This article was downloaded by: [Uni of Ontario Inst of Technology ] On: 27 August 2013, At: 10:46 Publisher:

Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Vehicle System Dynamics: International Journal of Vehicle Mechanics and Mobility

Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/nvsd20

Longitudinal Oscillations of Vehicle/Trailer Combinations Induced by Overrun Braking

R. S. SHARP
a a

Mechanical Engineering Department, University of Leeds, Leeds, LS2 9JT Published online: 27 Jul 2007.

To cite this article: R. S. SHARP (1982) Longitudinal Oscillations of Vehicle/Trailer Combinations Induced by Overrun Braking, Vehicle System Dynamics: International Journal of Vehicle Mechanics and Mobility, 11:1, 43-61 To link to this article: http://dx.doi.org/10.1080/00423118208968687

PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the Content) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions

Vehicle System Dynamics 1 1 (1982), pp. 43-61.

Longitudinal Oscillations of Vehicle/Trailer Combinations Induced by Overrun Braking

R. S. SHARP*
Downloaded by [Uni of Ontario Inst of Technology ] at 10:46 27 August 2013

SUMMARY
A mathematical model for the representation of longitudinal oscillations which can occur in car/trailer systems in braking, when the trailer brakes are applied through compression of the towing hitch, is described. The model is used to show how the trailer braking system parameters affect the steady deceleration performance ofthe vehiclecombination, and the stability,in the linear system sense, of the steady motions. The sensitivity of the stability to other system design parameters is also examined. Digital simulation of the motions occurring in response to a step input of car braking torque is reported, with the results confirming the predictions of the linear stability analysis, and also showing the influence of backlash in the trailer brake actuating mechanism. The system is shown to be capable of self-excitation in a "shunting" mode, in which the car and trailer motions are in antiphase, with the stability/damping property critically dependent on drawbar damping. and only weakly dependent on other system parameters. The characteristic frequency of the "shunting" mode oscillations is shown to be controllable via the stiffness of the trailer brake linkage, but this frequency is closely related to the steady drawbar deflection which occurs in uniform deceleration. The model behaviour described provides a basis for the design of relevant systems whose longitudinal dynamic characteristics will be satisfactory.

1. INTRODUCTION Overrun braking systems involve actuation of the trailer brakes through longitudinal loading of the ball hitch joining the trailer to the towing vehicle. They are common on trailers of medium weight. Small trailers typically d o not have a braking system of their own, while towards the other extreme, the trailer brakes are actuated directly from the foot pedal in' the towing vehicle.
Mechanical Engineering Department, University of Leeds, Leeds, LS2 9JT.

44

R. S. S H A R P

Downloaded by [Uni of Ontario Inst of Technology ] at 10:46 27 August 2013

Hales [I] noted the occurrence of low frequency longitudinal oscillations in some of the vehicle systems in question, and remarked that the oscillations may be violent, and will normally become apparent as the vehicle combination slows to a stop. Hales developed a simple model of the longitudinal motions of an appropriate car and trailer, and found a lightly damped "shunting" mode to be predicted by his model. However, even for no damping of the longitudinal motion of the trailer hitch, this shunting mode remained stable, with implications that the oscillations which are observed to occur, are forced oscillations arising from non-uniformity in the braking effort through time, unevenness in the road surface, or other things. In Hales' analysis, it was assumed that the braking force between the trailer wheels and the road surface is directly proportional to the longitudinal load on the hitch. This is considered too simplistic a representation for accuracy in the results, and a new model, including'wheel spin freedoms and a description of the development of longitudinal force by a tyre through the mechanism of longitudinal slip [2], is developed in this paper. The model is utilised to examine the steady state deceleration of a car and trailer, in particular relating it to that of thecar alone (as did Hales), thestability of such steady motions, and its dependence on the brake system design. Also the actual time history of certain motions are obtained by digital simulation. It may be anticipated at the outset that if very little braking effort is applied to the trailer wheels, the behaviour of the system will approach that when the trailer has no brakes. The steady state deceleration performance will compare badly with that of the towing vehicle alone, but there will be no tendency towards shunting oscillations. On the other hand, as the trailer brakes are made more effective for a given hitch load, the steady state performance may be made to approach that of the car alone, but an increasing tendency to oscillation can be expected. This fundamental conflict between good braking performance, and freedom from oscillation has received special attention.

NOTATION
masses o f car and trailer polar moments o f inertia o f (lumped) wheels wheel radii longitudinal displacements & and i , respectively angular.displacements o f wheels 8, and 8 , respectively perturbation variables employed in stability analysis longitudinal tyre forces longitudinal hitch force braking torques applied to wheels

8',

LONGITUDINAL OSCILLATIONS BY OVERRUN BRAKING spring rates stiffness of drawbar in tension (simulation only) damper coefficient ratio of braking torque to drawbar compression ratios of braking force to longitudinal slip forward speed steady state acceleration steady state acceleration of car alone steady value of (x, - +),the drawbar compression V - Ke, (steady state) V - RIB, (steady state) drawbar clearance (or backlash) distance (simulation only) time car trailer

A Ac D

Downloaded by [Uni of Ontario Inst of Technology ] at 10:46 27 August 2013

E F
C

t suffices
C

2. T H E CAR/TRAILER SYSTEM MODEL

The physical model of the c a r h a i l e r system employed is shown diagrammatically in Fig. 1.

- ---i--t(t) I
I

i
I

CAR
i c

TRAILER

1 9
I
I
I I

HMc

I L - - - A

/////////// Fig. 1. Diagrammatic representation of system model.

The model involves consideration of the structures of car and trailer as rigid, the lumping of the four car wheels into one, and the lumping of the two trailer wheels into one. The two parts of the vehicle are allowed independent longitudinal freedoms, with displacements x, and x, measured from some initial condition, and wheel spin freedoms with displacements defined by 8, and 8, are allowed. The road is considered flat and level. The input to the system is a braking torque r,, see Fig. 2, which leads to a decrease in the angular velocity of the car wheels, and a longitudinal retarding force, F,. In due course, the hitch mechanism is compressed and applies the ,leads to a force, F, , a t the trailer tyre contact area as trailer brakes. The torque, 7, with thecar.

46

R. S. SHARP

Downloaded by [Uni of Ontario Inst of Technology ] at 10:46 27 August 2013

Fig. 2. Forces and torques acting during braking.

The spring, s, is included in the model only when it is employed for stability analysis. This is simply a device for avoiding the occurrence of a zero frequency rigid body mode, which would upset the computer routines used for finding eigenvalues, s has been chosen numerically small enough not to influence significantly the motions of interest. It is considered that the purpose of the model and the results obtained from it justify the reduction of the real system to one with only four degrees of freedom.

3. EQUATIONS O F MOTION
By reference to Fig. 2, the equations of motion for the model can be seen to be: for the translation of the car = - Fh - Fc - s {q - f (t))

w&

for the translation of the trailer M,x, = Fh - Ft for the rotation of the car wheels I,& = Fc - rc and for the rotation of the trailer wheels 1,8, = F , R - T , Taking the hitch spring and damper to have linear characteristics,

and assuming a similar linear relationship between compression of the trailer drawbar and the braking torque applied to the trailer wheels,

Over the full range of operation, which of course does not only involve braking, it may be expected that the drawbar stiffness and damping, and the trailer braking

LONGITUDINAL OSCILLATIONS BY OVERRUN BRAKING

47

torque resulting from drawbar compression will be strongly non-linear; but for small perturbations from a steady decelerating condition, it is considered reasonable to treat these functions as linear, and to use numerical studies to determine the sensitivity of the system behaviour to parameter changes. The longitudinal tyre forces are represented as linear functions of longitudinal slip,
Downloaded by [Uni of Ontario Inst of Technology ] at 10:46 27 August 2013

F== c k C {(k- r t e c ) / k }
and

(7)

These functions are a good representation of actual behaviour for steady state behaviour with longitudinal slip values less than about 0.1, [2]. Ignoring the lags which occur between the generation of slip and the occurrence of the braking forces, which are associated with the compliance of the tyre structure, [3], is justifiable in view of the low frequency of the oscillations which are the subject of the investigation. Eliminating F,, using (9,Fc using (7), F, using (8), and 7, using (6) from equations (1) --- (4), and using the substitutions X, = k, X, = i , H, ,= 8,, and HI = 8, leads to the following equation set:

~ j < -I, (C + C,,/X)

X, = (k

+ s) x, - C X, - kx,
(9)

- (ckcrt/Y) Hc = s f(t)

- C X, - kx, -I~,k +,(C + Ck,/X,) X, + kx,

- (CkIS/X,) HI = 0
(1 1)

4. STEADY STATE DECELERATION

8, = A/R, ,(x,

For steady braking without oscillations, & = x, = A (say), 8, = A I R , x,) = D (say), the drawbar compression, and & = i , = V . Also,

48 letting V-KO,= E, and Vwith s = 0 in (9) become:

R. S. SHARP

Ret= F, the equations of

motion (9)

---

(14)

Downloaded by [Uni of Ontario Inst of Technology ] at 10:46 27 August 2013

- 7, (WR, + &/k)-' and a simple computer program has been employed to

It is easily shown that with no trailer, the acceleration, A,, is given by A, =

find the deceleration ratio A/&, and the corresponding drawbar compression, D, for various parameter combinations of interest.

5. SYSTEM STABILITY
In order to linearise the equations of motion to allow stability analysis b y eigenvalue/vector methods, the motion variables are first replaced by perturbation variables according to the following relations: x,=Vt+xg &=V+$
%=j%
x, = v t + x ;

i, = v *, = K;

+ i;

8, = V/R, + 8: e, = e: 8, = V/R, + 8;
and

8,=&

The tyre braking forces Fc and F, described by equations (7) and (8) then take the forms Fc=Ckc {(<-R,~:)/(v+<)) and F,=Ckt {(j(-R,e:)/(v+i:)}

LONGITUDINAL OSCILLATIONS BY OVERRUN BRAKING

49

which are linearised by ignoring the perturbation velocities with V, such that they become Fc = Ckc(i; -k g : ) /V and
Downloaded by [Uni of Ontario Inst of Technology ] at 10:46 27 August 2013

< and g i n comparison

F,=Ck,(i:-R,~:)/V

Using these relations, equations (I) to (6), and equations (16), the equations of motion become

h2< + (C f G,/V)Z 4- (k + s)$ -Ck' -k g -( ~ , k * = s[f(t) - Vt] - c;tS- k d + M,x: + (C + Ck,/V)i; + kx; - (c,,R,/v)& = 0
-( c k c ~ / v+ )~ ~ e+:(ckcg/v$: = -7,
and

- r d - (Ck,R,/V/i:

+ rx; 4- I,@

(c,,R:N)~

=0

Stability is determined by the homogeneous form of these equations, so that the right hand side expressions s [f(t) -Vt] and T, d o not require definition, and standard manipulations of the homogeneous equations [4], yield the following as the dynamic matrix

standard digital computer methods have been employed to find the eigenvalues, indicating stability-damping and natural frequency properties, and eigenvectors, indicating the corresponding mode shapes.

50

R. S. SHARP

6. DIGITAL SIMULATION For digital simulation of such a small system, there is little advantage in insisting that the system is linear, and, of course, the spring s is not needed. Normally in practice, some clearance in the mechanism through which the drawbar actuates the trailer brakes will occur, such that when the car brakes are first applied, the drawbar clearance, E, is used up before the trailer brakes are applied at all. It was thought that this clearance distance may be a significant design parameter, and its effect has been included in the simulation study by replacing equation (5) by the relations

Downloaded by [Uni of Ontario Inst of Technology ] at 10:46 27 August 2013

= C (& - i t )

for O<(X,-&)<E

= C (& - i,) - k, (x, - &) for (x, - &) < 0

the first of which describes the situation when the drawbar clearance has been taken up, and the trailer brakes are being applied, the second of which applies when the drawbar clearance is at neither extreme position, and the third of which applies when the drawbar is in tension, and includes the stiffness of the drawbar in tension, k,. Also equation (6) is replaced by

=O

for

(x,-&)<E.

Equations (9) to (14) with these changes introduced are in the preferred form for numerical solution. Responses to step inputs of car brake torque have been obtained.

7. PARAMETER VALUES
If all the parameters in the problem were varied freely, and steady state, stability, and simulation results were obtained, the essential physics of the problem would be obscured by the mass of information available. In an attempt to arrive economically at the essence of the problem, the following stratagem has been devised. It will be imagined that the towing vehicle is a typical car, having a mass, of 1000 kg, with (lumped) wheel inertia, I, of 5.0 kgm2 and (lumped) braking force coefficient, C,,, of 400,000 N. The trailer will have half this mass, half the wheel inertia, and half the brake force coefficient. Both car and trailer will have the same wheel radius, R, = R,~ 0 . 3 5 m. and forward speeds from 10 to 40 m/s will be considered of interest. Free choice of the drawbar design parameters, stiffness, k,

w,

LONGITUDINAL OSCILLATIONS BY OVERRUN BRAKING

51

and damping coefficient, C, and the trailer brake system design parameter, r, will be allowed, and the dependence of the system behaviour on practicable values of these three parameters, and the speed, will be studied. This datum system and some variations from it will be investigated.

8. RESULTS
Downloaded by [Uni of Ontario Inst of Technology ] at 10:46 27 August 2013

8.1 Steady state deceleration The form of equations (15), which describe the steady state braking behaviour, make it clear that the drawbar damping, C, has no influence on this part of the problem, and that the forward speed, V, does not influence the acceleration, A, or the drawbar compression, D. Also, proportional variations in k, the drawbar stiffness, and r, the trailer braking torque/drawbar deflection ratio, can be seen to change the drawbar deflection in proportion, and to leave the deceleration ratio unaffected. The ratio of r to k , which is the braking torque to hitch force ratio, determines the deceleration ratio, large r/k corresponding with deceleration ratios approaching unity. The dependence ofdeceleration ratio on r/k, regardless of the values of V and k , is shown in Fig. 3, in which the hitch stiffnesses required to give specific steady state hitch deflection values for 5m/sec2 deceleretion are also indicated. Beyond a certain point, there is obviously little value in increasing r further, but the implications of designing to achieve high deceleration ratios require consideration. Three particular cases, with some variations, will be examined. These are (i) r / k = 1.225 m, k = 11550 N/m, (ii) r / k = 2.975 m, k = 5480 N/m, and (iii) r/k = 8.23 m, k = 2125 N/m, giving deceleration ratios of 0.9, 0.95, and 0.98 respectively, and each giving a steady state hitch deflection of I cm per lrn/sec2 of deceleration.

8.2 Stability of the steady state motion The hitch damper coefficient, C , is critically important in determining the stability of the steady state motions, and calculations for a wide range of damping values have consequently been carried out. Results are recorded in root locus plot form in Fig. 4. The three cases (i), (ii), and (iii), with V = 20 m/sec can be seen to give very similar shunting mode oscillation frequencies, this being the result of choosing k in each case to give 1cm hitch deflection for lm/sec2 deceleration. They are also very similar in respect of their damping, case (iii) being marginally the least damped for each value of C . The effects of halving k, and changing V for case (ii) are also shown in Fig. 4. The real parts of the eigenvalues, the damping parameters, are affected little by these changes, although for low damper coefficients (C values), increasing V is significantly destabilising. In Fig. 4, the curves for V = 10, V = 20, and V = 40 have been separated artificially by using k = 5398 for V = 10, and k = 5562 for V = 40.

R. S. SHARP

Downloaded by [Uni of Ontario Inst of Technology ] at 10:46 27 August 2013

Fig. 3. Dependence o f steady state deceleration ratio and hitch stiffness o n trailer brake torque to hitch force ratio (hitch deflection quoted is for 5m/s/s deceleration).

The influences of several other parameter changes on the shunting mode stability are shown in Fig. 5. The cases treated are variations ofcase (ii)above and are defined as follows: case (iv) Ck,= 160,000 N; case (v) I, = 4.0 kg m2, r / k = 3.0 m, k = 5358 N/m; case (vi) I, = 2.0 kg m2, r / k = 2.95 m, k = 5317 N/m; case (vii) M, = 800 kg, r/k = 3.77 m, k = 4224 N/m; case (viii) M, = 400 kg, r/k = 2.335 m, k = 5158 N/m; and case (ix) R, = 0.28 m, r/k = 2.44 m, k = 5081 N/m. In each of these six cases, the first named parameter has been reduced to 80% of its standard value, r/k is such that a steady state deceleration ratio of 0.95 is achieved, and k has been chosen to separate the curves from each other on the root locus plot and to thereby make them distinguishable. The main impression obtainable form the results in Fig. 5 is of insensitivity of the system stability to parameter changes, although the damping is significantly increased when either the car mass o r the trailer mass is reduced, for a given damper coefficient, C. With any of the parameter base sets, adequate damping of

LONGITUDINAL OSCILLATIONS BY OVERRUN BRAKING

Downloaded by [Uni of Ontario Inst of Technology ] at 10:46 27 August 2013

C A S E ----. CASE CASE ------- ( 1 1 ......-...(11)

me..-..

(1) (1 1 ) 4 (111) U ITH K HALVED WITH V 10 3 ( 1 1 ) UITH V 40

REAL PART -5 -4 -3
1 1 1 1 .

-2

-1

Fig. 4. Root locus plots for the shunting mode for variations in hitch damping.

the shunting mode simply depends on having a sufficiently large damper coefficient. The shunting mode eigenvectors show the displacements of car and trailer are very nearly in antiphase for light damping, but as the damping increases, the phase difference changes gradually form 1800 to around 1100 (for C = 3200). The amplitude ratio (x,/q) reduces from 9 to 3 with increasing damping for the case (i) design, from 19 to 3 for case (ii), and from 48 to 3 for case (iii). In particular, for the

R . S. SHARP

Downloaded by [Uni of Ontario Inst of Technology ] at 10:46 27 August 2013

-CASE
----CASE
----.

(1 V)

5..
4 ..

CASE (V) (V1) ------- CASE ( V 1 1 ) -..-..--.-. CASE ( V l l l ) - - - - - - - CASE (1x1

3..
2.1*-

REAL PART -5 -4 -3
1 1 1 4 1

-2

-1

Fig. 5. Root locus plots for the shunting mode for variations in hitch damping.

lightly damped systems which have C small, the shunting mode involves much more oscillation in the trailer than in the car, especially when the trailer braking system is designed to give a steady state deceleration ratio near to unity. The trailer motions are then only weakly coupled, in a dynamic sense, t o those of the car, which supports the contention that a more elaborate model of the towing vehicle is unnecessary for the present purposes.

LONGITUDINAL OSCILLATIONS BY OVERRUN BRAKING

55

Downloaded by [Uni of Ontario Inst of Technology ] at 10:46 27 August 2013

8.3 Simulated motions The simulation studies conducted have concentrated on the case (ii) vehicle combination with r / k = 2.975 m, and k = 5480 N/m, giving a steady state deceleration ratio of 0.95. They have been aimed firstly at determining the extent to which the linearised analysis of stability gives an accurate prediction of the behaviour, and secondly at determining the influences of the non-linear features, drawbar clearance, and differential drawbar stiffness (k, # k). All the results included are for initial velocities of both car and trailer of 40 m/s, and for a step input of car braking torque of 1000 Nm. Also initially x, and x,, the car and trailer displacements, were zero. Car and trailer decelerations, the drawbar compression, and the longitudinal tyre slip values are shown in the figures 6-1 1. Figs. 6(a) and (b) are for E = 0 (no backlash), and C = 20 Ns/m. In 6(a), k, = 100 k, while in 6(b), k, = k. Figs. 7and 8are also fore = 0, but have C = 100, and C = 500 Ns/m respectively. Figs. 9, 10, and 11 result from taking e = 0.1 in each case, and C = 100,500, and 2500 Ns/m respectively. In Fig. 6, for C = 20, the oscillations build up with time as predicted by the linear stability analysis. The frequency of the oscillations is near that given by the
lo 8
U

M C E C E R A T I O N , il/S/S

-------CAR

TRAILER

2 0

Fig. 6a. Simulation results for C = 20. e = 0,k l= 548000.

56

R. S. SHARP

Downloaded by [Uni of Ontario Inst of Technology ] at 10:46 27 August 2013

b. Simulation results for C = 20, e = 0,k l= 5480.

eigenvalues when k, = k (6(b)), but is considerably greater fork, = 100 k as would be expected. In accordance with the shunting mode eigenvectors, the bulk of the oscillatory motion isconfined to the trailer when k, = k,although that ofthecar is not negligible. The effect of having k, = 100 k coupled with the low damping and the consequent high amplitude oscillations is to cause large forces to be transmitted between car and trailer through the drawbar in tension over a part of the oscillation cycle. The result is that variations in the car deceleration are much greater, and of course these variations represent a source of discomfort to the car passengers, and are best minimised. The linear system stability resultsare accurately represented also when C = 100, and C = SOO(Figs. 7and 8) in terms of both frequency and damping. In particular, it is confirmed by the simulation results that the hitch damping is extremely effective in stabilising the shunting oscillations. This is also shown to be the case when backlash of 0.1 m is introduced into the hitch mechanism, Figs. 9,10 and 11. After the car brakes are applied, there is some delay before the trailer brakes work, which delay increases with the hitch damping. The oscillation frequency is lowered by the clearance, and, since the steady

LONGITUDINAL OSCILLATIONS BY OVERRUN BRAKING

57

Zr

A~t~~~~ITI~N.fl/~I~

------- CAR

TRAILER

Downloaded by [Uni of Ontario Inst of Technology ] at 10:46 27 August 2013

::: I:

1IHE.S
TIRE S L I P . 1

Fig. 7. Simulation results for C = 100, t = 0,kl= 548000

state drawbar compression is 0.1 m greater than before, the plunging ofthe drawbar is increased. The non-uniformity of the car's deceleration is also increased somewhat. The mean deceleration over the six second simulation period is slightly less with the backlash than without it, but the differences are too small to mean very much in practical terms.

9. CONCLUSIONS
Longitudinal oscillations induced by braking in cadtrailer combinations in which the trailer brakes are of the overrun type are self excited vibrations, and require no periodic unevenness in either road surface o r braking effort for their occurrence. The oscillations are stabilised very effectively by hitch damping, and will only be troublesome when very little hitch damping is incorporated into a system. Then, the dynamic coupling between car and trailer motions is weak, the bulk of the oscillatory motion involving the trailer responding to fluctuations in the trailer

58
2

R. S. SHARP

,4CCELERATION, ti/S.fS

-------CAR TRAILEQ

Downloaded by [Uni of Ontario Inst of Technology ] at 10:46 27 August 2013

Fig. 8. Simulation results for C = 500, c = 0.

tyre longitudinal force. The behaviour of interest is then insensitive to any variations of the towing vehicle. If the oscillations are large enough for the drawbar to be loaded in tension over part ofthe cycle, and the drawbar is very stiff in tension, undesirable large changes in towing vehicle deceleration will occur. Self excitation is encouraged by high vehicle speed, and by trailer brake system designs which give steady state deceleration ratios near to unity, but the system sensitivity to these things is very small compared with that to hitch damping. Clearance or backlash in the drawbar mechanism slightly deteriorates the: shunting behaviour, and may have an adverse effect on the limit braking perfor-mance achievable, because of the resulting lag between the application of the car and the trailer brakes, which lag increases with increasing drawbar damping.

LONGITUDINAL 0SCILLATIONS.BY OVERRUN BRAKING

59

ACCELERATION, M/S/S

-------CAR TRAILER

-11

Downloaded by [Uni of Ontario Inst of Technology ] at 10:46 27 August 2013

Fig. 9. Simulation results for C = 100, c = 0. I

Downloaded by [Uni of Ontario Inst of Technology ] at 10:46 27 August 2013

60

Fig. 10. Simulation results for C = 500, e = 0.1

CAR

R. S. SHARP

LONGITUDINAL OSCILLATIONS BY OVERRUN BRAKING

ACCELERATION. tl/S/S CAR TRAILER

61

Downloaded by [Uni of Ontario Inst of Technology ] at 10:46 27 August 2013

Fig. I I. simulation results for C = 2500, t = 0. I

REFERENCES 1. Hales, F. D., "The basic dynamics of overrun brake systems". Proc. Conf. Science in Motor Vehicles, Belgrade, 1975, pp. 253-266. 2. Pacejka. H. B. "The tyre as a vehicle component" Ch. 7 of Mechanics of the Pheumatic Tyre, S. K. Clark (ed)., N.B.S. Monograph 122, Washington, D.C., 1971. 3. Pacejka, H. B. "Tyre factors and vehicle handling", Int. Jour. Veh. Des., Vol. I, No. 1, 1979, pp. 1-24. 4 . Sharp, R. S. "Eigenvalues and eigenvectors in linear vibration theory", Proc. 1st. Brit. Conf, on Teaching of Vibration and Noise in Higher Education, Sheffield City Polytechnic, 1976, pp. 131-152.