Induction

See how a changing magnetic field drives a current.

A current makes a magnetic field: that was the magnetism lesson. Michael Faraday was sure it must work the other way round too, and spent years trying to get a current out of a magnet. Strong magnets did nothing. On 29 August 1831 he found what he'd been missing: the magnetic field has to be changing. Most of the world's electricity is still made with what he found that day.

Drag the magnet through the coil. The top graph records the flux, the amount of the magnet's field passing through the coil, and the bottom one the voltage that drives current round the circuit. Blue is magnetic (the flux, and the coil's own field) and amber is electric (the voltage, and the current it drives). The magnet is treated as thin, all the coil's turns as sitting in one place, and the circuit as a plain 20 Ω whatever the number of turns.
  • Drag the magnet slowly into the coil and stop with it inside. The flux graph climbs and stays up. The voltage graph leaves zero only while you're moving, and comes straight back to zero when you stop, with the strongest field the coil will ever get sitting right inside it.
  • Now pull it out quickly. The flux falls, so the voltage swings the other way, and further than before because you moved faster. The needle swings to the other side and the bulb flares.
  • Press "Slow pass", wait for it to finish, then "Fast pass". The flux graph makes the same hump both times, 0.31 mWb high. Going four times as fast squeezes the hump into a quarter of the time, so its sides are four times as steep, and the voltage peaks at about 0.95 V instead of 0.24 V.
  • Look at one pass on both graphs together. Where the flux graph climbs, the voltage is below zero. At the top of the hump, where the flux graph is level for an instant, the voltage crosses zero. Where the flux falls, the voltage is above zero.
  • Slide the turns from 500 to 1,000 and run "Fast pass" again. The flux graph is the same. The voltage doubles.
  • Tick "Turn the magnet round" and run a pass. Both graphs flip over.
  • Watch the blue arrows through the coil and the blue "push back" arrow on the magnet during a pass. The push is always against the way the magnet is moving, coming in and going out.

Flux

What matters is the amount of the magnet's field passing through the coil, and how fast that amount changes. So the amount needs a name and a number. It's the magnetic flux, \(\Phi\) (capital phi), the same "field times area" you met with surface integrals. For a flat loop of area \(A\) in an even field \(B\):

\[ \Phi = BA\cos\theta \]

\(\theta\) is the angle between the field and the loop's normal, the line sticking straight out of its face. Face the loop square-on to the field and \(\cos\theta = 1\): it catches everything. Turn it edge-on and \(\cos\theta = 0\): the field slides past and none goes through. If you picture the field as lines, the flux counts the lines that thread the loop.

Field is measured in teslas (T) and area in square metres, so flux is in T·m², called the weber (Wb). The magnet in the figure puts about 0.31 mWb (milliwebers), which is 0.00031 Wb, through each turn of the coil when it sits in the middle.

Faraday's law

What the figure shows is that the flux itself drives nothing. Only its rate of change does, and that is the whole law:

\[ \mathcal{E} = -N\frac{d\Phi}{dt} \]

\(\mathcal{E}\) is a curly E, nothing to do with \(\varepsilon_0\). It's the voltage that appears round the loop: the energy handed to each coulomb of charge that's driven once round it, in volts, just as a battery's voltage is. Its usual name is the electromotive force, or EMF, though it isn't a force. \(\frac{d\Phi}{dt}\) is the derivative of the flux, the slope of the top graph, in webers per second. \(N\) is the number of turns: each turn has the same changing flux through it and gets the same voltage, and the turns are joined end to end, so their voltages add. The minus sign is the subject of the next section.

The voltage lesson said a round trip always comes to nothing. That was for the field of charges sitting still. The electric field made by a changing magnetic field is the exception: it has no height map, and the energy each coulomb gains on a lap comes from whatever is changing the field.

So the bottom graph in the figure is the top graph's slope, turned upside down and multiplied by \(N\). That's why holding the magnet still gives nothing, and why a fast pass beats a slow one. Here it is with the numbers from the fast pass.

In a fast pass the magnet moves at 0.4 m/s. On the steepest part of the hump, the flux through each turn is changing by about 1.9 mWb every second.

\[ \frac{d\Phi}{dt} \approx 1.9 \text{ mWb/s} = 0.0019 \text{ Wb/s} \]

That would give 0.0019 V round one turn. The coil has 500 of them in a row.

\[ \mathcal{E} = 500 \times 0.0019 = 0.95 \text{ V} \]

The circuit's resistance is 20 Ω, so by Ohm's law from the circuits lesson, \(I = \frac{\mathcal{E}}{R}\):

\[ I = \frac{0.95}{20} = 0.0475 \text{ A} \approx 48 \text{ mA} \]

And the power going into the bulb and the wire is voltage times current.

\[ P = \mathcal{E} I = 0.95 \times 0.0475 \approx 0.045 \text{ W} \]

A slow pass is a quarter of the speed, so every slope on the flux graph is a quarter as steep and the voltage peaks at \(\frac{0.95}{4} \approx 0.24\) V. The power goes as the voltage squared, so it's a sixteenth.

There are three ways to change \(\Phi = BA\cos\theta\), and all three work: change the field through the loop (move a magnet, or switch an electromagnet on and off), change the loop's area, or turn the loop. Faraday's law doesn't care which.

Lenz's law: the minus sign

The current that flows in the coil makes a magnetic field of its own, as every current does. The minus sign says which way that field points: against the change in flux. This is Lenz's law, after Heinrich Lenz, who worked it out in 1834.

Push the magnet's N end towards the coil and the flux through the coil grows. The coil's current runs the way that makes its own field point back out at the magnet, so the coil's near face becomes an N pole and repels the magnet coming in. Pull the magnet away and the flux shrinks. Now the current runs the other way, the near face becomes an S pole, and the coil tugs on the magnet as it leaves. Either way, you have to push against the coil. That's the blue arrow in the figure.

It has to be this way round. Suppose the coil pulled the approaching magnet in instead. The magnet would speed up, the flux would change faster, the current would grow, the pull would grow, and you'd have a magnet accelerating and a bulb shining with nobody doing any work. Energy would be coming from nowhere. As it is, the books balance exactly. At the peak of the fast pass the bulb and wire take 0.045 W. Work is force times distance, so the work done each second is force times the distance covered each second: power is force times speed. So at 0.4 m/s your hand must be pushing against

\[ F = \frac{P}{v} = \frac{0.045}{0.4} \approx 0.11 \text{ N} \]

about the weight of an 11 gram letter. Every joule the bulb gives out is a joule of work your hand did against that push. A generator is hard to turn for the same reason, and harder the more current is drawn from it.

The generator

Moving a magnet in and out is an awkward way to keep a current going. Turning a coil in a steady field is better, because turning can go on for ever.

You're looking along the axle. The white bar is the coil seen end on, and the two rings are its long sides: a dot is current coming towards you, a cross is current going away. The blue field lines are the ones passing through the coil, one for each milliweber of flux, to the nearest whole line. The coil's length along the axle stays the same, so a bigger area is a wider coil. Drag the amber point to turn it by hand. The field is drawn perfectly even, and the bulb as following the voltage instantly.
  • Watch the blue lines. Their number is the flux, one line for each milliweber. There are most of them, five, when the bar stands upright, square-on to the field, and none when the bar lies along the field.
  • Now watch the voltage at those two moments. It's zero when the flux is greatest, and at its peak when the flux is zero. A cosine is changing fastest as it passes through zero.
  • Pause it, then drag the amber point round slowly, and then fast. You are the engine now: the voltage follows how fast you turn as well as where the coil is.
  • Press "Turn it" and double the turning rate from 1.00 to 2.00 turns a second. The waves come twice as often, and they're twice as tall as well, 6.28 V instead of 3.14 V.
  • Put the rate back to 1.00 and double the area, the field or the turns instead. The peak doubles each time, to 6.28 V, with the waves no closer together. A bigger area or a stronger field also doubles the blue lines through the coil. More turns doesn't: each turn catches the same flux as before, and there are twice as many turns catching it.
  • Watch the dot and the cross in the rings. They swap every half turn: the current goes one way round the coil, then the other.

Let the coil turn steadily, at \(\omega\) radians a second, so that \(\theta = \omega t\). (One full turn is \(2\pi\) radians, so \(\omega = 2\pi \times\) turns a second.) Then Faraday's law gives the voltage in three lines.

The flux through each turn follows a cosine.

\[ \Phi = BA\cos\omega t \]

Differentiate. The derivative of \(\cos\) is \(-\sin\), and the chain rule brings out the \(\omega\).

\[ \frac{d\Phi}{dt} = -BA\omega\sin\omega t \]

Faraday's law multiplies by \(-N\). The two minus signs cancel.

\[ \mathcal{E} = -N\frac{d\Phi}{dt} = NBA\omega\sin\omega t \]

The sine swings between \(-1\) and \(1\), so the voltage swings between \(-NBA\omega\) and \(+NBA\omega\). With the figure's first settings, 100 turns, 0.5 T, 100 cm² (which is 0.010 m²) and one turn a second:

\[ NBA\omega = 100 \times 0.5 \times 0.010 \times 6.283 \approx 3.14 \text{ V} \]
\[ \Phi = BA\cos\omega t \qquad\qquad \mathcal{E} = NBA\omega\sin\omega t \]

A voltage that swings both ways drives a current that does too: alternating current, AC. Almost every power station does this, whether it burns gas, splits uranium or catches wind (solar panels are the exception): something turns a shaft, and the shaft turns coils and magnets past each other. In Australia and Europe the generators turn so that the voltage goes through 50 full swings a second, and in North America 60. Turning faster gives a higher voltage as well as a faster swing, because \(\omega\) multiplies the peak. That's the chain rule at work: the faster the flux goes round its cycle, the steeper its graph.

The transformer

Faraday's first success, that day in 1831, used no moving magnet. He wound two separate coils of wire on one iron ring, joined the first to a battery and the second to a meter. With the current flowing steadily, nothing. But at the instant he connected the battery, the needle kicked, and when he disconnected it, it kicked the other way. The first coil's current makes a field in the iron, and when that current changes, the flux through the second coil changes.

Feed the first coil (the primary) with alternating current and the flux in the iron never stops changing, so the second coil (the secondary) gives a steady alternating voltage. That's a transformer. The iron carries nearly all the flux through both coils, so every turn of both has the same \(\frac{d\Phi}{dt}\) and gets the same voltage. A coil's whole voltage is then just its number of turns times that:

\[ \frac{V_s}{V_p} = \frac{N_s}{N_p} \]

\(V_p\) and \(N_p\) are the voltage across the primary and its number of turns, and \(V_s\) and \(N_s\) the same for the secondary. A secondary with a twentieth of the turns gives a twentieth of the voltage: 230 V becomes 11.5 V. Nothing is gained for free. A good transformer wastes very little, so the power going in, \(V_pI_p\), is close to the power coming out, \(V_sI_s\). If the voltage goes down 20 times, the current goes up 20 times.

Transformers are the reason mains electricity is AC. A cable of resistance \(R\) carrying a current \(I\) turns \(I^2R\) of power into heat. Suppose a town needs 1 MW through a line of 5 Ω. At 10,000 V that takes \(\frac{1{,}000{,}000}{10{,}000} = 100\) A, and the line wastes \(100^2 \times 5 = 50{,}000\) W, a twentieth of everything sent. Step the voltage up to 100,000 V first and the same power needs only 10 A, wasting \(10^2 \times 5 = 500\) W. So power crosses the country at hundreds of thousands of volts and is stepped down, transformer by transformer, on the way to the socket. With a steady current none of this works: the flux doesn't change, and the secondary gives nothing at all.

The coil in the first figure isn't what makes the voltage. Take the wire away and the changing magnetic field still makes an electric field, curling round it in empty space; a wire is only a way of noticing. (In the language of the last lesson, it's a changing magnetic potential that makes this electric field.) So a changing magnetic field makes an electric field. If a changing electric field also made a magnetic one, each could keep the other going with no charges or magnets anywhere near. It does, and the next lesson shows what that travelling disturbance is.

Problems

The flux through a single loop of wire rises steadily from 0 to 0.060 Wb in 0.20 s. What voltage appears round the loop while it's rising? What would a coil of 50 such turns give?

Answer

A steady rise means a constant slope: \(\frac{d\Phi}{dt} = \frac{0.060}{0.20} = 0.30\) Wb/s, so the voltage is 0.30 V in size. Fifty turns in a row give \(50 \times 0.30 = 15\) V.

You push the N end of a magnet towards one face of a coil. Which pole does that face of the coil become? What does it become as you pull the magnet back out? And what would the coil do if its ends weren't connected to anything?

Answer

Going in, the face becomes an N pole and repels the magnet. Coming out, it becomes an S pole and pulls the magnet back. Both resist what you're doing, which is Lenz's law. With its ends unconnected, the voltage still appears across them, but no current can flow, so the coil makes no field of its own and you feel no push at all. No energy is being taken, so none has to be supplied.

A coil of 200 turns, each of area 4.0 cm², sits square-on to a field of 0.20 T. The field is switched off and falls steadily to zero in 5.0 ms. What voltage appears across the coil?

Answer

The area is \(4.0 \times 10^{-4}\) m², so the flux through each turn starts at \(\Phi = BA = 0.20 \times 4.0 \times 10^{-4} = 8.0 \times 10^{-5}\) Wb and ends at zero. It changes at \(\frac{8.0 \times 10^{-5}}{5.0 \times 10^{-3}} = 0.016\) Wb/s, so \(\mathcal{E} = 200 \times 0.016 = 3.2\) V. Switching off faster would give more: that's how a car's ignition coil turns 12 V into a spark.

Set the generator figure to 100 turns, 0.5 T, 100 cm² and 1.50 turns a second. Work out \(\omega\) and the peak voltage, and check them against the readout. Then find the voltage at the moment \(\theta = 30^\circ\). How fast would the coil have to turn to peak at 12 V?

Answer

\(\omega = 2\pi \times 1.50 = 9.425\) rad/s. The peak is \(NBA\omega = 100 \times 0.5 \times 0.010 \times 9.425 \approx 4.71\) V. At \(30^\circ\), \(\sin 30^\circ = 0.5\), so \(\mathcal{E} = 4.71 \times 0.5 = 2.355\) V, about 2.36 V.

For 12 V, \(\omega = \frac{12}{NBA} = \frac{12}{100 \times 0.5 \times 0.010} = 24\) rad/s, which is \(\frac{24}{2\pi} \approx 3.8\) turns a second.

A transformer steps 230 V mains down to 11.5 V. Its primary has 1,000 turns. How many turns has the secondary? If the secondary supplies 4.0 A, what current does the primary draw, taking the transformer as wasting nothing?

Answer

\(N_s = N_p \times \frac{V_s}{V_p} = 1{,}000 \times \frac{11.5}{230} = 50\) turns. The power out is \(V_sI_s = 11.5 \times 4.0 = 46\) W, and the same power goes in, so \(I_p = \frac{46}{230} = 0.20\) A. The voltage went down 20 times and the current up 20 times.

A metal rod 0.20 m long slides at 3.0 m/s along two rails joined at one end by a 1.5 Ω resistor, so that rod, rails and resistor make a loop. A field of 0.50 T passes straight through the loop. As the rod slides, the loop's area grows. Find the voltage, the current, and the power turned to heat in the resistor. Then find the force needed to keep the rod moving, using the fact that a wire of length \(L\) carrying a current \(I\) across a field \(B\) feels a force \(BIL\) (the magnetic force \(qvB\) added up over all the moving charges in it). Do the books balance?

Answer

When the rod is a distance \(x\) along the rails, the loop's area is \(Lx\) and its flux is \(\Phi = BLx\). Only \(x\) changes, at the rod's speed \(v\), so \(\frac{d\Phi}{dt} = BLv = 0.50 \times 0.20 \times 3.0 = 0.30\) Wb/s, and \(\mathcal{E} = 0.30\) V in size.

\(I = \frac{\mathcal{E}}{R} = \frac{0.30}{1.5} = 0.20\) A, and the resistor turns \(P = \mathcal{E} I = 0.30 \times 0.20 = 0.060\) W into heat.

The force on the rod is \(BIL = 0.50 \times 0.20 \times 0.20 = 0.020\) N, and by Lenz's law it points against the motion. To keep the rod going you push with 0.020 N at 3.0 m/s, doing work at \(Fv = 0.020 \times 3.0 = 0.060\) W. That's exactly the heat in the resistor.

Learn more

An unhandled error has occurred. Reload 🗙