3.4 Dynamics of Circular Motion
Moving on a circle needs a net force $mv^2/r$ towards the centre. Friction, tension, the normal force and gravity supply it — on a flat road, on a banked curve, in a conical pendulum and at the top and bottom of a loop.
A Formula 1 car takes a corner at over $250\ \text{km/h}$; a bus on a hill road crawls round a hairpin bend; a velodrome cyclist leans into a track banked at more than $40^\circ$; a child's swing ride whirls its seats outward on long chains. In Section 2.4 you learnt the kinematics: a body moving with speed $v$ on a circle of radius $r$ has an acceleration $v^2/r$ directed towards the centre, even if its speed never changes. Newton's second law now asks the question that kinematics cannot: what force produces that acceleration?
The answer is never a new, mysterious force. It is always one of the familiar ones — friction, tension, the normal force, gravity — or a combination of their components. In this section you will learn to find which force is doing the job and how large it must be, and so answer the classic JEE Main questions: the fastest safe speed on a level curve, the right banking angle for a highway, the range of speeds a banked road with friction can handle, and the angle of a conical pendulum. Motion in a vertical circle is introduced qualitatively; Chapter 4 completes it with energy.
Centripetal force and its sources
Recall the kinematics. For uniform circular motion of radius $r$, speed $v$ and angular speed $\omega$ (so $v=\omega r$), the acceleration is
$$a_c=\frac{v^2}{r}=\omega^2r,$$directed towards the centre at every instant. By Newton's second law the net force on the body must point the same way.
The centripetal force on a body moving on a circle is the component of the net force directed towards the centre. For a body of mass $m$ moving with speed $v$ on a circle of radius $r$ it must have magnitude
$$F_c=\frac{mv^2}{r}=m\omega^2r .$$"Centripetal" describes a role, not a kind of force: in each problem it is supplied by real forces — tension, friction, the normal force, gravity — or by components of them. It is never drawn as an extra arrow on a free-body diagram.
Resolve all forces on the body along the radial direction (towards the centre) and the tangential direction (along the velocity). Then
$$\sum F_\text{radial}=\frac{mv^2}{r},\qquad\sum F_\text{tangential}=m\frac{dv}{dt}.$$For uniform circular motion the tangential sum is zero; for non-uniform circular motion the net force has both components, and its magnitude is $m\sqrt{(v^2/r)^2+(dv/dt)^2}$. Any force perpendicular to the plane of the circle must balance (if the circle stays in a fixed plane).
Who supplies the centripetal force? A stone whirled on a string: the tension. A car on a flat bend: static friction from the road on the tyres. A train on a curved track: the sideways push of the rails on the wheel flanges. The Moon around the Earth: gravity. A rider on the wall of a spinning "rotor" drum: the normal force of the wall. A plane banking to turn: the horizontal component of the lift on its wings.
What happens when the force fails. If the string breaks, nothing pulls the stone towards the centre any more, so by the first law it moves off along the tangent with the speed it had — not radially outward. Mud flies off a bicycle tyre tangentially for the same reason.
The view from the rotating frame. A passenger in a turning car feels "thrown outward". In the ground frame (an inertial frame) there is no outward force: the passenger's body simply tends to go straight while the car turns under it, and the seat or door pushes them inward. In the frame of the car, which is accelerating, Newton's laws work only if we add a pseudo force $m\omega^2r=mv^2/r$ directed away from the centre — the centrifugal force. Use it only when you deliberately work in the rotating frame; then the body is "at rest" and the real forces plus the centrifugal force add to zero. Never mix the two pictures: either $\sum F=mv^2/r$ inward (ground frame) or $\sum F+F_\text{cf}=0$ (rotating frame).
In a frame rotating with angular speed $\omega$, a body at distance $r$ from the axis appears to feel a centrifugal force $m\omega^2r$ directed radially outward. It is a pseudo force — no other body exerts it, and it has no third-law partner — introduced so that Newton's laws can be used in the non-inertial rotating frame.
A puck on a string moves in a horizontal circle on smooth ice. The number of revolutions per second is doubled, with the same radius. The tension in the string becomes
A vehicle on a level circular road
On a flat road the forces on a car are its weight $mg$ (down), the normal force $N$ (up) and the friction of the road on the tyres (horizontal). The tyres roll without slipping, so this is static friction, and it points sideways, towards the centre of the curve. Vertically $N=mg$; horizontally $f=mv^2/r$. Since $f\le\mu_sN=\mu_smg$:
A vehicle can take a level curve of radius $r$ without skidding provided
$$v\le v_{\max}=\sqrt{\mu_srg},$$where $\mu_s$ is the coefficient of static friction between tyres and road. $v_{\max}$ does not depend on the mass of the vehicle. Above it, friction cannot supply $mv^2/r$ and the vehicle skids towards the outside of the curve. (Here we treat the vehicle as a particle and ignore toppling.)
Because $v_{\max}\propto\sqrt{\mu_s}$, a wet road that halves $\mu_s$ reduces the safe speed by a factor $\sqrt2$ — about $30\%$, not $50\%$. And because $v_{\max}\propto\sqrt r$, a sharper bend (smaller $r$) must be taken more slowly.
A loaded truck and a small car take the same level curve; the coefficient of static friction is the same for both. Compared with the car, the truck's maximum safe speed is
Banked roads
Relying on friction alone is risky — it changes with rain, oil and tyre wear. Engineers therefore bank curves: the outer edge of the road is raised so that the road surface is tilted at an angle $\theta$ towards the centre. Now the normal force, being perpendicular to the tilted road, leans towards the centre, and its horizontal component $N\sin\theta$ can supply the centripetal force on its own.
Without friction. Take axes horizontal (towards the centre) and vertical — not along the slope, because the acceleration is horizontal. With only $mg$ and $N$:
$$N\sin\theta=\frac{mv^2}{r},\qquad N\cos\theta=mg\quad\Longrightarrow\quad\tan\theta=\frac{v^2}{rg}.$$At this one speed $v_0=\sqrt{rg\tan\theta}$ — the design speed or optimum speed — no friction is needed at all, and $N=mg/\cos\theta$, larger than the weight.
With friction. If the car goes faster than $v_0$, it tends to slide up the bank (outward), so friction acts down the slope. Taking $f$ down the slope:
$$N\sin\theta+f\cos\theta=\frac{mv^2}{r},\qquad N\cos\theta-f\sin\theta=mg .$$Solving, the friction needed and the normal force are
$$f=m\left(\frac{v^2}{r}\cos\theta-g\sin\theta\right),$$ $$N=m\left(g\cos\theta+\frac{v^2}{r}\sin\theta\right).$$$f$ is positive (down the slope) for $v\gt v_0$ and negative (up the slope) for $v\lt v_0$. The car holds as long as $|f|\le\mu_sN$. Setting $f=+\mu_sN$ gives the top speed; setting $f=-\mu_sN$ gives the lowest speed below which the car would slide down into the curve.
For a curve of radius $r$ banked at angle $\theta$, with coefficient of static friction $\mu_s$ (vehicle treated as a particle):
$$v_0=\sqrt{rg\tan\theta}\qquad(\text{no friction needed}),$$ $$v_{\max}=\sqrt{rg\,\frac{\tan\theta+\mu_s}{1-\mu_s\tan\theta}},$$ $$v_{\min}=\sqrt{rg\,\frac{\tan\theta-\mu_s}{1+\mu_s\tan\theta}}\quad(\text{and }v_{\min}=0\text{ if }\mu_s\ge\tan\theta).$$The vehicle can take the curve at any speed from $v_{\min}$ to $v_{\max}$. With $\mu_s=0$ both reduce to $v_0$; with $\theta=0$, $v_{\max}=\sqrt{\mu_srg}$ (Result 3.16). If $\mu_s\tan\theta\ge1$ there is no upper limit. None of these depends on the mass.
The dividing form $\dfrac{\tan\theta\pm\mu_s}{1\mp\mu_s\tan\theta}$ is $\tan(\theta\pm\lambda)$, where $\lambda$ is the angle of friction of Section 3.3: $v_{\max}=\sqrt{rg\tan(\theta+\lambda)}$ and $v_{\min}=\sqrt{rg\tan(\theta-\lambda)}$. Friction acts like extra banking of $\pm\lambda$. Railway tracks are banked by raising the outer rail by a height $h$ (the superelevation or cant); for rails a distance $d$ apart, $\sin\theta=h/d$, and for small angles $h\approx d\,\dfrac{v^2}{rg}$.
Keep the bank at $15^\circ$ and set the speed to $v=5\ \text{m/s}$. With $\mu_s=0.20$ the car slides down into the curve. What happens if you raise $\mu_s$ to $0.30$?
A car takes a banked curve at a speed below its design speed $v_0$ and does not slip. The friction of the road on the tyres acts
Conical pendulum and motion in a vertical circle
Tie a small bob to a string of length $L$ and swing it so that it moves in a horizontal circle while the string sweeps out a cone of half-angle $\theta$ about the vertical. This is a conical pendulum. The bob moves on a circle of radius $r=L\sin\theta$ — not $L$. Only two forces act: the weight $mg$ and the tension $T$ along the string. Vertically there is no acceleration; horizontally the net force is centripetal:
$$T\cos\theta=mg,\qquad T\sin\theta=m\omega^2(L\sin\theta).$$A bob on a light string of length $L$ moving uniformly in a horizontal circle with the string at angle $\theta$ to the vertical has
$$T=m\omega^2L=\frac{mg}{\cos\theta},\qquad\cos\theta=\frac{g}{\omega^2L},$$ $$\text{period }\ P=\frac{2\pi}{\omega}=2\pi\sqrt{\frac{L\cos\theta}{g}},\qquad\tan\theta=\frac{v^2}{rg}.$$The motion is possible only for $\omega\gt\sqrt{g/L}$; as $\omega$ grows, $\theta$ approaches (but never reaches) $90^\circ$ and the tension grows without limit. The angle does not depend on the mass. The period depends only on the height $h=L\cos\theta$ of the pivot above the plane of the circle.
Notice that $\tan\theta=v^2/(rg)$ is the same relation as for the frictionless banked road: in both cases a single contact force (tension or normal force), tilted at $\theta$ to the vertical, together with gravity, provides the centripetal force. The same equation describes a cyclist leaning into a flat turn and an aircraft banking to turn — and a ball rolling round inside a smooth bowl.
With the string length $L=1.00\ \text{m}$, set $\omega=3.0\ \text{rad/s}$. At what angle to the vertical does the string settle?
Motion in a vertical circle — a first look. When a body moves round a vertical circle (a stone whirled on a string, a car over a hump, a roller-coaster loop), gravity has a component along the path, so the speed changes from point to point; working out how it changes needs the energy methods of Chapter 4. But at the top and bottom points the radial equation alone tells us a great deal, because there the weight is purely radial:
- At the top (centre below the body), weight and tension (or the track's normal force, for a body on the inside of a loop) both point down, towards the centre: $T+mg=\dfrac{mv^2}{r}$. Since $T\ge0$, the body stays on its circle only if $v_\text{top}\ge\sqrt{gr}$ — the critical speed; at exactly $\sqrt{gr}$ gravity alone provides the centripetal force and the string goes slack.
- At the bottom (centre above), tension points up and weight down: $T-mg=\dfrac{mv^2}{r}$, so $T=mg+\dfrac{mv^2}{r}$ — the string is tightest here, and a pilot pulling out of a dive feels much "heavier".
- Over a hump (car on top of a convex bridge, centre below): $mg-N=\dfrac{mv^2}{r}$, so $N=mg-\dfrac{mv^2}{r}$. The car feels lighter, and leaves the road when $v\gt\sqrt{gr}$ — for a hump of radius $20\ \text{m}$, at $14\ \text{m/s}$ ($50\ \text{km/h}$). In a dip (centre above), $N=mg+\dfrac{mv^2}{r}$.
Chapter 4 shows, using energy, that a body on a string needs at least $\sqrt{5gr}$ at the bottom to get round the top.
1. Drawing "the centripetal force" as an extra arrow. In the ground frame the free-body diagram contains only real forces (weight, normal, tension, friction); their resultant is $mv^2/r$. And do not add a centrifugal force unless you have chosen to work in the rotating frame.
2. Resolving along the slope on a banked road. The acceleration is horizontal, so resolve horizontally and vertically. That gives $N=mg/\cos\theta$ at the design speed — not $mg\cos\theta$, which is the answer for a block resting on an incline.
3. Friction's direction on a bank. Above $v_0$ the car tends to slide up and out, so friction acts down the slope; below $v_0$ it acts up the slope. Check $\tan\theta$ against $\mu_s$ before using $v_{\min}$: if $\mu_s\ge\tan\theta$, $v_{\min}=0$.
4. Wrong radius in the conical pendulum. The bob's circle has radius $L\sin\theta$, not $L$. With the correct radius, $T=m\omega^2L$ exactly.
5. Units and square roots. Convert km/h to m/s ($\div3.6$) before using $v^2/r$; and remember $v_{\max}\propto\sqrt{\mu_s r}$ — halving $\mu_s$ divides the speed by $\sqrt2$, quadrupling the radius only doubles it.
A bob on a $1.0\ \text{m}$ string moves as a conical pendulum with the string at $30^\circ$ to the vertical. The radius of the bob's circle is
Exercises
Take $g=9.8\ \text{m/s}^2$ unless told otherwise. Treat vehicles as particles.
- 1 Find the net force on a $1200\ \text{kg}$ car moving at a steady $15\ \text{m/s}$ round a circular track of radius $45\ \text{m}$.
- 2 A $0.50\ \text{kg}$ ball on a $1.2\ \text{m}$ string moves in a horizontal circle on a smooth table at $3.0$ revolutions per second. Find the tension.
- 3 A string that breaks at $80\ \text{N}$ whirls a $0.40\ \text{kg}$ stone in a horizontal circle of radius $0.60\ \text{m}$ on a smooth surface. Find the greatest possible speed.
- 4 In a fairground "rotor", riders stand against the wall of a drum of radius $2.5\ \text{m}$ and the floor drops away; $\mu_s=0.40$ between rider and wall. Find the least angular speed (in rad/s and rpm) that keeps them from sliding down.
- 5 A coin $20\ \text{cm}$ from the centre of a turntable starts to slip when the turntable reaches $40\ \text{rpm}$. Find $\mu_s$.
- 6 A $0.50\ \text{kg}$ particle moves on a circle of radius $2.0\ \text{m}$. At one instant its speed is $4.0\ \text{m/s}$ and increasing at $3.0\ \text{m/s}^2$. Find the net force on it and its angle with the radius.
- 7 A $60\ \text{kg}$ passenger sits in a car turning on a circle of radius $30\ \text{m}$ at $12\ \text{m/s}$. Find the centrifugal force in the car's frame, and the real force the seat exerts on the passenger horizontally.
- 8 A $0.10\ \text{kg}$ puck on a smooth table moves in a circle of radius $0.40\ \text{m}$, held by a string that passes through a hole at the centre to a hanging $0.50\ \text{kg}$ mass at rest. Find the speed of the puck.
- 1 Find the maximum speed at which a car can take a level curve of radius $80\ \text{m}$ if $\mu_s=0.50$.
- 2 What least coefficient of friction lets a car take a level curve of radius $100\ \text{m}$ at $90\ \text{km/h}$?
- 3 A cyclist rides round a level circular track of radius $15\ \text{m}$ at $6.0\ \text{m/s}$. At what angle to the vertical must she lean?
- 4 Find the banking angle for a curve of radius $200\ \text{m}$ designed for $90\ \text{km/h}$.
- 5 A broad-gauge track (rails $1.676\ \text{m}$ apart) curves with radius $400\ \text{m}$. By how much must the outer rail be raised for trains at $72\ \text{km/h}$ to need no sideways force from the rails?
- 6 A frictionless track of radius $60\ \text{m}$ is banked at $30^\circ$. Find the speed at which a car can go round it.
- 7 A curve of radius $100\ \text{m}$ is banked at $20^\circ$; $\mu_s=0.40$. Find the maximum and minimum safe speeds.
- 8 Repeat part 7 for an oily road with $\mu_s=0.10$.
- 9 Find the normal force on a $1500\ \text{kg}$ car taking the curve of part 7 at its design (no-friction) speed, and compare it with the car's weight.
- 1 A conical pendulum has a string $0.80\ \text{m}$ long at $30^\circ$ to the vertical. Find its period.
- 2 If the bob in part 1 has mass $0.50\ \text{kg}$, find the tension.
- 3 A bob on a $1.2\ \text{m}$ string is whirled as a conical pendulum at $1.0$ revolution per second. Find the angle of the string with the vertical.
- 4 A string can bear a tension of at most three times the weight of its bob. Find the largest angle it can make with the vertical as a conical pendulum, and the corresponding $\omega$ for $L=0.50\ \text{m}$.
- 5 A $1000\ \text{kg}$ car goes over a hump of radius $25\ \text{m}$ at $10\ \text{m/s}$. Find the normal force at the top, and the speed at which the car would lose contact.
- 6 A bucket of water is swung in a vertical circle of radius $1.0\ \text{m}$. Find the least speed at the top for the water to stay in.
- 7 A $70\ \text{kg}$ pilot pulls out of a dive along a vertical circle of radius $400\ \text{m}$ at $100\ \text{m/s}$. Find the force of the seat on the pilot at the lowest point, as a multiple of the weight.
- 8 An $800\ \text{kg}$ car passes through a dip of radius $30\ \text{m}$ at $12\ \text{m/s}$. Find the normal force at the lowest point.
- 1 A car starts from rest on a level circular track of radius $50\ \text{m}$ and speeds up with a constant tangential acceleration of $2.0\ \text{m/s}^2$; $\mu_s=0.50$. At what speed does it begin to skid?
- 2 Two identical coins sit on a turntable, at $10\ \text{cm}$ and $20\ \text{cm}$ from the centre; $\mu_s=0.25$. Which slips first, and at what angular speed?
- 3 A velodrome track of radius $25\ \text{m}$ is banked at $42^\circ$. Find the speed at which a cyclist needs no sideways friction.
- 4 (Modelling) A highway exit ramp has radius $60\ \text{m}$ and is banked at $10^\circ$. Designers assume $\mu_s=0.30$ on a wet road. Find the maximum safe speed in km/h and suggest a posted speed limit.
- 5 A small ball rolls without friction round the inside of a hemispherical bowl of radius $0.50\ \text{m}$, in a horizontal circle; the line from the centre of the bowl to the ball makes $60^\circ$ with the vertical. Find its angular speed.
- 6 Explain why the design speed of a banked road does not depend on the mass of the vehicle, and why, on a level road, a car skids outward rather than "flying off" radially.
E1. JEE Main · MCQ A car can take a level curve at a maximum speed $v$ on a dry road. On a wet day the coefficient of friction is halved. The maximum safe speed on the same curve becomes (A) $v/2$ (B) $v/\sqrt2$ (C) $\sqrt2\,v$ (D) $v/4$
E2. JEE Main · Numerical A circular race track of radius $90\ \text{m}$ is banked at $45^\circ$. Taking $g=10\ \text{m/s}^2$, find the speed (in m/s) at which a car needs no friction to go round it. Give the answer to the nearest integer.
E3. JEE Main · MCQ A curve of radius $30\ \text{m}$ is banked at $45^\circ$, and the coefficient of static friction between tyres and road is $0.5$. Taking $g=10\ \text{m/s}^2$, the range of speeds at which a car can take the curve without slipping is (A) $10\ \text{m/s}$ to $30\ \text{m/s}$ (B) $0$ to $30\ \text{m/s}$ (C) only $17.3\ \text{m/s}$ (D) $10\ \text{m/s}$ to $17.3\ \text{m/s}$
E4. JEE Main · Numerical A small bob tied to a string of length $0.8\ \text{m}$ moves as a conical pendulum with the string at $60^\circ$ to the vertical. Taking $g=10\ \text{m/s}^2$, find its angular speed in rad/s. Give the answer to the nearest integer.
E5. JEE Main · MCQ A car moves over the top of a convex bridge whose surface is an arc of radius $40\ \text{m}$. Taking $g=10\ \text{m/s}^2$, the greatest speed at which it can cross the top without losing contact with the road is (A) $10\ \text{m/s}$ (B) $40\ \text{m/s}$ (C) $14\ \text{m/s}$ (D) $20\ \text{m/s}$
E6. JEE Main · Numerical Two identical coins are placed on a horizontal turntable at distances $5\ \text{cm}$ and $10\ \text{cm}$ from its axis. The coefficient of static friction between each coin and the turntable is $0.49$. The turntable's angular speed is increased slowly from zero. Taking $g=10\ \text{m/s}^2$, find the angular speed (in rad/s) at which the first coin starts to slip. Give the answer to the nearest integer.
Research question. How much sideways acceleration do drivers actually use on real roundabouts and bends — and how close is it to the friction limit $\mu_sg$?
Why it matters. Highway engineers in many countries design curves using a "side friction factor" — the fraction of $g$ that the tyres may be asked to supply sideways — usually only about $0.1$ to $0.2$, far below the $0.6$–$0.9$ that dry tyres can give. The gap is a safety margin for rain, worn tyres and driver error. With a phone accelerometer you can measure what drivers really use and test the model $a=v^2/r$.
What to do.
- Only as a passenger, with the driver's agreement and the phone fixed flat in a holder, record the sideways acceleration with a free sensor app while the car goes round a few roundabouts and bends at normal speed. Note the speedometer reading at mid-curve (or log speed with a GPS app).
- Find the radius of each curve from an online map: fit a circle to three points on the road and measure its radius with the map's scale tool.
- Compute $v^2/r$ for each curve and compare with the measured sideways acceleration. Make a table and a graph of measured against predicted values.
- Express each measured acceleration as a fraction of $g$. How does it compare with the design range $0.1$–$0.2$ and with a dry-road $\mu_s\approx0.7$?
- Estimate, for one curve, the maximum safe speed in the dry and in the wet ($\mu_s\approx0.3$), using Result 3.16.
Tools: Phone with an accelerometer/GPS logging app, phone holder, online map with a scale tool, spreadsheet.
Hand in: The table and graph, the safe-speed estimates, and a short paragraph on whether drivers stay inside the engineers' margin.
Going further. Many curves are slightly banked. Measure (or estimate) the bank angle of one curve with the phone's inclinometer while parked safely, and work out its design speed $v_0=\sqrt{rg\tan\theta}$. Why do race tracks and velodromes use much steeper banking than highways?
🧪 Self-check quiz
Q1. For a car taking a level circular curve at constant speed, the centripetal force is provided by
Q2. On a frictionless banked road, the speed at which a vehicle can take the curve depends on
Q3. As the angular speed of a conical pendulum is increased, the angle of its string with the vertical
Q4. A bucket of water is swung in a vertical circle of radius $0.9\ \text{m}$ ($g=10\ \text{m/s}^2$). The least speed at the top for the water not to fall out is
Q5. A car takes a banked curve more slowly than the design (no-friction) speed. The friction of the road on the tyres acts