🔥 JEE Main 2027 Session 1 is in January — start the JEE Mains Physics Test Series →

CLASS 12 · MATHEMATICAL TOOLS

Calculus for Electromagnetism

Electromagnetism runs on exactly two calculus moves you already know from kinematics: differentiation gives you a rate (how fast flux is changing, which is EMF) and integration gives you a total (how much charge has flowed, or how a non-uniform distribution adds up). This page applies both squarely to flux, EMF, and charge, and derives the one result, Faraday's law, that the rest of the induction chapter is built on.

01 · See it

Watch it happen

0.760 Wb-turns
Flux-linkage Φ at this t
0.400 Wb-turns/s
Slope, dΦ/dt (blue curve)
-0.400 V
Induced EMF, −dΦ/dt

Here the loop stays flat and fixed while the field ramps up at a steady rate, so the flux line is straight and its slope (and hence the EMF) never changes. Drag the dB/dt slider and watch the line's steepness, and the EMF readout, move together.

Switch between a loop sitting still in a ramping field and a loop spinning in a steady field. Either way, watch the red tangent line on the Φ(t) curve: its slope is dΦ/dt, and the EMF readout is always just that slope with its sign flipped. Where the blue curve is flattest, the EMF is weakest; where the curve is steepest, the EMF peaks.

02 · Derive it

Where it comes from

Definite integration, applied to a genuine EM problem. Suppose a long straight wire of length LL carries a current whose density is not uniform but varies across its cross-section, or more simply, suppose a coil’s induced EMF itself varies with time as a current flows through a resistor and you want the total charge that passed through. Charge is current accumulated over time, exactly the “area under a curve” idea from the graphs page, so:

q=∫t1t2I(t) dtq = \int_{t_1}^{t_2} I(t)\,dt

This is definite integration: you are not just asking for “the antiderivative,” you are evaluating that antiderivative at an upper bound, evaluating it again at a lower bound, and subtracting. That subtraction is what turns a family of antiderivatives into one specific number tied to one specific interval, which is exactly what a physical quantity like charge has to be.

Rate of change of magnetic flux, derived properly. Start from the definition of flux through a flat loop of area AA sitting in a field BB at angle θ\theta to the loop’s normal:

Φ=BAcos⁡θ\Phi = BA\cos\theta

Any of BB, AA, or θ\theta can depend on time. Faraday’s experimental finding was that whenever the flux through a circuit changes, an EMF appears around that circuit, and the size of that EMF is set by how fast the flux is changing, not by how large the flux is. In the language of the graphs page: the induced EMF is the slope of the flux-time graph, not the height of the curve itself. For a coil of NN turns, each turn links essentially the same flux, so the effects add:

EMF∝− d( NΦ )dt=−NdΦdt\text{EMF} \propto -\,\dfrac{d(\,N\Phi\,)}{dt} = -N\dfrac{d\Phi}{dt}

Choosing SI units makes the constant of proportionality exactly 1, and the minus sign is Lenz’s law, built in so that the induced effect always opposes the change that caused it:

Faraday's law of electromagnetic induction
ε=−NdΦdt=−Nddt(BAcos⁡θ)\varepsilon = -N\dfrac{d\Phi}{dt} = -N\dfrac{d}{dt}\big(BA\cos\theta\big)

Expand that derivative with the product rule and every induction scenario in the syllabus falls out as a special case. If only BB varies (A and θ\theta constant, a coil sitting still in a changing field):

ε=−NAcos⁡θ dBdt\varepsilon = -NA\cos\theta\,\dfrac{dB}{dt}

If only AA varies (a loop of fixed shape being stretched or a rod sliding to enlarge a circuit, in a steady field head-on):

ε=−NB dAdt\varepsilon = -NB\,\dfrac{dA}{dt}

And if only θ\theta varies (a coil spinning at constant angular speed ω\omega in a steady field, exactly how an AC generator works, with θ=ωt\theta = \omega t):

ε=−NBA ddt(cos⁡ωt)=NBA ωsin⁡ωt\varepsilon = -NBA\,\dfrac{d}{dt}(\cos\omega t) = NBA\,\omega\sin\omega t

Three completely different physical setups, one single derivative, just applied to whichever factor in BAcos⁡θBA\cos\theta happens to be doing the changing.

Rate of change of electric flux, set up (not derived here). Electric flux through a surface is defined the same way as magnetic flux, swap BB for EE: ΦE=∫E⃗⋅dA⃗\Phi_E = \int \vec{E}\cdot d\vec{A}. Ampere’s law, in its original form, related the magnetic field circulating around a loop only to the conduction current piercing it, and that works fine for a steady current in a wire. It visibly breaks down for a circuit like a charging capacitor, where current flows in the wires but nothing physically crosses the gap between the plates, and yet a magnetic field is still measured looping around that gap. What is changing in that gap is the electric flux, as charge piles up on the plates, and Maxwell showed dΦE/dtd\Phi_E/dt there behaves exactly like a current for the purposes of generating a magnetic field. This “displacement current” term, and the full corrected form of Ampere’s law, is derived properly on the Electromagnetic Waves page; the point to take from here is just why a rate of change of flux, the same calculus move as Faraday’s law, is the quantity that had to enter the fix.

03 · Break it

Where the shortcut stops working

The most common shortcut is assuming EMF only appears when the field itself is changing. But Φ=BAcos⁡θ\Phi = BA\cos\theta has three factors, and the derivative can be non-zero because AA or θ\theta is changing even while BB sits perfectly constant. A coil spinning at constant angular speed in a steady, unchanging field is the textbook case: dB/dt=0dB/dt = 0 everywhere, yet the EMF is very much non-zero, because cos⁡(ωt)\cos(\omega t) is what’s doing the changing. This is exactly how an AC generator produces an alternating EMF from a perfectly constant magnet.

A second shortcut is forgetting the sign entirely and just reporting a magnitude. That throws away Lenz’s law, which the minus sign in ε=−N dΦ/dt\varepsilon = -N\,d\Phi/dt is encoding. The sign tells you the direction of the induced current (via the induced field it sets up), and an exam question asking “in which direction does the induced current flow” cannot be answered from the magnitude alone.

A third shortcut, when the flux-time relationship is not a straight line, is averaging the start and end flux and calling that “the” slope. That gives you only the average rate over the whole interval (a secant), not the instantaneous EMF at a specific moment (a tangent), exactly the secant-versus-tangent trap from the graphs page. If the question asks for the EMF “at t=2 st = 2\text{ s},” you need the derivative evaluated at that instant, not an average smeared across the whole interval.

04 · Master it

Apply it under exam conditions

A flat circular coil of N=120N = 120 turns and radius r=0.05 mr = 0.05\text{ m} sits face-on to a uniform magnetic field that increases with time as B(t)=0.2+0.6tB(t) = 0.2 + 0.6t (in tesla, tt in seconds). The coil’s total resistance is R=4 ΩR = 4\ \Omega. Find (a) the induced EMF at any time (it will turn out to be constant here), and (b) the total charge that flows through the coil during the first 2 s2\text{ s}.

Setting up. The coil’s area is fixed and it stays face-on to the field, so only BB changes with time; θ=0\theta = 0 throughout, so cos⁡θ=1\cos\theta = 1.

A=πr2=π(0.05)2=7.854×10−3 m2A = \pi r^2 = \pi (0.05)^2 = \textcolor{#e08a1e}{7.854\times10^{-3}\text{ m}^2}

(a) Differentiate the flux-linkage to get the EMF:

Φ(t)=B(t)A=(0.2+0.6t)(7.854×10−3)\Phi(t) = B(t)A = (0.2 + 0.6t)(7.854\times10^{-3})ε=−NdΦdt=−NAdBdt=−(120)(7.854×10−3)(0.6)\varepsilon = -N\dfrac{d\Phi}{dt} = -N A\dfrac{dB}{dt} = -(120)(7.854\times10^{-3})(0.6)ε=−0.5655 V (magnitude 0.565 V, constant)\varepsilon = \textcolor{#e08a1e}{-0.5655\text{ V (magnitude }0.565\text{ V, constant)}}

It comes out constant because dB/dt=0.6 T/sdB/dt = 0.6\text{ T/s} is itself constant, a straight-line Φ(t)\Phi(t) graph has the same slope everywhere.

(b) Charge is current integrated over time, and current is ε/R\varepsilon/R. Since ε\varepsilon here does not depend on tt, the definite integral is just a constant times the interval width, but written out properly as an integral:

q=∫02∣ε∣R dt=∣ε∣R[t]02=0.56554(2−0)q = \int_0^2 \dfrac{|\varepsilon|}{R}\,dt = \dfrac{|\varepsilon|}{R}\big[t\big]_0^2 = \dfrac{0.5655}{4}(2 - 0)q=0.2827 Cq = \textcolor{#e08a1e}{0.2827\text{ C}}

Cross-check with the flux-linkage shortcut, which only works because ε\varepsilon is constant here: q=NΔΦ/Rq = N\Delta\Phi/R. The flux per turn rises from B(0)A=1.571×10−3 WbB(0)A = 1.571\times10^{-3}\text{ Wb} to B(2)A=1.100×10−2 WbB(2)A = 1.100\times10^{-2}\text{ Wb}, a change of ΔΦ=9.425×10−3 Wb\Delta\Phi = 9.425\times10^{-3}\text{ Wb} per turn, giving q=(120)(9.425×10−3)/4=0.2827 Cq = (120)(9.425\times10^{-3})/4 = 0.2827\text{ C}, matching the integral exactly.

05 · FAQs

Quick answers

What does it actually mean for flux to 'change'?+

Flux, Φ = B·A·cos(θ), bundles three things together: the field strength B, the loop's area A, and the angle θ between the field and the loop's normal. 'Changing flux' just means this product is changing, and it does not care which of the three is doing the changing. A field ramping up, a loop physically growing or shrinking, or a loop tilting or spinning all change Φ exactly the same way, as far as the calculus is concerned.

Why is there a minus sign in EMF = −dΦ/dt?+

The minus sign is Lenz's law folded into the formula: the induced EMF always drives a current whose own magnetic field opposes the change in flux that created it, never reinforces it. Drop the sign and you still get the right magnitude, but you lose the direction, which is exactly why exam answers ask for both.

Why does a changing electric flux matter if this chapter is about magnetism?+

Ampere's original law related a magnetic field only to the conduction current threading a loop, and it quietly breaks down wherever charge is accumulating instead of flowing, such as between capacitor plates. Maxwell's fix was to show that a changing electric flux there acts exactly like an extra current term, since both are rates of change that can source a magnetic field.

Is 'rate of change of electric flux' the same kind of calculus as Faraday's law?+

Yes, structurally identical: both are d(flux)/dt applied to a surface. The physical payoff is different (one gives an EMF via Faraday's law, the other patches Ampere's law via displacement current) but the calculus move, differentiating a flux integral with respect to time, is the same tool used twice.

When do I need definite integration instead of just multiplying?+

The instant the quantity you are integrating is not constant across the interval. A uniform current for 5 seconds gives charge = current × time, no calculus needed. A current that itself varies with time needs q = ∫I dt evaluated between the two time bounds, because 'multiply by the width' only works on a rectangle, not a curve.