Practice 1.

Practice work #1
Measurements of physical quantities, measurement error
Goals: get the practice of calculating the main types of errors
1. Theory
The goal of any research is to establish links between various phenomena
and parameters. The quantitative relationship between the investigated quantities is
obtained as a result of measurements.
Measurement, the process of associating numbers with physical quantities
and phenomena. Measurement is fundamental to the sciences;
to engineering, construction, and other technical fields; and to almost all everyday
activities. For that reason the elements, conditions, limitations, and theoretical
foundations of measurement have been much studied. See also measurement
system for a comparison of different systems and the history of their development.
Measurements most commonly use the International System of Units (SI) as
a comparison framework. The system defines seven fundamental units: kilogram,
metre, candela, second, ampere, kelvin, and mole.
To measure the physical quantity A - is to know how many times in A (the
value of a physical quantity) contains the unit of measurement a . Then it can be
represented as A=na.
Before starting the measurement, it is necessary to determine the unit scale
of the instrument. The accuracy of the device is defined as half the scale of the
instrument
Error is the difference between the true value and the measured value of a
quantity such as displacement, pressure, temperature, etc. The better electronic
instrumentation systems are designed to limit the error, which is inevitable in any
measurement, to a value that is acceptable in terms of the accuracies required in a
engineering analysis or in the control of a process. Errors can occur due to the
following causes:
1. Accumulation of accepter error in each element of the instrumentation
system.
2. Improper functioning of any element in the system.
3. Effect of the transducer on the process.
4. Dual sensitivity of the transducer.
 5. Other sources of error.
Example 1. Determine the unit scale of bezel filled with water located on the
left in Figure 1.

Figure 1 - Beakers filled with water

Choose two nearest numbers to the water level, located on the scale of the
device. For example, V 1=900 мл and V 2=1000 . Count the number of divisions
between them, we obtain N = 5. Then the value of the scale division of the beaker
is
V 2−V 1 1000−900
Z= = =20 ml
N 5

Determine the amount of water in this beaker:

V =V 1+ n∙ Z=900+2.5∗20=950 ml .

The accuracy of the scale is:

Z 20
T acc = = =10 ml .
2 2

Write down the amount of water in the beaker taking into account the
accuracy of the measurement:
V =950 ±10 ml .

Task 1: Determine the scale division value, the measurement accuracy:

1 1.Scale division
value:
2.Measurement
accuracy:
3.Instrument
indication:

2 1.Scale division
value:
2.Measurement
accuracy:
3.Instrument
indication:

3 1.Scale division
value:
2.Measurement
accuracy:
3.Instrument
indication:

4 1.Scale division
value:
2.Measurement
accuracy:
3.Instrument
indication:
 5 1.Scale division
value:
2.Measurement
accuracy:
3.Instrument
indication:

Measurements generally provide approximate values.

As a result of a single measurement of a physical quantity X, the value Xexp is

obtained, which differs from the true value, which is unknown. As a result of
repeated measurements of a physical quantity, we obtain a series of values
x 1 , x 2 ,… , x N . The most probable value of the measured quantity is usually taken as
the arithmetic mean of the measured values, determined by the formula:
N

∑ xi
x 1 , x 2 , … , x N i=1
⟨ x ⟩= =
N N

(or the mean value is found using a graph of the dependence of physical
quantities). The measurement error will be determined by the formula:
|x T − ⟨ x ⟩|
e= ∙100
xT

After calculating the arithmetic mean of the measured quantity, the next step is to
determine the absolute and relative errors of the measurements. The absolute value
of the difference between the arithmetic mean and each of the individual
measurement results is called the absolute error of an individual measurement and
is determined as
∆ x i=|⟨ x ⟩ −x i|.

Often, the mean absolute error is determined as the arithmetic mean of the absolute
errors of individual measurements (determined with a large number of
measurements - more than ten):
N

∆ x 1 , ∆ x 2 , … , ∆ x N i=1 ∑ ∆ xi
⟨ ∆ x ⟩= = .
N N
its true value is enclosed. The range of values of the measured quantity from 〈x〉 -
The absolute error indicates two values of the measured quantity, between which

∆x to 〈x〉 + ∆x is called the confidence interval.

For example, as a result of measurements and subsequent calculations, the
diameter of a wire is determined:
⟨ d ⟩=2.4 мм и ⟨ ∆ d i ⟩ =0 ,1 мм .
This means that the true value of the diameter is d = 2.4 ± 0.1 mm, meaning it lies
in the interval between 2.3 and 2.5 mm.
The relative error of an approximate value of a certain quantity is defined as the
ratio of its absolute error to the exact (mean) value of that quantity. The relative
error is commonly expressed as a percentage:
⟨ ∆ xi ⟩
e i= ∙ 100 % .
⟨ xi ⟩
Example 2. When measuring the period of a pendulum, the following results were
obtained (in seconds): 3.1; 3.2; 3.0; 3.5; 3.2; 3.3; 3.1; 3.4; 3.0; 3.2; 3.3; 3.0.
Determine the true value of the period of oscillation.
Let's determine the average value of the period of the pendulum oscillation:
12

∑ Ti 3 ,1+3 ,2+3 , 0+3 , 5+3 , 2+3 ,3+ 3 ,1+3 , 4 +3 , 0+3 , 2

⟨ T ⟩= i=1 = =3,191 ( 6 ) ≈3 ,19 sec .
12 12

The rounding of the intermediate result has been done to three decimal places.
Now, let's calculate the absolute errors of individual measurements:

∆ T 1=|3 , 19−3 ,1|=0 , 09

∆ T 2=|3 , 19−3 ,2|=0 , 01

∆ T 3=|3 ,19−3 , 0|=0 ,19

∆ T 4 =|3 , 19−3 , 5|=0 ,31

∆ T 5=|3 ,19−3 ,3|=0 , 11

∆ T 6=|3 ,19−3 , 2|=0 , 01

∆ T 7=|3 ,19−3 ,2|=0 , 09

∆ T 8=|3 ,19−3 , 4|=0 , 21

∆ T 9=|3 ,19−3 , 0|=0 ,19

∆ T 10=|3 , 19−3 ,2|=0 , 01

 ∆ T 11 =|3 ,19−3 , 3|=0 ,11

∆ T 12=|3 , 19−3 , 0|=0 , 19

We calculate the average error of all measurements of the pendulum oscillation

period.
12

∑∆Ti
⟨ ∆ T ⟩= i=1 =¿
12
0 , 09+0 , 01+0 , 19+ 0 , 31+ 0 ,11+ 0 , 01+ 0 , 09+0 , 21+ 0 ,19+ 0 , 01+ 0 ,11+0 , 19
¿ =¿
12
¿ 0 , 12 (6 ) ≈ 0 ,13 sec.

The true value of the period of the pendulum oscillation will be expressed as a
range: T =3 , 19 ±0 , 13 seconds and will lie in the interval T min =3 ,06 seconds to
T max=3 ,32 seconds.

Task 2. The results of measurements of the disk diameter are presented in the
table. Determine the true value of the disk diameter.

№ d, мм ⟨ d ⟩ , мм ∆ d i , мм ⟨ ∆ d ⟩ , мм
1 42,4
2 42,6
3 42,8
4 42,7
5 41,9
6 41,8
7 42,0
8 42,4
9 41,9
10 41,9
11 42,0

The average value of the disk diameter:

N

∑ di
⟨ d ⟩= i=1
N

Absolute errors of the disk diameter measurements:

∆ d i=|⟨ d ⟩ −d i|

The average error of the disk diameter measurements:

 N

∑ ∆ di
⟨ ∆ d ⟩ = i=1
N

The true value of the disk diameter will be expressed as a range:

d= ⟨ d ⟩ ± ⟨ ∆ d ⟩

With a small number of conducted measurements (up to 10), for a reliable

estimation of errors, the root mean square deviation or standard deviation is
calculated:

σ=
√ ∑ ∆ xi2
N −1

The root mean square error of the mean value is calculated using the formula:
σ
m=
√N
A confidence probability p (0.6, 0.8, 0.95, 0.99, or 0.999) is given. Then, using the
"Student's distribution table," it is necessary to find the corresponding value of the
Student's coefficient t for the given number of measurements N and calculate the
error with the specified probability p using the formula X погр=m ∙t .
Table 1 - Student's distribution coefficients
Количеств Значение доверительной вероятности p
о 0,6 0,8 0,95 0,99 0,999
измерений
N
2 1,376 3,078 12,706 63,657 636,61
3 1,061 1,886 4,303 9,925 31,598
4 0,978 1,638 3,182 5,841 12,941
5 0,941 1,533 2,776 4,604 8,610
6 0,920 1,476 2,571 4,032 6,859
7 0,906 1,440 2,447 3,707 5,959
8 0,896 1,415 2,365 3,499 5,405
9 0,889 1,397 2,306 3,355 5,041

Example 3. Calculate the mean and its error for the series of numbers - 10, 11, 12,
13.
№ Х ∆ x i=|⟨ x ⟩ −x i| 2
∆ xi
1 10 1.5 2.25
2 11 0.5 0.25
3 12 0.5 0.25
4 13 1.5 2.25
 ∑ x =46 ∑ ∆ x 2=5

46
Let's determine the arithmetic mean: ⟨ x ⟩= 4 =11,5 .

It is necessary to calculate the absolute errors for each individual value of the
quantity and determine the sum of squares: ∑ ∆ x 2=5 Начало формы
Let's determine the root mean square deviation:
σ=
√ 5
4−1
=1, 29

Let's determine the root mean square error of the mean value:
1 ,29
m= =0,645
√4
Let’s set the confidence probability p=0.9. From the table, we find the value of the
Student's coefficient (N=4) is 1.639.
Let's calculate the error with the given probability: Xerror
=0.645×1.638=1.057≈1.06
The result of the calculations will be expressed as Xtrue=11.5±1.06, meaning it
lies in the interval between 10.44 and 12.57.
Task 3. Let's consider the values obtained as a result of four measurements: 2.8;
2.79; 2.84; and 12.57.
№ Х ∆ x i=|⟨ x ⟩ −x i| 2
∆ xi
1
2
3
4
∑ x =¿ ¿ ∑ ∆ x 2=¿ ¿

Let's determine the root mean square deviation:

σ=
√ ❑ =¿
4−1

Determine the root mean square error of the mean value:

m= ❑ =¿
√4
Let's set the confidence probability p=____. From the table, we find the values of
the Student's coefficient (for N=4) = ____.
Let's calculate the error with the given probability. X error=m∗t=¿ ¿
The result: X true=¿ ______________ ± ______________
Interval: _____________________ and ____________________
Results of an experiment are often beneficial to represent graphically. Graphs
allow for a visual presentation of research outcomes, revealing connections
between studied phenomena and facilitating their analysis. With the help of graphs,
patterns in the development of phenomena can be easily traced. Moreover, they are
easier to comprehend compared to tables.
It should be remembered that only well-executed graphs are easy to read.
Therefore, it is necessary to adhere to the requirements for plotting graphs:
 Graphically represent only the key data that requires attention;
 Avoid overloading the graph with unnecessary lines;
 Plot in a way that the most important aspects and connections are visible
(use different colors and line markings);
 Graph construction begins with presenting the results of the work in the
form of a table, based on which the graph is plotted.

When constructing a graph, it's necessary to pre-select the scale, mark the
coordinate axes, and only then proceed to plot the experimental points on the
graph. By drawing a smooth curve through the experimental points, you obtain the
graph of the investigated relationship.
When choosing the scale for the graph, consider the following principles:
A) Experimental points should not overlap, so it's better to choose a scale that
spreads the points with a reasonable interval.
B) The scale should be simple. It's easiest if one unit of the measured quantity
(or 10, 100, 0.1, etc., relative to the unit) corresponds to 1 cm.
C) Powers of ten are more convenient to relate to the unit of measurement.
Then, the divisions on the graph can be marked with numbers like 1, 2, 3 or
10, 20, 30, rather than 10000, 20000, etc.
 Figure 2 - Graph of a certain physical process
On the coordinate axes, it is advisable to indicate the name or symbol of the
quantity and the units of measurement (for example: t, sec).
Task 4. A coil of wire has electrical resistance that changes with temperature, as
shown in the table:
Resistance, 5.5 6.0 6.4 6.9 7.4 8.0
Ohms
Temperature, ℃ 10 20 30 40 50 60

Please create a graph showing the dependence of the wire coil resistance on
temperature. The resistance of the coil at 0℃ is _______ Ohms. The coil has a
resistance of 7 Ohms at a temperature of t = _______℃.

Task4 In one stack, there are N elements. Based on the measurement results,
perform calculations:
The thickness of 11,0 21,5 32,0 43,0 54,0 64,0
all elements in the
stack is D
millimeters.
The number of 10 20 30 40 50 60
elements is N.
The arithmetic
mean of the stack
thickness,
denoted as ⟨d⟩.
Начало формы

Absolute
measurement
error, Δd
Square of the
absolute
measurement
error, Δd2
Root mean square
deviation, σ
Root mean square
error of the mean
value, m
Confidence
probability, p
Student's
coefficient, t
Error with a given
probability, δ
Result of
thickness
measurements for
the sheet

REFERENCES:
1. James W. Dally, William F. Rilley, Kenneth McConnel. Instrumentation for
engineering measurements. – USA