AC generator

Every time you flip a switch and a bulb glows, somewhere a coil is spinning inside a magnet. An AC generator (alternator) is the machine that converts mechanical energy — turbine blades turned by falling water, expanding steam, or wind — into the alternating electrical energy that runs homes, factories and cities. It is not a new law of physics but a direct, practical payoff of Faraday's law of electromagnetic induction: rotate a coil in a magnetic field, the flux through it keeps changing, and a changing flux always induces an EMF. Because the rotation is steady and repetitive, the induced EMF comes out sinusoidal — the familiar $\varepsilon=\varepsilon_0\sin\omega t$ that underlies every AC circuit calculation later in this unit. This article builds the construction, derives that EMF expression from first principles, and works through the numbers examiners actually ask for: peak EMF, RMS EMF and frequency.

The core insight

Rotating a coil of $N$ turns and area $A$ at angular speed $\omega$ in a uniform field $B$ changes the flux linked with it sinusoidally, and Faraday's law converts that changing flux into an alternating EMF $\varepsilon=\varepsilon_0\sin\omega t$ with peak value $\varepsilon_0=NBA\omega$ .

The core idea

Principle. An AC generator works on electromagnetic induction: as a coil rotates in a uniform magnetic field, the magnetic flux linked with it changes continuously, and this changing flux induces an alternating EMF across the coil (Faraday's law), with the induced EMF's direction at every instant governed by Lenz's law. No external battery is needed — the mechanical work done turning the coil against the magnetic torque is what supplies the electrical energy delivered to the circuit.

Construction — four essential parts. The armature is a coil of many turns of insulated wire wound on a soft-iron core; it is the part in which the EMF is induced, and in the basic textbook picture it is the part that rotates. The field magnet — a strong permanent magnet or an electromagnet — supplies the uniform magnetic field $B$ in which the armature turns. Two slip rings, attached to the rotating shaft and turning with it, provide a continuous electrical path from the spinning coil to the outside world. Two brushes, stationary carbon (or graphite) contacts, press lightly against the slip rings and connect them to the external circuit. The soft-iron core inside the armature is not decorative: it concentrates the field lines through the coil, boosting the flux linkage and hence the EMF for a given $B$ and $N$ .

Working. Picture a rectangular coil of $N$ turns, each of area $A$ , mounted so it can spin about an axis lying in its own plane, perpendicular to a uniform field $B$ . As an external agent (a turbine, a hand-crank, falling water) turns the shaft at constant angular velocity $\omega$ , the angle between the coil's normal and $B$ changes steadily with time, so the flux linked with the coil rises and falls once every rotation. By Faraday's law this changing flux drives an EMF around the coil; because slip rings (not a split-ring commutator) connect the coil to the circuit, that EMF — and the current it drives in a resistive load — reverses direction every half-turn, which is exactly what "alternating" current means. The derivation block below turns this word picture into the standard formula.

Figure

Rectangular coil rotating between the poles of a magnet with slip rings and brushes carrying current to an external resistor

Derivation

Derivation AEMF induced in a rotating coil

The goal is to obtain the exact time-dependence of the EMF produced by a coil of $N$ turns and area $A$ spinning at constant angular velocity $\omega$ inside a uniform field $B$ , starting only from the definition of magnetic flux and Faraday's law, so that the peak value $\varepsilon_0=NBA\omega$ and the shape $\varepsilon=\varepsilon_0\sin\omega t$ both fall out of the same short calculation rather than being quoted.

Step 1. Set up the flux. Let the coil rotate about an axis that lies in its own plane and is perpendicular to $B$ . At any instant, let $\theta$ be the angle between the coil's normal (the vector perpendicular to its plane) and $B$ . The flux linked with one turn is $BA\cos\theta$ , so for $N$ turns the total flux linkage is $\phi=NBA\cos\theta$ .

Step 2. Put in the rotation. Take $t=0$ at the instant the coil's plane is perpendicular to $B$ (flux maximum, $\theta=0$ ). Because the coil turns at constant angular speed, $\theta=\omega t$ at all later times, so $\phi(t)=NBA\cos\omega t$ .

Step 3. Apply Faraday's law. The induced EMF is the negative rate of change of flux linkage, $\varepsilon=-\dfrac{d\phi}{dt}$ . Differentiating $\phi(t)=NBA\cos\omega t$ with respect to $t$ gives $\dfrac{d\phi}{dt}=-NBA\omega\sin\omega t$ , so

\varepsilon = -\frac{d\phi}{dt} = NBA\omega\sin\omega t.

Step 4. Read off the standard form. Writing $\varepsilon_0=NBA\omega$ for the constant coefficient,

\varepsilon = \varepsilon_0\sin\omega t, \qquad \varepsilon_0 = NBA\omega

where: $N$ is the number of turns, $B$ is the field strength (T), $A$ is the coil's area (m²), $\omega$ is the constant angular speed of rotation (rad/s), and $\varepsilon_0$ is the peak EMF (V). The EMF (and hence the current in a resistive load) swings positive and negative once every full rotation, so the AC frequency equals the coil's rotation frequency, $f=\omega/2\pi$ .

Worked example

A rectangular coil of $N=100$ turns, each of area $A=0.05\ \text{m}^2$ , rotates at a constant angular speed $\omega=50\ \text{rad/s}$ in a uniform magnetic field $B=0.4\ \text{T}$ . Find (a) the peak EMF, (b) the RMS EMF, and (c) the frequency of the AC generated.

Step 1. Peak EMF from the derived formula: $\varepsilon_0=NBA\omega=100\times0.4\times0.05\times50$ . Multiplying in order, $100\times0.4=40$ , then $40\times0.05=2$ , then $2\times50=100$ , so

\varepsilon_0 = 100\ \text{V}

Step 2. RMS EMF. For a sinusoidal EMF the RMS value is the peak divided by $\sqrt2$ : $\varepsilon_{rms}=\dfrac{\varepsilon_0}{\sqrt2}=\dfrac{100}{1.414}=70.7\ \text{V}$ . This is the value an AC voltmeter across the terminals would actually read.

Step 3. Frequency of rotation and of the AC: $f=\dfrac{\omega}{2\pi}=\dfrac{50}{6.283}=7.96\ \text{Hz}$ , meaning the coil (and the current direction) completes about 8 full cycles every second.

Final answer: peak EMF $\varepsilon_0=100\ \text{V}$ , RMS EMF $\varepsilon_{rms}=70.7\ \text{V}$ , frequency $f\approx7.96\ \text{Hz}$ .

Competitive stretch

Suppose the same coil is instead required to deliver the standard supply frequency of $50\ \text{Hz}$ , keeping $N$ , $B$ , $A$ fixed. The needed angular speed is $\omega=2\pi f=2\pi\times50=314.16\ \text{rad/s}$ , and the resulting peak EMF would jump to $\varepsilon_0=NBA\omega=100\times0.4\times0.05\times314.16=628.3\ \text{V}$ — over six times larger than in the original example. This shows the coupling built into $\varepsilon_0=NBA\omega$ : for a fixed coil design, spinning faster to hit a standard frequency also forces up the peak voltage, so real generators choose $N$ , $B$ and $A$ together to land on both a standard frequency and a standard voltage simultaneously.

Common traps

Mixing up the phase convention: taking $\theta=0$ at flux-maximum gives $\phi=NBA\cos\omega t$ and $\varepsilon=NBA\omega\sin\omega t$ ; taking $\theta=0$ at flux-zero flips cosine and sine. Either is valid, but stay consistent within one problem.
Dropping the factor $N$ in $\varepsilon_0=NBA\omega$ — a common arithmetic slip when multiplying four quantities together.
Confusing angular speed $\omega$ (rad/s) with frequency $f$ (Hz); the link is $\omega=2\pi f$ , and substituting one for the other directly into $\varepsilon_0=NBA\omega$ gives a wrong answer.
Thinking slip rings and a split-ring commutator do the same job — slip rings (continuous rings) preserve the alternating output; a split-ring commutator reverses connections every half-turn to force a unidirectional (pulsating DC) output instead. The AC generator described here uses slip rings, not a commutator.

Ask the tutor: if you want to see how the phase convention changes the EMF formula from sine to cosine, or work another peak/RMS/frequency numeric, just ask.

Why is the induced EMF sinusoidal and not, say, triangular or constant?

It follows directly from the geometry of rotation. The flux linked with the coil is $\phi=NBA\cos\omega t$ because the projection of the coil's area onto the field direction varies as a cosine of the rotation angle for uniform circular motion. Faraday's law differentiates this flux with respect to time, and the derivative of a cosine is a (scaled) sine. So as long as the coil turns at constant angular speed in a uniform field, the induced EMF is forced to be sinusoidal — it is a direct consequence of calculus applied to circular motion, not an assumption.

Does the armature always rotate, or can the magnet rotate instead?

In the basic principle taught here, the coil (armature) rotates inside a fixed field, and that is the picture to use for deriving $\varepsilon=\varepsilon_0\sin\omega t$ . In real power-station alternators, it is usually easier to keep the armature coil stationary and rotate the magnetic field (the rotor) instead, because large currents can then be drawn from fixed terminals without relying on sliding brush contacts, which wear out and cannot handle very high currents efficiently. The flux-change principle and the resulting EMF expression are identical either way; only which part physically spins changes.

Why do we need a soft-iron core inside the armature coil?

A soft-iron core increases the magnetic flux linked with the coil for the same external field B, because iron is far more easily magnetised than air and concentrates field lines through the winding. Since the induced EMF depends on the flux (through $\phi=NBA\cos\omega t$ here, with B effectively the field within the core), a stronger flux linkage for the same coil and rotation speed gives a larger EMF. It does not change the sinusoidal shape or the formula $\varepsilon_0=NBA\omega$ in principle — it simply lets a practical generator reach a useful EMF without needing an impractically large B, A or N.