|Views: 845|Likes: 5

Published by teodor1974

See More

See less

Steering in bicycles and motorcycles

J. Fajans

a)

Department of Physics, University of California, Berkeley, Berkeley, California 94720-7300

Received 12 April 1999; accepted 12 November 1999

Steering a motorcycle or bicycle is counterintuitive; to turn

right

, you must steer

left

initially, andvice versa. You can execute this initially counter-directed turn by turning the handlebars explicitly

called counter-steering

or by throwing your hips to the side. Contrary to common belief,gyroscopic forces play only a limited role in balancing and steering

D. E. H. Jones, Phys. Today

23

4

, 34–40

1970

. ©

2000 American Association of Physics Teachers.

Centrifugal forces will throw your bike over on its side if you steer the handlebars in the direction of a desired turnwithout ﬁrst leaning the bike into the turn. Indeed, bicyclecrashes are often caused by road obstacles like railroadtracks or sewer grates turning the front wheel and handlebarsabruptly. Leaning the bike into the turn allows gravitationalforces to balance the centrifugal forces, leading to a con-trolled and stable turn. Thus steering a bike involves a com-plicated interaction between centrifugal and gravitationalforces, and torques applied to the handlebars, all mediated bythe bike geometry.

I will use the word bike to refer to bothmotorcycles and bicycles.

One method of establishing the proper lean is counter-steering, i.e., explicitly turning the handlebars counter to thedesired turn, thereby generating a centrifugal torque whichleans the bike appropriately. Counter-steering is employedby both motorcyclists and bicyclists, though most bicyclistscounter-steer unconsciously. You may have noticed, how-ever, that while on a bicycle, it is surprisingly difﬁcult to rideclear of a nearby high curb or sharp drop. This is becauseyou must steer towards the edge to get away from the edge.It is easy to directly demonstrate counter-steering on a bi-cycle. While riding at a brisk pace

possibly downhill toavoid the complications of peddling

, let go with your lefthand while pushing the right handlebar with the open palmof your right hand. Since your hand is open, you can onlyturn the handlebar left, but the bike will turn right.The process of making a counter-steered right turn is il-lustrated in Fig. 1.

In this description, right and left are inthe frame of the rider.

The turn can be broken into ﬁvesomewhat arbitrarily divided steps:

a

You initiate the turn by applying a torque to thehandlebars, steering the front wheel to the left.

b

The wheel steers to the left. The rate at which the steer-ing angle increases is set primarily by the moment of inertia

I

s

of the wheel, fork, and handlebars around thesteering axis, and by the ‘‘trail’’

described later.

Asthe bike is now turning to the left, a centrifugal torqueleans both you and the bike frame to the right. Gyro-scopic action also leans the bike to the right, but, as Iwill show later, its effect is negligible.

c

Transmitted by the fork, the increasing lean attempts tolean the front wheel over as well. For the ﬁrst time,gyroscopic action becomes important, as the wheel re-sponds to this ‘‘leaning’’ torque by attempting to steerto the right, thus counteracting the steering torque. Thesteering angle stops increasing.

d

The leaning torque overcomes the steering torque andthe wheel steering angle decreases. Note that the leancontinues to increase because the bike is still turningleft.

e

As the bike has now acquired substantial leaning ve-locity, the lean increase cannot end instantly. Driven bythe still increasing lean, the wheel steering angle passessmoothly through zero and then points right. The cen-trifugal torques reverse direction, eventually halting thelean increase and balancing the gravitational torques.As no more leaning torque is applied to the wheel, thesteering angle stabilizes, and the bike executes the de-sired right turn.Alternately, the required lean can be generated by throw-ing your hips in the direction counter to the turn. Throwingyour hips is how a bike is steered no-hands. The sign of theeffect is subtle, but a half-hour session in an empty parkinglot should convince you that while riding no-handed, yousteer the bike by leaning your shoulders in the direction of the desired turn. Since angular momentum is conserved by asudden shift of your shoulders, your hips move the oppositeway, thereby leaning the bike the opposite way as well. Withthe bike now leaning, the bike’s ‘‘trail’’ becomes important.As the steering axis is not vertical, the point of contact of thewheel with the road ‘‘trails’’ the intersection of the steeringaxis with the road

see Fig. 1

a

. The trail makes the bikeself-steer: when the bike leans to the left, the front wheelsteers left; when the bike leans to the right, the front wheelsteers right. This effect is easily demonstrated by standingbeside a bicycle and leaning it from side to side.

The trail isthe single most important geometric parameter which entersinto the handling of a bicycle.

The complete hip-turn sequence is similar to the counter-steer sequence illustrated in Fig. 1, except that you initiatethe turn by throwing your hips left. The bike leans left aswell, and the trail steers the wheel left. Centrifugal torquesthen lean the center of mass to the right, and gyroscopicforces eventually steer the wheel right.A mathematical model is necessary to be more precise. Asthe geometry is complex and the constraints nonholonomic,an exact model is very complicated. I will use a simpliﬁedmodel, good for small leans and steering angles, and willignore some of the details. Since similar models have beenreported before,

1,2

I will only sketch the derivation.Equating the time derivative of the vertical angular mo-mentum to the torque applied to the handlebars

N

s

gives

I

0

˙

I

s

¨

N

s

Mgb

L

Mb

v

2

L

2

,

1

654 654Am. J. Phys.

68

7

, July 2000 © 2000 American Association of Physics Teachers

where

and

are the lean and steering angles,

is thewheel’s angular rotation frequency,

I

0

is the moment of in-ertia of the wheel around its rotation axis,

M

is the totalmass,

g

is the acceleration due to gravity,

b

is the horizontaldistance from the rear wheel to the center of mass,

is thetrail,

L

is the wheelbase, and

v

is the velocity of the bike

seeFigs. 1

a

and 1

c

. The ﬁrst term on the left-hand-side

LHS

comes from changes in the direction of the rotationalangular momentum

L

, and the second term comes from theangular momentum around the steering axis. The second andthird terms on the right-hand-side

RHS

come from the trail.The second term is responsible for the wheel steering to-wards the lean, as described above. The third term attemptsto straighten the wheel at high velocity, and comes from acastering effect. Good intuitive derivations of the trail termsare given in Ref. 2.Considering the wheel only, equating the time derivativeof the angular momentum around the lean axis to the torque

N

f

exerted on the wheel by the fork gives

I

0

˙

v

I

0

L

I

w

¨

N

f

,

2

where

I

w

is the moment of inertia of the wheel, fork, andhandlebars around the lean axis. The ﬁrst term on the LHScomes from changes in the direction of

L

due to changes inthe steering angle, and the second from changes in the direc-tion of

L

as the bike goes around in a circle of radius

L

/

. The third term on the LHS comes from the angularmomentum of the wheel around the lean axis.The wheel applies an equal and opposite torque,

N

f

, tothe center of mass through the fork and frame. Equating thetime derivative around the lean axis to the torques yields

I

¨

N

f

hM

v

2

L

hMg

v

I

0

L

hbM

v

L

˙

,

3

where

h

is the distance between the lean axis and the centerof mass, and

I

is the moment of inertia around the lean axis.The second term on the RHS is the centrifugal torque actingon the bike as it travels in a circle of radius

, the third termis the gravitational torque, and the fourth term is the gyro-scopic action of the rear wheel.The last term on the RHS of Eq.

3

is an unusual ﬁctitioustorque whose origin is not obvious in the derivations in Refs.1 and 2. The origin of this torque, called the kink torque laterin this paper, is described in the Appendix.Equations

1

–

3

are a reasonable, but approximate,model of bike dynamics. Left out are the ﬁnite thickness of the tires, friction, effects other than the trail of the steeringaxis not being precisely vertical, etc. In particular, an effectdue to the deformation of the tire into a cone-like shape isoften thought to be important.

3,4

As mentioned above, theequations are valid only for small

and

.

Some of theangles in subsequent ﬁgures are sufﬁciently large that higherorder terms are needed for fully accurate solutions.

Recentmeasurements by Jackson and Dragovan on an instrumentedbicycle validate a similar mathematical model.

5

The equa-tions are easy to solve numerically. I used Mathcad’s adap-tive Runge–Kutta routines;

6

the Mathcad worksheets can bedownloaded from my website.

7

The solutions to Eqs.

1

–

3

exhibit growing oscillations.These oscillations are discussed later in the text, and can besuppressed by adding a damping term,

˙

, to the RHS of Eq.

1

. Physically, this damping could come from passiveresistance from the rider’s arms on the handlebars, or fromactive responses from the rider.As an example, take a bicycle with

h

1.25m,

L

1.0m,

0.02m,

b

0.33m,

I

0

0.095kgm

2

,

I

s

0.079kgm

2

,

I

w

0.84kgm

2

,

I

163kgm

2

, and

0.65Js. Assume that your mass and the bike mass sum to

M

100kg

Ref. 8

and that you travel at the brisk speed of

v

7 m/s. You would begin a typical counter-steered rightturn

25m

by torquing the handlebars left

positive

and end the turn by counter-steering in the opposite direction

torquing the handlebars right.

Figure 2

a

shows thehandlebar torque time history used in this example. Othertime histories are possible, but for this time history, at least,you would

never

torque the handlebars in the direction of thedesired right turn until the end of the turn. Moreover, thetorques are very small; you would never apply a force greaterthan 0.9 N

for handholds 0.5 m apart

, the equivalent of aweight of 0.092 kg.Assuming that the torque follows the curve shown in Fig.2

a

, the lean and steering angle responds as shown in theFig. 2

b

. Figure 2

c

shows the torques that cause the lean.

Fig. 1. A counter-steered right turn, as described in the text. The bikegeometry is shown in

a

and

c

. The center of mass is represented by theﬁlled circle at the location of the seat. The arcs around the steering axis andthe lean axis show the direction and approximate magnitude of the torqueapplied to the handlebars and the net leaning torque.655 655Am. J. Phys., Vol. 68, No. 7, July 2000 J. Fajans

The centrifugal torque initiates the lean, with help from thekink torque. The lean angle reaches a steady state when thecentrifugal and gravitational torques balance. The contribu-tions from the remaining torques in Eqs.

2

and

3

, includ-ing all the gyroscopic torques, are not visibly different fromzero, and have been omitted from the ﬁgure.Figure 2

d

shows the contributions of the trail torques andof the gyroscopic torque (

I

0

˙

) to changing the steeringangle

Eq.

1

. While the gyroscopic torque is non-negligible, it is much smaller than the trail torques and some-what smaller than the handlebar torque. Thus, in accord withJones’ observation

9

that a bike equipped with a gyro nullingcounter-spinning wheel behaves much like a normal bicycle,the ‘‘feel’’ of the bike is dominated by the trail.A counter-steered turn on a motorcycle is qualitativelysimilar. For a motorcycle with

h

0.60m,

L

1.54m,

0.117m,

b

0.77m,

I

0

0.77kgm

2

,

I

s

0.57kgm

2

,

I

w

4.0kgm

2

,

I

118kgm

2

, and

3Js, ridden by arider whose mass, with the mass of the bike, totals

M

300kg,

10

at

v

20m/s, the turn parameters for a sharp

11

right turn

200m

are shown in Fig. 3. The handlebartorques used in this example are higher than for the bicycle:12.7 Nm, or a force of 12.7 N for handholds 0.5 m apart, theequivalent of a weight of 1.3 kg. Contrary to the assertion inRef. 3, gyroscopic action plays no role in leaning the bike.However, as shown in Fig. 3

d

, and in agreement with Ref.3, it does play a role in steering the front wheel back towardsthe desired direction.Scaling relations show why gyroscopic action does not

Fig. 2. A counter-steered turn on a bicycle.

a

Torque applied by the riderto the handlebars.

b

Steering

and lean

angles

dotted lines indicateequilibrium angles

.

c

Leaning torques.

d

Steering torques.Fig. 3. A counter-steered turn on a motorcycle.

a

Torque applied by therider to the handlebars.

b

Steering

and lean

angles

dotted lines indi-cate equilibrium angles

.

c

Leaning torques.

d

Steering torques.656 656Am. J. Phys., Vol. 68, No. 7, July 2000 J. Fajans

Filters

1 hundred reads

Adventura liked this

Murali Ramanath liked this

Va5h33r liked this

jules_LA995768 liked this

arbiter007 liked this