Angular Momentum & Its Conservation

A figure skater winds into her final spin turning lazily — maybe twice a second, arms and free leg stretched wide. Then she pulls everything in tight, and with nobody touching her, no motor, no push against the ice, she blurs to five times the spin rate. Ask her to slow down and she simply opens her arms again. Whatever she is trading against, it is being exchanged at a fixed total — and that total is this module's quantity: angular momentum, $L = I\omega$ , the rotational currency that completes the dictionary begun three modules ago ( $F \to \tau$ , $m \to I$ , and now $p \to L$ ). Angular momentum earns its keep through its conservation law, one of the small handful of exam superpowers in the whole syllabus. When no net external torque acts, $L$ is frozen — so a body that reshapes itself must trade moment of inertia against spin rate, $I_1\omega_1 = I_2\omega_2$ . That one line runs the skater, the tucking diver, the cat landing on its feet, spinning turntable problems, coupling clutch plates — and, at the cosmic end, the collapse of a burnt-out star into a pulsar spinning hundreds of times a second, the skater's trick performed with a stellar corpse. There is a sting in the tail that examiners adore: conservation of $L$ is not conservation of energy. The skater who halves her moment of inertia doubles her kinetic energy — her muscles pay the difference, honestly. Boards test the law itself (CBSE/ICSE/NEET: skater and disc problems), JEE adds the particle form $L = mvr$ and turntable variants, AP and IB-HL probe the "L conserved, K not" distinction. All of it is here.

Key takeaways

Angular momentum of a body spinning about a fixed axis: $L = I\omega$ — the rotational twin of $p = mv$ . Units: $\mathrm{kg\,m^2/s}$ .
For a single particle: $L = mvr_\perp$ — momentum times the perpendicular distance from the axis to the particle's line of motion (JEE lane; even a straight-moving particle carries $L$ ).
The deepest form of the rotational second law: $\tau_{\text{net}} = \dfrac{\Delta L}{\Delta t}$ — it survives even when $I$ changes, which $\tau = I\alpha$ quietly does not.
Angular impulse: $\tau\,\Delta t = \Delta L$ — the rotational twin of impulse, for sudden events.
Conservation: if the net external torque about an axis is zero, $L$ about that axis is constant: $I_1\omega_1 = I_2\omega_2$ — the skater's law.
$L$ conserved does not mean kinetic energy conserved: $K = \dfrac{L^2}{2I}$ rises when $I$ falls — the skater's muscles supply the difference.
Key idea: $L = I\omega$ is rotation's conserved currency — no external torque, no change; a reshaping body trades $I$ against $\omega$ at fixed $L$ , while its energy $L^2/2I$ moves in whatever direction the reshaping pays for.

The core insight

About a fixed axis, a rigid body carries angular momentum $L = I\omega$ and obeys $\tau_{\text{net}} = \Delta L/\Delta t$ ; when the net external torque vanishes, $L$ is conserved — so reshaping trades moment of inertia against spin rate ( $I_1\omega_1 = I_2\omega_2$ ) while the kinetic energy $L^2/2I$ changes by exactly the internal work done.

Where this lives today

Spacecraft steer with this module. Reaction wheels aboard Hubble and JWST spin up one way so the observatory turns the other — total $L$ frozen, orientation traded internally without a gram of propellant; "momentum dumping" (unloading accumulated wheel spin against magnetic torquers) is a standing chore of every mission. Helicopters are a conservation problem solved in hardware: torque the main rotor one way and the fuselage would counter-rotate — the tail rotor exists purely to supply the balancing external torque. Divers, gymnasts and cats run mid-air $I$ -management: tuck to spin fast, stretch to stop the somersault at entry, the cat twisting its two body halves against each other to land feet-down with total $L$ still zero. And at the observatory end, millisecond pulsars — stellar corpses spinning hundreds of times a second exactly as Example 4 predicts — serve as clocks so stable they are used to hunt gravitational waves: the skater's trick, keeping time for the universe.

The physical picture

Watch the skater with dictionary eyes. Linear mechanics has a "quantity of motion," $p = mv$ , guarded by a conservation law when no external force acts. Rotation needs its own: multiply each particle's momentum by its lever arm about the axis — $m_i v_i \times r_i$ — and sum. With every particle sharing the body's spin ( $v_i = r_i\omega$ ), the sum is $\sum m_i r_i^2\,\omega = I\omega$ . The same swarm bookkeeping as ever, and the same $\sum mr^2$ making its third appearance: resister of $\alpha$ , warehouse of $K$ , and now the exchange rate between shape and spin.

Figure

The skater's trade — I down, omega up, the product frozen

Why should this quantity be conserved? Because of the law you met two modules ago in its deeper form. The rotational second law, honestly stated, is $\tau_{\text{net}} = \Delta L/\Delta t$ : torque is the rate of change of angular momentum, exactly as force is the rate of change of $p$ . Set the net external torque to zero and $L$ cannot budge — no matter what happens inside the body. Internal forces (her muscles hauling her arms inward) come in third-law pairs whose torques cancel; they can reshape the body, redistribute its mass, change $I$ drastically — but they cannot touch the total $L$ . So when $I$ drops, $\omega$ must rise in exact inverse proportion. The skater is not creating spin; she is concentrating it. Now the sting: follow the energy. Kinetic energy in terms of $L$ is $K = \tfrac12 I\omega^2 = L^2/2I$ — and at fixed $L$ , smaller $I$ means more energy. The arms-in skater is more energetic than the arms-out skater, and no theorem is violated: pulling her arms in means hauling circling masses inward against their tendency to fly outward, and that haul is genuine muscular work — exactly the $\Delta K$ . Run the film backwards: letting the arms fly out, her muscles act as brakes, absorbing energy as she slows. Conservation of $L$ and conservation of $K$ are independent claims; torque-free guarantees the first always, the second only for a body that never reshapes.

The math: the currency, the law, the conservation

The angular momentum of a rigid body about its fixed rotation axis:

L = I\,\omega \qquad (1)

where: $I$ — moment of inertia about the axis; $\omega$ — angular speed in $\mathrm{rad/s}$ ; units $\mathrm{kg\,m^2/s}$ (equivalently $\mathrm{J\,s}$ ). Feel a number: the skater of the concept module, arms out ( $I = 4.74\ \mathrm{kg\,m^2}$ ) at $2$ revolutions per second ( $\omega = 12.6\ \mathrm{rad/s}$ ), carries $L \approx 60\ \mathrm{kg\,m^2/s}$ — and keeps carrying it through every pose until the ice's small friction torque slowly collects it. For a single particle — JEE's favourite extension — the definition is momentum times lever arm:

L = m\,v\,r_\perp \qquad (2)

where: $r_\perp$ — the perpendicular distance from the axis (or reference point) to the particle's line of motion, exactly the lever-arm reading you learned for torque. Two readings worth owning. A particle circling at radius $r$ : $r_\perp = r$ , so $L = mvr = mr^2\omega$ — the atomic case of (1). A particle moving in a straight line past a point: it still has $L = mvr_\perp$ , constant as it cruises (same $v$ , same perpendicular drop to the same line) — angular momentum does not require anything to be rotating, only to have motion with a lever arm. The dynamics, in its deepest fixed-axis form:

\tau_{\text{net}} = \frac{\Delta L}{\Delta t} \qquad (3)

where: $\tau_{\text{net}}$ — net external torque about the axis. For a rigid body ( $I$ constant), $\Delta L/\Delta t = I\,\Delta\omega/\Delta t = I\alpha$ , recovering last module's law — but (3) is the senior statement, because it stays true when the body reshapes and $I$ itself changes, which $\tau = I\alpha$ cannot handle. Multiplying by the time interval gives the angular impulse form, $\tau\,\Delta t = \Delta L$ — the tool for sudden events (a flywheel hit by a brief brake, a particle striking a door), the rotational twin of $F\Delta t = \Delta p$ . Set the external torque to zero and (3) delivers the module's crown:

\tau_{\text{net}}^{\text{ext}} = 0 \quad \Longrightarrow \quad L = \text{constant}, \qquad I_1\,\omega_1 = I_2\,\omega_2 \qquad (4)

where: states 1 and 2 are any two moments of the torque-free motion. The power of (4) is that it holds across reshapings, couplings, and internal chaos — arms pulled in, discs dropped onto discs, people walking on turntables — because internal torques cancel in pairs. And the energy rider, the sting examiners test:

K = \frac{L^2}{2I} \qquad (5)

where: at fixed $L$ , energy runs inversely to $I$ — halve the moment of inertia and the kinetic energy doubles, the difference paid (or absorbed) by whatever internal agency did the reshaping.

you want	use	notes
$L$ of a spinning body	$L = I\omega$ , eq. (1)	about a named axis, rad/s
$L$ of a particle (JEE)	$L = mvr_\perp$ , eq. (2)	lever arm to the LINE of motion
torque acting over time	$\tau\Delta t = \Delta L$	sudden events, brakes, strikes
reshaping / coupling / walking on turntables	conservation, eq. (4)	check $\tau_{\text{ext}} = 0$ first, about YOUR axis
energy before vs after	$K = L^2/2I$ , eq. (5)	$K$ changes even as $L$ holds

Derivation AFrom the swarm to the conservation law

We derive $L = I\omega$ from the particle definition, upgrade the second law to $\tau = \Delta L/\Delta t$ , and extract the conservation law with its exact conditions — the three-step spine of the module. The strategy mirrors the two previous modules' swarm sums; the step to watch is the upgrade (Step 3), where the deeper law parts company with $\tau = I\alpha$ .

Step 1. Define per particle, mirroring torque. Torque is force times lever arm; define each particle's angular momentum as momentum times lever arm: $L_i = m_i v_i r_i$ for a particle circling at radius $r_i$ (its velocity is tangential, so the lever arm to its line of motion is $r_i$ itself).

Step 2. Sum with the shared spin. Rigidity gives $v_i = r_i\omega$ :

L = \sum_i m_i r_i^2\,\omega = \left(\sum_i m_i r_i^2\right)\omega = I\omega

The third job of $\sum mr^2$ : the exchange rate converting spin rate into angular momentum.

Step 3. Upgrade the second law. For a rigid body, take last module's $\tau_{\text{net}} = I\alpha$ and read $I\alpha = I\,\Delta\omega/\Delta t = \Delta(I\omega)/\Delta t$ — legal because $I$ is constant for a rigid body. So $\tau_{\text{net}} = \Delta L/\Delta t$ . The general proof (valid even while $I$ changes) runs the same swarm sum on the particle law $F = \Delta p/\Delta t$ and cancels internal torques pairwise — the honest bookkeeping is the deep-dive article's business; what matters here is the scope: equation (3) governs reshaping bodies, coupling clutches, and walking-on-turntable systems, where $\tau = I\alpha$ has no single $I$ to speak of.

Step 4. Extract conservation, with its fine print. If the net external torque about the chosen axis is zero over an interval, then $\Delta L = 0$ across it: $I_1\omega_1 = I_2\omega_2$ . The fine print, each clause an exam trap: external (internal torques always cancel — muscles, friction between clutch plates, the man's footsteps on the turntable are all internal to the right system); about the chosen axis (gravity exerts no torque about a vertical spin axis, which is why skaters and turntables qualify); net (a bearing's friction torque, however small, collects $L$ steadily — conservation is exact only while the inventory is empty). Verified: dimensions $\mathrm{kg\,m^2/s}$ throughout; the rigid limit of (3) returns $I\alpha$ ✓; the two-disc coupling of worked example 2 conserves $L$ exactly while losing 50% of $K$ — consistency of (4) with (5) ✓. Standard treatment: NCERT Ch. 7, HC Verma Ch. 10, Kleppner Ch. 7.

Go deeper · Derivation B · 1 minThe skater's energy bill — L frozen, K rising, muscles paying

Derivation BThe skater's energy bill — L frozen, K rising, muscles paying

We settle the module's subtlest point with numbers: when a torque-free body pulls its mass inward, its kinetic energy increases — and we locate exactly who pays. The strategy: hold $L$ fixed (no external torque), write the energy in the $L$ -form (5), compare poses, then find the payment mechanism physically. Watch for the reflex this derivation must defeat: "no external torque, so nothing can change" — angular momentum cannot; energy can.

Step 1. The trade at fixed $L$ . Pose 1: moment of inertia $I_1$ , spin $\omega_1$ , so $L = I_1\omega_1$ . Pose 2 (arms in): $I_2 < I_1$ . Conservation gives $\omega_2 = \dfrac{I_1}{I_2}\,\omega_1$ — spin rises by the full ratio.

Step 2. The energy ledger, via $K = L^2/2I$ . With $L$ identical in both poses:

\frac{K_2}{K_1} = \frac{L^2/2I_2}{L^2/2I_1} = \frac{I_1}{I_2} > 1

Halve the moment of inertia, double the kinetic energy. The same ratio that boosts $\omega$ boosts $K$ — because $K = \tfrac12 L\omega$ and $L$ is fixed while $\omega$ grows.

Step 3. Who pays. Pulling the arms in means dragging circling masses toward the axis. A circling mass resists that — left alone it would fly outward, not inward — so the muscles must haul it in along a spiral path, exerting inward force through inward displacement: positive work. Sum the haul over both arms and the total muscular work is exactly $K_2 - K_1$ . Nothing is free: the "extra" spin energy is groceries, burned in the biceps. Reversed, arms flying outward: the muscles resist the outrush, doing negative work — absorbing precisely the energy the spin loses.

Step 4. Numbers, using the concept module's skater. $I_1 = 4.74$ , $I_2 = 1.035\ \mathrm{kg\,m^2}$ (ratio 4.58), $\omega_1 = 4\pi = 12.6\ \mathrm{rad/s}$ (2 rev/s). Then $L = 59.6\ \mathrm{kg\,m^2/s}$ ; $\omega_2 = 57.6\ \mathrm{rad/s} = 9.2$ rev/s; $K_1 = \tfrac12 L\omega_1 = 375\ \mathrm{J}$ ; $K_2 = \tfrac12 L\omega_2 = 1716\ \mathrm{J}$ . Her arms did $\approx 1340\ \mathrm{J}$ of work in the second or so of the pull — briefly over a kilowatt, which is why the move is genuinely athletic. Verified: $K_2/K_1 = 1716/375 = 4.58 = I_1/I_2$ ✓; $I_1\omega_1 = 4.74 \times 12.57 = 59.6 = I_2\omega_2 = 1.035 \times 57.6$ ✓. The energy bookkeeping matches Kleppner Ch. 7's treatment of the reshaping body.

Figure

The two-disc clutch — L crosses the coupling intact, half the energy does not

Limiting cases

case	result	meaning
rigid body, $\tau = 0$	$\omega$ constant	first law again — flywheels coast
reshaping body, $\tau = 0$	$I\omega$ constant, $\omega \propto 1/I$	the skater; $K = L^2/2I$ changes
$I \to I/2$ at fixed $L$	$\omega \to 2\omega$ , $K \to 2K$	muscles pay the doubling
two bodies couple (clutch)	$L$ conserved, $K$ always lost	rotational perfectly-inelastic collision
particle in a straight line	$L = mvr_\perp$ constant	no torque about the reference point
$\tau\,\Delta t$ brief and large	$\Delta L = \tau\Delta t$	angular impulse — strikes and jerks
stellar core collapse, $R \to R/10^4$	$\omega \times 10^8$ -ish ( $I \propto R^2$ )	pulsars: the skater at cosmic stakes

Common trap"No external torque, so nothing changes — the spin rate, the energy, all of it stays put"

Conservation of angular momentum freezes exactly one number: $L = I\omega$ . It does not freeze $\omega$ (the skater quintuples hers), and it does not freeze kinetic energy ( $K = L^2/2I$ rises as $I$ falls — her muscles pay; and when two spinning discs couple, $K$ always drops while $L$ crosses intact, the lost energy heating the clutch plates). Treat "conserved" as an accounting statement about one specific currency, not a general "nothing happens" claim. The partner error runs the other way: applying $I_1\omega_1 = I_2\omega_2$ where an external torque does act — a turntable with bearing friction, a wheel gripped from outside, a diver after re-contact with the board. The checklist before writing (4): is every torque about my axis either zero-lever-arm (gravity along the axis, axle reactions) or internal to my chosen system? If the ice, the bearing, or a hand intrudes, the currency leaks and the equation lies.

Worked examples

1 · Board level — the skater, quantified (CBSE/ICSE/NEET pattern, 3 marks)

A skater spins at $2.0$ revolutions per second with arms outstretched, her moment of inertia about the spin axis being $4.74\ \mathrm{kg\,m^2}$ . She pulls her arms in, reducing it to $1.035\ \mathrm{kg\,m^2}$ . Ignoring friction, find (a) her new spin rate, (b) the ratio of her final to initial kinetic energy.

Step 1. Check the licence: gravity acts along the vertical spin axis (no torque about it), the ice is taken frictionless, and her muscles are internal. $L$ is conserved.

Step 2. (a) $I_1\omega_1 = I_2\omega_2$ — and since the rate appears on both sides linearly, revolutions per second may be used directly here: $\omega_2 = \dfrac{4.74}{1.035} \times 2.0 = \boxed{9.2\ \text{rev/s}}$ .

Step 3. (b) $\dfrac{K_2}{K_1} = \dfrac{L^2/2I_2}{L^2/2I_1} = \dfrac{I_1}{I_2} = \boxed{4.6}$ — her kinetic energy has more than quadrupled.

Step 4. The marking-scheme sentence: "the increase in kinetic energy is supplied by the work done by the skater's muscles in pulling her arms inward; angular momentum, not energy, is the conserved quantity." That sentence is frequently the allotted mark.

2 · Board level — dropping one disc onto another (NCERT clutch pattern)

A disc of moment of inertia $0.40\ \mathrm{kg\,m^2}$ spins freely at $100\ \mathrm{rad/s}$ . An identical stationary disc is dropped coaxially onto it, and friction between the faces brings them to a common angular speed. Find (a) the common speed, (b) the kinetic energy lost, and state where it went.

Step 1. System = both discs. The friction between the faces is internal; no external torque acts about the common axis. $L$ is conserved: $I_1\omega_1 = (I_1 + I_2)\,\omega$ .

Step 2. (a) $\omega = \dfrac{0.40 \times 100}{0.80} = \boxed{50\ \mathrm{rad/s}}$ — the rotational twin of a perfectly inelastic collision.

Step 3. (b) $K_i = \tfrac12 (0.40)(100)^2 = 2000\ \mathrm{J}$ ; $K_f = \tfrac12(0.80)(50)^2 = 1000\ \mathrm{J}$ . Lost: $\boxed{1000\ \mathrm{J}}$ — exactly half, dissipated as heat (and a little sound) by the slipping friction between the faces while they ground to a common speed.

Step 4. Note the pairing: $L$ crossed the coupling intact; $K$ could not — internal friction cannot touch the first but freely eats the second. For unequal discs the surviving fraction is $I_1/(I_1+I_2)$ ; equal discs give the memorable 50%.

3 · Competitive level — the man on the turntable (JEE-Main pattern, ~2.5 minutes)

A circular platform (uniform disc, $M = 100\ \mathrm{kg}$ , $R = 2.0\ \mathrm{m}$ ) rotates freely at $0.50\ \mathrm{rad/s}$ with a $60\ \mathrm{kg}$ man standing at its rim. He walks to the centre. Find the new angular speed, treating the man as a point mass.

Step 1. Inventory: bearings frictionless, gravity axial — no external torque. His footsteps are internal to (man + platform). Conservation applies.

Step 2. Initial: $I_1 = \tfrac12 MR^2 + mR^2 = \tfrac12(100)(4) + 60 \times 4 = 200 + 240 = 440\ \mathrm{kg\,m^2}$ . So $L = 440 \times 0.50 = 220\ \mathrm{kg\,m^2/s}$ .

Step 3. Final (man at centre, $r = 0$ — his contribution vanishes): $I_2 = 200\ \mathrm{kg\,m^2}$ .

Step 4. $\omega_2 = L/I_2 = 220/200 = \boxed{1.1\ \mathrm{rad/s}}$ — more than double, from a walk of two metres. Timing note: ~2 minutes. Marks die in $I_1$ : forgetting the man's $mR^2$ , or giving the platform $MR^2$ instead of $\tfrac12 MR^2$ . And the energy rider is a favourite part (b): $K$ rises from $55$ to $121\ \mathrm{J}$ — the man's legs did the work, walking inward against the outward tendency.

4 · Competitive level — the pulsar, or the skater at cosmic stakes (JEE stretch, ~3 minutes)

A star's core of radius $7.0 \times 10^5\ \mathrm{km}$ rotates once every 30 days. It collapses, without shedding mass or angular momentum, into a neutron star of radius $10\ \mathrm{km}$ . Estimate the new rotation period, taking the core as a uniform sphere throughout ( $I \propto MR^2$ ).

Step 1. Conservation with $I = \tfrac25 MR^2$ : $R_1^2\,\omega_1 = R_2^2\,\omega_2$ , so periods scale as $T_2 = T_1\left(\dfrac{R_2}{R_1}\right)^2$ .

Step 2. Radius ratio: $\dfrac{R_2}{R_1} = \dfrac{10}{7.0\times10^5} = 1.43\times10^{-5}$ ; squared: $2.04\times10^{-10}$ .

Step 3. $T_1 = 30 \times 86{,}400 = 2.59\times10^6\ \mathrm{s}$ , so $T_2 = 2.59\times10^6 \times 2.04\times10^{-10} = \boxed{5 \times 10^{-4}\ \mathrm{s}}$ — half a millisecond.

Step 4. Read the astonishment honestly: a leisurely once-a-month rotation becomes ~2000 rotations per second, purely because $I \propto R^2$ collapsed by ten orders of magnitude at fixed $L$ . Real millisecond pulsars spin at just this scale — the conservation law confirmed nightly by radio telescopes. (Real collapses shed some mass and magnetize; the order of magnitude survives all corrections.) Timing note: ~2.5 minutes, and entirely about ratio discipline — no need for $M$ or actual $I$ values. The trap is scaling $T$ linearly with $R$ instead of $R^2$ ; options offering $10^{-1}$ s are pricing that.

Exam traps

Check the torque inventory before conserving. Bearing friction, an external hand, ice resistance — any external torque about the axis leaks $L$ . "Frictionless" and "freely rotating" are the licence words.

External about your axis. Gravity torques a tilted top but not a vertical spin axis; axle reactions have zero lever arm. The check is axis-specific.

$\omega \propto 1/I$ , not $1/\sqrt I$ or $1/I^2$ . And in ratio problems, rev/s may replace rad/s only because the conversion cancels — in any absolute calculation ( $L$ , $K$ ), radians per second, always.

Don't conserve $K$ across reshapings or couplings. The skater gains energy (muscles pay), the clutch loses it (friction eats) — in both, $L$ is the honest currency. Writing $\tfrac12 I_1\omega_1^2 = \tfrac12 I_2\omega_2^2$ is the classic planted wrong route.

All the moment of inertia, every state. Turntable problems: platform plus man (at his current radius), before and after. Dropping the passenger's $mr^2$ from one side is the standard mark-loser.

Particle $L$ uses the lever arm to the line of motion. $L = mvr_\perp$ , not $mvr$ to wherever the particle happens to be. A straight-flying particle has constant, generally nonzero $L$ .

$\tau = I\alpha$ dies when $I$ changes. Reshaping bodies obey $\tau = \Delta L/\Delta t$ ; with $\tau = 0$ that is the conservation law. Trying to run $I\alpha$ through a reshaping is a category error the stretch questions probe.

Origins · 1 minEmmy Noether and how this was discovered

Emmy Noether

1882–1935 · Erlangen & Göttingen

Why is angular momentum conserved at all? The deepest answer physics possesses was proved in 1918 by Emmy Noether: every continuous symmetry of nature's laws begets a conservation law. Momentum is conserved because the laws are the same here as there (translation symmetry); energy because they are the same today as tomorrow; and angular momentum because they are the same facing north as facing east — the universe has no preferred direction, and $L$ is the bookkeeping of that indifference. Noether proved this while officially barred from a professorship because she was a woman — she taught for years at Göttingen under David Hilbert's name, Hilbert remarking acidly that a faculty senate was not a bathhouse. Einstein called her the most significant creative mathematical genius since higher education for women began; her theorem is now load-bearing in every corner of physics, from the skater's spin to the conservation laws that particle physicists use to discover new particles by what fails to balance. When you write $I_1\omega_1 = I_2\omega_2$ , you are cashing a cheque Noether signed: rotational symmetry, paid out as a frozen number.

History

1609

— Kepler's second law: planets sweep equal areas in equal times — angular momentum conservation observed in the sky before the concept exists.

1687

— Newton's Principia proves the equal-areas law follows from a central force: zero torque, constant L, in geometric dress.

1765

— Euler's rigid-body treatise makes L = Iω and its dynamics explicit.

1918

— Noether's theorem roots the conservation law in rotational symmetry itself.

1967

— Jocelyn Bell detects the first pulsar: a collapsed star spinning seconds-fast, the conservation law performing at cosmic stakes.

Going further

Texts. NCERT Physics Class 11, Ch. 7 (angular momentum and conservation) · HC Verma, Concepts of Physics Vol. 1, Ch. 10 · Halliday–Resnick–Walker Ch. 11. For the vector story — why $\vec L$ and $\vec\omega$ need not even be parallel: Angular Momentum of a Rigid Body (UG deep dive).

Provenance. All results are standard. A verifier should recompute: the skater ( $\omega_2 = 4.74/1.035 \times 2.0 = 9.16$ rev/s; $K$ ratio $4.58$ ; $L = 59.6\ \mathrm{kg\,m^2/s}$ , $K_1 = 375$ J, $K_2 = 1716$ J); the discs ( $\omega = 50\ \mathrm{rad/s}$ , $2000 \to 1000$ J, 50%); the turntable ( $I_1 = 440$ , $L = 220$ , $\omega_2 = 1.1\ \mathrm{rad/s}$ ; $K$ : $55 \to 121$ J); the pulsar ( $T_2 = 2.59\times10^6 \times (1.43\times10^{-5})^2 = 5.3\times10^{-4}$ s).

Prerequisites. Newton's Second Law for Rotation · Conservation of Linear Momentum · Linear Momentum & Impulse. Leads to. Rolling With & Without Slipping · Rotational Collisions (archetype) · Angular Momentum of a Rigid Body (UG deep dive).

Why is angular momentum conserved — is it just because torques happen to cancel?

There are two honest answers, at two depths. The working answer: the rotational second law says $\tau_{\text{net}} = \Delta L/\Delta t$ , and internal forces — however violent — come in third-law pairs whose torques cancel exactly; so when nothing external twists the system, $L$ has no way to change. The deep answer is Noether's theorem: angular momentum conservation is the bookkeeping of rotational symmetry — the fact that physics has no preferred direction in space. Turn your whole experiment 30° and it runs identically; that indifference, made mathematical, is the conservation law. The two answers connect: a genuinely external torque marks something in the environment that breaks the symmetry (a preferred direction — the brake pad's location, the ice's grip), and precisely then the currency leaks.

Where does the skater's extra kinetic energy actually come from?

From her muscles, paid in the most literal sense. Her circling arm masses, left to themselves, would drift outward — pulling them inward means exerting an inward force through an inward displacement, which is positive mechanical work, summed along the spiral path each hand traces. That work equals exactly the kinetic-energy gain $K_2 - K_1 = \tfrac{L^2}{2}\left(\tfrac1{I_2} - \tfrac1{I_1}\right)$ — for the worked skater, about 1.3 kJ in roughly a second, briefly more than a kilowatt of power output, which is why the move takes visible effort and training. Opening the arms runs the account backwards: the muscles resist the outward rush, absorbing energy, and the spin slows. The conserved quantity throughout is $L$ ; energy simply follows the work that the reshaping costs or yields.

What is the difference between τ = Iα and τ = ΔL/Δt — aren't they the same equation?

For a rigid body, yes: constant $I$ lets you pull it through the rate of change, and $\Delta(I\omega)/\Delta t = I\alpha$ . The difference is jurisdiction. When the body reshapes — skater, diver, collapsing star, a system of parts moving relative to each other — there is no single constant $I$ , and $\tau = I\alpha$ becomes ill-posed. But $\tau = \Delta L/\Delta t$ never blinks: torque changes the total angular momentum, however the body chooses to carry it. That is why the conservation argument runs through $L$ and not through $\omega$ or $\alpha$ . The hierarchy is worth remembering at exam speed: $\tau = \Delta L/\Delta t$ is the law; $\tau = I\alpha$ is its rigid special case; $I_1\omega_1 = I_2\omega_2$ is its torque-free integral.

How can a particle moving in a straight line have angular momentum? Nothing is turning.

Because angular momentum measures motion with a lever arm, not visible circling. A particle cruising past a point at perpendicular distance $r_\perp$ carries $L = mvr_\perp$ about that point — and since $v$ and the perpendicular drop to its (fixed) line of motion never change, this $L$ is constant: conservation, trivially satisfied, because no force means no torque. The payoff for taking this seriously is Kepler: a planet is "a particle trying to move straight" continually bent by a force pointing at the Sun — zero lever arm, zero torque — so its $L$ about the Sun is exactly conserved, and equal-areas-in-equal-times follows. It also powers JEE problems where a flying particle strikes and sticks to a rod or disc: just before impact its $mvr_\perp$ about the hinge is the entire angular-momentum budget of the collision (the Rotational Collisions archetype develops the family).

In the two-disc problem, why is energy lost but angular momentum not? Both feel the same friction.

Because the friction between the faces is internal to the two-disc system, and internal interactions have unequal power over the two currencies. Torque-wise, the friction acts as a third-law pair: the top disc drags the bottom one forward, the bottom drags the top backward, equal and opposite about the same axis — net zero, so $L$ passes through the coupling untouched. Energy-wise there is no such protection: while the faces slip against each other, friction times slip-angle is honest heat, extracted from the kinetic store. The linear twin makes it familiar: a perfectly inelastic collision conserves momentum but always loses kinetic energy. Coupling clutches, dropped discs, and catching a spinning ball on a turntable are all rotational perfectly-inelastic collisions — $L$ crosses, $K$ pays a toll fixed by the mismatch: $\Delta K = \tfrac12\tfrac{I_1 I_2}{I_1+I_2}(\omega_1-\omega_2)^2$ for two coupling bodies.

Why does a helicopter need a tail rotor?

Conservation, enforced in hardware. The engine torques the main rotor; by Newton's third law the rotor torques the fuselage back, and with nothing else acting, the body of the helicopter would counter-spin under the blades (total $L$ of rotor-plus-fuselage staying fixed). The tail rotor supplies a sideways thrust at a long lever arm — an external torque on the fuselage that cancels the reaction, pinning the cabin's orientation. Watch the confirmation in the failure mode: news footage of tail-rotor failures shows the fuselage beginning to whirl opposite to the blades, conservation reasserting itself instantly. Twin-rotor helicopters (Chinook) solve the same books differently: two main rotors spinning opposite ways, their torques cancelling internally — the same physics as the cat, which twists front and rear body halves against each other to land feet-first with total $L$ still zero.

Both linear momentum and angular momentum are conserved — do I ever need both at once?

Yes, and the pattern is worth owning early: they are independent laws, each with its own no-external-agent condition, and rich problems invoke them separately about different things. A shell exploding mid-flight: linear momentum of the fragments is conserved (no external horizontal force), and angular momentum about the centre of mass is conserved too (internal explosion torques cancel) — two equations, different unknowns. A particle striking a hinged rod: linear momentum is not conserved (the hinge delivers an external impulsive force) but angular momentum about the hinge is (the hinge force has zero lever arm there) — choosing the hinge as the reference point is exactly what disarms the unknown reaction. That selective deployment — pick the point where the intruding force has no lever arm — is the master move of the Rotational Collisions archetype ahead, and the direct heir of "choose the pivot to kill the unknown" from the torque module.