1.

INTRODUCTION
The market offers a multitude of different devices to measure and control structures.
The specific application is often decisive because each piece of equipment has its own
advantages and disadvantages. The choice of the tool is not random but how not to
deceive? Over the years, these tools have evolved to keep up with increased standards.
quality: manual measuring tools still exist for simple applications of
control, but thanks to the evolution of electronics and computer science, the new devices more
sophisticated and so-called 'automatic' systems have managed to carve out a prominent place in the industry and the
metrology laboratories.

Knowing how to adapt a device to a specific type of work is therefore important and essential.
Knowledge of the tool must therefore be perfect.

2. METROLOGICAL CHARACTERISTICS OF AN INSTRUMENT

OF MEASURE
Lame tology is the science of themeasureIt defines the principles and methods for
guarantee and maintain trust in the measures resulting from themeasurement processIt is about
a transversal science that applies in all fields where quantitative measurements are
carried out.

To take a good measurement, perfect mastery of the device both in handling,

that are required in its metrological quality; not to forget obviously the methods that relate to it.
apply.

The metrological quality of ameasuring instrument is the set of data that characterizes the
quality of the measurement made by the device in question.

The main characteristics of measuring instruments (or metrological properties of

measurement devices) are defined within the framework of the International Vocabulary ofmetrologyand
include, among others:

sensitivity;
the accuracy
loyalty.
It should be noted that the term 'precision' is not used and constitutes a language error in
Metrology. This term is indeed absent from the International Vocabulary of Metrology.1.

Sensitivity: it is the smallest variation of the quantity to be measured that the instrument can detect.
capable of detecting

Fidelity: it is the ability of an instrument to always give the same value of a

measured grandeur several times (it is a primary quality)

Aside from operating conditions, the fidelity of a device is essentially related to two types
of characteristics: accuracy and fidelity.
A device is accurate if it is both correct and reliable.
The accuracy of a measuring device can also be affected by external causes:
operational error
error caused by influencing factors (temperature, pressure, etc.)
reference or calibration error,
error ofhysteresis,
fineness error etc.

The accuracy: it is the ability to provide an exact value of the measured quantity.

The accuracy error is the overall error resulting from all the causes for each of the
measurement results taken in isolation.
It is therefore the device's ability to produce results that are not tainted by error.
In the case of multiple measurements, it is the difference between the average result and the true value.

3. TERMINOLOGY

3.1 Measurement

Measurement is the set of operations aimed at determining the value of a quantity;

In surveying, the term 'measure' is generally used for both the method and the result.
Let us remember that one must not confuse measurement with calculation of a value.

3.2. True value of a quantity

The true value of a quantity is the value that perfectly characterizes a quantity.
defined; in the conditions that exist at the time this quantity is examined.
The true value of a quantity is an ideal notion that is generally not known. For example,
We will never be able to know the true value of the length of the classroom until a
approximation close; this value actually varies depending on certain conditions, such as by
example, temperature (expansion of materials).
Examples of true values:
The elevations in a closed path = 0
The int
of a triangle = 200g
The measure of a right angle = 100gr
Etc...

3.3. Conventionally true value of a quantity

It is an approximate value of the true value of a quantity such that; for the purpose to which
this magnitude is used, the difference between these 2 values can be neglected.
One generally determines the conventionally true value of the quantity by means of
methods (such as the calculation of the arithmetic mean of the measurements) and with the help of
instruments of suitable precision for each specific case.
4. NOTIONS OF FAULT AND ERROR
The topometric measurements, like all physical measurements, are inevitably inaccurate.
These inaccuracies come from:
Measuring instruments
The senses of the observer
And sometimes measurement methods
In topometry, they are classified into two main categories:
The mistakes
The mistakes

4.1. Mistakes

An error is an inaccuracy whose order of magnitude is significant relative to the precision.

sought in moderation.
Mistakes come from carelessness, clumsiness, or forgetfulness. Mistakes are not
susceptible to any mathematical analysis; the only way to detect them is to carry out some
superfluous measures or repetitions of measures.

Examples:
Errors of 1m on a leveling staff reading
Forgotten chain reach in distance measurement

It is therefore essential to eliminate them. For this, we use:

Direct control that involves repeating the measurement using the same method
Indirect control with a measurement or calculation using a different method.
Note: (without control or verification in topography, the measurement is not valid!!!)

4.2. Errors

[Link]

A systematic error is an error that, during several measurements made in the same
conditions remain constant in absolute value and in sign. It is due to a determined cause:
Example:
Chain expansion error
Vertical collimation error
Apparent Level Error
- Etc…
In most cases, the causes are known, so they must be eliminated.

Application: We measure several times with a 30m tape the length of a calibration base.
whose determined value is 120.428m. Complete the table below

Measured distances 120.551 120.548 120.547 120.559

Systematic Errors
Average systematic errors
 [Link]

An accidental error is an error that varies unpredictably in absolute value and in

signal when performing a large number of measurements of the same value of a quantity in
practically identical conditions.
These are observation errors, which have no systematic character. It is accepted by
definition that their influence cancels out in the final result, if one repeats a very large
number of times the operations (theoretically, an infinite number of times)
As a result, this means that:
Accidental errors are equally positive and negative
To every positive error, there corresponds a similar negative error.
Example: Measurement of mm on a target or a chain.
Application: We measure several times with a 30m tape the length of a calibration base.
of which the determined value is 151.128m. Complete the table.

Measured distances 151.133 151.124 151.120 151.135 151.140 151.138

Accidental Errors

4.2. True errors

The true error 'e' is the difference between the measured value and the true value of the quantity.
measured.

e = measurement result - true value

For a large number 'n' of measurements of the same quantity 'x', we have:
e1=x1-x
e2=x2-x
ei=xi-x
en=xn-x
ei 0
The true errors being of random signs and of the same order of magnitude, their sum is to some extent
almost none.

4.3. Apparent errors

The apparent error 'v' is the difference between the measured value and the most likely value of the
measured magnitude. It is also called deviation in direct measurement and residual in indirect measurement.

v = result of the measurement - most probable value of the measured quantity

v1=x1-x
v2=x2-x
vi=xi-x
vn=xn-x
vI 0
 4.4. Dispersion indices (reliability errors)

The fidelity of a measuring instrument is the quality that characterizes its ability to provide,
for the same value of the measured quantity, consistent indications among themselves.
The dispersion of indications of the phenomena presented by an instrument that gives in
a series of measurements of the same value of the measured quantity, taken under the conditions
well defined, different indications.

This dispersion is quantitatively expressed by the range of dispersion or by a

dispersion index also called fidelity error.
The different measures of dispersion are:
The probable errorp
The arithmetic mean ea
The mean squared error 'mse' or standard deviation
-
Dispersion indices are units of measurement for random errors. It is demonstrated (calculations
of probabilities) that they are related by the relationship:

4. emq = 5ea=[Link]
In topometry, although it is possible to use either of the indices; one adopts
The standard deviation or the mean squared error to characterize the precision of an instrument
measurement or the result of a measurement.
[Link]

The likely error is the equiprobable gap which has a probability of 1/2 of not being exceeded in
absolute value.
In practice, the number of measurements taken is generally insufficient for the
process provides a significant value of the probable error.

2 5
3. = ea 6
ep= emq

[Link]

The arithmetic mean error 'eais the arithmetic mean of the absolute values of the errors

For 'n' true errors:

n
ei
i 1
ea
n
With:

'e' or 'v': the different residues

‘’n’’ : nombre de mesures
For 'n' apparent errors:
 n
vi
i 1
ea
n  0.5

[Link]'mse'orstandarddeviation

The mean square error 'mse' or 'standard deviation' is equal to the square root of the
arithmetic mean of the squares of the errors.
The standard deviation is a measure of the dispersion of values around their arithmetic mean.

In the case of true errors:

n
ei 2
i 1
n
With:
In case of apparent errors: 'e' or 'v': the different residues
‘’n’’ : nombre de mesures

n
vi2
i 1
n 1

Note: The standard deviation of a device is usually provided by the manufacturer. But after
After several years of use, with the depreciation of the device and wear, this value changes.
(decreases). The standard deviation must therefore, in this case, be determined experimentally, through a large
number of measurements of a quantity.

The standard deviation of an arithmetic mean (AM) is:

d'a Measure
MA
n (for more explanation see point 3.9)

4.5. Frequency curves of accidental errors

(insert this curve here)

8
Tolerance T = [Link]= 3 = 2.7
Note: The previous formula T = 2.7 is valid for true deviations 'e'
For the apparent discrepancies, it becomes T = 2.58

5. CART TYPE BASED ON INDEPENDENT DIRECT MEASUREMENTS

5.1. Any Function

Let G be a quantity that depends on various independent direct measurements X, Y, and Z, whose deviations -
y
respective types are x, and z
G f(X,Y,Z)
It is easily demonstrated that the standard deviation G east

2 2 2
2 F F F
G . X 2 . Y 2 . Z
2

X Y Z

Example:
In the measure of the two sides of a rectangle, let them be L and l the standard deviations on the

values X and Y of the sides. What is the standard deviation? s from the surface?

S L.l

2 2
2 S 2 S 2
S
. L . l
L l

2
S l 2. L  2L. 2
l
2
S l2. L  L. l
2 2 2
 5.2. Linear function

Let G = a.X + b.Y + c.Z with: X, Y, Z: independent measures

X , Y , Z corresponding standard deviations

a, b, c: the coefficients of X, Y, Z

2 2 2
2 G 2 G 2 2
G . X . Y G . Z
X Y Z

2
G a2. X  2b 2 . Y  c 2 2. Z
2

5.3. The quantity G is the sum of 'n' quantities measured with the same degree.
of precision

For example, it can be about 'n' angles of a polygon or the 'n' spans of a ribbon of a
length

G = X1+X2+….Xn

2 2 2
2 G 2 G 2 G 2
G . X1  . X2  ... . Xn
X1 X2 Xn

2 2 2 2
G 1. X1 1. X2  ... 1. Xn or 2
X1
2
X2
2
Xn

(meme precision)
2 2
G n. X G n. X X .n

So so G X n

5.4. The quantity G is the arithmetic mean of 'n' measured quantities with
the same degree of precision

For example, it may involve a length or an angle measured 'n' times in the same.
conditions

X 1 X 2 X 3 ... X n
G
n
 2 2 2
2 G 2 G 2 G 2
G
. X1  . X2  ... . Xn
X1 X2 Xn
2 2 2
1 2 1 2 1 2 2 2 2
. X 1 . X2 ... . Xn or X1 X2 Xn
n n n
(meme precision)

So
2 2 2
2 G 2 G 2 G 2
G
. X1  . X2  ... . Xn
X1 X2 Xn
2 2 2
1 2 1 2 1 2 2 2 2
. X1  . X2  ... . Xn or X1 X2 Xn
n n n
(meme precision)

2 2 2 2
X1 X2 Xn X

so:
2
2 1
G n. . 2
X
n
2 122
G n.. X
n2
2 12
G . X
n
2 2
1 X X
G . 2
X
X
n n n n

X
G
n

[Link]