Chapter 12.

1
12.1
Braking systems
Julian Happian-Smith

The aim of this chapter is to: The chapter commences with a review of the function
of a brake system together with an outline of the prin-
 Aid the designer to understand the legal require-
cipal components and their possible configurations. The
ments of braking systems;
subject of legislation is reviewed and its importance as
 Understand the basic requirements for braking
a tool to aid the designer of a brake system is highlighted.
systems to be successful;
Straight forward kinematic and kinetic analyses are used
 Understand the design process for achieving to address the fundamentals of the braking problem as
an efficient braking system; a precursor to the analysis of brake proportioning, ad-
 Appreciate the material requirements for efficient hesion utilization and other related issues. A case study is
braking systems; built into this section of the chapter that illustrates the
 Understand current developments in braking control application of the theory and so reinforces un-
systems. derstanding. The selection of appropriate materials from
which to manufacture the friction pair is reviewed
and problems linked to thermo-mechanical behaviour
12.1.1 Introduction highlighted. The chapter concludes with a brief summary
of more advanced topics, often linked to modern chassis
The safe and reliable use of a road vehicle necessitates the control, that integrate the braking system with other
continual adjustment of its speed and distance in response chassis systems.
to change in traffic conditions. This requirement is met in
part by the braking system, the design of which plays a key
role in ensuring a particular vehicle is suitable for a given 12.1.1.1 The functions and conditions
application. This is achieved through the design of of use of a brake system
a system that makes as efficient use as possible of the finite
amount of traction available between the tyre and the road In order to understand the behaviour of a braking system
over the entire range of operating conditions that are likely it is useful to define three separate functions that must be
to be encountered by the vehicle during normal operation. fulfilled at all times:
The purpose of this chapter is to introduce the reader (a) The braking system must decelerate a vehicle
to the basic mechanics associated with the deceleration in a controlled and repeatable fashion and when
behaviour of a road vehicle and provide insight to the appropriate cause the vehicle to stop.
many issues that must be addressed when selecting the
brake rotor and friction materials. A complete coverage is (b) The braking system should permit the vehicle
not feasible within the confines of a single chapter and so to maintain a constant speed when travelling
a set of references and additional reading is provided at downhill.
its end that points the interested reader to further (c) The braking system must hold the vehicle stationary
sources of information. when on a flat or on a gradient.

Introduction to Modern Vehicle Design; ISBN: 9780750650441

Copyright Ó 2001 Julian Happian-Smith. All rights of reproduction, in any form, reserved.
 CHAPTER 12.1 Braking systems

When simply stated, as above, the importance of the role (2) Modulation system
played by the brakes/braking system in controlling the This embraces those elements of the brake system
vehicle motion is grossly understated. Consideration of which are used to control the level of braking effort
the diverse conditions under which the brakes must applied to each brake. Included in this system are
operate leads to a better appreciation of their role. These the driver, pressure limiting/modulating values and,
include, but are not limited to, the following: if fitted, anti-lock braking systems (ABSs).
(3) Transmission system
 slippery wet and dry roads.
The components through which energy travels to
 rough or smooth road; the wheel brakes comprise the transmission
 split friction surfaces; system. Brake lines (rigid tubes) and brake hoses
 straight line braking or when braking on a curve; (flexible tubes) are used in hydraulic and air brake
 wet or dry brakes; systems. Mechanical brakes make use of rods,
 new or worn linings; levers, cams and cables to transmit energy. The
 laden or unladen vehicle; parking brake of a car quite often makes use of
 vehicle pulling a trailer or caravan; a mechanical transmission system.
 frequent or infrequent applications of short or (4) Foundation brakes
lengthy duration; These assemblies generate the forces that oppose
 high or low rates of deceleration; the motion of the vehicle and in doing so convert
 skilled or unskilled drivers. the kinetic energy associated with the longitudinal
motion of the vehicle into heat.
Clearly the brakes, together with the steering compo-
nents and tyres, represent the most important accident There are four main stages involved in the design of
avoidance systems present on a motor vehicle which a brake system. The first, and perhaps most fundamental
must reliably operate under various conditions. The stage, is the choice of brake force distribution between
effectiveness of any braking system is, however, limited the axles of the vehicle. This is primarily a function of the
by the amount of traction available at the tyre–road vehicle dimensions and its weight distribution. Next is
interface. the design of the transmission system and this activity
embraces the sizing of the master cylinder together with
12.1.1.2 System design methodology the front and rear wheel cylinders. Additional compo-
nents, such as special valves that modulate the hydraulic
The primary functions of a brake system, listed above, pressure applied to each wheel are physically accounted
must be fulfilled at all times. In the event of a system for at this stage. The foundation brakes form the focus of
failure, the same functions must also be performed albeit the third stage of the process. As well as being able to
with a reduced efficiency. Consequently the braking react the applied loads and torques, the foundation brakes
system of a typical passenger car comprises a service must be endowed with adequate thermal performance,
brake for normal braking, a secondary/emergency brake wear and noise characteristics. The last phase in the
used in the event of a service brake failure and a parking process results in the incorporation of the pedal assembly
brake. Current practice permits service brake compo- and vacuum boost system into the brake system. To
nents to be used in the secondary/parking brake systems. accomplish this design task, the engineer requires access
Irrespective of the detail design considerations all to several fundamental vehicle parameters. These
brake systems divide into the following subsystems: include:

 laden and unladen vehicle mass;

(1) Energy source  static weight distribution when laden and unladen;
This includes all those components which generate,  wheelbase;
store or release energy required by the braking  height of centre of gravity when laden and unladen;
system. In standard passenger cars muscular pedal  maximum vehicle speed;
effort, applied by the driver, in combination with
 tyre and rim size;
a vacuum boost system comprise the energy source.
 vehicle function;
In the event of a boost failure, the driver can still
apply the brakes by muscular effort alone. Alter-  braking standards.
native sources of energy include power braking It is essential to recognize that each of the preceding
systems, surge brakes, drop weight brakes, electric stages are closely linked and that the final design will take
and spring brakes. many iterations to realize. Thus any formal methodology

362
 Braking systems CHAPTER 12.1

must be designed so as not to compromise the overall the secondary circuit. Conversely, if the secondary circuit
system quality that could result from design changes at develops a leak then the secondary piston moves forward
the component level. By way of example, a reduction in until it stops against the end of the master cylinder bore.
package space could lead to smaller diameter wheel This then allows trapped fluid between the two pistons to
brakes having to be fitted to the vehicle. This will change become pressurized and so the primary circuit remains
the brake force distribution unless checked, by say operative.
resizing the wheel cylinders, and in the worst case this
could lead to premature wheel lock and a violation of the
governing legislation. 12.1.1.3.4 Regulating valves
The dynamics of the braking process gives rise to need for
some means of reducing the magnitude of the brake force
12.1.1.3 Brake system components generated at the rear of a vehicle under the action of
and configurations increasing rates of deceleration. This need arises form the
load transfer that takes place from rear to front during
The principal components put together comprise a con- any braking event. This function is realized through the
ventional braking system that is outlined below together incorporation of some form of brake pressure regulating
with possible brake system layouts. The discussion of the valve into the rear brake circuit. The exact nature of the
components begins with the pedal assembly and moves valve depends upon the detail design but they fall into
through the brake system finishing with the foundation three generic types;
or wheel brakes.
 Load sensitive pressure regulating valve: Valves of
this type are fitted to vehicles that experience large
12.1.1.3.1 Pedal assembly
in-service changes in axle load. The valve is anchored
A brake pedal consists of an arm, pad and pivot attach- to the vehicle body and is also connected to the rear
ments. The majority of passenger cars make use of hanging suspension through a mechanical linkage. This
pedals. A linkage is connected to the pedal and this trans- permits the valve to sense the relative displacement
mits both force and movement to the master cylinder. between the body and suspension and adjust the
valve performance to effect control over the rear line
12.1.1.3.2 Brake booster pressure and so enable the rear brakes to compensate
The brake booster serves to amplify the foot pressure for the change in axle load.
generated when the brake pedal is depressed. This has the  Pressure-sensitive pressure regulating valve: Other-
effect of reducing the manual effort required for actua- wise known as a pressure limiter, this type of valve
tion. Boosters are invariably combined with the master isolates the rear brake circuit when the line pressure
cylinder assembly. A vacuum booster employs the nega- exceeds a predetermined value. They find applica-
tive pressure generated in the intake manifold of a spark tion on vehicles that are characterized by a low centre
ignition engine, whereas a hydraulic booster relies upon of gravity and a limited cargo volume.
the existence of a hydraulic energy source and typically  Deceleration-sensitive pressure regulating valve: This
finds application in vehicles powered by diesel engines class of valve finds wide application. The actuation
that generate only a minimal amount of intake vacuum. point is determined by the rate of deceleration of the
vehicle and this is typically of the order of 0.3g.
12.1.1.3.3 Master cylinder A benefit of this type of valve is that it does provide for
a degree of load-sensitive operation as the overall
The master cylinder essentially initiates and controls the
deceleration of the vehicle is the function of the vehicle
process of braking. The governing regulations demand that
weight and the line pressure. They are also sensitive to
passenger vehicles be equipped with two separate braking
braking on a slope. Mathematical models of this class of
circuits and this is satisfied by the so-called tandem master
valve are developed later in the text and their influence
cylinder. A tandem master cylinder has two pistons housed
on the performance of a brake system is demonstrated.
within a single bore. Each section of the unit acts as a single
cylinder and the piston closest to the brake pedal is called
the primary piston whilst the other is called the secondary 12.1.1.3.5 Foundation brakes
piston. Thus, if a leak develops within the primary circuit, Foundation, or wheel brakes, divide into two distinct
the primary piston moves forward until it bottoms against classes, namely disc (axial) and drum (radial) brakes.
the secondary piston. The push rod force is transmitted Modern vehicles are invariably fitted with disc units on
directly to the secondary piston through piston-to-piston the front axle and there is a growing tendency to fit
contact, thus allowing the secondary piston to pressurize similar units to the rear axle. If drum brakes are fitted to

363
 CHAPTER 12.1 Braking systems

1 Brake circuit 1 such documents forms a very important part of the brake
2 Brake circuit 2
II variant X variant
engineer’s database of information.
There are many arguments both for and against legis-
Figure 12.1-1 Common brake system layouts. lation. However, despite some inevitable drawbacks
standards and legislation form a necessary and desirable
part of today’s society and they are here to stay. Poorly
the rear axle then these are typically of the Simplex type written legislative documents may smother initiative and
which employs a leading and trailing shoe configuration restrict technical progress by enforcing unrealistic stan-
to generate the required brake torque. The torque output dards and by failing to recognize the advance of technology.
of this type of drum brake is not sensitive to change in The phrasing of the documents is somewhat complex
vehicle direction. On vehicles fitted entirely with disc which can lead to difficulties in understanding their con-
brakes, then a small drum unit is often employed to act as tent. This is, to a large extent, unavoidable because they are
a parking brake on the rear axle of the vehicle. Issues legal documents and they must attempt to cover all
surrounding the selection of the materials used to man- eventualities, prevent ambiguity and close any loopholes.
ufacture both discs and drums together with their fric- By default, proof of compliance with a national or in-
tion material partners are discussed in more detail later in ternational standard generates an overhead which is
the text. transferred to the consumer as an added cost. It is essential
that this process does not inhibit either new or small
manufacturers from entering the marketplace. Finally,
12.1.1.3.6 Brake system layouts
national legislation can be used as an economic weapon
Legislative requirements demand a dual circuit trans- (termed a technical barrier to trade), particularly by those
mission system to be installed on all road vehicles. Of countries operating an approval system, to protect their
the five possible configurations, two have become stan- industry from worldwide intrusion into the local market.
dard and these are known as the II and X variants shown The job of the importer is made very difficult through the
in Figure 12.1-1. The II design is characterized by sep- use of standards and test procedures that favour the home
arate circuits for both the front and rear axles whilst in industry or by withholding interpretations, by allowing the
the X configuration, each circuit actuates one wheel at test authority the use of subjective judgement and by
the front and the diagonally opposed rear wheel. The II showing a lack of co-operation in test scheduling and in the
design is often found on vehicles that are rear heavy and issue of approval documents.
the X layout has application on vehicles that are front Technical standards and legal requirements need to
heavy. be kept under review in order to force quality upwards.
They must set realistic standards for new and in-service
vehicles that result in real improvements in safety and
12.1.2 Legislation environmental protection. Bearing in mind the necessary
legal constraints, they should be straightforward to
Without exception, motorized road vehicles, whether understand and interpret as well as be universally
cars, buses or lorries, represent a potentially lethal hazard acceptable to encourage free trade and so prevent the
to other road users and pedestrians. Also the rise in production of trade barriers. Standards and legislation
‘green thinking’ during the past decade has led to serious must also be applicable to all types of vehicle and should
consideration of the impact of road vehicles on the en- not preclude innovative design by being so inflexible as
vironment in which we all must live. It is one of the many to limit technical advance; ideally they must actively
responsibilities of a government to ensure that all road encourage the use of new technology. Given time,
vehicles are as safe as possible and that any adverse ef- companies incorporate the formal test procedures in
fects of the vehicle on the environment are minimized. their design programmes and develop close working
This task is achieved through legislation which, in so far relationships with the national approval bodies. In
as the brakes are concerned, primarily sets the minimum principle this leads to improved export performance
standards for the performance of the systems and their since approval obtained in the country of manufacture is
components that combine to arrest the motion of a automatically valid for all others. Also the clearly
vehicle in a controlled manner. A design engineer has to defined ground rules quite often act as an aid to product
take into account many factors associated with the development.
mechanics of braking when designing a new brake With regard to the braking system, legislation first
system. In addition to these elements, conformity to the appeared in the form of the Motor Cars Order of 1904.
legislative requirements of the country or countries in Since this time, the range and complexity of the vehicles
which the vehicle is to operate is absolutely essential. that populate the road network has markedly increased.
Thus a working knowledge of the content and scope of Inevitably this has been accompanied by a similar

364
 Braking systems CHAPTER 12.1

1.0

0.9 k = 0.8 k = z + 0.07

0.85

0.8

0.7

Adhesion utilization (k)

0.6

0.5 z = 0.8
k=z

0.4

0.3
k = z + 0.05

0.2

k = 0.2
0.1

0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
Deceleration (z)

Figure 12.1-2 Adhesion utilization diagram for a category M1 vehicle.

increase in the size and complexity of the regulations that must be satisfied by a road vehicle relates to its use
pertaining to braking. The first major change away from of available tyre–ground adhesion. Manufacturers have to
the self-certification process came when the UK com- submit adhesion utilization curves that demonstrate
menced along the road of Type Approval, favoured by compliance with the limits defined in figures IA and IB of
continental Europe, by the incorporation of the Eco- EEC Directive 71/320 Annex II. The adhesion utilization
nomic Commission for Europe (ECE) Regulations. These diagram that refers to a category M1 vehicle (passenger
voluntary regulations attempt to harmonize vehicle leg- vehicle with seating capacity up to eight including the
islation and they provide for the reciprocal acceptance driver) is shown in Figure 12.1-2, and is derived from
and notification of vehicle systems and component ap- that contained within EEC Council Directive 71/320/
proval. The relevant braking legislation is contained in EEC as last amended by 91/422/EEC.
Regulation 13. The next major change occurred when the The Directive uses the letter k to indicate adhesion
UK became a member of the Common Market in which utilization and z for deceleration and it states that for all
the acceptance of EEC Directives is binding on all categories of vehicle for values of adhesion utilization
member states. Member states are not allowed to impose between 0.2 and 0.8,
more stringent standards than those contained in the
Directives. They are, however, free to demand additional z  0:1 þ 0:85ðk  0:2Þ
standards with regard to matters not covered by EEC
legislation. EEC Directives differ from the corresponding For category M1 vehicles, the adhesion utilization
ECE Regulations in one major aspect: Approvals issued curve of the front axle must be greater than that of the
by one member must be accepted by all others. The rear for all load cases and values of deceleration between
objective of either is, however, the same, namely the 0.15g and 0.8g. Between deceleration levels of 0.3g and
harmonization of differing technical requirements. In 0.45g, an inversion of the adhesion utilization curves is
1978, it became mandatory in the UK for all new cars to allowed provided the rear axle adhesion curve does not
be type approved to the EEC Braking Directives. A exceed the line defined by k ¼ z by more than 0.05. The
similar exemption was granted in 1980 for vehicles ap- above provisions are applicable within the area defined by
proved to ECE Regulation 13. The current EEC Di- the lines k ¼ 0.8 and z ¼ 0.8.
rective on braking is Directive 71/320/EEC as last Compliance of the braking system to the constraints
amended by Directive 91/422/EEC. defined in Figure 12.1-2 ensures that the rear wheels do
Of the many requirements laid down in the EEC not lock in preference to the front wheels and that the
regulations, perhaps the single most important aspect proportion of braking effort exerted at the front of

365
 CHAPTER 12.1 Braking systems

the vehicle is limited so that the braking system does not distance travelled by the vehicle moving with constant
become too inefficient. A detailed interpretation of this velocity U is:
requirement is outlined later in the text.
S1 ¼ Ut1 (12.1.1)
12.1.3 The fundamentals of braking In part 2 the vehicle is decelerated at a constant rate
until such time as the vehicle comes to rest. The distance
12.1.3.1 Kinematics of a braking vehicle travelled is:

12.1.3.1.1 Kinematic analysis of a braking Ut2 U2

vehicle S2 ¼ ¼ (12.1.2)
2 2a
The distance travelled by a vehicle when braking, either
during a stop or snub (when the final velocity is non-zero) Thus, the total stopping distance is simply
is a basic measure of the effectiveness of a brake system.
Before addressing issues coupled to the forces which act U2
on a vehicle during a braking manoeuvre it is worthwhile Si ¼ S1 þ S2 ¼ Ut1 þ (12.1.3)
2a
to first consider the kinematic behaviour of the vehicle. A
straightforward kinematic analysis assuming straight line The preceding analysis assumes the vehicle de-
(one-dimensional) motion and constant deceleration celeration is achieved instantaneously and is sustained
provides a ready indication of stopping distance. Pre- for the duration of the stop. No account is taken of
dictions of stopping distance made by this analysis find driver reaction time, initial system response time, de-
use in accident reconstruction. celeration rise time, change in deceleration during the
With reference to Figure 12.1-3, the total distance period of actual braking and, if applicable, release time.
travelled is made up of two parts. In part 1, the total These factors are now defined with reference to

Deceleration time history

2.5
Deceleration (m s–2)

2.0
1.5
1.0
0.5
0
0 5 10 15 20
Time (s)

Velocity time history

35
30
Velocity (m s–1)

25
20
15
10
5
0
0 5 10 15 20
Time (s)

Displacement time history

400
Displacement (m)

300

200

100

0
0 5 10 15 20
Time (s)

Figure 12.1-3 Kinematics of a simple stop.

366
 Braking systems CHAPTER 12.1

0.5

0.4

Deceleration g
0.3

0.2

0.1

0
2 4 6 8 10 12 14 16
Time (s)

t0 t1 t2 t3 t4 t5

Figure 12.1-4 Typical measured deceleration time history.

Figure 12.1-4, which illustrates the characteristics of and gravity, are not taken into account in the following
a typical measured deceleration time history. analysis, which is again built around one-dimensional
particle kinematics. With reference to Figure 12.1-5, the
12.1.3.1.2 Driver reaction time following expressions for the distance travelled during
each stage of the stop can be derived. Note that as
The time taken by the driver to respond to the danger,
a stop is the subject of the analysis, the effect of release
formulate an avoidance strategy and physically move
time ðt5  t4 Þ is ignored.
his/her right foot from the accelerator to the brake,
Distance travelled during reaction time S1:
ðt1  t0 Þ.

12.1.3.1.3 Initial system response time S1 ¼ Uðt1  t0 Þ (12.1.4)

The time taken from the point at which the brake pedal
Distance travelled during the initial system response
begins to move to the time at which a braking force is
time S2:
generated at the tyre road interface, ðt2  t1 Þ.

12.1.3.1.4 Deceleration rise time S2 ¼ Uðt2  t1 Þ (12.1.5)

The elapsed time for the deceleration to reach the value,
The distance travelled during the deceleration rise
determined by the driver, for the stop in question,
time S3, assuming this to be linear, can be shown to be:
ðt3  t2 Þ.

12.1.3.1.5 Braking time af ðt3  t2 Þ2

S3 ¼ Uðt3  t2 Þ  (12.1.6)
The time taken from the point at which fully developed 6
braking is reached to the time at which the vehicle stops
or brake release begins, ðt4  t3 Þ. where af is the value of the mean fully developed de-
celeration. The distance travelled during the course of
12.1.3.1.6 Release time braking, S4, under the deceleration of af, which is as-
sumed to remain constant, can be expressed as:
The elapse time between the point at which brake release
starts to the time at which brake force generation ceases, " #
ðt5  t4 Þ. 1 2 a2f ðt3  t2 Þ2
S4 ¼ U þ  Uaf ðt3  t2 Þ
The stopping time and distance are measured from 2af 4
the time t0 to the time t4 or t5, if appropriate, whilst the (12.1.7)
braking time and distance is measured from time t1. In-
clusion of these factors modifies the simple kinematic Thus, the total stopping distance, Ss, is simply
analysis with the result that the predicted stopping times
and distances are increased. 4
X
Additional deceleration forces, such as those arising Ss ¼ Si (12.1.8)
from engine drag, aerodynamic drag, rolling resistance i¼1

367
 CHAPTER 12.1 Braking systems

Deceleration (m s–2)
af

t0 t1 t2 t3 t4 Time (s)

u
Velocity (m s–1)

t0 t1 t2 t3 t4 Time (s)
Displacement (m)

Braking distance S b

Stopping distance Ss

t0 t1 t2 t3 t4 Time (s)
Braking time

Stopping time

Figure 12.1-5 Four-stage stop simulation.

and the braking distance, Sb, is: where

4
X M ¼ Vehicle mass Tr ¼ Rear axle braking
Sb ¼ Si (12.1.9) force
P ¼ Vehicle weight
i¼2
g ¼ Acceleration due to Rf ¼ Front axle load
gravity Rr ¼ Rear axle load
12.1.3.2 Kinetics of a braking vehicle Tf ¼ Front axle braking q ¼ Angle of incline
force
A general equation for braking performance can be easily
derived through application of Newton’s second law to Note that the front and rear braking force terms, Tf
a simplified free-body diagram of a vehicle in the di- and Tr, represent the sum of all the effects that combine
rection of its travel (Figure 12.1-6). Assuming x is posi- to generate the forces which act between the front and
tive in the direction of travel, then: rear axles and ground. These include the torque gener-
X ated by the brakes together with rolling resistance ef-
Fx ¼ M€x (12.1.10) fects, bearing friction and drive train drag.
If an additional variable for linear deceleration, d, is
and so defined such that
Tf  Tr  D  P sin q ¼ M€x (12.1.11) d ¼ €x (12.1.12)

368
 Braking systems CHAPTER 12.1

x which leads to

T Tx v2  v2f
ðxf  x0 Þ ¼ ¼ 0 (12.1.18)
M M 2

where x is the distance travelled during the brake

Tf
application.
Rf When considering a stop, the final velocity vf is zero
P
and so the stopping distance x is, from equation 12.1.18,
Tr given by:
Rr θ

Figure 12.1-6 Free body diagram of a braking vehicle. Mv20

x ¼ (12.1.19)
2T

then equation 12.1.11 becomes: and the time, tb, taken to stop the vehicle is, from
equation 12.1.16,
Md ¼ Tf þ Tr þ D þ P sin q ¼ T (12.1.13) Mv0 v
tb ¼ ¼ 0 (12.1.20)
T d
in which T is the sum of all those forces that contribute to
the overall braking effort. Thus, from equation 12.1.19 the distance required to
By considering the case of constant deceleration, stop the vehicle is proportional to the square of the initial
straightforward and fundamental relationships can be velocity and, from equation 12.1.20, the time taken to
derived that yield an appreciation of the physics which stop the vehicle is proportional to the initial velocity.
governs all braking events. To achieve maximum deceleration and hence mini-
From equation 12.1.13, the linear deceleration of mum stopping distance on a given road surface, each axle
a vehicle can be expressed as: must simultaneously be on the verge of lock. If this is
so then, if z is the deceleration as a proportion of
T dv g; z ¼ dg, and the brake force T is equal to the product of
d ¼ ¼  (12.1.14) the vehicle weight and the coefficient of tyre–ground
M dt
adhesion, Pm, then from equation 12.1.13,
in which v is the forward velocity of the vehicle. Since Pz ¼ T ¼ Pm (12.1.21)
the deceleration is assumed constant, then the total
brake force is also constant and so equation 12.1.14 can from which it can be deduced that
be integrated with respect to time between the limits of
the initial velocity, v0, to the final velocity, vf, to z ¼ m (12.1.22)
determine the duration of the braking event, tb.
On rearranging, equation 12.1.14 becomes: This represents a limiting case in which it is clear that
the maximum deceleration cannot exceed the value of
ð vf ð tb tyre–ground adhesion. A deceleration in excess of 1g
T
dv ¼  dt (12.1.15) therefore implies that the tyre–ground adhesion has
v0 M 0
a value greater than unity and this is quite realizable with
certain types of tyre compound.
which leads to As already indicated, the primary source of
retardation force arises from the foundation brake. Sec-
T
v0  vf ¼ t (12.1.16) ondary forces which contribute to the overall braking
Mb performance include:
The fact that velocity and displacement are related by  Rolling resistance, expressed by a coefficient of
v ¼ dx=dt permits an expression for stopping distance to rolling resistance. The total rolling resistance is
be derived from equation 12.1.14 through substitution independent of the load distribution between axles
for dt and integration between v0 and vf as before. On and the force is typically equivalent to a nominal
rearranging, equation 12.1.14 becomes: 0.01g deceleration.
ð ð vf  Aerodynamic drag which depends on dynamic pressure
T xf and is proportional to the square of the vehicle speed. It
dx ¼  vdv (12.1.17)
M x0 v0 is negligible at low speeds, however, aerodynamic drag

369
 CHAPTER 12.1 Braking systems

may account for a force equivalent to 0.03g when trav- of slip taking place at the tyre–road interface and so brake
elling at high speed. force and slip co-exist. The longitudinal slip of the tyre is
 Gradient makes either a positive (uphill) or negative defined as a ratio:
(downhill) contribution to the total braking force
experienced by a vehicle. This force is simply the slip velocity in contact patch
slip ¼
component of the total vehicle weight acting in the forward velocity
plane of the road. v  ur
¼ (12.1.24)
 Drivetrain drag may either help or hinder the braking v
performance of a vehicle. If the vehicle is deceler-
ating faster than the components of the drivetrain where v is the forward velocity of the vehicle, u is the
would slow down under their own friction then angular velocity of the wheel and r the wheel radius.
a proportion of the brake torque generated by the Useful information can be obtained by plotting brake
wheel brakes must be used to decelerate the rotating force coefficient against slip (Figure 12.1-7). During
elements within the drivetrain. Thus, the inertia of straight line braking, no lateral forces are generated which
the elements of the drivetrain effectively adds to the means that all of the force that is potentially available
mass of the vehicle and so should be considered in within the tyre–ground contact patch can be used to de-
any rigorous brake design programme. Conversely, celerate the vehicle. The uppermost characteristic illus-
the drivetrain drag may be sufficient to decelerate trates the brake coefficient derived from both the adhesive
the rotating elements and so contribute to the overall and hysteretic mechanisms and it increases linearly with
vehicle braking effort and this is often the case during increase in slip up to around 20% slip. On dry roads, the
braking manoeuvres involving a low rate of adhesion component dominates the production of friction
deceleration. coupling. The peak coefficient, denoted by mp, defines the
maximum braking force that can be obtained for a given
tyre–road friction pair. At higher values of slip this co-
12.1.3.3 Tyre–road friction efficient decreases to its lowest value of ms at 100% slip,
The brake force, Fb, which acts at the interface between which represents the full lock condition. The maximum
a single wheel and the road is related to the brake torque, brake force, corresponding to mp, is a theoretical maximum
Tb, by the relationship: as the system becomes unstable at this point. Once a wheel
is decelerated to the point at which mp is achieved, any
Tb disturbance about this point results in an excess of brake
Fb ¼ (12.1.23)
r torque that causes the wheel to decelerate further. This
leads to an increase in slip and this in turn reduces the
where r is the radius of the wheel. The brake force on brake force leading to a rapid deceleration to the full lock
a vehicle can be predicted using equation 12.1.23 as long
as all the wheels are rolling. The brake force Fb cannot
increase without bound as it is limited by the extent of Stable Unstable
the friction coupling between the tyre and the road.
1.0
The friction coupling that gives rise to the brake μp
force characteristic reflects the combination of tyre and
μb
road surface materials together with the condition of the
0.8
surface. The best conditions occur on dry, clean road
Lateral force coefficient μl
Brake force coefficient μb

surfaces on which the brake force coefficient, defined as

the ratio of brake force to vertical load, mb, can reach 0.6
values between 0.8 and unity. Conversely, icy surfaces
reflect the poorest conditions and on ice the brake force
coefficient can lie between 0.05 and 0.1. On wet surfaces 0.4
or on roads contaminated by dirt, the brake force μl at a constant value
coefficient typically spans the range 0.2–0.65. of side slip angle
Hysteresis and adhesion are the two mechanisms re- 0.2
sponsible for friction coupling. Surface adhesion comes
about from the intermolecular bonds which exist be-
tween the rubber and the aggregate in the road surface. 0
0 20 40 60 80 100
Hysteresis, on the other hand, represents an energy loss Brake slip (%)
in the rubber as it deforms when sliding over the aggre-
gate. Each of these mechanisms relies on a small amount Figure 12.1-7 Brake force against wheel slip.

370
 Braking systems CHAPTER 12.1

condition. It is worthwhile to note that ABSs make use of

Table 12.1-1 Prototype vehicle parameters
this phenomenon.
The negotiation of a bend requires a vehicle to develop Parameter Symbol Units Value
a lateral force in the tyre–ground plane through the de-
formation of the tyre carcass brought about by a slip angle. Mass M kg 980
Wheel base I m 2.45
Lateral forces, characterized by the coefficient m1, being Height of centre of gravity h m 0.47
a function of slip angle, and longitudinal forces, charac- above ground
terized by mb thus co-exist and compete for the finite Static axle loads (% of total) Front Ff – 65
amount of force that is available within the contact patch. Rear Fr – 35
A typical lateral force coefficient is shown on Figure 12.1-7 Fixed brake ratio Front xf – 75
xf
for a given value of slip angle. It has a maximum value when R ¼ ð%Þ Rear xr – 25
xr
the brake slip is zero and falls with increase in brake slip.
The minimum value occurs when the wheel has locked and
is unable to generate further lateral force. axle are equal and so the following analyses treat axle
loading rather than the individual wheels.
The vertical load due to the vehicle mass is simply:
12.1.4 Brake proportioning and
P ¼ Mg (12.1.25)
adhesion utilization
where g is the acceleration due to gravity.
The vertical loads carried by the front and rear wheels of Taking moments about the rear tyre–road contact
a rigid, two axle vehicle are not, in general, equal. In gives:
order to efficiently utilize the available tyre–road adhe-
sion the braking effort must be apportioned between the Pb
front and rear of the vehicle in an intelligent and con- Ff ¼ (12.1.26)
l
trolled fashion. Failure to do so could result in any of the
following: and moments about the front tyre–road contact provides
 A vehicle being unable to generate the necessary
a value of the vertical load acting at the rear axle given by:
deceleration for a given pedal pressure. Pa
 Front axle lock, in which the vehicle remains stable Fr ¼ (12.1.27)
l
yet suffers from a loss of steering control.
 Rear axle lock that causes the vehicle to become The influence of the fore–aft location of the centre of
unstable. gravity on the vertical wheel loads, due to change in
loading conditions is readily apparent from equations
The terms front wheel lock and rear wheel lock can
12.1.26 and 12.1.27.
be alternatively labelled front axle over brakes and rear
The maximum payload of a passenger car forms only
axle over brakes, respectively. In this section factors
a fraction of the unladen vehicle mass and the inherent
influencing the fore–aft axle loads are identified and their
space restrictions limit the extent to which the centre of
effect on braking examined, concentrating on vehicles
gravity can move. For a light rigid truck, such as a transit
that have a fixed brake ratio. Devices used to modify the
type van or pickup, there exists considerable scope for
braking ratio are discussed in the latter part of this section
and their effect on braking performance evaluated.
The theory developed in Sections 12.1.4.2–12.1.4.4
and 12.1.4.8 is applied to the analysis of the braking re-
quirements of a fictitious two axle road vehicle. This
vehicle is assumed to be unladen and is described by the
parameters given in Table 12.1-1.

12.1.4.1 Static analysis

Ff Fr
P
A simple representation of a two axle road vehicle is
shown in Figure 12.1-8, which has a mass, M, concentrated a b
at the centre of gravity. The centre of gravity is assumed
l
to lie on the longitudinal centreline of the vehicle and
the road surface is flat with no camber. As a consequence
of this, the loads on the two wheels mounted on each Figure 12.1-8 Static axle loads.

371
 CHAPTER 12.1 Braking systems

varying the payload and so the change in mass from unladen y

to laden can be severe. The distribution and size of the
payload, which also governs the location of the centre of θ
gravity, may change on a daily basis or throughout a delivery
x
cycle in which the payload is discharged in stages.

12.1.4.2 Braking with a constant D

brake ratio
Mg
The object of this analysis is to show how braking affects
the vertical loads carried by the front and rear axles. This
Rf
in turn leads to a way of determining the maximum de- Tfr Tf Rr Trr Tr
celeration attainable by a vehicle under specified condi- a b
tions that does not result in axle lock.
If the front and rear axles are to be on the point of l
locking, then the braking forces Tf and Tr acting at each
axle must be in proportion to the vertical loads being Figure 12.1-9 Free body diagram of the decelerating vehicle.
carried, Rf and Rr. The magnitude of the braking force
generated by each axle, up to the point at which it locks,
is a function of the design of the braking system. If the aerodynamic drag and rolling resistance forces
Changes in load transfer between the front the rear are assumed to be negligible, then equation 12.1.29
axles occur during braking and so a variable brake effort reduces to:
ratio is required to provide ideal braking. The situation is
Md ¼ D þ Tf þ Tr þ Tfr þ Trr (12.1.29)
in reality complicated by the following:
Md ¼ Tf þ Tr (12.1.30)
 Change in vehicle weight.
 Change in weight distribution. By defining z to be the vehicle deceleration as a pro-
 The effect of gradients (positive and negative). portion of g:
 Cornering, in which some proportion of the total d
force present at the tyre–ground interface is used z ¼ (12.1.31)
g
to generate lateral forces.
 Varying road surfaces and weather conditions. then equation 12.1.30 takes the form:
 Split friction surfaces where the coefficient of adhe-
Mgz ¼ Tf þ Tr ¼ Pz (12.1.32)
sion changes from port to starboard.
In the vertical direction
Consider the case for a vehicle with a fixed brake ratio,
X
R ¼ xxfr , on a flat road that has a uniform coefficient of M€y ¼ Fy ¼ Rr þ Rf  Mg ¼ 0 (12.1.33)
tyre–ground adhesion m. In the analysis that follows, the
governing equations of motion for the decelerating ve- as y = 0 ms2, whilst in the q direction, taking moments
hicle are derived by the direct application of Newton’s about the centre of gravity of the vehicle leads to:
second law to the free body diagram of the vehicle rather
than through D’Alembert’s method which is adopted in I€q
X
the EEC Council Directive 71/320/EEC. This permits ¼ Mcg ¼ Rf a  Rr b  Tf h  Tr h ¼ 0
easy extension of the model to embrace additional de-
grees of freedom or to take account of the presence of (12.1.34)
a trailer. The free body diagram of the decelerating
as €q ¼ 0 rad s2.
vehicle is shown in Figure 12.1-9.
Manipulation of equations 12.1.33 and 12.1.34 results
In the x direction
in the following expressions for the front and rear
X dynamic axle loads:
M€x ¼ Fx
(12.1.28) Mgb h
¼ D  Tf  Tr  Tfr  Trr Rf ¼ þ ðTf þ Tr Þ (12.1.35)
l l
which through combination with equation 12.1.12 Mga h
Rr ¼  ðTf þ Tr Þ (12.1.36)
becomes: l l

372
 Braking systems CHAPTER 12.1

which can be combined with equation 12.1.32 and the If, however, the ratio has been set so that the rear
static axle loads, equations 12.1.26 and 12.1.27, to give axle locks in preference to the front, then the brake
force generated at the rear axle when about to lock is
Pzh given by:
Rf ¼ Ff þ (12.1.37)
l
Pzh Tr ¼ mRr
Rr ¼ Fr  (12.1.38)  
l Pzh (12.1.45)
¼ m Fr 
The above are in accord with those given in the EEC l
Directive and they show that a change in axle load in
In this case, the brake force that is generated at the
favour of the front axle occurs during a braking ma-
front axle is not necessarily the limiting value and its
noeuvre. In order for each axle to be simultaneously on
magnitude is found from the brake ratio as:
the verge of locking, the brake force generated at each
axle must be in direct proportion to the vertical axle load.
x
This means that to fully utilize the available tyre–ground Tf ¼ T r f
adhesion, the braking system must support an infinitely xr  (12.1.46)
Pzh xf
variable brake ratio. ¼ m Fr 
l xr
Consider first the case of a vehicle in which the brake
ratio is fixed. If the ratio has been set so that the front
which leads to a total brake force of:
axle locks in preference to the rear, then the brake force
generated at the front axle when about to lock is given by:
T ¼ Pz ¼ Tf þ Tr
   
Tf ¼ mRf Pzh xf Pzh
  ¼ m Fr  þ m Fr 
Pzh (12.1.39) l xr l
¼ m Ff þ  
l Pzh 1
¼ m Fr  (12.1.47)
l xr
During the same braking event, the rear axle is also gen-
erating a brake force that has not exceeded its limiting value and this can be solved for the deceleration as a proportion
and this is found by considering the vehicle brake ratio: of g to be:

xf T lmFr
R ¼ ¼ f (12.1.40) z ¼ (12.1.48)
xr Tr Pðlxr þ mhÞ
from which Direct solution of equations 12.1.44 and 12.1.48 for z
  is straightforward, however, greater insight to the me-
xr Pzh xr chanics of the braking process can be gained through the
Tr ¼ T f ¼ m Ff þ (12.1.41)
xf l xf following graphical solution that deals with each axle in
turn.
leading to a total brake force of: The adhesion force acting between the front tyres
and ground depends upon the ratio of the tangential
T ¼ Pz ¼ Tf þ Tr forces at the front and rear wheels due to the brake
   
Pzh Pzh xr torques. It is therefore linked to the fixed brake ratio
¼ m Ff þ þ m Ff þ (12.1.42) and so the front adhesion force, Tf, as a proportion of
l l xf
the total is given by:
which reduces to:
  Tf ¼ x f T
Pzh 1 (12.1.49)
T ¼ Pz ¼ m Ff þ (12.1.43) ¼ xf Pz
l xf

This equation can be rearranged to yield the maximum which when normalized to the vehicle weight, P,
value of deceleration as a proportion of g as: becomes:

mFf Tf
z ¼ (12.1 44) ¼ xf z (12.1.50)
Pðlxf  mhÞ P

373
 CHAPTER 12.1 Braking systems

Equation 12.1.50 is shown in Figure 12.1-10 labelled as adhesion coefficient of unity. The lower value of de-
Tf =P. The total available braking force at the front of the celeration, b, is that value of deceleration that would first
vehicle is, from equation 12.1.43, give rise to wheel lock for the given value of tyre–ground
  adhesion. In this instance, the rear axle will lock first and
Pzh the tyres will be unable to generate the braking force
Txf ¼ m Ff þ (12.1.51)
l required by the rear brakes at higher levels of de-
celeration. Direct solution of equation 12.1.48 leads to
which when normalized to the vehicle weight becomes: a limiting value of deceleration of z ¼ 0:79 g.
 
Txf m Pzh 12.1.4.3 Braking efficiency
¼ F þ (12.1.52)
P P f l
The efficiency with which a brake system uses the
available tyre–ground adhesion, h, can be conveniently
This represents the maximum braking force, defined as the ratio of the deceleration, z, to the tyre–
expressed as a proportion of the total vehicle weight, that ground adhesion coefficient, m, that is:
could be sustained between the front tyres and road
surface for a given set of vehicle parameters. It is shown z
h ¼ (12.1.55)
in Figure 12.1-10 as the line Txf=P. m
Application of the same procedure to the rear axle of
There are two expressions for h; one for the case in
the vehicle results in the following normalized expression
which the front axle is about to lock and the other for the
for the rear adhesion force Tr:
case in which the rear axle is about to lock. To determine
Tr which is applicable recall that:
¼ Xr z (12.1.53)
P
h  1:0 (12.1.56)
and this is labelled Tr/P in Figure 12.1-10. Similarly, the
total available braking force at the rear axle, normalized For the case of front axle lock, h can be written using
to the vehicle weight, is, from equation 12.1.47: equation 12.1.44 as:
  z
Txr m Pzh h ¼
¼ Fr (12.1.54) m
P P l lmFf
Pðlxf  mhÞ
and this is the line Txr/P in Figure 12.1-10. ¼ (12.1.57)
m
The point of intersection of lines Tf/P and Txf/P,
labelled a, represents the solution to equation 12.1.44 F
¼  f 
and the point of intersection of lines Tr/P and Txr/P, mh
P xf 
labelled b, is the solution to equation 12.1.48. The data l
used to generate Figure 12.1-10 are that of the prototype
vehicle defined in Section 12.1.4.4. The vehicle is as- For the rear axle lock case, application of equation
sumed to be braking on a road that has a tyre–ground 12.1.48 results in the second expression for efficiency

1.00
Brake force/vehicle weight (kN/kg)

0.90
0.80
Tf /P
0.70
0.60 Txf /P
0.50 Tr /P
0.40 Txf /P
0.30
0.20
0.10
0.00
0 0.2 0.4 0.6 0.8 1.0 1.2
Deceleration (g)

Figure 12.1-10 Normalized brake force against deceleration.

374
 Braking systems CHAPTER 12.1

z 1.2 Optimum
h ¼
m Front axle
lmFr Rear axle
1.0
Pðlxr þ mhÞ
¼ (12.1.58)
m
Fr 0.8 Front
¼  

Deceleration g
axle
mh lock
P xr þ
l 0.6
a

A measure of the efficiency with which a vehicle, 0.4

having a particular brake design with fixed brake ratio, Rear
axle
performs over a variety of road surfaces can be shown lock
0.2
by a graph of efficiency, h, against tyre–ground
adhesion coefficient, m, Figure 12.1-11. This has been
generated using the prototype vehicle data. Both axles 0
0 0.2 0.4 0.6 0.8 1.0
are on the verge of lock and the system is 100% effi- Tyre–ground adhesion coefficient
cient when the vehicle is braked on a road surface that
has a tyre–ground adhesion coefficient of 0.52 in- Figure 12.1-11 Deceleration as a function of tyre–ground
adhesion.
dicated by the point a. On road surfaces below this
value of adhesion, the vehicle is limited by front axle
lock, equation 12.1.57, and the efficiency falls to curves that lie below the optimum line and it is clear that
a minimum of 87%, whilst on road surfaces with an the vehicle is governed by rear axle lock on road surfaces
adhesion coefficient greater than 0.52 the vehicle is with a high tyre–ground adhesion coefficient and that the
limited by rear axle lock, equation 12.1.58, and this system is least efficient in this area.
falls to a minimum of 79%. Data presented above the
line defining 100% efficiency has no physical meaning 12.1.4.4 Adhesion utilization
and can be ignored.
An alternative means of presenting efficiency data can Adhesion utilization, f, is the theoretical coefficient of
be achieved by plotting deceleration, z, against tyre– adhesion that would be required to act at the tyre–road
ground adhesion, m, for the cases of front and rear axle interface of a given axle for a particular value of de-
lock defined by equations 12.1.44 and 12.1.48, re- celeration. It is therefore the minimum value of tyre–
spectively. This method of presentation has the advan- ground adhesion required to sustain a given de-
tage that the brake engineer can obtain a comparison of celeration and is defined as the ratio of the braking
possible deceleration levels attainable on different road force to the vertical axle load during braking.
surfaces along with a measure of the system efficiency. In
this case, the efficiency is the gradient of the curve drawn 1.20 100%
on the deceleration–adhesion space. A line with unit Front axle
1.15
gradient represents optimum performance. Rear axle
Recall that for front wheel lock, 1.10

1.05
lmFf
z ¼ (12.1.59) a
Braking efficiency

Pðlxf  mhÞ 1.00

0.95
and for rear wheel lock,
0.90

lmFr 0.85
z ¼ (12.1.60)
Pðlxr þ mhÞ Front Rear
0.80 axle axle
The two curves that define the limiting deceleration of lock lock
0.75
the front and rear axles, derived from equations 12.1.59
and 12.1.60, respectively, are shown in Figure 12.1-12 for 0.70
0 0.2 0.4 0.6 0.8 1.0
the prototype vehicle. They intersect the optimum line Tyre–ground adhesion coefficient
at the point a which indicates 100% efficiency. Mean-
ingful information is taken from those portions of the Figure 12.1-12 Efficiency as a function of tyre–ground adhesion.

375
 CHAPTER 12.1 Braking systems

1.2 Optimum k Front axle For the front of the vehicle the adhesion utilization
k limit Rear axle is defined by:
1.0

Tf
Adhesion utilization

0.8 Front ff ¼ (12.1.61)

axle Rf
lock
0.6
a The vertical axle load is defined by equation 12.1.37
0.4 and the front axle brake force, expressed as a proportion
Rear
axle
of the total is xfPz, leads to:
0.2 lock
xf Pz
ff ¼ (12.1.62)
0 Pzh
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 Ff þ
l
Deceleration (g)
Similarly, for the rear of the vehicle:
Figure 12.1-13a Adhesion utilization, datum prototype vehicle.
Tr
fr ¼ (12.1.63)
Rr

The vertical axle load is defined by equation 12.1.38

1.2
and the rear axle brake force, expressed as a proportion
1.0 of the total is xrPz, leads to:
Adhesion utilization

0.8 xr Pz
fr ¼ (12.1.64)
Pzh
0.6 Optimum k Fr 
l
k limit
0.4
Front axle
Using the data that describe the prototype vehicle
0.2 leads to Figure 12.1-13a. The optimum line has unit gra-
Rear axle dient and defines the ideal adhesion utilization charac-
0 teristic in which the brake system remains 100% efficient
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2
Deceleration g
over all possible values of deceleration. The upper limit on
allowable adhesion utilization, defined in the EEC Braking
Figure 12.1-13b Adhesion utilization, modified prototype vehicle. Directive, Section 12.1.2, is shown for reference purposes.
The remaining two lines define the axle adhesion

1.20

1.15

1.10

1.05
Braking efficiency

1.00

0.95

0.90

0.85 100%

0.80 Front axle

Rear axle
0.75

0.70
0 0.2 0.4 0.6 0.8 1.0
Tyre–ground adhesion coefficient

Figure 12.1-13c Brake system efficiency of modified prototype vehicle.

376
 Braking systems CHAPTER 12.1

characteristics for the vehicle. The point labelled a, at the brake force coefficient increases in a roughly linear
which the curves cross, intersect the optimum line of ad- fashion to its maximum value, mp, at 20% longitudinal
hesion utilization indicating that at this value of de- slip. Further increase, due to increase in applied brake
celeration both axles are on the verge of lock. The axle torque, causes the wheel to decelerate rapidly to a con-
having the highest adhesion utilization coefficient for dition of full lock and the brake force coefficient takes
a given value of deceleration is that which limits the a value of ms at 100% longitudinal slip. The ratio of mp/ms
braking performance of the vehicle and, in this case, depends upon the nature of the road surface in question
braking is limited by front axle lock up to a deceleration of and it takes its highest value under wet or icy conditions.
0.52g. Thereafter, braking is limited by rear axle lock. It is This leads to a possible scenario in which a vehicle is
also possible to find from this diagram the maximum de- capable of generating its maximum braking potential
celeration for a given coefficient of adhesion utilization. when one axle is locked and the second is on the verge of
Comparison of the adhesion utilization diagram lock. This contrasts with the generally accepted idea that
derived for the prototype vehicle with the legislative maximum deceleration occurs when the first axle is
requirements outlined in Section 12.1.2 shows that the about to lock and is dependent upon the vehicle weight
vehicle brake system does not meet the minimum stan- distribution and the fixed braking ratio.
dard, as the front axle adhesion curve does not lie above If the front axle is locked and the rear axle is about to
that of the rear axle for all values of deceleration between lock then the total brake force is given by:
0.15g and 0.8g. This can be remedied by changing the
   
fixed brake ratio in favour of the rear axle and this causes Pzh Pzh
the point a to move up the optimum adhesion line. The Pz ¼ ms Ff þ þ m p Fr  (12.1.65)
l l
limiting deceleration is set at 0.8g which leads to a new
fixed brake ratio of xxfr ¼ 0:803
0:197. This in turn results in the Similarly, if the rear axle is locked and the front axle is
modified adhesion diagram shown in Figure 12.1-13b, on the verge of lock, then the total brake force is:
which this satisfies the legislative requirements. The
   
modified vehicle is governed by front axle lock up to Pzh Pzh
a deceleration level of 0.8g, achieved at the expense of Pz ¼ mp Ff þ þ m s Fr  (12.1.66)
l l
the overall brake system efficiency (Figure 12.1-13c).

12.1.4.5 Wheel locking

12.1.4.6 Effect of axle lock on
The role of tyre–ground friction and the dependency of vehicle stability
the brake force coefficient on the degree of longitudinal
slip has been outlined in Section 12.1.3.3. From 0% to When an axle locks, there is reduced friction in both the
approximately 20% longitudinal slip, the magnitude of longitudinal and lateral directions and so the ability of the

Fx FR
α
a Tf /2 Tf /2

Fy
x
l

b
y

Sr /2 Sr /2

Tr /2 Tr /2

Figure 12.1-14 Front axle lock.

377
 CHAPTER 12.1 Braking systems

vehicle to generate the lateral forces required to maintain initial slip angle a. Thus, when the front axle is locked,
directional control and stability is severely impaired. the vehicle is unable to respond to any steering inputs and
Irregularities in the road surface or lateral forces can so its forward motion continues in a straight line.
cause the vehicle to deviate from its direction of travel.
The nature of the ensuing motion, which is rotational 12.1.4.6.2 Rear axle lock
about the vehicle vertical axis, depends on which axle has
Assume now that the fixed brake ratio associated with
locked together with the vehicle speed, tyre–ground
the same vehicle has been changed such that the rear
friction coefficient, yaw moment of inertia of the vehicle
axle locks in preference to the front as depicted in
body and the vehicle dimensions. By considering the two
Figure 12.1-15. If the vehicle is subject to the same lat-
cases of front and rear axle lock it is possible to derive
eral disturbance, then this can only be reacted by a side
useful insight into the stability problem:
force generated between the front wheels and ground and
the resulting moment about the vehicle centre of gravity
12.1.4.6.1 Front axle lock has a magnitude of Sfa. In contrast, this yaw moment now
Any disturbance in the lateral direction due to gradient, has a destabilizing effect as it causes the longitudinal axis
sidewind or left to right brake imbalance produces a side of the vehicle to move away from the direction of travel,
force Fy that acts through the centre of gravity of the thereby increasing the vehicle slip angle a. This in turn
vehicle, as shown in Figure 12.1-14. leads to a rise in lateral force at the front of the vehicle
The resultant force FR that is due to the inertia force causing an increase in yaw acceleration.
Fx caused by the braking event and the lateral force Fy It is thus preferable, from a safety point of view, for
gives rise to a slip angle a. This slip angle represents the the front axle to lock in preference to the rear as this is
difference between the longitudinal axis of the vehicle a stable condition and the driver is able to regain di-
and the direction in which the vehicle centre of gravity is rectional control of the vehicle simply by releasing the
moving. The lateral force Fy must be balanced by the side brakes. If the rear axle has locked and the vehicle has
forces generated in the tyre–ground contact patches. As begun to spin, driver reaction must be rapid if control of
the front axle is locked, no side force is generated by the the situation is to be regained.
front wheels and the resulting side force is developed In a collision situation, a frontal impact, linked to front
solely by the still rolling rear wheels. This gives rise to axle lock, will usually result in less serious occupant
a total moment of Srb. This yaw moment has a stabilizing injury than the possible side impact that could well be
effect since it causes the longitudinal axis of the vehicle associated with the uncontrolled yawing of the vehicle
to align with the direction of travel, thereby reducing the that results from rear axle lock.

Sf /2

S f/ 2 Fx
FR
a α
Tf /2 Tf /2

Fy
x

b
y

Tr /2 Tr /2

Figure 12.1-15 Rear axle lock.

378
 Braking systems CHAPTER 12.1

It is therefore feasible to apply the preceding ideas to vehicle body to be rigid and that the front and rear sus-
the formulation of a fixed brake ratio that will invariably pension spring rates, kf and kr, are linear. The spring rates
lead to front axle lock and this is commonly applied to used are the axle rates.
the design of brake systems found on passenger vehicles. The opposed spring forces generated during a braking
The fixed brake ratio is chosen such that for the unladen event are equal to the load transfer that takes place and
case both front and rear axles are on the verge of lock so are equal to
when the vehicle undertakes a 1g stop on a road surface Pzh
that has a tyre–ground adhesion coefficient of unity. yr 
l
Under such conditions, the brake ratio is equal to:
and this causes the vehicle to go down at the front and
move upwards at the rear as shown in Figure 12.1-16.
xf Ff þ Ph
¼ l
(12.1.67) Thus, on the assumption of linear springing, the com-
xr Fr  Ph
l pression travel at the front is:

Pzh
l
and on all surfaces where the tyre–ground adhesion is less yf ¼ (12.1.68)
kf
than unity, the braking will be limited by front axle lock.
The effect of axle lock on vehicle stability may also be and the corresponding travel at the rear is:
assessed through the formal derivation of the equation of
Pzh
motion associated with the yawing of the vehicle. Anal- l
ysis of the same cases of axle lock leads to identical yr ¼ (12.1.69)
kr
conclusions regarding the behaviour of the vehicle with
the added benefit that measures of yaw acceleration, The pitch angle, q, in degrees, adopted by the vehicle
velocity and displacement can be deduced. body is therefore given by:
 
yf þ yr 360
12.1.4.7 Pitch motion of the vehicle q ¼
l

2p
(12.1.70)
body under braking
The transfer of load from the rear to the front axle that Vertical and longitudinal movement of the vehicle
takes place during a braking event will cause the vehicle body centre of gravity occurs as a result of the body pitch
body to rotate about its lateral axis. This pitching motion motion and this in turn causes a small change in the
also results in a change in the height of the vehicle centre overall centre of gravity of the vehicle. The extent of
of gravity. Both of these quantities can be determined as movement of the vehicle body centre of gravity, initially
a function of vehicle deceleration using the notation in located a distance ab from the front axle at a height hb
Figure 12.1-16. The following analysis assumes the above ground, depends upon its location within the

Body centre of
δhb gravity

hb kr yr
θ
yf
kf

ab bb

Figure 12.1-16 Determination of vehicle body pitch angle.

379
 CHAPTER 12.1 Braking systems

structure, the suspension rates and the rate of de- from which it can be shown that:
celeration. An indication of the extent of this movement
can be seen in Figure 12.1-16. Under severe braking Ff zh
conditions, the vertical displacement, dhb, of the vehicle xfv ¼ þ (12.1.77)
P l
body centre of gravity equates to approximately 5% of its
original height. A detailed account of the relevant theory and
can be found in Reimpell and Stoll (1996) and from this
the change in height is given by: Fr zh
xrv ¼  (12.1.78)
P l
Fbf F
dhb ¼ yf þ yr br (12.1.71)
Fb Fb A situation giving rise to the need for a variable braking
ratio might result from a given vehicle design in which
where the maximum deceleration using a fixed braking ratio is
too low. In practice the introduction of a regulating valve
Fbf ¼ Fsf þ Faf (12.1.72) into the braking system helps to optimize the braking
efficiency over a wide range of operating conditions. Al-
Fbr ¼ Fsr þ Far (12.1.73) though such devices do not permit a continuously vari-
Fb ¼ Fbf þ Fbr (12.1.74) able braking ratio, they do offer a means of improving the
overall braking performance. Mathematical models of
in which Fb is the vehicle body weight, Fbf,r are the brake deceleration sensitive pressure regulating valves are now
reaction loads applied to the front and rear of the vehicle derived.
body, Faf,r are the unsprung weights of the front and rear
axles and Fsf,r are the front and rear axle loads. If the 12.1.4.8.1 Deceleration-sensitive pressure
loads due to the unsprung axle masses are ignored then limiting valve
a corresponding expression for the change in the height A typical valve design is shown in Figure 12.1-17. At
of the overall centre of gravity of the vehicle, dh, can be a predetermined deceleration, determined by the mass
found using: of the ball and the angle of installation, the inertial force
acting on the ball causes it to roll up the valve body and
Fsf Fsr close the valve thereby isolating the rear brakes. These
dh ¼ yf þ yr (12.1.75)
P P valves are gradient sensitive but do act in a favourable
manner. On a rising slope the valve closes at higher levels
in which P is the total vehicle weight. of deceleration allowing the rear brakes to contribute
more to the total braking effort, whilst on a falling slope
the rear brakes are isolated sooner reflecting the load
12.1.4.8 Braking with a variable transfer to the front of the vehicle caused by the
braking ratio gradient.
The effect on performance brought about by the in-
If a vehicle is to achieve maximum retardation, equal to clusion of a regulating valve in the rear brake line can be
the value of the tyre–ground adhesion coefficient, equa- assessed by deriving equations which define the brake
tion 12.1.22, then the brake system must be designed ratio for all possible values of deceleration. These may
with a continuously variable brake ratio. This must be then be used in the equations for efficiency and adhesion
equal to the ratio of the dynamic load distribution be- utilization, derived earlier, which quantify the brake
tween the front and rear for all values of deceleration. system performance. In the following analysis it is as-
Thus the variable brake ratio, Rv, is defined as: sumed the valve isolates the line to the rear brakes when
the vehicle deceleration has reached a certain value of
deceleration, zv. Note that the mechanism through which
xfv cut-off is achieved depends upon the chosen valve type
Rv ¼
xrv and this determines the actual value of zv.
R Figure 12.1-18 shows a typical front to rear brake
¼ f (12.1.76)
Rr force characteristic. For all values of deceleration less
Ff þ Pzh than zv, the brake force is apportioned between the front
l
¼ and rear axles in the fixed ratio R. Once the deceleration
Fr  Pzh
l has exceeded zv, the line pressure to the rear brakes is
held constant and so they can no longer generate

380
 Braking systems CHAPTER 12.1

D2

D1
1

3 α
1 Ball
2 Support surface
3 Control opening
4 Brake fluid
5 To rear brakes 2
6 Differential piston 4

Figure 12.1-17 Deceleration-sensitive pressure limiting valve (Limpert, 1992).

additional braking force. Consequently the brake ratio the rear brakes can no longer generate additional braking
changes from its original value. force. As the front brakes are able to respond to further
In region 1, for z  zv, the proportion of braking effort increase in line pressure then the rate of deceleration can
at the rear of the vehicle is: increase above zv and the brake ratio changes, being equal
to the slope of the dashed line. When in region 2, the
xrv ¼ xr (12.1.79) brake force at the rear, Tr, is constant and is:

The proportion of braking effort at the front of the Tr ¼ Pzv xr (12.1.82)

vehicle is therefore:
Simultaneously, the brake force acting at the front of
xfv ¼ 1  xrv
(12.1.80) the vehicle, Tf, increases and is equal to the difference
¼ 1  xr ¼ xf between the total brake force, Pz, and that sustained at
the rear axle
and so the brake ratio Rv is:
xfv x Tf ¼ Pz  Pzv xr (12.1.83)
Rv ¼ ¼ f (12.1 81)
xrv xr
Thus, in region 2, the brake ratio is defined by:
In region 2, z > zv, the valve actuates and isolates the
Tf Pzxfv Pz  Pzv xr
rear brakes from any further increase in line pressure and Rv ¼ ¼ ¼ (12.1.84)
Tr Pzxrv Pzv xr

from which

2 z  zv xr
xfv ¼ (12.1.85)
Front axle brake force Tf

z
z = zv
and

zv xr
xrv ¼ (12.1.86)
1 z
The incorporation of such a valve into the brake system
of the prototype vehicle results in improved adhesion
Rear axle brake force Tr
utilization and efficiency. Biasing the fixed brake ratio in
favour of rear axle lock improves front axle adhesion up to
Figure 12.1-18 Typical limiting valve brake force distribution. the point of lock. This can be set, through the fixed brake

381
 CHAPTER 12.1 Braking systems

ratio, to lie within the deceleration range of 0.35g to 0.45g. efficiency can be seen through comparison of Figure
The deceleration-sensitive pressure limiting valve is 12.1-19b (valve fitted) to Figure 12.1-13c (no valve).
chosen to actuate at the point of lock and this results in an
adhesion utilization diagram that has the form shown in
12.1.4.8.2 Deceleration-sensitive pressure
Figure 12.1-19a. The fixed brake ratio has been changed to
modulating valve
R ¼ xxfr ¼ 0:73
0:27 which results in a critical deceleration of
0.4 g and the valve is assumed to actuate at this level of A pressure modulating valve, or reducer valve, differs
deceleration. With reference to Figure 12.1-19a, the ve- from a pressure limiting valve as once the activation point
hicle is now governed by front axle lock over all values of has been exceeded, they do not isolate the rear brakes
deceleration as the front axle adhesion lies above that but for higher pressures the rear brake pressure increases
of the rear. The brake system now makes much better use at a lower rate than that of the front brakes. The main
of the available adhesion when executing low/moderate g advantages of this type of valve are that the rear pressure
stops as, in comparison to Figure 12.1-13b, the front axle can be increased even after the front brakes have locked
adhesion now lies closer to the optimum line. However at and the front and rear line pressures lie close to the op-
higher rates of deceleration, the front axle adhesion utili- timum values. A typical pressure modulating valve
zation has reduced. The beneficial effect of the valve on characteristic is shown in Figure 12.1-20.
The effect on system performance of a pressure
modulating valve may be assessed in a similar fashion to
1.2 that of a limiting value. As before, it is assumed that the
valve is actuated once the vehicle deceleration has
1.0 exceeded a certain value of deceleration, zv, and the
Adhesion utilization

exact value of zv is determined by the mechanics of the

0.8 valve.
In region 1, the vehicle deceleration is less than zv and
0.6
so the vehicle brakes in accordance with the fixed brake
ratio assigned to the system,
0.4
Optimum k
k limit xrv ¼ xr (12.1.87)
0.2 Front axle
Rear axle xfv0 ¼ xf (12.1.88)
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 and
Deceleration g
(a) xfv x
Rv ¼ ¼ f (12.1 89)
xrv xr
1.2
In region 2, the vehicle deceleration is greater than zv
1.15
and the valve has actuated. The front and rear brake
1.1

1.05
Braking efficiency

0.95

0.9
Front axle brake force Tf

0.85

0.8 100%
Front axle z = zv
0.75
Rear axle
0.70
0 0.2 0.4 0.6 0.8 1 1
Tyre–ground adhesion coefficient
(b)

Figure 12.1-19 (a) Adhesion utilization, modified vehicle fitted Rear axle brake force Tr
with deceleration-sensitive pressure, (b) brake system efficiency,
modified vehicle fitted with deceleration sensitive. Figure 12.1-20 Typical modulating valve brake force distribution.

382
 Braking systems CHAPTER 12.1

forces increase in accordance with the slope of the valve The substitution of the modulation valve for the
characteristic and this causes the overall vehicle brake limiting valve into the brake system of the prototype
ratio to vary, being equal to the slope of the dashed line. If vehicle enables improvements to be made to the adhe-
the slope of the brake force characteristic in region 2 is sion utilization at high rates of deceleration. This is due
x2
defined as xf2 then the brake force at the rear axle, Tr, is: to the ability of the valve to increase the line pressure to
r

Tr ¼ Pzv xr þ ðPz  Pzv Þxr2 (12.1.90) the rear brakes at a reduced rate. By appropriate choice
of the slope of the brake force characteristic in region 2,
and the brake force at the front axle is: the front axle adhesion curve can be forced to move
closer to the optimum. In this example, setting the ratio
T f ¼ T  Tr x2
for region 2 to be xf2 ¼ 0:88
0:12 causes the front and rear
r
¼ Pz  ðPzv xr þ ðPz  Pzv Þxr2 Þ (12.1.91) adhesion curves to cross at a deceleration of 0.8g
and gives rise to the adhesion utilization diagram of
Thus, the overall brake ratio, defined by the slope of the Figure 12.1-21a.
dashed line, is: The adhesion behaviour is identical to that shown in
Figure 12.1-19a up until valve actuation. Thereafter the
Tf Pzxfv Pz  ðPzv xr þ ðPz  Pzv Þxr2 Þ front axle adhesion converges on that of the optimum at
Rv ¼ ¼ ¼
Tr Pzxrv Pzv xr þ ðPz  Pzv Þx2 0.8g. Decelerations greater than 0.8g lead to rear axle
(12.1.92) lock but this is strictly admissible according to the ad-
hesion utilization requirement specified in the governing
from which
EEC braking directive. The fact that the front axle ad-
hesion now deviates little from the optimum illustrates
z  zv xr  ðz  zv Þxr2 the positive advantage that can be gained through the
xfv ¼ (12.1.93)
z introduction of a bias valve into the brake system.
and Comparison of Figures 12.1-11, 12.1-13c, 12.1-19b and
12.1-21b illustrates this point by showing the progressive
zv xr þ ðz  zv Þxr2 refinement of the brake system efficiency during the
xrv ¼ (12.1.94)
z design process.

1.2
Adhesion utilization

1.0
Optimum k
0.8
k limit
0.6
Front axle
0.4 Rear axle
0.2
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2
Deceleration g

(a)
1.20
1.15
1.1
Braking efficiency

1.05
1 100%
0.95 Front axle
0.9 Rear axle
0.85
0.8
0.75
0.7
0 0.2 0.4 0.6 0.8 1
Tyre–ground adhesion coefficient

(b)

Figure 12.1-21 (a) Adhesion utilization, modified vehicle fitted with deceleration-sensitive pressure modulating valve, (b) brake system
efficiency, modified vehicle fitted with deceleration sensitive.

383
 CHAPTER 12.1 Braking systems

12.1.5 Materials design have a relatively high MOT to prevent thermal degra-
dation of the surface although, due to the nature of its
12.1.5.1 Materials requirements composition, the MOTof the pad material will always be
lower than that of the disc. A low conductivity for the
for braking systems pad or shoe material is desirable to minimize conduction
of heat to other components of the system, in particular
In any conventional foundation brake, the relative rota- to the hydraulic fluid. The material should be reasonably
tion of the so-called ‘friction pair’ under the action of the wear resistant but not excessively so since wear can be
brake system activating force is responsible for generat- beneficial in promoting a uniform contact pressure
ing the frictional retarding torque required to slow the distribution and preventing ‘hot spotting’. Likewise
vehicle. Most friction pairs consist of a hard, usually the elastic modulus of the material should be relatively
metallic, rotating component and a relatively compliant low to give good conformity with a roughened or
‘friction’ material in the form of a brake pad or shoe. The thermally distorted rotor surface. Finally, as for the
materials requirements for the rotating and stationary rotor, the friction material should be cheap and easy to
components of the friction pair are therefore quite dif- manufacture.
ferent as discussed below. The friction material selected to meet the above re-
Any rotor material must be sufficiently stiff and strong quirements is invariably a complex composite consisting
to be able to transmit the frictional torque to the hub of a variety of fibres, particles and fillers bonded together
without excessive deformation or risk of failure. How- in a polymeric matrix such as phenolic resin. For many
ever, the stresses arising from thermal effects are much years, asbestos fibres were an important element of
higher than purely mechanical stresses and are more friction materials due to their excellent thermal and
likely to give concerns over disc integrity. Thus the rotor friction properties. For health and safety reasons, asbes-
material should have high volumetric heat capacity tos has now largely been replaced by other less harmful
ðr$cP Þ and good thermal conductivity (k) in order to fibres, e.g. Kevlar. The exact composition of any friction
absorb and transmit the heat generated at the friction material must be tailored to the application and knowl-
interface without excessive temperature rise. Further- edge of the formulation is proprietary to the supplier.
more the maximum operating temperature (MOT) of
the material should be sufficiently greater than the
maximum expected temperature rise to ensure integrity
12.1.5.2 Cast iron rotor metallurgy
of the rotor even under the most severe braking condi- The overwhelming majority of rotors for conventional
tions. Ideally the rotor material should have a low co-
automotive brakes is manufactured from grey cast iron
efficient of thermal expansion (a) to minimize thermal
(GI). This material, also known as flake graphite iron, is
distortions such as ‘coning’ of a disc. It should also have
cheap and easy to cast and machine in high volumes. It has
low density (r) to minimize the unsprung mass of the
good volumetric heat capacity due mainly to its relatively
vehicle. It should be resistant to wear since generally it is high density, and reasonable conductivity due largely to
far easier and cheaper to replace the friction pads or the presence of the graphite (or carbon) flakes. The co-
shoes than the rotor itself. Finally, and most importantly, efficient of thermal expansion is relatively low and the
the rotor should be cheap and easy to manufacture. material has an MOT well in excess of 700  C (but note
The brake pad or shoe represents the stationary part of
that martensitic transformations at high temperatures can
the foundation brake assembly. Normally a proprietary
composite friction material is bonded to a steel backing
plate or shoe platform. The primary function of the
Table 12.1-2 Tensile strength and conductivity of some common
friction material is generally considered to be the pro-
cast irons
duction of a stable and predictable coefficient of friction
to enable reliable and efficient braking of the vehicle over Min. tensile Thermal conductivity
a wide range of conditions. In fact, it is the combined Grade strength (MPa) at 300 8C (W/m K)
tribological characteristics of both rotor and stator
materials (i.e. the ‘friction pair’) which are responsible 400/18 SG) 400 36.2
for the generation of the frictional torque. As for the 250 GI 250 45.4
rotor, the friction material must have sufficient structural
integrity to resist the mechanical and thermal stresses. 200 GI 200 48.1
This is particularly important for the bond between the 150 GI 150 50.5
friction material itself and the steel structure which
supports it, as a complete failure here could have *
Spherical graphite iron.
disastrous consequences. The friction material should

384
 Braking systems CHAPTER 12.1

lead to hot judder problems). Although the compressive

Table 12.1-3 Physical properties of three candidate disc materials
strength is good, the tensile strength is relatively low and
the material is brittle and prone to microcracking in ten- r cP r,cP k
sion. As the proportion of flake graphite in GI is increased, Disc kg J kgL1 kJ mL3 WmL1 a 310L6
the tensile strength reduces but thermal conductivity in- material mL3 KL1 KL1 KL1 KL1
creases as shown in Table 12.1-2. Note that spheroidal
graphite iron (SG) has a higher tensile strength than GI High carbon 7150 438 3132 50 10
cast iron
but a much reduced conductivity which explains why it is
rarely used for brake rotors. Generic 20% 2800 800 2240 180 17.5
Currently GI grades used for disc brakes fall into two SiC-
categories reflecting two different design philosophies reinforced Al
MMC
(MacNaughton and Krosnar, 1998):
Carbon–carbon 1750 1000 1750 40–150 0.7
1. Medium carbon GI (e.g. Grade 220)
composite
These irons are used for small diameter discs such as on
small- and medium-sized passenger cars. Such discs will
run hot under extreme conditions, and good strength and has been directed at investigating light-weight alterna-
thermal crack resistance at high temperatures are tives to cast iron. Two such alternatives which have re-
therefore required. ceived serious attention are aluminium metal matrix
2. High carbon GI (e.g. Grade 150) composites (MMCs) and carbon–carbon composites,
These grades tend to be used for larger vehicles where typical properties for each of which are displayed in Table
space constraints are not as content limited. Discs are 12.1-3. together with corresponding properties for a high
larger and, with the improved conductivity due to the carbon cast iron (Grieve, et al. 1995).
high carbon, will run cooler. Strength retention at high Aluminium MMCs normally incorporate 10–30% by
temperature is therefore not as critical and manufactur- volume silicon carbide particle reinforcement within
ability improves with the higher carbon content. a silicon-containing alloy matrix. The resulting composite
Alloying elements can be applied to all grades of cast iron has much lower density than cast iron and much im-
with the general effect of improving strength but at ex- proved conductivity. Thus the thermal diffusivity
pense of thermal properties and manufacturability. The ðk=r,cP Þ is much higher which opens the possibility of
most commonly used elements and their effects are as lighter discs running cooler by being able to rapidly
follows: conduct heat away from the friction interface. However,
aluminium MMCs have a low MOT (c. 500  C) and
 chromium increases strength by stabilizing pearlitic there are serious consequences if this MOT is exceeded
matrix at high temperatures (preventing martens- since complete surface disruption may then occur leading
itic transformations) but tends to promote formation to extremely rapid pad wear. Ideally higher re-
of bainitic structures which cause casting/machining inforcement contents or alternative reinforcing materials
difficulties and can reduce pad life; (e.g. alumina) should be used to increase the MOT but
 molybdenum similar to chromium; the former causes severe casting difficulties whilst alu-
 copper increases strength without causing manufac- mina reinforcement results in poorer thermal properties.
turing difficulties; It can be seen from Table 12.1-3 that carbon–carbon
 nickel as for copper but more expensive; composites have an even lower density than aluminium
 titanium reported to influence friction performance MMCs and can have a conductivity almost as high. Their
but rarely used at significant levels. MOT is also very high, raising the possibility of using thin
rotors which run much hotter and lose heat by radiation
12.1.5.3 Alternative rotor materials as well as by conduction/convection. Also the very low
coefficient of thermal expansion of carbon minimizes
Although GI is a cheap material with good thermal thermal distortions. Thus there is the potential for very
properties and strength retention at high temperature, its significant weight savings with carbon–carbon composite
density is high and, because section thickness must be discs. However, the material has a poor low temperature
maintained for both manufacturability and performance, friction performance and moreover is currently much
cast iron rotors are heavy. Currently there are significant more expensive than metallic alternatives. Hence, it is
incentives to reduce rotor weights in order to: (a) reduce likely to remain confined to high performance race car
emissions by improving the overall fuel consumption of applications for the foreseeable future.
the vehicle, and (b) aid refinement and limit damage to When considering alternative materials or designs for
roads by reducing the unsprung mass. Thus much effort disc brakes, reference can be made to the so-called

385
 CHAPTER 12.1 Braking systems

Heat input

MOT of material
temperature
Disc

Heat dissipation
(a) Strategy I (b) Strategy II (c) Strategy III

Figure 12.1-22 The ‘bucket-and-hole’ analogy.

‘bucket-and-hole’ analogy in which the rate of water Strategy III

flow into the bucket is taken to represent the heat flow Even smaller diameter bucket but much deeper with
into the disc and the height of the water level in the bucket moderately sized hole (see Figure 12.1-22c). This
represents the maximum temperature of the disc implies a material with high MOT which can be allowed
surface. A hole in the bucket represents the ability of the to run much hotter than current designs and so lose
disc to lose heat to the surroundings. The volume of the significant amounts of heat by radiation as well as more
bucket is therefore the heat capacity of the disc whilst the moderate amounts by conduction/convection. Carbon–
height is the MOTof the disc material. The question then carbon composites are a possibility here but, as men-
is how close does the level of water in the bucket get to tioned above, these are currently too expensive for
overflow! mass produced vehicles. High-temperature steels with
Consideration of the ‘bucket-and-hole’ analogy and good strength retention at temperatures well in excess of
with reference to the typical material properties of Table 1000  C may also be candidate materials under this
12.1-3, three distinct strategies for brake rotor materials heading. Such discs could perhaps be made much thinner
can be identified (Grieve et al., 1996): and without vents, and therefore also save significant
weight. However, there would be concerns over compati-
Strategy I
ble friction materials and heat transfer to other compo-
Large diameter and relatively deep bucket with small
nents in the underbody wheel arch area if discs were
hole (see Figure 12.1-22a). This implies a high volu-
allowed to run much hotter than is currently the practice
metric heat capacity to store heat during braking and
with cast iron.
a relatively high MOT but only moderate conductivity of
heat away from the rubbing surfaces. Current GI discs
represent such a system but some steels may also meet 12.1.5.4 Disc materials/design
these criteria. evaluation
Strategy II Ultimately, any new brake material or design must be
Smaller diameter and relatively shallow bucket but large validated by experimental trials on actual vehicles to
hole (see Figure 12.1-22b). This implies smaller volumetric allow accurately for model-specific parameters such as
heat capacity and a relatively low MOT. Hence, it is im- the effect of body trim on rotor cooling. However, much
portant to have high conductivity to transfer heat to other can be learnt about potential new rotor materials or
parts of the rotor and then the surroundings in order to designs by numerical simulations of critical brake tests
prevent temperature build-up at the rubbing surfaces. using finite element (FE) analysis. Such techniques re-
Aluminium MMC may meet these criteria but recent re- quire the rotor and/or stator geometry to be broken
search (Grieve et al., 1998) suggests that this can only be down into a number of small non-overlapping regions
successfully achieved for currently available MMCs if the known as elements which are assumed to be connected
brake rotor is redesigned to increase its thermal mass and to one another at certain points known as nodes. A 2D
cooling capability. Other materials that may be successful axisymmetric FE idealization can be used as a first ap-
with appropriate development include high reinforcement proximation but, for more accurate simulation of the
content MMCs and coated alloy discs but again there are heat flow and stresses, a 3D model is desirable such as
manufacturing, integrity and cost issues to be resolved. the 10  segment model of a brake disc and hub shown

386
 Braking systems CHAPTER 12.1

3 1
Disc

Figure 12.1-23 Finite element model of 10  segment of vented disc and hub.

in Figure 12.1-23. Note that in order to accurately and the vehicle is subsequently left to stand at the end of
simulate the heat loss from the rotor, it is sometimes the descent; (ii) a repeated high speed autobahn stop
necessary to include the wheel and other components in with the rotor allowed to cool only moderately between
the model. stops. The former test determines the ability of the
The heat input to the system is estimated from the- design to limit temperature build-up in the rotor by heat
oretical consideration and applied over the rubbing sur- transfer to the atmosphere whilst the high-speed re-
face. The heat loss to the surrounding is specified by peated stop examines the ability of the rotor material to
convective and sometimes radiative heat transfer condi- withstand repeated thermal cycling and the ability of the
tions along relevant boundaries of the model. The tem- friction pair to resist ‘fade’ under these severe conditions.
peratures predicted by a thermal analysis can be used as Friction performance cannot easily be predicted by
input conditions to a structural analysis in order to pre- the FE approach and there remains a requirement for
dict thermal deformations and stresses. If the pad is in- dynamometer testing to determine the fade-and-wear
cluded in the model, the contact pressure distribution characteristics of every new friction pair. The dyna-
(and hence the distribution of heat input) can be esti- mometer can either be a full-scale device or a small
mated leading to the possibility of a fully coupled sample rig in which the geometry and loading conditions
thermal-structural analysis (Brooks et al., 1994). are scaled to give an accurate representation of the actual
In addition to details of geometry and material prop- brake. These tests will not only give data on friction
erties, accurate date on heat loss to other components and performance over a wide range of conditions but can also
to the atmosphere are vital to allow accurate predictions be used to determine the MOT of the pad and rotor
of rotor temperatures using FE methods. Such data can be materials by progressively increasing the temperature at
generated by conducting the so-called ‘cooling tests’ on the rubbing interface until some form of failure occurs
actual vehicles fitted with representative brake rotors (Grieve et al., 1996).
carrying rubbing or embedded thermocouples. The rotor
surface is first heated to a predetermined temperature by
dragging the brakes and then allowed to cool whilst the 12.1.6 Advanced topics
vehicle is driven at constant velocity. By comparing the
experimental rate of cooling with that predicted by 12.1.6.1 Driver behaviour
the FE simulation for different boundary conditions,
optimized heat transfer coefficients can be derived which The driver of a vehicle plays a key role during any braking
are then assumed to apply for different rotor materials event since his/her reactions to external stimuli have
and factored for the varying air stream velocity under a direct bearing on his/her ability to maintain complete
different test conditions. control over the vehicle trajectory and deceleration rate.
Two very different vehicle brake tests are often sim- A knowledge of how the driver interacts with these ex-
ulated to critically examine the maximum temperatures ternal stimuli and the way in which the vehicle responds
and integrity of new rotor materials or designs: (i) a long to the control signals generated by the driver is vital to
slow Alpine descent during which the brakes are dragged the future development of safe road transport systems.

387
 CHAPTER 12.1 Braking systems

Many experimental studies, including Newcomb the control unit, its associated software and the array of
(1981), Newcomb and Spurr (1974), Mortimer (1976) sensors that combine to define the overall effectiveness of
and Spurr (1972), have been undertaken that have led to the system. The controller must operate in closed-loop
improved understanding of driver behaviour during fashion, be able to take into account the in-use variation of
braking. These have focused on the study of limb dy- the system parameters and fail safe.
namics, pedal effort, braking kinematics and response to
external stimuli such as obstacles and road signs. This has
12.1.6.3 Anti-lock braking systems
given rise to the development of mathematical models
that embody a representation of the driver into a model Under normal braking conditions, the driver of a vehicle
of the vehicle dynamics. Any such model, typified by makes use of the linear portion of the brake slip vs brake
McLean et al. (1976), contains elements that describe force characteristic (Figure 12.1-7). The brake force
the dynamics of the vehicle, the braking system, the coefficient, m, builds from zero in the free rolling state to
neuro-muscular system and force characteristics of the a maximum, mp, at around 20% slip and within this region
driver and finally the motion detection system/sensory the wheel is both stable and controllable. When braking
characteristics of the driver together with feedback loops under extreme conditions the driver may demand
as appropriate to the model in question. The adaptive a brake torque that is greater than that which is capable
nature of the driver that is captured in such models re- of being reacted by the wheel. This results in a torque
quires enhancement but simulation of vehicle braking imbalance that causes the wheel slip to increase and the
performance with the driver can yield deceleration wheel rapidly decelerates to the full lock condition and in
characteristics that match closely those from this state, the brake force coefficient is approximately
experiment. 0.7 mp. If the front wheels have locked, then steering
control is lost and if rear wheel lock takes place then the
12.1.6.2 Brake by wire vehicle becomes unstable. Simultaneously, the ability of
the vehicle to generate side force markedly reduces
The driver behind brake-by-wire systems has arisen from (Figure 12.1-7), and this explains why limiting wheel
the ongoing development of modern braking systems slip, thereby avoiding wheel lock, is more critical for
such as anti-lock and traction control systems (TCSs) steering and directional stability of the car than for
along with the need to effect their seamless integration stopping distance alone.
within the overall chassis control strategy. There are two The purpose of ABS is to control the rate at which
strategies currently receiving attention. individual wheels accelerate and decelerate through the
The first utilizes a conventional hydraulically actu- regulation of the line pressure applied to each founda-
ated braking system, that includes the brake fluid, tion brake. The control signals, generated by the con-
brake lines and conventional actuators, together with troller and applied to the brake pressure modulating
a significant number of electro-hydraulic components unit, are derived from the analysis of the outputs taken
(Jonner et al., 1996). from wheel speed sensors. Thus, when active, the ABS
The second relies upon a full electro-mechanical system makes optimum use of the available friction between
(Bill, 1991; Maron et al., 1997; Schenk et al., 1995) in the tyres and the road surface.
which the brake force is generated directly by electro-
mechanical foundation brake actuators. The electro- 12.1.6.4 Traction control systems
mechanical system potentially requires little maintenance
due to the removal of the hydraulic fluid as the means of Traction control systems aim (TCSs) to control and
energy transmission and this conveniently combines with maintain vehicle stability during acceleration manoeu-
a reduction in the amount of hardware demanded by the vres, by, for example, preventing wheel spin when ac-
brake system which in turn leads to an overall weight re- celerating on a low friction surface or on a steep up-grade.
duction. Such systems may also contribute towards the This is achieved by the optimization of individual wheel
enhancement of passenger safety as the location of the torques through the control of some combination of fuel
pedal assembly within the vehicle can be optimized so that mixture, ignition and driven wheel brake torque. TCSs
the likelihood of lower leg injury is minimized during are able to utilize components used in ABS and in-
impact events. As with all advanced control systems, it is tegration of the two systems is becoming commonplace.

388
 Braking systems CHAPTER 12.1

12.1.7 References and further reading

Automotive brake systems. (1995). Pub. Barton D.C. (Ed.), Advances in on Braking of Road vehicles. paper C41/
Robert Bosch GmbH. Distributed by Automotive Braking Technology. PEP. 76.
SAE. ISBN 1-56091-708-3. Grieve, D., Barton, D. C., Crolla, D. A., Mortimer, R. G. (1976). Implications of
Bill, K. (1991). Investigations on the Chapman, J., & Buckingham, J. T. some characteristics of drivers for brake
behaviour of electrically actuated friction (1998). Design of a light weight system performance. Proc ImechE Conf
brakes for passenger cars. EAEC 91021. automotive disc brake using finite Braking of Road Vehicles187–195,
Brake handbook. (1981) (2nd edition). element and Taguchi techniques. In: Proc Loughborough.
Alfred Teves GmbH. IMechE Part D. 245–254. Newcomb, T. P. (1981). Driver behaviour
Brake technology and ABS/TCS systems. Jonner, W.-D., Winner, H., Dreilich, L., & during braking. SAE/IMechE Exchange
(1999). Pub SAE. SP-1413ISBN Schunck, E. (1996). Electrohydraulic Lecture 1981. SAE 810832.
0-7680-0345-8. brake system – the first approach to Newcomb, T. P., & Spurr, R. T. (1967).
Brooks, P.C., Barton, D.C., Crolla, D.A., brake-by-wire technology. Detroit: SAE Braking of Road Vehicles. Chapman and
Lang, A.M. and Schafer, D.R. (1994). A 960991. Hall Ltd. Out of print.
study of disc brake judder using a fully Limpert, R. (1992). Brake design and Newcomb, T. P., & Spurr, R. T. (1974).
coupled thermo-mechanical finite safety. SAE. ISBN 1-56091-261-8. Aspects of driver behaviour during braking.
element model. Proceedings FISITA 94 MacNaughton, M. P., & Krosnar, J. G. XV Congress FISITA. paper A. 1.4.
Conference, Paper No 945042, Beijing, (1998). Cast iron – a disc brake material Puhn, F. (1995). Brake handbook. HP.
October. of the future? In D. C. Barton, & M. J. Books. ISBN 0-89526-232-8.
Grieve, D., Barton, D.C., Crolla, D.A., Haigh (Eds.), Automotive Braking: Reimpell, J., & Stoll, H. (1996). The
Buckingham, J.T. and Chapman, J. Recent Developments and Future Trends Automotive Chassis: Engineering
(1995). Investigation of light weight PEP. Principles. Arnold.
materials for brake rotor applications. Maron, C., Dieckmann, T., Hauck, S., & Schenk, O. E., Wells, R. L., & Miller, I. E.
IoM Conference on Materials for Lean Prinzler, H. (1997). Electromechanical (1995). Intelligent braking for current
Weight Vehicles, University of Warwick, brake system: Actuator control and future vehicles. Detroit: SAE 950762.
November. development system. Detroit: SAE
970814. Spurr, R.T. (1972). Driver behaviour during
Grieve, D., Barton, D. C., Crolla, D. A., braking. Proc. Symp. on Psychological
Buckingham, J. T., & Chapman, J. McLean, D., Newcomb, T. P., & Spurr, R. T. Aspects of Driver Behaviour, paper 1-2,
(1996). Investigation of light weight (1976). Simulation of driver behaviour Holland.
materials for brake rotor applications. In during braking. Proc IMechE Conference

389