Fairground Ride

Safety Report
King Yeung Justin Cheng
Department of Mechanical Engineering
Imperial College London
Date:14/11/2024
Wordcount: 1973
Time spent: 28hrs
Predict Mark:
Abstract
The experiment investigates the safety of drilling a hole in the connecting rod
of a fairground ride. The experiment is concerned about the yield failure and
fatigue failure of the rod. So the safety of the ride is determined by SF y and
SF f . Two tests are performed to investigate the material and geometric
property of the rod before and after the modification. In the materials test, a
tensile test is done on a steel sample of the material of the rod until failure. In
the geometric test, a tensile test is performed on an aluminium rod of same
dimension as the rod after modification, with strain gauges to measure strain
at different positions. Results of the materials test show that the σ y of the
steel sample is 303 ± 3.04 MPa, the σ UTS is 474 ± 4.75 MPa. The σ f is calculated
to be 316 ± 3.17 MPa. Hence the safety factors SF y =9.82 ±0.0984 and
SF f =10.2 ± 0.1022 before the modification. The geometric test showed that the
SCF is3.41 ± 0.154, SF y =2.88 ±0.159∧SF f =2.99 ± 0.165 after the modifications.
The fairground ride is concluded to be unsafe with the modification due to the
safety factor being lower than 5.

Table of Content

2
s
Nomenclature..............................................................................................................iv
1) Introduction...........................................................................................................1
2) Materials and Methods.......................................................................................2
a) Materials Test...................................................................................................2
b) Geometric Test................................................................................................3
3) Result....................................................................................................................5
a) Original data.....................................................................................................5
b) Materials test....................................................................................................5
4) Geometric Test................................................................................................6
5) Discussion............................................................................................................8
6) Conclusion............................................................................................................9
End Matter....................................................................................................................9
Acknowledgements.....................................................................................................9
List of References......................................................................................................10
Appendices.................................................................................................................11
Appendix A..............................................................................................................11
Appendix B..............................................................................................................11
Appendix C.............................................................................................................13

3
Nomenclature

𝐸 Young’s modulus
F Load/Force

𝑙
Fc Centripetal force
Extensometer gauge length

𝑚
l0 Initial extensometer gauge length

𝑟
Mass

𝑉
Radius
Voltage

𝜀
Δ𝑙 Extension
Strain

𝜎
εt True strain
Stress
σA Working stress after modification
σf Fatigue stress
σ op Operating Stress
σ UTS Ultimate tensile stress

𝜔
σy Yield stress
Angular velocity
A Surface Area
BV Bridge Voltage
FE Finite element
FEA Finite element analysis
GF Gauge factor
PEN Professional Engineering Skills
SF f Safety factor against fatigue
SF y Safety factor against yield
SCF Stress Concentration Factor
UTS Ultimate Tensile Stress

4
 1)Introduction
This report investigates the safety of drilling a 25 mm diameter hole through
the connecting rod of a fairground ride. The fairground ride consists of a 500
kg car and a connecting rod. With cross-sectional area of the connecting rod
being 80 mm x 10 mm in. The 500 kg car spins at 30 rpm with a 5 m radius
during its operation.

5 m radius

80 mm

Figure 1. Schematic of the fairground ride showing the proposed modification

Considering the forces acting on the rod, only the centripetal force will cause
stress to the rod as it is the only force that acts along the axis. So in this
investigation, the operating stress is calculated by
Fc
σ op=
A
( 1)
Two failures are concerned. Yielding and fatigue. Yielding refers to the
material deformed plastically, and the system has failed as the dimensions
are changed. Fatigue failure is due to stress varying between 0 and maximum
stress during its normal operation. Small cracks may form and grow,
eventually causing failure. Fracture failure is not considered as it happens
after yielding.
Two tests are performed to find the safety factors( SF f and
SF y) without the hole and the safety factors with the hole. The first test is the
material properties test, performing a tensile test on a steel sample of the
material of the rod. Find out σ y and σ f of the material. The second test is the
geometric test. By using strain gauges, strain at different positions of the rod
is measured and hence the stress concentration factor is found. The report
will talk about the two tests performed and the results of them and finally the
conclusion of the investigation.

2)Materials and Methods

a) Materials Test
Procedures
1. Measure the diameter of the sample using micrometer by taking the
average of 3 measurements
2. Apply load to the sample using a universal testing machine until failure
3. Record data of load, extension, gauge length extension, time, stress,
and strain throughout the process
Tensile test is performed on a steel sample, putting the sample under
controlled tension until failure, using the Instron Universal Tensile Testing
Machine shown below

Figure 2. Schematic of Instron Universal Tensile Testing Machine

The machine will output the time, load, and displacement of the crosshead.
2
And the video extensometer will output the extension of the gauge length,
which is the distance between two dots marked on the sample.
The values of stress and strain can then be calculated by
Δl
ε=
l0
(2)
F
σ=
and A.
( 3)
The stress-strain graph has three regions, the elastic region, the Lüders band
and the plastic region. The minimum point of the Lüders band is taken as the
yield stress instead of the end of the elastic region. Causing σ y as well as SF y
to be smaller, resulting in a more conservative calculation of the safety of the
connecting rod.
σ UTS is also found in the stress-strain curve, being the maximum point of the
graph. It is used to calculate σ f by the equation:
2
σ f = σ UTS
3
( 4)
Where the formula is suitable for most steels(see Appendix B).

b) Geometric Test
Before adding the hole, stress is uniform throughout the body of the sample.
However, with the hole introduced, the stress is no longer uniform, as the
cross-section area is smaller at the hole, causing stress to be higher near the
hole. Therefore, the objective of the geometric test is to investigate how the
yield strength and fatigue strength will decrease with the addition of the hole.
Procedures
1. Take the average of 3 measurements of the thickness and the width of
the rod
2. Put strain gauges on the block at specified positions
3. Apply a small load to the plate and zero the strain and load output
4. Apply a load of 4.5 kN, taking a reading for each strain gauge
5. Unload the plate
6. Zero the strain gauges and the load output
7. Repeat the measurement
In step 4, only 4.5 kN is applied. This is to remain in the elastic region, where
stress is proportional to strain with the ratio being Young’s modulus.

3
 Front Back

Figure 3. Positioning and numbering of strain gauges on rod

Strain gauges are used in this experiment to measure strain at a certain point.
As the resistivity changes as it elongates with the rod, hence the voltage
across it. The value of strain is given by
4V
ε=
GF × BV
(5 )
Where BV = 2.5V and GF = 2.1 in this experiment
Nine strain gauges are attached to the plate. With 1 and 2 on the inner wall of
the hole. Which has maximum strain as they are closest to the hole. 6 and 7
are holes that are far enough from the hole to be assumed unaffected by it,
meaning that 6 and 7 are expected to show the same results as a rod without
the hole. 9 and 8 are expected to show the same results as 6 and 7 as they
are at the same position as 6 and 7 at the back.
The force applied to the plate stops increasing at 4500 N to ensure that the
rod is still in the elastic region, having stress proportional to strain. Therefore,
we can obtain the Stress Concentration Factor by the equation
σ maximum E × ε maximum ε maximum
SCF= = =
σ remote E × ε remote ε remote
( 6)
In the equation, the Young’s modulus (E) is cancelled out. This means that
finding the SCF this way is independent to the young’s modulus, hence the
material property. Therefore aluminium is used in the experiment instead of
steel. As aluminium is less stuff than steel. With the same amount of force
applied, aluminium elongates more, resulting in a smaller percentage error, as
the error is an absolute fixed error.

4
 3)Result
All calculations and uncertainties are in Appendix B and C respectively.

a) Original data
The F c is calculated to be 24700 N. With that, the operating stress (σ op) is
calculated to be 30.875 MPa. σ opwill also be the same as σ remoteof the modified
rod.

b) Materials test
From the data collected in the tensile test, a stress strain curve can be plotted.

600

500

400 UTS(474)
Stress (MPa)

300

200 Yield Stress

(304)
100
00
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
00
0
0
-5 0 5 10 15 20 25 30 35 40
-100

Strain (%)

Figure 3. Stress strain curve of tensile test using extensometer data

Note that the strain is from the extension of the gauge length measured by the
video extensometer instead of the displacement of the crosshead as the grips,
load cell and crosshead extends as well, causing the displacement of the
crosshead which accounts for all the above to be inaccurate.
The yield stress is the minimum point of the Lüders band (red dot) and the
ultimate tensile stress (orange dot) is the maximum point of the graph. The
yield stress (σ y) and ultimate tensile stress (σ UTS ¿is 303 ± 3.04 MPa and
474 ± 4.75 MPa respectively. The fatigue stress (σ f ¿ is calculated to be
316 ± 3.17 MPa.
The failure strain reported is actually an estimate since it is the engineering
strain and not the true strain. The reported failure strain is 19.4062 % ± 0.39 %.
The true strain can be calculated by
ε t=ln (1+ ε )=17.7361 %
(7 )
The difference can be explained by the assumption of engineering stress. It is
assumed the cross-section area remains the same throughout the
5
experiment. The stress is calculated using the original cross-section area. But
as the material elongates, the cross-section area decreases, causing the true
stress to be larger than engineering stress at the same value of strain.
Therefore the actual strain at the point of fracture is less than the one reported
in the stress strain curve.

400

f(x) = 2200.47442973443 x + 22.2916101019821

350
f(x) = 163.18931910492 x − 53.2307123281806
300

250
Stress (MPa)

200

150

100

50

0
-0.5 0 0.5 1 1.5 2 2.5 3

Strain (%)

Figure 4. Stress strain curve of elastic region of tensile test of data from
crosshead and extensometer
The above graph shows the elastic region of the stress strain curve with
extension data from extensometer and crosshead. With linear regression, the
data follows a linear pattern, aligning with the properties of elastic elongation.
With the gradient of the regression being Young’s modulus. Which is 220 GPa
for the extensometer data and 16.3 GPa for the crosshead data. Comparing
to Young’s modulus of steel from Engineering toolbox, which is 200 GPa.
Using the extension measured by the extensometer is shown to have a much
higher accuracy than measuring the extension from the crosshead.
With the the σ y and σ UTS obtained, the σ f can be calculated, and Safety
Factors can then be calculated by
σy
SF y =
σ op
(8)

6
 σf
SF f =
σ op
(9)

To be SF y =9.82 ±0.0984 and SF f =10.2 ± 0.1022

c) Geometric Test
The SCF is obtained to be 3.41 ± 0.154
The safety factor after drilling the hole can be calculated
σy
SF y = =2.88 ± 0.159
σ remote × SCF
( 10 )
σf
SF f = =2.99 ± 0.165
σ remote × SCF
( 11)
There are two measures to check for the accuracy of the geometric test. One
is to find the Young’s modulus of aluminium using the data in the experiment
using eqn.
F
A
E=
4 V remote
GF × BV
(12)
Young’s modulus of aluminium is calculated to be 71.4 ± 5.88 GPa. Compared
to Young’s modulus from the engineering toolbox (The Engineering ToolBox,
2003), which is 69 GPa, the two values are similar with only a 3.48%
difference. Therefore suggesting the result from the geometric test to have a
high accuracy.
The other measure is to compare the data from strain gauges 3, 4 and 5 with
the FEA to see if it complies.

7
 3.5
Exper-
3.0 imen-
tal
2.5
2.0
Stress ratio

1.5
1.0
0.5
0.0
0 2 4 6 8 10121416182022242628
Distance to the circumference of the hole[mm]

Figure 5. Comparison of experimental data and finite element analysis

In the graph, the four experimental datum are stress ratios at maximum strain
region, strain gauges 3, 4 and 5 from left to right. A downward trend in stress
ratio with distance to circumference of hole increasing can be observed.
However, the FEA data does not lie between the error bars of the
experimental data. Showing the inaccuracy of the experimental data with a
tendency to have a higher stress ratio than the finite element analysis.

4)Discussion
With the safety factor of both yield of fatigue obtained, it is possible to
determine the safety of the ride after drilling the hole. In the Review of the
Design Review Process for Fairground Rides:
As a baseline, factors of at least 2 on life are normally required for
components with any degree of criticality, and larger margins up to 5 for more
critical elements.
It would be beneficial to identify a "criticality level" for each ride. This is a
normal approach in most industries. It would be based on the potential for
major injuries or multiple fatalities.
(Packer et al., 2007, p.14)
In this experiment, many assumptions were made to simplify the experiment,
8
ignoring real life conditions. Including corrosion, air resistance, strength of
joints, friction at joints, etc. Therefore a larger value safety factor should be
required.
Also considering the failure of the fairground ride would cause major injuries
to the riders and bystanders, the requirement of 5 in SF f should be taken.
Therefore the modification of the fairground ride should not be propagated
considering the SF f of the ride with the hole is only 2.99 ± 0.165.
This is aligned with the expectations. As the safety factor reduces by the
factor of SCF after the modification, the ride is not expected to be safe with
the hole in place with the SF f reducing by a factor of 3.41(SCF).
Not only does the experiment have limitations on evaluating the real situation,
but the experiment itself has limitations evaluating accurately as well.
Inspecting the data from the geometric test, the voltage at 6 and 8, 7 and 9 is
not the same. With 8 and 9 having a smaller value compared to 6 and 7. This
suggests that the force is not applied vertically. Possible reasons are the rod
being tilted, manufacture error for the rod, density not being uniform across
the volume, and rod bending during the experiment. This would explain the
differences in thickness along the vertical of the rod and the variations in
voltages recorded.
The experimental result have higher stress ratio than the FEA. Especially
when the experimental data is from a point further away from the centre. This
suggests a high possibility of the rod being tilted, with the side of strain gauge
5 and 6 under more stress.
Not only is there data errors, there is also many instrumental errors appear,
below is a list of them:
Strain gauge instrument error: ± 2%
Load cell instrument error: ± 1%
Micrometer instrument error: ± 0.005 mm
Calliper: ± 0.01 mm
Ruler: ± 1 mm
Extensometer: ± 1% Gauge Factor: ± 5%
For uncertainty analysis, see Appendix C.

5)Conclusion
This research investigated the safety of drilling a hole in the connecting rod of
a fairground ride.
Two experiments were conducted, the materials test and the geometric test.
Results of the materials test shows that, σ y =303 ±3.04 MPa, σ UTS=474 ± 4.75
9
MPa. σ f is calculated to be 316 ± 3.17 MPa. In the geometric test, the SCF is
found to be 3.41 ± 0.154. With the results of the two tests, the safety factor of
both before and after drilling the hole in the connecting rod was found. With
SF y =9.82 ±0.0984 and SF f =10.2 ±0.1022 before the modification. And
SF y =2.88 ±0.159 and SF f =2.99 ± 0.165 after the modification. Including errors
from various sources, instrumental errors, machine errors, etc. With reference
to the health and safety executive review on design review process of
fairground rides, the modification can be deemed unsafe, due to the safety
factor of fatigue being lower than 5.

Acknowledgements
Graduate Teaching Assistants
Idris Kevin Mohammed

List of References
Forrest, P.G. (1962) Fatigue of Metals, Oxford, Pergamon Press.
Packer, R., Commandeur, B., B., Linkens, D. & Ward. I. (2007) A review of
the design review process for fairground rides, Tech. Rep., Atkins Limited for
the Health and Safety Executive, Epsom, Surrey, UK.
Mohammed, I. (2024) Case Study Brief: Safety of a Fairground Ride, London:
Department of Mechanical engineering, Imperial College London
The Engineering ToolBox (2003). Young's Modulus, Tensile Strength and
Yield Strength Values for some Materials. [online]
Available at: https://www.engineeringtoolbox.com/young-modulus-d_417.html
[13/11/2024].Bibliography

10
Appendices
Appendix A
This appendix will show the raw data obtained from the experiment
Table 1. Raw data of measurements of diameter of sample in materials test
Diameter 1 7.98
Diameter 2 7.97
Diameter 2 7.96
Average diameter 7.97

Table 2. Raw data of measurements of dimensions of rod in geometric test

Thickness 1 Thickness 2 Thickness 3 Average thickness
(mm) (mm) (mm) (mm)
10.75 10.55 10.32 10.54
Width 1 Width 2 Width 3 Average width
(mm) (mm) (mm) (mm)
80 80 80 80

Table 3. Geometric Test raw data

Force Force Force Average
reading 1 reading 2 reading 3 Force (N)
(N) (N) (N)
F 4481.1 4499.3 4497.0 4492.5
Strain Voltage Voltage Voltage Average Averages
Gauge reading reading reading Voltage (mV)
Number 1(mV) 2(mV) 3(mV) (mV)
1 0.325 0.320 0.323 0.323 V maximum :
2 0.351 0.341 0.343 0.345 0.334
3 0.222 0.216 0.213 0.217
4 0.151 0.153 0.148 0.151
5 0.138 0.138 0.136 0.137
6 0.110 0.108 0.108 0.109 V remote :
7 0.106 0.104 0.103 0.104 0.098
8 0.092 0.095 0.090 0.091
9 0.087 0.089 0.087 0.088

Appendix B
This appendix will show the derivation and calculation of values in the report.
Original operating stress (σ op)

11
 2π
ω=30 RPM=30× =π rad /s
60
2
F c =mω r
( 13 )
2
F c =500 × π × 5=24700 N
F 24700
σ op= = =30.875 MPa
A 80 × 10× 10−6
(14 )

Fatigue Stress
The fatigue stress formula is an acceptable engineering estimate for zero-
tension loading of most steels (Forrest,1 1962, p. 212)
Derivation of True stress eqn(7)
l l
ε t=∫ d ε =∫
l0 l0
dl
l
=ln
l
l0 ()
=ln ( 1+ ε )

( 15 )
Calculation of cross section area of rod in geometric test

A=10.54 ×80 ¿ 843.2 mm2

Calculation of SCF
For the V maximum , the average V of strain gauges 1 and 2 is taken. And for V remote
, the average V of strain gauges 6, 7, 8 and 9 is taken.
V maximum
SCF= 0.334
V remote ¿
0.098
( 16 )
Calculation of Young’s modulus (E) of aluminium from experimental data in
geometric test
F 4492.5
σ
E= A 843.2
ε¿ ¿ ¿ 71.4 GPa
4 V remote 4 ×0.098
( 17 )
GF × BV 2.5 ×2.1

Appendix C
This appendix will show the calculations of uncertainties in the experiment.
The uncertainty of the average value is calculated by the formula
12
 Maximum possible value−Minimum possible value
umber of measurements
(18 )
Materials Test
The micrometer used to measure the diameter of the steel sample has an
instrument error of ± 0.005 mm. The uncertainty of the diameter measured
( 7.98+ 0.005 )−( 7.96−0.005 )
¿± ¿ ± 0.0001 mm
3
The uncertainty of the cross section area
0.0001
¿± ×2
7.97
¿ ± 0.00251 %

Hence the uncertainty of All σ y , σ UTS , σ f , SF y ∧SF f

¿ ± 1 %+0.00251 %¿ ± 1.00251 %

Uncertainty of 𝜀
± 1 %+1 %=± 2 %
Geometric Test
The calliper used to measure the thickness of the rod has an uncertainty of
± 0.01 m. The uncertainty of the average value
( 10.75+ 0.01 )−(10.32−0.01)
¿± ¿ ± 0.0001 mm
3
The ruler used to measure the width of the rod has an uncertainty of ± 1 mm.
The uncertainty of the average value
( 80+ 1 )−( 80−1 )
± ¿ ± 0 mm
3

The uncertainty of the cross-section area of the rod

0.015 0
± ×100 % + × 100 %¿ ± 0.142 %
10.54 80
The load cell has an uncertainty of ± 1 %. The uncertainty of the force applied
to the rod
( 4499.3+4499.3 × 1% )− ( 4481.1−4481.1× 1 % )
¿± ¿ ± 36.0 N
3
The strain gauge has an instrument error of ± 2 %. The uncertainty of V maximum

13
 ( 0.351+ 0.351× 2 % )− ( 0.320−0.320 × 2% ) 0.00740
¿± ¿ ± 0.00740 mV ¿ ± ×100 %
6 0.334
¿ ± 2.22 %

The uncertainty of V remote

( 0.110+0.110× 2 % ) −( 0.087−0.087 ×2 % ) 0.002245

¿± ¿ ± 0.002245 mV ¿ ± ×100 %
12 0.098
¿ ± 2.29 %

The uncertainty of V 3
( 0.222+ 0.222× 2 % )− ( 0.213−0.213 × 2% )
¿± ¿ ± 5.009 %
( 0.217 ) ×3

The uncertainty of V 4
( 0.153+ 0.153 ×2 % )−( 0.148−0.148 ×2 % )
¿± ¿ ± 4.728 %
( 0.151 ) ×3

The uncertainty of V 5
( 0.138+ 0.138 ×2 % )−( 0.136−0.136 ×2 % )
¿± ¿ ± 4.106 %
( 0.137 ) × 3

The gauge factor has an uncertainty of 5%. The absolute uncertainty

¿ ± 2.1× 5 %¿ ± 0.105

Finally, the uncertainty of Young’s modulus of aluminium calculated using the

data in the experiment can be calculated by

¿± ( 4492.5
36
+0.142 % +
0.002245
0.098
+5 % )¿ ± 8.23 %

And the uncertainty of the SCF

¿ ± 2.22 %+2.29 %¿ ± 4.51 %¿ ± 3.41× 4.51 %¿ ± 0.154

14
And the uncertainty of SF y

¿ ± 4.51 %+ 1.00251%¿ ± 5.51251 %¿ ± 0.159

And the uncertainty of SF f

¿ ± 5.51251 %¿ ± 0.165

15