Momentum, Friction, and the Physics of a Turn

For nearly two thousand years, the dominant view, credited to Aristotle, was straightforward and matched everyday experience: keeping something moving requires a continuous force, since every moving object, left alone, eventually stops. Galileo dismantled this using a cleverly designed experiment rather than pure argument: releasing a ball on one side of a smooth double-inclined ramp, letting it roll down one slope and up the other, he found it always climbed back to very nearly its starting height, regardless of how gently the second slope was angled. Flatten that second slope all the way to horizontal, and the logical conclusion follows: with the height it 'needs' to reach never achievable on a flat surface, the ball would have to keep rolling forever, seeking a height it can never reach. What actually stops a real ball on a real flat surface is friction, an opposing force Aristotle's framework never isolated as a separate cause. Remarkably, ancient Indian natural philosophy had independently developed related ideas centuries earlier: the Vaisesika school described several distinct kinds of force, including nodana (force from continuous pressure, such as wind on a sail) and vega, a persistent tendency of a moving body to continue in a straight line, opposed by contact with surrounding matter, a concept genuinely close to inertia. The astronomer Bhaskara, writing around 1150 CE, even described a notion of 'instantaneous motion' (tatkalikigati), anticipating by centuries the exact idea of instantaneous velocity built from calculus in the previous chapter.

Newton's own statement of the second law was not F = ma, but something more general: force is the rate of change of momentum, where momentum, p = mv, is the product of an object's mass and velocity, a vector pointing the same way as velocity. Two everyday observations motivate momentum as the genuinely relevant quantity: stopping a heavy truck needs a much larger force than stopping a light car moving at the same speed, and stopping a fast-moving object needs a much larger force than stopping a slow one of the same mass. Momentum, the product of both, captures both effects in one quantity, and force is defined as how quickly that product changes: F = dp/dt. For an object of fixed mass, this simplifies neatly, since d(mv)/dt = m(dv/dt) = ma, recovering the familiar F = ma, but the momentum-based version is genuinely more general, remaining correct even for situations where mass itself is changing, such as a rocket burning and ejecting fuel as it flies. Consider a badminton shuttlecock of mass 0.005 kg, struck so its velocity changes from 4 m/s to -12 m/s (reversing direction) in 0.01 s during a smash: the change in momentum is 0.005 x (-12 - 4), or -0.08 kg m/s, so the average force on the shuttlecock is -0.08 / 0.01, or -8 N, eight newtons in the direction of the reversed velocity.

Multiply both sides of F = Δp/Δt by Δt, and a genuinely useful quantity falls out: impulse, defined as force multiplied by the time it acts for, exactly equal to the change in momentum it produces. This is especially valuable whenever the force itself is difficult to measure directly, large and brief, such as a bat striking a ball, but the resulting change in momentum can still be calculated from before-and-after velocities alone. This is the precise, quantitative version of the cricket-catch and airbag reasoning already met earlier: for a fixed, required change in momentum, a longer contact time (Δt) demands a smaller average force, while a shorter contact time demands a larger one, since their product, the impulse, must stay the same. A goalkeeper of a mass 0.43 kg football, saving a shot arriving at 20 m/s and catching it to a complete stop in their arms over 0.15 s, experiences an impulse of magnitude 0.43 x 20, or 8.6 kg m/s, and therefore an average force of 8.6 / 0.15, about 57 N; the very same save, made rigidly with stiff, unmoving arms bringing the ball to rest in just 0.02 s instead, would require an average force of 8.6 / 0.02, a punishing 430 N, on the same hands.

Combine the second and third laws, and a genuinely powerful result appears. If two objects A and B interact, exerting forces on each other for the same time interval Δt, the third law guarantees those forces are equal and opposite: Fᴬᴮ = -Fᴮᴬ. By the second law, each force equals the rate of change of that object's own momentum, so Δpᴬ = -Δpᴮ over that same interval, meaning any momentum object A gains, object B loses, exactly. Add the two momentum changes together and they cancel completely: the total momentum of the pair, pᴬ + pᴮ, is exactly the same after the interaction as before it. This is the law of conservation of momentum: the total momentum of an isolated system, one with no external forces acting on it, never changes, no matter how complicated the internal interactions within it are. A cricketer of mass 60 kg standing at rest on a frictionless skateboard throws a 0.16 kg cricket ball forward at 25 m/s: since the system started with zero total momentum, it must still have zero total momentum immediately after the throw, so the cricketer and skateboard, combined mass roughly 60 kg, must recoil backward with velocity v where 60v + (0.16)(25) = 0, giving v ≈ -0.067 m/s, a small but genuine recoil in the opposite direction.

Push gently on a heavy box and it does not move; push harder, and eventually it does. Before it moves, static friction is not some fixed value, it is a self-adjusting force that exactly matches whatever push is applied, right up to a maximum limit, after which it can adjust no further and the object begins to slide. This limiting value follows an experimentally observed rule: (fₛ)ₘₐₓ = μₛN, where N is the normal (perpendicular) contact force between the surfaces and μₛ, the coefficient of static friction, is a number depending only on the nature of the two surfaces in contact, not on the area touching or how hard the push is. Once sliding actually begins, a slightly smaller coefficient takes over, kinetic friction, fₖ = μₖN, with μₖ consistently a little less than μₛ, which is exactly why a stuck object suddenly 'gives' and accelerates the moment it starts to slide, the resisting friction has just dropped. A useful related idea is the angle of repose: tilt a surface holding an object at rest, and the object stays put only as long as the tilt angle stays below θₘₐₓ = tan⁻¹(μₛ), since beyond that angle, gravity's component along the slope finally exceeds the maximum static friction available. A wooden crate that just begins to slide on a ramp tilted to exactly 20° has revealed its own coefficient of static friction directly: μₛ = tan 20°, about 0.36, no separate force measurement required at all.

A car taking a circular turn is undergoing exactly the accelerated circular motion from the previous chapter, needing a centripetal force of magnitude mv²/R directed toward the centre of the turn. On a flat, unbanked road, the only horizontal force available at all is friction between tyres and road surface, so friction alone must supply this entire centripetal force: f = mv²/R. Since static friction has a hard ceiling, μₛN = μₛmg on a flat road, there is a genuine maximum safe speed for a given turn: setting f equal to its maximum value gives vₘₐₓ = √(μₛRg), a speed that, notably, does not depend on the vehicle's mass at all. A motorcyclist taking an unbanked turn of radius 25 m, with a tyre-road coefficient of static friction of 0.4, has a maximum safe speed of √(0.4 x 25 x 9.8), about 9.9 m/s, roughly 35.6 km/h; taking the same turn any faster risks the tyres losing their grip entirely, since no more friction can be supplied beyond that limit, whatever the rider's skill or the bike's mass.

Tilting the road surface itself, banking, offers an elegant way around friction's speed limit. On a banked curve, the normal force N, now angled rather than purely vertical, has a horizontal component that can contribute directly to the centripetal force, without needing friction at all. There exists one particular optimum speed, v₀ = √(Rg tan θ), where the horizontal component of the normal force alone provides exactly the centripetal force needed, with zero friction required, and zero sideways wear on the tyres; a vehicle can travel this speed on ice and still hold the curve perfectly. Below this speed, friction acts up the slope to prevent sliding inward; above it, friction acts down the slope to prevent sliding outward, and including friction's maximum contribution raises the true maximum banked-curve speed higher still than the frictionless optimum. This is precisely why velodromes, purpose-built for cycling at high speed around tight curves, bank steeply, sometimes 40° or more, letting cyclists take a curve at racing speed with the track's own tilt doing most of the centripetal work, rather than relying on tyre grip alone. A racetrack of radius 200 m, banked at 20°, has an optimum, friction-free speed of √(200 x 9.8 x tan 20°), about 26.7 m/s, roughly 96 km/h, a speed at which the track's banking, on its own, provides exactly the right amount of inward pull.