Everything so far has moved along a line or a curve. But wheels, doors, skaters, planets and molecules also turn. You don't need new physics for that. Every idea from the last few lessons has a twin for rotation, and the equations are the same ones with the letters changed.
- Slide the push towards the hinge. The same force gives less and less torque, and at the hinge it gives none.
- Drag the amber tip round so the force leans along the door. Only the part square to the door, in amber, does any turning. Point it straight along the door and nothing happens.
- Find two quite different pushes that give the same torque.
- Now slide the weight out towards the edge. The torque hasn't changed, but the door is far more sluggish. Double the weight's distance and the moment of inertia goes up four times.
- Switch to "Spread along the door", which is a real door. It behaves like the single weight placed 0.58 m out.
- The amber dots mark the door's edge every tenth of a second. Why do the gaps grow?
Turning, in the language of angles
For something turning about a fixed axis, the angle \(\theta\) plays the part of position. Measure it in radians and the rest follows as before: the angular velocity \(\omega\) is how fast the angle changes, and the angular acceleration \(\alpha\) is how fast that changes.
Every part of a turning door turns at the same rate, but the parts further out travel faster. A point at distance \(r\) from the axis covers an arc \(r\theta\), so its speed is \(v = r\omega\) and its acceleration along the arc is \(a = r\alpha\).
What makes something turn
A force's turning effect is its torque, \(\tau\). It depends on how far from the axis the force acts, and on how much of the force is square to the line from the axis:
That's why door handles are on the far side from the hinges, and why a longer spanner shifts a tighter nut. Torque is measured in newton metres.
How strongly something resists being turned is its moment of inertia, \(I\). To see where it comes from, take the figure's single weight, a mass \(m\) at distance \(r\) from the hinge, and push on the door right at the weight, so the push is at distance \(r\) too.
The part of the push square to the door accelerates the weight along its arc. That's Newton's second law.
Its acceleration along the arc is \(r\) times the angular acceleration.
Multiply both sides by \(r\), so that the left side becomes the torque.
Call \(mr^2\) the moment of inertia, and this is Newton's second law for turning.
Push somewhere else and the result still holds, with the push's distance in the torque and the weight's distance in \(I\).
Mass counts for more the further out it is, and by the square of the distance. For a body made of many pieces, add up \(mr^2\) for each piece. For a solid one, the sum becomes an integral. A door of mass \(M\) and width \(w\), with its mass spread evenly, has \(\frac{M}{w}\) kilograms in each metre of width, so:
The same integral for other shapes, each turning about its centre, gives \(MR^2\) for a hoop, \(\frac{1}{2}MR^2\) for a solid disc and \(\frac{2}{5}MR^2\) for a solid ball. The hoop's is the largest because all of its mass is as far out as it can be.
The dictionary
Swap each quantity on the left for the one on the right, and every formula for motion along a line becomes a true formula for turning.
So the equations of motion come across too. From rest under a steady torque, \(\theta = \frac{1}{2}\alpha t^2\), which is how the figure times the door's quarter turn.
Angular momentum
The twin of momentum is angular momentum, \(L = I\omega\), and the twin of \(F = \frac{dp}{dt}\) is \(\tau = \frac{dL}{dt}\). So with no torque from outside, \(L\) can't change. But unlike mass, the moment of inertia can change while something is moving.
- Press "Pull the arms in". The spin rate climbs, while the amber bar for angular momentum doesn't move.
- Set the masses 1.00 m out, then 0.50 m. The moment of inertia falls from 2.50 to 1.00, so the spin rate has to rise by the same factor, from 2.00 to 5.00 rad/s.
- Watch the bottom bar. The rotational energy goes up as the arms come in. Where does it come from?
- Move the angular momentum slider. That's a push from outside. Nothing you do with the arms can change it.
as long as no torque acts on it from outside.
This is how a skater spins up by pulling in her arms, and how a diver tucks to turn faster. It also happens to dying stars. When the core of a big star, thousands of kilometres across, collapses to a ball about 20 km across, it's left turning many times a second: a pulsar.
Angular momentum stays the same but energy doesn't. Rotational energy is \(\frac{1}{2}I\omega^2\), which can be written \(\frac{L^2}{2I}\), so with \(L\) fixed, a smaller \(I\) means more energy. The skater supplies it. It takes a strong inward pull to draw her arms in while they're whirling round, and a pull through a distance is work.
Rolling
A rolling wheel is doing both kinds of motion at once, and it has both kinds of kinetic energy: \(\frac{1}{2}mv^2\) for moving along and \(\frac{1}{2}I\omega^2\) for spinning. If it doesn't slip, its rim turns at the same speed as the wheel travels, so \(v = r\omega\).
- Before you press Race, guess the order they'll finish in.
- At the finish, look at the bars. All three are the same length, but each is split differently.
- Change the slope. The times change. The order and the splits never do.
Let one of them roll down through a height \(h\). Write its moment of inertia as \(I = kmr^2\), where \(k\) is 1 for the hoop, \(\frac{1}{2}\) for the disc and \(\frac{2}{5}\) for the ball. The potential energy it loses becomes both kinds of kinetic energy, with \(\omega = \frac{v}{r}\):
The \(r^2\) cancels in the last term, leaving \(\frac{1}{2}kmv^2\). So:
The mass has cancelled, and so has the radius. A marble and a cannonball tie. Only the shape is left, through \(k\): the more of its mass a body carries far from its centre, the more of the energy goes into spin, and the less is left for getting down the slope.
Angular momentum is conserved for the same kind of reason momentum is: the laws of physics don't depend on which way you're facing. It's why a shrinking cloud of gas flattens into a disc, which gives solar systems and spiral galaxies their shape. In quantum mechanics it comes only in fixed steps, and electrons carry a fixed amount of it, called spin, that isn't anything turning at all. The way electrons fill atoms, and so the whole of chemistry, follows from those rules.
Problems
You tighten a wheel nut with a spanner 0.30 m long, pushing at its end with 150 N, square to the spanner. What torque is that? What if you slip a pipe over the handle to make it 0.60 m long? And what if you push on the 0.30 m spanner at \(60^\circ\) to it instead of \(90^\circ\)?
Answer
\(\tau = 0.30 \times 150 = 45\) N·m. With the pipe, \(0.60 \times 150 = 90\) N·m. At \(60^\circ\), only \(150\sin 60^\circ \approx 130\) N is square to the spanner, so \(\tau \approx 0.30 \times 130 = 39\) N·m.
In the first figure, put the 4 kg weight 0.50 m from the hinge and push with 6 N, square to the door, 0.80 m from the hinge. Find the torque, the moment of inertia, the angular acceleration and the time for a quarter turn. Check them against the figure.
Answer
\(\tau = 0.80 \times 6 = 4.8\) N·m. \(I = 4 \times 0.50^2 = 1.0\) kg·m². \(\alpha = \frac{4.8}{1.0} = 4.8\) rad/s². A quarter turn is \(\frac{\pi}{2}\) radians, and \(\theta = \frac{1}{2}\alpha t^2\), so \(t = \sqrt{\frac{\pi}{4.8}} \approx 0.81\) s.
A real door has a mass of 20 kg spread evenly across its 0.80 m width. What's its moment of inertia about the hinges? You push with 10 N at the handle, 0.75 m from the hinges and square to the door. How long does it take to swing through a quarter turn?
Answer
\(I = \frac{1}{3} \times 20 \times 0.80^2 \approx 4.27\) kg·m². \(\tau = 0.75 \times 10 = 7.5\) N·m, so \(\alpha = \frac{7.5}{4.27} \approx 1.76\) rad/s². Then \(t = \sqrt{\frac{\pi}{1.76}} \approx 1.3\) s.
A skater spins at 2.0 turns a second with her arms out, when her moment of inertia is 4.0 kg·m². She pulls her arms in and it falls to 1.6 kg·m². How fast does she spin now? By what factor has her rotational energy changed?
Answer
\(I\omega\) stays the same, so \(\omega\) goes up by \(\frac{4.0}{1.6} = 2.5\) times, to 5.0 turns a second. Energy is \(\frac{L^2}{2I}\) with \(L\) unchanged, so it's also 2.5 times bigger. In numbers: \(\omega\) goes from \(4\pi \approx 12.6\) to \(10\pi \approx 31.4\) rad/s, and \(\frac{1}{2}I\omega^2\) from about 316 J to about 790 J (using \(4\pi\) and \(10\pi\) themselves, not the rounded figures). She did the difference as work.
A solid ball and a hoop roll from rest down a ramp 0.50 m high. How fast is each going at the bottom? Compare them with a block sliding down with no friction.
Answer
Ball: \(v = \sqrt{\frac{2 \times 9.81 \times 0.50}{1.4}} \approx 2.65\) m/s. Hoop: \(\sqrt{\frac{2 \times 9.81 \times 0.50}{2}} \approx 2.21\) m/s. The block has nothing going into spin, so \(k = 0\) and \(v = \sqrt{2 \times 9.81 \times 0.50} \approx 3.13\) m/s. It wins.
On a \(20^\circ\) slope, a block sliding without friction would accelerate at \(g\sin 20^\circ\). Rolling divides that by \(1 + k\). Find the hoop's acceleration and its time to cover 2 m from rest, and check against the race.
Answer
\(a = \frac{9.81 \times \sin 20^\circ}{2} \approx 1.68\) m/s². From \(x = \frac{1}{2}at^2\), \(t = \sqrt{\frac{2 \times 2}{1.68}} \approx 1.54\) s.