Visual walkthrough — Magnetic flux Φ = ∫B·dA

Step 1 — Draw the field as lines you can count

WHAT. Before any formula, picture a magnetic field. We draw it as a set of parallel arrows — field lines — all pointing the same way, evenly spaced. Look at the field as an invisible "flow" that these arrows make visible.

WHY lines and not numbers? Because our whole goal is counting. If we agree that a stronger field = more tightly packed lines, then "field strength" becomes something you can literally count: lines per square metre. That trick — turning a strength into a density of countable things — is what makes flux visual.

PICTURE. In the figure, the arrows are the field $\mathbf{B}$. Where they crowd together the field is strong; where they spread out it is weak. The little number "lines per box" is our stand-in for the strength $B$.

Step 2 — Hold a surface face-on: flux is just $B\times A$

WHAT. Now push a flat surface — think of a wire loop's opening — straight into the field so the surface faces the field head-on. Every arrow that reaches the surface passes through it.

WHY multiply? If each square metre catches $B$ lines, and the surface has $A$ square metres, then total lines caught $=B\times A$. That's the definition of area: how many unit tiles fit inside. We just count tiles and multiply.

PICTURE. The green square in the figure is fully "lit up" by arrows — none miss it, none slide past.

$\Phi =\underset{\begin{array}{c}\text{lines per}\\ \text{square metre}\end{array}}{\underbrace{B}}\times \underset{\begin{array}{c}\text{number of}\\ \text{square metres}\end{array}}{\underbrace{A}}$

Read it as: (density of lines) × (how much area) = (total lines through). That total is the flux $\Phi$.

Step 3 — The normal: give the surface a pointing spike

WHAT. Before we tilt anything, we must give the surface a direction. A flat surface, on its own, doesn't point anywhere. So we attach a spike sticking straight out of it, perpendicular to its plane. That spike is the normal, written $\hat{\mathbf{n}}$ (the little hat means "length 1 — it only shows direction").

WHY do we need it? Because "how the surface is tilted" has to be measured against something. We can't measure the tilt using the flat plane (a plane doesn't have a single direction). The one honest direction a flat surface owns is the perpendicular spike. So the normal becomes the surface's official direction.

PICTURE. In the figure the purple spike $\hat{\mathbf{n}}$ rises out of the middle of the surface. When the surface faces the field, $\hat{\mathbf{n}}$ and $\mathbf{B}$ point the same way — remember that, it's the "face-on" reference.

Step 4 — Tilt it: only the shadow counts

WHAT. Rotate the surface by an angle $\theta$ — measured between the field $\mathbf{B}$ and the normal $\hat{\mathbf{n}}$. Now the surface no longer faces the field squarely. Fewer arrows pass through.

WHY does tilting lose lines? Watch the field lines in the figure: as you rotate, the surface presents a narrower shadow to the incoming lines. The lines only "see" the shadow — the projection of the surface onto a face-on plane. So the effective catching area shrinks.

PICTURE. The figure shows the tilted surface (coral) and, dropped beneath it, its shadow (dashed) cast by the field. The shadow width is the real slab of area the lines pierce.

How big is that shadow? Drop the tilted surface's edge onto the face-on direction and you form a right triangle:

• the tilted surface is the hypotenuse, length $A$,
• the shadow is the adjacent side to the angle $\theta$.

The ratio "adjacent over hypotenuse" is exactly what cosine means. So

$\text{shadow area}=A\cos \theta .$

Multiply the seed formula by this shrink factor:

$\Phi =\underset{{\text{lines / m}}^2}{\underbrace{B}}\underset{\text{true area}}{\underbrace{A}}\underset{\begin{array}{c}\text{fraction}\\ \text{facing field}\end{array}}{\underbrace{\cos \theta }}$

Step 5 — Every case at once: the three tilts

WHAT. Let's check the tilt formula at every important angle, so no situation surprises you later.

WHY check the extremes? A formula you trust is one you've watched behave at its edges. Cosine runs smoothly from $1$ down to $0$ down to $-1$; each value has a physical picture.

PICTURE. Three surfaces side by side in the figure:

The negative case matters. When $\theta$ passes $90^{\circ }$, cosine goes negative. Physically the normal now points against the field — the lines enter through the back of the surface. The sign of the flux records which face the lines go in through. Flux isn't just "how many lines," it's "how many, and from which side."

Step 6 — Compress it into a dot product

WHAT. We have two arrows now: the field $\mathbf{B}$ and the area vector $\mathbf{A}=A\hat{\mathbf{n}}$. The quantity $BA\cos \theta$ — (length of one)(length of other)(cosine of angle between) — is the definition of the dot product $\mathbf{B}\cdot \mathbf{A}$.

WHY switch to a dot product? Three payoffs, all visible:

1. It hides the angle. We no longer measure $\theta$ by hand; the dot product extracts "how much they align" automatically.
2. It returns a single number (a scalar) — exactly right, because flux is one number, not a direction.
3. It carries the sign for free — anti-aligned arrows give a negative dot product, matching the flipped-surface case from Step 5.

PICTURE. The figure shows $\mathbf{B}$ and $\mathbf{A}$ as two arrows from one point, the angle $\theta$ between them, and the projection of one onto the other — the dot product is that projection times length.

$\Phi =\mathbf{B}\cdot \mathbf{A}=\underset{B}{\underbrace{|\mathbf{B}|}}\underset{A}{\underbrace{|\mathbf{A}|}}\cos \theta$

Step 7 — When the field or surface won't sit still: slice it

WHAT. Real fields get stronger in one corner; real surfaces curve like a dome. Then no single $B$ and no single $\theta$ describe the whole thing. Fix: chop the surface into tiny patches $d\mathbf{A}$, each so small that over it the field looks uniform and the patch looks flat.

WHY does slicing rescue us? On a tiny enough patch, Step 6 applies perfectly — one field value, one normal, one shadow. We compute $\mathbf{B}\cdot d\mathbf{A}$ on each patch (a tiny flux), then add them all up. As the patches shrink to zero size, that sum becomes an integral. This is the whole idea of calculus: hard-because-it-varies → easy-tiny-pieces → sum → limit.

PICTURE. The figure shows a curved surface tiled with little patches; each has its own outward spike $d\mathbf{A}$ pointing a slightly different way, and its own local field arrow. One patch is magnified to show it's just a Step-6 problem.

$\Phi ={\sum }_{\text{patches}}\mathbf{B}\cdot d\mathbf{A}\underset{\text{patches}\to 0}{\to }{\int }_S\mathbf{B}\cdot d\mathbf{A}$

Term by term: $\mathbf{B}$ is the local field on a patch, $d\mathbf{A}$ is that patch's tiny area-vector (size × its own normal), the dot product is the tiny flux through it, and ${\int }_S$ says "sum over the whole surface $S$."

Step 8 — The degenerate seal: any closed surface gives zero

WHAT. Wrap the surface completely shut — a bag, a sphere — so it has an inside and an outside. Add up the flux over the entire closed surface (the $\oint$ symbol means "integrate over a closed surface").

WHY zero? Magnetic field lines never start or stop; they form closed loops (there are no magnetic monopoles — see Gauss's Law for Magnetism). So any line that pokes into the bag must poke back out somewhere. Entering counts negative (line comes in through the surface, opposite the outward normal), leaving counts positive. They cancel in pairs.

PICTURE. The figure shows a closed loop of field threading a sphere: one arrow enters (flux $-$), the same line exits (flux $+$). Net contribution: nothing.

${\oint }_S\mathbf{B}\cdot d\mathbf{A}=0$

The one-picture summary

Here is the whole journey in a single frame: a line-density field (Step 1), a surface with its normal spike (Step 3), tilted so only its shadow catches lines (Step 4), read as a dot product (Step 6), and — for non-uniform cases — sliced into patches summed by an integral (Step 7).

Active recall