Faraday's Law & Generators · Guide · glidepathscience.org
GuideFaraday's Law & Generators
Contents
This Guide explains the ideas and equations behind every Walkthrough step, Concept Challenge question and Predict round, in the order a textbook takes them. Part 1 prepares the Intro track and Part 2 the Advanced track; each takes about a quarter of an hour to read. Each section ends with a worked example that is not one of the sim’s own problems.
Part 1 · The essentials
Math you’ll need: algebra, a little trigonometry, and the idea of a rate of change; calculus appears only in boxes you may skip.
A changing field makes a current
In 1831 Michael Faraday wound two coils on an iron ring and connected one to a battery and the other to a galvanometer. A steady current in the first coil did nothing to the second. But at the moment he closed the switch, and again when he opened it, the galvanometer’s needle kicked. He found the same thing with a bar magnet and a coil: a magnet resting inside the coil did nothing, while a magnet moving in or out drove a current.
That is the whole surprise of this topic. A magnetic field, however strong, does not by itself make a current in a wire. A change does: a magnet moving, a coil turning, a loop growing, a field switched on or off. The current produced this way is the induced current, and what drives it round the circuit is an induced EMF, measured in volts like a battery’s. Every generator on every power grid works this way, and so does a guitar pickup, a card reader and the brakes on some trains.
Magnetic flux
To say what “changes”, we need a measure of how much field passes through a loop. Think of field lines crossing the flat surface the loop encloses. A strong field puts more lines through it; a bigger loop catches more; and a loop tilted away from the field catches fewer, because less of its area faces the field. The magnetic flux through a flat loop of area $A$ in a uniform field $B$ is
$Φ_B = BA\cos θ$
where $θ$ is the angle between $\vec B$ and the loop’s normal $\hat n$, the direction perpendicular to its plane. Face on to the field, $θ$ = 0 and the flux is $BA$; edge on, $θ$ = 90° and the flux is zero, however strong the field. Looking along the field, a tilted loop appears foreshortened to $A\cos θ$: that is exactly the area the field lines cross. Flux is measured in webers, 1 Wb = 1 T·m².
Some books write the flux $Φ_m$; this sim writes $Φ_B$, to pair with the electric flux $Φ_E$ of the Flux Counter, which counts electric field through a surface in the same way. One difference: magnetic field lines never begin or end, so the flux out of any closed surface is zero (Gauss’s law for magnetism). Through an open surface bounded by a loop, it can be anything.
Faraday’s law
Faraday’s discovery, in modern form, is that the EMF round a loop equals the rate at which the flux through it changes. For a coil of $N$ turns, each threaded by the same flux, the EMFs of the turns add in series:
$ε = −N\,ΔΦ_B/Δt$
Read it carefully. The EMF depends on how fast the flux changes, not on how much flux there is. A magnet resting inside a coil threads it with a large, steady flux and makes no EMF at all. Pull it out and the flux falls: an EMF appears for exactly as long as the flux is changing. The minus sign records the direction of that EMF, which is the subject of the next section.
The flux can change in any of the ways $BA\cos θ$ can: the field can change (a magnet moves, a current is switched), the area can change (a loop stretches), or the angle can change (a coil turns).
There is a second, quieter consequence. Since $εΔt = NΔΦ_B$, the area under a trace of EMF against time is the total change in flux linkage. Push a magnet into a coil quickly and the pulse is tall and narrow; push it slowly and the pulse is low and wide; but the areas are the same. Through a circuit of resistance $R$ the charge that flows is
$Q = NΔΦ_B/R$
whatever the speed, so a meter that measures charge (a ballistic galvanometer) measures the flux change directly.
Lenz’s law: opposing the change
Which way does the induced current flow? Lenz’s law answers: the induced current flows in the direction that opposes the change in flux that produced it. That is the meaning of the minus sign in Faraday’s law.
Push the north end of a magnet toward a coil. The flux through the coil grows. The induced current makes its own magnetic field, and it flows the way that makes that field oppose the growth: the coil’s end facing the magnet becomes a north pole, and repels it. Now pull the magnet away. The flux shrinks, and the induced current flows the other way, to prop the flux up: the coil’s end becomes a south pole and pulls back on the leaving north end. Either way the coil resists the motion.
Notice what Lenz’s law does not say. It does not say the induced field always points against the magnet’s field. When the flux is shrinking, the induced field points along the magnet’s field, to keep the flux from falling. The rule is about the change, not the field.
Why must it be so? Suppose the current flowed the other way, so that the coil attracted an approaching magnet. The magnet would speed up, the flux would change faster, the current would grow, and the attraction would grow: the magnet would gain kinetic energy and the circuit would gain heat, all from nothing. Lenz’s law is energy conservation. Because the coil always pushes back, you must do work to move the magnet, and that work is exactly the energy the current delivers.
Motional EMF: a rod on rails
A conducting rod of length $L$ (some books write $l$) slides with speed $v$ along two rails, across a uniform field $B$ perpendicular to the rails’ plane, and a resistor joins the rails’ far ends. The rod, rails and resistor make a loop whose area grows by $Lv$ every second, so its flux grows at $BLv$, and Faraday’s law gives
$ε = BLv$
The same answer comes from the force on charges. Each free charge in the rod moves with it at $v$ across the field and feels a magnetic force $qvB$ along the rod. Pushing charge along the rod’s length $L$ against nothing else, this force does work $qvBL$ per charge: an EMF of $BLv$. The rod is a battery made by motion (Force on a Moving Charge shows the same force on a single charge).
With a resistance $R$ in the circuit, a current $I = ε/R$ flows. That current crosses the field in the rod, so the field pushes on the rod with $F = ILB$, and by Lenz’s law this force points against the motion. To keep the rod moving steadily, you must push with that same force, supplying power $Fv$. Where does it go? The current heats the resistor at $I^2R$, and
$P = Fv = I^2R$
holds exactly, since $Fv = ILBv = Iε = I^2R$. Every joule you put in becomes heat. Push twice as fast and the EMF, the current and the force all double, so the power quadruples.
Terminal speed and the energy books
Now stand the rails upright and let the rod fall. At first it falls freely, but as it speeds up, the EMF $BLv$, the current $BLv/R$ and the braking force $B^2L^2v/R$ all grow. The rod stops speeding up when the braking force balances its weight, $B^2L^2v/R = mg$, at the terminal speed
$v_t = mgR/(B^2L^2)$
A larger resistance means less current at a given speed, so the rod must fall faster before the forces balance. The approach is quick: the speed closes the gap to $v_t$ exponentially, with a time constant $mR/(B^2L^2)$. Once at the terminal speed, gravity does work at $mgv_t$ every second, the rod’s kinetic energy no longer changes, and all of that power heats the circuit.
Generators
A generator turns a coil in a magnetic field. A coil of $N$ turns and area $A$, turning at angular speed $ω$ in a uniform field $B$, has its normal at angle $θ = ωt$ to the field, so its flux linkage is $NBA\cos ωt$ and its EMF is
$ε = NBAω\sin ωt$, peak $ε_0 = NBAω$
with $ω$ in radians per second, $ω = 2πf$. A coil turning at 600 rpm makes 10 turns a second, $f$ = 10 Hz, and $ω$ = 62.8 rad/s. Power stations make alternating current at 60 Hz (50 Hz in much of the world) the same way. Slip rings carry this alternating EMF out to the circuit.
When is the EMF largest? Not when the flux is largest. Face on to the field, the flux is at its maximum and, for an instant, not changing: the EMF is zero. A quarter turn later the coil’s plane lies along the field: the flux is zero, but it is changing fastest, and the EMF is at its peak. Flux and EMF are a quarter cycle apart.
Now connect a load, so that a current flows. The current crosses the field in the coil’s sides and the field pushes back on them: a counter-torque, by Lenz’s law, against the turning. With a resistance $R$ in all, the power delivered goes as $\sin^2 ωt$, whose average is ½, so the average power is
$\langle P\rangle = ε_0^2/2R$, $\langle τ\rangle = \langle P\rangle/ω$
and whatever turns the crank must supply that average torque. Open the circuit and the current, and the counter-torque, vanish: only friction is left. This is why a hand generator is easy to turn with nothing connected and hard to turn with a lamp lit. Run the same machine backward, feeding it current, and it is a motor (the Loop Torque Lab).
Eddy currents: the magnet in the copper pipe
Induced currents need no wires. In any solid conductor a changing flux drives currents that swirl within the metal: eddy currents. Drop a strong magnet down a copper pipe and it drifts down slowly, taking seconds to fall a distance it would fall in half a second through plastic. Copper is not magnetic; a magnet held still inside the pipe is not held up at all. The braking comes from the motion.
Picture the pipe’s wall as a stack of rings. Below the falling magnet, the flux through each ring is growing; above it, it is shrinking. So currents circle the pipe below the magnet one way and above it the other way, and by Lenz’s law both push back on the magnet: the ring below repels it, the ring above pulls it back. The faster the magnet falls, the larger the currents and the drag, until the drag balances the weight and the magnet falls at a steady speed. Then its gravitational energy, $Mgv$ every second, all becomes heat in the copper.
Eddy currents brake trains and roller coasters without contact, and the same currents waste energy as heat in the iron of transformers, which is why their cores are made of thin insulated sheets.
Part 2 · Going further
Math you’ll need: calculus throughout: integrals over surfaces and along paths, derivatives, and one first-order differential equation.
Flux as an integral, and a passing magnet
In a non-uniform field the flux is an integral over any surface the loop bounds, $Φ_B = \int\vec B\cdot\hat n\,dA$, or, with the vector potential, a line integral round the loop itself, $Φ_B = \oint\vec A\cdot d\vec l$. For a small magnet, a point dipole of moment $m$, on the axis of a loop of radius $a$ at a distance $z$ from its plane, either route gives
$Φ_B = μ_0ma^2/[2(a^2 + z^2)^{3/2}]$
This is less than the field at the center times the area, because the dipole’s field weakens away from its axis and reverses outside a cone. If the magnet moves along the axis at a steady speed, $ε = −Nv\,dΦ_B/dz$. The flux peaks as the magnet crosses the loop’s plane, so the EMF is zero there and changes sign; the slope, and so the EMF, is largest at $z = ±a/2$, where
$ε_{max} = \tfrac34(\tfrac45)^{5/2}Nμ_0mv/a^2$
A magnet dropped through a coil makes two pulses of opposite sign: the flux rises as it enters and falls as it leaves. Each pulse’s area is the same flux change, but the magnet is faster when it leaves, so the second pulse is taller and narrower. A real magnet of finite size changes these numbers by a few percent; the sim’s engine uses the uniformly magnetized cylinder, with the dipole as a check.
The induced electric field
When a loop sits still and the field through it changes, nothing is moving, so what pushes the charges? A changing magnetic field makes an electric field, and Faraday’s law in its general form says so directly:
$\oint \vec E\cdot d\vec l = −dΦ_B/dt$
Inside a long solenoid of radius $R$ whose field changes at $dB/dt$, symmetry makes $E$ circle the axis with the same strength all the way round. A circle of radius $r$ gives $E\cdot2πr = πr^2\,dB/dt$ inside, and outside, where the field is zero, the circle still encloses all of the solenoid’s changing flux:
$E = \frac r2\frac{dB}{dt}$ inside, $E = \frac{R^2}{2r}\frac{dB}{dt}$ outside
These lines of $E$ close on themselves. No charge makes them, and no potential describes them: carry a charge once round the circle and the field does work on it. That has a startling consequence. Clip two voltmeters to the same two points of a ring around a changing flux, one with its leads on one side, one on the other, and they read different values, both correctly: each reads $\int\vec E\cdot d\vec l$ along its own path, and the two paths differ by the EMF (Romer, 1982). “The voltage between two points” has a meaning only when the field is conservative.
The coasting rod: an exponential, and an impulse
Launch the rod on level rails at $v_0$ and let it go. The only horizontal force is the field’s, so
$m\,\frac{dv}{dt} = −\frac{B^2L^2}{R}v$, $v = v_0e^{−t/τ}$, $τ = mR/(B^2L^2)$
Unlike friction, this force fades with the speed, so the rod never quite stops; yet it goes only a finite distance, $x = \int_0^∞ v\,dt = v_0τ$, and all of its kinetic energy, $\tfrac12mv_0^2$, becomes heat. Taking the first deceleration as constant would give half that distance.
The charge that flows has a neat meaning. $Q = \int I\,dt = BLx/R$, and the field’s impulse on the rod is $\int ILB\,dt = BLQ$, which must equal the momentum lost, $mv_0$. So $Q = mv_0/(BL)$ whatever the resistance: the charge measures the momentum.
Machines: steady torque, commutators and three phases
A turbine applies a steady torque $τ$ to a generator driving a resistance $R$. The rotor speeds up until the average counter-torque, $(NBA)^2ω/(2R)$, balances the turbine, at
$ω = 2Rτ/(NBA)^2$
A heavier load (smaller $R$) means a slower turbine. Because the counter-torque goes as $\sin^2 θ$, the speed ripples a little about this average; the turbine’s work equals the heat plus the rotor’s kinetic energy.
A split ring (a commutator) reverses the coil’s connection each half turn, so the output is $ε_0|\sin ωt|$: one way, but dropping to zero twice a turn, with average $2ε_0/π$. More coils on more segments smooth it: two coils at 90° on four segments, the brushes always on the coil with the larger EMF, average $(2\sqrt2/π)ε_0$ = 0.900 $ε_0$ and never fall below 0.707 $ε_0$. Three coils at 120°, each into its own load, deliver a total power that does not vary at all as the rotor turns, so the torque on the turbine is steady: one reason power grids use three phases.
Eddy currents ring by ring
Slice the pipe (mean radius $a$, wall $w$, conductivity $σ$) into rings of width $dz$. A ring at distance $z$ from the falling magnet has EMF $v\,dΦ/dz$ and resistance $2πa/(σw\,dz)$, so it dissipates $v^2(dΦ/dz)^2σw\,dz/(2πa)$. Summed over the rings, the power is $kv^2$ with $k = (σw/2πa)\int(dΦ/dz)^2dz$, and for a point dipole the integral gives Levin, da Silveira and Rizzato’s result
$k = \tfrac{45}{1024}μ_0^2m^2σw/a^4$, $v_t = Mg/k$
The current circling the pipe in each half is $σwvΦ_0/(2πa)$, with $Φ_0 = μ_0m/(2a)$ the flux through a ring at the magnet. At the terminal speed it does not depend on the metal at all: a better conductor brakes harder, the magnet falls more slowly, and the two cancel.
A real magnet is not a point. Its flux is spread over its length, so the drag is less than the dipole formula says: 0.84 of it for the standard magnet in the ¾-inch tube, and much less for a long magnet, whose flux changes only near its two ends. A thick wall can be summed in layers. A lengthwise slit forces the currents to turn back along the cut; in a thin-sheet model a zero-width slit keeps about 77 % of the drag, a result to compare with a real slotted tube, since real slots are wider.
The same reasoning works for a sheet. A magnet whose axis is normal to a thin sheet, moving at a low speed $v$ at height $h$ above the sheet’s mid-plane, feels
$F = 3μ_0^2m^2σtv/(128πh^4)$
for a point dipole, proportional to $v$, as long as $v ≪ 2/(μ_0σt)$, beyond which the eddy currents’ own field matters.
Faraday’s disk, and the limits of the flux rule
A copper disk of radius $a$ turns at $ω$ in a uniform field $B$ along its axle, with brushes touching the axle and the rim. Each bit of the disk at radius $r$ moves at $ωr$, and its charges feel $qωrB$ along the radius, so
$ε = \int_0^a ωrB\,dr = \tfrac12Bωa^2$
appears between the brushes, steadily: a DC generator with no commutator (Faraday built one in 1831). But no fixed circuit’s flux changes here. The EMF is the magnetic force on moving charges, and the “flux rule” works only if the circuit is drawn through the moving material, sweeping out area as the disk turns. Feynman collects such cases as exceptions to the flux rule. The force law, $\vec F = q(\vec E + \vec v\times\vec B)$, never fails.
Reference
Common mistakes
Part 1
- Taking flux to be the field’s strength: it is the field through an area, and an edge-on loop has none.
- Measuring the angle from the loop’s plane instead of its normal: use $\cos θ$ with $θ$ from $\hat n$.
- Thinking a steady field, or a large flux, makes a current: only a change does.
- Thinking the induced field always opposes the magnet’s field: it opposes the change, and props up a shrinking flux.
- Thinking the induced current helps the motion: it always resists it, which is why you do work.
- Expecting a generator’s EMF to peak where the flux is largest: it peaks where the flux is zero.
- Taking the peak power $ε_0^2/R$ for the average: the average is half of it.
- Thinking copper is magnetic: it brakes a moving magnet only through induced currents.
Part 2
- Thinking an EMF needs a field at the wire: a ring outside a solenoid feels the full EMF of its changing flux.
- Treating magnetic braking like friction: the force falls with the speed, and the coast is exponential.
The equation sheet
| Part 1 | |
|---|---|
| flux | $Φ_B = BA\cos θ$ |
| Faraday | $ε = −N\,ΔΦ_B/Δt$ |
| charge | $Q = NΔΦ_B/R$ |
| rod on rails | $ε = BLv$; $I = ε/R$; $F = ILB$ |
| power | $P = Fv = I^2R$ |
| terminal speed | $v_t = mgR/(B^2L^2)$ |
| generator | $ε = NBAω\sin ωt$; $ε_0 = NBAω$ |
| its load | $\langle P\rangle = ε_0^2/2R$; $\langle τ\rangle = \langle P\rangle/ω$ |
| Part 2 | |
|---|---|
| a dipole on the axis | $Φ_B = μ_0ma^2/[2(a^2 + z^2)^{3/2}]$; $ε_{max} = \tfrac34(\tfrac45)^{5/2}Nμ_0mv/a^2$ at $z = a/2$ |
| induced E | $\oint \vec E\cdot d\vec l = −dΦ_B/dt$: $E = (r/2)\,dB/dt$ inside, $R^2(dB/dt)/(2r)$ outside |
| coasting rod | $v = v_0e^{−t/τ}$, $τ = mR/(B^2L^2)$, $x = v_0τ$, $Q = mv_0/(BL)$ |
| turbine | $ω = 2Rτ/(NBA)^2$ |
| split ring | $\langle ε\rangle = 2ε_0/π$ |
| pipe | $k = \tfrac{45}{1024}μ_0^2m^2σw/a^4$, $v_t = Mg/k$ |
| sheet | $F = 3μ_0^2m^2σtv/(128πh^4)$ |
| Faraday’s disk | $ε = \tfrac12Bωa^2$ |
Symbols: $Φ_B$ magnetic flux (OpenStax writes $Φ_m$); $B$ the field; $A$ an area; $θ$ the angle between $\vec B$ and the normal; $N$ turns; $ε$ an EMF, $ε_0$ a generator’s peak; $Q$ charge; $R$ resistance (in the induced-E formula, the solenoid’s radius); $L$ the rod’s length between the rails (some books write $l$); $v$ speed; $I$ current; $F$ force; $P$ power; $m$ a mass, or a magnet’s moment in A·m²; $M$ a magnet’s mass; $ω$ angular speed in rad/s; $τ$ a torque, or a time constant; $a$ a loop’s or pipe’s radius; $z$ a distance along the axis; $r$ a distance from it; $σ$ conductivity; $w$ a wall’s and $t$ a sheet’s thickness; $h$ height above a sheet; $k$ a drag coefficient.
Sources and further reading
- OpenStax, University Physics Volume 2, chapter 13, “Electromagnetic Induction,” CC BY-NC-SA 4.0 (openstax.org): the Intro track follows its order of topics and its notation, except $Φ_B$ for $Φ_m$ and $L$ for $l$.
- D. J. Griffiths, Introduction to Electrodynamics, chapter 7, §7.2 (4th ed.), for Faraday’s law and the induced electric field.
- Y. Levin, F. L. da Silveira and F. B. Rizzato, “Electromagnetic braking: A simple quantitative model,” Am. J. Phys. 74, 815 (2006), for the pipe.
- R. H. Romer, “What do voltmeters measure? Faraday’s law in a multiply connected region,” Am. J. Phys. 50, 1089 (1982); R. P. Feynman, The Feynman Lectures on Physics, Vol. II, §17-2, “Exceptions to the flux rule.”
- This sim’s Field Manual (H) for the model and its limits; the Flux Counter for electric flux; Force on a Moving Charge for the magnetic force and the copper pipe race; the Jumping Ring Lab and the Loop Torque Lab for rings on iron cores and motors.
Faraday's Law & Generators v1.1 · © 2026 Dr. David Meier · CC BY-NC-SA 4.0 · glidepathscience.org