Visual walkthrough — Biot-Savart law — magnetic field from current element

We answer one question: a long straight wire carries current $I$. How strong is the magnetic field a distance $a$ away, and which way does it point?

---

Step 1 — Draw the setup and name every symbol

WHAT. Lay the wire along a vertical line. Current $I$ (the amount of electric charge flowing per second, measured in amperes) flows upward. Pick the point $P$ where we want the field, sitting a perpendicular distance $a$ to the side of the wire.

WHY. Before any formula we must fix a picture and give a name to every distance and direction. A formula full of letters means nothing until each letter points at something on a drawing.

PICTURE. In the figure below:

• the white vertical line is the wire;
• $a$ (amber) is the shortest distance from $P$ to the wire — measured along the perpendicular, meeting the wire at a point we call the foot $O$;
• $y$ (cyan) is how far up the wire we are from the foot $O$;
• the little cyan chunk of wire is one current element.

---

Step 2 — Draw the arrow $\vec{r}$ from the wire piece to $P$

WHAT. From the current element at height $y$, draw a straight arrow to the point $P$. Call this arrow $\vec{r}$. Its length is $r$.

WHY. The magnetic field a piece of wire makes at $P$ depends on how far $P$ is from that specific piece and in what direction $P$ lies. That "from-the-piece-to-$P$" information is exactly what $\vec{r}$ stores. Crucially $\vec{r}$ is not measured from the foot $O$ — it starts at the actual chunk of wire, and it is a different arrow for every chunk.

PICTURE. The right triangle jumps out: the wire piece, the foot $O$, and the point $P$ form a right-angled triangle. The vertical leg is $y$, the horizontal leg is $a$, and the slanted side (the hypotenuse) is $r$.

Because it is a right triangle, Pythagoras gives the hypotenuse:

$r=\underset{\text{hypotenuse of the }a,y\text{ triangle}}{\underbrace{\sqrt{a^2+y^2}}}$

• $a^2$ — the horizontal leg squared (fixed, same for every piece),
• $y^2$ — the vertical leg squared (grows as we climb the wire),
• the square root undoes the squaring to give an actual length.

---

Step 3 — Find the angle $\theta$, and see why we even need it

WHAT. Let $\theta$ be the angle between the wire piece $d\vec{l}$ (pointing up) and the arrow $\hat{r}$ (pointing toward $P$).

WHY. The Biot–Savart law says a current element throws out no field straight along its own direction and the most field sideways. So the answer must depend on the angle between "the way the wire points" and "the way $P$ is." That angle is $\theta$. We need a trigonometric ratio to turn a picture-angle into a number.

WHY $\sin$ and not $\cos$ or $\tan$? On a right triangle, $\sin$ of an angle is opposite over hypotenuse. We want the part of the geometry that is across from $\theta$. Looking at the triangle, the side opposite the angle $\theta$ (formed at the wire piece) is the horizontal distance $a$, and the hypotenuse is $r$. So:

$\sin \theta =\frac{\text{opposite}}{\text{hypotenuse}}=\frac{a}{r}=\frac{a}{\sqrt{a^2+y^2}}$

• $a$ — the fixed sideways distance (opposite side),
• $\sqrt{a^2+y^2}=r$ — the hypotenuse,
• the ratio is a pure number between $0$ and $1$ that measures how "sideways" $P$ is from the piece.

PICTURE. The angle $\theta$ sits at the current element, opened between the upward wire and the slanted line to $P$.

---

Step 4 — Write the field of ONE piece

WHAT. Plug $r$ and $\sin \theta$ into the magnitude form of the Biot–Savart law. The field from one element is

=\frac{\mu_0 I}{4\pi}\cdot\frac{dy\cdot \dfrac{a}{\sqrt{a^{2}+y^{2}}}}{a^{2}+y^{2}} =\frac{\mu_0 I}{4\pi}\,\frac{a\,dy}{(a^{2}+y^{2})^{3/2}}$$ Term by term: - $\dfrac{\mu_0}{4\pi}$ — a fixed constant of nature ($\mu_0=4\pi\times10^{-7}\ \text{T·m/A}$); it just sets the units and strength. - $I$ — the current (bigger current, bigger field). - $dl=dy$ — the tiny length of *this* piece. - $\sin\theta=\dfrac{a}{\sqrt{a^2+y^2}}$ — the "sideways-ness" from Step 3. - $\dfrac{1}{r^2}=\dfrac{1}{a^2+y^2}$ — the distance weakening, same spreading as [[Coulomb's Law]]. - Combining the $\sqrt{a^2+y^2}$ (from $\sin\theta$) with $(a^2+y^2)$ (from $r^2$) gives $(a^2+y^2)^{3/2}$ downstairs — the power $\tfrac32$ is just $1$ (from $r^2$) plus $\tfrac12$ (from the square root). **WHY.** Everything is now written in **one** variable, $y$. That is the whole point of Steps 2–3: to trade the picture for a formula that depends only on how high up the wire we are. **PICTURE.** The tiny field arrow $d\vec B$ at $P$ points **out of the page** (right-hand rule: fingers curl from $d\vec l$ to $\hat r$). Every piece of the wire, above and below, gives a $d\vec B$ pointing the *same* way out of the page. ![[deepdives/dd-physics-1.8.22-d2-s04.png]] > [!intuition] Why we may soon add plain numbers, not arrows > All the little $d\vec B$ arrows are parallel (all out of the page at $P$). Parallel arrows add like ordinary numbers — no sideways cancellation. That is what makes the next step a simple sum of magnitudes. --- ## Step 5 — Add up every piece (the integral) **WHAT.** "Adding up infinitely many tiny pieces" is exactly what the symbol $\displaystyle\int$ means — a long, stretched-out **S** for "Sum." We sum from the bottom of the wire ($y=-\infty$) to the top ($y=+\infty$): $$B=\int dB=\frac{\mu_0 I a}{4\pi}\int_{-\infty}^{\infty}\frac{dy}{(a^{2}+y^{2})^{3/2}}$$ - The constants $\dfrac{\mu_0 I a}{4\pi}$ come out front (they do not change from piece to piece). - The $\displaystyle\int_{-\infty}^{\infty}$ says: run $y$ over the whole infinite wire and total the contributions. **WHY.** One piece makes a whisper of field; the wire is the *chorus* of all pieces. Integration is the machine that sums a whisper repeated infinitely many times into a finite, real number. **PICTURE.** Think of stacking the $dB$ contributions as a curve: it is tallest at the foot ($y=0$, closest, most sideways) and dies away up and down the wire. The **area under that curve** is the total field $B$. ![[deepdives/dd-physics-1.8.22-d2-s05.png]] The standard result of this sum is $$\int\frac{dy}{(a^{2}+y^{2})^{3/2}}=\frac{y}{a^{2}\sqrt{a^{2}+y^{2}}}$$ Check the limits: - as $y\to+\infty$, the fraction $\dfrac{y}{a^2\sqrt{a^2+y^2}}\to \dfrac{y}{a^2\cdot y}=\dfrac{1}{a^2}$; - as $y\to-\infty$, it $\to -\dfrac{1}{a^2}$; - the total is the top minus the bottom: $\dfrac{1}{a^2}-\left(-\dfrac{1}{a^2}\right)=\dfrac{2}{a^2}$. --- ## Step 6 — Collect the answer **WHAT.** Put the number $\dfrac{2}{a^2}$ back: $$B=\frac{\mu_0 I a}{4\pi}\cdot\frac{2}{a^{2}}=\boxed{\dfrac{\mu_0 I}{2\pi a}}$$ - one factor of $a$ on top cancels one of the two $a$'s on the bottom; - the $2$ and the $4\pi$ combine to $2\pi$; - the field ends up shrinking as $1/a$ — **not** $1/r^2$. **WHY it fell to $1/a$ (not $1/r^2$).** Each piece gave $1/r^2$, but there are *more far-away pieces* the further out you look, and their contributions add up. Summing infinitely many $1/r^2$ whispers along an infinite line softens the fall-off to $1/a$. This is a genuine surprise that only the integral reveals. **PICTURE.** The finished field: concentric circles wrapping the wire, spaced further apart (weaker) as you move out — the $1/a$ decay made visible. ![[deepdives/dd-physics-1.8.22-d2-s06.png]] --- ## Step 7 — The degenerate & finite cases (never leave a gap) **WHAT.** Real wires are finite. If $P$ sees the two ends of the wire at angles $\theta_1$ and $\theta_2$ (each measured from the perpendicular foot up to an end), the same integral, stopped at finite limits, gives $$B=\frac{\mu_0 I}{4\pi a}\left(\sin\theta_1+\sin\theta_2\right)$$ **WHY / cover all cases.** - **Infinite wire** — both ends go to $90^\circ$: $\sin90^\circ+\sin90^\circ=2$, giving $\dfrac{\mu_0 I}{4\pi a}\cdot 2=\dfrac{\mu_0 I}{2\pi a}$. ✔ Matches Step 6. - **Semi-infinite wire** (one end at $P$'s level, one end at infinity): $\sin 0^\circ+\sin 90^\circ=0+1=1$, giving **exactly half** the infinite-wire field, $\dfrac{\mu_0 I}{4\pi a}$. - **$a\to 0$** (right on the wire): $B\to\infty$ — the formula warns you it breaks down at the wire itself, where a real wire has thickness. **PICTURE.** The two end-angles $\theta_1,\theta_2$ drawn on a short wire, with a slider showing them opening toward $90^\circ$ as the wire grows. ![[deepdives/dd-physics-1.8.22-d2-s07.png]] --- ## The one-picture summary Every step in a single frame: the right triangle ($a$, $y$, $r$), the angle $\theta$ with $\sin\theta=a/r$, one out-of-page $d\vec B$, the sum along the wire, and the boxed result. ![[deepdives/dd-physics-1.8.22-d2-s08.png]] > [!recall]- Feynman retelling — the whole walkthrough in plain words > Stand next to a straight wire with electricity running through it. Chop the wire into a zillion tiny bits. Each bit throws a puff of magnetic "swirl" toward you. How big a puff? It's stronger if the bit is close (that's the $1/r^2$), and it's biggest for the bit *right beside* you and vanishes for bits far up or down that are aiming almost straight at you (that's the $\sin\theta$). We measured the distance to each bit with the right triangle whose sides are $a$ (sideways), $y$ (up the wire), and $r$ (the slanted line). Every puff points the same way — out of the page — so we just pile them up. "Piling up infinitely many tiny things" is what the integral sign does. When we finish the pile, the two $a$'s and the $2$ tidy up and we get $B=\mu_0 I/(2\pi a)$: the field fades as one-over-distance, gentler than a single point would, because there's always more wire far away pitching in. > [!mnemonic] The shape of the derivation > **Triangle → sine → one-piece → sum → tidy.** Draw the triangle, take the sine for sideways-ness, write one piece, integrate the pile, clean up the constants. --- ## Connections - [[1.8.22 Biot-Savart law — magnetic field from current element (Hinglish)|Parent topic]] — the law and its other worked cases. - [[Ampère's Law]] — gets this same $\mu_0 I/2\pi a$ in one line by symmetry, skipping the integral. - [[Right-Hand Rule]] — fixes the out-of-page direction of every $d\vec B$. - [[Magnetic Field of a Solenoid]] — the same integrating trick, stacking loop fields. - [[Coulomb's Law]] — the $1/r^2$-per-piece cousin without the cross product. - [[Lorentz Force]] — what this field then does to a moving charge. - [[Magnetic Dipole Moment]] — the far-field character of loop versions of this integral.