Evaluation of The Thermal Impact From Battery Packs From Electrical Vehicles in Underground Mining Environment
Evaluation of The Thermal Impact From Battery Packs From Electrical Vehicles in Underground Mining Environment
Evaluation of The Thermal Impact From Battery Packs From Electrical Vehicles in Underground Mining Environment
net/publication/332980968
Evaluation of the Thermal Impact from Battery Packs from Electrical Vehicles in
Underground Mining Environment
CITATIONS READS
2 1,407
1 author:
Daniel de Almeida
Laurentian University
1 PUBLICATION 2 CITATIONS
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Evaluation of the thermal impact from battery packs from electrical vehicles in underground mining environment View project
All content following this page was uploaded by Daniel de Almeida on 10 May 2019.
Mining Environment
by
Laurentian University
Sudbury, Ontario
Title of Thesis
Titre de la thèse EVALUATION OF THE THERMAL IMPACT FROM BATTERY PACKS FROM
ELECTRICAL VEHICLES IN UNDERGROUND MINING ENVIRONMENT
Name of Candidate
Nom du candidat De Almeida, Daniel
Degree
Diplôme Master of Science
APPROVED/APPROUVÉ
Dr. JA Scott
(Committee member/Membre du comité)
Approved for the Faculty of Graduate Studies
Approuvé pour la Faculté des études supérieures
Dr. David Lesbarrères
Monsieur David Lesbarrères
Dr. Jie Liu Dean, Faculty of Graduate Studies
(External Examiner/Examinateur externe) Doyen, Faculté des études supérieures
I, Daniel De Almeida, hereby grant to Laurentian University and/or its agents the non-exclusive license to archive
and make accessible my thesis, dissertation, or project report in whole or in part in all forms of media, now or for the
duration of my copyright ownership. I retain all other ownership rights to the copyright of the thesis, dissertation or
project report. I also reserve the right to use in future works (such as articles or books) all or part of this thesis,
dissertation, or project report. I further agree that permission for copying of this thesis in any manner, in whole or in
part, for scholarly purposes may be granted by the professor or professors who supervised my thesis work or, in their
absence, by the Head of the Department in which my thesis work was done. It is understood that any copying or
publication or use of this thesis or parts thereof for financial gain shall not be allowed without my written
permission. It is also understood that this copy is being made available in this form by the authority of the copyright
owner solely for the purpose of private study and research and may not be copied or reproduced except as permitted
by the copyright laws without written authority from the copyright owner.
ii
iii
iv
Abstract
One of the main aspects which governs the size of ventilation facilities in underground mines
is the amount of heat load generated in the underground environment. This heat load comes from many
different sources, one of which is the heat contributed by underground diesel machinery operation.
One strategy to mitigate the heat and other emissions from such equipment is to substitute these units
This study presents a heat load evaluation of the Lithium-iron Phosphate battery system used
in a prototype electric mining vehicle. The set of equations which governs the heat generation from
these devices have been developed by previous researchers and is used in this thesis to calculate the
heat generation and loss. However, in the mining industry, the current methodology for heat load
calculation from electric vehicles (EVs) is usually based on the rated power or on a simple power loss
equation. This strategy might lead to incorrect estimations of the heat load from this type of machinery.
Experimental and simulation work has been conducted as a means to evaluate the heat flux
from the Lithium-iron Phosphate battery system. The battery was tested through charging and
discharging it under different levels of current within the 10% to 90% range of its maximum capacity.
The test was performed firstly with a single cell and then with a module. Furthermore, the battery
system was set in operation under different environment temperature settings. These current and
temperature levels represent the range of possible conditions in which the prototype will face in service.
Through the estimation of the heat released from the other main electrical components in the vehicle,
it was possible to calculate the heat impact of these units in the surrounding environment.
v
Acknowledgements
I would like to express my deep gratitude towards my thesis supervisor, Dr. Markus Timusk
of the Bharti School of Engineering at Laurentian University, who provided me with assistance,
encouragement, and inspiration to pursue the required knowledge to develop this thesis. This was a
great opportunity to develop myself intellectually and professionally under his guidance. Without his
I would also like to thank John Le, from Tracks and Wheels, who opened the doors of his
company to me, and Dr. Seyed Mahdi Mousavi, who helped me with each step of my research. Both
individuals provided me with invaluable assistance, which was essential for the development of this
Finally, I would also like to acknowledge the Laurentian University staff, especially Dr.
Ramesh Subramanian, to Damien Duff from Centre for Excellence in Mining Innovation – CEMI, and
again, to Dr. Markus Timusk, for their financial and academic support.
vi
Table of Contents
Abstract v
Acknowledgements vi
Table of Contents vii
Figures and Tables x
1 Introduction 1
1.1 Mobile Machinery in Underground Mining 2
1.2 Scope 3
1.3 Objectives and Thesis Structure 4
2 Background Work 5
2.1 Overview 5
2.2 Mine Ventilation and Mining Mobility 6
2.3 Heat Sources Underground 8
2.4 Propulsion of Machinery in Underground Mines 11
Diesel Internal Combustion (IC) Propulsion 11
Hybrid Diesel-Electric Propulsion 12
Hydrogen Fuel Cell Propulsion 13
Full Electrical Propulsion 15
2.5 Previous Evaluations of Heat Load applied to Diesel and Electric Vehicles 20
2.5.1 Machinery Heat Load Estimation based on Efficiency Calculations 20
2.5.2 Machinery Heat Loss based on Power Consumption and Psychometric Conditions 22
2.5.3 Machinery Heat Loss based on Infrared Thermo Images 24
2.5.4 Estimation of the Heat Generation from Components in a BEVs 25
3 Experimental Set-up 28
3.1 Introduction 28
3.2 Physical Setup 29
3.2.1 Chamber 31
3.2.2 Ceramic Support Bricks 32
3.3 Battery Control Units 32
3.3.1 Battery Balance Charger/Discharger 32
3.3.2 Resistor Load Bank 34
3.4 Battery Systems 36
vii
3.4.1 Battery Unit 37
3.4.2 Internal Structures 40
3.5 Instrumentation and Control 43
3.5.1 Battery Management System (BMS) 44
3.5.2 Data Logging System 44
3.5.3 Anemometer – TSI/Alnor 9545 VelociCalc Air velocity meter 47
3.5.4 Thermal Camera 48
3.6 Experimental Procedure 50
3.7 Experimental Results 52
3.8 Discussion 57
4 Mathematical Treatment 59
4.1 Overview 60
4.2 Physical Properties of the Core Domain 61
4.3 Heat Generation from the Core Domain 62
4.4 Collector Domain 64
4.5 Air Gap Domain 64
4.6 Boundary Conditions 65
4.7 Natural Convection 66
4.8 Forced Convection 67
5 Simulation and Experimental Results 69
5.1 Overview 69
5.2 Methodology 70
5.3 Single Cell Analysis 70
5.3.1 Important Considerations about the Simulation and the Experiment 72
5.3.2 Uphill Scenario 73
5.3.3 Flat scenario 79
5.3.4 Charging scenario 81
5.3.5 Downhill scenario 85
5.3.6 Single cell – Conclusion 89
5.4 Module Analysis 93
5.5 Estimation of Overall Machinery Heat Generated 98
6 Conclusions 109
7 References 112
viii
8 Appendix A – Single Cell Study Results 124
8.1 Charging – 25°C 125
8.2 Charging – 30°C 128
8.3 Charging – 35°C 130
8.4 Downhill – 25°C 132
8.5 Downhill – 30°C 134
8.6 Downhill – 35°C 136
8.7 Flat – 25°C 138
8.8 Flat – 30°C 140
8.9 Flat – 35°C 142
8.10 Uphill – 25°C 144
8.11 Uphill – 30°C 146
8.12 Uphill – 35°C 148
9 Appendix B – Module Case Study Results 150
9 Charging – 25°C 150
9.1 Charging – 30°C 153
9.2 Charging – 35°C 155
9.3 Charging – 40°C 157
9.4 Downhill – 25°C 159
9.5 Downhill – 30°C 161
9.6 Downhill – 35°C 163
9.7 Downhill – 40°C 165
9.8 Flat – 25°C 167
9.9 Flat – 30°C 169
9.10 Flat – 35°C 171
9.11 Flat – 40°C 173
9.12 Uphill – 25°C 175
9.13 Uphill – 30°C 177
9.14 Uphill – 35°C 179
9.15 Uphill – 40°C 181
ix
Figures and Tables
Figure 1-1: Electric mine truck manufactured by Tracks and Wheels, Commander 5EV (Tollinsky, 2017).
...................................................................................................................................................................... 1
Figure 1-2: Energy Flow Diagram for Diesel Mobile Equipment. ............................................................... 2
Figure 1-3: Thesis methodology block diagram. .......................................................................................... 5
Figure 2-1: Energy Cost Comparison between the Total Mine Operation and the Mine Ventilation
(adapted from CIPEC, 2005). ....................................................................................................................... 7
Figure 2-2: Typical heat load underground mines depending on each source (adapted from Vergne, 2008).
...................................................................................................................................................................... 8
Figure 2-3: Heat sources founded in Craighton underground mine (adapted from Enderlin, 1973). ........... 9
Figure 2-4: Average heat sources from August – 1977 to March - 1978 founded in Blyvooruitizicht
underground mine (adapted from Deglon P. and Hemp R., 1980). ............................................................ 10
Figure 2-5: Major components in a fuel-cell vehicle (adapted from Fuel Cell Research Lab, nd). ............ 14
Figure 2-6: Heat generation impact from electric machinery (adapted from McPherson, 1986). ............ 21
Figure 2-7: Energy input destination for a combined cycle in combustion vehicles (US Department of
Energy, n.d.)................................................................................................................................................ 26
Figure 2-8: Energy input destination for a combined cycle in electric vehicles (US Department of Energy
n.d.). ............................................................................................................................................................ 26
Figure 3-1: Schematic for the experimental set-up. .................................................................................... 28
Figure 3-2: Battery heat loss experiment set-up for a single cell inside the chamber. ................................ 30
Figure 3-3: Battery heat loss experiment set-up. ........................................................................................ 30
Figure 3-4: Front view of the industrial oven. ............................................................................................ 31
Figure 3-5: EFUEL PFC Heavy Duty 1200W/max 50A Power Supply. .................................................... 33
Figure 3-6: Balance Charger/Discharger. ................................................................................................... 33
Figure 3-7: Resistor Load Bank. ................................................................................................................. 34
Figure 3-8: Arrangement of resistors for single cell analysis under Flat Condition. .................................. 35
Figure 3-9: Resistor arrangement for single cell analysis under Uphill condition. ..................................... 35
Figure 3-10: Arrangement of the resistors for a module analysis under Flat Condition. ............................ 36
Figure 3-11: Resistors arrangement for module analysis under Uphill Condition. .................................... 36
Figure 3-12: Discharge curves for CALB CAM 72Ah at 1C, 0.5C, and 0.3C (CALB, 2014). .................. 37
Figure 3-13: Single cell CALB CAM 72 Ah external view (CALB, 2014). ............................................. 39
Figure 3-14: One module of the battery pack with 6 cells with the plastic frame used in the EV. ............. 39
Figure 3-15: Internal components of the battery cell CALB CAM 72Ah. The opening stages can be seen
from picture (a) to (c). ................................................................................................................................. 40
Figure 3-16: Internal Components of the battery core. The core is exposed in (a), and its contents in (b),
and (c). ........................................................................................................................................................ 41
Figure 3-17: Battery core internal Layers. (a) – Positive Electrode, (b) – Negative Electrode, (c) –
Separator, and (d) – Exterior Polyethylene / Contact Layer. ...................................................................... 41
Figure 3-18: Internal Battery view – Schematic (H x L x D – 216.8 x 135.2 x 29.2 mm). ........................ 42
Figure 3-19: Terminal representation locate inside the air gap................................................................... 43
x
Figure 3-20: BMS Jr from Orion. ............................................................................................................... 44
Figure 3-21: PXIe-1082 chassis with NI TB-4353 front mounting terminal block attached. ..................... 45
Figure 3-22: 10 K-type thermocouples location around the battery cell. Schematic not in scale. .............. 46
Figure 3-23: K-type thermocouples around the battery module. Schematic is not in scale. ....................... 46
Figure 3-24: Setup for thermocouples and voltage tap connected around the battery cell. ........................ 47
Figure 3-25: Anemometer – TSI/Alnor 9545 VelociCalc and its probe stick (TSI, n.d.). .......................... 48
Figure 3-26: Thermal Camera – FLIR E5 (FLIR, n.d.). ............................................................................. 49
Figure 3-27: Experiment, battery surface temperature evolution, Uphill single-cell case. ......................... 53
Figure 3-28: Experiment, battery surface temperature evolution, Uphill module case. ............................. 53
Figure 3-29: Experiment, battery surface temperature evolution, Flat single-cell case. ............................. 54
Figure 3-30: Experiment, battery surface temperature evolution, Flat module case. ................................. 54
Figure 3-31: Experiment, battery surface temperature evolution, Charging single-cell case. .................... 55
Figure 3-32: Experiment, battery surface temperature evolution, Charging module case. ......................... 55
Figure 3-33: Experiment, battery surface temperature evolution, Downhill single-cell case. .................... 56
Figure 3-34: Experiment, battery surface temperature evolution, Downhill module case.......................... 56
Figure 3-35: Temperature Coefficient curve adapted from (Forgez et al., 2010). ...................................... 57
Figure 4-1: Battery subdomains and special parameters............................................................................. 60
Figure 4-2: Effective thermal conductivity for parallel (a), and series (b) elements (Chen; Wan; Wang,
2005). .......................................................................................................................................................... 61
Figure 4-3: Battery faces schematic. ........................................................................................................... 66
Figure 5-1: Block diagram of the FEA model. ........................................................................................... 70
Figure 5-2: Mesh Distribution for the CAD single cell model. .................................................................. 71
Figure 5-3: Final visual simulation of the temperature distribution along the external battery surface for
44.2A discharging. Total running time of 78 minutes. ............................................................................... 74
Figure 5-4: Experimental temperature distribution on the external battery surface for 44.2A discharging.
Total running time of 78 minutes................................................................................................................ 75
Figure 5-5: Experimental Heat generated and lost during battery discharge for the Uphill condition. ...... 76
Figure 5-6: Simulated temperature and Experimental temperature obtained at the centre of the front face
during Uphill discharging at 40°C. ............................................................................................................. 77
Figure 5-7: Voltage and Open-voltage during Uphill discharging at 40°C. ............................................... 78
Figure 5-8: Heat generation Comparison between the Ohmic Heating term and the Reversible Entropic
term for Uphill 40°C. .................................................................................................................................. 78
Figure 5-9: Final visual simulation of the temperature distribution along the external battery surface for
16.9A discharging. Total running time of 205 minutes. ............................................................................. 79
Figure 5-10: Experimental temperature distribution on the external battery surface for 16.9A discharging.
Total running time of 205 minutes. ............................................................................................................. 80
Figure 5-11: Experimental Heat generated and lost during battery discharge for the Uphill condition. .... 80
Figure 5-12: Simulated temperature and Experimental temperature obtained at the centre of the front face
during flat discharging at 40°C. .................................................................................................................. 81
Figure 5-13: Final visual simulation of the temperature distribution along the external battery surface for
16.7A discharging. Total running time of 195 minutes. ............................................................................. 82
xi
Figure 5-14: Experimental temperature distribution on the external battery surface for 17.8A discharging.
Total running time of 195 minutes. ............................................................................................................. 83
Figure 5-15: Experimental heat generated and lost during battery charging for the charge condition. ...... 84
Figure 5-16: Voltage and Open-voltage during charging at 40°C. ............................................................. 84
Figure 5-17: Simulated temperature and Experimental temperature obtained at the centre of the front face
during charging at 40°C. ............................................................................................................................. 85
Figure 5-18: Final visual simulation of the temperature distribution along the external battery surface for
31.0A charging. Total running time of 114 minutes. .................................................................................. 86
Figure 5-19: Experimental temperature distribution on the external battery surface for 31.0A charging.
Total running time of 114 minutes. ............................................................................................................. 87
Figure 5-20: Experimental Heat generated and lost during battery charging for the Downhill condition. . 88
Figure 5-21: Simulated temperature and Experimental temperature from the centre of the front face during
downhill charging at 40°C. ......................................................................................................................... 88
Figure 5-22: Average current and pack heat generated during Charging mode. ........................................ 90
Figure 5-23: Average current and pack heat generated during Downhill mode. ........................................ 90
Figure 5-24: Average current and pack heat generated during Flat mode. ................................................. 91
Figure 5-25: Average current and pack heat generated during Uphill mode. ............................................. 91
Figure 5-26: Mesh Distribution for the CAD multi-cell model. ................................................................. 93
Figure 5-27: Temperature distribution in the meridional section of the module during Uphill operation at
40°C. ........................................................................................................................................................... 95
Figure 5-28: Experimental temperature distribution on the module during Uphill operation at 40°C. ...... 96
Figure 5-29: Experimental Heat generated and lost during module discharging calculated per cell for the
Uphill condition. ......................................................................................................................................... 96
Figure 5-30: Simulated temperature and Experimental temperature discharging at 40°C. ........................ 97
Figure 5-31: Simulation for the losses in the electric motor used in the Commander 5 EV during nominal
operation (~200 amperes), courtesy of the manufacturer. .......................................................................... 99
Figure 5-32: Motor controller heat loss curve. ......................................................................................... 100
Figure 5-33: Simulation for the losses in the hydraulic electric motor used in the Commander 5 EV during
nominal operation (~20 amperes), courtesy from the manufacture. ......................................................... 101
Figure 5-34: Estimation of the machinery average efficiency under Charging regime. ........................... 103
Figure 5-35: Estimation of the machinery average efficiency under Downhill regime. ........................... 103
Figure 5-36: Estimation of the machinery average efficiency under Flat regime. .................................... 104
Figure 5-37: Estimation of the machinery average efficiency under Uphill regime. ................................ 104
Figure 8-1: Final visual simulation of a temperature gradient along the external battery surface for 16.7A
charging. Total running time of 207 minutes. ........................................................................................... 125
Figure 8-2: Thermal picture from the battery at 207 minutes for 16.7A charging, 25°C ambient. .......... 126
Figure 8-3: Experimental Generated heat and Lost during battery charge for the charging condition. .... 126
Figure 8-4: Simulated temperature and Experimental temperature obtained during charging at 25°C. ... 127
Figure 8-5: Final visual simulation of a temperature gradient along the external battery surface for 16.5A
charging. Total running time of 210 minutes. ........................................................................................... 128
Figure 8-6: Thermal picture from the battery at 210 minutes for 16.5A charging, 30°C ambient. .......... 128
Figure 8-7: Experimental Generated heat and Lost during battery charge for the charging condition. .... 129
xii
Figure 8-8: Simulated temperature and Experimental temperature obtained during charging at 30°C. ... 129
Figure 8-9: Final visual simulation of a temperature gradient along the external battery surface for 18.4A
charging. Total running time of 188 minutes. ........................................................................................... 130
Figure 8-10: Thermal picture from the battery at 188 minutes for 18.4A charging, 35°C ambient. ........ 130
Figure 8-11: Experimental Generated heat and Lost during battery charge for the charging condition. .. 131
Figure 8-12: Simulated temperature and Experimental temperature obtained during charging at 35°C. . 131
Figure 8-13: Final visual simulation of a temperature gradient along the external battery surface for 29.7A
Downhill. Total running time of 116 minutes........................................................................................... 132
Figure 8-14: Thermal picture from the battery at 116 minutes for 29.7A charging, 25°C ambient. ........ 132
Figure 8-15: Experimental Generated heat and Lost during battery charge for the Downhill condition. . 133
Figure 8-16: Simulated temperature and Experimental temperature obtained during Downhill at 25°C. 133
Figure 8-17: Final visual simulation of a temperature gradient along the external battery surface for 28.4A
Downhill. Total running time of 122 minutes........................................................................................... 134
Figure 8-18: Thermal picture from the battery at 122 minutes for 28.4A charging, 30°C ambient. ........ 134
Figure 8-19: Experimental Generated heat and Lost during battery charge for the Downhill condition. . 135
Figure 8-20: Simulated temperature and Experimental temperature obtained during Downhill at 30°C. 135
Figure 8-21: Final visual simulation of a temperature gradient along the external battery surface for 29.8A
Downhill. Total running time of 116 minutes........................................................................................... 136
Figure 8-22: Thermal picture from the battery at 116 minutes for 29.8A charging, 35°C ambient. ........ 136
Figure 8-23: Experimental Generated heat and Lost during battery charge for the Downhill condition. . 137
Figure 8-24: Experimental Generated heat and Lost during battery charge for the Downhill condition. . 137
Figure 8-25: Final visual simulation of a temperature gradient along the external battery surface for 17.2A
discharging. Total running time of 200 minutes. ...................................................................................... 138
Figure 8-26: Thermal picture from the battery at 200 minutes for 17.2A discharging, 25°C ambient. .... 138
Figure 8-27: Experimental Generated heat and Lost during battery discharge for the flat condition. ...... 139
Figure 8-28: Experimental Generated heat and Lost during battery discharge for the flat condition. ...... 139
Figure 8-29: Final visual simulation of a temperature gradient along the external battery surface for 16.9A
discharging. Total running time of 205 minutes. ...................................................................................... 140
Figure 8-30: Thermal picture from the battery at 205 minutes for 16.9A discharging, 30°C ambient. .... 140
Figure 8-31: Experimental Generated heat and Lost during battery discharge for the flat condition. ...... 141
Figure 8-32: Experimental Generated heat and Lost during battery discharge for the flat condition. ...... 141
Figure 8-33: Final visual simulation of a temperature gradient along the external battery surface for 16.9A
discharging. Total running time of 205 minutes. ...................................................................................... 142
Figure 8-34: Thermal picture from the battery at 205 minutes for 16.9A discharging, 35°C ambient. .... 142
Figure 8-35: Experimental Generated heat and Lost during battery discharge for the flat condition. ...... 143
Figure 8-36: Simulated temperature and Experimental temperature obtained during discharging at 35°C.
.................................................................................................................................................................. 143
Figure 8-37: Final visual simulation of a temperature gradient along the external battery surface for 44.3A
discharging. Total running time of 78 minutes. ........................................................................................ 144
Figure 8-38: Thermal picture from the battery at 78 minutes for 44.3A discharging, 25°C ambient. ...... 144
Figure 8-39: Experimental Generated heat and Lost during battery discharge for the Uphill condition. . 145
xiii
Figure 8-40: Simulated temperature and Experimental temperature obtained during discharging at 25°C.
.................................................................................................................................................................. 145
Figure 8-41: Final visual simulation of a temperature gradient along the external battery surface for 44.3A
discharging. Total running time of 78 minutes. ........................................................................................ 146
Figure 8-42: Thermal picture from the battery at 78 minutes for 44.3A discharging, 30°C ambient. ...... 146
Figure 8-43: Experimental Generated heat and Lost during battery discharge for the Uphill condition. . 147
Figure 8-44: Simulated temperature and Experimental temperature obtained during discharging at 30°C.
.................................................................................................................................................................. 147
Figure 8-45: Final visual simulation of a temperature gradient along the external battery surface for 43.7A
discharging. Total running time of 79 minutes. ........................................................................................ 148
Figure 8-46: Thermal picture from the battery at 79 minutes for 43.7A discharging, 35°C ambient. ...... 148
Figure 8-47: Experimental Generated heat and Lost during battery discharge for the Uphill condition. . 149
Figure 8-48: Simulated temperature and Experimental temperature obtained during discharging at 35°C.
.................................................................................................................................................................. 149
Figure 9-1: Final visual simulation of a temperature gradient along the external battery surface for 16.5A
charging. Total running time of 210 minutes. ........................................................................................... 150
Figure 9-2: Thermal picture from the battery at 210 minutes for 16.5A charging, 25°C ambient. .......... 151
Figure 9-3: Experimental Generated heat and Lost during battery charge for the charging condition. .... 151
Figure 9-4: Simulated temperature and Experimental temperature obtained during charging at 25°C. ... 152
Figure 9-5: Final visual simulation of a temperature gradient along the external battery surface for 16.4A
charging. Total running time of 210 minutes. ........................................................................................... 153
Figure 9-6: Thermal picture from the battery at 210 minutes for 16.4A charging, 30°C ambient. .......... 153
Figure 9-7: Experimental Generated heat and Lost during battery charge for the charging condition. .... 154
Figure 9-8: Simulated temperature and Experimental temperature obtained during charging at 30°C. ... 154
Figure 9-9: Final visual simulation of a temperature gradient along the external battery surface for 16.5A
charging. Total running time of 210 minutes. ........................................................................................... 155
Figure 9-10: Thermal picture from the battery at 210 minutes for 16.5A charging, 35°C ambient. ........ 155
Figure 9-11: Experimental Generated heat and Lost during battery charge for the charging condition. .. 156
Figure 9-12: Simulated temperature and Experimental temperature obtained during charging at 35°C. . 156
Figure 9-13: Final visual simulation of a temperature gradient along the external battery surface for 16.5A
charging. Total running time of 210 minutes. ........................................................................................... 157
Figure 9-14: Thermal picture from the battery at 210 minutes for 16.5A charging, 40°C ambient. ........ 157
Figure 9-15: Experimental Generated heat and Lost during battery charge for the charging condition. .. 158
Figure 9-16: Simulated temperature and Experimental temperature obtained during charging at 40°C. . 158
Figure 9-17: Final visual simulation of a temperature gradient along the external battery surface for 30.9A
discharging. Total running time of 112 minutes. ...................................................................................... 159
Figure 9-18: Thermal picture from the battery at 112 minutes for 30.9A charging, 25°C ambient. ........ 159
Figure 9-19: Experimental Generated heat and Lost during battery discharge for the Downhill condition.
.................................................................................................................................................................. 160
Figure 9-20: Simulated temperature and Experimental temperature obtained during discharge at 25°C. 160
Figure 9-21: Final visual simulation of a temperature gradient along the external battery surface for 31.0A
charging. Total running time of 111 minutes. ........................................................................................... 161
xiv
Figure 9-22: Thermal picture from the battery at 111 minutes for 31.0A charging, 30°C ambient. ........ 161
Figure 9-23: Experimental Generated heat and Lost during battery charge for the Downhill condition. . 162
Figure 9-24: Simulated temperature and Experimental temperature obtained during charging at 30°C. . 162
Figure 9-25: Final visual simulation of a temperature gradient along the external battery surface for 30.1A
charging. Total running time of 111 minutes. ........................................................................................... 163
Figure 9-26: Thermal picture from the battery at 111 minutes for 31.0A charging, 35°C ambient. ........ 163
Figure 9-27: Experimental Generated heat and Lost during battery charge for the Downhill condition. . 164
Figure 9-28: Simulated temperature and Experimental temperature obtained during charging at 35°C. . 164
Figure 9-29: Final visual simulation of a temperature gradient along the external battery surface for 31.0A
charging. Total running time of 111 minutes. ........................................................................................... 165
Figure 9-30: Thermal picture from the battery at 111 minutes for 31.0A charging, 40°C ambient. ........ 165
Figure 9-31: Experimental Generated heat and Lost during battery charge for the charging condition. .. 166
Figure 9-32: Simulated temperature and Experimental temperature obtained during charging at 40°C. . 166
Figure 9-33: Final visual simulation of a temperature gradient along the external battery surface for 16.8A
discharging. Total running time of 206 minutes. ...................................................................................... 167
Figure 9-34: Thermal picture from the battery at 206 minutes for 16.8A discharging, 25°C ambient. .... 167
Figure 9-35: Experimental Generated heat and Lost during battery discharge for the flat condition. ...... 168
Figure 9-36: Simulated temperature and Experimental temperature obtained during discharging at 25°C.
.................................................................................................................................................................. 168
Figure 9-37: Final visual simulation of a temperature gradient along the external battery surface for 16.8A
discharging. Total running time of 206 minutes. ...................................................................................... 169
Figure 9-38: Thermal picture from the battery at 206 minutes for 16.8A discharging, 30°C ambient. .... 169
Figure 9-39: Experimental Generated heat and Lost during battery discharge for the flat condition. ...... 170
Figure 9-40: Simulated temperature and Experimental temperature obtained during discharging at 30°C.
.................................................................................................................................................................. 170
Figure 9-41: Final visual simulation of a temperature gradient along the external battery surface for 16.8A
discharging. Total running time of 205 minutes. ...................................................................................... 171
Figure 9-42: Thermal picture from the battery at 205 minutes for 16.8A discharging, 35°C ambient. .... 171
Figure 9-43: Experimental Generated heat and Lost during battery discharge for the flat condition. ...... 172
Figure 9-44: Simulated temperature and Experimental temperature obtained during discharging at 35°C.
.................................................................................................................................................................. 172
Figure 9-45: Final visual simulation of a temperature gradient along the external battery surface for 16.2A
discharging. Total running time of 206 minutes. ...................................................................................... 173
Figure 9-46: Thermal picture from the battery at 206 minutes for 16.2A discharging, 40°C ambient. .... 173
Figure 9-47: Experimental Generated heat and Lost during battery discharge for the flat condition. ...... 174
Figure 9-48: Simulated temperature and Experimental temperature obtained during discharging at 40°C.
.................................................................................................................................................................. 174
Figure 9-49: Final visual simulation of a temperature gradient along the external battery surface for 41.1A
discharging. Total running time of 84 minutes. ........................................................................................ 175
Figure 9-50: Thermal picture from the battery at 84 minutes for 41.1A discharging, 25°C ambient. ...... 175
Figure 9-51: Experimental Generated heat and Lost during battery discharge for the Uphill condition. . 176
xv
Figure 9-52: Simulated temperature and Experimental temperature obtained during discharging at 25°C.
.................................................................................................................................................................. 176
Figure 9-53: Final visual simulation of a temperature gradient along the external battery surface for 41.1A
discharging. Total running time of 84 minutes. ........................................................................................ 177
Figure 9-54: Thermal picture from the battery at 84 minutes for 41.1A discharging, 30°C ambient. ...... 177
Figure 9-55: Experimental Generated heat and Lost during battery discharge for the Uphill condition. . 178
Figure 9-56: Simulated temperature and Experimental temperature obtained during discharging at 30°C.
.................................................................................................................................................................. 178
Figure 9-57: Final visual simulation of a temperature gradient along the external battery surface for 40.9A
discharging. Total running time of 85 minutes. ........................................................................................ 179
Figure 9-58: Thermal picture from the battery at 85 minutes for 40.9A discharging, 35°C ambient. ...... 179
Figure 9-59: Experimental Generated heat and Lost during battery discharge for the Uphill condition. . 180
Figure 9-60: Simulated temperature and Experimental temperature obtained during discharging at 35°C.
.................................................................................................................................................................. 180
Figure 9-61: Final visual simulation of a temperature gradient along the external battery surface for 40.9A
discharging. Total running time of 85 minutes. ........................................................................................ 181
Figure 9-62: Thermal picture from the battery at 85 minutes for 40.9A discharging, 40°C ambient. ...... 181
Figure 9-63: Experimental Generated heat and Lost during battery discharge for the Uphill condition. . 182
Figure 9-64: Simulated temperature and Experimental temperature obtained during discharging at 40°C.
.................................................................................................................................................................. 182
xvi
1 Introduction
The levels of safety and comfort for personnel are of significant concern in underground mines.
Internal combustion (IC) engines, such as those used in mobile mining equipment, emit a considerable
amount of waste heat, hazardous gases, and particulate matter into the environment. Once they have
been released into the working environment, mitigation of these emissions is mainly achieved through
costly operations, such as ventilation and refrigeration. As these costs continue to increase, the industry
(Tracks and Wheels of Sudbury, Ontario) which involves the design and production of an Electric
Vehicle (EV), referred to as the Commander 5EV, presented in Figure 1-1. This general-purpose utility
vehicle is designed to transport five passengers, including the driver, and has a cargo bed suitable for
a maximum payload of 6350kg (14000 lbs). The machine is powered by two electric motors (one per
axle), has a battery pack rated at 100kWh, and can provide a range and performance that is intended to
be acceptable to mining operators. One of the principal motivations for the development of the
Commander 5EV is to replace diesel-powered vehicles and thereby reduce emissions of contaminants
and heat in the underground environment, thereby providing a more economical and safer alternative.
Figure 1-1: Electric mine truck manufactured by Tracks and Wheels, Commander 5EV (Tollinsky,
2017).
1
In addition, the primary focus of this thesis is on the heat generation from EV’s, specifically
from battery systems. The study also estimates the heat generated from other components in EV’s
(electric motors, motor controllers, DC/DC converters, etc). Combining all sources of heat generation
from EV’s will then help in providing an overall estimate of heat load on the underground environment.
Since the 1960’s, diesel engines have been widely used for mobile equipment in underground
mining (McGinn, 2007). However, despite this historic presence, the mining industry is currently
looking for alternatives due to three main issues. The first issue is emissions of toxic gasses and
particulate matter which must be removed from the local environment. The second issue is related to
increased risk of overheating IC engines, which are approaching their operational limits in some of the
hottest operating environments currently experienced. The third concern is the heat load on mine
cooling systems from relatively inefficient IC engines. It is estimated that approximately two-thirds
of the energy input into an internal combustion engine is expected to be lost to the environment in the
form of waste heat (McPherson, 1993). A breakdown of the energy flow can be seen in Figure 1-2. In
comparison, the heat generated from electric vehicles is approximately three times lower when
compared to a regular internal combustion engine (McPherson, 1993). This is principally due to the
2
Currently, diesel engines still have a solid presence in the mines underground, however with
the enforcement of strict safety regulations, increasing commercial advantages of electric machinery,
1.2 Scope
This study focuses on the heat load of battery electric vehicles in underground mines, with
special attention on battery packs in electric vehicles. In relation to the other components, such as
electric motors, motor controllers, DC/DC converter, hydraulic pumps, air conditioner, and heater, the
heat load is calculated based on predictions and estimations. This means that predictions, and
estimations were made for different expected driven conditions, following information provided by the
In relation to the battery pack, the heat generation and loss are based on a well-known set of
equation. As for the heat generation, these calculations are specifically related to the chemistry used,
which in this case is Lithium-iron Phosphate. This chemistry is becoming the standard in mining
electric vehicle applications, due to its good balance between cost, reliability, safety, energy capacity,
and energy density. The heat loss is calculated differently, and it is based on the traditional method of
The set of tests conducted examines the heat magnitude and distribution over the exterior
surface of the battery for a single cell and for module with 8 multiple cells. The battery is tested under
different regimes of currents for discharging and charging, that simulates the demands that these units
are expected to face when in service in the underground mining environment. The results obtained
through the experimental phase are then compared to a simulation built in a finite-numerical software,
3
1.3 Objectives and Thesis Structure
The overall objective of this work is to gain an understanding of the heat generated from
electric vehicle operations in the context of underground mines. The specific focus being on the battery
systems in terms of heat loads. This objective was achieved through a review of the literature,
The literature review examines existing studies relating to the generation of heat from electric
vehicles in underground mines, focusing on battery systems. The section begins with the economic
aspects of mine ventilation, and their impact on the entire mining operation. Then, studies related to
underground heat sources are presented for different mine sites, with a focus on contribution of diesel
machinery. Additional studies related to alternative methods of underground propulsion are then
discussed. This section concludes with a review of relevant studies related to the emission of heat from
The experimental objective of this work is to estimate the heat generated by the battery system
of an electric mining vehicle under a range of different driving, charging conditions and ambient
temperatures. The experiments were followed by a series of simulations in order to replicate the
experimental results and validate assumptions of the equations and parameters that govern the heat
flux and heat generation process in battery systems. Two overall cases were modeled using a finite
element analysis tools, the single cell case and the multi-cell module. Both cases were simulated under
the same conditions presented during the experimental phase. Lithium-iron Phosphate for single cell
case, the 3D model was generated reflecting the internal geometric features of the real battery. For the
4
Figure 1-3: Thesis methodology block diagram.
2 Background Work
2.1 Overview
This section begins by exploring the main topic of mine ventilation and cooling. It then
progressively narrows down to the various contributors to the heat load in mines, sources of this heat
and in particular the contribution of mobile equipment propulsion methods for mining machinery. A
5
review of current studies about the heat impact on machinery in underground mines is presented in this
chapter. A brief review of academic work about the heat and emissions effects of diesel engines to
their surroundings, the alternative technologies for propulsion of mining machinery, methods for
predicting the heat loss from electric machines and battery systems is also presented.
The main task of mining ventilation is to dissipate air contaminants, provide fresh-air to the
miners, and lower the mine temperature. Needed to keep the work environment safe and comfortable
for underground miners, mine ventilation can often represent the highest operating expenses among
the mine operations, see Figure 2-1. This is especially true in the case of mines in tropical areas, deep
(hot) mines and in some sites in temperate climates (Vergne, 2008). Thus, any measure toward the
mitigation of air contaminants and the heat load in underground environments can result in
In a recent study conducted by the Canadian Federal Government (CIPEC, 2005), ten different
underground mining sites were surveyed in order to evaluate the cost contribution of their internal
operations. Figure 2-1 shows that mining ventilation represents, on average, to 37% of the total cost
per mass of ore hoisted. This result confirms the statement from Euler de Souza, a specialist in mine
ventilation, in which it is affirmed that the cost for ventilation can range from 25% to 40% of the total
energy cost, and it can represent from 40% to 50% of the total power consumption in a typical
underground mine (De Souza, 2015). It should be noted that these numbers are affected by a broad
range of factors.
6
Figure 2-1: Energy Cost Comparison between the Total Mine Operation and the Mine Ventilation
(adapted from CIPEC, 2005).
Mine ventilation is a vital component of and has considerable impact on mining operations.
Together with removal of heat, the main task of main ventilation is to dissipate air harmful
contaminants, such as the ones from diesel engines. In some cases, the removal of diesel emissions is
estimated to represent as much as 70% of the total air demand (Hartman et al., 1997). By replacing
diesel units by equivalent electrical versions (which do not have any gaseous or particulate emissions),
40% to 60% of ventilation demand was reduced (Varaschin J., 2016). This result is possible because
compared to IC units, electrical vehicles generate less heat (estimated in 40% of IC units), and do not
emit no harmful gases. Consequently, the total energy consumed by the fans was also decreased in
70% (Varaschin J., 2016). According to the same study, the implementation of electric units can lead
to savings as high as 20%, despite the fact that these units are, on average, 25% more expensive than
their diesel counterparts. Even though the introduction of electric units in underground environments
can represent a new paradigm in underground mines, mining ventilation is still needed to dissipate
7
2.3 Heat Sources Underground
The most significant sources of heat in underground mining environments include geothermal,
air flow losses, waste heat from machinery, and air auto-compression. The heat distribution and
magnitude vary from mine to mine. Some aspects to determine the impact of each heat source include
site location, depth, productivity, geothermal gradient, machine efficiency, type of material excavated
and size of excavated material. Figure 2-2 presents a typical distribution of heat loads in an
underground mine. This is a general assumption and does not represent a specific mining site.
However, it illustrates the general distribution of the heat. The heat sources are distributed among
several categories, in which “Others” is referred to Explosives, Ground Water, Backfill, Lighting, and
Personnel combined.
Figure 2-2: Typical heat load underground mines depending on each source (adapted from Vergne,
2008).
In a specific heat balance study (Enderlin, 1973) conducted in seven hot underground mines,
the average heat proportion obtained was 48% - geothermal, 20% - equipment, 19% - water, 11% -
auto-compression, and 2% - blasting and personnel. As a side note, in this study several devices are
included under the label of “Equipment”, such as: pumps, haulage motors, transform stations, fans,
8
underground hoists, electrical load centers, mechanical refrigeration units, and slushers. The exception
is the results expressed for Creighton in Sudbury, Ontario. In this mine, heat from diesel equipment is
expressed separately. The result for the contribution shares from each class of equipment among the
electrical/mechanical devices are 46% (diesel equipment), 36% (electrical fans), 11% (stationary
electrical load centers), and 7% (pumps). The impact of electro-mechanical equipment, pumps, fans,
electrical load centers and diesel equipment, on the total heat load generated from all sorts of heat
Figure 2-3: Heat sources founded in Craighton underground mine (adapted from Enderlin, 1973).
Company Limited, the average result for the heat distribution between August – 1977 and March 1978
can be seen in Figure 2-4. The results were obtained from measurements at the shaft longwall stope,
within an error margin between -6% and 22%. In this study, five major sources of heat were identified,
namely Rock, Electrical Power, Workers, Diesel equipment, and Explosives. The Electrical Power
category includes winches, fans, lighting, hoist, level pumps and the level cooling plant. It is important
to emphasize that each mining site is unique, and Blyvooruitzicht is no different. This because the heat
contribution of all electrical components surpasses in more than 13 times the heat released from diesel
9
units. The reason for this result is that this mine site employs in a large number high power electrical
machinery, especially in relation to the hoist, winches, and the fans, in detriment of diesel locomotives.
Figure 2-4: Average heat sources from August – 1977 to March - 1978 founded in Blyvooruitizicht
underground mine (adapted from Deglon P. and Hemp R., 1980).
In conclusion, the current section had provided information about the main sources of
underground heat. It was confirmed that, typically, geothermal, auto-compression and diesel machines
are the main sources of heat in this environment. However, the proportional contributions of each
source can vary significantly from site to site. Furthermore, it was shown that the mitigation of these
heat loads is beneficial to the mine operation since the cost with mine ventilation and cooling is
considerable. This cost is intrinsically correlated to the depth and productivity of the mine (Vergne,
2008). Hence, any mitigation of the heat sources is valid and needed to keep the operation profitable.
One of the current possibilities available to the industry is the replacement of thermally inefficient IC
10
2.4 Propulsion of Machinery in Underground Mines
In underground mining, there are many different types of mobile equipment, of which some of
the most commonly employed are load-haul-dump (LHD) vehicles, personnel carriers, locomotives,
and haul-trucks. Listed below are the power-trains currently available for these units for mining
applications:
Diesel Internal Combustion (IC) powertrains are the most common type of propulsion for
mining vehicles. Diesel IC units usually have longer range than the alternatives. This because, as a
fuel, diesel has a relatively high energy capacity when compared to other sources. Other advantages of
this technology include competitive cost of acquisition and operation, and compared to gasoline, diesel
is safer to store, making it the preferred alternative to gasoline as a fuel for mining vehicles.
Furthermore, refueling these vehicles is a quick and simple task. Nowadays, in order to be used in
underground mining applications, diesel engines must to comply to Tier 4 regulation for emissions.
These emissions can include carbon monoxide and dioxide, nitrogen and sulfur oxides, hydrocarbons,
and diesel particulate matter, in which the last one, is considered to be a carcinogenic substance (Burke
J., 2017). As a result, the emission of hazardous gases over time has been greatly reduced, however,
they cannot be eliminated. In addition, diesel engines not only emit toxic gases, but also considerable
amounts of heat to the environment, they generate excessive noise and vibration, and they have high
maintenance cost (these are complex machines, which require specialized mechanics). These negative
11
aspects are, to a certain degree, consequences of the low powertrain efficiency (30% to 35% for diesel
units). Lastly, the operation of diesel machines in underground environments represents a challenging
task in financial terms, since the oil prices tend to be considerably volatile (Paraszczak et al., a, 2014;
“Diesel to Electric: What is the Future of Mining?”, 2016; Wood A., n.d.).
Currently, several methods capable of decreasing diesel emissions are commercially available.
However, these technologies have limited impact on the pollution control, some are expensive to
implement, others reduce the machinery`s performance, and in some cases, they can even increase the
machinery heat load. For this reason, the mining industry is seeking for alternatives.
combining the diesel engine with an electric power-train, better mileage can be achieved, mostly
because the electric motor allows for the use of regenerative braking, and they have higher efficiency
then IC engines. Usually in hybrid technology, the IC engine can be used to charge the battery pack,
or to directly power the electric motor. This strategy allows the IC engine to run on the nominal
capacity, reducing the possibility of internal components wearing (Fleet M., 2012), maximizing its
performance and reducing the fuel consumption. Consequently, the machine durability and robustness
are improved, and the consumption of lubricants and fuel is reduced. As a result, even though these
machines have higher capital cost than regular diesel machines, their cost of operation is reduced
(Anon., 2014). It is estimated that the fuel savings are between 45% to 60% in comparison to regular
diesel mining machines (Norris J., 2013). Other sources claim that the fuel consumption reduction is
less dramatic, 20% reduction for hybrid vehicles using AC motor, and 60% for SR (Switched-
Reluctance) motors (Wood A., 2014), or even low as 15% (Demers et al, 2010). Conversely, the fuel
savings are strictly correlated to the type of machine, duty cycle, and production rate analyzed. The
12
reduce consumption of lubricants is also a result of less movable parts. The explanation for this is,
since electric motors have high torque at lower RPMs, there is no need for gear box and transmission.
Less movable parts results in a higher reliability overall, which in turn leads to higher productivity
because less time is spent during maintenance (Ozdogan, M., 2015). Due to the higher machine
efficiency overall, a small fraction of the energy input is wasted in the form of heat loss. Consequently,
less power is required to propel the vehicle, which allows for the diesel engine to be downsized
Fuel cell propulsion has been investigated as a viable alternative to diesel mining vehicles since
the early 2000`s. The initial studies started with the proposal of a locomotive, and later with the
development of the Caterpillar R1300 LHD (Barnes D. and Miller A., 2005). In the fuel cell
application, hydrogen is absorbed and safely stored in the proton-exchange membrane (PEM) fuel cell,
which is combined with the metal-hydride bed. This process requires low temperature and pressure.
When the PEM is heated, the hydrogen is oxidized by the oxygen in the air, generating electricity and
water in the process (Barnes D. and Miller A., 2005). The electrochemical reaction can be sustained as
long as hydrogen and oxygen are continuously provided. Also, a small battery pack might be needed
in order to store energy provided by the regenerative breaking if is available. See Figure 2-5 for a
13
Figure 2-5: Major components in a fuel-cell vehicle (adapted from Fuel Cell Research Lab, nd).
Compared to diesel IC machines, fuel cell vehicles do not emit pollutants and do not release
considerable amounts of heat to the environment. In comparison to electric vehicles, they are flexible,
capable of extended use, and possess a considerably fast refueling process. (Miller A. et al, 2012). In
mining applications relating to locomotives and LHDs, the heavy weight of the metal-hydride bed is
beneficial as a mean to provide more traction to the unit, whereas a negative aspect of fuel cells is the
The development of fuel cell technology faces numerous challenges in underground mining
applications. The process of absorption and release of hydrogen from the metal-hydride bed requires
close temperature control to minimize the volume of hydrogen leaked to the environment, which in the
underground mining ambient can be a safety and financial issue as hydrogen is a flammable and
expensive fuel. Usually, wasted heat from the battery and fuel cells are used during the warm up phase
in order to generate electricity. The thermodynamic efficiency of this process is around 50% (Valicek
P. and Fourie F., 2014). Furthermore, fuel cells need to be used under the proper temperature range,
since irreversible damaging can occur if they are freeze (“Fuel Cell Technology”, 2001). These
characteristics add complexity in the operation of fuel cell vehicles. One of the biggest disadvantages
14
of this technology is how to provide hydrogen fuel underground since it is flammable, volatile, difficult
to contain and store, and susceptible to leakage when refueling the unit.
In the end, hydrogen can’t be considered as a very efficient energy source when in a stored
medium. With electricity being easily available underground, the development of fast charging centers,
and higher energy power capacity batteries, hydride storage devices are becoming less common
compared to EVs.
In the next paragraphs, the three subcategories of underground EV’s commonly available at the present
time are examined. Herein, this class of vehicle includes battery electric vehicles (BEVs), power
cables, and overhead trolley lines. It has been estimated that the implementation of electric mining
vehicles can bring savings in the ventilation system of around 70% or more (Varaschin, 2016;
Paraszczak et al., b, 2014). This is mainly due to the absence of emissions, the use of regenerative
breaking (which also can be used in hybrid and fuel cell vehicles), and the higher efficiency of these
units in comparison to diesel propulsion. Hence, the efficiency of EVs is assumed to be three times
higher than diesel machines, which for the same application, can correspond to three times less heat
released by EVs (MacPherson, 2009; Paraszczak et al., b, 2014). In one study by (Varaschin, 2016), it
was assumed that the energy efficiency for a diesel IC unit was 37%, and for the EV, 92%. Being more
efficient also translates in a reduced power consumption, adding to the fact that electricity is cheaper
than diesel fuel (evaluated for the US market in the period between 1999 and 2014) (Varaschin, 2016),
which presents a supportive financial case for EVs. In comparison to diesel vehicles, mining EVs
consume 70% less energy, and there is an estimation that 30% of the total energy spent can be
recovered through the regenerative breaking (“Diesel to Electric: What is the Future of Mining?”, 2016;
15
Jensen, S., 2013). These are not the only benefit for EVs in general. They usually tolerate better
overloads and can achieve higher speeds of operation in relation to IC units, two parameters that have
BEVs offer the advantages such as these: no emissions of noxious gases and particulate matter,
good mobility, higher speeds when hauling, low maintenance, high torque available at a low RPM.
Unlike diesel engines and fuel cell technologies, BEVs don’t compete with the underground personnel
for the oxygen consumption. As a matter of fact, the use of regenerative breaking reduces the heat load
to the environment by redirecting the kinetic energy back to the battery pack (Paraszczak et al., b,
2014). However, their batteries have low energy capacity compared to diesel fuel. Lead acid batteries
usually have 40 W/kg of power, Lithium-iron phosphate batteries possess around 90 to 110 Wh/kg,
whereas some experimental batteries can reach up to 250 W/kg. In contrast, diesel fuel specific energy
is 13 000 W/kg (Paraszczak et al., a, 2014). The low energy capacity of batteries impacts on BEVs
having a considerable battery size in order to provide reasonable vehicle`s range. In a one reported
case, BEV Load Haul Dump (LHD) vehicles having 1.5-2 tons of batteries were able to be operated
between 2 to 2.5 hours, requiring 2 hours of charging time (Greenhill & Knights, 2013). In another
application, it was possible to identify that the BEV Haulmaster 800-20EB using LFP batteries was
capable of operating continuously for 4 hours, requiring a lower charging time of 1 to 2 hours
(Paraszczak et al., b, 2014). Nonetheless, the technology is improving over time. For instance, it is
feasible in economical and productive terms to convert the Caterpillar CL115 LHD from having a
diesel motor to being powered by an electrical battery. In the space available in the engine bay, it is
possible to place a battery pack that can last half a shift (in the case of lead acid technology), a single
shift (in the case of NaMx chemisty), and two shifts (for Li-ion batteries) without the need of
recharging (Schatz R. et al, 2015). Due to the fact that long charging times is one of the downsides of
battery electric vehicles, an alternative strategy would be to swap batteries after their usage, which is
16
a practice that takes about 15 minutes (Paraszczak et al., b, 2014). Even though the practice of battery
replacement can improve the productivity of BEVs, this strategy imposes an issue that extra batteries
need to be purchased, thus decreasing the financial advantages of purchasing these machines. As for
the battery chemistry, the decision lies on the total weight, energy capacity, and cost related to each
technology, especially for Li-ion technology, which can be considerably high. It is estimated that
switching from 4 to 8 hours battery capacity doubles its size, weight and price (Leonida, C., 2017). In
economic terms, it is expected that lead-acid BEVs are 25% more expensive their diesel counterparts,
and operational savings can range from marginal to 20% (Varaschin, 2016). The use of BEVs impose
some limitations. For instance, a dedicated charging station is required to charge batteries, and in case
of fast-charging stations, a significant investment might be needed. In addition to this, the battery pack
has a limited life expectancy of approximately 12 000 hours, which usually coincides with the life
Beside BEVs, underground electric vehicles used in mining can also be cable-powered (or
tethered) or trolley-powered, where the latter has a historical presence in underground mines, dating
back 30 years. Today, even though tethered electric LHDs account for 5% to 10% of the total
underground LHDs (Leonida C., n.d.), BEVs are the most common type of EVs (Jacobs, 2013).
Comparing a diesel with a tethered electric LHD (Sandvik LH514 and Sandvik LH514E) shows that
the electric version is 20% more expensive to purchase and 17% more expensive to maintain (Moore,
2010; Jacobs, 2013). The higher maintenance cost in this case is due to the frequently required cable
repairs, as seen in Table 2-1. Nevertheless, when the mine ventilation savings of using BEVs are
included to the financial calculation, the use of these units result in 30% cheaper operations (Jacobs,
2013). Both trolley and cable-powered units present limited mobility and flexibility and are usually
recommended for cases in which the machine operates for the most part in confined locations (such as
with the LHD cable-powered vehicle), or when a specific path is constantly used (as when a truck
17
trolley is used). In the case of the trolley technology, large investments in infrastructure are required,
which in turn reduces its applicability in underground mines. Usually these machines have small diesel
engines, which are used when the machines need to be relocated to different sections of the mine,
where the electrical power infrastructure is non-existent. In conclusion, it is expected that with the
current advancements in battery technologies, these technologies will soon be phased out (Paraszczak
Table 2-1: Maintained cost comparison between a diesel LHD and a tethered electric LHD (Jacobs,
2013)
Table 2-2 presents a summary of the main characteristics of each propulsion technology
available for mining vehicles. EVs embody the best balance of efficiency, cost of operation, reliability,
and performance among the options available. In addition, they represent a solution for the elimination
of hazard emissions altogether, and to reduce the heat impact of mining vehicles in the given
environment. It is important to note, that for battery-electric vehicles, the heat load is highly dependent
on the duty cycle (loaded, regeneration, charging, etc). This research aims to aim to determine the
magnitude of the heat loads for a range of different duty cycles for these vehicles.
18
Table 2-2: Summary of the main characteristics of mining vehicles propulsion alternatives.
19
2.5 Previous Evaluations of Heat Load applied to Diesel and Electric Vehicles
There are several possible approaches in evaluating the heat load from machinery in the
underground mining environment. These approaches include using an overall efficiency calculation,
estimations of power consumption, direct measurement using infra-red imaging, and summation of
estimates and models from individual heat producing components of the vehicle. The following
One method for determining the heat load for mobile equipment is by efficiency calculation of
the machinery (or constituent components). This method has been applied in studies to both diesel and
In the case of diesel machinery, the efficiency calculation approach to heat load estimation is
based on using the engine fuel consumption under a rated driving cycle in one shift for diesel
machinery. The total heat load is approximately distributed into three different areas: the radiator, the
engine, and the exhaust gases. Water vapour is emitted from the exhaust pipe through combustion.
This means that heat is lost in sensible and latent forms. Because water vapour is converted into liquid,
heat impact will be felt in underground environments over an extended period of time. This approach
has been described in several different text books (McPherson, 1986; Vergne, 2008; Marks, 2012).
The heat emitted (HE) can be calculated using fuel consumption (Fc) given in a specific time
frame (t), the combustion efficiency (Ce), and the calorific power of diesel fuel (Cpw), as demonstrated
20
(𝐹𝑐 𝐶𝑒 𝐶𝑝𝑤 ) (2-1)
𝐻𝐸 =
𝑡
be 34000 kJ/l, and the ratio between the heat emitted and the mechanical work to be 2.83 kJ/s
(McPherson, 1986). The rated power for the machine is the sum of the heat emitted plus the mechanical
work. Therefore, the calculated ratio shows that diesel engines have a rated power of 26.1% efficiency.
This result is expected since diesel engines are summed to normally have efficiencies ranging from
When applied to electric vehicles, the efficiency calculation approach to heat load
determination involves estimating of the mechanical work. In this methodology, heat loss is calculated
based on both electric power-train efficiency and useful energy. Part of the useful energy is used as
mechanical work, while the remaining energy is converted into heat through frictional losses
Figure 2-6: Heat generation impact from electric machinery (adapted from McPherson, 1986).
The heat load approach from efficiency calculation approach was applied by Marks (2012),
which found that electric motors emit 2.6 times less heat into the environment than diesel motors, while
other studies demonstrate an emission of 3 times less (McPherson, 1986; Vergne, 2008; Paraszczak,
2014; and Halim and Kerai, 2013). Based on these two ratios and knowing the efficiency rate of the
21
diesel engine (i.e. 26.1%), we can estimate the efficiency of electric units to be between 75.3% and
71.5%. However, knowing that electrical units that have 70% (Halim and Kerai, 2013) of the rated
power of diesel engines can accomplish the same mechanical work, the final efficiency for the
electrical unit will be between 59.4% and 64. 8% according to these estimations.
This method is based on general assumptions that might not be applicable to all machines under
all operations in underground mines. These assumptions include the combustion efficiency and the
ratio between the mechanical work and the heat emitted. However, it is a practicable method that allows
2.5.2 Machinery Heat Loss based on Power Consumption and Psychometric Conditions
The evaluation of the heat load is based on experimental data. For a specific study evaluating
the heat load of a mining BEV (Halim and Kerai, 2013), five different points in front of and behind the
vehicle in a tunnelling section were studied measuring the dry bulb temperature (DB), wet bulb
temperature (WB), and barometric pressure. Psychometric equations calculated heat loss from the
machinery. The study additionally considered whether the heat emitted is equal to the rated power.
The calculation for power consumption was based on the voltage (V=1000 volts), current (I=100
amperes), and power factor (cosӨ=0.85), in accordance with the three-phase power consumption
equation (2-2):
In this case, the electric vehicle was measured while it was parked. This measurement did not
impact the results since the motor drive-shaft is disengaged, allowing for continuous motion of the
22
The study concluded that the average heat emitted, and the power consumed in the electric
vehicle, are 145 kW and 147 kW respectively. It is important to note that due to the absence of moisture,
electric vehicles emit only sensible heat. These numbers slightly differ due to the precision
measurement technique. The study results supported the prior belief that heat emitted is equal to the
power consumed when the machines do not undergo work against gravity. This contradicts the
manufacturer of the LHD514E which asserted the heat emission to be approximately 107 kW
(Rakochy, 2012). The sources of this heat emission were determined to be the drive-train (97,2 kW of
heat loss) and the thermal efficiency (9 kW of heat loss). The manufacturer did indeed use the
efficiency calculation method but did not provide an adequate rationale for the values used in the
calculation of the thermal efficiency and drive-train losses. The manufacturer estimated the value of
the heat loss be 1.36 lower than that value estimated in the study by Halim and Kerai (2013). If
manufacturer estimated the efficiency for the electric machine to be approximately 73%, matching the
previously theorised efficiency range of 71.5% to 75.3%, then both studies would produce the same
result.
Calculating power consumption to determine heat loss is also used for other devices, such as
electrical equipment (Hartman, Mutmanky, and Wang, 1982). This method is used by ASHRAE to
measure heat loss from electrical equipment that operates inside buildings (Hosni, Jones, Xu, 1999),
as well as by The Engineering Tool Box (Electrical Motors and Heat Loss, n.d.)), which states that if
the electric motor is used in a closed environment, all power input is eventually converted into heat.
It is important to point out that, if no work against gravity is conducted, all the energy
consumed by any machine in a close environment will inevitably be converted into heat. This means
that, if the machine is driven to a location with a different depth within the underground mining
environment, the power consumption will not be equal to the heat load. Also, this method does not
23
take into consideration the time for the heat generated to dissipate to the environment, hence the
and the measurement points need to be well defined in order to encompass the region affected by the
Infra-red thermal imaging can be used indirectly to determine heat emission, through the use
of heat transfer equations in combination with the temperature readings. This method is an alternative
thermocouples. Infra-red cameras can be used to determine the degree of general temperature
homogeneity by identifying hot and cold spots, thus enabling comparison between simulations and
experimental results. Infra-red cameras also allow for evaluation of the heat generated from the
In previous studies, both radiators from IC vehicles (Menéndez-Díaz et al., 2014), (Calisir et
al., 2015) and batteries have been evaluated using thermal cameras (Niculuta, 2012). The disadvantage
of this methodology is that thermal images are not adequate to calculate heat emission parameters such
as the airflow magnitude and the emissivity, which is an estimated value. Acquiring these values may
nonetheless be challenging depending on the scope of the analysis. The magnitude of the airflow
however is required to determine whether natural or forced convections equations will be used to
calculate heat transfer. The value of the emissivity of most materials is already known and tabulated.
However, past studies have calculated heat radiation by painting the studied objects with a colour
whose emissivity is well-known, such as black (Forgez et al., 2010). This procedure is invasive and
24
affects the measurement, especially if radiated heat is the main contributor to the overall heat transfer.
Furthermore, the process of painting is time consuming and, depending on the situation, can make the
object of study useless after the experiment. The other issue is that, for complex geometries, heat
transfer from the faces will interfere with each other, complexifying the analysis and the acquisition of
the thermal-images. All this considered, the use of infra-red thermal cameras is an alternative to the
use of thermocouples if the intention is to calculate the heat loss using heat transfer equations. This
method may be used not only to discover the temperature distribution but also the heat load
contribution from each machine`s component. However, it is ideal to use this measuring technology in
cases in which the emissivity of the surfaces is well known, and the geometry of the object is not
complex.
Although this thesis focusses on calculating the heat generated from the battery pack, it is also
beneficial to calculate the heat produced by the other main components in a BEV in order to put the
battery heat contribution into context. To obtain the values required for the heat transfer calculation,
data from the manufacturer of the component used in the Commander 5EV must be used. This section
presents the parameters and curves obtained from the manufacturers of each component1.
Figure 2-8 presents the energy contribution of a combined cycle of city and highway duty
1
The specific names and model numbers of specific electrical components have not been reported for
the purposes of commercial confidentiality.
25
Figure 2-7: Energy input destination for a combined cycle in combustion vehicles (US Department of
Energy, n.d.).
Figure 2-8: Energy input destination for a combined cycle in electric vehicles (US Department of
Energy n.d.).
26
Without considering regenerative breaking and charging, the efficiency of electric vehicles is
approximately 65%. Again, this estimation is reasonable since the efficiency of electric vehicles
ranges, as mentioned earlier, from 71.5% to 75.3%. Taking regenerative breaking into consideration
though, the efficiency would be as high as 82%, far superior than the 28% to 32% range of efficiency
for diesel machines. This current method is used in Section 5.5 to estimate the heat generation for each
27
3 Experimental Set-up
3.1 Introduction
The focus of the experiment is to evaluate heat transfer during charging and discharging of an
individual battery cell and a battery module (multiple cells working together) in a range of simulated
operating conditions. Calculation of the heat flux is based on the heat transfer equations for the
processes of convection and radiation. Chapter 4 explains in detail the equations for the heat loss and
ceramic base between the battery set and the chamber surface. This is done to facilitate and minimize
the number of equations used to calculate the heat loss of the batteries. The description of the set-up
28
3.2 Physical Setup
To estimate the heat transfer and heat generation from the batteries during various states of
charge, seven parameters were measured: cell temperature, chamber temperature (ambient
temperature), cell voltage, cell open voltage, cell current, state of charge, and air velocity. K-type
thermocouples attached to the battery surface read the cell temperature, while one K-type thermocouple
inside the chamber read the environment temperature. These readings were sent to a data logging
system, which recorded the results by means of a LabVIEW software. The BMS (Battery Management
System), coupled directly to the positive and negative battery terminals, recorded the remaining
variables except for the airflow speed. The BMS measures the cell voltage, open circuit voltage, and
internal resistance, while the current sensor measures the current rate. The state of charge was
estimated internally by using the Coulomb-counting method to evaluate the battery profile. All
parameters measured and estimated for each second by the BMS, were provided in a single file at the
end of the operation in the laptop. Finally, the anemometer situated vertically to the airflow direction,
measured air speed inside the chamber (See Figure 3-2 and Figure 3-3).
29
Figure 3-2: Battery heat loss experiment set-up for a single cell inside the chamber.
30
3.2.1 Chamber
An industrial oven with an integrated, closed loop, temperature controller was used to simulate
the operating environment of the batteries (See Figure 3-4). The required temperatures include 25°C,
30°C, 35°C, and 40°C. Two ports were located on each side of the chamber. The left port (from the
front view of the chamber) served as the ventilation exhaust, while the right port was used for all
necessary wiring connections. A single non-profiling ¼ DIN channel controls and adjusts the
temperature based on temperature readings from a type J thermocouple located on the top interior of
the chamber. On Table 3-1 is shown the main specifications of the chamber Oven Blue M DC 146.
31
3.2.2 Ceramic Support Bricks
Two high-porous refractory ceramic bricks were located between the metal base inside the
chamber and the battery. The bricks were 125 mm apart from each other, providing space for the
bottom thermocouple. The bricks that were employed are characterised by a low thermal conductivity,
thus considerably reducing the effects of heat conduction at the contact points. This simplifies the heat
loss calculation to convection and radiation. Table 3-2 presents the ceramic brick physical properties.
Table 3-2: Low Thermal Mass, Ceramic Brick (Carbon trust, 1993).
Thermal Conductivity 0.3 W/(mK)
Specific Heat 1000 J/(kgK)
Density 130 kg/m³
Dimensions (258 x 134 x 150) mm (L x W x H)
The following section presents the devices used in the experimental phase which are responsible
The charging system, as presented in Figure 3-5, converts AC input (ranging from 100VAC to
240VAC) to DC output. The voltage output ranges from 15VDC to 30VDC, with a maximum current
capacity of 50A.
32
Figure 3-5: EFUEL PFC Heavy Duty 1200W/max 50A Power Supply.
The power supply was responsible for directing the electricity from the power grid to the
The Balance Charger/Discharger controlled the proper voltage and current to the battery. When
used to charge battery modules, it ensured that the voltage of the cell was within the necessary voltage
range, which for Lithium-iron phosphate batteries is generally between 2.8V and 3.6V. It operates a
of 40A charging. Its software can verify the voltage, the current, the state of charge, and the internal
resistance in each cell. The unit also has the capacity to discharge batteries, but this feature was not
33
used during the experiment due to the limited current capacity. Instead, a resistor bank was used for
The resistor bank, presented in Figure 3-7, is composed of three 2.5Ω Ohmite C300K2R5E in
the first column and nine 0.63Ω Ohmite C300KR63E in the remaining columns. In the experiment, 2.7
Ω of resistance was measured in the first column and 0.83 Ω in the other columns, with the wiring
imposing an extra resistance of 0.02 Ω. These high-power resistors, which have a maximum capacity
of 300 W capacity, can used under intermittent cycles. To protect against extreme temperature
oscillations, the wires are embedded in ceramic with an enamel coating. Fans additionally were used
The connecting wires in the resistor load bank may be rearranged for various setups equating
to a range of effective resistances for the discharging processes. Figures 3-8 and 3-9 exhibit the set up
for Flat and Uphill discharging respectively. The current target of the Flat condition is 17 amperes,
34
Figure 3-8: Arrangement of resistors for single cell analysis under Flat Condition.
To simulate the Uphill condition, two battery cells were connected in series and then connected
to six 0.63Ω and three 2.5Ω resistors in parallel. An extra battery cell was used in series with the battery
Figure 3-9: Resistor arrangement for single cell analysis under Uphill condition.
35
The same resistor arrangement was used for a module with 6 cells connected in series. The
most suitable arrangement can be seen in Figures 3-10and 3-11, where the final current values are 14.5
amperes and 35 amperes for the Uphill and Flat conditions respectively. The limited number of
resistors in the resistor bank caused a difference in current between the single cell and the module
during the discharging process. The limitations of the charger caused similar deviations in the current
value during the charging process. During the discharge process, the resistor bank is connected directly
Figure 3-10: Arrangement of the resistors for a module analysis under Flat Condition.
Figure 3-11: Resistors arrangement for module analysis under Uphill Condition.
The experimental phase is divided into two parts, the heat transfer analysis of the single cell
battery and the heat transfer of the battery module. The single cell analysis aims to provide a more
complete description of the heat generation based on the temperature distribution over the battery cell,
36
and as for the module analysis, how this heat generation is affected when multiple cells are used closed
The experiments used the CALB CAM 72Ah battery cells. These batteries are based on
Lithium-iron Phosphate chemistry, and they are the same used in the Commander 5EV. In comparison
to other lithium battery chemistries, LiFePO4 technology is safer to operate if abused. As well, not
only has it got considerable specific energy and power, but it also can be used for a high number of
cycles (see Table 3-3). These attributes make this a common choice for modern mining EVs. Figure 3-
12 shows the discharge voltage curve as specified by the manufacturer under different rates of current,
Figure 3-12: Discharge curves for CALB CAM 72Ah at 1C, 0.5C, and 0.3C (CALB, 2014).
37
Table 3-3: CALB CAM 72Ah specification (CALB, 2014).
Capacity 72 Ah
Voltage 3.2 V nom
Cycle Life ≥2000 Cycles
Internal Resistance ≤1mΩ
Maximum Constant
Charging (Constant 72 A
Charging Current
Current-Constant Voltage)
CC To CV Voltage 3.65 V
Maximum Constant
144 A
Discharging Current
Discharging Cut-off
Discharging 2.5 V
Voltage
576 A @ 10 s
Pulse Discharge
288 A @ 30 s
Externally, the battery cell has a prismatic geometry (Figure 3-13), in which its internal content
is held by an aluminum alloy case 1 mm thick. A thin blue plastic layer surrounds the case, giving it
an emissivity of 0.78 (Omega, accessed on April 5th, 2017). The two electric battery connectors
(terminals) located at the top are made of different materials: aluminum and copper. The air-port vent
prevents the pressure to build up inside the case, and it is located between the top connectors.
38
Figure 3-13: Single cell CALB CAM 72 Ah external view (CALB, 2014).
As for in Figure 3-14, it is presented the module with 6 cells connected in series, with a total
voltage of approximately 19V, in the same arrangement used during the experiment.
Figure 3-14: One module of the battery pack with 6 cells with the plastic frame used in the EV.
39
3.4.2 Internal Structures
As for the internal structure of the Lithium-iron Phosphate battery cell, it is represented by the
core, enclosed by the aluminum case, with an air gap between the top of the core and the top of the
battery shell. In the air gap is located the aluminum, and copper extensions from the electrodes, which
are physically connected to each top terminal. The battery core has 80 layers of the negative electrode,
and 80 layers of the positive electrode for which each element is separated by the PP separator, with
the polyethylene layer evolving the whole content (Figure 3-15 and Figure 3-16).
Figure 3-15: Internal components of the battery cell CALB CAM 72Ah. The opening stages can be
40
Figure 3-16: Internal Components of the battery core. The core is exposed in (a), and its contents in
Figure 3-17: Battery core internal Layers. (a) – Positive Electrode, (b) – Negative Electrode, (c) –
The Positive Electrode consists of an aluminum sheet, coated by LiFePO4, while the Negative
Electrode is a sheet of copper, coated by graphite. Every electrode is placed parallel to each other inside
41
the core, with a PP Separator in between. Then, these electrodes can trap the liquid electrolyte inside
them since they have porous structure. Table 3-4 presents physical and geometric properties for each
material.
Table 3-4: Specification for the Layers inside the battery core (Chen, Wan, Wang, 2005).
Thermal
Material Thickness Density Heat Capacity length height
Conductivity
- (mm) (kg/m³) (J/kg K) (W/m K) (mm) (mm)
Aluminum foil 0.064 2702 903 238 100 165
Carbonaceous electrode 0.067 1347.33 1437.4 1.04 100 165
PP separator 0.036 1008.98 1978.16 0.3344 100 165
Lithium electrode 0.079 2328.5 1269.21 1.58 100 165
Copper foil 0.026 8933 385 398 100 165
Figure 3-18 presents the internal layout of the battery. In this schematic, the view of the
terminals in the gap volume is omitted, to facilitate the visualization. In the battery core, each segment
of PP separator, Positive electrode, PP separator again, and Negative electrode together, represents one
pouch cell. A total of 80 pouch cells is compressed inside the case. It is the number of pouch cells that
Figure 3-18: Internal Battery view – Schematic (H x L x D – 216.8 x 135.2 x 29.2 mm).
42
Figure 3-19 presents the terminals with a prismatic geometry. This is a simplified
representation, since there are several layers of copper, or aluminum for the opposite side, compressed
together, and attached to the top battery cells connectors, as can be seen in Figure 3-17. The dimensions
for w’, w”, and h” are 8.5, 32, 28.8 mm, respectively. Additionally, these connectors are located at the
The CALB CAM 72 Ah is the actual battery cell selected as energy storage for the Commander
5EV. The machinery has a total of 50 modules, and each module has 8 battery cells. One of these
modules is the one selected for this study, as shown in Figure 3-14, with 6 cells connected in series.
The Battery Management System, Data Logging System, Anemometer, and Thermal Camera
are explained in detail in terms of features, usage requirements and the results provided by each device.
43
3.5.1 Battery Management System (BMS)
The BMS Jr manufactured by Orion, presented in Figure 3-20, not only obtains readings but
also balances the battery pack. Capable of evaluating from one to sixteen Lithium-iron Phosphate
chemistry cells in series with 48V nominal and 60V maximum, it measures the state of charge,
discharge current limit, charge current limit, cell voltage, open cell voltage, and current. The CANBUS
communication protocol is used to communicate with the BMS, meaning that the device also requires
a can adapter. The BMS is powered by 12V input, which is supplied by the DC Power Supply AB5PS-
D.
A National Instruments PXIe-1082 data acquisition platform fitted with an NI TB-4353 front
mounting isothermal terminal block used for thermocouple inputs (Figure 3-21) to obtain temperature
readings. LabVIEW, the chassis software, recorded the temperature values sent from each K-type
thermocouple.
44
Figure 3-21: PXIe-1082 chassis with NI TB-4353 front mounting terminal block attached.
K-type thermocouples were attached to the surface of the cells, according to the layouts
presented in Figure 3-22 and Figure 3-23 for single cell and module respectively. For the single cell
setup, thermocouples were placed in the centre of an area of interest on the front panel. The same was
performed on the bottom and top surfaces, but the vent port locations required that thermocouple T9
was located on the side of the battery. The final temperature value used to calculate the heat loss is the
average temperature reading of the front and sides of the area of interest. T10 was used to evaluate the
45
Figure 3-22: 10 K-type thermocouples location around the battery cell. Schematic not in scale.
In the module thermocouple arrangement, one thermocouple is in the centre of the front face
Figure 3-23: K-type thermocouples around the battery module. Schematic is not in scale.
46
Figure 3-24 shows that the thermocouples were located at the back of the battery cell to allow
Figure 3-24: Setup for thermocouples and voltage tap connected around the battery cell.
The Anemometer probe, located inside the chamber adjacent to the battery cell, measures the
air flow speed (Figure 3-25), as well as the mass flow, temperature, relative humidity, wet bulb, and
dew point. The values measured are recorded in a data acquisition system.
47
Figure 3-25: Anemometer – TSI/Alnor 9545 VelociCalc and its probe stick (TSI, n.d.).
A thermal camera (FLIR model E5 depicted in Figure 3-26) was employed to capture the
temperature distribution of the battery surface. The image from the experiment was used to validate
the simulation image. The infrared picture has also the advantage of providing a far more detailed
result than the evaluation of the local temperature by a thermocouple. The downside of this
methodology is that the image can only be obtained at the end of the procedure when the battery unit
reaches its peak maximum temperature. This because photographs of the battery may not be taken
during the experiment as this would require the chamber to be open, thus interfering with temperature
reading results. Furthermore, it can be used to evaluate the temperature over surfaces with emissivities
48
within the range of 0.1 to 1 (see Table 3-5), which includes the battery cell, which has an emissivity
49
3.6 Experimental Procedure
1. Battery was placed inside the chamber and over the ceramic base;
2. Charging/Discharging cables were connected between the power supply and the resistor bank;
3. K-Thermocouples were attached to the battery surface and internally to the chamber at the top,
4. Anemometer was placed inside the chamber and connected to the laptop;
5. BMS was connected to the negative and positive terminals of the battery;
8. Waiting time until the battery reached temperature equilibrium with the chamber;
9. Charging process: Charging process was initiated by giving a command to the power-supply
via laptop / Discharging process: Emergency bottom was opened allowing the battery to be
10. The temperature was started being recorded by the thermocouple and the voltage/current by
the BMS;
11. Once the desired SoC was reached accordingly to the BMS, the system was disengaged;
13. Chamber was opened, and a thermal picture of the battery/module was taken.
There are generally four modes of operation for the battery which result in heat generation:
charging, downhill driving (regeneration charging), flat driving and uphill driving modes. These modes
represent the basic conditions in which the Commander 5EV is going to be used underground. The
current target for each mode is 16.9 amperes, 30.8 amperes, 17.0 amperes, and 42.3 amperes
50
respectively. These values were estimated by the electric motor manufacturer based on the power
requirement for each condition. Experiments were conducted to simulate these operating conditions.
In the single cell experiment the battery was charged, for charging and downhill mode, at a
current of 17.4 amperes, and 29.4 amperes on average respectively. The battery was then discharged
in two different conditions, one at 17.0 amperes on average in a condition simulating driving over a
flat surface, and the other at 44.1 amperes on average in an uphill simulation as though the vehicle is
driving up the ramp. These values do not precisely match those from the target due to accuracy
limitations from the resistor bank and the balance changer/discharger unit. Each driving condition is
conducted in 25°C, 30°C, 35°C, and 40°C ambient temperature controlled by the oven chamber.
Table 3-7: Test matrix for single cell charging and discharging cases.
Usage Mode Operating Condition Current Rate Ambient Temperature
Charging Simulates battery charging when vehicle 17.4 amperes 25°C, 30°C, 35°C,
is not in use. and 40°C
Downhill Simulates the regenerative breaking 29.4 amperes 25°C, 30°C, 35°C,
Charging system when driving downhill. and 40°C
Flat Simulates the requirement for driving the 17.0 amperes 25°C, 30°C, 35°C,
Discharge machine over flat surface. and 40°C
Uphill Simulates the requirement for driving the 44.1 amperes 25°C, 30°C, 35°C,
Discharge machine upward on 20° amperes and 40°C
In the module heat evaluation, 6 cells were connected in series to form a pack (See Figure 3-
14). The pack was charged in charging and downhill modes and discharged in flat and uphill modes
at 16.5 amperes, 31.0 amperes, 16.8 amperes, and 41.0 amperes respectively. As explained earlier, the
slight difference between the target and the actual results were due to the same reasons as in the case
of single cell experiment. The module was tested in the same range of ambient temperature as the
single cell case. In both single cell and module experiment procedures, the module was
charged/discharged within the range of 10% to 90% State of Charge (SoC). For the experimental
results, see appendix A for single cell, and appendix B for module.
51
Table 3-8: Test matrix for module charging and discharging cases.
Usage Mode Operating Condition Current Rate Ambient Temperature
Charging Simulates module charging when vehicle 16.5 amperes 25°C, 30°C, 35°C,
is not in use. and 40°C
Downhill Simulates the regenerative breaking 31.0 amperes 25°C, 30°C, 35°C,
Charging system when driving downhill. and 40°C
Flat Simulates the requirement for driving the 16.8 amperes 25°C, 30°C, 35°C,
Discharge machine over flat surface. and 40°C
Uphill Simulates the requirement for driving the 41.0 amperes 25°C, 30°C, 35°C,
Discharge machine upward on 20° amperes and 40°C
The results for the evolution of the average surface temperature throughout the experiment for
each case are presented below, Figures 27 to 34. The results for the same mode but for different ambient
temperatures, 25°C, 30°C, 35°C, and 40°C are combined in the same graph to facilitate visualization
of the impact of the ambient temperature on the temperature developed on the surface of the battery /
module. In addition, it is worth mentioning that the graphs correlate the surface temperature with the
state of charge, rather than the surface temperature over time, because each experiment exhibited
different charge or discharge rate. Thus, the normalization of all the results based on SoC facilitates
52
Uphill Mode
Figure 3-27: Experiment, battery surface temperature evolution, Uphill single-cell case.
Figure 3-28: Experiment, battery surface temperature evolution, Uphill module case.
53
Flat Mode
Figure 3-29: Experiment, battery surface temperature evolution, Flat single-cell case.
Figure 3-30: Experiment, battery surface temperature evolution, Flat module case.
54
Charging Mode
Figure 3-31: Experiment, battery surface temperature evolution, Charging single-cell case.
Figure 3-32: Experiment, battery surface temperature evolution, Charging module case.
55
Downhill Mode
Figure 3-33: Experiment, battery surface temperature evolution, Downhill single-cell case.
Figure 3-34: Experiment, battery surface temperature evolution, Downhill module case.
56
3.8 Discussion
As expected, the overall surface temperature of the battery increased during the experiment.
The variation in the rate of increment in temperature is due to the second term in the heat generation
equation for lithium battery cells. This second term referred to as the reversible entropy, discussed in
detail in Section 4.3. The current as well as the difference between the voltage and the open voltage
The temperature coefficient required in equation (4-4) is obtained from a commercial Lithium-
iron Phosphate battery cell (Forgez et al., 2010), whose respective curve can be seen in Figure 3-35.
To not compromise the battery’s life-span, the temperature coefficient curve is only used within the
Figure 3-35: Temperature Coefficient curve adapted from (Forgez et al., 2010).
The average temperature evolution along the cell/battery module system generally follows the
“shape” of the temperature coefficient curve (Figures 3-27 to 3-34). With regards to the charging
processes (charging and downhill mode), the general average temperature surface curve follows a small
decrement in the average surface temperature from 10% to 15% SoC, increasing over time until it
57
reaches a plateau around 60%, decreasing again until 80%, and increasing once more until 90% SoC
has been achieved. We can deduce from equation (4-4) that since the Ohmic Heating term is assumed
to be constant for this study curve, the heat generation variation is going to be dictated by the Reversible
Entropic term. As for the discharging processes (Flat and Uphill modes), the SoC is decreased over
time. This means that at the beginning of the process, the temperature is dramatically increased due to
the high value for the temperature coefficient (Figures 3-22 to 3-25). Usually a plateau level is reached
around 60% as a delay reflection of the lower temperature coefficient level, which stands around 80%
SoC (Figure 3-30). After 40% SoC, the battery continuously increases its temperature dramatically
because of the Ohmic Heat term, which has a considerably higher magnitude in comparison to the
Table 3-9: Average Heat Generation in terms of Ohmic Resistance and Reversible Entropy.
The average surface temperature results in the current study showed that there is an agreement
with the findings obtained in a prior study for both charging and discharging processes in the case
where the chamber temperature is constant (Niculuta, 2011). In addition to this, it was observed that
the magnitude of the ambient temperature was not enough to considerably affect the surface
temperature increment of the battery/module for a given current level. In the case of the single cell,
the average temperature increases for uphill, flat, charging and discharging cases, resulting in 7° C, 3°
C and 2° C increments occur for the same modes. These results show that the current level is the
Furthermore, even though both the battery and the module use similar current levels, which
lead to similar heat generation, it did not translate into similar surface temperature because both cases
58
had different surface emissivity. For instance, the emissivity of the battery pack module is 0.78
(Omega, accessed on April 5th, 2017), versus 0.95 (Emissivity Values of Common Materials, 2007) of
the electrical tape used to cover the single cell. The insulating effect is assumed to be insignificant
because it has low thermal resistance (vinyl ester: 0.25W / mK, air 0.0262W / mK) (Thermal
conductivity of common materials and gases, n.d.) and has a negligible thickness. This new emissivity
value affected the amount of radiative heat loss emitted by the single battery, which comparable to the
The understanding of temperature variation over time can be valuable in the daily use of these
machines underground. As mentioned at the beginning of this section, the temperature increment is not
constant and sometimes there is a possibility of a temperature decrease for a specific range of SoC, in
which its surface temperature variation over time strongly depend on each battery chemistry. These
are repeatable and consistent results, as expressed by Figures 3-22 to 3-29, which proves that the
heat generation model. This knowledge can also be useful to proper size and operate the battery cooling
system (if necessary), minimizing power consumption by the cooling pumps and to avoid thermal
4 Mathematical Treatment
59
4.1 Overview
The section describes the mathematical treatment which is used as a basis for the analysis of
the battery thermal performance. The method, which is also followed by numerous other researchers,
(Niculuta, 2012; Chen, Wan, Wang, 2005) divides the internal structure of the battery into different
subdomains. This strategy allows direct mathematical and virtual modeling of the battery, given the
specificity of each region and the equations which governs the heat transfer. See Figure 4-1 for the
Most of the studies researched had divided the battery cell into three domains, the case, core,
and the air gap (Niculuta, 2012) and (Chen, Wan, Wang, 2005). In this study, the battery connectors
will be included as part of the evaluation as a mean to obtain a more accurate result. Inside the battery
core, and under charging and discharging regimes, the transient heat is generated and propagated
through three dimensions, as expressed by the energy balance equation (4-1). This equation describes
the heat transfer from the battery core. The radiative heat term is not considered because it is assumed
60
that the internal layers are opaque. This heat is transferred through the internal layers, to the air gap at
the top of the core, to the aluminum case, and then, it is released to the outside in the form of radiative
𝜕𝑇 𝜕 𝜕𝑇 𝜕 𝜕𝑇 𝜕 𝜕𝑇 (4-1)
𝜌𝐶𝑝 = (𝑘𝑥 ) + (𝑘𝑦 ) + (𝑘𝑧 ) + 𝑄
𝜕𝑡 𝜕𝑥 𝜕𝑥 𝜕𝑦 𝜕𝑦 𝜕𝑧 𝜕𝑧
𝜌, Cp, k, and Q represent density, heat capacity, thermal capacity and heat generated in volumetric
units respectively.
The battery core is composed by several thin layers of different materials as seen in Figure 3-
16. When running a simulated model for heat transfer analysis, this level of detail can be excessively
burden computational resources. For this reason, a common practice among those who study batteries
from the thermal aspect is to simplify the internal structure into a single component, with a specific
heat capacity and thermal conductivity (Niculuta, 2012; and Chen, Wan, Wang, 2005). The plane XZ
is the one used as a reference for the equations because the layers are parallel to it. See Figure 4-2.
Figure 4-2: Effective thermal conductivity for parallel (a), and series (b) elements (Chen; Wan; Wang,
2005).
The battery orientation used for the calculation can be seen in Figure 4-1. Then, the calculation
for the effective thermal conductivity normal to the plane XZ, or ktrans, is given by equation (4-2 (a));
and for planes XY, and YZ, or klong, is given by equation (4-2 (b)). To calculate the equation (4-1), it
is also required to obtain the effective 𝜌 and 𝐶𝑝 terms, in which both can be calculated through the
61
equations (4-2 (c)) and (4-2 (d)). L is the material thickness, k is its thermal conductivity, A is the layer
front area in the plane XZ, v is the layer volume, and i is the index for each layer.
From equations (4-2) and the values provided by Table 3-4, the parameters ktrans, klong, 𝜌, and
𝐶𝑝 are calculated, given 1.145 W/mK, 23.9 W/mK, 2133 kg/m³, and 1335 J/kgK. These final values
The most detailed model available to describe the heat generated in lithium-ion battery cells
was developed by Thomas and Newman (2003), and it is described in equation (4-3):
𝜕𝑉𝑜𝑐 𝜕𝑐 (4-3)
𝑄̇ = 𝐼 (𝑉 − 𝑉𝑜𝑐 ) + 𝐼 (𝑇
𝑎𝑣𝑔 ̅𝑗𝑎𝑣𝑔 ) 𝑗 𝑑𝑣
̅𝑗 − 𝐻
) − ∑ ∆𝐻𝑖 𝑟𝑖 − ∫ ∑ (𝐻
𝜕𝑇 𝑖 𝑖 𝜕𝑡
This expression has four main terms: Ohmic heating (or resistive dissipation), reversible entropic heat,
heat by chemical reaction and heat of mixing. Where 𝑄̇ is the heat generation over time; I is the current,
which can be either positive (charging), or negative (discharging); V is the cell voltage; 𝑉𝑜𝑐 is the open-
circuit potential or equilibrium cell voltage; T is the temperature, which in general is the average
𝜕𝑉𝑜𝑐
between the surface and the core; 𝜕𝑇
is the temperature coefficient; ∆𝐻𝑖 is the enthalpy variation; 𝑟𝑖
62
materials j; t is time; and v is the volume. Usually, the temperature coefficient is obtained
The fourth term represents the heat generated by the concentration variation across the
electrolyte and the porous inside the internal layers as starting from the last component of equation (4-
The third term considers the side reactions that can contribute to heat generation or
consumption. So, as the fourth term, it can have positive or negative value. Since these reactions are
commonly related to slow process such material aging, this term can also be neglected (Bernarda;
The formula ends up taking the shape of equation (4-4) through the elimination of the third and
𝜕𝑉𝑜𝑐 (4-4)
𝑄̇ = 𝐼 (𝑉 − 𝑉𝑜𝑐 ) + 𝐼 (𝑇 )
𝜕𝑇
Ohmic Heating Reversible
(Joules Effect) or Entropic Heat
Resistive
Dissipation
This equation was originally developed by (Bernardi; Pawlikowski; Newman, 1985). The first
term represents the Ohmic Heating from Joule`s effect, and other internal resistances. It has positive
magnitude. The second term is the mathematical interpretation of entropy change due to
electrochemical reactions. It can be either negative or positive. According to some sources (Pesaran;
Vlahinos: Burch, 1997), and (Khasawneh, 2011) it always has a considerably lower magnitude when
compared to the first term, and for complete charge and discharge, such as in EV applications for
example, it can be ignored. However, in some other studies (Thomas, Newman, 2003; Forgez et al.,
2010; Niculuta, 2012), it is stated the opposite since they have the same magnitude as the Resistive
Dissipation. Thus, this study follows the last approach in order to obtain more accurate results.
63
As pointed out by Niculuta (2012), the first term in equation (4-4) can be described as in
equation (4-5).
The heat generation from the collectors, 𝑄̇𝑐𝑜𝑙 , is given by equation (4-6):
𝑅𝐼 2
𝑄̇𝑐𝑜𝑙 = ( 𝑣 ), where 𝑅 = 𝑟"ℎ (4-6)
The material resistance (R) is calculated based upon the material resistivity (r”), which is assumed to
be 1.68x10−8 Ωm for copper, and 2.65x10−8 Ωm for aluminum (Giancoli, 1995). As for the remain
parameters in the equation (4-6), v is the collector volume, and h is the collector height.
The air inside the gap domain is static, or a solid like material. This allows the use of equation
(4-1) for determining the energy balance. This is done to facilitate the simulation through the
elimination of fluid dynamics mathematical modeling. Thus, the effective thermal conductivity is
It is assumed that the air gap is a vertical enclosure where occurs internal natural air convection
from the bottom to the surrounding faces. Thus, equation (4-8) can be used as a mean to calculate the
Nusselt value. 28.8mm is assumed to be the height of the air volume, and 27.3mm the depth (Incropera;
DeWitt, 2002):
0.29 (4-8)
𝑃𝑟
𝑁𝑢𝑔𝑎𝑝 = 0.18 ( 𝑅𝑎 )
0.2 + 𝑃𝑟 𝐿
64
Requirements for the use of equation (4-8) are listed in equation (4-9).
𝐻 (4-9)
1< <2
𝐿
10−3 < 𝑃𝑟 < 105
3
𝑅𝑎𝐿 𝑃𝑟
{ 10 <
0.2 + 𝑃𝑟
temperature surface, 𝑇2 is top surface temperature, 𝑣 is the volumetric thermal expansion coefficient,
which is equal to 2/(𝑇1 – 𝑇2 ), v” is the kinematic viscosity, α is the thermal diffusivity, and Pr is the
(dimensionless) Prandtl number, in which the values for the last two parameters are taking from the
The heat transfer from the battery to the surrounding ambient environment occurs mainly
through radiation and convection. In this study, the heat transfer due to conduction between the battery
and the base support is assumed to be of negligible magnitude and not considered.
The heat transfer through Radiation and Convection is given by equations (3-11) and (3-12)
respectively:
constant, which is equal to 5.67 x 10−8 W/m². K, and A represents the surface area from where the heat
65
is being transferred from. For equation (4-12), ℎ𝑐𝑜𝑛𝑣 is the convection heat transfer coefficient, which
𝑁𝑢 𝑘𝑎𝑖𝑟 (4-13)
ℎ𝑐𝑜𝑛𝑣 =
𝐿
In this case, different correlations for the Nusselt calculation are used for each battery side. The
equations for each face are as described below following the orientation seen in Figure 4-3 (Incropera;
DeWitt, 2002).
1/4 (4-14)
0.670 𝑅𝑎𝐿
𝑁𝑢 = 0.68 +
[1 + (0.492/𝑃𝑟)9⁄16 ]4⁄9
66
● Top Face
1/4 (4-15)
𝑁𝑢 = 0.54 𝑅𝑎𝐿
For the top surface, the characteristic length, L, is given by equation (4-16):
𝐴𝑠 (4-16)
𝐿=
𝑃
● Bottom Face
1/4 (4-17)
𝑁𝑢 = 0.27 𝑅𝑎𝐿
For the bottom surface, the characteristic length is calculated as described by equation (4-17)
as well. For each surface side, the Rayleigh number is calculated the same manner as given by equation
(4-18):
which is equal to 2/(𝑇𝑠 – 𝑇∞), v” is the kinematic viscosity, α is the thermal diffusivity, and Pr is the
(dimensionless) Prandtl number. The values for the thermal diffusivity and Prandtl number are taken
from the average temperature between the surface and the environment.
The methodology used is the same as developed by Ellison (1969) for forced convection. In
this method, the heat transfer coefficient hc is calculated based on a function for the average
temperature between the surface and the environment, as described by equation (4-19):
67
(4-19)
𝑉
ℎ𝑐 = 𝑓 √
𝐿
Where V is the air velocity, and L is the characteristic length, whereas the values for f, the temperature
Table 4-1: Sample values for Temperature Difference Coefficient (Ellison, 1969).
68
5 Simulation and Experimental Results
5.1 Overview
The general goal of this section is to complement the experimental results described in Chapter
3 through calculation and simulation of the test conditions. The section will examine surface
temperature distribution and average surface temperature over time for the single battery and the multi-
cell battery modules subjected to different environmental temperatures and electrical duty cycles. The
methodology is like what was done by other researchers working in this issue including (Niculuta,
2012; Chen, Wan, Wang, 2005). The simulation is based on the heat flux and heat generation equations
for lithium battery cells, the mathematical basis for these calculations are described in the previous
section. Parameters such as voltage, open voltage, current, cell temperature, and environment
temperature were obtained from the experiment for each second and used as input for the simulation,
see Figure 5-1. Heat generation and heat loss in the experiment are calculated based on these same
parameters, which are imported into the model via spread-sheets. Due to the inevitable damage to the
batteries, and the safety measures imposed on the experiment, it was not possible to insert a
thermocouple inside of each cell (Forgez et al, 2010). Instead, this value is estimated based on the
results extracted from the simulation itself. Once the contribution of the battery is calculated, then it is
possible to address the total heat impact of the machinery by adding the thermal impact of the battery
69
Figure 5-1: Block diagram of the FEA model.
In the actual battery, there are multiple layers (also known as the pouch cells) of dissimilar
materials…For the modeling, a simplification that was introduced by previous researchers (Niculuta,
2012; Chen, Wan, Wang, 2005) is used to optimize the simulation. It involves the representation of the
multiple layers of pouch cells as a single structure (referred here as the battery core) (See Chapter
3.4.2) in which the material properties can summarize into a single component as described by
equations 4.2. After combining all pouch cells together, the resulting structure is anisotropic with
5.2 Methodology
Simulations for this work were conducted using the Finite Element Modeling (FEM) software
COMSOL MULTIPHYSICS (version 3.5a) using the heat transfer pack for 2D and 3D environments.
COMSOL is a well-known, robust, and powerful software which allows the user to incorporate
physical equations to better describe a phenomenon, such the heat generation in lithium battery cells.
The single cell the FEM model employed a 3D mesh size with a total of 17023 tetrahedral
elements presented in Figure 5-2. Special attention was given to the thickness of the aluminum battery
70
case, hence, the dense mesh. For this analysis, assumed that a negligible amount of heat is transferred
at the base of the model. In all the remain surfaces, only radiative and convective heat transfer occurs.
The mesh is considerably denser in the center of the frontal and back face to better describe the heat
flow on these regions. The battery cell modeled for the FEM simulations is the CALB CAM 72Ah
(previously described in section 3.4.1 and 3.4.2), see Figure 4-2 and Figure 4-3.
Figure 5-2: Mesh Distribution for the CAD single cell model.
71
5.3.1 Important Considerations about the Simulation and the Experiment
The simulations were conducted using the same ambient temperature conditions and battery
parameters (Voltage, open-circuit voltage, current) from the experimental work described in Chapter
3. When the battery was tested inside the chamber, the air temperature was set at mean value throughout
the operation. During the experimental test described in Chapter 3, the environmental chamber
temperature control unit worked to maintain a steady environmental temperature during the test. As
stated earlier, due to limitations on the ability of the heating system to keep maintaining a perfectly
steady temperature, some temperature fluctuation was observed. It is important to note that the
simulations described in this section used the actual observed ambient temperatures in during the
experimental work rather than the setpoint. This measure is intended to reflect a more representative
Prior to calculating the convective heat transfer from the battery surfaces, it was first required
to determine if the air flow regime is laminar (natural convection) or turbulent (forced convection).
This was performed by determining the Reynolds number (Re). If this number is higher then 5 x 105,
then air flow regime is turbulent, otherwise it is laminar. This parameter is calculated based on equation
5-1.
𝑣𝑙 (5-1)
𝑅𝑒 =
𝑉
For the thermal test chamber used in the experiment test, room temperature air is drawn in to
a heating compartment from the outside, then it is preheated with resistive heating elements, blown
inside the test chamber, and finally expelled to the outside again. According to measurements taken
with an anemometer, the airflow inside the chamber never exceeded 0.3 m/s. This resulted in a
72
Reynolds value always lower than 280 for the side of the battery side and 2550 for the front of the
battery. Corresponding to these Reynolds values, the set of equations for natural convection were used
As can be seen in equation (4-4), the battery internal temperature is required for the heat
generation calculation. The core temperature was measured in another study (Niculuta, 2012) by
internally inserting a thermocouple in a single experiment. The average between the core and surface
temperatures was thus used as a constant in the reversible entropy heat term. This methodology is
problematic because it does not consider the temperature difference between the core and the surface
over time. However, a solution proposed for this issue was to estimate the internal temperature based
on the temperature difference between the core and the battery`s surface obtained in the simulation.
Section 5.3 provides summarised results only for the Uphill, Flat, Charge, and Downhill cases
in 40°C ambient temperature. The results for scenarios involving 25°C, 30°C, and 35°C are provided
The simulation demonstrates the discharge from 90% to 10% SOC within 78 min on average
for the single cell Uphill scenario. As was in the experiment, the simulation is conducted in 25°C,
30°C, 35°C, and 40°C, with currents of 44.3 amperes, 44.3 amperes, 43.7 amperes, and 44.2 amperes
used in each case respectively. After the discharge process, the battery temperature gradually decreases
until it again reaches the thermal equilibrium within the test chamber again. This response reflects what
The results of the FEM simulation are presented in Figure 5-3 below. It presents the surface
temperature distribution after discharging at the ambient temperature of 40°C. The temperature
73
difference between the side and the front surfaces is negligible, but there exists a temperature difference
of 1°C between the upper section, where the air gap is located, and the lower section, where the core
is located. The figure also presents the heat generated by the battery terminals inside the air gap. The
copper terminal, located at the internal top right, has a higher surrounding temperature than the
aluminium terminal at the internal top left, since the copper material has higher thermal conductivity.
Figure 5-3: Final visual simulation of the temperature distribution along the external battery surface
Figure 5-4 presents the observed temperature distribution of the cell during the period of peak
heat generation (5 W) at the completion of the process for the Uphill condition at 40°C, resulting in an
74
average of 1.45W. As well, the mean value increases consistently in heat loss at an average of
0.50W/cell2.
Figure 5-4: Experimental temperature distribution on the external battery surface for 44.2A
In Figure 5-5, the heat loss result is approximated using a 6th degree polynomial fit. This
provides a smoothed result minimizing the effects of the chamber temperature fluctuations.
2
The thermal image presents a region with lower temperature in the upper right side, which is the result
of the insulation tape not being attached directly to the surface on the cell. This tape is holding the
thermocouple in place and that does not mean that the temperature surface is reduced on the surface of
the battery in this region.
75
Figure 5-5: Experimental Heat generated and lost during battery discharge for the Uphill condition.
It was observed that the lower portion of the cell, particularly the sides and the base, produced
the highest temperature in the cell as presented in Figure 5-3, this result is in agreement with what was
observed in the experimental case (Figure 5.4). A cooler portion was also observed in the top section
of the battery which coincided exactly with the position of the air gap, see Figure 3-18. Knowing that
the temperature distribution around the cell is not homogeneous, the temperature curve described in
Figure 5-6 is related to the weighted arithmetic mean of the temperature from each thermocouple in
relation to the area of the surrounding region as pointed out in Figure 3-22. According to the
simulation, the battery cell surface temperature ranges from 45.3°C to 44.4°C at the completion of the
process. The average temperature for the whole simulation varies within a 6°C range, independently
76
Figure 5-6: Simulated temperature and Experimental temperature obtained at the centre of the front
As explained earlier, because the temperature inside the chamber oscillates around a mean
value, this also affects the temperature of the surface of the cell (Figure 5-6). Nevertheless, since the
frequency for the temperature variation is sufficiently high in comparison to the experiment length, the
The most significant impact on heat generation from the battery pack is the Ohmic Heating
term equation (4-4). This term is the magnitude of the heating produced during battery charging or
discharging which then impacts the current level and creates the difference between voltage and open
voltage (Figure 5-7). Consequently, the overall impact of the reversible Entropic term is almost
negligible. The average Ohmic Heating term is 1.52W on average, while the Reversible Entropic term
77
is -0.06W on average (Figure 5-8). However, the Reversible Entropy term has the most significantly
Figure 5-8: Heat generation Comparison between the Ohmic Heating term and the Reversible Entropic
78
5.3.3 Flat scenario
The Flat condition (representing the battery duty while the EV is driving on flat ground)
requires approximately 205 minutes to discharge from a state of charge of 90% to 10%. The
experiment was conducted in ambient temperatures of 25°, 30°C, 35°C, and 40°C with average
discharge currents of 17.3 amperes, 16.9 amperes, 16.9 amperes and 16.7 amperes respectively. Figure
5-9 shows the thermal distribution on the surface of the battery cell at the completion of the process,
which is considerably more homogeneous than the last scenario, having a maximum surface
temperature variation of 0.5°C. As for Figure 5-10, it presents the temperature distribution over the
Figure 5-9: Final visual simulation of the temperature distribution along the external battery surface
79
Figure 5-10: Experimental temperature distribution on the external battery surface for 16.9A
The maximum heat generated at the completion of the process is almost 1.5W, which is nearly
0.4W of heat loss. However, the average value for the whole process is 0.16W of heat generated and
0.15W of heat lost (Figure 5-11). The heat accumulation, less than in the Uphill condition, produces
Figure 5-11: Experimental Heat generated and lost during battery discharge for the Uphill condition.
80
Figure 5-12: Simulated temperature and Experimental temperature obtained at the centre of the front
The temperature decreases during the middle of the process as seen in the temperature curve
in Figure 5-12, unlike in the previous case when the temperature remained constant between 60% to
40%. This results in a negative heat generation of 80% to 40% SoC. The final average cell temperature
is approximately 43.3°C in the experiment and 43.75°C in the simulation, which shows that both
The for the batteries used in this work, the charging process typically required 195 minutes to
be completed from 10% to 90% SoC. The Charging scenarios are conducted in 25°C, 30°C, 35°C, and
40°C using currents of 16.7 amperes, 16.5 amperes, 18.4 amperes, and 17.8 amperes respectively. The
81
variations in these charging currents are due to imprecisions in the charging rate selection at the power
supply. Figure 5-13 presents the surface temperature distribution at the completion of this scenario.
The distribution varies within the range of approximately 1°C. Even though the Charging and Flat
processes have similar current rates 39.5% more heat is generated and 41.1% more heat is lost on
average during the charging process. Figure 5-14, it presents the surface temperature distribution over
Figure 5-13: Final visual simulation of the temperature distribution along the external battery surface
82
Figure 5-14: Experimental temperature distribution on the external battery surface for 17.8A
During this process, heat generation reaches a peak at approximately 65% SoC with 0.84W
(Figure 5-15), as well as a maximum mean heat loss average of 0.4W. Average heat generation and
heat lost is 0.23W and 0.22W respectively. Even though heat generation is considerably higher in this
scenario than the Flat discharge case, the temperature difference in the experiment reached a peak of
2.5°C (see Figure 5-17). Figure 5-16 shows the curve for the difference between the open circuit
voltage and the nominal voltage. This difference is an important component in the Ohmic Heating term
in equation (4-4). Comparing Figure 5-7 to Figure 5-16, the voltage difference is considerably reduced
in the case of the charging mode. This difference may be a result of the current level used in the
experiment.
83
Figure 5-15: Experimental heat generated and lost during battery charging for the charge condition.
84
Figure 5-17: Simulated temperature and Experimental temperature obtained at the centre of the front
As mentioned before, in electric vehicles downhill driving charges the battery pack through the
regenerative braking system. This process requires 114 minutes on average to be completed (to charge
a battery from a SOC of 0% to 100%). The experiment was conducted in 25°C, 30°C, 35°C, and 40°C
with currents of 30.9 amperes, 31.0 amperes, 31.0 amperes, and 31.0 amperes. In Figure 5-18, the
maximum temperature gradient is 0.5°C, which is lower than the previous case, where the temperature
reaches 1°C during charge since the experiment length is only half the charging time. The temperature
distribution in the lower portion of the cell, including the lower sides and bottom, is similar in all the
85
ambient cases. On the upper portion however, the aluminum and copper connectors create a different
heat pattern. As for Figure 5-19, it presents the temperature distribution over the battery surface at the
Figure 5-18: Final visual simulation of the temperature distribution along the external battery surface
86
Figure 5-19: Experimental temperature distribution on the external battery surface for 31.0A charging.
The heat generation reaches maximum peak at 65% SoC with 1.67W (Figure 5-20), two times
higher than during the charging, while the maximum heat loss decreases to 70% SoC with 0.5W. The
maximum temperature difference between the beginning and the completion of the experiment is 3.5°C
(Figure 5-21), 1°C higher than during Charging mode. In both Charging and Downhill modes, the heat
generating term commences as a negative number and increases over time. This results in a delay in
87
Figure 5-20: Experimental Heat generated and lost during battery charging for the Downhill condition.
Figure 5-21: Simulated temperature and Experimental temperature from the centre of the front face
88
5.3.6 Single cell – Conclusion
The previous section regarding the single cell case presented results for the discharging and
charging conditions for 40°C ambient temperature. Here, Table 5-1 summarises all the results obtained
for the single cell study case for the other ambient temperatures (25°C, 30°C, 35°C and 40°C). The
Charging, Downhill, Flat, and Uphill modes used current averages of 17.3 amperes, 29.4 amperes, 16.9
amperes, and 44.1 amperes and generated heat of 97.3 W, 251.9 W, 58.9 W, and 564.1 W respectively.
Based on the results presented in Table 5-1, and Figures 5-22 until 5-25, the current level impacts
battery cell heat generation. By contrast, the environment temperature does not significantly impact
heat generation.
Table 5-1: Average Heat lost and generated from the battery pack for each environment temperature.
89
Figure 5-22: Average current and pack heat generated during Charging mode.
Figure 5-23: Average current and pack heat generated during Downhill mode.
90
Figure 5-24: Average current and pack heat generated during Flat mode.
Figure 5-25: Average current and pack heat generated during Uphill mode.
91
In the discharging cases, increasing the current by 2.6 times (from 16.9 amperes in the Flat
scenario and 44.1 amperes in the Uphill case) also increases the average heat generation by 9.6 times
(from 58.9 W to 564.1 W for the whole pack). An increment of 70% in the absolute magnitude of the
current from the Charging to the Downhill mode increases heat generation by 2.6 times. Furthermore,
the charging process generates 1.65 times more heat even though, in absolute numbers, the current for
charging is 1.02 times higher than that used during discharging Flat.
As can be seen in Table 5-1, the environmental temperature has limited effect on heat
generation in Lithium-iron Phosphate battery cells because of the magnitude of the environmental
temperature. Instead, the gap between voltage and open circuit voltage, the current level used, and the
chemical reaction inside the core all contribute considerably more to heat generation. The strongest
impacts though on heat generation are the values of the current and the difference between the battery
voltage and the open-circuit voltage equation (4-4). Typically, an increase in the current level increases
the difference. In the Charging, Downhill, Flat, and Uphill cases, these differences are typically
The current level also indirectly affects heat loss by increasing the amount of heat transferred
from the core to the outer layers. The speed of heat propagation from the core of the cell depends on
the heat capacity (1350 J/kgK) and the thermal conductivity in the y direction (1.145 W/mK), as seen
in Figure 3-18. However, since thermal conductivity in the xz directions is higher (23.9 W/mK), heat
tends to flow more easily in these directions. The difference between the heat loss and the heat
generated causes heat to accumulate inside the battery. Thus, heat will continuously be released into
92
5.4 Module Analysis
For the multi-cell battery module, 15623 tetrahedral elements were used in the mesh to describe
the cross-sectional 2D plane as presented in Figure 5-26. The mesh is refined in the areas representing
the case and the air gap between each cell. Each cell is modeled with the core in the middle, the
aluminum case around it, and an air gap between the cells. Around the cell, it is assumed that
3
For further information about the mathematical Heat Transfer and Heat Generation equations used in
the FEM in each domain of the model, and boundary conditions, see Chapter 4.
93
The results for the heat generation study of the module are reported for the case of Uphill 40°C,
since overall, the findings in this section are somewhat similar to the ones for the single cell case. In
the Commander 5 EV, each module has eight Lithium-iron Phosphate cells connected in series. Due
to the limitations of the number of cells available for the experiment, six cells instead were connected
in series. In the vehicle, a total of fifty modules are used, which are distributed between the front and
back of the machine. The use of modules allows for convenient maintenance. Since the computer which
runs the simulations has limited memory capacity, a 2D model is used to run the simulations instead
of a 3D model. The simulation uses the meridional cross section plane of the module, which is located
halfway down from the top of the battery cell. This region has not got the highest temperature of the
cell surface, which instead is more likely to be observed in the lower part of the cell due to the core
location. The model is designed with a 4mm thick air gap between each cell. The simulation inputs, as
seen in Figure 5-1, come from the data collected during the experimental phase. One of these
parameters is the surface temperature, which is obtained using thermocouples located on the center of
The Uphill 40°C operating mode was chosen for the module simulation since this condition
resulted in the highest heat generation for the single cell case. For the other scenarios, see Appendix
B. Moreover, due to limitations of the power supplies and resistor load bank, the module was charged
and discharged with slightly different currents than the ones used during the single cell experiment.
The currents used were on average 16.5 amperes (4.9% higher), 31.0 amperes (5.3% lower), 16.8
amperes (0.7% higher), and 41.0 amperes (7.2% lower), and they are referred to charging, Downhill,
The widest temperature difference on the surface of the module in the Uphill 40°C scenario
was observed to be approximately 1°C. As expected, the highest temperatures are reached in the cells
located in the interior of the module. The virtual model accurately presents these temperature results
94
obtained from this experiment (Compare Figure 5-27, Figure 5-28, and Figure 5-29). The temperature
is increased by 3°C during the experiment as can be seen in Figure 5-28. In this scenario, heat
Figure 5-27: Temperature distribution in the meridional section of the module during Uphill operation
at 40°C.
95
Figure 5-28: Experimental temperature distribution on the module during Uphill operation at 40°C.
Figure 5-29: Experimental Heat generated and lost during module discharging calculated per cell for
the Uphill condition.
96
Figure 5-30: Simulated temperature and Experimental temperature discharging at 40°C.
Table 5-2 presents the average heat generated and lost calculated during the experiment in each
Table 5-2: Average Heat lost and generated from the battery pack for each environment temperature.
97
The results contained in Table 5-2 confirm that charging processes typically generate more
heat than discharging processes that proceed at the same rate, and that the heat generated is not
considerably influenced by the ambient temperature. Furthermore, even though heat is generated at the
same level as in the single cell case, heat loss is lower since the cell is covered with black tape which
This section calculates the heat emission from each component in the electric vehicle
(Commander 5EV), except for the battery pack, which is evaluated in depth in Chapter 4. The major
heat emitting components that are evaluated include two electric motors, two motor controllers, a
hydraulic motor (which controls the steering and the brakes), a motor controller for the hydraulic pump,
a DC/DC converter, and two chargers. The vehicle also includes an enclosed cab with a heat and air
conditioning (HVAC) which, in addition to being responsible for creating a comfortable environment
inside the cab, is also required for the cooling system. Heat losses from these components is presented
in Figure 2-8, except for the charger and electric motors. This current study also assumes the idle losses
to be zero, even though there is a small drive shaft connecting the electric motor to its respective driving
axle.
Electric motor losses are correlated to internal frictions between the movable parts and to the
electric losses (based on P = I²R). These losses can be summarized into the electric motor efficiency
4
Black tape was used to secure the thermocouples to the cells. In order to have a consistent emissivity
for the entire cell, and therefore have a consistent infra-red temperature measurement, it was decided
that a consistent surface finish was desirable, therefore the entire cell was covered in a single layer of
tape.
98
Figure 5-31: Simulation for the losses in the electric motor used in the Commander 5 EV during
nominal operation (~200 amperes), courtesy of the manufacturer.
The manufacturer of the electric motors estimates that both motors require 22 kW, 53 kW, and
40 kW of power consumption for driving on Flat, Uphill, and Downhill conditions respectively. Table
5-3 shows the expected power consumption, RPM, and vehicle speed for each driving scenario, and
Table 5-3: Result for each electric motor losses for different driving scenarios.
The current value is used to calculate the losses for both the electric motor and the motor
controller. The motor controller manufacturer has provided the curve for the power loss of this
99
Figure 5-32: Motor controller heat loss curve.
The average electric current used for Flat, Uphill, and Downhill conditions are 68, 169, and 123
amperes respectively, giving 711 W, 2269 W, and 1488 W of total energy loss as presented in Table
5-4.
Table 5-4: This table summarizes the power loss estimation of each major component in the
Commander 5EV.
100
Another component that needs to be included for heat evaluation is the hydraulic electric motor
pump, dedicated to brakes and steerage. The pump heat loss calculation is based on the nominal power
requirement of 6.3kW at 325V. The curve for the total heat losses, which is the sum of the moving and
electrical losses, is presented in Figure 5-33. The rotation of the shaft at 4500 RPM and at 20 amperes
Figure 5-33: Simulation for the losses in the hydraulic electric motor used in the Commander 5 EV
during nominal operation (~20 amperes), courtesy from the manufacture.
As well, a DC/DC converter is required in the vehicle to convert battery power from 325V to
a lower voltage for vehicle accessories (24Vdc). With a current of 2kW with 93% efficiency, it
generates 140W in heat loss. The machine is also equipped with an HVAC system, which is used for
the air conditioner and heater. The air conditioner has a capacity of 35000Btu/hr. The heater is rated
at 5kW, if all this energy is converted to electric heat. The system utilises two electric compressors that
each use 2.5kW and have 95% efficiency, resulting in a total of 250W of heat loss. Finally, the charger,
according to the manufacturer, is rated at 11kW which is divided between two units that have each
101
90% efficiency at nominal condition, producing a total heat loss of 1100W. It is important to note that
when on charge mode, the other major components are not be accessible, as seen in Table 5-4.
The total heat produced by the Commander 5EV can be estimated based on the values
presented in Table 5-4, while the results for the total heat generated by the battery pack heat generated
is displayed in Table 5-5, and it is presented in the graphic form in Figures 5-34, 5-35,5-36 and 5-37.
The average battery heat expressed in this section is obtained from the module scenario (the single cell
case has different emissivity in comparison to the original batteries due to the use of black electrical
tape), and its results are extrapolated for the battery system. As for the remain components, some of
the calculations are based on the rated power, and some others on the device efficiency. Both are
summarised on Table 5-4. In the power loss calculation, the transmission loss is not included since, in
the case of the Commander 5EV, each electric motor is connected directed to each axle via a small
Table 5-5: Total Heat generated by the Personnel Carrier Truck with no use of the air conditioner.
102
Figure 5-34: Estimation of the machinery average efficiency under Charging regime.
Figure 5-35: Estimation of the machinery average efficiency under Downhill regime.
103
Figure 5-36: Estimation of the machinery average efficiency under Flat regime.
Figure 5-37: Estimation of the machinery average efficiency under Uphill regime.
104
Figure 2-8 shows that the efficiency of an electric vehicle ranges from 65% (if there is no
regenerative braking), and 82% (when there is regenerative braking). However, in this calculation, the
charging efficiency is included. Nonetheless, if it is assumed only the parasitic losses, which includes
cooling and electric devices, and the electric drive system losses, which count the losses in the electric
motor and the transmission, the efficiency increases to 81.5% (without regen) and 98.5% (with regen).
In the case of the Commander 5 EV, the average overall efficiencies estimated for the whole vehicle
are 89.0% (Charging), 78.0% (Downhill), 88.5% (Flat), and 80.7% (Uphill) when the air-conditioning
Table 5-5 presents the average machine`s efficiency for each charging/discharging condition,
and for different environment temperature. Relative to the other components of the system, the battery
pack is a very efficient device, since its heat generated comprehend no more than 6% of the total heat
developed, being 3.4%, 3.0%, 2.2%, and 5.3% of the total for charging, Downhill, flat, and Uphill
mode, respectively. In these calculations, it is assumed that all the components are being used
simultaneously, except during charging (see Table 5-4). For this reason, the efficiency drops between
Charging and Downhill modes. As for the discharging current, when is increased, the machine`s
efficiency is slightly decreased. The average efficiency for Charging, Downhill, Flat, and Uphill are
As a side note, it is included Table 5-6 about the average efficiency under air conditioning
usage.
105
Table 5-6: Total Heat generated by the Personnel Carrier Truck with air conditioner usage.
Although the use of air conditioning in electric vehicles decreases considerably the efficiency
of the machine, since it consumes 10.26kW, its heat impact is barely noticeable due to the compressor
high efficiency (95%). When in use, the system efficiency drops in 0.8%, 1.3%, and 0.6% for Downhill,
The impact of air conditioning usage is considerably lower when compared to the heater, as
can be seen in Table 5-7. However, it is unlikely that heaters will be used with a frequently in
underground environment. The side effect of using HVAC system is that the machine`s range is severe
impact by it. This information can be inferred by the efficiency reduction, especially when the heat is
in operation.
106
Table 5-7: Total Heat generated by the Personnel Carrier Truck with heater usage.
From Table 5-7, the efficiency decreases expressively under Downhill, Flat, and Uphill regime
to 65.6%, 65.6%, and 73.8%, respectively, which represents a decrement on the efficiency in relation
Another important point is that the regenerative brake only charges the battery if the state of
charge is below 90%. This limit is set to not damage the Lithium-iron Phosphate batteries, since
crossing it can induce to memory effect, which reduces the battery life span. Driving the machine from
the top to the bottom of the mine, starting with nearly completed battery capacity, can lead to cross this
limit under the Downhill driving process. If it does, the machine has a backup system which directs
the current from the battery to a resistor mounted at the back of the machine. This resistor is coupled
to a back radiator through a water line, and it is rated at 25kW. Since during Downhill, 40.3 kW can
be produced on average, the braking resistors have the capacity to slow down the machine by
107
transforming 25 kW of kinetic energy into thermal heat. Thus, it is highly recommended to not charge
the machine at the top of the shaft if the intention is to drive it down the mine.
108
6 Conclusions
The goal of this study is to evaluate heat generated from battery cells that are designed to
be used in electric vehicles and how the thermal signature is influenced by both the
driving/charging condition and the ambient temperature. These topics and questions are important
for the mining industry in operating electric mining underground vehicles. From the mining
perspective, the movement towards EVs technologies is a way to mitigate the heat generated by
diesel engines, which is generally the third major source of heat, and to provide a healthier work
environment for miners with no diesel particulate matter or hazardous gases emitted. Indeed,
financial savings in mining operation and a healthier environment in accordance with the most
recent mining regulations can be achieved when underground machines emit a lower rate of heat.
Electric machines are becoming a viable alternative to traditional diesel vehicles. Higher
efficiency, high torque at low RPM, reduced motor size, reduced mechanical complexity, and lack
of hazardous emissions are among their benefits. Currently, the costs of acquisition and operation
are currently moderately higher than for diesel units with similar bucket capacity. Furthermore,
electric machines have reduced range, have a longer charging time, and require charging
infrastructure, all consequences of the low energy capacity of current battery pack models. Yet,
this may change soon. With new improved cell chemistries being introduced to the market, EVs
From the mine operator’s perspective, one of the EVs’ strongest advantage beside the
aspect of no emissions is the considerably lower heat impact on the immediate environment in
comparison to IC vehicles. Current methods used to estimate heat impact involve an approximate
estimation of heat generation based on the general assumption of the efficiency value. These
109
methods do not describe changes in the heat load as shaped by different driving and charging
requirements. These strategies can cause misleading assumptions that can either overestimate or
underestimate the total heat generation. Another strategy estimates that the heat generation of these
machines is equal to the power consumption when not working against gravity. This strategy is
also equivocal since vehicles are typically used in underground mines up and down slopes, and
some other machines, such as scoop trams, almost always work against gravity. All of these
considerations must be raised if there is an intention to properly evaluate the heat impact of electric
However, calculating the efficiency of the machine is a valid strategy to determine how heat
generation is affected by different driving conditions and environmental temperatures. This study
primarily evaluates the battery pack as it is a critical component in EVs. The heat generation of
the other electrical components is estimated based their efficiencies and rated powers. The study
consists of two analysis focusses: the single cell and the module. One of the purposes of the model
study is to specify interference between individual cells. In both cases, the cells are tested in
situations typically found in a mining environment, being Charging, Downhill, Uphill, and Flat
conditions, all of which are tested in 25°C, 30°C, 35°C, and 40°C ambient temperatures. The
thermal impact is calculated based on the heat generated by the Lithium-iron Phosphate cells, and
The results show that there is a considerable difference between the heat generated and the
heat lost for high discharging/charging currents. This is a consequence of the low thermal
conductivity in the battery core. Furthermore, the heat flux has a preferable direction, but due to
the magnitude of the temperature gradient around the battery cell, these temperature differences
for each face are not expressive. This difference between the heat generated and the heat loss is
110
expressed by the heat accumulated, which is released into the environment after the process of
charging/discharging. For this reason, analysing the heat generation is the most advantageous
The findings of this study show that the current level has a more significant impact on the
battery heat generation than the environmental temperature found in underground mines. The heat
generated by the single cell under the Downhill operating condition is 2.6 times higher on average
than during the Charging condition, and the Uphill condition produces 9.6 times more heat on
average than the Flat mode. The Charging, Downhill, Flat, and Uphill scenarios use current levels
of 17.3 amperes, 29.4 amperes, 16.9 amperes, and 44.1 amperes, producing heat averages of 97.3
W, 251.9 W, 58.9 W, and 564.1 W respectively. However, the heat generation varies considerably
over time because of the reversible entropic term, which can be positive or negative depending on
the state of charge and the current level. The study also determined that the battery produces 39.5%
more heat during the Charging process than the Flat discharge, even though in both scenarios the
battery has a similar absolute current rate. By contrast, for the extreme case of Uphill mode at
40°C, there is a minor temperature difference of only 0.5°C between the external cells in the
The analysis of the heat generated by the Commander 5 EV, a personnel carrier mining
electric truck designed by Tracks and Wheels, shows that the battery pack, with 400 cells in total,
does not generate more than 6% of the overall heat generated by the vehicle. On average it
produces 3.4%, 3.0%, 2.2%, and 5.3% of the heat in the Charging, Downhill, Flat, and Uphill
modes respectively. The average total heat generated and the efficiency for the vehicle in the
aforesaid scenarios are respectively 2.4 kW (89.0% efficiency), 8.9 kW (78.0%), 2.5 kW (88.5%),
111
and 10.3 kW (80.7%). This means that higher current levels generally reduce the efficiency of the
machine.
There are a multitude of areas that could serve as fertile ground for future work. Future
studies can use the results of this work to optimize the usage of electric vehicle in underground
mines. These aspects can include: optimal paths for electrical machine usage, the most suitable
charging locations, and the financial gains from the substitution of diesel machines by their electric
counter parts.
In addition, future studies may expand the range of temperature and current. The
temperature range between 25°C and 40°C has no considerable impact on the thermal signature of
Lithium-iron Phosphate battery cells, however, there is a possibility the conclusions may differ for
temperatures under 25°C and above 40°C. Furthermore, the current level requirements vary with
each electrical machine. It is important to increase their scope since each electrical machine has
different current requirements. By increasing their range, the findings can be applied to different
machines used underground. Another important point is the investigation of the difference between
using a constant environmental temperature and the actual measured temperature in the simulation.
Overall, the proposed ideas for future work may expand the scope of understanding of both
7 References
112
Al Hallaj S., Maleki, H., Hong, J.S., and Selman, J.R. (1999), Thermal modeling and design
Anon (2014); JoyGlobal P&H, Generation 2 Wheel Loader Product Overview, P&H Mining
AUSIMM. 2012. Cost Estimation Handbook. 2nd ed. The Australian Institute of Mining and
Barnes D., Miller A. (2005). Fuelcell-Powered Front-End Loader Mining Vehicle. Retrieved on April
Bandhauer, T.M., Garimella, S., & Fuller, T.F. (2011). A Critical review of Thermal Issues in Lithium-
Baz-Dresch JJ, Bickel KL, Watts WF (1993). Evaluation of catalyzed diesel particulate filters in an
underground metal mine. Minneapolis, MN: US Department of the Interior, Bureau of Mines,
RI 9478.
Bernardi D., Pawlikowski E., Newman J. (1985) A General Energy Balance for Battery Systems,
Betounay M.C. (2012). The Green Mining Vehicle-Green Energy Road Map. April 12, 2018, from
http://www.mdec.ca/2012/S4P3-Young.pdf
Buchmann, I. (2016, February 21). BU-205: Types of Lithium-ion. Retrieved March 02, 2016, from
http://battery university.com/learn/article/types_of_lithium_ion.
Burke J. (2017). GMSG Recommended Practices for Battery Electric Vehicles in Underground
content/uploads/2017/04/20160726_UG_Mining_BEV-GMSG-WG-v01-r01_published.pdf
CALB, China Aviation Lithium Batteries. (2014). CAM72F I, Aluminum alloy shelled, rechargeable
https://evwest.com/support/CAM72FI%20Final.pdf
113
Calisir T., Baskaya S., Yazar H. O., & Yucedag S. (2015). Experimental Investigation of Panel
Radiator Heat Output Enhancement for Efficient Thermal use Under Actual Operating
Carder D.K. (1999). Performance evaluation of exhaust after-treatment devices used for emissions
control on diesel engines employed in underground coal mines [Thesis]. Morgantown, WV:
Catenacci, M., Verdolini, El, Bosetti, V., Fiorese, G., & Ameli, N. (2012). Going Electric: Survey on
the Future of Battery Technologies for Electric Vehicles. SSRN Electronic Journal SSRN
Journal.
“Cat Loader Powered by Hydrogen Fuel Cells” (2007). Sudbury Mining Solutions Journal.
Chan, C. C. (2007). The State of the Art of Electric, Hybrid, and Fuel Cell Vehicles. Proceedings of
Chen, D., Jiang, J., Kim, G., Yang, C., & Pesaran, A., (2016). Comparison of different cooling methods
for Lithium-iron Phosphateion battery cells. Applied Thermal Engineering, 94, 846-854.
Doi:10.1016/j.applthermaleng.2015.10.015
Chen, S., Wan, C., & Wang, Y. (2005). Thermal analysis of lithium-ion batteries. Journal of power
CIPEC (2005): “Benchmarking the Energy Consumption of Canadian Underground Bulk Mines.”
Canadian Industry Program for Energy Conservation. Natural Resources Canada, 2005.
Deglon, P. & Hemp, R.,(1980). Second International Mine Ventilation Congress: Reno, NV, November
4-8, 1979. New York, NY: Society of Mining Engineers of American Institute of Mining
Demers M., Labelle D., Wilson R. and Matikainen O. (2010, October 5). Hybrid Underground Loader
– 2010 Project Update, presentation given at the MDEC 2010 Conference, Toronto, Ont.,
Canada.
114
De Souza E., Cost Saving Strategies in Mine Ventilation (2015). Accessed on 1st February, 2018.
http://www.airfinders.ca/wp-content/uploads/2015/05/Cost-Saving-Strategies-in-Mine-
Ventilation.pdf
Diesel to Electric: What is the Future of Mining? (2016, June). Accessed on 11 th April 2018.
http://www.skidsteers.net.au/BlogPostView.asp?BlogPostID=22
Electrical Motors and Heat Loss. (n.d.). Retrieved March 15, 2016, from
http://www.engineeringtoolbox.com/electrical-motor-heat-loss-d_898.html
Emissivity Values of Common Materials (2007) Retrivied October 10, 2017, from
http://support.fluke.com/find-ales/Download/Asset/3038318_6251_ENG_A_W.PDF.
EPA – United State Environmental Protection Agency. (2002). A Comprehensive Analysis of Biodiesel
Enderlin, W., (1973), Evaluating Underground Heat Sources in Deep Mines, M. S. thesis, Mont. Coll.
Minl. Sci. Tech., Butte; also USBM, O.F.R.,Research Contract No. GO 122137, June. 78-86.
Ellison G.N., Thermal Computational for Electronic Equipment, Van Nostrand Reinhold, New York,
1969.
Fleet, M. (2012); Personal communication on “Hybrid Drive Electric Mining Loaders”, Longview,
Texas, USA
Forbush S. (2001). Diesel emissions reduction program. Unpublished paper presented at the Mining
Forgez, C., Vinh Do D., Friedrich G., Morcrette M., and Delacourt C.. (2010). Thermal Modeling of a
Web.
Fuel Cell Research Lab (nd). Fuel Cell Hybrid Bus Program. Retrieved July 17, 2018 from
115
http://www.me.udel.edu/research_groups/prasad/research/fcb.html
https://www1.eere.energy.gov/hydrogenandfuelcells/tech_validation/pdfs/fcm04r0.pdf
Garcia-Valle, R., & Lopes, J. A. P. (2013). Electric Vehicle Integration into Modern Power Networks.
Gautam, D., Musavi, F., Edington, M.m Eberl, W., & Dunford, W. G. (2011). “An automotive on-
board 3.3 kW battery charger for PHEV application.” (2011) IEEE Vehicle Power and
Propulsion Conference.
Hall.
Halim, A., Kerai, M. (2013). Ventilation Requirements for ‘Electric’ Underground Hard Rock Mines
Hartman, H. L., Mutmansky, J. M., & Wang, Y. J. (1982). Mine ventilation and air conditioning. New
Hartman, H.L., Mutmansky, J.M., Ramani, R.V., and Wang, Y.J., (1997). Mine Ventilation and Air
Conditioning, Third Edition. John Wiley & Sons, New York, U.S.A., 730 p.
Hodkinson, R., & Fenton, J. (2000). Current EV design approaches. Lightweight Electric/Hybrid
Hosni, M.H., Jones, B.W., and Xu, H. (1999). Experimental results for heat gain and
Hsu, J. S. (2000). A Machine Approach for Field Weakening of Permanent-Magnet Motors. SAE
Elsevier/Newnes.
Hussain, I. (2003). Electric and hybrid vehicles: Design fundamentals. Boca Raton, FL: CRC Press.
Hussain, I. (2011). Electric and hybrid vehicles: Design fundamentals. Second Edition. Boca Raton,
Incropera, Frank P., and David P. DeWitt (2002). Fundamentals of Heat and Mass Transfer. New
Jacobs, W. (28 October, 2013). Electric LHDs in Underground Hard Rock Mining: A Cost/Benefit
Jacobs, W., Hodkiewicz, M. R., & Braunl, T. (2014). A cost-benefit analysis of electric loaders to
reduce diesel emissions in underground hard rock mines. 2014 IEEE Industry Application
Jang J.C., Rhi S.H. (2010). Battery Thermal Management System of Future Electric Vehicles with Loop
https://www.oemoffhighway.com/electronics/article/11224086/electrification-of-underground-
mining-equipment
Klein H, Lox E, Kreuzer T, Kawanami M, Ried T, Bächmann K (1998). Diesel particulate emissions
of passenger cars: new insights into structural changes during the process of exhaust after-
treatment using diesel oxidation catalysts. Warrendale, PA: Society of Automotive Engineers,
Kumar, L., & Jain, S. (2014). Electric propulsion system for electric vehicular technology: A review.
Leonida, C. (2017, June). Battery electric: The death of diesel underground? MiningMagazine, 48-54.
Nensen, P., & Lundkvist, A. (2005). From 167 GWH to 72 GWH - Ventilation on Demand in LKAB's
Iron Ore Mine, Malmberget. Proc IPPC Conference, Innovative Energy Efficiency Examples of
Different Industrial Secotros - Energy Efficiency in the Cement, Metal and Petrochemical
Industry.
Niculuta, M.C. (2012). Thermal Analysis of a LiFePO 4 Battery Pack, Master’s Thesis.
Norris, J. (2013): Personal Comunication with Ozdogan M. and Ozdogan H., P&H LeTourneau
Marks, J R, (2012). Airflow specification for metal and non-metal mines, in Proceedings of 14th
US/North American Mine Ventilation Symposium (eds: F Calizaya and M Nelson), pp 191-195,
Marinov, A., & Valchev, V. (2009). Power loss reduction in electronic inverters trough IGBT-
regulatory structure in Canada and the United States of America: in Queensland Mining
Industry Health & Safety Conference, 2007 Step Change to a Safety Future, Townsville, 11 pp.
<http://www.qrc.org.au/conference/_dbase_upl/mcginn_Controlling%20Diesel%20Emissions.
pdf>
Thermal analysis of a stoneware panel covering radiators. Applied Energy, 131, 248-256.
118
Miller, A. R., van den Berg G., Barnes D.L., Eisele R.I., Tanner D.M., Vallely J.M., and Lassiter D.A.
(2012). Fuel Cell Technology in Underground Mining. The Southern African Institute of Mining
and Metallurgy.
MK Powered. (2010). Sealed Valve Regulated (SVR) Gelled Electrolyte Lead-acid Batteries:
<http://www.mkbattery.com/pdf/technical_manual.pdf>
Monem, M. A., Trad, K., Omar, N., Hegazy, O., Mantels, B., Mulder, G., Bossche P., Mierlo, J. V.
Moore, P. 2010. Plugging the Gap Underground. Mining Magazine, November, 40-46.
Mollenhauer, K., & Tschöke, H. (2010). Handbook of diesel engines. Berlin: Springer. 420-425.
OMEGA. (n/a). Emissivity of Common Materials. Retrieved on April 5 th, 2017, from
https://www.omega.com/literature/transactions/volume1/emissivityb.html
Ozdogan, M., and Ozdogan H. (2015); Diesel-Electric Mining Loaders – A Case Study, IMCET 2015,
Antalya, Turkey.
Norton P, Vertin K, Clark NN, Lyons DW, Gautam M, Goguen S, et al. [1999]. Emissions from buses
with DDC 6V92 engines using synthetic diesel fuel. Warrendale, PA: Society of automotive
Paas N (1999). Dry system technology management system. In: Proceedings of the Mining Diesel
Paraszczak, J., Laflamme, M., Fytas, K. 2013. Electric load-haul-dump machines: real alternative for
Paraszczak, J., Svedlund, E., Fytas, K., & Laflamme, M., a, (2014). Electrification of loaders and
in Deep Metal Mines. International Conference on Mine Planning and Equipment Selection.
Dresden, Germany.
Pesaran A.A. (2001). Battery Thermal Management in EVs and HEVs Issues and Solutions. Advanced
Pesaran A.A. (2002). Battery Thermal Models for Hybrid Vehicle Simulations, Journal of Power
Pesaran A. A., Vlahinos A., and Burch S. D.. (1997). Thermal Performance of EV and HEV Battery
Pignon, J.. (2005). Diesel Engines: Design and Emissions. Johnson Matthey Technology Review, a
Rao Z., Wang S (2011). A Review of Power Battery Thermal Energy Management. Renewable and
Quilliam, J. H. (1974). The Ventilation of South African gold mines. Cape Town: Chamber of Mines
Regan, R. (2008). Guideline for the Management of Diesel Engine Pollutants in Underground
Environments. Mine Safety Operations Division, New South Wales Department of Primary
http://www.resourcesandenergy.nsw.gov.au/__data/assets/pdf_file/0011/419465/MDG-29.pdf>
Regan, R. (2013, June 14). Safety Bulletin - Diesel Emissions in Mines. Retrieved March 27, 2016,
03-Diesel-emissions-in-mines.pdf>
Righettini G. and Mousset-Jones P. (2004). Ventilation Savings with Fuel Cell Vehicles A Cost-Benefit
120
Ryan M. & Rakochy D. (2012, October 2-4). Considerations when deciding between and electric
Loader and a diesel Loader. Paper presented at MDEC – Mining diesel emissions council,
SAE Hybrid Committee (2011). SAE Charging Configurations and Ratings Terminology: Technical
<http://www.sae.org/smartgrid/chargingspeeds.pdf>
ouvertes.fr/tel-01298416/document
Sandivik. (2010). LH514E: Technical specification. Retrieved March 24, 2016, from
<http://www.miningandconstruction.sandvik.com/sandvik/5100/SAM/Internet/cl01023.nsf/All
docs/Products%2a5CLoad%2aand%2ahaul%2amachines%2a5CUnderground%2aloaders%2a
%2a28LHDs%2a29%2a2ATORO%2a1400E/$FILE/Technical%20specification%20Sandvik%
20LH514E-06.pdf>
Santiago, J. D., Bernhoff, H., Ekergård, B., Eriksson, S., Ferhatovic, S., Waters, R., & Leijon, M.
Sayadi A. R., Lashgari A., & Paraszczak J. (2012). Hard-rock LHD cost estimation using single and
Schatz R., Nieto A., Dogruoz C., and Lvov S. (2015). Using modern battery systems in light duty
mining vehicles, International Journal of Mining, Reclamation and Environment, 29:4, 243-
Schnakenberg, G. H., & Bugarski, A. D. (2002). Review of technology available to the underground
121
mining industry for control of diesel emissions (United States of America, U.S. Department of
Health and Human Services). Pittsburgh, PA: U.S. Dept. of Health and Human Services, Public
Health Service, Centers for Disease Control and Prevention, National Institute for Occupational
Sengupta, M. (1990). Mine environmental engineering. Boca Raton, FL: CRC Press. 21-23.
Smil, V. (2010). Prime Movers of Globalization: The History and Impact of Diesel Engines and Gas
Soylu, S. (2011). Electric vehicles: the benefits and barriers. Rijeka: In Tech.
Suberu, M. Y., Mustafa, M. W., and Bashir, N. (2014). “Energy Storage Systems for Renewable energy
Power Sector Integration and Mitigation of Intermittency”. Renewable and Suitainable Energy
Takahashi M., Tobishima S., Takei K., and Sakuari Y. “Characterization of Lithium-iron
Takano Y., Takeno M., Hoshi N., Chiba A., Takemoto M., Ogasawara S., and Rahman M. A., “Design
and analysis of a switched reluctance motor for next generation hybrid vehicle without PM
The carbon trust. Energy Efficiency Office, UK Government. Good Practice Guide 76-Continuous
Steel Reheating Furnaces: Specification Design and Equipment. 1993. Retrieved on May 24,
Thermal Conductivity of common Materials and Gases (n.d.) Retrieved August 12, 2018, from
https://www.engineeringtoolbox.com/thermal-conductivity-d_429.html
Thomas K. E. and Newman J. (2003). Thermal Modeling of Porous Insertion Electrodes Journal of
Tollinsky N., Electric Vehicle Purpose-built for Deep Mining. January 27, 2017. Sudbury Mining
122
Solutions Journal. Retrieved on March 19, 2018, from
http://www.sudburyminingsolutions.com/electric-vehicle-purpose-built-for-deep-mining.html.
alnor/datalogger/9545/main/tsi-alnor_9545_datalogger.jpg
Valicek, P. Fuel cell technology in underground mining (2014). The 6th International Platinum
Conference, ‘Platinum–Metal for the Future’, The Southern African Institute of Mining and
Metallurgy.
Varaschin, J., & De Souza, E. (2015, June 20). The 15th North American. Economics of diesel fleet
Varaschin J. (2016), The Economic Case for Electric Mining Equipment and Technical Considerations
Vergne, J.N., (2008). Hard rock miners handbook. Ed. Edmonton: Stantec Consulting Ltd, 174.
Victoria A. Fisher, M.P.H (July, 2012). Diesel Exhaust Exposure in Miners Linked to Lung Cancer.
newsletter/2012-07/research-publications/diesel-exhaust
Vo, T. T., Chen, X., Shen, W., & kapoor, A. (2015). “New charging strategy for lithium-ion batteries
based on the integration of Taguchi method and state of charge estimation.” Journal of Power
West, J. (1994). DC, induction, reluctance and PM motors for electric vehicles. Power Engineering
Where the Energy Goes: Gasoline Vehicles, US Department of Energy. Retrieved on July 19, 2017,
from https://www.fueleconomy.gov/feg/atv.shtml#data-sources
Where the Energy Goes: Electric Cars, US Department of Energy. Retrieved on July 19, 2017, from
https://www.fueleconomy.gov/feg/atv-ev.shtml#data-sources.
123
Wood, A. (2014); Personal Communication with Ozdogan M. and Ozdogan H., Joy Global (UK)
Wood, a. (n.d.); Electric Drive Mining Class Wheel Loaders as Primary Production Tools. Retrieved
https://www.academia.edu/11694100/Electric_Drive_Mining_Class_Wheel_Loaders_as_Prim
ary_Production_Tools
Woof M., “Technology for Underground Loading and Hauling Systems Offers Exciting Prospects,”
World Health Organization, International Agency for Research on Cancer. (June 12, 2012). IARC:
DIESEL ENGINE EXHAUST CARCINOGENIC [Press release]. Retrieved November 30, 2017,
from https://www.iarc.fr/en/media-centre/pr/2012/pdfs/pr213_E.pdf
Yong, J. Y., Ramachandaramurthy, V. K., Tan, K. M., & Mithulananthan, N. (2015). A review on the
state-of-the-art technologies of electric vehicle, its impacts and prospects. Renewable and
Yonghuang Y. (2015). A Study on the Thermal Behavior and Capacity Recovery of Lithium-ion
Batteries and Their Thermal Management Using Heat Pipes. (Doctoral dissertation). Retrieved
Zeraoulia, M., Benbouzid, M. E., & Diallo, D. (2006). Electric Motor Drive Selection Issues for HEV
Propulsion Systems: A Comparative Study. IEEE Trans. Veh. Technol. IEEE Transactions on
124
In the current appendix is presented the results for the single cell case during 25°C, 30°C, and
35°C ambient temperature. The case for 40°C is excluded in this chapter since it is already mentioned
during section 5.3. Some of the results obtained from the infrared camera might be slightly divergent
from the thermocouples. This can be caused by uncertainties in the camera measurement, capture
range, and the fact that the chamber needs to be open to capture the picture. The maximum temperature
obtained in the thermal imaging is the maximum temperature in the target area, as for the temperature
scale, it is the result of the temperature range in the whole field of view.
Figure 8-1: Final visual simulation of a temperature gradient along the external battery surface for
16.7A charging. Total running time of 207 minutes.
125
Figure 8-2: Thermal picture from the battery at 207 minutes for 16.7A charging, 25°C ambient.
Figure 8-3: Experimental Generated heat and Lost during battery charge for the charging condition.
126
Figure 8-4: Simulated temperature and Experimental temperature obtained during charging at 25°C.
127
8.2 Charging – 30°C
Figure 8-5: Final visual simulation of a temperature gradient along the external battery surface for
16.5A charging. Total running time of 210 minutes.
Figure 8-6: Thermal picture from the battery at 210 minutes for 16.5A charging, 30°C ambient.
128
Figure 8-7: Experimental Generated heat and Lost during battery charge for the charging condition.
Figure 8-8: Simulated temperature and Experimental temperature obtained during charging at 30°C.
129
8.3 Charging – 35°C
Figure 8-9: Final visual simulation of a temperature gradient along the external battery surface for
18.4A charging. Total running time of 188 minutes.
Figure 8-10: Thermal picture from the battery at 188 minutes for 18.4A charging, 35°C ambient.
130
Figure 8-11: Experimental Generated heat and Lost during battery charge for the charging condition.
Figure 8-12: Simulated temperature and Experimental temperature obtained during charging at 35°C.
131
8.4 Downhill – 25°C
Figure 8-13: Final visual simulation of a temperature gradient along the external battery surface for
29.7A Downhill. Total running time of 116 minutes.
Figure 8-14: Thermal picture from the battery at 116 minutes for 29.7A charging, 25°C ambient.
132
Figure 8-15: Experimental Generated heat and Lost during battery charge for the Downhill condition.
Figure 8-16: Simulated temperature and Experimental temperature obtained during Downhill at
25°C.
133
8.5 Downhill – 30°C
Figure 8-17: Final visual simulation of a temperature gradient along the external battery surface for
28.4A Downhill. Total running time of 122 minutes.
Figure 8-18: Thermal picture from the battery at 122 minutes for 28.4A charging, 30°C ambient.
134
Figure 8-19: Experimental Generated heat and Lost during battery charge for the Downhill condition.
Figure 8-20: Simulated temperature and Experimental temperature obtained during Downhill at
30°C.
135
8.6 Downhill – 35°C
Figure 8-21: Final visual simulation of a temperature gradient along the external battery surface for
29.8A Downhill. Total running time of 116 minutes.
Figure 8-22: Thermal picture from the battery at 116 minutes for 29.8A charging, 35°C ambient.
136
Figure 8-23: Experimental Generated heat and Lost during battery charge for the Downhill condition.
Figure 8-24: Experimental Generated heat and Lost during battery charge for the Downhill condition.
137
8.7 Flat – 25°C
Figure 8-25: Final visual simulation of a temperature gradient along the external battery surface for
17.2A discharging. Total running time of 200 minutes.
Figure 8-26: Thermal picture from the battery at 200 minutes for 17.2A discharging, 25°C ambient.
138
Figure 8-27: Experimental Generated heat and Lost during battery discharge for the flat condition.
Figure 8-28: Experimental Generated heat and Lost during battery discharge for the flat condition.
139
8.8 Flat – 30°C
Figure 8-29: Final visual simulation of a temperature gradient along the external battery surface for
16.9A discharging. Total running time of 205 minutes.
Figure 8-30: Thermal picture from the battery at 205 minutes for 16.9A discharging, 30°C ambient.
140
Figure 8-31: Experimental Generated heat and Lost during battery discharge for the flat condition.
Figure 8-32: Experimental Generated heat and Lost during battery discharge for the flat condition.
141
8.9 Flat – 35°C
Figure 8-33: Final visual simulation of a temperature gradient along the external battery surface for
16.9A discharging. Total running time of 205 minutes.
Figure 8-34: Thermal picture from the battery at 205 minutes for 16.9A discharging, 35°C ambient.
142
Figure 8-35: Experimental Generated heat and Lost during battery discharge for the flat condition.
Figure 8-36: Simulated temperature and Experimental temperature obtained during discharging at
35°C.
143
8.10 Uphill – 25°C
Figure 8-37: Final visual simulation of a temperature gradient along the external battery surface for
44.3A discharging. Total running time of 78 minutes.
Figure 8-38: Thermal picture from the battery at 78 minutes for 44.3A discharging, 25°C ambient.
144
Figure 8-39: Experimental Generated heat and Lost during battery discharge for the Uphill condition.
Figure 8-40: Simulated temperature and Experimental temperature obtained during discharging at
25°C.
145
8.11 Uphill – 30°C
Figure 8-41: Final visual simulation of a temperature gradient along the external battery surface for
44.3A discharging. Total running time of 78 minutes.
Figure 8-42: Thermal picture from the battery at 78 minutes for 44.3A discharging, 30°C ambient.
146
Figure 8-43: Experimental Generated heat and Lost during battery discharge for the Uphill condition.
Figure 8-44: Simulated temperature and Experimental temperature obtained during discharging at
30°C.
147
8.12 Uphill – 35°C
Figure 8-45: Final visual simulation of a temperature gradient along the external battery surface for
43.7A discharging. Total running time of 79 minutes.
Figure 8-46: Thermal picture from the battery at 79 minutes for 43.7A discharging, 35°C ambient.
148
Figure 8-47: Experimental Generated heat and Lost during battery discharge for the Uphill condition.
Figure 8-48: Simulated temperature and Experimental temperature obtained during discharging at
35°C.
149
9 Appendix B – Module Case Study Results
As for the appendix A, the results here presented refer to the module case, related to the
charging, Downhill, flat, and Uphill modes, and for ambient temperature (25°C, 30°C, 35°C, and
40°C). The graph results provided are related to the average between the cells in the module. The
module is the result of 6 cells connected in series and operated simultaneously. For the whole battery
pack in the unit, the cell heat results need to be multiplied by 400, the total amount of cells. The
maximum temperature obtained in the thermal imaging is the maximum temperature in the target area,
as for the temperature scale, it is the result of the temperature range in the whole field of view.
9 Charging – 25°C
Figure 9-1: Final visual simulation of a temperature gradient along the external battery surface for
16.5A charging. Total running time of 210 minutes.
150
Figure 9-2: Thermal picture from the battery at 210 minutes for 16.5A charging, 25°C ambient.
Figure 9-3: Experimental Generated heat and Lost during battery charge for the charging condition.
151
Figure 9-4: Simulated temperature and Experimental temperature obtained during charging at 25°C.
152
9.1 Charging – 30°C
Figure 9-5: Final visual simulation of a temperature gradient along the external battery surface for
16.4A charging. Total running time of 210 minutes.
Figure 9-6: Thermal picture from the battery at 210 minutes for 16.4A charging, 30°C ambient.
153
Figure 9-7: Experimental Generated heat and Lost during battery charge for the charging condition.
Figure 9-8: Simulated temperature and Experimental temperature obtained during charging at 30°C.
154
9.2 Charging – 35°C
Figure 9-9: Final visual simulation of a temperature gradient along the external battery surface for
16.5A charging. Total running time of 210 minutes.
Figure 9-10: Thermal picture from the battery at 210 minutes for 16.5A charging, 35°C ambient.
155
Figure 9-11: Experimental Generated heat and Lost during battery charge for the charging condition.
Figure 9-12: Simulated temperature and Experimental temperature obtained during charging at 35°C.
156
9.3 Charging – 40°C
Figure 9-13: Final visual simulation of a temperature gradient along the external battery surface for
16.5A charging. Total running time of 210 minutes.
Figure 9-14: Thermal picture from the battery at 210 minutes for 16.5A charging, 40°C ambient.
157
Figure 9-15: Experimental Generated heat and Lost during battery charge for the charging condition.
Figure 9-16: Simulated temperature and Experimental temperature obtained during charging at 40°C.
158
9.4 Downhill – 25°C
Figure 9-17: Final visual simulation of a temperature gradient along the external battery surface for
30.9A discharging. Total running time of 112 minutes.
Figure 9-18: Thermal picture from the battery at 112 minutes for 30.9A charging, 25°C ambient.
159
Figure 9-19: Experimental Generated heat and Lost during battery discharge for the Downhill
condition.
Figure 9-20: Simulated temperature and Experimental temperature obtained during discharge at
25°C.
160
9.5 Downhill – 30°C
Figure 9-21: Final visual simulation of a temperature gradient along the external battery surface for
31.0A charging. Total running time of 111 minutes.
Figure 9-22: Thermal picture from the battery at 111 minutes for 31.0A charging, 30°C ambient.
161
Figure 9-23: Experimental Generated heat and Lost during battery charge for the Downhill condition.
Figure 9-24: Simulated temperature and Experimental temperature obtained during charging at 30°C.
162
9.6 Downhill – 35°C
Figure 9-25: Final visual simulation of a temperature gradient along the external battery surface for
30.1A charging. Total running time of 111 minutes.
Figure 9-26: Thermal picture from the battery at 111 minutes for 31.0A charging, 35°C ambient.
163
Figure 9-27: Experimental Generated heat and Lost during battery charge for the Downhill condition.
Figure 9-28: Simulated temperature and Experimental temperature obtained during charging at 35°C.
164
9.7 Downhill – 40°C
Figure 9-29: Final visual simulation of a temperature gradient along the external battery surface for
31.0A charging. Total running time of 111 minutes.
Figure 9-30: Thermal picture from the battery at 111 minutes for 31.0A charging, 40°C ambient.
165
Figure 9-31: Experimental Generated heat and Lost during battery charge for the charging condition.
Figure 9-32: Simulated temperature and Experimental temperature obtained during charging at 40°C.
166
9.8 Flat – 25°C
Figure 9-33: Final visual simulation of a temperature gradient along the external battery surface for
16.8A discharging. Total running time of 206 minutes.
Figure 9-34: Thermal picture from the battery at 206 minutes for 16.8A discharging, 25°C ambient.
167
Figure 9-35: Experimental Generated heat and Lost during battery discharge for the flat condition.
Figure 9-36: Simulated temperature and Experimental temperature obtained during discharging at
25°C.
168
9.9 Flat – 30°C
Figure 9-37: Final visual simulation of a temperature gradient along the external battery surface for
16.8A discharging. Total running time of 206 minutes.
Figure 9-38: Thermal picture from the battery at 206 minutes for 16.8A discharging, 30°C ambient.
169
Figure 9-39: Experimental Generated heat and Lost during battery discharge for the flat condition.
Figure 9-40: Simulated temperature and Experimental temperature obtained during discharging at
30°C.
170
9.10 Flat – 35°C
Figure 9-41: Final visual simulation of a temperature gradient along the external battery surface for
16.8A discharging. Total running time of 205 minutes.
Figure 9-42: Thermal picture from the battery at 205 minutes for 16.8A discharging, 35°C ambient.
171
Figure 9-43: Experimental Generated heat and Lost during battery discharge for the flat condition.
Figure 9-44: Simulated temperature and Experimental temperature obtained during discharging at
35°C.
172
9.11 Flat – 40°C
Figure 9-45: Final visual simulation of a temperature gradient along the external battery surface for
16.2A discharging. Total running time of 206 minutes.
Figure 9-46: Thermal picture from the battery at 206 minutes for 16.2A discharging, 40°C ambient.
173
Figure 9-47: Experimental Generated heat and Lost during battery discharge for the flat condition.
Figure 9-48: Simulated temperature and Experimental temperature obtained during discharging at
40°C.
174
9.12 Uphill – 25°C
Figure 9-49: Final visual simulation of a temperature gradient along the external battery surface for
41.1A discharging. Total running time of 84 minutes.
Figure 9-50: Thermal picture from the battery at 84 minutes for 41.1A discharging, 25°C ambient.
175
Figure 9-51: Experimental Generated heat and Lost during battery discharge for the Uphill condition.
Figure 9-52: Simulated temperature and Experimental temperature obtained during discharging at
25°C.
176
9.13 Uphill – 30°C
Figure 9-53: Final visual simulation of a temperature gradient along the external battery surface for
41.1A discharging. Total running time of 84 minutes.
Figure 9-54: Thermal picture from the battery at 84 minutes for 41.1A discharging, 30°C ambient.
177
Figure 9-55: Experimental Generated heat and Lost during battery discharge for the Uphill condition.
Figure 9-56: Simulated temperature and Experimental temperature obtained during discharging at
30°C.
178
9.14 Uphill – 35°C
Figure 9-57: Final visual simulation of a temperature gradient along the external battery surface for
40.9A discharging. Total running time of 85 minutes.
Figure 9-58: Thermal picture from the battery at 85 minutes for 40.9A discharging, 35°C ambient.
179
Figure 9-59: Experimental Generated heat and Lost during battery discharge for the Uphill condition.
Figure 9-60: Simulated temperature and Experimental temperature obtained during discharging at
35°C.
180
9.15 Uphill – 40°C
Figure 9-61: Final visual simulation of a temperature gradient along the external battery surface for
40.9A discharging. Total running time of 85 minutes.
Figure 9-62: Thermal picture from the battery at 85 minutes for 40.9A discharging, 40°C ambient.
181
Figure 9-63: Experimental Generated heat and Lost during battery discharge for the Uphill condition.
Figure 9-64: Simulated temperature and Experimental temperature obtained during discharging at
40°C.
182
183