + ∏ * RL * ∆L = ∏ ( Rs ∆s + RL ∆L)

180 180 180

Chainage of T₁ = chainage of B = T₃

Chainage of T₃ = Chainage of T₁ + ls

DEPARTMENT OF CIVIL ENGINEERING, Page 68

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

Chainage of T₂ = Chainage of T₃ +lL

= Chainage of T₁ + ls +lL

Relationship between elements of a compound curve

Given :- ∆ , Rs , RL and ∆s (or ∆L)

Required :- ∆L (or ∆s) , Ts and TL

in
Solution :-

we have , ∆L = ∆ - ∆s

n.
Ts = (ts +tL) Sin ∆L + ts

Sin ∆

TL = (ts +tL) Sin ∆s + tL

Sin ∆
tio
ts = Rs tan ∆s and tL = RL tan ∆L
u
2 2
ol

Case 2

Given :- ∆ , Rs , RL and Ts
us

Required :- ∆s , ∆L and TL

solution :- The shorter curve T₁T₃ is prolonged to a point E until the


central angle at Os is equal to ∆.
vt

Two straight lines with a total deflection angle of 72°30′ are to be


connected by a compound curve of 2 branches of equal length. Radius
of first are is 350m and that of second arc is 500m. Change vertex is
1525m. Find chainage of P.C , P.T, P.C.C

∆ = 72°30′

DEPARTMENT OF CIVIL ENGINEERING, Page 69

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

ls = lL

Rs = 350m

RL = 500m

ls = ∏ Rs ∆s = ∏ RL ∆L

180 180

in
∏ * 350 * ∆s = ∏ * 500 * ∆L

180 180

n.
∆s = 500 . ∆L = 1.43 ∆L

350

∆ = ∆s + ∆L

= 1.43 ∆L + ∆L
tio
= 2.43 ∆L = 72°30′
u
∆L = 29°50′

∆s = 1.43 * 29°50′ = 42°40′


ol

ls = lL = T * 350* 42°40′ = 260.63m

180
us

ts = Rs tan ∆s = 350 tan 42°40′ = 136.69m

2 2
vt

tl = Rs tan ∆L = 500 tan 29°50′ = 133.19m

2 2

MN = 269.88m

269.88 . = BM . = BN .

Sin 107°30′ Sin 29°50′ Sin 42°40′

DEPARTMENT OF CIVIL ENGINEERING, Page 70

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

The following data refer to a right hand compound curve

1) Total deflection angle = 80°

2) Radius of first arc = 200m

3) Radius of the second arc = 250m

4) Chainage of the P.I = 1504.80m

in
5) Deflection angle of first arc =50°

Determine chainages of starting, P.C.C and P.I

n.
∆ = 80° Rs = 200m RL = 250m ∆s = 50° ∆L = 30°

ts = Rs tan ∆s = 200 . tan 50° = 93.26m

2 2

tL = RL tan ∆L = 250 . tan 30° = 66.99m


tio
2 2
u
ls = ∏ Rs ∆s = ∏ * 200 * 50 = 174.53m

180 180
ol

lL = ∏ * 250 * 30 = 130.89m

180
us

MD = 93.26m DN = 66.99m MN = 160.25m

160.25 = BM = BN
vt

Sin 100° Sin 30° Sin 50°

BM = 81.36m BN = 124.66m

Ts = 93.26 + 81.36 = 174.62m

TL = 66.99 + 124.66 = 191.65m

Chainage at A = 1504.80 – 174.62 = 1330.18m

DEPARTMENT OF CIVIL ENGINEERING, Page 71

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

Chainage at P.C.C = 1330.18 + 174.53 = 1504.71m

Chainage at P.T = 1504.71 + 130.89 = 1635.6m

BM = 140.78M BN = 191.73M

Ts = 136.69 + 140.78 = 277.47m

TL = 133.19 +191.73 = 324.92m

in
Chainage at P.I = 1525m

P.C = 1525 – 277.47 = 1247.53m

n.
P.C.C = 1247.53 + 260.63 = 1508.16m

P.T = 1508.16 + 260.63 = 1768.79m


tio
Two straights AB and BC are intersected by a line KM. ∟MKA = 140°
and ∟KMC = 145° . Radius of the first arc is 600m and second arc is
400m. Find the chainage of tangent points and P.C.C given that the
chainage at P.I is 3145 m
u
RL=600m Rs = 400m
ol

P.I = 3145m

Rs = ∏ * 400 * 35 = 244.35m
us

180

lL = ∏ * 600 * 40 = 418.89m

180
vt

ts = 400 tan 35 = 126.12m

2 BK = BM = MK

Sin 35° Sin 40° Sin 105°

BK = 161.34m BM = 108.81m

tL = 145.59m Ts = 234.93m

DEPARTMENT OF CIVIL ENGINEERING, Page 72

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

ts + tL = 271.71m TL = 306.93m

Chainage at P.C = 3415 – TL = 3415 - 306.93 = 3108.07m

Chainage at P.C.C = 3108.07 +lL = 3526.96m

Chainage at P.T = 3526.96 + ls = 3771.31m.

Reverse Curve (Serpentive Curve)

in
It is compound of 2 circular arcs curving in opposite directions with a
common tangent at their junctions. The point at which the two arc join
is called the point of reverse curvature or contrary flexure (P.C.C),

n.
Reverse curves are used when the straights are parallel or intersect at a
very small angle. They are used in railway sidings and some time on
roads. They should be avoided as far as possible on highways and main
tio
railway line where speeds are necessarily high for the following reasons
like.

a) They involve a sudden change of cant from one side to the other.
u
b) The curves cannot be properly super elevated at the point of reverse
curvature.
ol

c) The sudden change in direction is objectionate.

Reverse Curve with parallel tangents


us
vt

DEPARTMENT OF CIVIL ENGINEERING, Page 73

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

The important parameter of reverse curves are R, ∆, T₁, R₂, ∆₂, and
T₂.

Most of the times the radius of both the simple curves will be equal (
R₁ = R₂ ) or they have equal central angle (∆₁ = ∆₂).

The important formula for a reverse curve is ∆ = ± (∆₁ -∆₂).

in
Reverse Curve becomes Traverse Tangents ( When radius are not
equal and equal )

n.
u tio
ol

In a reverse curve between 2 parallel straights , ∆₁ = ∆₂


us

R₁ = Smaller radius

∆₁ = Central angle of first curve

L = T₁T₂ distance
vt

V = Perpendicular distance between 2 straights

h = Distance between the 2 straights T₁ and T₂

Through E , draw a line BD parallel to the 2 tangents.

Since O₁T₁ and O₂T₂ are parallel to each other,

DEPARTMENT OF CIVIL ENGINEERING, Page 74

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

we have ∆₁ - ∆₂

T₁B = O₁T₁ - O₁B

= R₁ - R₁ cos ∆₁ = R₁ ( 1 - cos∆₁ ) O₁B = R₁ cos ∆₁ = R₂ Versin ∆₁

T₂D = O₂T₂ - O₂D = R₂ cos ∆₂

similarly ;

in
Again became T₁E is the long chord of first arc of T₂E that of second

T₁E = 2R₁ Sin ∆₁ OD = cos ∆

n.
2 R₂

T₂E = 2R₁ Sin ∆₂ tio 2R₂ Sin ∆₁

2 2

T₁T₂ = L = T₁E + ET₂ = 2(R₁ + R₂) Sin ∆₁

2
u
But, Sin ∆₁ = V/L
ol

L = 2( R₁ + R₂) V/L
us

L = √2V R₁+R₂

BE =R₁ Sin ∆₁ , ED = R₂ Sin ∆₂ = R₂ Sin ∆₂


vt

BD =
h = (R₁ + R₂) Sin ∆₁

R₁ = R₂ = R; we have,

V = 2R (1- cos ∆₁)

L= 4 R Sin ∆₁

DEPARTMENT OF CIVIL ENGINEERING, Page 75

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

l= √4RV

h = 2R Sin ∆₁

Two parallel railway lines are to be corrected by a reverse curve, each


section having the same radius. If the lines are 12m apart and the max
allowable radius . If however, both the radii arc 2 be different , Calculate
the radius of the second branch, if that of first branch is 60m . Calculate

in
the length of both branches.

R₁ = R₂

n.
V = 12m

h = 48m

tan ∆₁ = V

2 h
tio
∆₁ = 14°2′
u
2

∆₁ = 28°4′
ol

V = 2R (1 – cos ∆₁)
us

R = V .

2(1 – cos 28°4′)


vt

= 12 . = 51.1m

2(1 – cos 28°4′)

b) R₁ = 60m

∆₁ = 28°4′

h = (R₁ + R₂) Sin ∆₁

DEPARTMENT OF CIVIL ENGINEERING, Page 76

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

48 = (60 + R₂) Sin 28°4′

(60 + R₂) = 48 .

Sin 28°4′

R₂ = 42m

Length of the first branch = ∏ R₁ ∆

in
180

= ∏ * 60 * 28°4′ = 29.39m

n.
180

Length of second branch = ∏ R₂ ∆₂ .


tio 180

= ∏ * 42 * 28°4′ = 20.57m

180
u
ol
us
vt

DEPARTMENT OF CIVIL ENGINEERING, Page 77

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

UNIT-7

CURVE SETTING (TRANSITION AND VERTICAL)

Transition curves

A transiton curve is a curve of varying radius introduced between a straight and a

in
circular curve or between 2 branches of a compound curve or reverse curve. The main
functions of a tranisition curve are

1. To accomplish gradually the transition from the tangent to the circular curve, so that

n.
the curvature is increased from zero to a specified value.
2. To provide a medicine for the gradual introduction or change of the required super
elevation. tio
3. It mainly eliminates the derailment, overturning of vehicles and discomfort of
passengers.

Characterstics of transition curve(Requirement)

1. It should be tangential to the straight


u
2. It should meet the circular curve tangentially.
3. Curvature should be zero at the origin on straight
ol

4. Curvature at joint should be same as circular curve


5. Rate of increase of curvature should be same as that if cant.
6. Length should be such that full cant is attained at the junction with circular curve.
us

Length of transition curve

The length of the transition curve should be that full super elevation or cant is attained at the
end of ythe transition curve and applied at a suitable rate.
vt

The various methods for calculating length are

I. By rate of super elevation


If the rate of application of super elevation is 1 in n of the length of the curve
and h is superelevation then the length is given by

L= nh where h=GV2/gr
If G=1.5m g=9.81m/s

DEPARTMENT OF CIVIL ENGINEERING, Page 78

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

Then h=1.18 v2/R cm where V is in km/sec and R is in metres


Then L=1.18 nv2/R

II. By time rate


Let the rate of application of super elevation be X cm/s, then as x cm cant is
applied in 1s, full cant of h cm will be applied in h/x seconds= t(say)
Distance travelled= v*t
L= V*h/x= v* GV2/gr*1/x

in
L= GV3/gRx metres
Substuting G and g
L= 0.327V3/Rx metres

n.
III. By rate of change of radial acceleration(most scientific)

Radial acceleration= V2/r


tio
Rate of change of radial acceleration α= V2/r/t

t= V2/αr……………………..(1)
u
if the length of transition curve is L and speed of the vehicle is v, then

t= L/v……………………..(2)
ol

from 1 and 2, V2/αr= L/v

L=V3/αr
us

If α=0.3m/s2/s

Then L= V3/0.3r

V= v/3.6 m/s, then L= V3/14r


vt

For sharp curves, L= (gtanθ)3/2 √R/α

L=12.8 √R metres

DEPARTMENT OF CIVIL ENGINEERING, Page 79

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

in
n.
u tio
ol
us
vt

DEPARTMENT OF CIVIL ENGINEERING, Page 80

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

in
n.
u tio
ol
us
vt

DEPARTMENT OF CIVIL ENGINEERING, Page 81

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

in
n.
u tio
ol
us
vt

DEPARTMENT OF CIVIL ENGINEERING, Page 82

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

in
n.
u tio
ol
us
vt

DEPARTMENT OF CIVIL ENGINEERING, Page 83

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

in
n.
u tio
ol
us
vt

DEPARTMENT OF CIVIL ENGINEERING, Page 84

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

UNIT 8

AREAS AND VOLUMES

Areas

The degree of accuracy of the computed area depends on

1. The accuracy of the field measurements

in
2. The accuracy of plotting when the calculations are made from a plan.

3. The method adopted for the computations.

n.
Area is usually expressed in

1. Hectare (ha) = (100m)⁴ = 10⁴ m²


tio
2. 1 arc (a) = (10m)² = 10²m²

3. 1 square kilometre (km)² = (1000m )² = 10⁶m²

Areas from field Measurnments


u
a) Area consisting of regular boundary
ol

1)Area of triangle = √s( s-a)(s-b)(s-c)

= ½ bh S = a+b+c
us

= ½ ab Sin C

= a . = b . = c .
vt

Sin A Sin B Sin C

b = C . Sin B

Sin C

DEPARTMENT OF CIVIL ENGINEERING, Page 85

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

2) Area of a trapezoidal = ½ (sum of parallel sides) * altitude

= ½ (a + b)*h

Area from co-ordinates

a) Co-ordinates ( Latitudes and Departures)

Area of closed traverse ABCDE,

in
A = Area of (A a e E) + Area of (E e d D) + Area of (D d c C) +

Area of (B b c C).

n.
A = ½ ( y₁ + y₅) (x₅ - x₁) + (y₅ + y₄) (x₄ - x₅) + (y₃ + y₄) (x₃ - x₄) – (y₂ + y₁) (x₂ - x₁) –
(y₃ +y₂) (x₃ - x₂)→①

since, Area of a trapezoidal A a e E is given as ½ (aA + eE) * ae


tio
But aA = (y₁ - 0) eE = (y₅ - 0) ae = (x₅ - x₁)

Area of AaeE = ½ [(y₁ - 0) + (y₅ - 0)] (x₅ - x₁)


u
= ½ [(y₁ +y₅) (x₅ - x₁)] →②

Similarly,
ol

Area of EedD = ½ [(y₅ + y₄) (x₄ - x₅)] →③

Area of DdcC = ½ [(y₃ + y₄) (x₃ - x₄)] → ④


us

Area of AabB = ½ [(y₂ + y₁) (x₂ - x₁)] → ⑤

Area of BbcC = ½ [(y₃ +y₂) (x₃ - x₂)] →⑥


vt

① → A = 1/2[ y₁x₅ - x₁y₅ + y₅x₄ - y₄x₅ - y₄x₃ + y₂x₁ - y₁x₂ + y₃x₂ - y₂x₃]

A = ½[x₁(y₂ -y₅) + x₂(y₅ - y₁) + x₃(y₄ - y₂) + x₄(y₅ -y₃) + x₅(y₁ - y₄)]

If a traverse have n sides , then

A = ½ [ x₁(y₂-yn) + x₂(y₃ - y₁) +. . . . . +xn-1(yn – yn) + xn(y₁ - yn-₁)

A = 1/2 [x₁y₂ - y₁x₂ + x₂y₃ - y₂x₃ +. . . . . xn y₁- yn x₁]

DEPARTMENT OF CIVIL ENGINEERING, Page 86

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

Calculations of area from cross staff Survey

In cross – staff survey, area of a plot can be calculated by the direct


use of field notes. In this method, the chain line run through the centre
of the area so that offset to boundaries are taken in order of their
chainages. Cross staff is used for setting out perpendicular offsets. For
accurate work, an optical square or prism can be used. The plot is
divided into right angled triangles and trapezoidal and area of each

in
figure is calculated separately.

Area of a right angle triangle = ½ *b*h

n.
Area of a trapezoidal = ½ (a+b) * h

=(O₀ +O₂)2d + O₁ . 4d - O₀ .2d - O₂ . 2d

2 3

= O₀ . 2d + O₂ .2d + O₁ . 4d - O₀ . 2d - O₂ . 2d
tio
3 3

2 2 3 3 3
u
= O₀ ( 2d + (- 2d) + O₁ . 4d + O₂( 2d – 2d)
ol

2 3 3 2 3

= O₀ (6d – 4d) + O₁ . 4d + O₂ (6d – 4d)


us

6 3 6

= O₀ .2d + O₁ . 4d + O₂ . 2d

6 3 6
vt

= O₀ . d + O₁ . 4d + O₂ . d

3 3 3

= d [O₀ +4O₁ +O₂]

similarly for the next 2 divisions ,

DEPARTMENT OF CIVIL ENGINEERING, Page 87

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

= d [ O₂ + 4O₃ + O2]

The total area = d [ O₀ + 4O₁ +O₂ + O₂ + 4O₃ +O₄ + . . . . . . ]

d[O₀ + On + 4( O₁ + O₃ + O₅ + . . . ) + 2( O₂ +O₄ +O₆) .

in
Planimeter

There are 2 types of planimeter

n.
a) Amsler polar palnimeter

b) Rolling planimeter
tio
Amsler polar Planimeter is being commonly used to determine the areas
of figures plotted to scale, with irregular boundaries.

It consists of 2 bars PQ and QR hinged at „Q „. A tracing point „P‟ is


grined around the boundary of the area to be measured. The QR
u
boundary of the area to be measured . The QR terminate at R which is a
stationary point and is fixed on the papers by a needle point and
ol

weight.

The displacement of the tracing point is measured by the wheel W


us

fixed in a plan perpendicular to PQ. While PQ moves, the wheel partly


rotates and partly shades on the paper. The total normal displacements is
measured by the rotation, the axial component of the motion causes slip
and does not affect the measurement.
vt

The wheel W is geared to a dial which records the members of


revolutions made by the wheel. One revolution of the wheel is equal to
one divisions of the dial which has 10 equal divisions on its periphery.
The wheel is graduated on its periphery with 100 divisions, subdivide
into tenth by a vernier . Thus each reading consists of four digits. The
units being read on the dial, the thousands on the vernier dial can be

DEPARTMENT OF CIVIL ENGINEERING, Page 88

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

utilized to know the number of the index . The distance between P and
Q is adjustable and on the tracing arm opposite the index by means of
the clamp and tangent screw.

Use of Planimeter

The method for using the planimeter as follows :-

1) To obtain the area in the desired unit, adjust the tracing arm.

in
2) Fix the needle point on the paper outside the area to be measured in
such a manner that the tracing point can reach all parts of the
boundary of the area.

n.
3) Select any arbitrary point on the boundary and move the tracing point
from this point all around the boundary in a clockwise direction till the
tio
tracing point reaches the starting point.

The area of the plan is calculated from the following formula :

A = M (Rϝ – Rı ± 10N + C)
u
M = multiplying constant whose value is marked on the tracing arm
ol

Rϝ and Rı = Initial and final readings

N = No of compute revolutions of the dial


us

[+ = when in clockwise direction ]

[- = when in anticlockwise direction]

C = Constant of the instrument


vt

C = 0 , for the needle point outside area

Zero circle is a circle round the circumference of which if the tracing


arm is moved, the wheel W does not rotate at all , and , no change in
reading.

A₀ is the area of zero circle, which is equal to C * M.

DEPARTMENT OF CIVIL ENGINEERING, Page 89

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

Digital Plannimeter

It is an area – curvature which plots the apex of the polygon figures


without the need to trace straight lines. Very small areas and curved
contours are also measured. This it is a computer + digitizer. It consist of

i) A control panel = which offers a large number number of facilities for

in
calculation of area and length,

ii) a display screen – which displays numbers, upto eight fingures, measuring
units, memory and functions.

n.
iii) a high frictions roller – for greater precisions.

iv) Trace arm :- Which travels up and down on the surface to be


measured.
tio
v) Trace arm facing lever :-It release the trace arm and power gets
switched on when it is lifted.
u
vi) Trace lens :- a large eccentric lens for wider view.
ol

vii) Trace point :- for putting on top of line to be traced.

viii) LED indicator :- Shows continuous made tracing.


us

ix) Measuring mode shifting switch :- Point mode and continuous mode can
be attered.

x) Start Switch :- To start the operalers.

The working of a digital planimeter is easy. Point mode is used to


vt

measure straight lines and continuous mode for curved lines. For switching
on the power, trace arm fixing lever is lifted. Required units is set with
numbers key and horizontal and vertical scale ratios are set. Centre mark
of the trace lens is set at the initial point and start switch is prssed.
For a straight line measurement , the trace mark is set at the other
apex of the straight line and start switch is again depressed. For curved

DEPARTMENT OF CIVIL ENGINEERING, Page 90

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

lines other mode is pressed. To find area enclosed by the figure, ENDA/L
key, pressed and area, length etc in given units pushed down to switch
off the power.

This planimeter has some special features like

a) Simultaneous length and area measurement.

b) Metric and British units can be measured.

in
c) Wide measuring range

e) High resolutions

n.
f)High precisions.

Capacity Contours tio


The contour plan can be used to calculate the capacity of a reservoirs.
The area enclosed by the contours can be measured using planimeter. The
volume of water in between 2 contours will be equal to average areas
of two contours multiplied by contours interval. Total volume can thus be
u
calculated.
ol

If A₁ , A₂ , A₃ . . . . An are the area of successive contours,

h is the contours interval and


us

V is the capacity of reservoirs

then by trapezoidal formula ,

V = h ( A₁ + An + A₂ +A₃ + . . . . An – 1)
vt

By prismoidal Formula,

V = h ( A₁ + 4A₂ + 2A₃ + 4A₄ + . . . . + 2An-₂ + 4An-₁ +An)

When n is an odd members

DEPARTMENT OF CIVIL ENGINEERING, Page 91

Vtusolution.in
Vtusolution.in
SURVEYING-II(10CV44)

V = h ( A₁ + 4(A₂ +A₄ + . . . . )+ 2(A₃+A₅+….) +An)

Volumes

1) From cross – sections ( for calculating volume of catch work)

a) Average end area (trapezoidal) formula

in
V =d (A₁+An + A₂ + A₃ + . . . An-₁)

n.
b) Volume by prismoidal formula

V = d[ ( A₁ +An) + 4( A₂ +A₄ + . . . ) + 2(A₃ +A₅ +. . . )

3
tio
From spot levels (for finding volume of length earth work)

V = A ( ∑ h₁ + 2∑ h₂ + 3 ∑ h₃ + 4 ∑ h₄)
u
4
ol

A = Area of horizontal base of prism.

h₁ , h₂ are the depth of corners of truncated prism.


us

From contours (for calculating the reservoir capacity)

i) V = h [ A₁ + An + A₂ +A₃ + . . . An-₁)

2
vt

when A₁ , A₂ etc are whole area under each contour line.

h = contour interval.

DEPARTMENT OF CIVIL ENGINEERING, Page 92

Vtusolution.in

You might also like