FIZIKA SPhO Handouts

Mechanics: Rotation (Solutions)

FIZIKA SPhO Training

June 2025

Contents
1 Notes 2
1.1 Rotational Kinematics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.1.1 Rotational Analogues of the SUVAT Equations . . . . . . . . . . . . . . . 2
1.1.2 Rolling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.1.3 Instantaneous Centre of Rotation (ICOR) . . . . . . . . . . . . . . . . . . 3
1.2 Moment of Inertia (MOI) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.2.1 Naive Summation/Integration . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.2.2 Parallel Axis Theorem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.2.3 Perpendicular Axis Theorem . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.3 Rotational Dynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.3.1 Torque and Angular Momentum . . . . . . . . . . . . . . . . . . . . . . . 8
1.3.2 Rotational Analogue of N2L (Euler’s Rotation Equation) . . . . . . . . . 9
1.3.3 Rotational Kinetic Energy, Work and Power . . . . . . . . . . . . . . . . . 10
1.3.4 Rotational Collisions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.3.5 Physical Pendulums . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
1.4 Ideas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
1.4.1 Massive Pulleys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
1.4.2 Where Does Normal Force Act? . . . . . . . . . . . . . . . . . . . . . . . . 14
1.4.3 Inextensibility Constraints . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

2 Problems 17

3 Advanced Problems 32

4 Appendix 35
4.1 More Methods of Calculating MOI . . . . . . . . . . . . . . . . . . . . . . . . . . 35

1
FIZIKA SPhO Handouts

1 Notes
In real life, many objects are not well-modelled by point masses, since they have finite sizes. In
this case, the point of application of forces matter, and will lead to rotation of the object.

1.1 Rotational Kinematics

Just like in regular translational kinematics, we will first study rotation without caring about
where it comes from.

1.1.1 Rotational Analogues of the SUVAT Equations

In translational kinematics, we are concerned with the quantities x, v, a and t. We may draw an
analogy to rotational quantities as follows:

x ←→ θ, v ←→ ω, a ←→ α (1)

where θ is the angular displacement, ω is the angular velocity and α is the angular acceleration.

As a result, the following definitions hold:

dθ
ω := (2)
dt
dω
α := (3)
dt
and the corresponding rotational analogues of the SUVAT equations are:

ωf = ωi + αt (4)
1
θf = θi + ωi t + αt2 (5)
2
|ωf |2 = |ωi |2 + 2α · (θf − θi ) (6)

1.1.2 Rolling
We commonly discuss rolling, which is a combination of translation and rotation. In other
words, you may think of rolling as a sum of:
1. Translation of a fixed point on an object
2. Rotation of the object about that fixed point
For instance, for the ball below, the rolling motion is a sum of the translation of the CM and
rotation about the CM.

2
FIZIKA SPhO Handouts

For rolling without slipping, the condition is

vCM = Rω (7)

while for rolling with slipping, we cannot simply relate vCM and ω (i.e. vCM ̸= Rω).

Differentiating Equation (7) with respect to time, we can relate aCM and α for the non-slip case:

aCM = Rα (8)

Remark. When there is no slipping, friction is static and does not do work. When there is
slipping, friction is kinetic and does work. You need to account for this if you want to calculate
the frictional force or apply energy conservation!

1.1.3 Instantaneous Centre of Rotation (ICOR)

It can be more helpful to visualise rolling using the ICOR.

The velocity at any point (for instance, point P2 in the diagram above) is given by

vP2 = vtrans + vrot = vCM + ω × r2 (9)

Applying this equation, we can see that at point P0 , vCM and ω × r0 perfectly cancel out, if
there is no slipping. (Both have magnitude |vCM | and point in opposite directions.)

Hence, the point P0 is actually instantaneously stationary!

vP0 = 0 (10)

We usually call P0 the ICOR. You can equivalently visualise rolling as a fixed-axis rotation
about the ICOR.

Remark. This visualisation (the ICOR) and the visualisation of the translation of the CM +
rotation about the CM are equivalent.

Another nice geometrical fact is that any point on the rigid body moves perpendicular to
a line drawn to the ICOR. Using the diagram above,

P0 P2 ⊥ v2 (11)

Example 1.1 (Ricardo). An inextensible string is spooled around a cylinder at its inner radius
r. The loose end of the string is rolled around a pulley and pulled with speed v. The string
makes an angle θ with the vertical. If the outer radius of the cylinder is R, determine the CM
speed of the cylinder, assuming it rolls without slipping.

Let the CM speed of the cylinder be V . Because it rolls without slipping,

V = Rω

3
FIZIKA SPhO Handouts

Now, let’s draw all the relevant velocity vectors on the diagram.

Considering the velocities at point A, since point A moves with speed v along the string,
r v
ωr − V sin θ = v ⇒ V − V sin θ = v ⇒ V = r
R R − sin θ

Remark. This idea of taking a convenient point and summing velocity vectors is useful,
especially when the said point has its velocity constrained. In this case, the constraint is the
fact that the string is pulling point A, which forces it to have speed v along the string.
Example 1.2 (Kevin Zhou). A bicycle wheel is rolling without slipping. When it is photographed,
its spokes look blurred, except along a curve of special points. What is this curve?
For the points to not be blurred, we want the points to be moving parallel to the spokes.
This ensures that in the small time where the picture is captured, the point still lies along the
spoke and will hence appear clear.

If r is the position vector from the CM, then

v = vCM + ω × r
describes the velocity at r. We also want the velocity to point along the spoke, so v ∥ r.

Quantitatively, let the wheel have radius R, and set up coordinates such that the centre
of the wheel is at (0, R) and the bottom of the wheel (which is also the ICOR, as the wheel is
not slipping) is at (0, 0). Consider the velocity of some point (x0 , y0 ) on the curve.

The gradient between this point and the ICOR is xy00 . From Equation (11), the velocity vector
will be perpendicular to this line. The gradient of the normal line is − xy00 .

We can find the equation of this normal line. Since it also passes through (x0 , y0 ),
x0 y − y0 x0 x2 + y02
− = ⇒ y=− x+ 0
y0 x − x0 y0 y0

4
FIZIKA SPhO Handouts

This line must pass through (0, R). Thus,

2 2
x20 + y02 R R
 
R= ⇒ x20 + y02 − Ry0 = 0 ⇒ x20 + y0 − =
y0 2 2
R
Thus, our desired curve is a circle of radius 2 centered at (0, R2 ).

1.2 Moment of Inertia (MOI)

The rotational analogue of mass is the moment of inertia, or the MOI. The MOI depends on
the geometry of the shape, which influences its mass distribution.

MOIs are always calculated about an axis. Some common MOIs which you should know are:

Remark. By right, you aren’t expected to memorise MOIs. However, you are highly encouraged
to just memorise them (for these common shapes above), to speed up your working during
contests like SPhO! (You’d just need to memorise the numerical prefactors anyway.)

Now, let’s look at how to find these MOIs. Interested readers may also consult the Appendix for
more interesting but niche methods.

1.2.1 Naive Summation/Integration

In the case of point masses, the MOI is defined as
X
I := mi ri2 (12)
i

Thus, the MOI of a point mass M at a distance R away from the axis of interest is I = M R2 .

5
FIZIKA SPhO Handouts

However, as usual, most objects are continuous mass distributions instead. As an integral,
the MOI is defined as Z
I := r2 dm (13)

where r is the distance of the infinitesimal mass dm from the axis of interest.
Example 1.3. Determine the MOI of a uniform thin disk of mass M and radius R about its
central axis.
We can just imagine the disk to be made of many concentric rings, from radius 0 to radius R.
We can then integrate to sum up all the moments of inertia.

For a ring of mass M and radius R, all the mass is equidistant from the central axis, at
a distance R away. Thus, we can take out the r2 from the integral:
Z Z Z
Iring = r2 dm = R2 dm = R2 dm = M R2

For the disk, we can define our mass density per unit area as
M M
σ= =
A πR2
Hence, integrating over all the rings,
" #R
r4 πσR4
Z R
1
Z Z
Idisk = dI = (dm) r2 = σ (2πr) r2 dr = 2πσ = = M R2
0 4 0
2 2

Example 1.4. Determine the MOI of a uniform solid sphere of mass M and radius R about its
diameter.
The mass density per unit volume of the sphere is
M M 3M
ρ= = 4 3 =
V 3 πR
4πR3

Slice the sphere into many disks as such:

√
R2 − z 2 , thus dV = πr2 dz = π R2 − z 2 dz for each small disk.


We have r =

Integrating over all the disks, we get

Z R
1 1 2 3M  2 2
Z Z Z  2
Isphere = dI = (dm) r2 = R − z 2 ρ dV = 3
R − z 2
dz = ... = M R2
2 2 −R 8R 5

6
FIZIKA SPhO Handouts

1.2.2 Parallel Axis Theorem

To simplify MOI calculations, the parallel axis theorem states

I = ICM + M d2 (14)

where d is the distance between the axis of interest and the axis passing through the CM.

Remark. When applying the parallel axis theorem, one of the axes must be through
the CM. It is wrong to say that I2 = I1 + M d2 for any random points 1 and 2!
Example 1.5. Knowing that the MOI of a thin rod of mass M and length L about the axis
1
through its centre and perpendicular to its length is I = 12 M L2 , find the MOI for the same rod
about the axis through one of its ends and perpendicular to its length.
This is a very simple application of the parallel axis theorem:
 2
1 L 1
I = ICM + M d2 = M L2 + M = M L2
12 2 3
Let’s look at a more complicated application of the parallel axis theorem.
Example 1.6 (F=ma). The moment of inertia of a uniform equilateral triangle with mass m
and side length a about an axis through one of its sides and parallel to that side is 18 ma2 . Using
this result, calculate the moment of inertia of a uniform regular hexagon of mass m and side
length a about an axis through two opposite vertices.
Of course, you could have performed naive integration to solve this. However, as the question
wants you to use the result of the MOI of the equilateral triangle, we shall split the hexagon up
into six equilateral triangles as follows:

Triangles labelled A and B contribute different MOIs about the axis.

For A, we can directly use the result, so each triangle contributes

1 m 2 1
 
IA = a = ma2
8 6 48
For B, we need to use the parallel axis theorem.

7
FIZIKA SPhO Handouts

By the parallel axis theorem,

√ !2
3 1 m 2 m 1 2 1
   
ICM = I − M a = a − a = ma2
6 8 6 6 12 144
and by the parallel axis theorem again,
√ !2
3 1 m 1 2 1
 
2
IB = ICM + M a = ma + a = ma2
3 144 6 3 16
Thus, the final MOI of the hexagon is
1 1 5
   
Ihexagon = 4IA + 2IB = 4 ma2 + 2 ma2 = ma2
48 16 24
Remark. When calculating the MOI of B, we applied the parallel axis theorem twice. The first
time we applied it, we actually applied it in reverse, by finding ICM from I! This trick is very
useful for simplifying calculations.

1.2.3 Perpendicular Axis Theorem

Only for planar objects (i.e. objects in 2D), the perpendicular axis theorem states
Iz = Ix + Iy (15)
Usually, in Physics Olympiad, you’ll only apply this for symmetrical situations where Ix = Iy .
Example 1.7. Knowing that the MOI of a uniform thin disk of mass M and radius R is 12 M R2
about its central axis, find the MOI of the same disk about its diameter.
This is a very simple application of the perpendicular axis theorem.

By symmetry, Ix = Iy , and Iz = 21 M R2 . Thus,

1 1
Ix = Iy = Iz = M R2
2 4
where the subscripts indicate the axis which the MOI is calculated with.

1.3 Rotational Dynamics

Now that we understand the kinematics, let’s now analyse how rotation even comes about.

1.3.1 Torque and Angular Momentum

The torque, τ , is the rotational analogue of force. It is defined as
τ := r × F (16)
where r is the position vector (from the origin) of the point of application of F.

The angular momentum, L, is the rotational analogue of linear momentum. It is defined as

L := r × p (17)
where p is the linear momentum of the object.

Using F = ma and p = mv, Equations (16) and (17) are related by

dL
τ = (18)
dt

8
FIZIKA SPhO Handouts

From Equation (18), you can see that

τ =0 ⇒ L = constant (19)

This is the conservation of angular momentum, or COAM. When there is no net torque
about a given axis, the angular momentum is conserved!

1.3.2 Rotational Analogue of N2L (Euler’s Rotation Equation)

For a fixed-axis rotation (i.e. rotation about a stationary axis), we have

L = r × (mv) = r × (mω × r) = mr × (ω × r) = mr2 ω = Iω (20)

where the vector triple product and the fact that ω ⊥ r was invoked in the second last equality.

Thus, substituting this into Equation (18),

d
τ = (Iω) = Iα (21)
dt
Equation (21) is commonly referred to as Euler’s Rotation Equation, and can be seen as a
rotational analogue of N2L (by swapping m with I and a with α).

Remark. When applying Equation (21), it is crucial that you consider the same pivot/axis
for calculating both your torque and MOI!

You’ll usually use N2L to write equation(s) involving forces, and Equation (21) to write an
equation involving torques, and solve the equations. Let’s see an example on how to apply this.
Example 1.8 (SPOT TST 2022). Consider a yo-yo made up of two uniform solid disks of
radius R and each with mass M , connected rigidly by a light cylindrical axle of radius r < R,
such that the disks and axle all share a common axis. A thin light string is wound tightly around
the axle. The free end of the string is held fixed and the yo-yo is released from rest. Assume that
the string stays vertical as it is unwound from the axle. (i) Determine the downward acceleration
a of the yo-yo when it is released from rest. (ii) Determine the tension T in the string when the
yo-yo is released from rest.
Firstly, the MOI of the yo-yo about its central axis is given by
1
 
I = 2Idisk =2 M R2 = M R2
2
We can write N2L for the translational motion of the yo-yo:

2M g − T = 2M a

Taking torques about the centre of the yo-yo, we can write Euler’s Rotation Equation for the
rotational motion of the yo-yo:
T r = Iα = M R2 α
The last equation we need is the non-slip condition to relate a and α. Since the ICOR in this
case is the point on the yo-yo where the string is attached (at a distance r away from the centre),
the non-slip condition is
a = rα
Combining all the results above, you should eventually get
2r2 2M R2
a= g, T = g
2r2 + R2 2r2 + R2

9
FIZIKA SPhO Handouts

1.3.3 Rotational Kinetic Energy, Work and Power

When objects rotate, there is a kinetic energy associated with rotation.

Since there is a kinetic energy of Ktrans = 12 mv 2 associated with translation, by taking the
rotational analogues of m and v, the rotational kinetic energy is defined as
1
Krot := Iω 2 (22)
2
More generally, for an object that is rolling, we have
1 2 1
K = Ktrans + Krot = mvCM + ICM ω 2 (23)
2 2
Remark. To avoid confusion, always consider I and v of the CM when calculating total KE!

For a rolling body, the instantaneous power by a force F causing a torque τ is

Pinstantaneous = F · vCM + τCM · ω (24)

and the work done is

Z Z
W = Pinstantaneous dt = (F · dR + τCM · dθ) (25)

In Equations (24) and (25), the first term is referred to as translational work/power and the
second term is referred to as rotational work/power. Don’t forget the rotational work/power!

In rotation, the work-energy theorem also holds if we account for Krot . We have

Wtotal = ∆K = ∆Ktrans + ∆Krot (26)

Now, let’s look at an example where we must consider the rotational kinetic energy.
Example 1.9 (200 Puzzling Physics Problems). A fire hose of mass M and length L is coiled
R. The hose is sent rolling along level ground, with its centre of mass given
into a roll of radius √
initial speed v0 >> gR. The free end of the hose is held fixed. The hose unrolls and becomes
straight. How long does this process take to complete?

√
First, let’s rationalise what v0 >> gR means. Since v0 is large, this simply means that changes
in GPE are negligible compared to KE. Thus, when writing COE, we may drop all GPE terms.

Thus, accounting for the rotational kinetic energy, the COE equation is
1 1 1 1 1 1 2
   
Ktrans,i + Krot,i = Ktrans,f + Krot,f ⇒ M v02 + M R2 ω02 = mv 2 + mr ω 2
2 2 2 2 2 2
The non-slip conditions give v0 = Rω0 and v = rω, hence we can simplify the above equation:
s
3 3 M
M v02 = mv 2 ⇒ M v02 = mv 2 ⇒ v= v0
4 4 m

10
FIZIKA SPhO Handouts

Here, m is the mass of the portion of the hose that is still rolling. After the hose travels a
distance x, by proportionality, we have
x
 
m(x) = M 1 −
L
Substituting this into the expression for v, we have
r Z t Z Lr
dx v0 1 x 1 x 2L
v= =q ⇒ dt = 1 − dx ⇒ dt = 1− dx ⇒ t=
dt 1− x v0 L 0 v0 0 L 3v0
L

Remark. You might ask - why is energy conserved, if the hose is inelastically sticking to the
ground as it unrolls? This process actually dissipates no energy, because the circular part of the
hose is rolling without slipping, so the bottom part is the ICOR and has no velocity, hence no
work is done against friction!

1.3.4 Rotational Collisions

A useful quantity in rotational collisions is the angular impulse, Jθ . (There is actually no
universal symbol for it.) It is the rotational analogue of linear impulse, and is defined as
Z
Jθ = τ dt (27)

There also exists the angular impulse-momentum theorem:

Jθ = ∆L (28)

In rotational collisions, we usually deal with translation and rotation separately, to write separate
equations for COM and COAM. If the collision is elastic, we can also write COE.
Example 1.10 (Kleppner & Kolenkow). A rigid, massless rod of length L joins two particles,
each of mass m. The rod lies on a frictionless table and is struck by a particle of mass m and
velocity v0 as shown. After an elastic collision, the particle moves straight back. Find the
angular velocity of the rod about its CM after the collision.

Firstly, the MOI of the rod + 2 particles about its CM is

 2 !
L 1
I=2 m = mL2
2 2

Many quantities are conserved here. Suppose the projectile moves backwards with speed v1 , the
CM speed of the rod is v2 , and its angular velocity about its CM is ω.

By COM, we have
mv0 = −mv1 + 2mv2 ⇒ v0 + v1 = 2v2

11
FIZIKA SPhO Handouts

By COAM about the axis parallel to v0 and passing through the initial CM of the rod, we have
mv0 L 1 mv L √
√ = mL2 ω − √1 ⇒ v0 + v1 = 2Lω
2 2 2 2 2
By COE, we have
1 1 1 1 1
   
mv02 = mv12 + 2 mv22 + mL2 ω 2 ⇒ (v0 − v1 ) (v0 + v1 ) = 3v22
2 2 2 2 2
The COM and COE equations should be relatively simple to write. However, for COAM, you
need to be careful of the signs of each term! You should go back to Equations (17) and (20) to
determine the direction of the vectors, and assign different signs to different directions.

Now, solving the three simultaneous equations above, we get v2 = 47 v0 , and hence
√ √
2v2 4 2 v0
ω= =
L 7 L
Remark. You might ask - why do we take COAM about this axis? Why not take it, for instance,
about the point of collision? Actually, both are equally valid! However, the bodies that should
be in your system are different (to ensure τ = 0 on your system):
1. About the centre of the rod: Both the moving particle and the rod + 2 particles
2. About the collision point: Only the rod + 2 particles

1.3.5 Physical Pendulums

In oscillations, you learnt that for a simple pendulum,
r s
g L
ωsimple = , Tsimple = 2π (29)
L g
However, this only applies if the pendulum bob is a point mass, and if the string connecting the
pivot to the pendulum bob is light! Instead, for a general rigid body swinging freely about some
pivot point, we have a physical pendulum.

We can use Equation (21) on the torque due to gravity. Assuming θ << 1,
mgd
mgd sin θ = −Ipivot θ̈ ⇒ θ̈ + θ=0 (30)
Ipivot
where we take torques and calculate the MOI about the pivot O. The negative sign is essential
as the torque due to gravity decreases θ.

12
FIZIKA SPhO Handouts

This is the form of a SHM, with

s s
mgd Ipivot
ωphysical = , Tphysical = 2π (31)
Ipivot mgd

Using the parallel axis theorem, you can also write Ipivot = ICM + md2 to simplify this.
Example 1.11 (Ricardo). A physical pendulum consists of a disk of radius R and mass md ,
fixed at the end of a rod of mass mr and length l. (i) Find the period of the pendulum. (ii) How
does the period change if the disk is mounted to the rod by a frictionless bearing so that it is
free to spin?

(i) The MOI of the physical pendulum about its pivot is

1 1
 
Ipivot = Irod + Idisk = mr l2 + md R2 + md l2
3 2
where we have used the parallel axis theorem for the MOI of the disk.

Let θ be the angle of the rod with respect to the vertical. As usual, we assume θ << 1.
The net torque due to gravity about the pivot is
1
 
τ= mr + md glθ
2
Thus, being careful with signs, we write Euler’s Rotation Equation:
v
u1

1 1

1
  u mr l2 + 1 md R2 + md l2
mr l2 + md R2 + md l2 θ̈ = − mr + md glθ ⇒ T = 2π u 3  2 
3 2 2 1
t
2 mr + md gl

(ii) The difference here is that since the disk is free to rotate, we lose the 12 md R2 term. In other
words, we are essentially treating the disk as a point mass now.

With this correction, we instead have

v v 
u 1
u 1
2
u mr l + md l 2 u mr + md l
3 u 3
T = 2π u
t1  = 2π t  
1
2 mr + md gl 2 mr + md g

Remark. This example shows you a subtlety in calculating MOI. When we add up MOIs like
we did in (i), we must ensure that all the components we are adding up are rigidly connected
(i.e. we can treat the whole body as one rigid body).

13
FIZIKA SPhO Handouts

1.4 Ideas
Many tricky mechanics problems involve the use of the following ideas.

1.4.1 Massive Pulleys

In dynamics, we always made an assumption that all pulleys were massless (and small). However,
that is not the case in real life!

When we consider rotation, we usually model pulleys as uniform disks of finite mass and
radius. Consequently, the net force and torque on them need not be zero!
Example 1.12. Consider a pulley system with masses m1 and m2 , with m2 < m1 . The mass of
the pulley is M and its radius is R. Assuming the pulley and string doesn’t slip, determine the
acceleration of the masses.

The first, most important thing for you to realise is that the tension in the string is not
constant! This allows for a non-zero net torque on the pulley, allowing it to rotate.

Thus, writing N2L for each mass,

m1 g − T1 = m1 a, T2 − m2 g = m2 a
and taking torques about the centre of the pulley,
1
(T1 − T2 ) R = M R2 α
2
Combining with the non-slip condition a = Rα, we eventually get
m1 − m2
a= g
m1 + m2 + 21 M

1.4.2 Where Does Normal Force Act?

When it comes to rigid bodies of finite size, there can be some confusion on where exactly the
normal force acts. The example below illustrates this.
Example 1.13 (Ricardo). Consider a solid cube of mass m and dimensions b × b × b moving
linearly on the rough horizontal ground with initial horizontal velocity v0 . An external force F
acts in the same direction as the initial velocity at its edge, as shown below. The coefficient of
friction between the cube and the ground is µ. Determine the maximum value of F before the
cube topples.

14
FIZIKA SPhO Handouts

Intuitively, you might draw the FBD of the cube as such:

However, this makes no sense! If we take torques about the CM, then the forces mg and N
contribute no torque, while F and f both contribute clockwise torques. This would imply that
any value of F , no matter how small, would cause a non-zero clockwise torque about this point,
causing the tube to topple!

The correct FBD considers a different point of action for N . Suppose that the point of
action of N is shifted by some distance x (to the right, so that the torque by N opposes the
torque by F and f ).

From N2L, we know that

N = mg, F − µN = ma
Now, taking torques about the CM,
(F + f ) b (F + µmg) b F b
 
Nx = ⇒ mgx = ⇒ x= +µ
2 2 mg 2
Clearly, the point of action of N must still remain in the cube. Thus,
b
x≤ ⇒ F ≤ mg (1 − µ) ⇒ Fmax = mg (1 − µ)
2
Remark. This example shows you that if you wrongly assume the position at which your
normal force acts, you may get nonsensical results (in this case, it led us to conclude that even
the smallest force would lead to the cube toppling, which we know is not true from daily life).
In such a case, rethink where the normal force should act!

1.4.3 Inextensibility Constraints

When dealing with rotation problems involving rods, the inextensibility (or rigidity) of the rods
gives rise to an additional constraint equation.
Example 1.14 (USAPhO 1999). Consider two uniform rods each of mass M and length L in
a "V" shape, whose angle can vary. The rods are released from rest when their angle θ to the
horizontal is 45◦ . All hinges are frictionless and have negligible mass. The left end is fixed, and
the right end is free to slide on the horizontal surface without friction. Find the upward force
FN that the horizontal surface exerts on the right end just after release.

15
FIZIKA SPhO Handouts

We first label all forces with red and all accelerations with blue:

The motion of the 2 rods will be symmetric (we don’t expect the angles θ to be different). To
write our constraint equation, define the fixed left end as the origin. Then,

xCM,2 = 3xCM,1 ⇒ ax,2 = 3ax,1

yCM,1 = yCM,2 ⇒ ay,1 = ay,2

by geometry and then differentiating with respect to time. These are the constraint equations.

Additionally for the left rod, as θ = 45◦ , the symmetry gives that

ax,1 = ay,1

Also, the magnitude of a is

q αL αL
a= a2x,1 + a2y,1 = ⇒ ax,1 = ay,1 = √
2 2 2
Now, we can proceed with writing N2L for the right rod:
3αL αL
Fx = M ax,2 = M √ , M g + Fy − FN = M ay,2 = M √
2 2 2 2
Taking torques about the hinge at the fixed left end of the left rod,
1 M gL Fx L Fy L
M L2 α = √ − √ − √
3 2 2 2 2
Taking torques about the CM of the right rod,
1 FN L Fy L Fx L
M L2 α = √ + √ − √
12 2 2 2 2 2 2
We can solve these simultaneous equations to eventually get
7
FN = Mg
10

16
FIZIKA SPhO Handouts

2 Problems
Problems are arranged in roughly increasing difficulty.
Problem 2.1 (SPhO 2008). A hungry bear weighing 700 N walks out on a beam in an attempt
to retrieve a basket of food hanging at the end of the beam. The beam is uniform, weighs 200 N,
and is 6.00 m long; the basket weighs 80.0 N. (i) When the bear is at x = 1.00 m, find the tension
in the wire and the components of the force exerted by the wall on the left end of the beam. (ii)
If the wire can withstand a maximum tension of 900 N, what is the maximum distance the bear
can walk before the wire breaks?

Solution. (i) Clearly, since the bear is not accelerating in the vertical direction, the normal force
exerted on the beam by the bear has to balance the weight of the bear Wbear . Similarly, the
tension force exerted on the beam by the string holding the basket has to balance the weight of
the basket Wbasket . Hence, since the beam is stationary, we can equate the anticlockwise and
clockwise torques about the pivot of the beam to get
l
T l sin θ = Wbear x + Wbeam + Wbasket l
2
x 1
W bear + 2 beam + Wbasket
W
T = l = 343 N.
sin θ
(ii) Using our previously obtained equation, we can solve for the maximum distance x.

Tmax l sin θ − Wbeam 2l − Wbasket l

xmax = = 5.14 m.
Wbear

Problem 2.2 (SPhO 2005). A 15.0 m uniform ladder of mass 50 kg rests against a frictionless
wall and makes an angle of 60◦ to the horizontal. (i) Determine the horizontal and vertical forces
that the ground exerts on the base of the ladder when an 800 N firefighter is 4.00 m along the
ladder from the bottom. (ii) If the ladder is just on the verge of slipping when the firefighter is
9.00 m up, what is the coefficient of static friction between the ladder and the ground?
Solution. Considering the horizontal direction, the normal force by the wall must balance the
frictional force by the ground f . Hence, considering the torque about the contact point on the
ground,
l
f l sin θ = mg cos θ + W x cos θ
2
1
mg + xl W
f= 2 = 265 N
tan θ
which is the desired horizontal force. Then, considering the vertical direction,

Nground = mg + W = 1290 N

17
FIZIKA SPhO Handouts

which is the desired vertical force.

(ii) When the firefighter is now 9.00 m up,
1
2 mg + xl W
f= = 418.72 N
tan θ
Then, if the ladder is just on the verge of slipping, then

f = µNground
f
µ= = 0.324.
Nground

Problem 2.3 (SPhO 2010). An electric motor turns a flywheel through a drive belt that joins
a pulley on the motor and a pulley that is rigidly attached to the flywheel, as shown in the
figure. The flywheel is a solid disk with a mass of 80.0 kg and a diameter of 1.25 m. It turns on a
frictionless axle. Its pulley has a much smaller mass and a radius of 0.230 m. If the tension in the
upper, taut segment of the belt is 135 N and the flywheel has a clockwise angular acceleration of
1.67 rad s−2 , find the tension in the lower, slack segment of the belt.

Solution. Firstly, note that the tension in the belt is not constant, as this tension difference is
required for a net torque for the flywheel to have an angular acceleration!

Let the mass of the flywheel be M , radius of the flywheel be R and the radius of the pul-
ley be r. Consider torques about the centre of the flywheel. Writing Euler’s Rotation Equation,
1
τnet = Ttop r − Tbottom r = (Ttop − Tbottom ) r = Iα = M R2 α
2
We may hence calculate Tbottom :

M R2 α
Tbottom = Ttop − = 21.6 N
2r

Problem 2.4 (SPhO 2014). A yo-yo is pulled by its string along a horizontal surface without
slipping. The horizontal velocity at the end of the string remains equal to v. A bar is pivoted as
shown and remains supported by the yo-yo. Find the angular speed of the bar, ω, as a function
of the angle θ. The outer and inner radii of the yo-yo are R and r respectively.

18
FIZIKA SPhO Handouts

Solution. The speed of the yo-yo is

R
vcm = Rωyo-yo = v
R+r
since it rolls without slipping. Then, using some geometry, we can equate θ with the horizontal
distance x of the center of the yo-yo to the pivot as
θ R
tan =
2 x
Differentiating both sides and realising the rate of change of x is vcm and the rate of change of θ
is −ω,
1 θ dθ R dx
 
sec2 =− 2
2 2 dt x dt
1 θ 1 θ R
 
sec2 (−ω) = − tan2 v
2 2 R 2R+r
2v θ
ω= sin2
R+r 2
v
= (1 − cos θ).
R+r
where the last expression is obtained by a trigonometric identity, just to simplify the expression.
Problem 2.5 (SPhO 2012). A pendulum bob is constructed by taking a thin, uniform density
circular ring of mass M and radius R and affixing a straight, thin uniform rod of mass m and
length 2R across its diameter as shown in the diagram. The pendulum hangs in a vertical plane
from a frictionless pivot that can be attached to the ring at any point. The pivot allows the
pendulum to swing either in the plane of the bob or in the plane perpendicular to the bob.
Assume the angular amplitude is small. (i) What swinging configuration(s) give the maximum
period? (ii) What swinging configuration(s) give the minimum period? (iii) Find the ratio of
the maximum to the minimum period.

Solution. Let the moment of inertia through the centre of mass about the axis chosen be Icm . We
have that by the rotational analogue of Newton’s 2nd law, using the small angle approximation
sin θ ≈ θ and the parallel axis theorem,
τ = Iα
 
−(M + m)gR sin θ = Icm + (M + m)R2 θ̈
(M + m)gR
θ̈ ≈ − θ
Icm + (M + m)R2
Hence, the oscillation period is
s
Icm + (M + m)R2
T = 2π
(M + m)gR

19
FIZIKA SPhO Handouts

which increases with Icm , so the maximum and minimum period occur with the maximum and
minimum Icm respectively.
Due to symmetry, the centre of mass lies at the centre of the ring, which coincides with the
centre of the rod. Therefore, the moment of inertia about the centre of mass through an axis
perpendicular to the plane of the bob is
1
Iperp = M R2 + m(2R)2
12
1

= M + m R2
3

On the other hand, the moment of inertia about the centre of mass through an axis passing
through the centre and in the plane of the bob depends on the angle of the axis. Clearly,
the moment of inertia of the ring has no effect on the maximum and minimum angles due to
rotational symmetry. The moment of inertia of the ring about its diameter is
1
Iring,diam = M R2
2
since by perpendicular axis theorem and rotational symmetry, twice its value should be the
moment of inertia about an axis through its centre perpendicular to its plane, M R2 . Then, the
maximum moment of inertia for this configuration occurs when the axis is perpendicular to the
rod, when the rod contributes the maximum inertia,
1 1
Iplane (max) = M R2 + m(2R)2
2 12
1 1

= M + m R2
2 3
and the minimum moment of inertia for this configuration occurs when the axis is parallel to
the rod, when the rod contributes zero inertia,
1
Iplane (min) = M R2 .
2
Therefore, comparing the moments of inertia across all three cases, (i) the maximum period
occurs when swinging in the plane of the bob, (ii) the minimum period occurs when swinging in
the plane perpendicular to the bob with the pivot point on the rod, and (iii) the ratio of the
maximum and minimum period is
s
Tmax Imax + (M + m)R2
=
Tmin Imin + (M + m)R2
v s
u 2M + 4 m
u
3 12M + 8m
=t 3 = .
2M +m 9M + 6m

Problem 2.6 (Ricardo). Compute the MOI of a uniform solid cone of mass M , base radius R
and height H about (i) its symmetry axis (ii) the diameter of its base.
M 3M
Solution. The density of the cone is ρ = π 2
R H
= πR2 H
. We proceed by integration.
3

(i) A circular cross section of the cone parallel to the base with distance from the vertex x
R
and thickness dx has radius H x (by similar triangles), and so a moment of inertia about the

20
FIZIKA SPhO Handouts

symmetric axis of
1
dIsym = dm r2
2
1
= ρπr4 dx
2
ρπR4 4
= x dx
2H 4
Integrating,

ρπR4 4
Z H
Isym = 4
x dx
0 2H
ρπR4 H 5
=
2H 4 5
3M πR4 H
=
πR2 H 10
3
= M R2 .
10
(ii) Considering the same cross section as before, the moment of inertia of the cross section about
its diameter is
1 1
dIdiam, cm = · dm r2
2 2
1
= dm r2
4
since by perpendicular axis theorem and rotational symmetry, twice its value should be the
moment of inertia about an axis through its centre perpendicular to its plane, 12 dm r2 . Then,
by using parallel axis theorem,
1
dIdiam = dm r2 + dm (H − x)2
4
1 2
 
= r + (H − x)2 ρπr2 dx
4
!
R4 4 R2 2 2
= ρπ x + 2 (H − x) x dx
4H 4 H
! !
R4 R2 4 2R2 3
= ρπ + x − x + R2 x2 dx
4H 4 H 2 H

Integrating,
! !
R4 R2 2R2 3
Z H
4
Idiam = ρπ + x − x + R2 x2 dx
0 4H 4 H 2 H
! !
R4 R2 H 5 2R2 H 4 2H
3
= ρπ + − + R
4H 4 H 2 5 H 4 3
3M 1 4 1
 
= 2
π R H + R2 H 3
πR H 20 30
3 1
= M R2 + M H 2 .
20 10

Problem 2.7 (SPhO 2003). A marble rests on the top of a hemispherical bowl as shown in
the figure below. The radii of the bowl and the marble are R and a respectively. The marble is

21
FIZIKA SPhO Handouts

displaced from rest and it rolls down the bowl without slipping. At what height, in terms of R
and a, will the marble leave the hemisphere?

Solution. Let θ, Ω be the angle and angular velocity of the centre of the marble measured from
the centre of the bowl, with θ = 0 at the top and increasing in the direction of rolling, and let ω
be the angular velocity of the marble measured from its centre. Since the marble rolls without
slipping,

aω = (R + a)Ω
R+a
ω= Ω
a
After rolling an angle of θ, by energy conservation,
1 1
mv 2 + Iω 2 = mg(R + a)(1 − cos θ)
2 2
2
1 1 2 R + a
 
ma2 + · ma2 Ω2 = mg(R + a)(1 − cos θ)
2 2 5 a
7
(R + a)Ω2 = g(1 − cos θ)
10
10g(1 − cos θ)
Ω2 =
7(R + a)

When the marble leaves the hemisphere, the normal force by the hemisphere is zero, so we can
equate the radial component of gravity to centripetal force to get

m(R + a)Ω2 = mg cos θ

10
(1 − cos θ) = cos θ
7
10
cos θ = .
17
Therefore, the height is

h = (R + a) cos θ
10
= (R + a).
17

Problem 2.8 (SPhO 2016). Consider a rigid, axially symmetric wheel with mass M , moment
of inertia I about the axis of symmetry S, and radius R, moving in an upright position on a
horizontal plane. The coefficient of kinetic friction between the wheel and the plane is µk . (i)
The wheel is initially spun about S with constant angular speed ω0 and is then released. It slips
for time Ta and then rolls without slipping. Determine Ta and the centre of mass speed va of the
rolling wheel after it stops slipping. (ii) The wheel has an initial horizontal centre of mass speed
v0 in addition to an initial angular speed ω0 . Determine the time Tb at which the wheel rolls
without slipping and the centre of mass speed vb of the rolling wheel. (iii) The wheel has an

22
FIZIKA SPhO Handouts

initial horizontal centre of mass speed v0 and an initial backspin −ω0 . Determine the possible
motions of the wheel.

Solution. Suppose the initial speed of the centre of mass is v and the initial angular speed is ω
clockwise. By conservation of angular momentum about the instantaneous centre of rotation,
since there is no net torque about that point,

mvR + Iω = mva R + Iωa

I
 
= mR + va
R
mvR + Iω
va =
mR + RI
R(mvR + Iω)
=
mR2 + I
The wheel accelerates due to kinetic friction, so

ma = µk mg
a = µk g

where a denotes the magnitude of acceleration. Then, substituting the initial conditions given,
(i) in the first case
RIω0
va =
mR2 + I
and
va
Ta =
a
RIω0
=
µk g(mR2 + I)

while (ii) in the second case

R(mv0 R + Iω0 )
vb =
mR2 + I
and
|vb − v|
Tb =
a
I |Rω0 − v0 |
=
µk g(mR2 + I)

(iii) In the third case,

R(mv0 R − Iω0 )
vc =
mR2 + I

23
FIZIKA SPhO Handouts

and
vc − v
Tc =
a
I (Rω0 + v0 )
=
µk g(mR2 + I)
From the sign of the numerator, we can infer that there are three possible motions of the wheel:
1. If mv0 R > Iω0 , then the wheel slips and decelerates for time Tc , after which the wheel
rolls without slipping in its initial direction at speed vc .
2. If mv0 R = Iω0 , then the wheel slips and decelerates for time Tc , after which the wheel
stops rolling and comes to a rest.
3. If mv0 R < Iω0 , then the wheel slips and decelerates, comes to an instantaneous rest and
accelerates opposite to its initial direction for a total time Tc , after which the wheel rolls
without slipping opposite to its initial direction at speed −vc .
Problem 2.9 (Ricardo). A uniform solid disk of mass M and radius R rotates with initial
angular speed ω0 about its centre while lying on the ground. The coefficient of friction between
the disk and the ground is µ. (i) Find the total torque acting on the disk about its centre. (ii)
Find the time taken for the disk to stop rotating. (In engineering, this is called disk friction.)
Solution. (i) Let the height of the disk be H, so the density of the disk is ρ = πRM2 H . Consider a
cross-sectional ring of the disk with radius r and thickness dr. The frictional torque on the ring
is

dτ = µk dm gr
= µk grρ(2πrH dr)
= 2πµk ρgr2 H dr

Integrating,
Z R
τ= 2πµk gρr2 H dr
0
R3
= 2πµk ρgH
3
2π M
= µk gHR3
3 πR2 H
2
= µk M gR.
3
(ii) Solving for the angular acceleration,

τ = Iα
2
3 µk M gR
α= 1 2
2MR
4µk g
=
3R
Therefore, the time taken for the disk to stop rotating is
ω0
t=
α
3Rω0
= .
4µk g

24
FIZIKA SPhO Handouts

Problem 2.10 (Morin). A spool consists of an axle of radius r and an outside circle of radius
R which rolls on the ground. A thread is wrapped around the axle and is pulled with tension T
at an angle θ to the horizontal. (i) Given R and r, what should θ be so that the spool doesn’t
move? Assume that the friction between the spool and the ground is large enough so that the
spool doesn’t slip. (ii) Given R, r and the coefficient of friction µ between the spool and the
ground, what is the largest value of T for which the spool remains at rest? (iii) Given R and µ,
what should r be so that you can make the spool slip from the static position with as small a
T as possible? That is, what should r be so that the upper bound on T in (ii) is as small as
possible? What is the resulting value of T ?

Solution. (i) Balancing the horizontal forces, f = T cos θ. Therefore, balancing the torques by
tension and friction about the centre,

fR = Tr
T R cos θ = T r
r
θ = cos−1 .
R
(ii) Balancing the vertical forces,

mg = T sin θ + N
N = mg − T sin θ
r
Since the spool is at rest, cos θ = R. Therefore, on the verge of slipping,

f = µN
Tmax cos θ = µ(mg − Tmax sin θ)
µmg
Tmax =
cos θ + µ sin θ
µmg
= q
r r 2


R +µ 1− R
µmgR
= √ .
r + µ R2 − r 2
(iii)
p Considering the expression for Tmax in terms of θ, the denominator is cos θ + µ sin θ =
−1


2
1 + µ cos θ − tan µ , and so the minimum of Tmax occurs with the maximum of the
denominator, i.e. θ0 = tan−1 µ. Therefore,

tan θ0 = µ
R2
− 1 = µ2
r02
R
r0 = p
1 + µ2

25
FIZIKA SPhO Handouts

where the resulting maximum value of T is

µmg
Tmax,0 = p .
1 + µ2

Problem 2.11 (Ricardo). A uniform solid cylinder of mass M and radius R is rolling without
slipping on the ground with an angular speed of ω0 . It hits an edge of height h and rolls up.
The point of contact does not slip during the up-rolling. (i) Calculate the angular speed of the
cylinder immediately after the collision, assuming the collision time is very short. (ii) Calculate
the minimum value of ω0 such that the cylinder is able to roll up. (Hint: For a very short
collision time, what quantity is conserved? Why?)

Solution. (i) During the collision, the angular impulse by gravity about the pivot point is
negligible, since the collision time is very short. Hence, during the collision, angular momentum
about the pivot is conserved! Equating the angular momentum just before and after the collision,

M v0 (R − h) + Iω0 = M vR + Iω
1 1
M Rω0 (R − h) + M R2 ω0 = M R2 ω + M R2 ω
2 2
3
R − h
ω= 2 3 ω0
2R
3R − 2h
= ω0 .
3R
(ii) For the cylinder to just be able to roll up, the kinetic energy of the cylinder after the collision
must just be able to overcome the gravitational potential energy at the top, without any kinetic
energy remaining. Therefore,
1 1
M v 2 + Iω 2 = M gh
2 2
1 1 1
M R2 ω 2 + · M R2 ω 2 = M gh
2 2 2
4gh
ω2 =
3R2
(3R − 2h) 22 4gh
ω0,min =
9R2 3R√2
2 3gh
ω0,min = .
3R − 2h

Problem 2.12 (Morin). Consider the following "car" on an inclined plane. It is made out of a
board (a thin sheet) of mass m and two cylinders, each of mass m 2 and radius R. If it is released
from rest, and there is no slipping between any surfaces, find the acceleration of the board. (You
should find that the acceleration of the board could exceed g if θ can be freely varied between 0◦
and 90◦ . Can you rationalise why?)

26
FIZIKA SPhO Handouts

Solution. Rather than analysing forces, which can be complicated and tedious, we shall explore
the use of conservation of energy.
Clearly, both wheels have the same velocity at any point in time. Suppose at some time, the
speeds of both wheels are v, their accelerations are a and they have travelled a distance of x.
Since the wheels are rolling without slipping and the board is not slipping on the wheel, the
speed of the board is 2v, and so its acceleration is 2a and it has travelled a distance of 2x.
Therefore, considering the kinetic and gravitational potential energies and the conservation of
energy,
m 1m 2 1 2 1
 
2· gx sin θ + mg(2x) sin θ = 2 · v + Iω + m(2v)2
2 2 2 2 2
1 1 m
3mgx sin θ = mv 2 + R2 ω 2 + 2mv 2
2 2 2
11
3gx sin θ = v 2
4
Now, we can differentiate the conservation of energy equation with respect to time to get
dx 11 d  2 
3g sin θ = v
dt 4 dt
11 dv
3gv sin θ = v
2 dt
11
3g sin θ = a
2
6
a = g sin θ
11
and so the acceleration of the board is
12
aboard = 2a = g sin θ.
11
Notice that if sin θ > 11
12 , the acceleration of the board exceeds g. This is physically possible
since there is no slipping between any surfaces, and so static fraction between the wheels and
the board provides an extra downward component to the net force, allowing it to accelerate
faster than g.
This example illustrates that conservation of energy can be used to find accelerations through
the use of differentiation, and shall see more examples of this in the future solutions.
Problem 2.13 (200 Puzzling Physics Problems). Two identical dumbbells move towards each
other on a frictionless table as shown. Each consists of two point masses m joined by a massless
rod of length 2l. The dumbbells collide elastically. (i) Find the angular velocity of the rods right
after collision. (ii) Describe the subsequent evolution of the system.

27
FIZIKA SPhO Handouts

Solution. (i) Suppose the dumbbells reverse directions, with their centres travelling at speed v ′
(by symmetry) and rotating clockwise around their centres at angular velocity ω. Since the only
force acts between the two colliding masses, we can consider conservation of angular momentum
at the collision point to get
 
2(2mvl) = 2 2ml2 ω − 2mv ′ l
v + v ′ = lω

Then, considering conservation of energy,

1 1 1
   
2 (2m)v 2 = 2 (2m)l2 ω 2 + (2m)v ′2
2 2 2
2 ′2 2 2
v −v =l ω

Dividing the two equations obtained, we get v − v ′ = lω, so v ′ = 0, i.e. the dumbbells only
rotate after the collision, with angular velocity
v
ω= .
l
(ii) After the dumbbells collide, their centres stop moving and they both rotate clockwise at
angular velocity ω. After rotating half a revolution, they will collide again. By time reversibility,
they will stop rotating and start moving in their original directions at speed v as if they had
never collided, except flipped upside down.
Problem 2.14 (Ricardo). A uniform rod of mass m and length l is placed on a horizontal table
top with its CM a distance a from the perpendicular edge. The rod is released from rest from a
horizontal position and begins to rotate about the edge of the table. The coefficient of friction
between the rod and the table is µ. (i) Find the maximum angle that the rod makes with the
horizontal before slipping begins. (ii) Now, let only a quarter of the rod’s length be on the table.
What angle does the rod make with the horizontal when it starts to slip? The coefficient of
7
static friction between the rod and the table is µ = 13 .

28
FIZIKA SPhO Handouts

Solution. (i) By conservation of energy,

1 1
mv 2 + Iω 2 = mga sin θ
2 2
1 1 1
ma2 ω 2 + ml2 ω 2 = mga sin θ
2 2 12
2ga sin θ
ω2 = 2 1 2
a + 12 l
24ga sin θ
=
12a2 + l2
This can be differentiated with respect to time to obtain
d  2 24ga d
ω = (sin θ)
dt 12a2 + l2 dt
dω 24ga cos θ dθ
2ω =
dt 12a2 + l2 dt
12ga cos θ
ωα = ω
12a2 + l2
12ga cos θ
α=
12a2 + l2
which could also be obtained by considering the gravitational torque about the pivot. Now,
considering the tangential acceleration,

maα = mg cos θ − N
12ga cos θ
 
N = m g cos θ − a
12a2 + l2
mgl2 cos θ
=
12a2 + l2
Finally, considering the centripetal acceleration at the verge of slipping,

maω 2 = µN − mg sin θ
24ga sin θ mgl2 cos θ
ma 2 + l2
= µ − mg sin θ

12a 12a2 + l2
36a2 + l2 sin θ = µl2 cos θ
µl2
θ = tan−1 .
36a2 + l2
l 7
(ii) Substituting a = 4 and µ = 13 , we have
7 2
13 l
θ = tan−1  2
36 4l + l2
28
= tan−1 .
169

Problem 2.15 (Kleppner & Kolenkow). A uniform plank of length 2L leans nearly vertically
against a wall. The wall and the ground are frictionless. The plank starts to slip downwards.
Find the height of the top of the plank when it loses contact for the first time (either the wall or
the floor).

29
FIZIKA SPhO Handouts

Solution. Let the angle the ladder makes with the vertical be θ, so by some simple geometry,
the angle the line from the centre of mass to the intersection of the wall and the ground makes
with the vertical is also θ. The angular velocity of the plank is ω and the speed of the centre of
mass is Lω. By conservation of energy,
1 1
mv 2 + Iω 2 = mgL(1 − cos θ)
2 2
1 1 1
mL2 ω 2 + m(2L)2 ω 2 = mgL(1 − cos θ)
2 2 12
3g
ω2 = (1 − cos θ)
2L
Differentiating with respective to time, we have
d  2 3g d
ω = (1 − cos θ)
dt 2L dt
dω 3g dθ
 
2ω = sin θ
dt 2L dt
3g
ωα = ω sin θ
4L
3g
α= sin θ
4L
The plank loses contact with the wall when the normal force by the wall vanishes. Since the
normal force by the wall is the only horizontal component to the net force, when the ladder slips
from the wall, the horizontal acceleration ax = 0. Denoting the horizontal distance of the centre
of mass to the wall as x, using some geometry and differentiating the resulting expression, we get

x = L sin θ
dx dθ
= (L cos θ)
dt dt
2  2
d x d2 θ dθ
2
= (L cos θ) 2 − (L sin θ)
dt dt dt
ax = (L cos θ)α − (L sin θ)ω 2

Equating the horizontal acceleration to be zero when the plank slips from the wall, we have

α cos θ = ω 2 sin θ
3g 3g
sin θ cos θ = (1 − cos θ) sin θ
4L 2L
2
cos θ =
3
Similarly, the plank loses contact with the floor when the normal force by the ground vanishes.
The vertical components to the net force are from the normal force by the ground and the
gravitational force, so when the ladder slips from the ground, the vertical acceleration ay = −g.
Similarly denoting the vertical distance of the centre of mass to the ground as y, we get

y = L cos θ
dy dθ
= −(L sin θ)
dt dt
 2
d2 y d2 θ dθ
= −(L sin θ) 2 − (L cos θ)
dt2 dt dt
2
ay = −(L sin θ)α − (L cos θ)ω

30
FIZIKA SPhO Handouts

Equating the vertical acceleration to be −g when the plank slips from the ground, we also have
g
α sin θ + ω 2 cos θ =
L
3g 2 3g g
sin θ + (1 − cos θ) cos θ =
4L 2L L
9 cos2 θ − 6 cos θ + 1 = 0
1
cos θ =
3
Comparing both cases, the first case has the lower value of θ, so we finally have that the plank
loses contact with the wall first, and the height of the top of the plank when that happens is
4
H = 2L cos θ = L.
3

31
FIZIKA SPhO Handouts

3 Advanced Problems
These problems are way too difficult to be tested in a modern-day SPhO. If you have completed
all the previous problems and are down for a challenge, try these!
Problem 3.1 (SPhO 2011, Morin). (i) A stick of mass density per unit length ρ rests on a
circle of radius R. The stick makes an angle θ with the horizontal and is tangent to the circle at
its upper end. Friction exists at all points of contact, and assume that it is large enough to keep
the system at rest. Find the friction force between the ground and the circle.

(ii) A large number of identical sticks and circles lean on each other, as shown in the figure.
Each stick makes an angle θ with the horizontal and is tangent to a circle at its upper end. The
sticks are hinged to the ground, and every other surface is frictionless (unlike in (i)). What is
the normal force between a stick and the circle it rests on, very far to the right? (Assume the
last circle is fixed and cannot move.)

Solution. (i) Let N be the normal force between the stick and the circle, and Ff be the frictional
force between the ground and the circle. For rotational equilibrium, the frictional force between
the stick and the circle must also be Ff , for the torques to cancel.

Considering torques on the stick about the point of contact with the ground as the pivot,
l 1
 
Mg cos θ = N l ⇒ N = ρlg cos θ
2 2
where M = ρl is the mass of the stick and l is its length.

Balancing the horizontal forces on the circle gives

N sin θ ρlg sin θ cos θ
N sin θ = Ff + Ff cos θ ⇒ Ff = =
1 + cos θ 2 (1 + cos θ)

32
FIZIKA SPhO Handouts

 
θ
From the geometry of the set-up, l = R cot 2 . Doing some manipulation, we obtain
1
Ff = ρRg cos θ
2
(ii) Let’s be systematic about how we label our sticks and circles, to avoid confusion. Let Si be
the ith stick, and let Ci be the ith circle.

The normal forces that Ci feels from Si and Si+1 must be equal in magnitude, because the circle
is massless and frictionless (so this is the only way it can stay in equilibrium).

Considering torques on Si+1 about the point of contact with the ground as the pivot,
1 θ 1 θ
   
M gl cos θ + Ni l tan2 = Ni+1 l ⇒ M g cos θ + Ni tan2 = Ni+1
2 2 2 2
By definition, N0 = 0. As per (i), N1 = 12 M g cos θ. Applying this recurrence relation, we can
easily see that
1 θ θ θ
      
Ni = M g cos θ 1 + tan2 + tan 4
+ ... + tan 2(i−1)
2 2 2 2
In the limit i → ∞, we may write the infinite geometric series as
 
1 1
N∞ = Mg   
2 1 − tan2 θ 2

We may finally replace M with ρ using M = ρl, and doing some manipulation, we obtain
 
θ
ρRg cos3 2
N∞ =  
θ
2 sin 2

Problem 3.2 (EuPhO 2024). A puck (a small disc) with radius r and uniform density is
moving on a horizontal plane with the velocity v0 without rotation. The puck meets a fixed
half-circular wall of radius R >> r and starts to move along the wall. The coefficient of friction
with the wall is µ, and friction with the horizontal plane is negligible. (i) Find the velocity of the
puck ve when it leaves the wall. (ii) Sketch the graph of ve (µ). Indicate all important features.

33
FIZIKA SPhO Handouts

While this is a EuPhO problem, don’t be intimidated! Many competitors received full credit on
this problem (T1) that year, making it easy for EuPhO standards.
Solution. You may check the official solutions here (T1). Don’t forget to consider both cases of
exiting while still slipping and exiting while not slipping!
Problem 3.3 (Ricardo). A solid cylinder of radius R rolls without slipping with CM speed v0
on a horizontal ground. At point O, it rotates with point O as the rotational axis, and continues
rolling without slipping on an inclined plane of angle α afterwards. Determine the maximum
value of v0 such that the cylinder does not lose contact with point O.

Solution. When the cylinder is rotating about O, let the angle the line from the centre to the
pivot makes with the vertical be θ. By conservation of energy,
1 1 1 1
mv 2 + Iω 2 = mv02 + Iω02 + mgR(1 − cos θ)
2 2 ! 2 2
1  2  11 v 2 v 2
m v − v02 + mR2 − 0 = mgR(1 − cos θ)
2 22 R2 R2
4
v 2 = v02 + gR(1 − cos θ)
3
At the maximum value of v0 , at some point in its motion, the normal force by the inclined plane
vanishes. Therefore, considering the centripetal force at that point in time,

mv 2
= mg cos θ
R
v02 4
+ g(1 − cos θ) = g cos θ
R 3 r
1
v0 = gR(7 cos θ − 4)
3
Finally, the minimum of this expression occurs at the maximum θ, which is clearly α, so the
maximum v0 is
r
1
v0,max = gR(7 cos α − 4).
3

34
FIZIKA SPhO Handouts

4 Appendix
4.1 More Methods of Calculating MOI
There are many elegant ways you can calculate MOI, other than the methods introduced in
Section 1.2. Here are a few research papers you can read to find out more!
1. Scaling Method
2. Squashing Method
3. Three-Axis Theorem
4. Analysis of Differential Elements
While you don’t need to know all of these methods for Physics Olympiad, hopefully these give
you some insight into how different mass distributions from a given axis influence the MOI!

35