Uncertainty Calculations

Uncertainty is the ½ of the minimum reading of an instrument. It is also called as the Resolution
or Precision.

The minimum reading of some instruments are given below.

0 Rule

Meter rule(more than 5cm) = 0.1 cm (1mm) 0

Vernier caliper(less than 5cm)= 0.01cm Ruler

Micrometer screw gauge(less than 5mm) = 0.001cm(0.01mm)

Electronic balance = 0.01g

Stop watch = 0.01s rule

32.0cm 32.1cm

1. Single Measurement

Eg: Length of a book = 32.0cm

32cm

Reading, l = 32.0cm

Uncertainty = ½ of the minimum reading of the instrument

Uncertainty, ∆l = 0.05cm = 0.5mm

Final Reading = (32.0 + 0.05)cm

Range = (31.95 – 32.05) cm

%U = 0.05 x 100 = 0.16%

32.0
2. Repeated Readings (Average Measurement)

Eg:1 Time taken for the ball-bearing to drop from a height

t1 = 2.31s t2 = 2.25s

t3 = 3.21s anomolous t4 = 2.34s

t5 = 2.23s

Average time, t = 2.31 + 2.25 + 2.34 + 2.23

4

= 2.2825s

= 2.28s (cannot keep more than 2 d.p / 3s.f)

Uncertainty = Range = Max. – Min.

2 2

Uncertainty, ∆t = 2.34 – 2.23

2

= 0.055s

= 0.06s (2 d.p)

Final Reading = (2.28 + 0.06)s

Range = (2.22 – 2.34) s

%U = 0.06 x 100
2.28

= 2.6%

If the repeated readings give the same measurement

then uncertainty of the measurement = ½ of the minimum reading

Eg: 2 - 22.1cm, 22.1cm, 22.1 cm, uncertainty = ½ (1 mm) = 0.5 mm = 0.05 cm

3. Equations

a) Addition/Subtraction

Eg: L1 = (95.0 + 0.05) cm

L2 = (12.5 + 0.05)cm

Addition Subtraction

95.0cm 12.5cm 95.0cm

L L
12.5cm

L = 95.0 + 12.5 L = 95.0 - 12.5

L = 107.5cm L = 82.5cm

Uncertainty of L = U of L1 + U of L2

U of L = 0.05 + 0.05 U of L = 0.05 + 0.05

= 0.1cm = 0.1cm

Final Reading = (107.5 + 0.1)cm Final Reading = (82.5 + 0.1)cm

%U = 0.1 x 100 %U = 0.1 x 100

107.5 82.5
= 0.09% = 0.12%
b) Multiplication/Division

Eg: Area of rectangle / Density of an object

Area of a rectangle
Length, L = (32.4 + 0.05)cm
Breath, b = (6.68 + 0.005)cm

Area, A = 32.4 x 6.68

= 216.432cm2
= 216cm2
(Minimum significant figures of the given data, 3 s.f.)

A = L x b

% Uncertainty of A = %U of L + %U of b

U of A x 100 = U of L x 100 + U of b x 100

A L b

∆A = ∆L + ∆b
A L b
∆A = A{∆L + ∆b}
L b

∆A = 216{0.05 + 0.005}
32.4 6.68

= 216{1.54x10-3 + 7.49x10-4}

= 0.495 cm2

Final Reading = (216 + 0.495)cm2

%U = 0.495 x 100 = 2.29x10-3

216
Eg: Density of an object D = m
 V

% U of D = % U of m + % U of V

∆D = ∆m + ∆V
D m V

c) Power

Eg: Volume of a cube of length, l

V = l3

% U of V = 3 x %U of l

∆V = 3x ∆l
V l

d) Multiplied by a constant

Eg: Circumference, C / Diameter, d / Periodic Time, T

C = πd r = ½d

%U of C = ∆C x 100 %U of r = ∆r x 100
C r

U of C(∆C) = π x U of d(∆d) U of r = ½ x U of d

∆C = π x ∆d ∆r = ½ ∆d

Therefore %U of C = π∆d x 100

πd
= ∆d x 100
d
%U of C = %U of d Similarly, %U of r = %U of d

Eg: Periodic Time, T

Measurement = Time for 20 oscillations = T20 = (5.56 + 0.01)s

T = T20 = 5.56s
20 20
T = 0.28s

∆T = 1 x ∆T20
20

= 1 x 0.01
20
∆T = 5 x 10-4 s

But %U of T20 = 0.01 x 100 = 0.18%

5.56

%U of T = 5x10-4 x 100 = 0.18%

0.278
 Percentage Difference (AS syllabus)

1. Using Practical(P) & Theoretical Values(T)

(take the + value of difference)

%D = P - T x 100
T

2. Using two Practical Values ( P1 , P2 )

%D = P1 - P2 x 100
(P1 + P2) / 2

Eg 1: Find the % D of gravitational acceleration, g.

Practical Value, P1 = 9.78ms-2.

Theoretical Value, T = 9.81ms-2.

% D = 9.81 – 9.78 x 100 = 0.3%

9.81

Eg 2: Two practical values of constant, k are 2.23 and 2.19

%D = 2.23 - 2.19 x 100

(2.23+2.19)/2

= 0.04 = 1.8%
2.21
 Criticizing a Table

 Check the Units in the column headings

 Inconsistent Precision (no. of significant figures should be equal in a certain
column in a table)
 Precision is not high enough. (more significant figures can be measured)
Eg: 25.0 cm = 0.250m
 Average values are not considered. (Repeat readings are not considered)
 More no. of data should be taken. (not enough data)
 In oscillations, 20 oscillations should be considered

Advantage of a graph over an average value

 Anomalous readings can be identified.
 By considering a larger range, an gradient is calculated.
 Random errors can be identified.
Eg: If the points are scattered, then there is random error.
 Systematic errors can be identified.
Eg: If there is an intercept in a graph, then there is systematic error.
 Graph can be extrapolated and more information of the result can be
analyzed.
 The relationship between the variables can be found.
 Types of Errors

There are two types of errors.

1. Random Error
2. Systematic Error

Random Errors Eg: Reaction Time.

When using the stop watch to measure a time, reaction time occurs. Reaction time
is different person to person. Normal reaction time of a person is 0.1s.

Random Errors can be identified in a graph. If the points are scattered in a graph
we know that there is random error.

Random Errors can be eliminated by repeating many times a practical. So that we

can eliminate the anomalous readings and to consider the points which are close
together. As result points are much closer to the graph.

Systematic Errors Eg: 1. Zero errors.

2. Parallax errors (eye level).
3. The surface in bench is not horizontal.

Systematic Errors can be identified in a graph. If there is an intercept, we know

that there is systematic error.

Systematic errors cannot be eliminated by repeating the practical.

 The Difference between Accuracy and Precision

Diagram A – Précised but Not Accurate

Précised - This is because you can see all the points are very close
to each other.

Not Accurate- All the points are deviated from the centre of the
diagram.

If a graph is plotted it is as follows. All the points are closer to

the line. They are not scattered. Therefore there is no random
error (Precised). But the graph does not pass through the
origin. So there is systematic error (Not accurate).

x
x
x
Diagram B – Précised and Accurate

Précised - This is because you can see all the points are very close
to each other.

Accurate - All the points are close to the centre of the diagram.

If a graph is potted it is as follows. All the points are closer to

the graph. They are not scattered. So that there is no random
error (Precised) and the graph passes through the origin. So
there is no systematic error (Accurate).

x
x
x

Diagram C – Not Précised, Not Accurate

Not Précised - This is because you can see the points are scattered.

Not Accurate- All the points are not close to the centre of the diagram.

If a graph is potted it is as follows. All the points

are not closer to the line. They are scattered. So that there is
random error (Not Precised) and also the graph does not pass
through the origin. So there is systematic error (Not
accurate).

x x
x
Diagram D – Not Précised but Accurate

Not Précised - This is because you can see the points are scattered.

Accurate - You can see the points are very close to centre of the diagram.

If a graph is potted it is as follows. All the points are not closer

to the line. They are scattered. So that there is random error
(Not Precised). But the graph passes through the origin. So
there is no systematic error (Accurate).

x x
x
 How to get accurate Readings

1. Measuring the thickness, t of a coin (20mm)

Method 1 Method 2
Using one coin Using multiple coins

Measure the thickness using micrometer. t

% U of one reading = 0.01 x 100

20
= 0.05% Measure the thickness using micrometer.

%U can be reduced by taking three Measure thrice, at least two readings

readings, at least two readings perpendicular to each other.
perpendicular to each other and
calculating an average value. % U of one reading = 0.01 x 100
20x5
= 0.01%
* When measurement is larger, %U has
reduced. But uncertainty is the same.
Therefore this Method 2 is better than
Method 1.
2. Measuring the diameter, d of a marble

Method 1 Method 2
set square

Take three readings at least two

readings perpendicular to each d1 d2
other and calculate an average d = d2 - d1 meter rule
value. 2

%U is reduced since a longer length is

measured.

Method 2 is better.
3. Measuring height, h

eye

Base of the meter rule is short. Therefore, when measuring height, the rule may not
be vertical. So we use the set square as shown in the diagram, in order to keep the
rule vertical.
If the bench is not horizontal, the rule still may not be vertical and we cannot
eliminate this error. So it is called systematic error.

4. Measuring extended length, l

meter rule

Eye

In this practical, we need to check

1. the meter rule is in horizontal.

2. since there is a gap between the spring and the meter rule, a
set square is used to check the horizontal level as shown in the diagram.
3. When taking the reading, to avoid parallax error, the eye needs to place at the
 correct level. Otherwise there will be a systemic error

5. Measuring an angle of an inclined plane, ϴ

protractor
H

When measuring the angle,ϴ we cannot use the protractor, as the reference line
cannot be align on a bench as shown on diagram. Therefore trigonometry is used to
calculate the angle, ϴ.

Measure D and H and use tan ϴ = H

D

6. Taking a reading using Vernier Caliper or Micrometer

Before taking any reading check the zero error of the Vernier or Micrometer.
Otherwise there will be a systematic error in your reading.
 Additional Notes (not in the syllabus)

How to eliminate ZERO ERROR of Vernier

( - ) zero error
0 ( + ) zero error

0 0
Vernier zero Vernier zero

zero of main scale

Final Reading = Reading - ( + ) zero error

Reading - zero error

Final Reading = Reading - ( - ) zero error

Reading + zero error
 How to calculate zero error

( + ) zero error

0 D main scale

d
zero error 0 vernier scale

You can see the 2nd division of Vernier is coinciding with the main scale.
Therefore short method of calculating (+) zero error is as follows.

(+) zero error = 2(count from the 0th Vernier division) x least count

= 2 x 0.01

= 0.02 cm

Actually the ( + ) zero error is the distance between two zeros. It can be calculate
as follows.
( + ) zero error = D - d
= 2 mm - (2 x 0.9mm)
= 0.2 mm
= 0.02 cm
( - ) zero error

0 main scale
D

d 10
zero error 0 vernier scale

You can see the 3rd division of Vernier is coinciding with the main scale.
Therefore short method of calculating (-) zero error is as follows.

(-) zero error = 7(count from the 10th Vernier division x least count

= 7 x 0.01

= 0.07 cm

Actually the ( - ) zero error is the distance between two zeros. It can be calculate as
follows.
( - ) zero error = d - D
= (3 x 0.9 mm) - 2 mm
= 0.7 mm
= 0.07 cm