Data Acquisition To Analyze The Driver and His Race Car

CRANFIELD UNIVERSITY
Massimo Bottiglieri
DATA ACQUISITION TO ANALYSE THE DRIVER AND HIS RACE CAR
SCHOOL OF APPLIED SCIENCES
MSc THESIS
CRANFIELD UNIVERSITY
SCHOOL OF APPLIED SCIENCES
Academic Year 2006-2007
Massimo Bottiglieri
DATA ACQUISITION TO ANALYSE THE DRIVER AND HIS RACE CAR
Supervisor:
John Nixon
September 2007
This thesis is submitted in partial fulfillment of the requirements

for the degree of Master of Science in Motorsport Engineering and Management
Cranfield University 2007. All rights reserved. No part of this publication may be
reproduced without the written permission of the copyright owner.
Table of contents
Table of contents
Table of contents ........................................................................ 2
List of figures ............................................................................. 6
Acknowledgements...................................................................... 11
Preface.................................................................................... 13
Abstract ................................................................................... 14
Chapter I
Basic Electronics ..................... Error! Bookmark not defined.
Voltage ............................................................................................. 15
Current ............................................................................................. 15
Ohms Law ......................................................................................... 16
Kirchoffs Current law ........................................................................... 16
Capacitance ....................................................................................... 17
Strain Gauges...................................................................................... 19
Hall Effect ......................................................................................... 21
Piezoelectricity ................................................................................... 25
Voltage Divider Circuit........................................................................... 25
Chapter II
1
Position Sensors ................................................ Error! Bookmark not defined.

1.1
Linear potentiometers..................................................................... 27
1.1.1
Generic Features of wirewound Potentiometers .................................. 28
1.1.2
Hybrid Track Technology .............................................................. 28
1.1.3
The Operating Principle ............................................................... 29
1.2
Sensors.................................................................... 26
Linear Variable Differential Transformer ........... Error! Bookmark not defined.
1.2.1
Ratiometric Operation ................................................................. 35
1.2.2
Differential Operation ................................................................. 36
1.3
Rotary Potentiometers .................................................................... 37
1.4
Rotary Variable Displacement Transformers........................................... 38
1.5
Ride Height Sensor ......................................................................... 39
Proximity Sensor .................................................................................. 42

2.1
Inductive sensors ........................................................................... 42
2.2
Capacitive Sensors ......................................................................... 42
2.3
Ultrasonic Sensors.......................................................................... 42
2.4
Magnetic Sensors ........................................................................... 43
2.4.1
Reed Relay ............................................................................... 43
2.4.2
Wrapped Coil sensor.................................................................... 43
2.4.3
Hall Effect sensor ....................................................................... 43
2.5
Optical Types ............................................................................... 44
MSc Motorsport Engineering & Management
Table of contents
2.6
3
Temperature sensors............................................................................. 46
3.1
TDR or PRT .................................................................................. 46
3.2
Temperature sensing ICs ................................................................. 46
3.3
Thermocouples .......................................... Error! Bookmark not defined.
3.4
Infrared Sensors ............................................................................ 50
Strain and force sensors ......................................................................... 53

4.1
Loadcells and force measurement....................................................... 53
4.2
Accelerometers............................................................................. 53
4.2.1
Based on load cells ..................................................................... 53
4.2.2
Based on Hall Effect .................................................................... 53
4.2.3
Based on Piezoelectricity.............................................................. 54
4.2.4
Based on silicon technology ........................................................... 54
4.3
4.3.1
5
Applications ................................................................................. 44
Yaw Sensor .................................................................................. 56

Piezoelectric principle ................................................................. 56
Speed sensors ..................................................................................... 59

5.1
Timing........................................................................................ 59
5.2
Wheel speed sensing....................................................................... 60
5.3
GPS speed sensing.......................................................................... 61
Gas flow sensors .................................................................................. 62

6.1
Sprung flap .................................................................................. 62
6.2
Calorimetric techniques................................................................... 62
Pressure Sensors .................................................................................. 63
Voltage and Current Sensors .................................................................... 64
Lambda sensors ................................................ Error! Bookmark not defined.
Literature References .................................................................................. 68
Chapter III
1
Aerodynamics...................................................................................... 70
1.1
Basic Race Engineering................................................. 70

CL and CD coefficients ...................................................................... 72
Tire model ......................................................................................... 75

2.1
Forces acting on tire....................................................................... 75
2.2
Toe, camber and caster................................................................... 76
2.2.1
Toe ........................................................................................ 76
2.2.2
Camber ................................................................................... 77
2.2.3
Caster.................................................. Error! Bookmark not defined.
2.3
Friction Coefficient ........................................................................ 80
2.3.1
Basics ..................................................................................... 80
2.3.2
Circle of friction ........................................................................ 81
Table of contents
2.3.3
Slip Ratio ................................................................................. 82
2.4
Slip Angle ................................................. Error! Bookmark not defined.
2.5
Neutral Steering, Understeer and Oversteer .......................................... 86
Damping technology.............................................................................. 90
Conclusion ......................................................................................... 95
Literature References .................................................................................. 97
Chapter IV
Driving Techniques ..................................................... 98
Introduction ....................................................................................... 98
Seating Position ................................................................................... 99

2.1
Open Wheeler............................................................................... 99
2.2
Saloon car ................................................................................... 99
Pedals............................................................................................. 101
Braking and Accelerating ...................................................................... 103

4.1
4.1.1
Braking phases......................................................................... 104
4.1.2
Cadence Braking ...................................................................... 106
4.1.3
Threshold braking..................................................................... 106
4.1.4
Left foot braking ...................................................................... 106
4.2
5
Braking ..................................................................................... 103
Accelerating.............................................. Error! Bookmark not defined.
Cornering......................................................................................... 109
Literature References ................................................................................ 112
Chapter V
1
Dijon Prenois (France) ......................................................................... 113

1.1
Ferrari F430 GT3 FFSA: J.B. Scheier vs P. Ballay ................................... 114
1.1.1
Corners 1, 2 and 3 .................................................................... 115
1.1.2
Turn 4,5,6 and 7 ...................................................................... 118
1.2
Ferrari F430 GT3 FFSA: J.B. Scheier vs J.B. Scheier ............................... 123
1.3
Ferrari F430 GT3: P. Ballay vs P. Ballay .............................................. 127
Spa-Francorchamps............................................................................. 131
2.1
2.1.1
Data Analysis ......................... Error! Bookmark not defined.
Ferrari 430 GT3 FIA GT 24 hours of Spa .............................................. 131

Pascal Ballay vs Damien Chanard................................................... 132
Silverstone (UK) ................................................ Error! Bookmark not defined.

3.1
Ferrari F430 Challenge: Gentleman driver vs Professional Driver................ 149
3.1.1
Copse Maggotts Becketts Chapel.............................................. 150
3.1.2
Stowe The Vale Club Abbay.................................................... 154
Literature Reference ................................................................................. 162
Chapter VI
1
Diagnosis ................................................................163
ABS ASR not working.......................................................................... 163
Table of contents
1.1
Problem.................................................................................... 163
1.2
Theoretical background ................................................................. 163
1.3
Solution .................................................................................... 164
Speed and brake problems .................................................................... 167

2.1
Problem.................................................................................... 167
2.2
Theoretical Background ................................................................. 167
2.3
Solution .................................................................................... 168
2.3.1
Pit speed limiter ...................................................................... 168
2.3.2
ASR ...................................................................................... 168
Gearbox problems .............................................................................. 171

3.1
Problem.................................................................................... 171
3.2
Solution .................................................................................... 172
3.2.1
Step 1 : Looking at the values of every channel ................................. 172
3.2.2
Step 2: Replacing the broken item................................................. 173
3.2.3
Step 3 : Calibrating the rotary potentiometer ................................... 174
3.2.4
Step 4 : Checking the gearbox while engine worksError!
Bookmark
not
defined.
4
Engine Starting problems ...................................................................... 176

4.1
Problem.................................................................................... 176
4.2
Theoretical Background ................................................................. 176
4.2.1
Dashboard .............................................................................. 176
4.2.2
CAN-line ................................................................................ 177
4.3
Solution .................................................................................... 179
Literature References ................................................................................ 185
Conclusion ...............................................................................186
List of figures
List of figures
Fig. I.1: Kirchhoffs current law [38] .........................................................................................17
Fig. I.2: Capacitor [10]; (b) Capacitance [37] ............................................................................... 18
Fig. I.3: (a) Wire lay-out of strain gauge [30]; (b) Strain gauge parts [11]; (c) Strain gauges [29] .................20
Fig. I.4: (a) Working principle of strain gauges [11]; (b) Rosettes [11]..................................................20
Fig. I.5: Wheatstone bridge as shown by [29] ...............................................................................21
Fig. I.6: Hall Effect Principles [10] ............................................................................................23
Fig. I.7: Voltage divider circuit ............................................................. Error! Bookmark not defined.
Fig. II.1: (a) Linear Potentiometer by Penny & Giles [6,12]; (b) Linear Pot (rectangular housing) [17] ...........28
Fig. II.2: (a) Hybrid Track Linear Potentiometer [6,12]; (b) Polymeric track [5] ......................................29
Fig. II.3: Ladder principle of the hybrid track [5] ..........................................................................30
Fig. II.4: Kirchoffs first law applied on the ladder structure [5] ........................................................30
Fig. II.5: Electrical transformers principle of an LVDT in differential mode [8] ......................................32
Fig. II.6: (a) Cut away 3D view of LVDT [10]; (b) cut away of LVDT [8] ..............................................33
Fig. II.7 : (a) Core at left side ; (b) Core at right side [9] ................................................................. 34
Fig. II.8:Voltage vs Displacement of LVDT [8] ............................................................................... 34
Fig. II.9: Ratiometric Operation, Ratio vs Displacement [8] .............................................................. 35
Fig. II.10: Differential operation, Voltage vs Displacement [8]...........................................................36
Fig. II.11: Rotary potentiometer (internal) [11] ............................................................................ 37
Fig. II.12 : (a) Conventional rotary pot [9]; (b) Rotary Micropotentiometer by McLarenElectronics [17]..........37
Fig. II.13 : Rotary potentiometer using the hybrid track technology by Penny and Giles [6,12] ....................38
Fig. II.14 : RVDT by Penny & Giles [6,12]; RVDT by McLarenElectronics [17] ..........................................38
Fig. II.15: RVDT measures greater arc due to the wheel mounted on its shaft gearing ratio ....................39
Fig. II.16 : (a) IR Ride height sensor principle ; (b) IR Ride-height sensor by McLaren Electronics [17]............40
Fig. II.17: Possible position of Ride Height sensors on a sportscar .......................................................40
Fig. II.18 : (a) Laser Ride Height Sensor on F1 [13] ; (b) and (c) Laser Ride-height Unit from Optimess
[15] ; (d) Laser Ride-height Unit by Bosch [14] ............................................................................. 41
Fig. II.19: Bosch Motorsport range of inductive proximity sensors [14] .................................................42
Fig. II.20: Hall Effect Sensor by Bosch Motorsport [14] .................................................................... 44
Fig. II.21: Transmissive optical sensor principle ............................................................................44
Fig. II.22: Passive Wheelspeed Trigger Assembly [16]......................................................................45
Fig. II.23: Crankshaft angle inductive measurement via differential principle (two sensors) [28]..................45
Fig. II.24: PRT by Fluke [23]....................................................................................................46
Fig. II.25: Microscopic Temperature sensing IC [17]........................................................................47
Fig. II.26: Millivolts vs Temperature for each type of thermocouple [24] ..............................................48
Fig. II.27: (a) AD 595 CJC Thermocouple amp circuit diagram; (b) Thermocouple amplifier [25] ..................48
Fig. II.28: Compensating cables and connector block ......................................................................49
Fig. II.29: Thermocouples : (a) K Type by Magneti Marelli [19]; (b) J Type and (c) K Type by McLaren
Electronics [17]...................................................................................................................50
Fig. II.30: (a) Infrared temperature sensor [17]; (b) Triple Infra red temperature sensor [17] .....................50
Fig. II.31: IR temperature measurement of brake disk .................................................................... 51
Fig. II.32: (a) Application of IR sensors on tyre temperature measurement; (b) Mounted on a racecar [27] .....51
List of figures
Fig. II.33: (a) Irisys Thermal Imager [26] and its (b) measurement results [26] ....................................... 52
Fig. II.34: Hall Effect Accelerometer [28].................................................................................... 54
Fig. II.35: Accelerometer based on the piezoelectric principles [28] ...................................................54
Fig. II.36: Micromechanical accelerometer [28] ............................................................................55
Fig. II.37: (a) 3- and 2-axis accelerometers [14]; (b) 3-axis accelerometer [17] ......................................55
Fig. II.38: Knock sensor based on the accelerometer principles [28].................................................... 56
Fig. II.39: Yaw [28] and gyroscope [28] ......................................................................................56
Fig. II.40: Yaw sensor in fork-type based on piezoelectricity [28]; (a) straight line [28]; (b) Yaw [28]............57
Fig. II.41: Drum-type piezoelectric sensor [28] ............................................................................. 57
Fig. II.42: (a) Micromechanical sensor [28]; (b) Under the microscope .................................................58
Fig. II.43: Yaw sensor by Bosch [14] .......................................................................................... 58
Fig. II.44: 3-axis yaw sensors by McLaren Electronics [17] ................................................................58
Fig. II.45: (a) Optical (Infra Red) Lap trigger [33] and (b) Optical Lap marker receiver [33] of the Ferrari
430 Challenge racecar...........................................................................................................59
Fig. II.46: (a) Transponder, (b) Holder and (c) entire transponder data system with decoder [34] ................60
Fig. II.47: Principle of the inductive proximity sensor (wheelspeed measurement) [28].............................60
Fig. II.48: Magnetised wheel incorporated into bearing [28] .............................................................61
Fig. II.49: (a) GPS Systems by Racelogic [35] and (b) receiver/calculator [35] ........................................61
Fig. II.50: Sprung flap air flow rate sensor [28] ............................................................................. 62
Fig. II.51: Electrical diagram of the sensing wire principle [28]..........................................................63
Fig. II.52: (a) Membrane for low pressures [28]; (b) Diaphragm on chip with reference vacuum [28]; (c)
Metal Diaphragm for high pressures [28] .....................................................................................63
Fig. II.53: Pitot Pressure sensor [17]; (b) Hydraulic Pressure Sensor [19]; (c) Miniature pressure sensor
[17] .................................................................................................................................64
Fig. II.54: Different types of pressure measurements......................................................................64
Fig. II.55: Lambda Sensor [19] and its voltage output vs a ................................................................65
Fig. II.56: Schematic overview of implementation of the oxygen sensor [36]..........................................66
Fig. II.57: Lambda sensor schematically described [28] ...................................................................66
Fig. II.58: Conversion rate of the catalytic converter [36] ................................................................ 67
Fig. III.1: Lift, Drag and total force on a thin plate [5] .................................................................... 71
Fig. III.2: Distribution of measured pressure coefficients around a 2D generic car shape [5] .......................71
Fig. III.3: CL vs pitch angle at different ride heights .......................................................................73
Fig. III.4: CD vs pitch angle at different ride heights .......................................................................73
Fig. III.5: L/D vs pitch angle at different ride heights .....................................................................73
Fig. III.6: Ideal Ride Heights [6] ...............................................................................................74
Fig. III.7: Changes in ride height while turning and braking [6] ..........................................................74
Fig. III.8: Forces acting on tires [6] ...........................................................................................75
Fig. III.9: Bridgestone tire on Ferrari F1 car [11] ........................................................................... 76
Fig. III.10: (a) Toe-in and (b) toe-out [12] ................................................................................... 76
Fig. III.11: Definition of negative and positive camber; (b) Example of negative camber [10]......................78
Fig. III.12: Camber on a race car [12] ........................................................................................ 79
Fig. III.13: Caster [12] and [11] ................................................................................................79
Fig. III.14: Vertical load acting and resulting in maximum lateral and longitudinal forces [6]. .....................81
List of figures
Fig. III.15: (a) Ideal Circle of traction and (b) real circle of traction....................................................82
Fig. III.16: Slip ratio example [6] ..............................................................................................83
Fig. III.17: (a) Longitudinal friction coefficient variation (b) plotted for several normal forces [6]................83
Fig. III.18: Contact patch area vs vertical load [6] .........................................................................84
Fig. III.19: Slip Angle definition with aligning torque and lateral force [6].............................................84
Fig. III.20: Maximum lateral force versus slip angle in function of vertical load ......................................86
Fig. III.21: Oversteer and understeer .........................................................................................87
Fig. III.22: Neutral, Oversteer and understeer [11] ........................................................................87
Fig. III.23: The lateral force sequence in a corner [6] ..................................................................... 88
Fig. III.24: Neutral steer [6] ....................................................................................................88
Fig. III.25: Understeer and oversteer [6] ..................................................................................... 89
Fig. III.26: Oversteering old formula 1 car [11] .............................................................................89
Fig. III.27: Typical plot of damping force vs piston velocity [6]. .........................................................91
Fig. III.28: Typical force versus displacement curve for a damper [6]. .................................................91
Fig. III.29: 4 way dampers, as on the F430 GT3 [9].........................................................................92
Fig. III.30: Suspension movement Formula Renault 2.0 Eurocup at Magny-Cours (France) gathered via the
AIM data acquisition system (Maxime Martin as driver)....................................................................92
Fig. III.31: Damping force versus piston speed for (a) Low Speed damping, (b) High speed damping and
(c)combined damping [9]. ......................................................................................................93
Fig. III.32: Suspension analysis tool with the AIM data analysis software ...............................................94
Fig. IV.1: Cut away view of McLaren F1 to see seating position [11] ....................................................99
Fig. IV.2: (a) Surface Body/Seat; (b) Arm Position; (c) Leg Position [8] .............................................. 100
Fig. IV.3: (a) Normal set ; (b) Pedal set of Supercar Ferrari Enzo (F1 system, thus no clutch pedal); (c)
Pedals of seven time F1 champion M. Schumacher [11] ................................................................. 102
Fig. IV.4: (a) Normal Braking and accelerating [7]; (b) X-Y graph of gs perceived by the car. Note the
negative gs in combined behaviour (cornering-baking) ................................................................. 105
Fig. IV.5: A WRC driver using left foot braking in a Peugeot WRC of 2003. Note the larger brake pedal........ 107
Fig. IV.6: Data acquisition via AIM system on the Formula Renault 2.0 driven by Maxime Martin ................ 108
Fig. IV.7: Geometric compared to possible Racing line [4,8] ........................................................... 110
Fig. IV.8: Geometric racing line compared to Ideal Racing Line (Late Apex) [4,8].................................. 111
Fig. V.1: (a) Picture taken at Kessel Racing (CH); (b) The GT3 on the Dijon-Prenois track........................ 114
Fig. V.2: Turn 1, 2 and 3, blue for Jobo and red for Pascal ............................................................. 115
Fig. V.3: Time difference between Jobo and Pascal (middle line)..................................................... 116
Fig. V.4: Dijon-prenois track ................................................................................................. 116
Fig. V.5: Turn 4, 5, 6 and 7................................................................................................... 118
Fig. V.6: Negative ramp of deceleration, bad throttle application and wrong rpm at upshift (Pascal).......... 119
Fig. V.7: Johan Boris Scheier, Champon Clio Cup France 2006, Driver of the Morgan Aero in FIA GT with
Lafitte and FSSA GT3 driver with the Thierry Boutsen Energy Racing team .......................................... 120
Fig. V.8: Braking performance, Higher Gs for Jobo (blue trace) in longitudinal (X) and lateral
acceleration (Acc_Y) .......................................................................................................... 120
Fig. V.9: X-Y Graph for Gs, Better trail braking for Jobo (rounder at bottom) ..................................... 121
Fig. V.10: X-Y graph for Engine performance, RPM vs Speed for both drivers ....................................... 123
Fig. V.11: Pressure checker and damper unit (without bumpstop) [3] ................................................ 124
List of figures
Fig. V.12: Race 1 (red) vs Free Practice (blue). Huge steering problems ............................................. 125
Fig. V.13: Race 1 compared to Race 2. Best laps both by Jobo Scheier............................................... 126
Fig. V.14: Higher amount of full throttle application in Race 2. ....................................................... 127
Fig. V.15: Some extra red dots during combined cornering and braking, thus better trail braking .............. 127
Fig. V.16: Free Practice compared to First Qualify of Pascal. .......................................................... 128
Fig. V.17: Hystogram of throttle application. Best laps from Free Practice vs Qualify 1 (blue) from Pascal ... 129
Fig. V.18: X-Y of Gs. Best laps from Free Practice vs Qualify 1 (blue) from Pascal ................................ 130
Fig. V.19: Entire Best laps of Damien (Red) and Pascal (Blue) in Qualify 3........................................... 133
Fig. V.20: Raidillon Kemmel Les Combes. Red traces for Damien and Blue traces for Pascal. ................ 134
Fig. V.21: Rivage-Turn 6 Pouhon - Fagnes................................................................................ 136
Fig. V.22: Turns 5, 6, 7 and 8 at Spa........................................................................................ 137
Fig. V.23: Track of Spa seen form the sky ................................................................................. 138
Fig. V.24: Stavelot Blanchimont. Huge difference between Damien (red) and Pascal (blue).................... 138
Fig. V.25: Turn 9-10 (Stavelot) and 11-12 (Blanchimont)................................................................ 139
Fig. V.26: Chicane Source hairpin. Source taken in different ways. ................................................. 140
Fig. V.27: Chicane and Source Hairpin (both new since 2007).......................................................... 141
Fig. V.28: Throttle Application: Relatively 16% more full throttle for Damien compared to Pascal.............. 142
Fig. V.29: Brake application from both drivers (a) Damien Chanard ; (b) Pascal Ballay............................ 143
Fig. V.30: Speed visually demonstrated on track. Huge speed differences in the advantage of Damien (a)
compared to Pascal (b). Raidillon, before Les Combes, Blanchimont and before Chicane .............. 144
Fig. V.31: RPM versus Speed, generally higher revs for Damien (red dots) ........................................... 145
Fig. V.32: Oil pressure, Oil Temperature, Water Temperature and RPM. ............................................ 146
Fig. V.33: Damien (a) is more in 6th gear than Pascal (b). .............................................................. 147
Fig. V.34: (a) F430 GT during qualify 3 at 24 hours of Spa and (b) after the race .................................. 147
Fig. V.35: Ferrari F1 Clienti and Ferrari 250 GTO from 1962 ........................................................... 148
Fig. V.36: Massimo Bottiglieri and his Belgian driver with the Ferrari F430 Challenge at Silverstone............ 149
Fig. V.37: Copse Maggotts Becketts Chapel. Blue trace for the gentleman driver and red for the pro..... 150
Fig. V.38: Wheelspeeds, brakes, ABS and 3 math channels for the wheelspeed difference ....................... 151
Fig. V.39: Left right left of Becketts .................................................................................... 153
Fig. V.40: Stowe The Vale Club Abbay ................................................................................ 154
Fig. V.41: Wheelspeed channels and brake channels from the second sector of Silverstone. ..................... 155
Fig. V.42: Stowe The Vale - Club Abbay ................................................................................ 156
Fig. V.43: Brutal steering input at Vale and drift for Club (red) ................................................... 156
Fig. V.44: Last part of circuit, where Louis loses everything. Bridge- Priory Brooklands Luffield
Woodcote........................................................................................................................ 158
Fig. V.45: Bridge Priory Brooklands Luffield - Woodcote........................................................... 158
Fig. V.46: Throttle application for both drivers on their respective best lap ........................................ 160
Fig. V.47: Respective time elapsed for each sector per driver (blue Louis Machiels)............................... 160
Fig. V.48: Speed plotted on the track for each driver. (a) Louis Machiels, (b) professional driver............... 161
Fig. VI.1: ASR switch; central console at the right side.................................................................. 163
Fig. VI.2: (a) ABS system [2]; (b) Screen (Magneti Marelli) with alarm leds on dashboard [5]..................... 164
Fig. VI.3: (a) Scroll button on steering wheel [3] and (b) screen coming up the dashboard [5]................... 165
Fig. VI.4: Wheelspeeds, ABS and ASR channels; Note the 511.9 or Speed_RR ....................................... 166
List of figures
Fig. VI.5: Rear right wheel bearing ......................................................................................... 167

Fig. VI.6: Steering wheel of GT3 and Challenge and its buttons; a=scroll, b=radio, c=pit speed ................. 168
Fig. VI.7: ASR and wheelspeed problems at Spa .......................................................................... 169
Fig. VI.8: Renault Mgane Trophy BTCS of Thierry Boutsen Energy Racing ........................................... 171
Fig. VI.9: a general window showing most of the channels available on the Renault............................... 172
Fig. VI.10 Error list............................................................................................................. 173
Fig. VI.11: Input check : Barrel diag shows an error on the counter (23), just like all the dampers. ............ 173
Fig. VI.12: New rotary potentiometer installed ........................................................................... 174
Fig. VI.13: Calibration of the rotary pot via the general window of Vision 4......................................... 174
Fig. VI.14: (a) Conventional versus (b) CAN wiring [8] ................................................................... 177
Fig. VI.15: Graphical overview of a CAN-bus implementation [8] ...................................................... 178
Fig. VI.16: C-CAN-bus in the F430 Challenge (modified compared to the street version of the F430) [5]....... 180
Fig. VI.17: Visible original NQS dashboard [3] ............................................................................. 181
Fig. VI.18: (a) Installation of fuse and relay boxes in the F430 Challenge [5]; (b) B Section with covers
[5] ................................................................................................................................ 182
Fig. VI.19: List of details about each fuse and relay position in box B .............................................. 183
Fig. VI.20: Names and designation of each fuse and relay in the box [5]............................................. 183
Fig. VI.21: Fuse and relay boxes B in the front compartment [5] .................................................... 184
10
Acknowledgement
Acknowledgements
It is probably one of the most used clichs in the words of gratitude that traditionally
precede a thesis text, but it remains a fact that the accomplishment of an MSc degree is
never the work of a single person. Therefore, I would like to thanks a number of people
for their kind help in the realisation of my thesis, but also of my diploma of MSc
Motorsport Engineering and Management.
In the framework of this MSc thesis, I worked closely together with a number of people
at Francorchamps Motors Antwerpen. Therefore, I have to say a very special thanks
to my boss Guy Fawe, who supported me during this entire year and my thesis, and the
guys at the garage who were there to provide me with some practical tips. I really enjoy
to be part of the Ferrari Maserati family, a family I dreamed about for many years.
Additionally, I start working since the 1st May 2007 at Thierry Boutsen Energy Racing
where I have to thank Olivier Lain and Georges Kaczka for that great period of
learning and working. I had the chance to be on track almost every weekend working
with e.g. Rene Hilhorst (Toyota F1, Mygale F3,etc.) which gave me the ability to learn
a lot about race cars, but also about drivers and tracks. I would like to thank the entire
team for the funny moments, but also for the hard moments where I was able to
discover some aptitudes I thought I did not had. I will never forget my first podium in
Valencia or the victory in collaboration with JMB Racing during the 24 hours of Spa in
the Coupe du Roi.
I really would like to thank John Nixon for the patience he had during my thesis period.
I had a lot of work to do on track and could thus not write my thesis immediately. I
really appreciate that he trusted me, and tried to give me good advice. John, I hope you
will enjoy your retirement and will have a lot of pleasures with your stock car.
Moreover, I would like to say thanks to the entire teaching staff at Cranfield University
for that great year where I learned more than I thought I did
11
Acknowledgement
Apart from these professional relationships, there remain a few people that contributed
to this work, without realising it and certainly without expecting any credit for it.
However, I could never express enough gratitude towards my family, who has
supported me and motivated me all my life. A big part of the reward for obtaining this
MSc degree, is feeling how proud they are of me. Therefore, I dedicate this work to my
parents and my two brothers.
Again, thank you very much everyone. You have made this experience worthwhile.
ing. Massimo Bottiglieri - August 2007
12
Preface
Preface
The content of this thesis is as follows:
Chapter I will introduce some generalities and units used in the electronics, not really
worthwhile to read if the reader has got enough knowledge of electricity and electronics.
In Chapter II will provide a large overview of the main sensors used in Motorsport.
Therefore, many images are used to show the reader how these measurement tools look
like.
Some basics of race engineering are written down in Chapter III. One will be able to
read about aerodynamics, tires and dampers technology. Obviously, the writer had to
limit the amount of text and could thus only provide the reader with a simple overview.
Chapter IV will focus on the way a driver should drive. Here, only some basic driving
techniques will be explained and only the ones used for circuit racing. This excludes
thus rally where a totally different way of driving is used.
In Chapter V some data gathered during the thesis period of Massimo Bottiglieri will be
explained. Three tacks will be involved and two types of Ferraris will be used. Much
more data was gathered by the writer, but time and space was an issue.
Chapter VI will discuss the main problems solved via data acquisition software where
each problem with some theoretical background and the solution are explained Of
course, much more problems occurred, but only a selection of four was maintained to
keep the work in normal sizes.
13
Abstract
Abstract
The race engineer does not only use his experience in conjunction with the drivers
comment, but is now also able to use the modern technique: data acquisition systems.
These systems receive data gathered by sensors and store them in a data logger. After a
driving session, the engineer will be able to empty the data logger via a link between his
computer, which often is a LAN or USB connection. Once the data is stored on the hard
disk of the computer, the race engineer will open the data in dedicated data analysis
software like Wintax Magneti Marelli or Bosch Darab. Subsequently, he will use the
data to discuss with the driver about the past session. Consequently, the driver will be
able to see where it went wrong or well and will thus normally improve if he is not in
shortage of talent. Added, to the drivers comment, the data gathered can help the
engineer to provide the driver with a better set-up of the competition car. Additionally,
the engineer will have the possibility, depending on the amount of channels used, to
detect malfunctions on the car, with or without extra diagnosis software (Waycon or
Vision4).
To be able to get the entire profit of the data acquisition system, the user needs to be
aware of the sensors used, electromechanical engineering, driving techniques and race
engineering. In addition, he needs to have communication skills to talk to the driver and
computer skills to use the full potential of the software. Thus, it is not because one has
got data acquisition on his car that he will be able to turn a Spyker into a McLaren or a
Ferrari.
The reader of this work will notice a huge amount of references. Despite this huge
amount, one has to deliberately chose his references. Often references on the internet
are wrong and some pseudo engineers write inaccurate things about vehicle dynamics or
tires. Also one has to be aware that what is good for one car or driver is not necessarily
good for the other.
Finally, experience, intelligence, patience and being multidisciplinary will provide the
race engineer with a good establishment in the small and hard world of the Motorsport.
The higher one wants to be, the harder and faster he has to work.
14
I. Basic Electronics
Chapter I
1
Basic Electronics
Voltage
The volt [V] is the SI derived unit of electric potential difference or electromotive force
[1,2] It is named in honour of the Italian physicist Alessandro Volta (17451827), who
invented the voltaic pile, the first modern chemical battery.
The volt is defined as the potential difference across a conductor when a current of one
ampere dissipates one watt of power. Hence, its SI representation:
V=
2
W W . s J N . m kg . m2
=
= =
=
A A.s C A.s
A . s3
Current
The Ampere, [A] is a unit of electric current, or amount of electric charge per second.
The ampere is an SI base unit, and is named after Andr-Marie Ampre, one of the main
discoverers of electromagnetism.
The ampere is that constant current which, if maintained in two straight parallel
conductors of infinite length, of negligible circular cross section, and placed 1 meter
apart in vacuum, would produce between these conductors a force equal to 2107
newton per meter of length.
Electric current is the time rate of change or displacement of electric charge. One
ampere represents the rate of 1 coulomb of charge per second.
1A = 1
C
s
The ampere is defined first (it is a base unit, along with the metre, the second, and the
kilogram), without reference to the quantity of charge. The unit of charge, the Coulomb,
is defined to be the amount of charge displaced by a one ampere current in the time of
one second [1,2].
15
Ohms Law
Ohm's law states that, in an electrical circuit, the current passing through a conductor
between two points is proportional to the potential difference (i.e. voltage drop or
voltage) across the two points, and inversely proportional to the resistance between
them. In mathematical terms, this is written as :
R=
V
I
R = Resistance []
V = Potential difference [V]
I
= Current [A]
R is a constant, measured in ohms, called the resistance. The potential difference is also
known as the voltage drop, and is sometimes denoted by E or U instead of V [3].
Kirchoffs Current law
This law is also called Kirchhoff's first law, Kirchhoff's point rule, Kirchhoff's junction
rule, and Kirchhoff's first rule.
The principle of conservation of electric charge implies that:
At any point in an electrical circuit where charge density is not changing in
time, the sum of currents flowing towards that point is equal to the sum of
currents flowing away from that point.
A charge density changing in time would mean the accumulation of a net positive or
negative charge, which typically cannot happen to any significant degree because of the
strength of electrostatic forces: the charge buildup would cause repulsive forces to
disperse the charges.
However, a charge build up can occur in a capacitor (see next section)[38].
16
Fig. I.1: Kirchhoffs current law [38]
Capacitance
The capacitor's capacitance (C) is a measure of the amount of charge (Q) stored on each
plate for a given potential difference or voltage (V) which appears between the plates:
C=
Q
V
C = Capacitance [F]
Q = Charge [C]
V = Voltage [V]
In SI units, a capacitor has a capacitance of one Farad when one Coulomb of charge is
stored due to one Volt applied potential difference across the plates. Since the farad is a
very large unit, values of capacitors are usually expressed in microfarads (F),
nanofarads (nF), or picoFarads (pF).
A capacitor is an electrical device that can store energy in the electric field between a
pair of closely spaced conductors (called 'plates'). A capacitor consists of two
conductive electrodes, or plates, separated by a dielectric.
When current is applied to the capacitor, electric charges of equal magnitude, but
opposite polarity, build up on each plate.
Capacitors are used in electrical circuits as energy-storage devices. They can also be
used to differentiate between high-frequency and low-frequency signals and this makes
them useful in electronic filters [37].
17
Fig. I.2: Capacitor [10]; (b) Capacitance [37]
The capacitance is proportional to the surface area of the conducting plate and inversely
proportional to the distance between the plates. It is also proportional to the permittivity
of the dielectric (that is, non-conducting) substance that separates the plates.
The capacitance of a parallel-plate capacitor is given by:
.A
d
with A d 2
C = Capacitance [F]
= Permissivity of the dielectric [-]
A = Area of the plates [m]

d
= Spacing between plates [m]
18
Strain Gauges
The most common way of measuring strain or force is by the use of strain gauges
invented by Edward E. Simmons in 1938. These depend on the physical fact that the
resistance of a piece of wire is directly related to its physical shape. For any wire, its
resistance R is
R=
.L
A
R = Resistance []
= Resistivity of the wire [m]
L = Length [m]
A = Cross sectional area of wire [m]
Based on the this formula of Pouillet [28], if a wire is stretched, its length will increase
while its cross sectional area will decrease. Both phenomena will tend to increase the
resistance R of the wire. Strain gauges are films of metal shaped to maximise the effect
of strain. They are bonded to components so that they strain at the same rate as the
component [4].
19
(a)
(b)
(c)
Fig. I.3: (a) Wire lay-out of strain gauge [30]; (b) Strain gauge parts [11]; (c) Strain gauges [29]
If the stiffness of the component is known, the load can be calculated from the strain.
By installing rosettes (patterns of strain gauges at defined angles relative to each
other), the direction as well as the magnitude of the strain can be measured [4].
(a)
(b)
Fig. I.4: (a) Working principle of strain gauges [11]; (b) Rosettes [11]
A major problem with strain gauges is that they can be affected by parameters other
than strain. Temperature affects the resistance of the gauge, and will also cause the
components to expand, if it is metal, without load being applied. These problems are
20
overcome by connecting the gauges in a Wheatstone bridge configuration (Fig. I.5) as

described in [4].
Fig. I.5: Wheatstone bridge as shown by [29]
Strain gauges are very time consuming to install, and unless carefully protected can be
easily damaged. However, modern adhesives and coatings have helped remarkably with
these problems [4].
Hall Effect
The Hall effect refers to the potential difference (Hall Voltage) on the opposite sides of
an electrical conductor through which an electric current is flowing, created by a

magnetic field applied perpendicular to the current. Edwin Hall discovered this effect in
1879.
The ratio of the voltage created to the product of the amount of current and the magnetic
field divided by the element thickness is known as the Hall coefficient. It is a
characteristic of the material from which the conductor is made, as its value depends on
the type, number and properties of the charge carriers that constitute the current.
The Hall Effect comes about due to the nature of the current flow in a conductor.
Current consists of the movement of many small charge-carrying "particles" (typically,
but not always, electrons). These charges experience a force, called the Lorentz Force,
when a magnetic field is present that is not parallel to their motion. When such a
magnetic field is absent, the charges follow an approximately straight, 'line of sight'
21
path. However, when a perpendicular magnetic field is applied, their path is curved so
that moving charges accumulate on one face of the material. This leaves equal and
opposite charges exposed on the other face, where there is a dearth of mobile charges.
The result is an asymmetric distribution of charge density across the hall element that is
perpendicular to both the 'line of sight' path and the applied magnetic field. The
separation of charge establishes an electric field that opposes the migration of further
charge, so a steady electrical potential builds up for as long as the current is flowing
[32] .
For a simple metal where there is only one type of charge carrier (electrons) the Hall
voltage VH (Volts) is given by [32]:
VH =
I . B / d
n.e
The Hall coefficient RH is defined as:
RH =
VH
1
=
I .B / d
n .e
= Current across plate length [A]
B = Magnetic Flux density []

d
= Depth of the plate [m]
= electron charge [V]
= Bulk density of the carrier electrons
22
As a result, the Hall Effect is very useful as a means to measure both the carrier density
and the magnetic field.

One very important feature of the Hall Effect is that it differentiates between positive
charges moving in one direction and negative charges moving in the opposite. The Hall
Effect offered the first real proof that electric currents in metals are carried by moving
electrons, not by protons. The Hall effect also showed that in some substances
(especially semiconductors), it is more appropriate to think of the current as positive
"holes" moving rather than negative electrons [28,31,32] .
Fig. I.6: Hall Effect Principles [10]
The figure above demonstrates the Hall Effect for different directions of electric current
and magnetic field:
1. Electrons (not conventional current!)
2. Hall element, or Hall sensor
3. Magnets
4. Magnetic field
5. Power source
23
In drawing "A", the Hall element takes on a negative charge at the top edge (symbolised
by the blue color) and positive at the lower edge (red color). In "B" and "C", either the
electric current or the magnetic field is reversed, causing the polarization to reverse.
Reversing both current and magnetic field (drawing "D") causes the Hall element to
again assume a negative charge at the upper edge .
24
Piezoelectricity
Piezoelectricity is the ability of some materials (notably crystals and certain ceramics)
to generate an electric charge in response to applied mechanical stress. If the material is
not short-circuited, the applied charge induces a voltage across the material. The word is
derived from the Greek piezein, which means to squeeze or press [28].
The piezoelectric effect is reversible in that materials exhibiting the direct piezoelectric
effect (the production of electricity when stress is applied) also exhibit the converse
piezoelectric effect (the production of stress and/or strain when an electric field is
applied). For example, lead zirconate titanate crystals [31] will exhibit a maximum
shape change of about 0.1% of the original dimension.
Voltage Divider Circuit
The principle of the potentiometer is based on the voltage divider circuit, which itself is
based on the first law of Kirchoff.
Fig. I.7: Voltage divider circuit
Vout
R2
=
Vin R1 + R2
Vout= Voltage over R2 [V]
Vin = Voltage applied over entire circuit [V]
R1 = Resistance []
R2 = Resistance []
25
II. Sensors
Chapter II
Sensors
Sensors or transducers represent the link between the external world and the data
collection and analysis system. As the first element in the system, they contribute to the
accuracy and capability of the system to be usable or not. Their characteristic and
working principles must be understood to be able to implement or optimize them in a
data acquisition system. If one wants to use the data acquired, he needs to be sure that
he is reading correct data
Many years ago, race cars were very simple in their electrical design, many racing
vehicles having only an ignition system. In recent years, however, with the increasing
adoption of engine and vehicle management systems, the number of sensors utilized has
grown rapidly. One told that there were more than 1000 channels gathered on a 2007
Formula One. Although, only a few channels are used the engineer to discuss with
drivers and/or other engineers. In case of failure of the F1 though, a huge databank is
available to the engineers to check and eventually find a correlation between data and
failure.
A good definition of a sensor could be [4]:
A component or device which changes its electrical properties in a
consistent and repeatable manner in response to a single external
parameter.
This definition makes several assumptions which need to be considered before
specific sensor types are discussed. This work will be focused on sensors in
conjunction with electronic systems, thus electrical outputs.
In order to be usable, the sensor is required to be [4,7]
Consistent
Repeatable
26
II. Sensors
Linear response to some degree of accuracy across defined operating range
Response time (small enough)
The latter is a more subtle sensor performance characteristic. When the input
parameter to a sensor is modified, it will take a defined amount of time for the
output of the sensor to achieve the value corresponding to the last input level. The
response time fully depends on the technology, type and design of the sensor
employed and can be a few seconds or a few milliseconds. Therefore, the designer
of the system needs be aware of the rate of change of the parameter being measured
to be able to select an appropriate design [7].
As can be seen, lots of parameters have to be taken into account to obtain a correct
sensor in a correct data acquisition system. The following sections will provide the
reader with an overview of the technology available with some advantages and
disadvantages .
1
1.1
Position Sensors
Linear potentiometers
The word potentiometer comes from potential, which refers to voltage potential, or
difference. A potentiometer (sometimes called a pot) is a device for measuring voltage

difference. In fact, a potentiometer is simply a resistor with a variable resistance value.
Typically it is composed of a resistive material through which a current is passed.
Unlike a static resistor, however, the point at which the current exits the resistive
material is variable. Most often this is done by using a brush, which makes electrical
contact with the resistive material, and which can be moved using a handle. By moving
the brush, the voltage potential being measured can be modified, thus the device is a
user-adjustable resistance.
Usually, this is a three-terminal resistor with a sliding contact in the center (the wiper or
brush). If all three terminals are used, it can act as a variable voltage divider. If only two
terminals are used (one side and the wiper), it acts as a variable resistor. Its shortcoming
27
II. Sensors
is that of corrosion or wearing of the sliding contact, especially if it is kept in one

position. A positive point of this type of sensor is its competitive price [4,6,17] .
(a)
(b)
Fig. II.1: (a) Linear Potentiometer by Penny & Giles [6,12]; (b) Linear Pot (rectangular housing)
[17]
1.1.1
Generic Features of wirewound Potentiometers
These are the generic features as given by [5,6]:

9 Absolute Measurement, thus no loss of position upon power down.
9 Simple Electronics
9 Low Current, hence lower cost electronics.
9 Good electromagnetic compatibility (EMC), which is good for motorsport
applications.
9 Intrinsically Safe and thus able to operate in hazardous
areas
without
specific
approvals
9 Quick and easy installation
1.1.2
Hybrid Track Technology
The latest evolution of a regular linear potentiometer is the so-called Hybrid track
Technology by Penny & Giles and was developed during the late 1970's, with the
objective "to combine the advantages of both conductive plastic and wirewound
potentiometers to give optimum sensor performance".
28
II. Sensors
The main advantages are [5]:
Infinite resolution and long life, provided by the conductive plastic track
Stability under extremes of temperature and humidity, provided by the

wirewound element.
Linearity and resistance values are typically those of the wirewound element to
which zero width taps, capable of carrying full rated element current, may be
attached.
The hybrid track consists of a conventional precision wirewound potentiometer element

upon which has been mechanically and electrically bonded a thin film of high resistivity
conductive polymer track. The polymer track is composed of a plastic matrix in which
carbon particles are suspended. The density of these carbon particles controls the
resistivity of the polymer.
The polymer track is 'wiped' by a precious metal contact, giving a virtually infinite
resolution with considerable resistance to wear (since the majority of the current still
flows in the wire, the carbon content of the conductive plastic film is low, and the film
is therefore very hard).
Fig. II.2: (a) Hybrid Track Linear Potentiometer [6,12]; (b) Polymeric track [5]
1.1.3
The Operating Principle
The sheet resistivity of the polymer is chosen to give a track resistance at least three
orders higher than the wirewound element upon which it is deposited. This causes most
29
II. Sensors
of the potentiometer element current to be carried in the wire, with only a very small
proportion in the parallel polymer track.
The hybrid potentiometer element can be represented diagrammatically by a ladder
network of resistors:
Fig. II.3: Ladder principle of the hybrid track [5]
Rw
= resistance of each turn of resistance wire
Rp
= resistance of the polymer per unit length (each unit being the wire pitch)
Ri
= resistance at the interface between the wire and the polymer
Consideration of the current distribution in this ladder network on the figure below
shows that there is an initial increase in the current levels in the polymer film.
Fig. II.4: Kirchoffs first law applied on the ladder structure [5]
Because i1 is at least three orders greater than i2 or i3, the current flowing in the
wirewound element can be considered virtually constant and thus the voltage
increments in each turn of wire are constant, giving linearly increasing voltages at each
point of contact between the wire and the polymer film.
30
II. Sensors
Kirchoff's first law makes it evident that the voltage generated in the polymer film
increases from the low applied voltage end at a non-linear rate until the potential across
the interface between polymer and wire is zero. At this point, the current flow in the
polymer remains sensibly constant, and the voltage increases linearly with track length
until the mirror image of the effect just described is encountered at the high applied
voltage end of the track.
Changes in the resistivity of the polymer can be caused by wear and/or temperature and
humidity variations whilst in service. Contact wear reduces the cross sectional area of
the track, temperature variations cause further polymerisation of the plastic/carbon mix,
reducing its resistivity. Water absorption by the basic plastic matrix causes a physical
expansion of the polymer, reducing the density of carbon and thus increasing resistance
[5] . More details can be obtained in [5], but the main purpose of this work is to provide
the reader with an overview and basic principles.

The main features of the Hybrid Track Potentiometer as given in [5] are:
Due to the conductive plastic on the wire element, there is no metal to metal
contact which directly increases the operational track life and reduces
maintenance and service costs.
Self Compensating for Track Wear.
Linearity, Temperature, Humidity Stability
Virtually Infinite Resolution, hence, very small movements can be sensed.
Low Hysteresis (good for heat dissipation into compact space)
Self Cleaning, Long Life Contact Design
All these features are very important to motorsport where time, development and money
are very important. As told by Frank Williams: I will never sign a check if it doesnt
make my cars go faster!.
31
II. Sensors
1.2
Linear Variable Differential Transformer
The letters LVDT are an acronym for Linear Variable Differential Transformer [6,8].
They are non-contacting common type of electromechanical transducer that can convert
the rectilinear motion of an object to which it is coupled mechanically into a
corresponding electrical signal. LVDT linear position sensors are readily available that
can measure movements as small as a few millionths of a millimetre up to several
centimetres, but are also capable of measuring positions up to 0.5 m [9].
The transformer's internal structure consists of a primary winding centred between a
pair of identically wound secondary windings, symmetrically spaced about the primary.
The coils are wound on a one-piece hollow form of thermally stable glass reinforced
polymer, encapsulated against moisture, wrapped in a high permeability magnetic shield,
and then secured in cylindrical stainless steel housing. This coil assembly is usually the
stationary element of the position sensor [8,9,11].
Fig. II.5: Electrical transformers principle of an LVDT in differential mode [8]
An LVDT Displacement Transducer comprises 3 coils; a primary and two secondaries.

The transfer of current between the primary and the secondaries of the LVDT
displacement transducer is controlled by the position of a magnetic core called the
armature or core.
32
II. Sensors
Fig. II.6: (a) Cut away 3D view of LVDT [10]; (b) cut away of LVDT [8]
When the core is centred with respect to the two secondary windings, they will have the
same magnitude of induced voltages, but the polarity or phasing will be opposite
(resulting in a zero output). When the core is displaced from this null position, the
output amplitude on one secondary coil (Va) increases, while the output amplitude in the
other coil (Vb) decreases. These voltages can be used individually or combined to
produce an output signal proportional to position, dependant upon the method of
demodulation employed (Fig. II.8). The two main methods used are described below
[4,8,9,11]. The figures below demonstrate the output of each coil in schematic way .
33
II. Sensors
Note that with LVDTs, the phase of the output (compared with the excitation phase)
enables the electronics to know which half of the coil the armature is in [8,9].
(a)
(b)
Fig. II.7 : (a) Core at left side ; (b) Core at right side [9]
Fig. II.8:Voltage vs Displacement of LVDT [8]
34
II. Sensors
1.2.1
Ratiometric Operation
A high accuracy method of translating the LVDT output is to measure the secondary
voltages independently to generate a ratio of the difference divided by the sum of these
values.
Ratio =
Va Vb
Va + Vb
Fig. II.9: Ratiometric Operation, Ratio vs Displacement [8]
This configuration is commonly referred to as ratiometric operation and will provide

much higher system accuracy performance than operation in the differential mode.
Ratiometric operation provides [4,8]:
Improved immunity to LVDT supply voltage and frequency variations
Improved immunity to errors due to temperature effects on LVDT sensitivity
Improved frequency and phase response
Improved immunity to common-mode noise on LVDT lines
Improved transducer interchangeability
Additionally, the sum of the secondary output voltages (Va + Vb) is nominally constant
throughout the LVDT stroke range, so it can be used for system error detection in high
integrity systems [4,8].
35
II. Sensors
The strength of the LVDT sensor's principle is that there is no electrical contact across
the transducer position sensing element which for the user of the sensor means clean
data, infinite resolution and a very long life thanks to their robustness and their ability to
operate at high temperatures and in extreme environments [4,6,8,9] .
1.2.2
Differential Operation
LVDTs are normally available with either four or five wires, where the extra wire is the
centre-tap in the output.
When operating in the differential mode, this centre-tap
connection is often not used.

The output is taken across the whole transformer secondary, (see Fig. II.5). In this
connection configuration, when the core is displaced from the centre null position, the
output will increase in-phase with the input in one direction and anti-phase with the
input in the other.
Fig. II.10: Differential operation, Voltage vs Displacement [8]
To derive the position from the LVDT, a modulator is required to provide the primary
a.c. voltage in conjunction with a demodulator to translate the in-phase component
transformer secondary output (Fig. II.10) to a DC signal proportional to position.
When using LVDTs in this differential mode, the output will be directly affected by
changes in supply voltage, operating temperature and supply frequency and is therefore
of lower accuracy.
36
II. Sensors
Thus considering all the features and characteristics of the two operations seen before,
the Ratiometric one will be used.
1.3
Rotary Potentiometers
Both sensing techniques described earlier can be applied to the measurement of a rotary
position. Circular resistive tracks, similar to those in a standard potentiometer can be
used and will result in a positional sensor called rotary potentiometer. The construction
of such a wire-wound circular potentiometer is as follows:
Fig. II.11: Rotary potentiometer (internal) [11]
The resistive element (1) of the shown device is trapezoidal, giving a non-linear
relationship between resistance and turn angle. The wiper (3) rotates with the axis (4),
providing the changeable resistance between the wiper contact (6) and the fixed contacts
(5) and (9). The vertical position of the axis is fixed in the body (2) with the ring (7)
(below) and the bolt (8) (Fig. II.11) [11]. Some examples of conventional rotary pots are
illustrated below with their basic measurements .
(a)
(b)
Fig. II.12 : (a) Conventional rotary pot [9]; (b) Rotary Micropotentiometer by McLarenElectronics
[17]
37
II. Sensors
The Hybrid tracks discussed in the prior section of the linear potentiometers are also
manufactured by Penny & Giles (see figure below) in the rotary form. Multiturn (3, 5
and 10 turn) potentiometers have also been produced using hybrid tracks, but these have
become less popular in recent years [5].
Fig. II.13 : Rotary potentiometer using the hybrid track technology by Penny and Giles [6,12]
1.4
Rotary Variable Displacement Transformers
A rotary variable differential transformer (RVDT) is a type of electrical transformer

used for measuring angular displacement. More precisely, it is an electromechanical
transducer that provides a variable alternating current (AC) output voltage that is
linearly proportional to the angular displacement of its input shaft. When energized with
a fixed AC source, the output signal is linear within a specified range over the angular
displacement.
(a)
(b)
Fig. II.14 : RVDT by Penny & Giles [6,12]; RVDT by McLarenElectronics [17]
38
II. Sensors
RVDTs utilise brushless, non-contacting technology to ensure long-life and reliable,

repeatable position sensing with infinite resolution. Such reliable and repeatable
performance assures accurate position sensing under the most extreme operating
conditions.
Most RVDTs are composed of a wound, laminated stator and a salient two-pole rotor.
The stator, containing four slots, contains both the primary winding and the two
secondary windings. Some secondary windings may also be connected together.
Most RVDTs can only measure through a restricted movement arc (and are thus no
multiturn ones) which typically lies between 60 or 90 degrees. A way to counter this
problem if several revolutions are needed (steering wheel for example) is to use a
gearing ratio as shown on the figure below.
Fig. II.15: RVDT measures greater arc due to the wheel mounted on its shaft gearing ratio
1.5
Ride Height Sensor
The first models of this specialised sensor were mounted on the bottom of the car,
sending an infrared beam, which was reflected off the ground into a detector (Fig. II.16).
39
II. Sensors
(a)
(b)
Fig. II.16 : (a) IR Ride height sensor principle ; (b) IR Ride-height sensor by McLaren Electronics
[17]
Ride height sensors are used on race cars to get a better idea of how the aerodynamic
performance of the car relates to height off the road surface and its attitude on the road.
There can be positional sensors on the suspension as well, but these cannot tell the exact
height of the car off the road because they don not take account of tyre deflection.
Another use for the ride-height sensor is to use that data in conjunction with the
suspension data to determine tyre deflections.
Furthermore, in a formula car there is really only room for one sensor in the centre of
the car on the bottom at either end and thus it cannot take account of body roll and thus
get a good picture of what the tyres on either side of it are doing. On a GT or saloon car
though, there is the possibility to position a laser sensor very close to each wheel (Fig.
II.17).
Fig. II.17: Possible position of Ride Height sensors on a sportscar
40
II. Sensors
New technologies like laser ride height sensors are now used in Formula 1. There can
be seen on a picture of the bulkhead of a Ferrari F1 car (Fig. II.18), that the international
warning symbol for laser radiation is present. That unit is made by Optimess [15] and
works by the same principles as the IR beam. Consequently, a laser shines on the road
surface through a small hole in the bottom of the nosecone. The beam is directed not
exactly straight down, but at a small angle such that reflected light would not come back
at the laser diode, but at an array sensor next to it in the package. The sensor array is
like the sensor in a camera, but tuned to the wavelength of the laser. As the distance
from the ground of the sensor package changes, the spot of reflected light moves on the
sensor array due to the triangular shape created by that small angle.
(a)
(b)
(c)
(d)
Fig. II.18 : (a) Laser Ride Height Sensor on F1 [13] ; (b) and (c) Laser Ride-height Unit from
Optimess [15] ; (d) Laser Ride-height Unit by Bosch [14]
41
II. Sensors
Proximity Sensor
Proximity sensors do not measure position directly, but they detect the presence of, and
sometimes the distance to, objects within their sensing range. Several types are
available, each with their own characteristics.
Proximity sensors are used for a wide variety of Motorsport applications as will be
demonstrated later on in this section. Their use in the Motorsport is thus very important
and detailed information can be obtained at e.g. Bosch Motorsport [14], Pi Research [16]
or McLaren Electronics [16].
2.1
Inductive sensors
This type of proximity sensors are the most popular due to their robustness, simplicity
and price. Inductive sensors normally contain a resonant circuit, the frequency of which
depends on the inductance of a sensing coil. When a metallic object is brought within
the sensing range of the sensor, the inductance of the coil changes, thus modifying the
resonant frequency and causing the sensor output to change its state.
Fig. II.19: Bosch Motorsport range of inductive proximity sensors [14]
2.2
Capacitive Sensors
Capacitive sensors operate in similar manner to the inductive sensors, except that a
capacitive, rather than an inductive sensing element is used. They tend to be slightly
more expensive, but they have the enormous advantage to detect non metallic materials
as well as metals [4].
2.3
Ultrasonic Sensors
Ultrasonic sensors transmit a beam of ultrasound and detect the returning energy
reflected from an object in the beam path. They are more expensive than the other
proximity sensors discussed above, but they are able to detect any material, including
42
II. Sensors
liquids. Additionally, their sensitivity is not altered by the type of material, which is a
huge advantage. Nevertheless, the surface condition of the target may affect sensitivity,
smooth surfaces tending to reflect the beam away from the detector unless normal to the
beam. Ultrasonic proximity sensors can detect objects to ranges of several meters
(useless in Motorsport), but then tend to be larger and heavier than the short range units
[4].
2.4
Magnetic Sensors
There are currently three types of magnetic proximity switches which can used in
Motorsport. Some are less adequate to the harsh environment and are seldom used.
2.4.1
Reed Relay
The reed relay is a magnetic proximity sensor which detects the presence of magnetic
fields, normally generated by magnets. These tend to be too fragile for Motorsport
applications.
2.4.2
Wrapped Coil sensor
An alternative of the reed relay sensor may use a coil wrapped around a magnet within
the sensor. When the magnetic field generated by the magnet is distorted, usually by
moving a piece of steel through the field, a voltage is generated in the coil. One must
know that the sensor is insensitive to stationary objects.
2.4.3
Hall Effect sensor
The Hall Effect sensor is a semiconductor device which is sensitive to changes in local
magnetic field. The Hall Effect sensor works in a similar way to the inductive type.
Additionally, the response time of the Hall Effect sensor is faster than its inductive
concurrent, hence, can operate at higher frequencies. Frequency responses will vary
between 10 Hz and 2 kHz for an inexpensive Bosch [14] Hall Effect sensor. A much
more expensive McLaren Electronics one will have a frequency response between 15
Hz and 20 kHz [17]. These sensors are compact and robust and are thus widely used in
all kind of Motorsport applications.
43
II. Sensors
Fig. II.20: Hall Effect Sensor by Bosch Motorsport [14]
2.5
Optical Types
Two different types of optical sensor exist: Transmissive and Reflective type. More
often, the transmissive type is installed where a LED (light emitting diode) emits light.
An LDR (lightdependant resistor) will lower its resistance if it absorbs the emitted light
that can pass trough a hole or tooth gap of a wheel.
Fig. II.21: Transmissive optical sensor principle
One should note that for reflective optical sensors, that the measurement is affected by
colour and surface condition of the target. The optical sensors are not widely used in
Motorsport.
2.6
Applications
Proximity sensors are used for a wide range of Motorsport application such as the
generation of timing pulses for engine management by detecting the teeth on a timing
wheel on the crankshaft (Fig. II.23). Additionally, they are used for speed sensing and
44
II. Sensors
as detectors in antilock braking systems (ABS). These detectors for ABS are generally
used for an ASR or ESP system as well like on the Ferrari F430 GT3 or Challenge.
On the figure below, a general overview of the assembly of such a proximity sensor is
provided. There can be seen that several geometries are of huge importance to the good
functioning of the assembly.
Fig. II.22: Passive Wheelspeed Trigger Assembly [16]
Fig. II.23: Crankshaft angle inductive measurement via differential principle (two sensors) [28]
45
II. Sensors
Temperature sensors
The measurement of temperature is essential for the monitoring and control of a wide
range of Motorsport systems. The temperature of cooling water and lubricants has a
direct influence on the performance of power trains; while the temperature of incoming
air, brakes, tyres and even cockpit can influence vehicle performance. All these
temperatures vary widely from ambient to over 1000C. Therefore, it is not surprising
that a number of temperature sensors are used [4].
3.1
TDR or PRT
The simplest of the temperature sensor is obviously the temperature dependant resistor,
the also called TDR. Another name which is often used in the Motorsport world is the
platinum resistance thermometer or PRT. This, as its name implies, is a resistor whose
value varies as the temperature changes. They are simple, cheap, compact, robust and
light weight. In general, however, their resistance temperature response curve is
frequently not linear and they may only operate over a few tens of degrees temperature
range. In addition to these drawbacks, their response to changes in temperature is slow,
usually taking several seconds to achieve new temperature value [4,14,23] .
Fig. II.24: PRT by Fluke [23]
Despite these drawbacks, they are widely used for the measurement of air, oil and water
temperatures. These parameters tend to change slowly and thus the PRTs are suited for
that purpose. Thin film resistors have improved thermal response, but are physically
less robust. These devices can operate at temperatures up to 600C [23].
3.2
Temperature sensing ICs
In recent years, semiconductor devices have become available which are very similar in
specification and performance to TDRs. These rely on the level of conductivity across a
46
II. Sensors
semiconductor junction being temperature dependent. Temperature sensing ICs can

generate more linear outputs, either voltage or current, although at higher cost. There
should be noted that their maximum operating temperature is also limited to a 150C
[17]. Currently, they are widely used, but offer the promise of improved performance
compared with TDRs [4].
Fig. II.25: Microscopic Temperature sensing IC [17]
3.3
Thermocouples
For measurements across a wider range of temperatures, thermocouples are frequently

used. These depend, as their name implies, on the physical effect that if a junction is
made between two metallic alloys, a voltage will be generated which is dependant on
the temperature difference between the ends of the wires. For protection, the
thermocouple junction is frequently contained within a thin stainless steel tube, or in a
threaded metallic housing. The performance of thermocouples is dependant on the
composition of the metals used, and a range of alloys is available, of which the most
common are K, J and R types. An overview of the main properties of these common
types is given in the table below [4,24].
Table 1: Most common thermocouples [4,24]
Type Metal
K
Chrome-nickel (chromel) or
Temperature range
Output voltage
200C to +1100C
41V/C
-40C to +750C
55V/C
Up to 1500C
10V/C
aluminium-nickel (alumel) alloys

J
Iron-constantan
Alloys of Platinum and Rhodium
Other types of thermocouples exist like the B, E, N, S, T and an overview of all the
properties can be found in [24]. As can be seen on the figure below, linearity is not
always present, especially at the extremes of the temperature range.
47
II. Sensors
Fig. II.26: Millivolts vs Temperature for each type of thermocouple [24]
A significant problem with thermocouples is the low output voltages. This makes their
signal sensitive to external noise and interference, and optimum performance is

achieved by amplifying thermocouple signals as close to the sensor as possible.
Additionally, the output is dependant on the temperature difference between the two
ends of the wire, and thus normally on the temperature of the signal conditioning
equipment.
(a)
(b)
Fig. II.27: (a) AD 595 CJC Thermocouple amp circuit diagram; (b) Thermocouple amplifier [25]
48
II. Sensors
Modern thermocouple amplifiers incorporate a separate temperature sensing element

which adds an additional voltage to the output, equivalent to the thermocouples output
voltage at the amplifier temperature, relative to 0C. The element described is the so
called cold junction compensation (CJC) circuit. It ensures that the actual output
temperature can be measured directly. A simple example of the phenomenon is given
below.
4mV = 100C output of thermocouple is dependent of the cold junction (CJ)
temperature
Table 2: CJC example
Current temperatures at junctions
Temp diff
Volts
HJ = 100C
CJ = 25C
75C
3 mV
HJ = 100C
CJ = 50C
50C
2 mV
This means that if no CJC is applied that the direct interpretation of the figures will
provide the user with wrong temperatures.
Because the output is dependent on the composition of the thermocouple wires, it is
necessary to continue these to the thermocouple amplifier or to use special

compensating cables (Fig. II.28). These have the same electrical characteristics as the
thermocouple wires, but are less expensive. One must imagine that the thermocouple
used is a Platinum/Rhodium one These compensating cables must be joined to the
thermocouple wires via a connecting block made of a special alloy which is also
compatible with the used thermocouple cables [4,24].
Fig. II.28: Compensating cables and connector block
49
II. Sensors
Despite all the problems mentioned earlier, thermocouples are widely used because of
their repeatability and because they can be made into very compact and lightweight
sensors [4].
Some examples of thermocouples can found in the figures below.
Fig. II.29: Thermocouples : (a) K Type by Magneti Marelli [19]; (b) J Type and (c) K Type by
McLaren Electronics [17]
3.4
Infrared Sensors
Thermocouples, as seen in the section above, can measure only if there is contact with
the junction. To measure brake temperature, one should install the thermocouple to rub
against the brake disk. A more satisfactory arrangement is to use a non contact
temperature sensor which measures the infrared (IR) emissions. As the temperature of
the brake disk rises, the level of IR emissions increases. This can be directed onto an IR
sensitive photodiode, and hence converted into an electrical voltage. Such systems are
geometry sensitive, but are repeatable and accurate if properly engineered [4].
Fig. II.30: (a) Infrared temperature sensor [17]; (b) Triple Infra red temperature sensor [17]
50
II. Sensors
The measurement range of such an IR sensor like illustrated in Fig. II.30a will be
between 0 and 1000C. Of course the sensor will only withstand a temperature of about
100C [17].
Fig. II.31: IR temperature measurement of brake disk
The triple IR temperature sensor can be used to measure at three different points on e.g.
a race tyre. The schematic overview on the next page demonstrates the extra value that
this sensor can have on a race car. By means of the IR sensor, the temperature can be
measured in real-time and the chance to have a measurement error via a thermocouple
during a pitstop will be smaller.
Fig. II.32: (a) Application of IR sensors on tyre temperature measurement; (b) Mounted on a
racecar [27]
An article appeared in Racecar engineering [26] demonstrated the use of a thermal

imager by Irisys. The thermal imager (IR camera) linked to a PDA or Pocket PC, could
show images of the temperature of a tyre in a really accurate way. The huge advantage
is that one can measure the temperature at distance and thus does not disturb the
mechanics around the race car. Another advantage is that there can be spied on other
51
II. Sensors
race teams, without even touching their car. They often say that an image says more
than a thousand words Nevertheless, some problems concerning the emissivity on
shiny surfaces like steel brake disks are still a problem. Again calibration is one the
most important points in this measurement. If not correctly performed, one could have
an error of 100C on a 200C measurement! Better no data, than wrong data.
(a)
(b)
Fig. II.33: (a) Irisys Thermal Imager [26] and its (b) measurement results [26]
52
II. Sensors
Strain and force sensors
4.1
Loadcells and force measurement
If a beam of known stiffness is manufactured and fitted with strain gauges, it can be
used to monitor loads directly. Such a device is called a load cell and can be used for a
variety of tasks. A very important example of the use of load cells is the measurement
of corner weights as part of the suspension set-up procedure. Load cells are also used to
measure engine output power and for many other tasks in system monitoring [4].
4.2
Accelerometers
The basic principle of an accelerometer is the measurement of a force as a result of the

acceleration of a mass:
F = m .a
F = Force [N]
m = mass [kg]
a
4.2.1
= acceleration [m/s]
Based on load cells
If a weight is attached to a load cell beam, acceleration will cause the beam to bend.
Thus, this combination can be utilised as an accelerometer, a device having many uses
in performance monitoring. If the accelerometer is sensitive to acceleration in one
direction, it is termed a one axis device. In Motorsport, the use of one axis
accelerometers is rare and two and three axis device are used to be able to measure the
direction and the magnitude of the acceleration/deceleration that the race car is
experiencing [4,7].
4.2.2
Based on Hall Effect
The accelerometers based on the Hall Effect are manufactured on the principle of a
sprung magnetic mass. These sensors are most often used in ABS systems due to their
simplicity. One can see on the figure below that the acceleration of the car, thus the
53
II. Sensors
magnetic mass, will move the mass from left to right. This will result in a changed
magnetic field and will induce a Hall voltage.
Fig. II.34: Hall Effect Accelerometer [28]
4.2.3
Based on Piezoelectricity
Accelerometers are also constructed using piezoelectric load sensors. These tend to be
more compact, but they are also more expensive [4,28].
Fig. II.35: Accelerometer based on the piezoelectric principles [28]
4.2.4
Based on silicon technology
In the last few years, a new type of accelerometer has been produced which makes use
of similar technology to that used to produce electronic integrated circuits (IC). The
weight and support springs are machined out of a piece of silicon a few millimetres
square. If the weight moves due to acceleration, the gap between the weight and its
support will change. This will be related in a change in the capacitance between
electrodes on the weight and the supports [4,28].
54
II. Sensors
Fig. II.36: Micromechanical accelerometer [28]
Many of these units incorporate signal conditioning (more details in [4]) on the same
chip, to provide a voltage output proportional to the acceleration they are experiencing.
Multi-axis accelerometers are also available in that type of technology. Similar in size
and shape than a small IC chip, they are smaller, lighter and less expensive than the
conventional accelerometers described in the section above. Additionally, they represent
the first examples of a new range of smart sensors discussed in [31] and [28].
(a)
(b)
Fig. II.37: (a) 3- and 2-axis accelerometers [14]; (b) 3-axis accelerometer [17]
55
II. Sensors
Fig. II.38: Knock sensor based on the accelerometer principles [28]
4.3
Yaw Sensor
Yaw is a specific movement around the Z axis or vertical axis of the car. Of course
other yaw rates can be measured around the X and Y axis but are not denominated as
yaw but respectively as roll and pitch. The yaw rate is often used by the ESP
system (electronic stability program anti skid) and will also be used by the data
acquisition software to draw the circuit.
Fig. II.39: Yaw [28] and gyroscope [28]
uur
v y = v y .sin t
uur
a c = 2. v y . x . z
4.3.1
Piezoelectric principle
As told before, a pressure difference on a piezo crystal will induce a voltage if the
crystal is not shortcut. One should also know that a voltage applied on the crystal can
make that crystal vibrate.
56
II. Sensors
The sensor on the figure below is based on the piezoelectric principle where number 6
and 5 are both piezo crystals. The crystals at the bottom of the sensor (number 6) are
used to vibrate, while the crystals at the top measure the vibration induced by a rotation
(yaw, not due to number 6). Consequently, if the racecar is driving on a straight line, no
voltages will be measured at 5. Subsequently, if a rotation or yaw around the vertical
axis is occurring, the sensor will sense it as shown in Fig. II.40b.
(a)
(b)
Fig. II.40: Yaw sensor in fork-type based on piezoelectricity [28]; (a) straight line [28]; (b) Yaw [28]
Fig. II.41: Drum-type piezoelectric sensor [28]
A micromechanical type based on changes in capacity (electrical) exists as well. The

plates shown on the figure below are placed in vacuum to diminish the resistance
against movement. If Coriolis Forces are experienced by the plate, its position will
modify compared to the bottom. This change in height will imply a modification of the
capacity. One can notice on the figure b below that the scale at which this type of sensor
is manufactured is in microns and not millimetres
57
II. Sensors
Fig. II.42: (a) Micromechanical sensor [28]; (b) Under the microscope
Some examples of real life yaw sensors are shown below. As can be seen multi-axis
yaw sensors are available.
Fig. II.43: Yaw sensor by Bosch [14]
Fig. II.44: 3-axis yaw sensors by McLaren Electronics [17]
58
II. Sensors
5
5.1
Speed sensors
Timing
The simplest way to measure an average speed is to note the time taken by the race car
to travel between two points, where the so-called lap timing is the most popular version.
It has long been the practice for racing teams to time cars between other selected points,
called the intermediates or sector, to monitor performance. Many circuits enable this
process to be undertaken more easily by installing beacon systems at various points
around the track. These can be operated using either optical or microwave sensors.
Optical sensors are cheaper, but the onboard sensor must be visible from the side of the
track, and they are vulnerable to dirt on their lenses. Another drawback is that the
weather conditions (heavy sun in south of France for example) influence on the
efficiency of the transmission between beacon and receiver (in the car).
Fig. II.45: (a) Optical (Infra Red) Lap trigger [33] and (b) Optical Lap marker receiver [33] of the
Ferrari 430 Challenge racecar
Microwave units can be installed internally, and can achieve higher timing accuracies
because the timing beam can be made narrower. All beacons transmit a coded signal so
that the specific beacon can be identified. During the Ferrari Challenge for example, it
will always be the champion of the season before that will be responsible of installing
the unique optical beacon during each race weekend. The race organisation is
responsible for the microwave type beacon which is needed for the official timing, the
so-called transponder systems.
59
II. Sensors
(a)
(b)
(c)
Fig. II.46: (a) Transponder, (b) Holder and (c) entire transponder data system with decoder [34]
5.2
Wheel speed sensing
Speed may be calculated by integrating the output of the accelerometers over a period of
time, but this technique is prone to electronic drift, which can give rise to inaccurate
results. A more common technique is to measure road wheel rotational speed by the use
of a toothed wheel and a proximity sensor. One must know that this method is prone to
errors due to tyre wear and growth and in case of wheelspin or brake locking (as will be
demonstrated in the data analysis chapter later on). The first two are generally ignored,
and the latter are diminished by the use of more than one wheelsensor. A Ferrari
Challenge uses four wheelsensors and the software calculates an average.
Fig. II.47: Principle of the inductive proximity sensor (wheelspeed measurement) [28]
As discussed in the section handling about the proximity sensors, such a sensor is
placed to sense the presence or not of a ferrous timing disk attached to the wheel. The
frequency of the pulsed signal is related to the number of slots in the disk, the diameter
of the wheel and the wheel rotational speed.
60
II. Sensors
Other systems are also used where the bearing incorporates a magnetised wheel as
illustrated in the figure below. A Hall Effect Sensor senses the alternation of the North
and South magnetic pole of the wheel which results in a pulsed signal with a frequency.
Fig. II.48: Magnetised wheel incorporated into bearing [28]
5.3
GPS speed sensing
A more exotic speed measurement technique has been developed the last few years with
the availability of compact and accurate global positioning system (GPS) receivers.
Some of these plot the position of the vehicle several times a second, and calculate
speed from the resultant position changes. Portable GPS apparatuses like the well
known Tom-Tom are based on that principle. More recently, Racelogic [35] developed
several systems that measure speed directly by monitoring the change in frequency of
the transmission from the GPS via the Doppler Effect. This effect will result in a higher
frequency in front of the car and lower one at the back due to the movement of the
emitting object. The latter system is more expensive but errors due to other factors are
erased and may be useful in competitions such as rallying [4].
(a)
(b)
Fig. II.49: (a) GPS Systems by Racelogic [35] and (b) receiver/calculator [35]
61
II. Sensors
6
6.1
Gas flow sensors

Sprung flap
The most common way to measure gas flow on older cars was to use a simple moving
flap. The airflow was directed through a channel, which was closed off by a lightly
sprung flap, which was pushed aside by the air. The position of the flap was sensed by a
rotary position sensor to provide an indication of the airflow rate (e.g. m/s). While
simple and cheap, this system suffers from several disadvantages. It slightly restricts the
airflow but it is also sensitive to other parameters such as acceleration and braking
effects. Therefore, this kind of sensor is not often used assuming that the gas flow is
dependent on throttle position and / or engine speed (carburettor principle) [4,28].
Fig. II.50: Sprung flap air flow rate sensor [28]
6.2
Calorimetric techniques
With the development of computer based engine management systems, more
sophisticated gas flow measurements techniques were required in order to match engine
air throughput with the amount of fuel injected. Most of these use calorimetric
techniques. A silicon chip is exposed to the airflow which cools it. Its temperature is
raised to match that of the control chip, which is not exposed to the airflow, by an
electric current. Consequently, the magnitude of that current is indicative of the airflow
rate. Although simple in concept, the shaping of the gas flow passages within the sensor
is critical to its accuracy. The devices currently in use are the result of a huge amount of
research and development [4,28].
62
II. Sensors
Earlier versions used a wire sensing element which was less robust and reliable.
Nevertheless, they were based on the same principle of control wire and a cooled wire.
Fig. II.51: Electrical diagram of the sensing wire principle [28]
Pressure Sensors
Pressure sensors operate by measuring the deflection of a diaphragm, which is exposed

to pressure on one side. The sensing element may be made of strain gauges or
piezoelectric sensors, connected in a bridge configuration in similar manner to a load
cell. By appropriate selection of the materials, the body and the diaphragm will be able
to work within a full scale of a few Pascals and GigaPascals. In addition, the materials
used will also rule the fact if they can withstand any kind of medium, temperature
ranges and the size of the pressure sensors [4,19,28].
(a)
(b)
(c)
Fig. II.52: (a) Membrane for low pressures [28]; (b) Diaphragm on chip with reference vacuum [28];
(c) Metal Diaphragm for high pressures [28]
63
II. Sensors
These kinds of sensors usually use a DC excitation voltage and generate outputs of
about 100 mV at their maximum rated pressure. Sensor to plot pressures in combustion
chambers can be found having response times of a few microseconds.
(a)
(b)
(c)
Fig. II.53: Pitot Pressure sensor [17]; (b) Hydraulic Pressure Sensor [19]; (c) Miniature pressure
sensor [17]
Three different models are available to measure different types of pressures: Absolute
pressure, relative to atmosphere or gauge pressure, and differential pressure. A
schematic overview is given in the figure below. There should be seen that different
pressure differentials are working on the diaphragm.
Fig. II.54: Different types of pressure measurements
Voltage and Current Sensors
Voltages are really simple to measure within electrical systems when using a pair of
resistors to reduce the voltage within an appropriate range to the data acquisition system.
If high voltages have to be downsized or high spikes occur, it can be dangerous to the
data collection system. Therefore, one should prefer to isolate the voltage from the
system. This can be achieved by the use of an isolation amplifier (more details in [4]).
Alternatively, a Hall Effect sensor can be used, by converting the voltage into an
64
II. Sensors
equivalent current by the use of a know resistor. The output of the Hall Effect sensor
will be proportional to the magnetic field surrounding the device. If the current carrying
cable is surrounded with a ferrous guide to direct the magnetic field produced by the
current to the device, it becomes possible to measure the current without direct contact
with the cable itself. Consequently, complete galvanic isolation with no additional
voltage drop in the conductor is achieved [4].
Lambda sensors
The lambda sensors used in Motorsport are based on standard road vehicle designs, but
are ruggedised to withstand higher temperatures and vibrations.
Fig. II.55: Lambda Sensor [19] and its voltage output vs a

Air to Fuel ratio (A/F)
Engines burn organic matter, i.e. petrol, and because of this emit gases which harm the
environment. In order to alleviate this unpleasant effect the catalytic converter is
employed, which converts the emissions into gas less harmful to the environment.
In order to enable this, a Lambda sensor is installed between the engine and the catalytic
converter which measures the residual oxygen in the exhaust gas at the electrodes and
passes the information to the engine control unit. This processes the signal and adjusts
the optimum fuel-air mixture of the mixture controller.
Catalytic converters in Motorsport start to be required in many competition classes.
65
II. Sensors
Fig. II.56: Schematic overview of implementation of the oxygen sensor [36]
The sensor below is an oxygen sensor made of zircon dioxide (ZrO2), working
according to the principle of W. Nernst (a German scientist, 1864-1941). He discovered
that the normal potential of an electrode can be described in dependence of the acting
oxygen concentration.
Fig. II.57: Lambda sensor schematically described [28]
The sensor is so positioned in the exhaust gas stream that the outer electrode (outer
platinum layer) is immersed in the exhaust gas, while the inner electrode is in contact
with the ambient air.
Because the zircon dioxide ceramics material becomes conductive from about 300C for
oxygen ions, an electric voltage develops with differing oxygen content between outer
and inner electrode of maximum 1 V, which is picked off via the cable terminals.
66
II. Sensors
Fig. II.58: Conversion rate of the catalytic converter [36]
The oxygen concentration developing at the outer electrode depends to a high degree on
the fuel-air ratio and the chemical reaction at the electrode surface and, thereby,
produces in the Lambda () range the leap of the sensor voltage curve [36].
67
II. Sensors
Literature References
[1] Rudolf F. Graf, "Volt", Dictionary of Electronics; Radio Shack, 1974-75. Fort
Worth, Texas, ISBN B000AMFOZY
[2] SI Brochure, Table 3 (Section 2.2.2). BIPM (2006), July 2007.
[3] Handbook of Chemistry and Physics, Fortieth Edition, p.3112, 1958
[4] John H. Nixon, Motorsport Electronics & Data Acquisition, Cranfield
University, UK, 2006.
[5] Penny + Giles Controls Ltd, Hybrid Track Potentiometers, Technical Notes, UK,
August 2004
[6] Penny + Giles, www.pennyandgiles.com, July 2007
[7] Simon McBeath, Competition Data Logging, Haynes Publishing, UK, 2002,
ISBN 1 85960 653 9
[8] Penny + Giles Controls Ltd, LVDT Displacement Transducers, Technical Notes,
UK, October 2004
[9] RDPE, How it works, www.rdpe.com, July 2007
[10] Wikipedia pictures, en.wikipedia.com, July 2007
[11] Sensorland, www.sensorland.com, July 2007
[12] Penny + Giles, Motorsport measurement and control sensors, UK, 2006
[13] Autosport, www.autosport.com/gallery, July 2007
[14] Bosch Motorsport GmbH, www.bosch-motorsport.de, July 2007
[15] Elag, www.elag.com, July 2007
[16] PI Research, www.piresearch.co.uk, July 2007
[17] McLaren Electronics, www.mclarenelectronics.com, July 2007
[18] Gill Sensors, www.gillsensors.co.uk, July 2007
[19] Magneti Marelli holding S.p.A, Motorsport Sensors, 2006
[20] XAP Electronique, www.xap.fr, July 2007
[21] AIM Sports, www.aimsports.com, July 2007
[22] MoTeC, www.motec.com.au, July 2007
[23] Fluke, www.fluke.nl, July 2007
[24] Delta Controls Corporation, www.deltacnt.com, July 2007
[25] EAS Electronics Ltd, www.embeddedtronics.com, July 2007
68
II. Sensors
[26] The international journal, Racecar Engineering, Taking the heat, November
2005
[27] Solidstateracing, www.solidstateracing.com, July 2007
[28] Dr. Ir. P. Van De Bossche, Voertuigelektronica: Algemene Begrippen,
Erasmushogeschool Brussel, Belgium, 2005
[29] Omega, www.omega.com, July 2007
[30] Vishay, www.vishay.com, July 2007
[31] James Njuguna, Smartsensors in the Motorsport, Cranfield University, UK,
2006
[32] Hall, Edwin, On a New Action of the Magnet on Electric Currents, American
Journal of Mathematics, Volume 2, 1879
[33] Magneti Marelli S.p.A, Data acquisition, Display and Lap Trigger, 2007
[34] AMB Transponders, www.amb-it.com, July 2007
[35] Racelogic Ltd., www.racelogic.co.uk, July 2007
[36] NGK Spark Plugs, www.ngk.de, July 2007
[37] Zorpette, Glen, Super Charged: A Tiny South Korean Company is Out to Make
Capacitors Powerful enough to Propel the Next Generation of Hybrid-Electric
Cars, IEEE Spectrum 42 (1), ISSN 0018-9235, 2005
[38] Paul, Clayton R, Fundamentals of Electric Circuit Analysis, John Wiley & Sons,
ISBN
0-471-37195-5,
2001
69
III. Basic Race Engineering
Chapter III
1
Basic Race Engineering
Aerodynamics
Aerodynamics is surely one of the greatest points of attention to obtain a performing

racecar. If nowadays, it seems logical to see F1 cars with highly sophisticated
aerodynamic packages, according to [1], it has been an important consideration since
the early stages of Motorsport development. Indeed, back in the twenties, high
performance engine and tire developments were quite expensive and only affordable for
the official constructor teams. Consequently, at that time, race car aerodynamic package
improvements were felt to be a good and cheap way to improve race car performances.
Aerodynamics is basically the science of fluids flowing around a body. Furthermore,
when a race car races on a track, it interacts with the air surrounding it [4]. In order to
go around the car, the air has to accelerate or decelerate considering the mass
conservation. This change of air speed has the direct effect of creating forces on the car
which can easily being understood via the simplified Bernoullis equation. The equation
is only correct for a steady, inviscid, incompressible fluid neglecting the effect of
change in pressure with the change of height of the fluid [5]:
Pstatic +
1
V = Ptotal
2
Pstatic = Static Pressure [N/m], [Pa]

Ptotal
= Total Pressure [N/m], [Pa]
V = Freestream velocity of air [m/s]
= Air density [kg/m]
Considering a limited volume of air where the overall temperature and pressure
conditions remain constant and homogenous, the direct consequence of this equation is
that when a fluid accelerates its static pressure decreases and when it decelerates its
static pressure increases.
This will result in a gradient of pressure on a race car which is the origin of the drag
(longitudinal direction) and lift (vertical direction) forces. These forces have
70
respectively the effect of limiting the penetration of the car through the air and lifting it
(or pushing it downwards). These forces are schematically represented on a thin flat
plate profile. Furthermore, a typical static pressure variation around a road car is
provided on the figure below.
Fig. III.1: Lift, Drag and total force on a thin plate [5]
Fig. III.2: Distribution of measured pressure coefficients around a 2D generic car shape [5]
The lift force L is the direct consequence of the difference between the resultant forces,
generated in the vertical direction of the car, on the upper and lower surfaces of the car.
On the other hand, drag force D is generated in a slightly more complicated way. The
main sources of drag are:
9 Skin friction of the air on the car
9 Flow separation (wakes, separation bubble)
9 Tires aero-resistance
71
9 Lift induced drag (due to the wing interaction, thus one of the greatest)
9 Shape of the lower surface (under the car)
1.1
CL and CD coefficients
These coefficients are the dimensionless reductions of the lift and drag (longitudinal
drag) forces explained above and are speed independent. The expressions of these
coefficients are:
CL =
L
1
V A
2
CD =
D
1
V A
2
L = Lift Force (downwards preferably) [N]

D = Drag Force [N]
V = Air Freestream velocity [m/s]
= Air density [kg/m]
A = Reference Area [m] (projected frontal area)

For a good understanding of how the aerodynamic lift force is generated along a race
car, the CL is quite often separated into two different coefficients at the front and rear of
the car: CLf and CLr. Consequently, to achieve a good and coherent distribution of the
downforce along the car, these coefficients can be used. Normally, for good handling,
the aerodynamic package, such as front and rear wings, is designed to achieve a
front/rear lift distribution as close as possible of the inverse of the front/rear weight
distribution of the car [1,2]
One should remember that there is also a lateral aerodynamic force working on the car
and only measured for purposes like NASCAR or F1.
The following figures are results obtained in the wind tunnel at Cranfield University by
Massimo Bottiglieris group during the Aerodynamics Module in 2007. These graphs
are clear enough and are there to make the reader aware of the influences of aero on a
generic race car.
72
Lift Vs Pitch angle at different ride heights

-2.9
0
10
12
14
-3
-3.1
10
-3.2
CL
13
16
-3.3
19
-3.4
-3.5
-3.6
pitch angle [mm]
Fig. III.3: CL vs pitch angle at different ride heights
Drag Vs Pitch angle at different ride heights

0.83
0.82
Cd total
0.81
10
0.8
13
16
0.79
19
0.78
0.77
0.76
0
10
12
14
pitch angle [mm]
Fig. III.4: CD vs pitch angle at different ride heights
L/D Vs Pitch angle at different ride heights

4.4
4.3
10
4.2
L/D
13
16
4.1
19
4
3.9
3.8
0
10
12
14
pitch angle [mm]
Fig. III.5: L/D vs pitch angle at different ride heights
73
Some ideal heights of a car as told by Claude Rouelle [6] are given below. These ride
heights can be found again when racing around the track within any event happening.
Fig. III.6: Ideal Ride Heights [6]
The pitch speed and ride height variation will be conditioned by the change of front and
rear aerodynamic force and the way dampers are set to resist the force of these changes.
Fig. III.7: Changes in ride height while turning and braking [6]
74
2
2.1
Tire model
Forces acting on tire
The tire is one of the most important parameters to obtain high performances since it is
the only component which interfaces between the car and the road. It helps, by normal
deformation, to keep a permanent contact with the road, and it are the tires which
generate the necessary forces to accelerate, brake and corner. These forces are the
results of a normal force which generates lateral and longitudinal forces. As can be seen
on the figure below, a huge amount of forces are acting on the tires.
Fig. III.8: Forces acting on tires [6]
75
Fig. III.9: Bridgestone tire on Ferrari F1 car [11]
2.2
2.2.1
Toe, camber and caster

Toe
When a pair of wheels is set so that their leading edges are pointed slightly towards
each other, the wheel pair is said to have toe-in. If the leading edges point away from
each other, the pair is said to have toe-out. The amount of toe can be expressed in
degrees as the angle to which the wheels are out of parallel, or more commonly, as the
difference between the track widths as measured at the leading and trailing edges of the
tires or wheels.
(a)
(b)
Fig. III.10: (a) Toe-in and (b) toe-out [12]
Toe settings affect three major areas of performance: tire wear, straight-line stability
and corner entry handling characteristics.
For minimum tire wear and power loss, the wheels on a given axle of a car should point
directly ahead when the car is running in a straight line. Excessive toe-in or toe-out
causes the tires to scrub, since they are always turned relative to the direction of travel.
76
With the steering wheel centred, toe-in causes the wheels to tend to roll along paths that
intersect each other. Under this condition, the wheels are at odds with each other, and no
turn results. When the wheel on one side of the car encounters a disturbance, that wheel
is pulled rearward about its steering axis. This action also pulls the other wheel in the
same steering direction. If it is a minor disturbance, the disturbed wheel will steer only a
small amount, perhaps so that it is rolling straight ahead instead of toed-in slightly. But
note that with this slight steering input, the rolling paths of the wheels still do not
describe a turn. The wheels have absorbed the irregularity without significantly
changing the direction of the vehicle. In this way, toe-in enhances straight-line stability.
If the car is set up with toe-out, however, the front wheels are aligned so that slight
disturbances cause the wheel pair to assume rolling directions that do describe a turn.
Any minute steering angle beyond the perfectly centred position will cause the inner
wheel to steer in a tighter turn radius than the outer wheel and pull literally the car into
the turn rather than maintaining a straight line of travel. So it is clear that toe-out
encourages the initiation of a turn, while toe-in discourages it.
With four-wheel independent suspension, the toe must also be set at the rear of the car.
Toe settings at the rear have essentially the same effect on wear, directional stability and
turn-in as they do on the front. However, it is rare to set up a rear-drive race car toed out
in the rear, since doing so causes excessive oversteer, particularly when power is
applied. Front-wheel-drive race cars, on the other hand, are often set up with a bit of
toe-out, as this induces a bit of oversteer to counteract the greater tendency of frontwheel-drive cars to understeer.
Remember also that when driving torque is applied to the wheels, they pull themselves
forward and try to create toe-in.
The amount of toe-in or toe-out dialed into a given car is dependent on the compliance
of the suspension and the desired handling characteristics.
2.2.2
Camber
Camber is the angle of the wheel relative to vertical, as viewed from the front or the rear
of the car. If the wheel leans in towards the chassis, it has negative camber, if it leans
77
away from the car, it has positive camber. The cornering force (see later) that a tire can
develop is highly dependent on its angle relative to the road surface, and so wheel
camber has a major effect on the road holding of a car. It is interesting to note that a tire
develops its maximum cornering force at a small negative camber angle, typically
around -0.5. This fact is due to the contribution of camber thrust, which is an additional
lateral force generated by elastic deformation as the tread rubber pulls through the
contact patch [4,6].
(a)
(b)
Fig. III.11: Definition of negative and positive camber; (b) Example of negative camber [10]
Since the wheel is connected to the chassis by several links which must rotate to allow
for the wheel deflection, the wheel can be subject to large camber changes as the
suspension moves up and down. For this reason, the more the wheel must deflect from
its static position, the more difficult it is to maintain an ideal camber angle.
To maintain the ideal camber relative to the road, the suspension must be designed so
that wheel camber relative to the chassis becomes increasingly negative as the
suspension deflects upward. If the suspensions were designed so as to maintain no
camber change relative to the chassis, then body roll would induce positive camber of
the wheel relative to the road.
The best way to determine the proper camber for competition cars is to measure the
temperature profile across the tire tread immediately after completing some hot laps or
in real time via IR sensor as shown in the sensor chapter. Often the inboard edge of the
tire is slightly hotter than the outboard edge. Nevertheless, it is more important to
ensure that the tire is up to its proper operating temperature.
78
Fig. III.12: Camber on a race car [12]
2.2.3
Caster
Caster is the angle to which the steering pivot axis is tilted forward or rearward from
vertical, as viewed from the side. If the pivot axis is tilted backward (that is, the top
pivot is positioned farther rearward than the bottom pivot), then the caster is positive.
On the other hand if it is tilted forward the caster will be negative.
Fig. III.13: Caster [12] and [11]
Positive caster tends to straighten the wheel when the vehicle is travelling forward, and
thus is used to enhance straight-line stability.
The tilted steering axis has an important effect on suspension geometry. Since the
wheels rotate about a tilted axis, they have an effect causing the outside wheel in a turn
to gain negative camber, while the inside wheel gains positive camber. These camber
changes are generally favourable for cornering, although it is possible to overdo it.
Most cars are not particularly sensitive to caster settings. Nevertheless, it is important to
ensure that the caster is the same on both sides of the car to avoid the tendency to pull to
one side. While greater caster angles serve to improve straight-line stability, they also
cause an increase in steering effort. Three to five degrees of positive caster is the typical
range of settings [4,12].
79
2.3
2.3.1
Friction Coefficient
Basics
The friction coefficient of a tire will dictate the grip that will be available when related
to the normal force applied on it.
Fx = x . Fz
Fy = y . Fz
Fx, = Maximum longitudinal force available [N]
Fy = Maximum transversal force available [N]
x = Longitudinal friction coefficient [-]
y = Lateral friction coefficient [-]
Fz = Normal force acting on tire [N]
On the figure below one can easily recognize the parameters of the formula above.
Theory tells that the average of the four tires coefficient of friction equals the cars
lateral acceleration in g. Thus, if a 10 kN car is cornering at 2g as measured by the cars
lateral accelerometer:
average =
Fy
Fz
20 kN
=2
10 kN
80
(a)
(b)
Fig. III.14: Vertical load acting and resulting in maximum lateral and longitudinal forces [6].
2.3.2
Circle of friction
Working within the last 1% of traction means there is no reserve to call upon to gain
control of the car back. Even the pros very rarely recover a car that has lost control. It is
not because they do not know how, it is because there is no traction left to work with.
To see if the driver is performing well, an X-Y graph plotting the longitudinal and
lateral gs can be used. These figures have then to be compared with the real Circle of
traction of the tires [7]. Note that these are not often provided by the manufacturer. The
circle of traction is a circle centred on an X-Y axis graph. The radius of the circle is 1
unit. 1 unit is equal to the adhesion limit of a tire. To see how close one is to the limit,
the basic formula of Pythagoras can be used (R2 = x2 + y2).
81
(a)
(b)
Fig. III.15: (a) Ideal Circle of traction and (b) real circle of traction
As told by Claude Rouelle [6], a well established race engineer and founder of
Optimum G, the tire does not have a circle of friction but more an elliptical shape of
friction. Consequently, the grip available while combined turning and braking (or
turning and accelerating) is smaller than only turning, braking or accelerating. The
figure above (b), also demonstrates another real circle of traction, where a higher grip in
combined cornering and braking/accelerating is available compared to the elliptical one.
Nevertheless, a smaller amount of longitudinal force compared to the elliptical one is
available. This will result in different speeds on track.
2.3.3
Slip Ratio
Slip Ratio is defined as the difference between the forward vehicle speed and the speed
of the wheel on the track surface defined as .R [6]. Several other definitions are
provided in [3,4]. A good example of slip ratio is given in the figure below. There can
be seen that the higher the normal force, the higher the tractive force of tire is. The
higher Fz the higher the slip ratio will be at maximum tractive force. One should
remember that longitudinal friction coefficient is function of a huge amount of
parameters like e.g. the slip ratio, rubber of the tire, temperature, tire pressure or ground.
82
Fig. III.16: Slip ratio example [6]
(a)
(b)
Fig. III.17: (a) Longitudinal friction coefficient variation (b) plotted for several normal forces [6].
Furthermore, a very useful graph to plot is the contact patch area versus the vertical load
on the tire. There can be seen that the area does not grow entirely linearly with an
increasing load [6]. Thus with weight transfer, the contact patch area lost at the inside is
greater than the gain at the outside. Consequently, the total grip is smaller.
83
Fig. III.18: Contact patch area vs vertical load [6]
2.4
Slip Angle
During acceleration and braking phases, the rolling tire travels straight ahead and it is
only the longitudinal contraction/elongation of the tire thread which produces the
longitudinal force Fx.
When cornering, the situation is a bit different. Indeed, at the tire contact patch, the tire
slips out of the velocity vector plane with a certain angle, the so-called slip angle .
This slip angle generates a distortion of the rubber creating a lateral force Fy as function
of the normal force Fz applied on the tire. As the lateral force is not located at the centre
of the tire contact patch, it provides the effect of creating an aligning torque Mz around
this centre, realigning the tire into the wheel plane.
Fig. III.19: Slip Angle definition with aligning torque and lateral force [6]
84
As slip angle increases, events in the contact patch area are:

1. Lateral tire deformation
2. Torsion around a vertical axis
3. modification of the contact patch area
4. modification of the contact patch lateral forces
A typical relationship between lateral force and tire slip angle for a racing tire in
function of the normal force is shown below. Furthermore, some fundamental tire
parameters can be concluded:
9 At low slip angles ( 4< slip angle) the relationship between the slip
angle and the tire generated lateral force is linear. The linear relation
between these two parameters is called the cornering stiffness.
9 When = max lateral force, the maximum tire grip is reached.
9 When > max lateral force, Fy decreases when increases and the
associated level of grip as well to reach a constant value.
85
There should be notices that the shape of the curve of Fy versus is dependant of [3,6]:
9 Rim
9 Tire pressure
9 Camber
9 Toe initial slip angle
9 Vertical load
9 Speed
9 Tire construction and
compound
9 Ground surface structure
9 Ambient and track
temperature
9 Tire temperature (surface and gas)
9 Slip Ratio
Fig. III.20: Maximum lateral force versus slip angle in function of vertical load
2.5
Neutral Steering, Understeer and Oversteer
Neutral steer is when the car follows the steering input properly, by following the
desired line without any correction. In real world terms, understeer is when the car will
not turn any sharper, even if when turning the steering wheel more. At some point, the
front end may start to grip less even when the steering is turned sharply and the result is
the car continues in more of a straight line. Oversteer is when the car's rear tires lose
grip in a turn while the front tires are still gripping. If the rear end starts to slide out in a
turn, that is oversteer. These behaviours can be seen in the two figures below.
86
Fig. III.21: Oversteer and understeer
A graphical overview of the technical aspects of the complete definition of a neutral,

understeering and oversteering car are provided by [6] and explained in [3,6] and are
shown below.
Fig. III.22: Neutral, Oversteer and understeer [11]
87
Fig. III.23: The lateral force sequence in a corner [6]
Fig. III.24: Neutral steer [6]
88
Fig. III.25: Understeer and oversteer [6]
Fig. III.26: Oversteering old formula 1 car [11]
89
Damping technology
Dampers are another key factor for high performance race car since competition cars
have to race on different surfaces and tracks. Consequently, the two main objectives of a
good damper are:
1. Keep the wheel on the road without transferring any vertical acceleration to the
car
2. Limit the movement of the car due to the load transfers (so-called weight
transfer) when cornering, braking and accelerating to insure a better handling for
the driver (the also called low speed displacement).
Unlike springs, dampers do not react to a displacement but to a velocity of displacement.
As [3,4,8,9] state, dampers have an important range of velocity displacement between.
Considering a really simple approach where a damper reacts linearly to a velocity of
displacement, the force generated by the damper can be expressed with:
F = C .v
F = Force generated by damper [N]
C = Linear damping force [N/(mm/s)]
v
= Velocity of damper [m/s]
In reality, the damping factor is never linear and is a function of the damper setting at
different speeds. The best way to quickly compare dampers is to plot the force versus
the velocity of the damper as demonstrated below.
90
Fig. III.27: Typical plot of damping force vs piston velocity [6].
Fig. III.28: Typical force versus displacement curve for a damper [6].
There are different sorts of dampers with different adjustment capabilities: 4 ways, 3
ways and 2 waysOf course, the 4 ways technology is the best tool so as to shape the
damping force vs displacement velocity curve since it allows the independent change of
the nose, the knee and the slope shapes of this curve both in rebound and bump. The
four settings available on the damper are given in the figure below.
91
Fig. III.29: 4 way dampers, as on the F430 GT3 [9]
Fig. III.30: Suspension movement Formula Renault 2.0 Eurocup at Magny-Cours (France)
gathered via the AIM data acquisition system (Maxime Martin as driver)
An example of the outputs for Low and High speed damping are provided for a Formula
Matrix Damper by Sachs [9].
in the test there are two kinds of shocks that the modern damper can absorb:
Table 3: Damping speeds
Low Speed
High speed
Speed
Around 25 mm/s
125 mm/s and over
Physical interpretation
Weight Transfer
Track surface bumps / curbs
Control
Transition phases
Tire contact patch surface

consistency
92
(a)
(b)
(c)
Fig. III.31: Damping force versus piston speed for (a) Low Speed damping, (b) High speed damping
and (c)combined damping [9].
93
The Sachs 4-way adjustable Formula Dampers have two separated adjusters for bump
and rebound. The damping characteristics can be adjusted individually in low and high
piston speeds [9].
The low speed adjustment provides 16 positions position 1 = soft, position 16 =
hard.
The high speed adjustment provides 12 positions position 1 = soft, position 12 =
hard.
Often the set-up of the dampers are referred in clicks and written down as:
67
low speed (Ls) high speed (Hs)
rebound
11 5
low speed (Ls) high speed (Hs)
bump
The data as shown below can be used in conjunction with the other data analysis and the
drivers comment to fine tune the damper units. More details about fine tune of
suspensions and dampers can be found in [3,6,8].
Fig. III.32: Suspension analysis tool with the AIM data analysis software
94
Conclusion
As a race engineer in the make, the writer discovered a lot during the last few months.
After long theoretical researches in several books and magazines, the courses of the
MSc Motorsport Engineering and Management at Cranfield and the huge experience
during the past month on track with all kind of different cars, he was able to set a sort of
list to counter oversteer and understeer. Of course this is just a tool and might be
extended in the future as his experience keeps on growing (every part of the car has got
a colour).
Table 4: Counter understeer and oversteer (all references and experience)
Counter Understeer
Counter oversteer
Push, front tires slide out first.
Loose, rear tires slide out first. That kind
Usually slight understeer is safer.
of oversteer can be dangerous, especially

at high speeds.
Raise front tire pressure to stiffen sidewalls
Lower front tire pressure.
Lower rear tire pressure if pressure too
Raise rear tire pressure.
high
Install smaller front tires.
Install taller front tires.
Install taller rear tires.
Install smaller rear tires.
Install wider front tires.
Install narrower front tires.
Install narrower rear tires.
Install wider rear tires.
Use softer front compound if possible.
Use harder front compound if possible.
Use harder rear compound if possible.
Use softer rear compound if possible.
More front toe out.
More front toe in.
Reduce rear toe in slightly.
Increase rear toe in.
Increase front negative camber.
Reduce front negative camber.
Increase positive caster.
Reduce positive caster.
Soften front anti roll bar (compared to
Stiffen front anti roll bar.
rear)
Stiffen rear anti roll bar.
Soften rear anti roll bar.
Soften front dampers, harder in Bump.
Stiffen front dampers.
Stiffen rear dampers.
Soften rear dampers.
Lower front end.
Raise front end.
Raise rear end.
Lower rear end.
95
Soften front springs.
Stiffen front springs.
Stiffen rear springs.
Soften rear springs.
May need more front suspension travel.
May need more rear suspension travel.
Remove weight from front of vehicle.
Add weight to front of vehicle.
Add weight to rear of vehicle.
Remove weight from rear of vehicle.
Drive a different line.
Driver may be going in too deep.
Use weight transfer to drivers advantage.
Driver on the throttle too early?
Widen front track.
Reduce rear track.
If high speed,increase front wing
If high speed, increase rear wing
downforce.
downforce.
Too much front brake.
Too much rear brake.
Widen front track.
Reduce rear track.
Twitchy vehicle at limit and hard to keep

ahead of in the steering department.
Vehicle slides and is easy to drive at limit

but
does not corner to full potential.
Lower front and rear tire pressures slightly.
Raise front and rear tire pressures slightly.
Suspension may be too stiff.
Suspension may be too soft.
Dampers may be set too firm.
Dampers may be too soft.
Tires may be old or hard.
Roll centers may be too high.
Vehicle may not have enough suspension
Lower vehicle.
travel.
Vehicle may have a toe problem front or
Tires may be too hard.
rear
Increase negative camber front and rear
Widen track front and rear.
96
[1] Ph.D. Joseph Katz, Race Car Aerodynamics: Designing for speed, Bentley
Publishers, USA
[2] Simon McBeath, Competition Car Aerodynamics: A practical handbook,
Haynes Publishing, UK, 2005
[3] William F. Milliken, Douglas L. Milliken, Race Car Vehicle Dynamics,
Milleken & Milliken, USA, 1995
University, UK, 2006
[5] Ph.D. Kevin P. Garry, Motorsport Aerodynamics, Cranfield University, UK,
2006
[6] Ir. Claude Rouelle, Seminar about: Vehicle Dynamics, Race Car engineering,
Simulation and data acquisition, Optimum G, Denver, USA, 2007
[7] Simon McBeath, Competition Data Logging, Haynes Publishing, UK, 2002
[8] Allan Staniforth, Competition Car Suspension: A practical handbook, Haynes
Publishing, UK, 2006
[9] Sachs Race Engineering, The Formula Matrix Damper, 2007
[11] Google Images, images.google.be, August 2007
[12] ing. Luc Vercammen ,Voertuigtechnologie, Erasmushogeschool Brussel,
Belgium, 2005
97
IV. Driving Techniques
Chapter IV
1
Driving Techniques
Introduction
It is very easy to get caught up in the prestige of expensive parts for a race car or
changing the set-up, but the best investment one can ever make in racing is the time
spent tuning the driving skills of the person behind the steering wheel.
The goal of this chapter is to introduce the reader to many of the basic driving
techniques used in race car driving. There are numerous details to be conscious of while
racing on a track. One should focus on one or two techniques each time going out on the
track. As each technique becomes second-nature, there can be worked on a new one.
No matter how much the driver is telling how good he is, his race engineer or coach
have to keep their drivers feet on the ground. Good drivers, like all good athletes, have
a natural skill, and yet are also smart and/or humble enough to know that there are
known techniques they must practice if they are to be proficient. Even if they have
natural talent, one should never make the mistake of thinking that all the driver needs is
a better car or set-up to improve his performance. Driving skills can always be
improved. Even the Alonso's, Hamilton's, and Schumacher's of the pro-driving world
continually analyse their driving so they can improve.
As with any skill, knowing what to do is not the same as doing it. Practice, practice,
practice. Time in the car on the track, repetitively performing these techniques is the
only thing that will make the driver good at using them. Often one thinks of himself he
is doing something right, only to recognise several months later, that he could do it even
better.
Because there are so many things to remember and practice, the race engineer, data
engineer and drivers coach have to remember the driver at each session what to do,
even if the driver is already doing it.
98
Seating Position
One of the first things to prepare before even starting the ignition is to set a proper
seating position. This is often overlooked, resulting in poorer car control and premature
fatigue of the driver. If one looks at a variety of race cars, there can be seen that a
variety of seating positions. In the open-wheel cars, the also called Formula cars, it
appears that the driver is almost laying down with arms fully outstretched (they are not).
In a full-bodied saloon-type car, there can be seen that the driver is more erected and
almost cramped against the steering wheel [1,2,8].
2.1
Open Wheeler
The body of the open-wheel car is very shallow in height, and the cockpit is very
narrow. This shape determines much of the driver's position. The driver's legs are
relatively straight out with a slight bend in the knee, and the feet just barely below the
hips as can be seen on the figure below. The pedals in many of these cars are almost
touching each other. The pedals also require little more than a flexing of the ankle to go
from 0-100% depression. The driver's arms have little room for movement, but the
steering requires extremely little turning input by the driver. In the open-wheel car,
function (driver's seating position and controls operation) follows form (the shallow and
narrow cockpit).
Fig. IV.1: Cut away view of McLaren F1 to see seating position [11]
2.2
Saloon car
In a saloon type car, many things are completely opposite. The driver sits very erect,
and is very close to the steering wheel. In fact, the driver can almost lay his whole
99
forearm on the steering wheel. That position allows the driver to utilize more of the
shoulder and back muscles [1].
Dr Brock Walker [1] says there are only three points of contact between the driver and
the car: the seat, the steering wheel and the pedals. The drivers optimum position,
contained within a proper seating system, will eventually activate the amount of control,
sensitivity, strength and general influence that the driver will possess over both the
steering wheel and pedals. A controlled smooth, efficient manoeuvring inside the
cockpit relates directly to consistent results and faster lap times. A lot of feedback like
vibrations and g-forces will be received by the driver through his seat. Consequently, if
he is more sensitive to these, he will be able to interpret better what the car is doing.
Finally, if the driver is well positioned in his car he will be able to drive better, but also
to provide the team with a better feed-back [1,5].
These three main aspects will be related to the setting of the correct seating position.
Each of these are described after the illustrations below.
(a)
(b)
(c)
Fig. IV.2: (a) Surface Body/Seat; (b) Arm Position; (c) Leg Position [8]
Firstly as shown on Fig. IV.2a, sitting in the seat itself, the driver's back should be flat
against the back of the seat with the buttocks squarely tucked into the corner created at
the intersection of the seat back and bottom. The underside of the legs should be in
contact with the seat bottom. The purpose of this position is to provide as much surface
contact between the driver's body and the seat. This has safety benefits as well as
providing the driver with the most tactile feedback as possible. There should be noticed
that the safety belts have to be tighten up, to simulate a driving situation.
100
Secondly (Fig. IV.2b) is the arm position. When the driver is tightly strapped into the
seat as described above, the arms when fully extended should allow the wrists to rest at
the top of the steering wheel. This allows the arms to be slightly bent at the elbow when
fully extended for a turn. The purpose of this position is to prevent the arms from being
overextended during turns There should be thought about the fact that the driver is
completely stuck into his seat due to the belts, so the shoulders should not need to lift
from the seat back even to do a full arm crossover. Overextending the arms will cause
them to tire quickly, and will cause the driver to lose sensitivity to the vibrations in the
steering wheel, thus time at each turn [1].
Third is the leg position (Fig. IV.2c). When any of the pedals are fully depressed with
the ball of the foot on the pedal (not the toes), the leg should still be bent at the knee.
This is to prevent overextension as described for the arms. Additionally, one needs to be
sure that the knees are not against the underdash or steering column. In fact, there
should be several inches room to prevent injury in event of a collision. The right leg in
particular will need enough knee room to allow the ball of the foot to be on the brake
pedal, and the heel to be on the gas pedal for heel-toe downshifting [1,2,8].
Most people say that placing the driver into a good and comfortable position will
improve their lap times. Actually, it is the cheapest version of earning time and
improving the race car thanks to the better feedback of the driver [1,2,8].
Pedals
How many pedals are there? In a manual transmission car there should be four pedals:
Accelerator (gas)
Brake
Clutch
Rest (or "dead" pedal)
101
(a)
(b)
(c)
Fig. IV.3: (a) Normal set ; (b) Pedal set of Supercar Ferrari Enzo (F1 system, thus no clutch pedal);
(c) Pedals of seven time F1 champion M. Schumacher [11]
The rest pedal is a permanently positioned pedal pad at the far left to rest the left foot on.
It provides a place to stabilize the leg when not using the clutch. It is not a controlling
pedal, but is a full time tool to the driver.
In cars Ferraris equipped with a F1 gearshift system, there is no clutch pedal. Therefore,
the driver can use his left foot to brake. This technique is not the easiest one and is
prone to errors.
Every pedal must be used smoothly. Stabbing at the accelerator or brake pedal in
particular will cause sudden shifts in weight distribution on the car's suspension and will
unsettle the car. Inducing a weight transfer at the wrong moment will often result in a
loss of control. When applying either the brake or gas pedals, the driver must ease into
it allowing a smooth transition to the full pressure needed. "Easing" does not necessarily
imply being overly slow about it. A rapid but smooth and controlled transition is what is
102
wanted. One should avoid sudden jerky movements in either braking or accelerating
[2,6].
One of the pedal control techniques used in racing is the heel-toe downshift [2]. To
maximize the speed and smoothness through a corner, it becomes necessary to do some
cockpit acrobatics and operate the steering wheel, shifter, clutch, brake, and accelerator
all at the same time. The problem is that there are five functions to perform and only
four limbs to deal it with. Something has to do double duty, and it turns out to be the
right foot. To be short: prior to enter the corner, the right foot will need to operate the
brake pedal and the throttle at the same time between a downshift. Obviously, this
problem is not present if the driver is competing in F1 gearbox typed car.
One may need to use shims to raise either the brake pedal or gas pedal to allow
comfortable reach when operating both pedals. The brake and gas pedals should be
close to the same height at the time there is a need to operate both simultaneously.
When the brake pedal is fully depressed, its height should be roughly equal to the height
of the gas pedal when it is not pressed. This allows for more comfortable and
controllable heel-toe manoeuvres [1,2,6].
Braking and Accelerating
One of the keys to good race driving is smoothness, and this most certainly applies to
the use of the brake and accelerator pedals as already told above. This section is also
important for the data analysis chapter later on in the work.
4.1
Braking
Braking and accelerating when racing on a road race course, is treated something like
driving on a wet surface, thus gently and smoothly. Braking and accelerating are used in
conjunction with the corners. The driver brakes going into them and accelerates coming
out of them. Because the objective is to have the car moving as fast as possible through
the corner, the tires will be utilising most of the available traction (done right they
should be using 100% of the available traction). The driver must be very smooth with
103
the use of the brakes going into the corner and the accelerator coming out of the corner.
A sharp change in braking or power at these points will upset the car's traction balance
just as quickly as if it was driving on ice.
4.1.1
Braking phases
There are three phases in braking. First, braking begins with a rapid, but not instant,
application of as much braking force as possible. How rapid the brakes can be applied
will depend on the suspension in the car. The stiffer the springs and dampers, the more
rapidly maximum braking can be applied. Soft springs will have significant pitch which
will require a little longer and smoother ramp-up of braking to keep the car stable.
Secondly, once the car settles onto the front tires, one should try to minimise the length
of the braking zone, so it will require taking the tires to the edge of locking up. The
driver will need to be very aware of the vibrations in his foot from the pedal and in his
hands from the steering wheel to feel that small difference. The car will travel some
distance using a fairly constant brake pedal pressure.
The third phase is towards the end of the braking zone when the vehicle has been
slowed to near its final speed. Gradually releasing pressure off the pedal makes the
transition from full to zero braking force as smooth as possible. During braking, the
front tires are under heavy load which increases the available traction. A sudden release
of the brakes will abruptly reduce the load and reduce the traction potential of the front
tires which at this point is needed for turning into the corner.
The corner entry (turn-in of the figure below) is one of the points where the car will be
the most sensitive to sudden weight transfer transitions as though it were being driven
on ice. Indecisive braking resulting in a last second extra tap, or a sudden release of the
brake pedal will unsettle the car's handling and force the driver to slow down to gain
control and hopefully avoid a spin.
As the braking zone completes, and the driver eases off the brake pedal, he will have to
apply some throttle to reach a steady state of neither acceleration or deceleration.
Depending on the shape of the turn, the steady throttle zone will vary, but with a typical
late-apex corner, it will be from the turn-in to just before the apex.
104
Fig. IV.4: (a) Normal Braking and accelerating [7]; (b) X-Y graph of gs perceived by the car. Note
the negative gs in combined behaviour (cornering-baking)
The figure above shows graphically what the driver has to do to be sure of the cars
behaviour. Although something called trail braking is used by race drivers where
cornering is combined with braking. This behaviour can be checked via data analysis
and the X-Y graph explained above. There should be noticed that trail braking is utterly
difficult, considering the circle of friction seen the race engineering chapter. This circle
of friction is also changing constantly during a season, race weekend or even a race!
Maybe Michael Schumacher did disserve his salary
105
4.1.2
Cadence Braking
Cadence braking is a technique used to stop more quickly on a slippery surface. It

would normally be used to effect where traction is limited.
Cadence braking involves pumping the brake pedal fairly rapidly but deliberately, to
make the wheels lock and unlock. Because a locked wheel creates a smaller braking
force than one which is just on the point of locking, this technique may cause the
vehicle to stop more quickly, because the point of optimum braking is passed through
multiple times. In addition, by avoiding a total lock-up, steering control can partly be
retained. While cadence braking is effective on most surfaces, it is less effective than
keeping the wheels continually at the optimum braking point. However, this is almost
impossible to do manually. Note that the principle of cadence braking is automated in
anti-lock brakes (ABS), but done multiple times every second.
4.1.3
Threshold braking
Threshold braking or limit braking is a driving technique wherein the driver adjusts
control of the brake system in an attempt to maximize the braking force of the vehicle.
Done properly, it will delay braking when entering a corner. The optimal amount of
braking force is developed on a wheel at the point when the wheel just begins to slip
(optimal slip will be between 5-15%). As the tire begins to slide, the friction force
between the tire and driving surface transitions from the static friction range into the
kinetic friction (lower than static one) range. Threshold braking avoids sliding and tries
to keep the tire's percentage slip at the optimal value, the value that produces the
maximum frictional force.
4.1.4
Left foot braking
Left-foot braking is the technique of using the left foot to operate the brake pedal in a
race car leaving the right foot dedicated to the throttle pedal. It contrasts with the normal
practice of the left foot operating the clutch pedal, and the right foot operating the brake
and accelerator pedals.
106
Fig. IV.5: A WRC driver using left foot braking in a Peugeot WRC of 2003. Note the larger brake
pedal
At its most basic purpose, left-foot braking can be used to decrease the time spent
between the right foot moving between the brake and throttle pedals. It can also be used
to control load transfer to obtain a more oversteering car at the entry of a corner.
4.2
Accelerating
From this point to the corner's exit point, the use of the accelerator must be equally
smooth for the same reasons they were for braking. Through the turn, the car will have
settled with a certain loading on each tire. A sudden change in that with the accelerator
can also upset the available traction on one or more tires and cause a loss of control.
Controlled use of the accelerator is a matter of depressing and releasing it in smooth
motions. The figure below, demonstrates a gradual throttle application, and the braking
technique corresponds to the theory given before (especially threshold braking, while
using the left foot)
107
Fig. IV.6: Data acquisition via AIM system on the Formula Renault 2.0 driven by Maxime Martin
Whether the car is relatively easy to control or not, the introduction of 5 hp too much
and the driver could experience this like 200hp too much. Coming out of a turn, as soon
as the car begins to straighten out, the power has to be applied gradually while the race
car gets straighter. Indecisive on and off stabs will end up being slower than a smooth
increase (again unbalancing the car) [2,4,5].
108
Cornering
Every corner is made out of three parts:

1. entry
2. apex
3. exit.
The entry is where turning begins. The apex is the point where the car reaches the
furthest point on the inside of the turn, and the exit is where the car is driving straight
again.
The objective in driving through a corner, or a series of corners, is to have the fastest
possible speed at the exit of corner, or the last corner of a series. It is not necessarily to
have the fastest speed going into the corner, nor even the fastest speed in the middle of
the corner. The last corner exit before a straight is the most important segment. The
speed of the exit determines the speed during and at the end of the straight. If one can
increase the average speed of an entire straight, it will have greater impact on lap times
than a faster average over the shorter distance of the corner entry.
The path, or line the driver is going through a corner will determine the exit speed. In
general, the fastest line through a corner is the one that allows the greatest radius, or
straightest path. As a car can go faster around a large corner than it can around a tight
corner, the shortest path around a corner is rarely the fastest.
In the illustration below, the dotted line follows the path of the road. The solid line
indicates a path which maximizes the radius of the turn, or attempts to make the turn as
straight as possible [2-4,7].
109
Fig. IV.7: Geometric compared to possible Racing line [4,8]
As mentioned, the objective in any corner is often to have the highest exit speed. In
addition to increasing the corner radius, this also involves taking a line which allows the
earliest possible point of getting back into the throttle. To do this, the car must be
straightening back out on the corner exit path as early as possible.
While the geometric racing line is faster than the natural line of the road, there is still a
faster technique for most corners. The technique is called using a late apex. By
delaying the turn-in point, and beginning the turn with a slightly sharper bend, the car
can be aimed to apex later than the geometric apex point. This straightens out the
second part of the turn, allowing the driver to apply the throttle earlier. The driver will
have to slow down the car a little more at the turn-in phase, but the exit speed will be
higher. That exit speed will provide the driver with much more speed on the straight
which will result in a lower overall lap time.
The next figure shows the previously noted large radius path in the dotted line. The
solid colored line shows a path known as the late apex. The latter path moves forward
the point at which the car reaches the corner apex. The late apex straightens out the exit
path of the car, and therefore allows the driver to apply the throttle earlier. This
increases the exit speed, and in effect lengthens the straight which allows for higher
speed at the end of the straight.
110
Fig. IV.8: Geometric racing line compared to Ideal Racing Line (Late Apex) [4,8]
This approach works for corners which require hard accelerating cornering out of them,
which will be most of them. However, there are many types of corners, and
combinations of corners which require some data analysis, between driver and race/data
engineer to understand the best approach [2-4,7].
111
[1] Ross Bentley, Speed Secrets: Professional Race Driving Techniques, MBI
Publishing Company, USA, ISBN 0-7603-0518,1998
[2] Ross Bentley, Speed Secrets II: More Professional Race Driving Techniques,
MBI Publishing Company, USA, ISBN 0-7603-1510-8, 2003
[3] Ross Bentley and Ronn Langford, Speed Secrets 3, Inner Speed Secrets: Mental
Strategies to maximize your racing performances, MBI Publishing Company,
USA, ISBN 0-7603-0834-9, 2000
[5] Ross Bentley, Speed Secrets 4: Engineering the Driver, MBI Publishing
Company, USA, ISBN 0-7603-1510-8, 2003
ISBN 1 85960 653 9
[8] Illustrations, www.turnfast.com, July 2007
[11] Google Picture, images.google.be, August 2007
112
V. Data Analysis
Chapter V
1
Data Analysis
Dijon Prenois (France)
Dijon is one of the tracks where the French GT championship is raced. The FFSA
Superseries has got two different categories running at the same time: GT (GT1 and
GT2) and GT3. This track is an old Grand Prix track where one the most beautiful
racefights in the history of F1 happened in 1979 between Gilles Villeneuve and Ren
Arnoux.
Even with a history like that, the Dijon track is a very dirty track. This is not an
advantage for the GT3 cars. There should be noticed that they are passed by GT1 cars
for several times during a one hour race. Consequently, if a GT3 car receives a blue flag,
it has to leave its racing line (clean line) and literally drive on the dirt of the track. A
negative consequence of that event is that the tyres are full of used rubber and sand,
hence, have no grip at all during at least one lap.
1
Double
droite
de
Villeroy
2 : S des Sablieres
3 : Gauche de la Bretelle
4 : Parabolique
5 : Double gauche de la
Bretelle
6 : Courbe des Gorgeolles
7 : Virage de la Combe
8 : Courbe de Pouas
Length : 3801 meters
(gauche = left)
(droite = right)
113
V. Data Analysis
1.1
Ferrari F430 GT3 FFSA: J.B. Scheier vs P. Ballay
One should know that the car running during that weekend arrived at the track on
Thursday evening directly form Switzerland where it was modified. Indeed, the Ferrari
GT3 is based on a Ferrari Challenge. The GT3 kit developed by Loris Kessel Racing in
collaboration with Ferrari, has to be bought at and installed by Kessel Racing at Lugano
(CH). To gain some time when arriving at the Dijon track, there was asked to do a basic
set up of the car before leaving Switzerland. As will be shown later on, one can only
trust his own work.
(a)
(b)
Fig. V.1: (a) Picture taken at Kessel Racing (CH); (b) The GT3 on the Dijon-Prenois track
The data discussed here is the data gathered during the Free Practice session of Friday.
There can be seen that the difference between Pascal Ballay (1:32.843) and Johan Boris
Scheier alias Jobo (1:20.898) is huge! Subsequently, one can quickly calculate that the
gap between the best lap of each driver is about 12 seconds.
114
V. Data Analysis
1.1.1
Corners 1, 2 and 3
Fig. V.2: Turn 1, 2 and 3, blue for Jobo and red for Pascal
Before the entry to the first corner of the Double droite de Villeroy, there can be seen
that the negative ramp of the speed for Jobo is much higher than Ballays. This is not a
direct consequence of a lower brake pressure applied by Ballay but the fact that Pascal
releases the throttle for about 250 meters before applying pressure on the brakes. This
means that neither throttle nor brakes are applied in a straight line. Consequently, a
huge amount of speed and thus time will be lost compared to Jobo.
The speed at the entry of turn 1 is drastically different between both drivers. Jobo is
about 41 km/h faster at the moment he brakes and gives a slightly different steering
wheel angle. This difference in angle is firstly due to the fact that the slip angle of the
tyres has to be greater due to the greater lateral acceleration required by the tyres, but
also due to a light understeer. The latter cause was not felt by Pascal, but one may
accept his low speed as an explanation.
115
V. Data Analysis
Both driver apply throttle at the same place on the track for the exit of turn 1 and the
entire S des Sablier which follows, but only Jobo applies full throttle. Despite the fact
that the car is not going fast and that the car sticks to ground, Pascal only applies 20%
throttle. In addition, the gentle man driver is so afraid of the noise that he upshifts the
gears around 6700 rpm. Knowing that the car is a V8 made by Ferrari and is able to rev
at about 8750 rpm, one can see that the driver is not using the Italians entire engine
potential. When concentrating on the speed trace, there can be seen that the average
speed difference between both drivers is about 40 km/h. Deeper analysis demonstrates
that Pascal lost 5.1 seconds on the first 1570 meters of the track. This portion goes from
the startline till the end of turn 3 and the start of turn 4.
Fig. V.3: Time difference between Jobo and Pascal (middle line)
Fig. V.4: Dijon-prenois track
116
V. Data Analysis
During the second part of the S, Pascal starts to brake 12m after his throttle release.
This again will result in a loss of speed, thus time, but helps to balance the car because
Pascal has got the time to react to changes he induces to his car. The entry braking at
turn 3 is again too gentle: Pascal leaves his right foot of the throttle and starts breaking
25 metres later. The effects of this event can clearly be seen on the speed trace. The
trace is rounded at the top and the value decrease slowly instead of increasing (thus
throttle). Here again, Jobo is about 30 km/h faster at the start of the entry of the bend
(brakes on). Pascal has got the advantage of being slower, hence, he is able to brake
17m later and leaving the brakes for what they are 22 metres earlier. Consequently, his
speed at a certain moment matches Jobos, but Jobo applies throttle earlier than Pascal.
Consequently, it result in a speed advantage for Jobo equal to 17 km/h at the exit of the
Gauche de la Bretelle , even if he had to release the throttle for a while (and
countersteer) due to oversteer. Note that the exit of turn 3 is taken when the steering
wheel is taking its straight line position. Additionally, Pascal does not apply a throttle
of 100%, while Jobo does.
One does not have to have experience to notice that the steering wheel output of Jobos
run has got more spikes than Pascals. A reason for this behaviour is that Jobo is closer
to, or even on the limit of the car. This implies that he needs to correct the cars travel
constantly. Another reason for these corrections is that the car may have a bad set-up.
That point will be discussed later on in this work. In addition to the high correction rate
of Jobo, one can see that the steering angle on straights is not equal to zero. Figures
likes 3 or 4 degrees are acceptable, but 90 is a little over the limit. The cause, the
drawbacks and the method to repair that problem will be detailed in the chapter called
Diagnosis.
117
V. Data Analysis
1.1.2
Turn 4,5,6 and 7
Fig. V.5: Turn 4, 5, 6 and 7
The reference speed at the entry of the Parabolique is taken at the moment Jobo starts
to brake (like every other turn in this discussion) and is 32 km/h higher for the almost
professional driver. Again, Pascal does the same error by leaving the accelerator (not
abruptly once again) several metres before he starts to brake. Moreover, he starts to
brakes earlier than Jobo while being much slower than the latter one. While starting at
the same moment to apply some right foot on the throttle, Pascal needs 50 metres more
to be at 100% and makes the error of upshifting during his increase in throttle
application. It is an error firstly because the car will have the tendency to lose grip at the
rear due to a longitudinal (to the front) weight transfer. In addition to the first error, he
upshifts 2000 rpm too earlyThus, he would have been able to keep the car steadier by
not changing gear. Nevertehless, Pascal continues all the way long to upshift too early.
As a result of that, he manages to have the car in fifth gear while he should be in fourth
like Jobo who is changing gear at the correct engine revolutions.
118
V. Data Analysis
For the entry of turn 5, Pascal does not change his habit and refuses to go full out with
the throttle during 56m. A direct consequence of that strange behaviour can be seen on
the rpm and speed trace like illustrated on the next page. There can be seen that both
traces are much spikier for Jobo than Pascal. Consequently, the entry speed is 29 km/h
higher for Jobo, even if the straight line prior to the braking period is only about 200
metres long. One can see on the speed trace that the deceleration of Jobo is higher than
that of Pascal. This can be seen via the negative ramp of the trace on the figure below
(white cursor/line).
Fig. V.6: Negative ramp of deceleration, bad throttle application and wrong rpm at upshift (Pascal)
Another method is to look at the X channel which represents the longitudinal G. A

good reason for the higher deceleration is the higher brake pressure applied by Jobo. For
the GT3, Ferrari did not install a brake pressure sensor, but other lines can be used to
determine whether or not he was braking harder. The advantage here is that no ABS or
ASR was present during the entire session due to a failure of a wheelspeed sensor (see
chapter Diagnosis). This can be seen via the channels ABS_ON (if 0), ASR_OFF (if 1)
and ABS_ER (if 1). Therefore, the data engineer can immediately conclude that Jobo is
119
V. Data Analysis
better at braking and accelerating than Pascal viewing the following graphs.
Additionally, one can also see that Jobo is using the technique of trail braking much
better than Pascal. This can be seen on the X-Y graph later, which has a larger heart
form and is slightly longer in the vertical direction.
Fig. V.7: Johan Boris Scheier, Champon Clio Cup France 2006, Driver of the Morgan Aero in FIA
GT with Lafitte and FSSA GT3 driver with the Thierry Boutsen Energy Racing team
Fig. V.8: Braking performance, Higher Gs for Jobo (blue trace) in longitudinal (X) and lateral
acceleration (Acc_Y)
120
V. Data Analysis
Fig. V.9: X-Y Graph for Gs, Better trail braking for Jobo (rounder at bottom)
Still in turn number 5, there can be noticed that the difference between both drivers at
the lowest speed is 8 km/h, but Pascal still refuses to go flat out at the exit of that turn
while Jobo does. Jobo considers turn 6 as straight and does not brake while Pascal
releases the throttle (to diminish/keep the speed). Consequently, the slope of the speed
trace is steeper for Jobo than Pascal after the exit of turn 5. This implies that the top
speed at the entry of turn 7 is about 30 km/h higher for Jobo compared to his co driver.
One might have noticed the two spikes in the throttle trace for Jobo (blue) at the exit.
This is due to the fact that the car has a tendency to understeer (see Angle trace). Thus,
he forces the car to oversteer by applying too much throttle. The greatest difference
here between Jobo and Pascal is speed and therefore, Jobo has to correct a lot (too much)
with the steering wheel in turn 7. One may conclude that Pascal starts to brake 10 m
earlier than Jobo just after the light bend of turn 6 to enter turn 7.
While Jobo uses the throttle at 100% between turn 7 and the Courbe de Pouas, Pascal
does not even think about it. Moreover, the gentleman driver even dares to upshift while
being slow. Both drivers brake at the same moment and because of the throttle issue,
Pascal is 33 km/h slower at the entry of Pouas. The problem now for Pascal in Pouas
is that his revs are too low, but he does not dare to downshift because this would induce
a weight transfer. That kind of weight transfer in a bend is not desired by the driver
especially in fast bends like the one of Pouas. One should know that Pouas is a bend
which goes upward with a light reverse banking. This does not help the driver to keep
121
V. Data Analysis
the car on track. Jobo and Pascal start to push their right foot on the accelerator at the
same moment, but Jobo applies it at 100% and 180 m (!) earlier while still being in the
corner. This settles his car by pushing the outside of the car on the road surface. Due to
his early throttle application, he is able to keep a 30 kph difference after the corner exit
till the finish line. Although Pascal is not applying full throttle at the beginning, he
keeps that difference due to a lower aerodynamic resistance (lower speed) compared to
Jobo.
To conclude with that data we can see that Pascal:
Does not apply full throttle immediately after the exit of a turn
Leaves the throttle gently while not even braking several metres before the
entry of the bend only 19.78% instead of the 53.02% full throttle compared
to Jobo
His revs are too low (about 2000 rpm) when upshifting, so he does not use the
full potential of the engine (see figure below, blue dots). Revs are also too low
especially in 4th gea
122
V. Data Analysis
Fig. V.10: X-Y graph for Engine performance, RPM vs Speed for both drivers
The pressure applied on the brakes is not as high as Jobos
Uses less trail braking than Jobo (see X-Y graph before)
Goes slower than Jobo in bends which results in lower lateral accelerations
A general comment on this data is that there was no ABS or ASR working. The car did
not get a specific set-up of our team prior to drive. This was a sort of shakedown of the
race car. Therefore, Jobo noticed a lot of understeer.
1.2
Ferrari F430 GT3 FFSA: J.B. Scheier vs J.B. Scheier
As told before, Jobo felt some understeer on the car. During Race 1, which comes just
after the First Qualify and before the Second Qualify, he even managed to have the car
totally unstable, fighting with the steering wheel to keep the car on track with decent
times. This happened while being 2 to 3 seconds slower per lap, using his throttle less.
Deeper analysis of the Ferrari and its set-up before the second qualify resulted in a very
strange and disappointing conclusion: the car was 10 mm too low compared to the setup done in Switzerland. Therefore, the car became totally unstable because the front
spring-damper units were resting on their bump stops. This implies that neither the
dampers, nor the spring were changing status during the transient movements of the car.
This was not the greatest problem. Due to the very low front, the car did not have
enough grip (transition handling) at the back which gave the car a very oversteering
behaviour, especially when accelerating. The external front wheel was constantly
123
V. Data Analysis
jumping of the track and the front splitter hitting the tarmac while the driver tried to
place or settle the car in the mid corner.
This phenomenon did not happen during the free practice because the track was cleaner
thus more grip was available at the back. Moreover, the dampers were new and still full.
Indeed, the dampers used are gas dampers and when the damping rates are high, they
have a tendency to lose preload pressure (16 bars instead of 25 bars). Consequently,
their damping rate lowers and the car loses harshness during transient movements like
braking or accelerating (the so-called low speed damping rate). In addition to all these
problems, the ASR and ABS still did not work yet. If ASR would have been present, he
would have felt the same, but much lower.
Fig. V.11: Pressure checker and damper unit (without bumpstop) [3]
124
V. Data Analysis
Fig. V.12: Race 1 (red) vs Free Practice (blue). Huge steering problems
After the set-up modifications, where height of the car (higher), camber (more camber)
and toe (more out at front) of the wheels were modified, Jobo was able to go on track
with a competitive car. This had two immediate consequences: The driver felt mentally
better and safer, so he wanted to go for it [1]. Secondly, the car was better, so it had the
potential of being faster as well. There should be noticed that the data shown below is
Race 1 compared to Race 2. The second race was run with ASR and ABS thanks to the
new wheel bearing with its incorporated wheelspeed sensor. The ABS improves the
steerability at the entries while braking. One should know that the Ferrari GT3 is able
to be faster without ASR because throttle can be applied earlier without being disturbed
by the rear brakes. Indeed, ASR corrects the trajectory by working on the rear brakes
only. Nevertheless, it is advised to use the ASR during the one hour race and to
disconnect it for qualifying sessions. More details on the ASR system can be found in
the Diagnosis chapter.
Their can be seen that Jobo was almost a full second faster in the second race compared
to the first one. Firstly due to the fact that the ABS and ASR were working, but also due
125
V. Data Analysis
to fresher tyres compared to the set used in the first race. Indeed, the tyres in Race 1
had already done the laps ran by Pascal in that same race. Moreover one can see in the
traces on the next page that the Steering Angle in Race 2 is less spiky than the blue one
of Race 1. Thus the change in set-up worked
Fig. V.13: Race 1 compared to Race 2. Best laps both by Jobo Scheier.
Their can be seen that the understeer in the entry of the corners has disappeared. This
can be seen by the slope of the steering wheel, which is higher for the blue trace
compared to the red one. The oversteering undertsteering entrances and exits (unstable
in fact) in some bends have disappeared as well. The damper-spring units have enough
travel and the preload on the springs is lower as well. Consequently, it is easier to settle
the car, and the grip at the front compared to Race 1 is improved. One should remember
this phenomenon as the front of the car going from left to right and vice versa without
the back end of the car slipping. Only good drivers like Johan Boris Scheier are able to
describe such things to help their race engineer finding a solution. For other drivers like
126
V. Data Analysis
most of the gentleman drivers, the engineer will have to use a bunch of sensors on the
car and a lot of experience. Thanks to the better set-up, Jobo was able to accelerate
earlier, and to brake slightly harder in the corner entries, thus improving his trail
braking. The latter effect can be seen in the X-Y graph on the next page while the better
throttle usage can be seen in the figure just below. Of course, as told before, ABS and
ASR are of good help to both these improvements.
Fig. V.14: Higher amount of full throttle application in Race 2.
Fig. V.15: Some extra red dots during combined cornering and braking, thus better trail braking
1.3
Ferrari F430 GT3: P. Ballay vs P. Ballay
The laps discussed below are the best laps of Free Practice (1:32.843) and Qualify 1
(1:27.613). Both sessions had the same technical (same set-up and technical problems)
and track conditions. This increases the influence that the driver has in the
improvements or changes in data traces. One can easily calculate that Pascal improved
by approximately 5.2 seconds from one session to another. This event is not magic.
After the long debriefing with the race engineer, he discussed his driving with Jobo.
This helped him to improve, but also to have some basic knowledge of what he should
do in every situation, learning the reference points Jobo has.
127
V. Data Analysis
Fig. V.16: Free Practice compared to First Qualify of Pascal.
A quick eye on the data will demonstrate good improvements. Even though, this is not
yet a good performance, but Pascal, like any other driver, needs to take error by error
and not everything in one time. Thus after his First Qualifying session, Pascal was taken
back again to show him is errors. It is not a secret telling the reader of this work that it
took over one hour to discuss everything in detail. Of course, the race engineer has to be
gentle and enthusiast with the driver and his performance [5]. The contrary will have the
negative consequence of placing a lot of pressure on the gentleman driver and will also
place him in a context where he is not used to be. Often, gentlemen driver are great
business man or rich managers. They are not used to be told what to do. Especially not
by someone who is paid by him. There has to be a friendly relationship between the race
engineer and his driver. This improves the communication and the trust they will have
in each other.
The conclusions of the improvements are:
More 100% throttle application. An increase of about 11% on the overall time,
but relatively seen, it increased by more than 50% (Free Practice = 19.78%,
128
V. Data Analysis
Qualify 1 = 31.09%). There is still space for improvement, especially compared

to the 50% of Jobo.
Fig. V.17: Hystogram of throttle application. Best laps from Free Practice vs Qualify 1 (blue) from
Pascal
The trace of the throttle is more squared no throttle release (or less) without
any reason and without braking.
Revs of the engine are higher (blue dots figure below) and the moments of
upshift are more appropriate. Although, higher engine revs can be still be
obtained.
Less playing with gearbox.
Faster in each turn, except for the entry of the S des Sablier.
A greater deceleration, especially in turn 1, 5 and 7.
129
V. Data Analysis
Better use of the trail braking technique (more blue dots) added to fact that
Pascal is accelerating harder and braking harder than in his first practice session.
There can be seen that the lateral Gs are higher as well. the blue dots are
spread wider than de red ones.
Fig. V.18: X-Y of Gs. Best laps from Free Practice vs Qualify 1 (blue) from Pascal
130
V. Data Analysis
Spa-Francorchamps
Since the early 1920s, the Spa-Francorchamps circuit has resounded with a thousand
and one fabulous tales of those heroic times in which they drove on earth roads at the
wheel of awesome machines.
In 1939 human imagination gave birth to a unique bend: the Raidillon - Eau Rouge was
to become famous throughout the world. As motor sport has evolved, the circuit has
always been kept up to FIA requirements in terms of safety over the most prestigious
competitions such as Formula 1 and Endurance Sportscar Racing. It has gone back to its
roots with the Sports Prototypes. The track itself was modified to host the F1 Grand
Prix of 2007. The main track modification done during the extraordinary works is the
chicane before the well-known Source hairpin turn.
1 : Source (larger since 2007)
2 : Raidillon Eau Rouge
3 : Kemmel
4 : Les Combes
5 : Rivage
6 : Bruxelles
7 : Pouhon
8 : Fagnes
9-10 : Stavelot
11-12 : Blanchimont
13 : Chicane (new since 2007)
Length: 7004 meters
A detailed plan of the track can be found at the end of this chapter, where all heights
and lengths of the track are provided to the reader.
2.1
Ferrari 430 GT3 FIA GT 24 hours of Spa
During the FIA GT 24 hours of Spa, the car runned by Thierry Boutsen Energy Racing
was placed in the team of JMB Racing. The teams have competed in the new version of
131
V. Data Analysis
the Coupe Du Roi and won it. This race is a relay race where each of the three cars
(with two drivers in each car) has to run at least 45 minutes to maximum an empty fuel
tank. If one car cannot finish its relay, the next car will have to wait till the end of the
relay to start again. The car which could not finish its relay will not be able to drive
again. Consequently, the team will have to finish the race with two cars. The team
finishing with 3 cars will always win, even if they are 20 laps behind the one with two
cars.
The car which Massimo Bottiglieri was held responsible for was the 201C (203 during
qualify), with the two French drivers: Pascal Ballay and Damien Chanard. Both never
raced on the track of Spa and the latter one never drove a GT car before. Therefore,
there were some doubts about the fact if the latter driver would be able to keep the 515
hp car on track. At the end of the weekend, the entire team was surprised by the decent
performances of D. Chanard. He was able to set the best lap time of the 6 drivers (two
seconds faster!) during the race, he had the second best time of all 430 GT3s during the
race and he kept the car on track. The latter event was not really true for the gentle man
driver P. Ballay. He was able to spin twice and he managed to drive the car into the wall
at 45 minutes of the end. Luckily, he was able to drive the car till the box, where the
team was able to do the necessary repairs (only 3 minutes) to finish his relay.
2.1.1
Pascal Ballay vs Damien Chanard
The discussed data in this section is the one gathered during the third qualifying session.
The track was dry and both drivers were briefed four times about their driving errors.
One should know that no major modifications can be made on the car if the driver does
not drive in constant lap times. This would be like guessing what the car could have.
Therefore, only minor modifications were done the night before this session. These
modifications were:
The car was placed on the corner weights to check the static mass distribution
The damper pressures were checked (25 bar)
The rear wing angle of attack has been modified from 4.5 to 6. Both drivers
felt uncomfortable at the Raidillon and Blanchimont (really fast bends) due to
132
V. Data Analysis
the back moving losing grip. More downforce at the back should stabilise this.
Again the driver will feel more comfortable only by knowing they have more
downforce.
At this moment of the race weekend, Damien (2:32.482) was about 4.9 seconds faster
than Pascal (2:37.386). This is not bad for someone who never drove a GT car before,
especially against someone who did like Pascal. Note that they both had the same
amount of fuel, but that Damien drove the second part of the session, thus with old
tyres.
A specific feature of Spa is that there are two startlines, the F1 before the Source
hairpin and the Endurance one which is after the famous hairpin. Therefore the
discussion of the data will start with the Raidillon and not with the Source (24 hours
is an endurance race not?).
Fig. V.19: Entire Best laps of Damien (Red) and Pascal (Blue) in Qualify 3
133
V. Data Analysis
2.1.1.1 Raidillon Kemmel Les Combes

At the attack of Raidillon (which goes seriously upwards), there can be seen that
Damien is breaking 57 metres earlier than Pascal. This is not correct. Pascal brakes at
the good moment which is when the car starts to be compressed against the ground due
to the vertical acceleration. This will stabilise the car when braking, because the forces
on the tyres are less abrupt. One can clearly see that because Damien is breaking too
early, he has to apply some throttle again. This is really dangerous knowing the speed
he is travelling in such a dangerous bend. This is how both drivers do in the Raidillon:
Damien : Brakes Throttle Release Throttle Throttle Lots of weight transfers in
a bend!
Pascal : Slowly Releasing Throttle Brakes- Release brakes Brakes Nothing
Throttle
Fig. V.20: Raidillon Kemmel Les Combes. Red traces for Damien and Blue traces for Pascal.
134
V. Data Analysis
There can be seen that Damien is turning in slightly earlier than Pascal, which compared
to all other drivers on track is wrong. His speed trace is really strange due to his wrong
braking and throttle application. Even though, when exiting at the top of the Raidillon,
Damien has got a speed advantage of about 30 kph when he starts to apply full throttle
to go into the direction of Kemmel. On the other hand, Pascal is breaking his car too
much because he is afraid of this mythical curve. The problem is that this disadvantages
him for an entire straight until the bends of Les Combes (turn 4). One can use Wintax
to see that Pascal loses over 2.1 seconds from the entry of the Raidillon until the exit
of Les Combes. He loses so many seconds not only due to his lower speed, but also
due to his greatest problem. Indeed, again he is not using his throttle or his brakes
before the entry of Les Combes. He is gradually releasing the throttle 65 metres before
he start to brake gently. This low brake pressure can be seen on the small negative slope
of the speed trace and the Gs (on another graph). Then he suddenly decides to release
the brakes and not to use his throttle for about 80 metres before braking again an having
more or less the same speed in the curve as Damien. One can see on the speed trace that
he is not applying the same brake pressure as Damien (slightly smaller slope).
Damien is 13 kph faster than Pascal at the exit of the left of Les Combes. One can also
notice that Damien is applying a little bit of throttle to rotate the Ferrari between the left
and last right of turn 4. This is helped by the little brake application (Longitudinal
weight transfer) while turning the steering wheel to obtain a lateral mass transfer
between. Nevertheless, one can see that Damien has to increase his steering input due to
understeer. This is completely normal because that curve has got a negative camber
which will attract the car to the external part of the track.
135
V. Data Analysis
2.1.1.2 Rivage Turn 6 Pouhon- Fagnes
Fig. V.21: Rivage-Turn 6 Pouhon - Fagnes
The following important curve is the hairpin of Rivage. Pascal is again doing his usual
thing during 54 metres, again losing times and impulse to rotate the car. The angle trace
shows the reader that Damien is turning in slightly later (kind of late apex), but also that
he is suffering from understeer. This is meanly due to the fact that he stopped breaking
too early and that his car did not have enough load transferred to the front. This would
have lightened the rear and helped to rotate the car around its vertical axis. There can be
seen that Pascal is suffering form the contrary: He is struggling with oversteer because
applying less brake pressure but for longer, while having another raceline which is less
brutal and being faster at the entry and middle of the turn. In racing terms they say that
he has not enough back compared to the front. Viewing both drivers, one can conclude
that Pascal has got a better entry and Damien a better exit. If Damien would have
applied more throttle for a while and start to brake later, he would have less understeer
for sure. There can be seen that Damien has also got some understeer at the exit of
136
V. Data Analysis
Rivage, especially at the end when he is applying more throttle. This means that his
front wheels do not have enough grip due to greater amount of squat at the acceleration.
Fig. V.22: Turns 5, 6, 7 and 8 at Spa
The throttle application between turn 5 and 6 is more squared, starts earlier and ends
later for Damien. Consequently, the small negative (- 4kph) difference in speed he had
at the exit of Rivage became a positive one (+8kph) at the entry of turn 6. The
youngest driver starts to push on the accelerator 10 metres earlier which provides him
with an extra speed advantage. Hence, his huge increase of 0.2 of a second compared to
Pascal on the distance between the exit of 6 and the entry of Pouhon. Once more,
Pascal is doing his strange thing with the throttle Strange enough, Pascal is faster at
the entry of the first left at turn 7, due to a lower braking pressure. One can see that
Damien and Pascal are using another race line. Pascal tries to drive through the left in
one time, while Damien really turns twice. This implies that both drivers have to push
the accelerator on different moments. There can be concluded that Damien is faster at
the exit of the first left and the exit of the second left, but he did not win time compared
to Pascal which is positive for the latter one.
Despite all his efforts, Pascal still does the same mistake by gently releasing his throttle
73 metres too early. Luckily his entrance is slightly faster than Damiens, but he has
some speed difficulties in both curves of Fagnes. Consequently, he loses an enormous
0.3 seconds only on turn 8. Thus, from the exit of Les Combes till the exit of Fagnes,
Pascal loses an extra 95 hundreds of a second.
137
V. Data Analysis
Fig. V.23: Track of Spa seen form the sky
2.1.1.3 Double Right of Stavelot Blanchimont one and two
Fig. V.24: Stavelot Blanchimont. Huge difference between Damien (red) and Pascal (blue)
Strange enough, Pascal starts to brake later than Damien at the entry of the first right of
Stavelot, but brakes during a longer distance which destroys his speed for the exit.
Knowing he braked for too long, he tries to gain some time by pushing on the
accelerator. There can be seen that due to the heavy throttle application that Pascal is
suffering from an instantaneous oversteer which he has to correct with a fast
countersteer (see the spike in the blue steering wheel angle trace). Moreover, after his
throttle release, he applies a full throttle again. This is perfect to make pitch and squat
138
V. Data Analysis
the car constantly and to unstabilise it. Meanwhile, Damien was much gentler with the
throttle and applied it progressively. This gave him a more stable car and thus more grip
at the back to provide him with a better acceleration. Again, he overtook his inferior
speed and became faster than Pascal at the exit of the first right of Stavelot. For the
second right, Pascal releases his accelerator later, but applies it much later at the attack
of the exit compared to Damien. This will result in a speed difference between 7 and 10
kph during the straight of Blanchimont.
Fig. V.25: Turn 9-10 (Stavelot) and 11-12 (Blanchimont)
They often say that Blanchimont can be passed full throttle with a Formula 1, but this
is not the case for both drivers. The angle trace shows that there are two little bends at
Blanchimont. Pascal is doing a huge error by gently releasing his throttle 250 metres
before he starts to brake for the entrance of Blanchimont part 2. Both drivers brake for
the entry of the second part of Blanchimont, but Damien releases it earlier, gives a blip
(to settle car) and then gently applies full throttle. From the exit of Stavelot till the
entry of the Chicane, Pascal loses an extra 1.25 seconds, which is huge, especially
because it is almost straight line. This demonstrates that Pascal does not have the
courage to handle such fast bends (just like the Raidillon). One should know that
Blanchimont is a fast blind light bend and needs to be known very well. It looks
simple but it is not and as long as the driver does not put himself in danger In
addition, the difference in speed between Pascal and Damien stays between the 18 and
the 24 kph during the second part of Blanchimont with exactly the same race line.
139
V. Data Analysis
2.1.1.4 Chicane Source Hairpin (both modified since 2007)
Fig. V.26: Chicane Source hairpin. Source taken in different ways.
Thanks to the higher speed obtained through both parts of Blanchimont, Damien is
able to arrive 12 kph faster than Pascal when he starts to brake. On the other hand,
Pascal is able to brake later due to his lower speed. Meanwhile, the pressure Damien is
applying on his brakes is slightly higher than Pascals. Of course the influence of used
and new tyres plays a role in such small differences in deceleration. Thanks to the lower
deceleration, Pascal has got a greater speed during the first part of the chicane. There
can be seen that Damien is suffering of understeer compared to Pascal, which bring him
at the outside of the left curve of the chicane. Therefore, Damien has to steer earlier for
the right bend of the chicane to be able to keep the same raceline. While Damien is
correcting his trajectory, Pascal is going on the kerb at the exit of the right bend of the
Chicane. In addition, Pascal starts to push a bit on the accelerator which helped him to
spin the wheels at the back and lose lateral traction at the rear. This had to be corrected
with a huge countersteer to recuperate the GT3. Damien starts to be faster at the exit of
140
V. Data Analysis
the chicane firstly due to his better line, but also because he applied full throttle 44 m
earlier. At the start of the straight to the Source hairpin, Damien is 8 kph faster.
Fig. V.27: Chicane and Source Hairpin (both new since 2007)
Arriving at the Source (turn 1), Damien brakes 24m sooner than Pascal. The
difference between Pascal and Damien on that hairpin is huge. Their can be seen that
Pascal starts to turn earlier but is quite smooth. On the other hand, Damien starts to turn
later and more heavily. The little spike in the red trace starts when he leaves the brakes
and is a consequence of the track surface that has a little bump. Therefore, the car has a
tendency to spin and Damien has to correct a bit while turning in the hairpin. Pascal
does not have this problem because he is going larger, while Damien is closer to the
internal wall at the Source. Note that Pascal is still braking (an extra 14m). Damien
steers back earlier as well and again we can see a spike in his trace, demonstrating the
second bump on the road. Following the steering trace we can see that Damien is
running larger at the exit of Source because he has to apply a bit of a right steer to
come back on a normal race line (this is now possible since 2007). Thanks to that
particular line, Damien can push onto his accelerator 34 metres earlier. Therefore, his
lower speed at the entry is now becoming an advantage of 15 kph at the exit of the
hearpin. Again, Damien will save a lot of time compared to Pascal at the arrival of the
Raidillon.
141
V. Data Analysis
2.1.1.5 Conclusion of comparison

To conclude with Damien compared to Pascal:
Damien applies more throttle than Pascal and has got a better on/off mentality
between throttle and brakes. There can be seen on the histogram below that
Damien is procentually seen more often applying full throttle.
Fig. V.28: Throttle Application: Relatively 16% more full throttle for Damien compared to Pascal.
The two tracks plotted below are also useful to have a quick overview of the
braking behaviour on track. There can be seen that Pascal is braking for longer
distances and sometime on the wrong moments. Nevertheless, Pascal brakes at a
better moment to climb the Raidillon while the car is in compression, but for
too long. Additionally, the arrival at Les Combes is incorrect and the passage
through Blanchimont is not good compared to Damien.
142
V. Data Analysis
(a)
(b)
Fig. V.29: Brake application from both drivers (a) Damien Chanard ; (b) Pascal Ballay
Damien is more brutal with the steering wheel and is a fervent fan of the late
apex technique. (Source is a derivative of that technique).
The plotted tracks below show the speed that the cars had with both drivers
behind the steering wheel. An immediate remark is that Damien (7.6%) passes
the 230 kph more often than Pascal (1.6%). There can be seen that Damien is
much faster than Pascal in the section from turn 1 till Raidillon and the entire
Blanchimont section with the bends and straights.
143
V. Data Analysis
(a)
(b)
Fig. V.30: Speed visually demonstrated on track. Huge speed differences in the advantage of
Damien (a) compared to Pascal (b). Raidillon, before Les Combes, Blanchimont and before
Chicane
Another way to prove that Damien is running harder with his car is to watch at
the rpm of the normally aspired V8 engine versus the speed of the race car. One
can notice that there are more red dots at the top of the X-Y graph, especially in
5th and 6th gear for Damien.
144
V. Data Analysis
Fig. V.31: RPM versus Speed, generally higher revs for Damien (red dots)
The parameters like tyres temperature and pressure are not gathered by the data
acquisition system but via other measurement tools. During the entire weekend,
Damien had approximately 0.1 bar more pressure and a tyre temperature of
around 5C higher. This is indicative to the way he is using his tyres and thus
shows he is dissipating more energy due to his higher speed.
Both oil temperature and water temperature are 2 to 6C higher for Damien
(same track conditions), hence, is dissipating more Watts. The Oil pressure is
slightly lower as well, this especially due to the higher temperature of the
lubricant. Indeed, higher temperatures do decrease the viscosity of the engine oil.
Consequently, the build of oil pressure is more difficult. One can see that low
RPM will result in lower oil pressure (look at the Source at the end).
145
V. Data Analysis
Fig. V.32: Oil pressure, Oil Temperature, Water Temperature and RPM.
Comparing the gear engagement of two drivers with each other (one who knows
and the other doesnt) is pretty useful with the track plot below. Nevertheless,
this was not useful during the weekend for both drivers. The figures below only
demonstrate that Damien (17.5%) is more often in 6th gear than Pascal (15.3%).
146
V. Data Analysis
(a)
(b)
Fig. V.33: Damien (a) is more in 6th gear than Pascal (b).
(a)
(b)
Fig. V.34: (a) F430 GT during qualify 3 at 24 hours of Spa and (b) after the race
147
V. Data Analysis
Silverstone UK
The Silverstone track is a Grand Prix track just like Spa Francorchamps is. The British
circuit was the host of the Ferrari Days 2007 UK where the 60 years of Ferrari were
celebrated. A fantastic weekend, with historic race, FXX demonstrations, customer F1
car demos (F1 Clienti) and last but not least, the Ferrari Challenge Trofeo Pirelli Europe
and Copa Shell Europe.
Fig. V.35: Ferrari F1 Clienti and Ferrari 250 GTO from 1962
1 : Copse
2 : Maggotts
3 : Becketts
4 : Chapel
5 : Stowe
6 : The Vale
7 : Club
8 : Abbay
9 : Bridge
10 : Priory
11 : Brooklands
12 : Luffield
13 : Woodcote
Track Length : 5140 m
The Silverstone GP track is a really fast one, with difficult bends and complexes to
handle. The top speed reached at the entry of Stowe is around 250 kph for the F430
148
V. Data Analysis
Challenge. This Ferrari only uses the principle of ground effect without any wing or
spoiler, so is more difficult to handle in the curves.
Fig. V.36: Massimo Bottiglieri and his Belgian driver with the Ferrari F430 Challenge at
Silverstone
3.1
Ferrari F430 Challenge: Gentleman driver vs Professional Driver
The data discussed below will demonstrate the difference between a very good
gentleman driver (Louis Machiels) and a young professional driver (secret) from the
Trofeo Pirelli Europe. The problem with that data is the fact that both drivers did not
drive the same car. This may result in another set-up of the car with other tyre pressures,
different amount of fuel embarked or other static heights of the car regulated via the
spring-damper unit (thus eventual preload on springs). Both lap times were set during a
qualifying session, thus new tyres were used by both drivers. The difference between
both is that the pro only needed one unique lap to place his name on the pole position
(1:57.062, red), while the gentleman driver got his best time in the 5th lap (1:58.007,
blue). One should know that new tyres only need one warming up lap, and are then able
to provide the driver with the highest amount of grip during two laps. After these three
(warm up + 2 laps) laps, the performance of the tyres suddenly decreases and better
times will only be reached if the driver did errors during his hot lap. Any lap added to
these three will only wear out the tyre resulting in less performing tyres for the race.
149
V. Data Analysis
3.1.1
Copse Maggotts Becketts Chapel
Fig. V.37: Copse Maggotts Becketts Chapel. Blue trace for the gentleman driver and red for
the pro
Louis arrives at Copse at a speed of 227 kph while the pro has got an average speed 4
kph higher which advantages him for about 0.15 of a second. Louis is struggling with
his car at the entry of Copse. The speed trace shows some downward spikes, which can
indicate that the wheels have a tendency to lock up. This problem can be seen just at the
moment Louis starts to brake. The data shows (Fig. V.38) that the front left wheel speed
suddenly drops for a few kph (14 compared to the Right front tyre). It is normal to have
a difference due to the different radius the tyres are revolving on, but this does not
induce a sudden drop in wheelspeed. It looks like Louis has got some problems with his
front tyres. He does not have enough grip to turn the car in like he would like to.
Consequently, he has to destroy his speed to be able to reach the correct exit point.
While the pros car keeps its trajectory, Louis has to fight with an unstable and
understeering car and therefore slower. One can notice that Louis is even trying to
rotate the car by adding a 100% throttle to make the car spin at the back Some
reasons for that difference between him and the pro can be:
150
V. Data Analysis
Louis is breaking too hard There is not enough grip available to turn because
the longitudinal deceleration takes is all (circle of friction)
The car does not have the same pitch angle compared to the pros car lower
pitch angle (less nose down) than the pro. Having a higher pitch angle will result
in a lighter back at the moment of forward weight transfer. A direct consequence
is that the rear wheels will have a higher tendency to lock (which they not do on
the pros car). The latter effect will improve the yaw (rotation around the
vertical axis of the car) providing the possibility to have an eventual oversteer.
All brought together: the car turns better and has less understeer thanks to the
rotation of the car.
Fig. V.38: Wheelspeeds, brakes, ABS and 3 math channels for the wheelspeed difference
The force provided by the springs may be too high (spring rate and preload)
everyone has got the same springs. Therefore one may think that Louis is
driving on his bump stops. Consequently, the car jumps and the force to obtain
the grip at the tyres changes too much which unstabilises the car. A good
151
V. Data Analysis
solution to counteract that particular problem is to use packers. These will

decrease the tendency to touch the bumpstops.
The compressions are not damped enough which makes the wheel jumping of
the road surface. This implies that the wheels tend to lock when braked.
Everyone has got the same dampers, so only tyre pressure can play a role in the
damping ratio of the wheel damper unit (assuming the dampers are ok).
Harder anti roll bar for Louis. Having a softer anti roll bar can provide the car
with some extra mechanical grip at the front. This softness has to be compared
with the rear anti roll bar. Later knowledge told that both cars had the same anti
roll bar (21 mm diameter = hard and not the 19 mm = soft).
Of course Louis problem can be something else or a mixed event, but the fact that both
cars have different set-ups is not a case to forget.
While being faster at the entry of Copse, the pro gets his steering wheel back in position
earlier to go larger sideways and to search for grip at the front wheels and to put his car
in a light oversteer by pushing on the throttle 20m earlier than Louis. Only on that
portion of the corner, Louis is losing 0.3 of a second compared to the fast driver. This is
huge and is a proof of the fact that understeer really damages speed, hence lap times.
During the straight between Copse and Magotts, Louis trespasses the pro on the speed
trace at the moment the pro upshifts (at 8413 rpm). A good raison for this can be the
fact that Louis was oversteering less at the exit compared to the pro. But to be able to go
faster, Louis needs less drag or weight. This drag can be translated into several forms
where the most important once are:
Aerodynamic drag: Car height and pitch will result in different aerodynamic
drag. The closer to the ground the higher the drag, but the higher the downforce.
Consequently, this extra downforce can be a reason for the pro having more grip
in the curves. The higher the pitch, the more downforce and drag. This of course
is a general rule and some limits exist. Moreover, the reader needs to be aware
152
V. Data Analysis
about the L/D factor (lift/drag) which is in function of ride height and pitch
angle. Again what was told for the weight transfer and the springs can be correct
by admitting that the pitch angle was slightly higher for the pro and thus limiting
his understeer [6].
Tire drag: The more the toe of the tire is out, the higher the drag and thus the
slower the car goes. On the other hand, toe out helps to counter understeer. By
applying toe out the tyre rubs over the tracks heating the tyres for a better grip,
but slowing the car down. Additionally, toe out will provide the car with another
directionality when turning due to the internal wheel rubbing and pulling the car
at the interior of the track. Another form of tyre drag is a consequence of tyre
pressure. The higher the tyre pressure, the smaller the contact patch of the tyre
with the road surface and thus the lower the tire drag. Again there has been told
something about damping and tyre pressure that could influence the latter
kinematics feature earlier on[6]
Still during the straight, Louis loses his advantage on the pro at his shift from 4th to 5th
upshift, Louis is only 1kph slower instead of the 6 at the exit of Copse. When arriving at
Maggots, Louis gently starts to release the accelerator feeling that the car does not stick
to the ground (think about the ground effect and its downforce) and hence, loses speed.
In addition, his race line is not the same as the pro who goes straight on while Louis
starts to turn. This is also a reason for his throttle release because the pro does not have
to turn
Fig. V.39: Left right left of Becketts
At Maggots, Louis shifts a gear back down which is not ideal for the stability of the car
(forward weight transfer). Once more, there can be noticed that Louis is at the limit
(maybe even over) of the tires grip. Nevertheless, Louis applies full throttle earlier and
153
V. Data Analysis
for longer between the first left (3a) and the right (3b) of Becketts which speeds him up
to a 9kph extra compared to the pro. Due to his higher speed, the gentleman driver
needs to brake harder and during a longer distance to enter the 3b at a correct speed.
This results in a back end that becomes vivid (finally!) thanks to the weight transfer in
turn 3b. Therefore, Louis needs to heavily countersteer twice to keep the Challenge on
track. Between the exit of 3b and the entry of 3c, Louis is able to push on the throttle at
100% while the pro only tries Unluckily for Louis, he starts to brake later and for
longer again. There can be seen that Louis car has got the tendency to lock up the
wheels again. Turn 3c goes better for the pro and he is able to push fully on his
accelerator 17 m earlier and both driver pass Chapel with their right foot pushing on it!
There can be seen now that during the entire straight from Chapel till Stowe the speeds
are almost identical for both drivers. Moreover, one can conclude that gearshifting is a
really crucial event that can cost a lot of speed and thus time. Although, the ideas told
earlier can still be true. There is still one point not to forget: the wind!
3.1.2
Stowe The Vale Club Abbay
Fig. V.40: Stowe The Vale Club Abbay
154
V. Data Analysis
Arriving at the entry of Stowe, both drivers attain their top speed of 246 kph (data
acquisition speed, not measured via radar thus Doppler Effect). Once more, Louis
car demonstrates a tendency to lock up the wheels while braking hard (downwards
spikes in speed trace). It looks like he has got some troubles with grip again (and that
will be the problem during the entire lap) because his steering output is going from left
to right. Louis commented later on that only his front wheels were sliding. This means
that he did not had the back of the car sliding away (a real oversteer). Thus, the front
loses grip giving the car an understeer. Subsequently, the front tyres get a sudden grip
and the car oversteers due to a greater steering input than required. Therefore, Louis
needs to correct his steering input comparable to a countersteer. One may think that his
front wheels are not fully sticked onto the ground surface and that the spring-dampertyre pressure system has got some problems. There can be seen on the wheelspeed trace
below that the front right wheels wants to lock up, meaning that the forces applied on it
are to low or incorrectly applied (damping) to obtain a higher Fx.
Fig. V.41: Wheelspeed channels and brake channels from the second sector of Silverstone.
155
V. Data Analysis
All these problems can come due to his higher speed during the entire bend, which turns
around the 14 and 10 kph from entry till the start of the exit compared to the pro.
Thanks to the higher amount of grip available, the pro is able to accelerate faster and
catches his lower speed up to finish with a 1 to 2 kph difference during the straight
between Stowe and the chicane of The Vale. The pro starts to break 10 metres earlier
than Louis and his entry is slightly slower. Although, something strange is happening
here: the steering input of the pro is really brutal at the entry and especially at the start
of the exit of the Vale. A good explanation is that he is using the lateral weight transfer
to unbalance his car and rotate it around the corner by applying and extra brutal
throttle. He then releases his throttle which will result in a forward weight transfer.
Fig. V.42: Stowe The Vale - Club Abbay
(a)
(b)
Fig. V.43: Brutal steering input at Vale and drift for Club (red)
Consequently, the front wheels receive an extra grip while the rear wheels become
lighter and lose grip. Knowing that his steering wheel is still rotated, the car can have a
tendency to rotate around its vertical axis Additionally, he applies a continuously
156
V. Data Analysis
growing pressure on his throttle and looking at the steering input he is still in a sort of
small drift. One notices that his steering input is the contrary of the driving direction in
the middle of the Club. Meanwhile, Louis is struggling again with his car (especially
an understeer / oversteer problem) but is also doing an error by braking-releasingbraking in the mid corner (especially because his car behaves badly). Throwing his car
to drift it around this sequence of bends gave the pro a speed advantage of 3kph at the
exit of the Club, being slightly slower at the entry. Knowing that there is a straight line
coming after the club, tells one that the pro is again saving some time.
Approaching the chicane of Abbay, Louis brakes 10 metres earlier, but is slightly faster
between the exit of the first turn of the chicane and the exit of the other one, because he
releases his brakes earlier. On the other hand, the pro starts earlier to push on the
accelerator and catches up thanks to a more neutral-oversteering car. Hence, the pros
speed is 4 kph higher at the exit of the chicane which will provide him with an extra
time advantage
3.1.2.1 Bridge- Priory Brooklands Luffield - Woodcote
The complex discussed below is the part of the track where Louis loses a lot of time due
to his unstable Ferrari F430. While the pro has a more neutral oversteering behaviour
on his car, the gentleman driver struggles with an understeering oversteering car. Of
course, the driver can be at the source of this problem, but if the team can increase the
oversteering behaviour of the car it would have improved the performance of Louis as
well.
157
V. Data Analysis
Fig. V.44: Last part of circuit, where Louis loses everything. Bridge- Priory Brooklands Luffield
Woodcote.
There can be seen that Louis downshifts when entering Brooklands (10) while the pro
keeps his 4th gear engaged. This is a very intelligent move because to handle such a
sequence of slow bends, it is better to have a bit more torque to accelerate faster and to
stabilise the car. Louis is about 10 kph slower at the entry of Brooklands which
penalises him a lot. One can notice that at each entrance of each bend, Louis being
slower of several kph, but his exit at Brookfield is faster because he releases his brakes
earlier.
Fig. V.45: Bridge Priory Brooklands Luffield - Woodcote
158
V. Data Analysis
One can easily see that the pro is more oversteering than the gentleman driver. A good
reason therefore is that he is breaking later and stops braking later as well. This results,
as told before, in a lighter backend (weight transfer) and improves the ability to rotate
the car around its vertical axis. The pro accelerates for a longer period on little straight
between each bend which improves his average speed on that sector. At the hairpin of
Luffield, Louis is 10 kph faster but he is struggling with a huge understeering race car at
the exit (because he is to fast) and loses his ideal trajectory, thus reducing the perfect
longitudinal acceleration and destroys his speed advantage. Nevertheless, due to a better
gearshift sequence at the exit of the hairpin and the entrance of Woodcote, he manages
to reduce his speed disadvantage from 10 to 2 kph. Of course the possible difference in
car set up still can play a major role in that acceleration in a straight
3.1.2.2 Conclusion
Having largely discussed what happened in both laps, there are a few conclusions that
can be taken. One of these is that the pros car was slightly better having less understeer.
On the other hand, the pro uses driving techniques which are very difficult to imitate for
a gentleman driver (even a very good one), especially on a 490 hp car worth over
140k
The pro is procentually seen, applying full throttle for a higher amount of time
during his best lap compared to his concurrent, the gentleman driver. As told
before, the more one is pushing on his throttle, the faster he will be. (Of course
too much throttle in a bend is an error)
159
V. Data Analysis
Fig. V.46: Throttle application for both drivers on their respective best lap
Actually, the pro makes the greatest difference in the sector where his technique
is helped by his oversteering car. Of course he is also at the source of that
behaviour but saving 1.1 seconds only on the third sector demonstrates the
difference between a good and an excellent driver. Handling a race car in a
complex is the most difficult part of driving Even though, there can be seen
that the gentleman driver was 0.5 seconds faster in sector 2, but also lost 0.3
second in sector one and this in only one curve!
Fig. V.47: Respective time elapsed for each sector per driver (blue Louis Machiels)
160
V. Data Analysis
The traces below demonstrate the speed each driver had at a particular place on
the track. Their can be seen that the pro was slower at the start of the second
sector, but was much faster in sector 3
(a)
(b)
Fig. V.48: Speed plotted on the track for each driver. (a) Louis Machiels, (b) professional driver
The data discussed can help the race engineer to decide to do some modifications on the
car, but that data is not always available. In the case of Louis, this data came too late
and was thus of no helpfor that race at least. The race engineer is now aware of what
the other car could have as a set up to be faster. In the opinion of the writer, the car of
the pro was slightly lower having with a toe slightly more out at the front. He also
thinks that the pro was using packers. Of course these are just suppositions and can be
tested later on to check if a difference is perceived by the gentleman driver.
161
V. Data Analysis
Literature Reference
[1] Ross Bentley, Speed Secrets: Professional Race Driving Techniques, MBI
Publishing Company, USA, ISBN 0-7603-0518,1998
[2] Ross Bentley, Speed Secrets II: More Professional Race Driving Techniques,
MBI Publishing Company, USA, ISBN 0-7603-1510-8, 2003
[3] Sachs Race Engineering, Formula Matrix Dampers, 2007
[5] Ross Bentley and Ronn Langford, Speed Secrets 3, Inner Speed Secrets: Mental
Strategies to maximize your racing performances, MBI Publishing Company,
USA, ISBN 0-7603-0834-9, 2000
ISBN 1 85960 653 9
162
VI. Diagnosis
Chapter VI
Diagnosis
ABS ASR not working
1.1
Problem
As told before, the Ferrari GT3 was modified from a Challenge to a GT3 in Switzerland.
Once arrived at Dijon, the car had to run immediately with a few important (safety)
checks done prior to start the free practice. Although all the checks were done, the
driver complaint about a message written on his Magneti Marelli screen after 1 lap
only: ABS ER and ASR OFF. This meant that both ABS and ASR did not work on
the car. This can be really strange for the driver if he is used to drive with ABS
(eventually ASR). The first question asked to the driver was if he disengaged the ASR
with the switch placed on the dashboard. His answer was negative and this implied that
the engineer had to search for the real source of the problem.
Fig. VI.1: ASR switch; central console at the right side
1.2
Theoretical background
ABS is the acronym for Anti-Lock Braking System and was invented to be able to
maintain steering control while braking very hard. It is generally known by the common
driver, that when locking the front wheels while braking hard, that the car (even with an
important steering input) does not turn. This will imply that an obstacle right in front of
the driver will be impossible to counter. For a race driver it means that he will miss his
curve or even drive his car directly into the sand or the wall.
The idea behind ABS is very simple: wheelspeed sensors gather the speed at which each
wheel turns. Subsequently, the information is send to the ABS ECU (electrical control
unit) which calculates the deceleration of each wheel. If the deceleration is too
163
VI. Diagnosis
important compared to the other wheels (thus tendency to lock the wheels, the ABS
ECU will decrease the hydraulic pressure in the braking system of that particular wheel
(for the new generations of ABS) and will thus reaccelerate that wheel (due to impuls).
This calculation goes much faster then our body and brains are able to handle. Therefore,
the driver has to push on the brake pedal and keep pushing it to be able to use the ABS
correctly and pumping (cadence braking) is thus useless because the ABS goes
pumps much faster. Nevertheless, the highest braking performance will be obtained if
the wheel is just not locking-up, applying a threshold braking pressure[1]
(a)
(b)
Fig. VI.2: (a) ABS system [2]; (b) Screen (Magneti Marelli) with alarm leds on dashboard [5]
Since ABS came into widespread use in production cars (with "version 2" in 1978),
ABS has made considerable progress. Recent versions not only handle the ABS
function itself, but also traction control (ASR see next section), brake assist, and
electronic stability control, amongst others. There should be reminded that each ABS
system is calibrated for a car with specific tires, thus grip. It has been told by Ferrari
that the road version of the F430 is not allowed to drive with slicks on racetracks due to
the extra amount of grip (thus deceleration), meaning that the ABS will not work
correctly/efficiently.
1.3
Solution
What components can bring the ABS in problem: the ECU, the hydraulic device, the
wheelspeed sensors and the CAN information (see last section of the chapter). Checking
the car with the specific Diagnosis Software from Waycon for the GT3, told the writer
164
VI. Diagnosis
that one of the wheelspeed sensors gave a wrong signal to the ABS/ASR system. The
next step thus was to check these wheelspeed sensors if they where connected correctly.
A great help is to use the menu on the dashboard of the car by pushing on the scroll
button on the steering wheel.
(a)
(b)
Fig. VI.3: (a) Scroll button on steering wheel [3] and (b) screen coming up the dashboard [5]
There could be seen that three wheels had a speed of 2.7 and the last wheel had an
incredible speed of 511.9 kph. Checking the name of that wheel told that it was the rear
right wheelspeed sensor (Speed_RR) that had a problem. While the engineer checked
the data acquired during the session, the mechanics also noticed that the sensor wire was
broken.
165
VI. Diagnosis
Fig. VI.4: Wheelspeeds, ABS and ASR channels; Note the 511.9 or Speed_RR
There can be seen on the figure above that ABS_ON is never 1, that ASR_OF is always
on 1 and that ABS_ER does not want to be 0. Furthermore, the wheel speeds are shown
while the steering input is 90 clockwise shifted. The latter channel is important for the
ASR. While driving on a straight, the ECU thinks the driver is turning his steering
wheel for 90. Therefore, the ASR (if it worked) would have been very intrusive or
would give an error and automatically disconnect the ASR again. Therefore, four things
had to be done so that the ABS/ASR system would work correctly:
1. Replace the entire wheel bearing and wheel hub. Indeed, the entire block (over
1000 had to be replaced because the bearing is also the magnetic encoder wheel.
The wheel sensor makes one component as well with wheel hub/bearing
166
VI. Diagnosis
Fig. VI.5: Rear right wheel bearing
2. To be able to reconnect the ABS/ASR system, all the errors stored by the ECU
all over the car had to be erased via the Waycon Diagnosis Software, the also
called SD3 system by Ferrari.
3. Rotate (90) the angular position sensor placed around the steering rack back
into its correct position.
4. Calibrate the angular position sensor with the SD3 by following the instructions
given by the software.
2
2.1
Speed and brake problems

Problem
When driving at the 24h of Spa, the qualifying session happens at night as well. When
switching from driver, the tires also have to be changed. Pascal Ballay drives through
the entire pitlane at a speed of 60 kph with the pitspeed engaged (button on steering
wheel) and exits it without any problem. After a few seconds Pascal tells us he cannot
go faster because his pitspeed button got stuck. He feels like his engine is braked and it
takes him about one lap to tell the team that it is ok, that the car accelerates normally.
Finally, when passing in front of the pitlane, the race engineer noticed that the rear
brakes are red. This never happened before and the race engineer asks Pascal to be
careful with his brakes and to enter the pit. Pascal immediately answers with a I have
no brakes! when arriving at the Source. Luckily he was not driving fast.
2.2
Theoretical Background
One should know that the speed limiter on the F430 Challenge/GT3 is obtained via the
throttle (not via the ignition). This throttle is a drive-by-wire type and there is thus no
mechanical link with the engine. Consequently, the speed limit for the pit lane works by
167
VI. Diagnosis
modifying the position of the throttle in an electronic way. This means that more or less
fuel/air mixture will go into the cylinders of the engine to limit the revs of it and thus
the speed of the car.
One should know that ASR is the acronym for Anti Slip Regulator, which is another
name for Traction Control. Basically it detects any difference in the speed of the front
and rear wheels of the vehicle via the same wheelspeed sensors used by the ABS system.
The Traction Control will sense if the rear wheels are going around faster and will brake
them (also via the ABS system) if they do so.
2.3
2.3.1
Solution
Pit speed limiter
Checking the data provided the team with information telling that the speed limiter was
not engaged. Moreover, knowing that the rear brakes were red tells us that the rear
brakes were constantly braking. Thus, one may conclude that the pit limiter was not
working, it was rather the ABS/ASR system
Fig. VI.6: Steering wheel of GT3 and Challenge and its buttons; a=scroll, b=radio, c=pit speed
2.3.2
ASR
As told before, the rear brakes were red (and completely destroyed). Thus to brake only
the rear wheels, there is a need to have a difference in speed. Of course the specific
ASR software of the GT3 could have an bug
After a diagnosis with the SD3 there could be seen that one particular error was always
coming back telling that the magnetic wheels reference had a problem. Subsequently, it
could be that wheel (mounted on the wheels) that had a problem. Of course, the wheel
sensors could have a major problem. All this had been checked as quickly as possible,
168
VI. Diagnosis
but no major sign of default in that electro mechanic system was found. Therefore, only
the data acquisition could provide more information to solve that strange problem.
Fig. VI.7: ASR and wheelspeed problems at Spa
There can be seen on the figure above that the speed of the rear wheels is 13 kph higher
than the front wheels (at 129 kph). This is completely abnormal knowing that the car
is driving on a straight line and that the maximum difference noticed during the season
was around 3 kph. (almost 0 with the Challenge). Subsequently, if the measurement
system does not have a problem and admitting that the ASR software is correct, the
strange idea to check the tires comes up. Bingo! The tire supplier made an error when
mounting the tyres on the wheels. Michelin did the huge mistake to mount a front tyre
for a GT1 (higher diameter) on our GT3 car and to mount the front wheels (lower
diameter) of a GT3 on the rear GT3 wheels. A basic mechanical formula tells:
169
VI. Diagnosis
v = .R
R = wheelspeed radius [m]
= Angular frequency [s-1]
v
= speed on circumference of radius R [m/s]
Thus to obtain the same circumferential speed (equal to the speed of the car), the R.
needs to be equal. Furthermore, the data acquisition system was calibrated on a
particular wheel radius to have equal velocities on each wheel. Therefore, if the speed at
the back is higher on the data, it means that the wheel is turning more often than it
should (thus is higher). So, the rear wheels need to rotate more to have the same
speed as the front wheels. Consequently, the radius of the rear wheels is smaller than
was supposed to be. The ASR thinks the car is spinning its rear wheels and thus acts
braking the rear wheels resulting in the brakes suffering enormously. Of course it only
starts to brake once the speed difference is too great. If the ASR has to be sustained for
too long, it disconnects itself providing the driver with an ASR Error message. This is
what Pascal felt: his car was braked and then suddenly not anymore due to the
automatic disconnection of the traction control.
Five things had to be done urgently because each driver was obliged to drive at least 3
complete laps:
1. Dismount the rear tyres and mount them on the front rims while using the
correct rear tyres for rear wheels.
2. Changing the brake disks and pads at the rear. The pads faded and just 1mm of
braking material was present on the metallic support (instead of about 20 mm)
and the disks were completely destroyed.
3. Brake oil had to be changed due to the fact that it boiled and thus lost its features.
4. Purging the air out of the braking circuit was needed
5. Erasing the errors stored in the ABS/ASR ECU via the SD3.
170
VI. Diagnosis
3
3.1
Gearbox problems
Problem
The driver of the Renault Mgane Trophy BTCS, complained about problems to shift
gears up and down while driving. Knowing that his gearbox was a new one and
mounted on the car just a few days earlier, resulted in some doubts about the labour
delivered and the new gearbox itself. The gearbox used in the Mgane Trophy BTCS is
a mechanical gearbox which is electro-hydraulically actuated and controlled via little
gearshift paddles behind the steering wheel. This system is comparable with the Ferrari
system except for the fact that there is an extra paddle to command the clutch when
engaging the first gear.
Fig. VI.8: Renault Mgane Trophy BTCS of Thierry Boutsen Energy Racing
171
VI. Diagnosis
3.2
3.2.1
Solution
Step 1 : Looking at the values of every channel
Fig. VI.9: a general window showing most of the channels available on the Renault
There can be seen that the barrel voltage of in the little gear box check window is
equal to zero. This value is impossible to have and thus there must be a problem with
the rotational potentiometer. This sensor, tells the ECU and the gearbox electronics in
which gear the gearbox is engaged via the position of the gearbox command. Therefore,
the sensor was dismounted while still being connected to the data acquisition system.
While manually turning on the sensor, the value stayed on 0 mV. Consequently, the
potentiometer was definitly broken and had to be replaced.
172
VI. Diagnosis
Fig. VI.10 Error list
Fig. VI.11: Input check : Barrel diag shows an error on the counter (23), just like all the dampers.
3.2.2
Step 2: Replacing the broken item
The values on the next page are the values obtained after the replacement of the rotary
potentiometer. One can notice that the linear indicator is not in the green sector of the
window. Consequently, the mechanic and engineer know that the rotary potentiometer
is not calibrated well and will thus provide the ECU with errors which will result in the
inability to switch gears.
173
VI. Diagnosis
Fig. VI.12: New rotary potentiometer installed
3.2.3
Step 3 : Calibrating the rotary potentiometer
Fig. VI.13: Calibration of the rotary pot via the general window of Vision 4
174
VI. Diagnosis
To calibrate the sensor, one needs to push or pull the command that engages the gears.
This can be done by feeling if the wheels are free to turn (neutral) and while they are not.
Once the first gear is engaged, the mechanic needs to rotate the potentiometer until the
blue indicator falls into the limits of the green area of the Gear box check window.
Once this is obtained, the mechanic can tighten the screw which fixes the sensor to the
gearbox.
3.2.4
Step 4 : Checking the gearbox while engine works
On the figure below a few channels can be observed while the engine is running. There
can be seen, thanks to the RPM channel, that the engine is running (1901 rpm in idle).
To be able to shift in the first gear, there is a need to manipulate the control for the
clutch on the righthand side of the steering wheel. This manipulation can be measured
on the clutch itself via the channels INBV (sped of primary axis of gearbox), Clutch
Position and Clutch Lever.
During this test, something strange, but logic at the same time appeared. Once the
screen showed 1 there was an inability to go back into N while downshifting with
the gearshift paddle on the steering wheel. A better look at the rotation of the wheels
175
VI. Diagnosis
showed us that the gearbox was actually set in R and not in 1 while calibrating the
rotary pot. Why was the screen telling us that we were in first gear then? A simple
explanation for this is that the rotary potentiometer was calibrated to show 1 in a
certain position, hence the calibration was wrong. What happened is that the
manipulation to manually push the gearbox in neutral went wrong. This manual
downshift resulted in placing the gearbox in reverse or R. The calibration of the rotary
potentiometer was then placed into the green area of the gear box check window.
Furthermore, if the gearbox was in 1 the screen showed 2, if it showed 3 it was in
second. A supplementary problem occurred while we were trying to downshift from 2
on the screen to 1, which in reality was a downshift from first gear into reverse The
typical cracking noise of teeth rubbing against each other was well perceived by its
surround.
Finally, the last step was to correctly place the gearbox in first gear and to repeat the
manipulations for the calibration of the rotary pot.
Engine Starting problems
4.1
Problem
After a small crash in Val de Vienne, where front radiator and rear bumper had to be
replaced, the car did not want to start. That was a major problem because the car had to
run the 24h of Spa just a few days later. The mechanic anticipated and filled the car with
some litres fuel, checked all the fuses and relays of the car, checked if there was fuel
arriving to the engine and if the ignition coil ignited the spark plugs. These tests were
positive and everything was working and reported to the engineer.
4.2
4.2.1
Theoretical Background
Dashboard
What could be the problem then? The engineer connected the SD3 to check for errors.
This diagnostic demonstrated some annoying messages. When checking the gearbox for
error, the SD3 told the user that there was no message coming from the dashboard. Idem
176
VI. Diagnosis
for the ECU of the engine (left one). The problem here is that the dashboard (cruscotto
in Italian) controls [3]:
9 CAN messages
9 Brake pad wear / insufficient brake fluid
9 Fuel level signal
9 Engine oil temperature signal
While the other fault conditions and data are managed by the additional data acquisition
box. Thus there are two dashboards: the original one (without visible output) and the
Magneti Marelli data acquisition dashboard/screen. The latter one does not have a CAN
line, but the acquisition box is connected via CAN. Thus when using the SD3, it is not
the Magneti Marelli but the original Ferrari dashboard which is diagnosed. Therefore, to
check all the wiring of the dashboard, one needs to dismount both dashboards (installed
in front of each other).
4.2.2
CAN-line
In the old days of the automobile, everything was analog and wired with 40 m of
electric wires. With the modern technology coming, especially the electronic systems,
more wires were needed and in the year 2000, about 4 km of wires were installed in the
modern cars. Nowadays, the CAN technology (Controller Area Network) came to
diminish the amount of wires, but also the complicated wiring.
Fig. VI.14: (a) Conventional versus (b) CAN wiring [8]
It is clear, out of the two simple diagrams above, that CAN technology uses one serial
bus to transport all kinds of coded digital information. The conventional wiring needs a
wire for each message (often analog).
177
VI. Diagnosis
The physical medium is not really defined by standards and can be made out of
fibreglass or normal wires just as being wireless. A CAN line can often be recognized
via two wires who are twisted in many cases (not on the F430). The two wires are
named CAN-H and CAN-L and it is the measurement of a difference in potentials
between these two wires which is the basis of CAN messages. If noise is present, both
wires will be influenced but not the potential difference
On the figures below, there can be seen that one resistor is installed at each end of the
wires and that the wires are isolated as most as possible from electromagnetic
interferences. These resistors are always 120 and can be found in the car at the end of
each CAN line.
Fig. VI.15: Graphical overview of a CAN-bus implementation [8]
178
VI. Diagnosis
4.3
Solution
Because the mechanic told the engineer and the team manager he checked all the fuses
and relays, the engineer started to check the entire electrical system of the car i.e. the
CAN system. As the SD3 told the engineer that some ECUs did not receive
information of the dashboard (the Ferrari one, not the Magneti Marelli), he had to find
the CAN wires and its resistors. There can be seen on the schematic overview of the CCAN system of the F430 Challenge/GT3 below that the dashboard corresponds to the
NQS (Nodo Quadro Strumenti).
One could also see the alarm message on the Magnetti Marelli screen (which is
connected to the dashboard) and some values that were unreal:
9 Brake Pad Wear 5
9 Oil T = 0C
9 Fuel = 0 (while the mechanic inserted about 25 litres)
This improves the credibility of the SD3 diagnosis telling that the NQS is out of the
CAN system of the car. Because the NQS is sending and receiving can messages with
the Engine ECU, the car does not want to start Additionally, the engineer is unable to
enter into the ECU of the dashboard to check for error or problems, which is not a good
news for the CAN line of the NQS.
179
VI. Diagnosis
Fig. VI.16: C-CAN-bus in the F430 Challenge (modified compared to the street version of the F430)
[5]
The engineer started to check the resistors which are situated in the front compartment
and behind the driver seat (D on the Fig. VI.18). Pulling the connector out of the blue
resistor provided the engineer to measure both resistors. One should know that when
measuring the resistor only, the multimeter has to show 120. Moreover, because the
second 120 resistor is placed at the other end of the CAN line, one should measure
also 120 in the holes of the connector itself (between the CAN-L and CAN-H). These
two tests resulted in correct values. Only one last step has to be done in the
measurement of the CAN line. Both resistors have to be connected and the multimeter
has to measure on the line itself (on a connector, without disconnecting it).
180
VI. Diagnosis
This resulted in a logical 60. Indeed, two resistors in parallel provide a total resistance
in the system equal to:
Rp =
R1 . R2
R1 + R2
Rp = Total resistance for two parallel linked resistors []

R1 = Resistor 1 []
R2 = Resistor 2 []
Thus, if the these measurements are ok and that the fuses are allright, one might think
that the instrument panel (the Ferrari one) is broken. Therefore the Magneti Marelli
screen has to be dismounted and the original dashboard can be measured. A first check
is to see if the connectors are still connected and if the signals arriving are still ok. On
the pin-out below, the engineer can see what type of signal he must have (on connector
number B behind the instrument board). If one measurement pin of the multimeter is
connected to a mass on the car and the other into the pin of the connector, one can
measure the voltage at that certain pin.
Fig. VI.17: Visible original NQS dashboard [3]

Table 5: Pin-out name and functions (starting form top left going around CW from pin to pin)
PIN
Function
Programme instrument
Sensor signal fuel level
Sensor signal engine oil temperature
Sensor signal oil pressure
Mass of engine sensors
Diagnostic line L
Diagnostic line K
C CAN-L
181
VI. Diagnosis
C CAN-H
10
Mass
11
+15
12
+30
Suddenly, something strange happens when measuring pin number 12: There is no
voltage coming trough. This is pretty annoying because it is a +30 source on the
dashboard for the C CAN line and the clock. This means that no current or voltage is
passing, just like if the fuse would be broken. Checking in [5] where the fuses are
situated tells one they are in the front compartment (B). Once this is found, there is a
need to find the correct fuse. The list on the next page tells the engineer that the fuse is a
5A fuse on the position AF9.
(a)
(b)
Fig. VI.18: (a) Installation of fuse and relay boxes in the F430 Challenge [5]; (b) B Section with
covers [5]
182
VI. Diagnosis
Fig. VI.19: List of details about each fuse and relay position in box B
Fig. VI.20: Names and designation of each fuse and relay in the box [5]
183
VI. Diagnosis
Fig. VI.21: Fuse and relay boxes B in the front compartment [5]
Firstly, the cover has to be taken off, and the corresponding position found as told in [5].
What a surprise when pulling the fuse out of the fusebox: The fuse is broken!
Consequently, a new 5A mini fuse had to be placed on the correct place. Finally the car
started back again. One can conclude that one should always check the fuses himself
and never trust someone else. Do not watch at it, but measure the continuity of the wires
inside with a multimetre. This report has been written in a few lines, but the repair of
that problem took more than one day, a horrible day!
184
VI. Diagnosis
[1] Kurt M. Marshek, Jerry F. Cuderman II, Mark J. Johnson, Performance of Antilock Braking System equipped passenger vehicles Part III: Braking as a
function of tire inflation pressure, SAE Technical International, 2002-01-0306,
SAE 2002 World Congress, Detroit, 2002
[2] www.abs-education.org, August 2007
[3] Ferrari Training Centre, F430 Challenge, Maranello, Italy, 2006
[5] Ferrari SpA, F430 Challenge Interactive Technical Documentation (CD Rom),
Maranello, Italy, March 2006
ISBN 1 85960 653 9
[8] Dr. Ir. P. Van De Bossche, Voertuigelektronica: Algemene Begrippen,
Erasmushogeschool Brussel, Belgium, 2005
185
Conclusion
Conclusion
One should be aware of the problems that he can have due to data acquisition systems
and improvements he wants to bring to the car and driver. Therefore some conclusion
can be made after this thesis and working period:
9 The goals of the thesis are attained, but the writer would have liked to add more
information about the data loggers and their connections to the computer, which he
did in his presentation. In addition, an extra chapter about software only would have
been better as well, but one has to limit himself when writing. No appendices are
present, they would be copies of reference which is useless. One can consider that
the first chapters are more or less an appendix.
9 A huge amount of problems occurred with the data loggers and their connection.
This would have been good to insert in the work, but again the writer deliberately
placed boundaries concerning the subjects.
9 The writer managed to improve both drivers and cars during each race weekend. It
was easier to improve the car when a good driver was present in the car. By good
driver there is meant the one which is fast and has got an excellent feeling of the car.
Obviously, he needs to be able to communicate this to his race engineer.
9 The writer learned that it is useless to do major modifications to a car if the driver is
not running in constant lap times. This means he is not able to reproduce the same
braking distances, corner handlings and acceleration. In addition, it can demonstrate
that the driver struggles to have reference points. Therefore, it is better to work on
the driver first (while considering his comments) and then to work on the car if the
problems are still present.
9 During each race weekend, the writer managed to repair each malfunction on the car.
Of course, these were not always fast due to his shortage of experience. Obviously,
one can learn a lot of old mechanics and team managers and has to be open to any
comment, but never accept too quickly. There are a lot of race engineers that think
they know what they do, but often they use the wet finger method.
9 Never trust someone for 100%! If one has got a doubt about the system he is using or
the set up of the car, he better checks it himself (or stand next to the mechanic that
does it).
186
Conclusion
9 Data acquisition is a useful tool, but errors can be made if the engineer does not
correct the traces obtained. This means that when the driver locks his wheels, his car
continues its travel, but the data logger receives a speed equal to zero. Consequently,
because the distance is obtained via the speed, the data analysis software will give a
wrong total distance to the engineer. If suspension channels are present like on the
Formula Renault 2.0, one can easily detect a bump in the road before each braking
session and shift the traces to make them correspond again. The shorter the trace, the
more the car/driver had a tendency to lock-up the wheels.
9 An added value to data acquisition is a professional driver coach. During the thesis,
the writer had the honour to work with a 15 year old driver who had is own driver
coach and former Formula one driver Eric Bernard. Such (ex) professional drivers do
help both the race engineer and the driver a lot. They can correct the driver by just
watching them on track and thus report this to the race/data engineer. Consequently,
the data engineer can react faster and better on the data gathered.
9 The writer wrote some assumptions about car set-up, but these are just assumptions
and are not necessarily the reality. Of course, the writer needs to gather much more
experience to obtain an F1 professionalism in his analysis.
9 Books are not the only things that a race engineer needs to become one. A lot of test
and race sessions will teach much more than books only. Ideally, one should have the
literature with him while analysing data and comments.
187

Data Acquisition To Analyze The Driver and His Race Car

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Data Acquisition To Analyze The Driver and His Race Car

Uploaded by

Copyright:

Available Formats

CRANFIELD UNIVERSITY

DATA ACQUISITION TO ANALYSE THE DRIVER AND HIS RACE CAR

SCHOOL OF APPLIED SCIENCES

This thesis is submitted in partial fulfillment of the requirements

Basic Electronics ..................... Error! Bookmark not defined.

Ohms Law ......................................................................................... 16

Kirchoffs Current law ........................................................................... 16

Hall Effect ......................................................................................... 21

Voltage Divider Circuit........................................................................... 25

Position Sensors ................................................ Error! Bookmark not defined.

Generic Features of wirewound Potentiometers .................................. 28

Hybrid Track Technology .............................................................. 28

The Operating Principle ............................................................... 29

Linear Variable Differential Transformer ........... Error! Bookmark not defined.

Ratiometric Operation ................................................................. 35

Differential Operation ................................................................. 36

Rotary Potentiometers .................................................................... 37

Rotary Variable Displacement Transformers........................................... 38

Ride Height Sensor ......................................................................... 39

Proximity Sensor .................................................................................. 42

Inductive sensors ........................................................................... 42

Capacitive Sensors ......................................................................... 42

Magnetic Sensors ........................................................................... 43

Reed Relay ............................................................................... 43

Wrapped Coil sensor.................................................................... 43

Hall Effect sensor ....................................................................... 43

Optical Types ............................................................................... 44

MSc Motorsport Engineering & Management

TDR or PRT .................................................................................. 46

Temperature sensing ICs ................................................................. 46

Thermocouples .......................................... Error! Bookmark not defined.

Infrared Sensors ............................................................................ 50

Strain and force sensors ......................................................................... 53

Loadcells and force measurement....................................................... 53

Based on load cells ..................................................................... 53

Based on Hall Effect .................................................................... 53

Based on silicon technology ........................................................... 54

Yaw Sensor .................................................................................. 56

Speed sensors ..................................................................................... 59

Wheel speed sensing....................................................................... 60

GPS speed sensing.......................................................................... 61

Gas flow sensors .................................................................................. 62

Sprung flap .................................................................................. 62

Pressure Sensors .................................................................................. 63

Voltage and Current Sensors .................................................................... 64

Lambda sensors ................................................ Error! Bookmark not defined.

Literature References .................................................................................. 68

Basic Race Engineering................................................. 70

Tire model ......................................................................................... 75

Forces acting on tire....................................................................... 75

Toe, camber and caster................................................................... 76

Caster.................................................. Error! Bookmark not defined.

Friction Coefficient ........................................................................ 80

Circle of friction ........................................................................ 81

MSc Motorsport Engineering & Management

Slip Ratio ................................................................................. 82

Slip Angle ................................................. Error! Bookmark not defined.

Neutral Steering, Understeer and Oversteer .......................................... 86

Literature References .................................................................................. 97

Driving Techniques ..................................................... 98

Seating Position ................................................................................... 99

Saloon car ................................................................................... 99

Braking and Accelerating ...................................................................... 103