Torque & the Turning Effect

Push a heavy door at its handle and it swings open under two fingers. Push just as hard a hand-width from the hinge and you can lean your whole shoulder into it for nothing. Same door, same force — completely different result. Force alone evidently does not decide whether things turn; where you push and in which direction matter every bit as much. The quantity that scores all three at once is the torque, $\tau = rF\sin\phi$ , and it is to rotation what force is to straight-line motion: the thing that causes it, changes it, and — when balanced — leaves it alone. Humanity ran on this idea long before writing it down: levers moved temple stones, oars steered ships, and Archimedes turned it into the first quantitative law of mechanics twenty-two centuries ago ("give me a place to stand and I will move the Earth" is a torque calculation wearing a toga). Your daily life still runs on it — spanners, door handles, steering wheels, bottle caps and bicycle pedals are all machines for manufacturing torque from modest forces by renting a long lever arm. Exam-wise, this module is a rare all-boards module: IGCSE and A-Level test it as "moments" and the principle of balance; CBSE, ICSE and NEET want the $rF\sin\phi$ geometry; JEE and AP add the vector version $\vec\tau = \vec r \times \vec F$ and couples. The turning dynamics — how much angular acceleration a torque buys — waits two modules ahead; here we master the turning effort itself, and the geometry that examiners mine for marks.

Key takeaways

Torque measures a force's turning effectiveness about a chosen axis: $\tau = rF\sin\phi$ — force, distance from the axis, and angle all matter.
Two equivalent readings: full force × lever arm ( $r_\perp = r\sin\phi$ , the perpendicular distance from the axis to the force's line of action), or full distance × tangential component ( $F_\perp = F\sin\phi$ ).
A force pointing along the radius ( $\phi = 0$ or $180°$ ) has zero torque — pulling a door toward its hinge turns nothing.
Sign convention: anticlockwise positive, clockwise negative (or the right-hand rule for the vector $\vec\tau = \vec r \times \vec F$ , JEE/AP).
A couple — two equal, opposite, parallel forces separated by $d$ — produces pure turning: $\tau = Fd$ about any point, with zero net force.
Units: newton-metres (N·m) — dimensionally joules, but torque is not energy; never write J.
Key idea: torque = force × lever arm — slide the force along its own line and nothing changes; move its line closer to the axis and the turning dies.

The core insight

The turning effect of a force about an axis is $\tau = rF\sin\phi = r_\perp F = rF_\perp$ — read either as force times the perpendicular (lever-arm) distance from the axis to the force's line of action, or as radius times the force component perpendicular to it; radial components and forces through the axis turn nothing.

Where this lives today

Torque is the headline number of the machine world. Car and EV spec sheets quote peak torque (a family EV's motor: ~300 N·m from rest — the reason electric cars leap from traffic lights); engine dynamometers measure it directly, and gearboxes exist to trade rotation rate for torque like currency. Assembly lines run on torque wrenches — cylinder-head bolts are specified to, say, $90$ N·m, because too little torque loosens and too much strips threads; aircraft maintenance logs record the torque applied to every critical fastener. Your own body is a torque catalogue: biceps pulling a forearm about the elbow work at a mere ~4 cm lever arm (which is why lifting 10 kg at the hand costs the muscle hundreds of newtons), and physiotherapy quantifies joints in newton-metres. Even the doorknob placement in every building is this module's Example 1, solved by carpenters centuries before it was written as $rF\sin\phi$ .

The physical picture

Interrogate the door properly, because it contains the whole law. Three separate changes each kill your push's turning power. Slide your hand toward the hinge: same force, small $r$ — barely turns. Keep the distance but angle your push along the door toward the hinge line: you are now pulling the door through its own plane — zero turning, however hard you shove. Push perpendicular to the door at the far edge: maximum effect. Torque is the product that scores all three, $\tau = rF\sin\phi$ , and the $\sin\phi$ is not decoration — it extracts the only part of the force with any interest in turning.

Figure

Torque geometry — one product, two equivalent readings

The two readings of $rF\sin\phi$ are worth owning separately, because different problems hand you different geometry. Reading one: keep the whole force, but measure the lever arm — the perpendicular distance from the axis to the force's line of action. This reading carries a free theorem: you may slide a force anywhere along its own line without changing its torque (the line, and hence the perpendicular drop to it, is unchanged). Weight hanging from a beam? The line of action is vertical through the load; the lever arm is the horizontal distance — however the rope is drawn. Reading two: keep the whole distance $r$ , but keep only the force's tangential component $F\sin\phi$ ; the radial component points through the axis, has zero lever arm, and is the hinge's problem, not the rotation's. And one non-negotiable clarification, because the units invite the error: torque is measured in newton-metres, which is dimensionally a joule — but torque is not energy. A joule of work is force through a distance moved along the force; a newton-metre of torque is force beside a distance, at right angles to it, and a stuck bolt under 100 N·m of torque receives zero energy while nothing rotates. The convention: torque in N·m, never in J.

The math: the definition, the sign, and the couple

The working definition, in its fixed-axis form and its vector dress:

\tau = r\,F\sin\phi = r_\perp F = r\,F_\perp, \qquad \vec\tau = \vec r \times \vec F \qquad (1)

where: $r$ — distance from the axis to the point of application; $\phi$ — angle between $\vec r$ and $\vec F$ ; $r_\perp = r\sin\phi$ — the lever arm; $F_\perp = F\sin\phi$ — the tangential component. Feel a number: a 200 N push at the handle of a 0.9 m-wide door, perpendicular, is $\tau = 0.9 \times 200 = 180$ N·m; the same push 0.1 m from the hinge is 20 N·m — nine times weaker at the same effort. For fixed-axis work, use a sign: anticlockwise positive, clockwise negative (a convention, but announce it and keep it). Several torques about the same axis add algebraically:

\tau_{\text{net}} = \sum_i \pm\, r_{\perp i}\,F_i \qquad (2)

where: the signs carry each force's turning sense. For the vector version (JEE/AP): $\vec\tau$ points along the axis by the right-hand rule — fingers sweep from $\vec r$ to $\vec F$ , thumb gives $\vec\tau$ ; in components, for $\vec r = x\hat i + y\hat j$ and $\vec F = F_x\hat i + F_y\hat j$ (all in a plane),

\tau_z = x F_y - y F_x \qquad (3)

where: the memorable cross-multiplied difference — Worked Example 3 drills it. A-Level and IGCSE students: your syllabi call all this moments, use only the perpendicular-distance form $\tau = F d$ , and examine it through the principle of moments (balance: total clockwise moment = total anticlockwise moment) — the vector form is beyond both specs, flagged here for the JEE/AP lane. Finally, the configuration with a name of its own. A couple is a pair of equal-magnitude, opposite, parallel forces whose lines of action are separated by $d$ — your two hands on a steering wheel, thumb and finger on a tap:

\tau_{\text{couple}} = F\,d \qquad \text{about ANY point} \qquad (4)

where: the net force is zero (no pushing anywhere) yet the turning is real and — remarkably — the same about every pivot, which is proved in Derivation B and exploited by every screwdriver.

Figure

The couple — zero net force, pure turning, any pivot

you want	use	notes
turning effect of one force	$\tau = rF\sin\phi$ , eq. (1)	either reading; slide along the line freely
several forces, one axis	signed sum, eq. (2)	declare anticlockwise +
torque from components (JEE/AP)	$\tau_z = xF_y - yF_x$ , eq. (3)	right-hand rule for direction
two hands / tap / wheel	couple: $\tau = Fd$ , eq. (4)	pivot-independent; net force zero
balance problems (AL/IG "moments")	$\sum\tau_{\text{cw}} = \sum\tau_{\text{acw}}$	the equilibrium module completes this

Derivation AWhy the lever arm is the whole story

We establish that the three expressions in (1) are the same number, and that sliding a force along its line of action changes nothing — the two facts that turn messy torque geometry into quick reading. Strategy: pure right-triangle trigonometry on the door figure. Watch which perpendicular is which: the lever arm is a distance from the axis to a line; the tangential component is a piece of the force. Two different right triangles, one shared angle $\phi$ .

Step 1. Split the force. At the push point, resolve $\vec F$ into a component along the radius ( $F\cos\phi$ , pointing at or away from the axis) and one perpendicular to the radius ( $F_\perp = F\sin\phi$ ). The radial component's line of action passes through the axis: whatever it does (stress the hinge, stretch the door), it cannot turn the body about that axis. Only $F_\perp$ turns, acting at the full distance $r$ : torque $= r \times F\sin\phi$ .

Step 2. Now read the same product the other way. Drop a perpendicular from the axis onto the force's line of action; basic trigonometry in the figure's second right triangle gives that distance as $r_\perp = r\sin\phi$ . Grouping the product the other way, $\tau = (r\sin\phi) \times F = r_\perp F$ : the whole force times the lever arm. Same number, two readings — take whichever the diagram makes easy.

Step 3. The sliding theorem. Move the application point anywhere along the force's own line of action. The line has not moved, so the perpendicular drop from the axis — the lever arm — is identical, and by Step 2 so is the torque. This is why "a 40 N weight hangs from the beam at the 30 cm mark" needs no drawing of where along the string to take the force: the line is vertical through the 30 cm mark, full stop.

Step 4. The two zero-torque geometries, now obvious: force through the axis ( $r_\perp = 0$ : the line of action passes through the pivot) and force along the radius ( $\phi = 0$ or $180^\circ$ , same thing). Every "why doesn't this force appear in the moments equation?" answer is one of these two sentences — and choosing the pivot to make an annoying unknown force pass through it is the single best trick in equilibrium problems, two modules ahead. Verified: at $\phi = 90^\circ$ both readings give $\tau = rF$ (maximum); at $\phi = 0$ both give zero; the 200 N door push at $r = 0.9$ m and $\phi = 60^\circ$ gives $\tau = 0.9\times200\times0.866 = 156$ N·m by either grouping ✓. Standard geometry (NCERT Ch. 7; Kleppner Ch. 7).

Go deeper · Derivation B · 1 minThe couple — pure turning, the same about every pivot

Derivation BThe couple — pure turning, the same about every pivot

We derive the couple's torque $\tau = Fd$ and its startling pivot-independence. The strategy: place two equal and opposite parallel forces, compute the signed sum (2) about an arbitrary pivot, and watch the pivot's position cancel. This result is why "torque of a couple" needs no axis specified — and why a body feeling only a couple spins without travelling.

Step 1. Set up. Forces $+F$ (up) acting at position $x_1$ and $-F$ (down) at $x_2$ , with separation $d = x_2 - x_1$ — think of a horizontal rod pushed up at one point, down at another. Net force: $F - F = 0$ ; the body's centre of mass will not accelerate anywhere.

Step 2. Torque about an arbitrary point $P$ at position $x$ , anticlockwise positive. With signed lever arms $(x_i - x)$ , an upward force at $x_i$ contributes $+F(x_i - x)$ and a downward force contributes $-F(x_i - x)$ — the sign rule handling both sides of the pivot automatically. Here the up-force acts at $x_2$ and the down-force at $x_1$ :

\tau_P = F(x_2 - x) - F(x_1 - x) = F(x_2 - x_1) = F\,d

Step 3. Read the cancellation: the pivot position $x$ appeared once with each sign and vanished. The couple's turning effect is $Fd$ about the end of the rod, about its middle, about a point a mile away — about anywhere. A couple has no favourite axis; it is "pure torque," the rotational analogue of a free force.

Step 4. Consequences worth carrying. A steering wheel gripped with two hands at diameter $d$ and tangential pushes $F$ each feels $\tau = Fd$ — no net shove on the steering column's bearings. A single force can never be a couple (it always has a net push); and a couple can never be balanced by a single force, only by another couple — an A-Level favourite statement-question, and the deep reason two-pin supports exist in engineering. Verified: numbers — two 25 N hands on a 0.38 m steering wheel: $\tau = 25\times0.38 = 9.5$ N·m about the centre; recompute about the rim point where one hand grips: near force arm 0, far force arm 0.38 m: $25\times0.38 = 9.5$ N·m ✓ pivot-independent. Matches the general treatment (Kleppner Ch. 7; NCERT Ch. 7).

Limiting cases

case	result	meaning
$\phi = 90^\circ$	$\tau = rF$ , maximum	push perpendicular, at the far end — spanner design
$\phi = 0$ or $180^\circ$	$\tau = 0$	force along the radius: pulling the door toward its hinge
line of action through the axis	$\tau = 0$	zero lever arm — the pivot-choice trick
$r \to$ 2×	$\tau \to$ 2× at the same $F$	why longer spanners loosen stubborn nuts
net force zero, forces separated	couple: $\tau = Fd$ , any pivot	pure spin, no shove
single force through the COM of a free body	pushes without turning	zero torque about the COM

Common trap“The bigger force wins the turning contest — and torque is basically energy since both are newton-metres”

Two errors, both mark-killers. First: turning contests are decided by torque, not force, and lever arm can overwhelm any strength deficit — a child of 300 N at a see-saw's end (2 m) out-turns an adult of 700 N sitting 0.5 m from the pivot ( $600$ vs $350$ N·m), and two fingers at a door handle beat a shoulder at the hinge nine to one. Whenever an exam offers "the larger force produces the larger moment" as an option, look for the lever arms first. Second: torque is not energy despite sharing the unit N·m. Energy is force times distance moved along the force (a scalar that accumulates); torque is force times perpendicular offset (a turning effort that exists even in complete stillness). A seized bolt under a mighty torque gains no energy at all until it actually turns — only then does work flow, at the rate torque × angle turned. Write torque in N·m, never joules, and never "convert" between a torque and an energy: the shared unit is a dimensional coincidence, policed by the $\sin\phi$ in one and the $\cos$ -projection in the other.

Worked examples

1 · Board level — the spanner and the angled pull (CBSE/ICSE/NEET pattern, 3 marks)

A mechanic pulls the end of a $30\ \mathrm{cm}$ spanner with a force of $50\ \mathrm{N}$ . Find the torque about the nut when the pull is (a) perpendicular to the handle, (b) at $60^\circ$ to the handle, (c) along the handle.

Step 1. Perpendicular: $\tau = rF = 0.30 \times 50 = \boxed{15\ \mathrm{N\,m}}$ — the spanner's best.

Step 2. At $60^\circ$ : $\tau = rF\sin 60^\circ = 0.30 \times 50 \times 0.866 = \boxed{13\ \mathrm{N\,m}}$ . (Equivalently: lever arm $= 0.30\sin60^\circ = 0.26$ m, times 50 N — same 13 N·m.)

Step 3. Along the handle: $\phi = 0$ , so $\tau = \boxed{0}$ — pulling toward or away from the nut turns nothing, however hard.

Step 4. Marking-scheme sentence: "only the component of force perpendicular to the handle produces a turning effect." That sentence, plus the unit N·m, carries the method marks.

2 · Board level — principle of moments on a see-saw (IGCSE/A-Level "moments" pattern)

A uniform see-saw is pivoted at its centre. A $30\ \mathrm{kg}$ child sits $2.0\ \mathrm{m}$ from the pivot. Where must a $40\ \mathrm{kg}$ child sit to balance it? Take $g = 9.8\ \mathrm{m/s^2}$ .

Step 1. The beam is uniform and pivoted at its centre, so its own weight acts through the pivot — zero lever arm, out of the books (Derivation A, Step 4).

Step 2. Balance = principle of moments about the pivot: clockwise = anticlockwise.

(30\times9.8)\times 2.0 = (40\times9.8)\times x

Step 3. The $g$ cancels (it always does in pure balance): $x = \dfrac{30\times2.0}{40} = \boxed{1.5\ \mathrm{m}}$ on the other side.

Step 4. Sanity: heavier child, shorter arm — moments trade inversely, the lever's ancient law. (Full rigid-body equilibrium, with support forces and off-centre pivots, is completed in the Rigid-Body Equilibrium module.)

3 · Competitive level — vector torque from components (JEE-Main pattern, ~2.5 minutes)

A force $\vec F = (3\hat i + 4\hat j)\ \mathrm{N}$ acts at the point $\vec r = (2\hat i + 1\hat j)\ \mathrm{m}$ . Find the torque about the origin, and state its direction.

Step 1. Planar cross product, equation (3): $\tau_z = xF_y - yF_x = (2)(4) - (1)(3) = 8 - 3 = 5$ .

Step 2. So $\vec\tau = \boxed{5\hat k\ \mathrm{N\,m}}$ — magnitude 5 N·m, along $+z$ (out of the page): an anticlockwise turning in the $xy$ -plane by the right-hand rule.

Step 3. Cross-check via lever arm: $|\vec r| = \sqrt5$ , $|\vec F| = 5$ , so $rF = 5\sqrt5 \approx 11.2$ N·m at most; our 5 N·m implies $\sin\phi = 5/(5\sqrt5) = 1/\sqrt5$ — the force lies mostly along $\vec r$ , wasting most of its magnitude radially. The components quietly contained that geometry. Timing note: ~90 seconds. The formula $xF_y - yF_x$ with signs intact is all; the classic slip is the order ( $yF_x - xF_y$ flips the direction — and "clockwise, into the page" is sitting among the options). Memorize: $x$ before $y$ , "cross-multiply and subtract."

4 · Competitive level — JEE/AP stretch: net torque and the couple hiding in the door (~4 minutes)

A door $0.90\ \mathrm{m}$ wide is pushed at its outer edge with $40\ \mathrm{N}$ perpendicular to the door, while someone on the other side pushes back with $40\ \mathrm{N}$ perpendicular at a point $0.30\ \mathrm{m}$ from the hinge. (a) Find the net torque about the hinge. (b) Show the same answer arrives via a couple-plus-force decomposition.

Step 1. Direct signed sum about the hinge (anticlockwise positive for the first push): $\tau = 40\times0.90 - 40\times0.30 = 36 - 12 = \boxed{24\ \mathrm{N\,m}}$ anticlockwise. The door turns toward the weaker position, not the weaker force — the forces are equal.

Step 2. Decomposition view: two equal and opposite 40 N forces separated by $d = 0.60$ m form a couple of $Fd = 24$ N·m — and a couple's torque is pivot-independent (Derivation B), so about the hinge it is still 24 N·m. The net force on the hinge from this pair: zero.

Step 3. Read what the decomposition bought: no lever-arm bookkeeping at all — the couple's $Fd$ ignored the hinge's location entirely. On multi-force JEE problems, spotting an embedded couple often replaces four signed terms with one product. Timing note: ~3 min. Couples are the examiner's favourite "zero net force but things still turn" trap — the assertion-reason pattern "net force zero ⇒ no turning" is FALSE, and this example is the two-line counterexample to carry into the exam hall.

Exam traps

Radial components turn nothing. Any force (or component) whose line passes through the axis contributes zero torque. Before summing moments, delete these terms out loud — and choose the pivot to kill unknown forces (the hinge force never appears in moments about the hinge).

Lever arm is to the line, not to the point. $r_\perp$ is the perpendicular distance from the axis to the force's line of action — extend the line if needed. Measuring to the application point (and then also multiplying by $\sin\phi$ ) double-counts the geometry.

Sign chaos. Declare "anticlockwise positive" in writing, then give every term its sign from the turning sense, not from the force's up/down direction. A downward force can turn either way depending on which side of the pivot it acts.

N·m, not J — and not "Nm⁻¹" either. Dimensional-analysis MCQs bait all three. Torque and energy share dimensions but not identity; work done by a torque is $\tau \times \theta$ (radians), which is where the joules actually appear.

"Zero net force ⇒ zero turning" is false. Couples turn with no net force (Example 4). The converse trap too: a single force through the centre of mass shoves without turning. Force balance and torque balance are independent books — rigid-body equilibrium needs both, as the equilibrium module formalizes.

AL/IG scope line. A-Level (CIE 9702) and IGCSE examine moments: $\tau = Fd$ with perpendicular distances, principle of moments, couples as $Fd$ . The $\sin\phi$ resolution appears in AS mechanics contexts, but $\vec r \times \vec F$ components (Example 3) are JEE/AP territory — don't import vector machinery into a moments answer that wants two products and a balance.

Distance from the wrong point. In multi-pivot problems (wheelbarrows, crowbars), every lever arm must be measured from the same chosen axis. Mixing distances-from-the-load with distances-from-the-wheel in one equation is the classic wheelbarrow error.

Origins · 1 minArchimedes and how this was discovered

Archimedes

c. 287–212 BC · Syracuse

Torque's founding document is two millennia older than Newton. Archimedes' On the Equilibrium of Planes states and proves the law of the lever — weights balance at distances inversely proportional to their magnitudes — deriving it from symmetry axioms with full Greek rigor, and making "force times distance" the first quantitative law of mechanics. He wielded it as engineering: compound pulleys that let one man haul a loaded ship, and the war machines that held the Roman siege of Syracuse at bay for two years ("give me a place to stand and I will move the Earth" is the lever-arm scaling read as a boast). What Archimedes had was the statics of torque — balance. What he lacked was its dynamics: that an unbalanced torque produces angular acceleration in proportion to a mass-weighted $r^2$ sum, a closure that waited nineteen centuries for Euler. The word "torque" itself is the newest thing here — James Thomson, Lord Kelvin's brother, proposed it only in 1884, from the Latin torquere, to twist.

History

250

BC — Archimedes' On the Equilibrium of Planes proves the law of the lever — the first quantitative mechanics.

1638

— Galileo's Two New Sciences analyses beams and levers, carrying moment reasoning into the new science of motion.

1687

— Newton's Principia supplies the force laws whose rotational bookkeeping torque performs.

1765

— Euler's rigid-body dynamics completes the story: unbalanced torque drives angular acceleration (

\tau = I\alpha

, two modules ahead).

1884

— James Thomson coins "torque"; the machine age adopts it as the currency of every rotating shaft.

Going further

Texts. NCERT Physics Class 11, Ch. 7 (moment of force; couple) · HC Verma, Concepts of Physics Vol. 1, Ch. 10 · Halliday–Resnick–Walker Ch. 10–11. For the full fixed-axis dynamics this module feeds ( $\tau = I\alpha$ , work and power of torques): Torque & Fixed-Axis Dynamics (UG deep dive).

Provenance. All results are standard. A verifier should recompute: the spanner triple (15, 13, 0 N·m — with $0.3\times50\times0.866 = 12.99$ ); the see-saw balance $x = 1.5$ m with $g$ cancelling; Example 3's $\tau_z = 8-3 = 5$ N·m ( $+\hat k$ ) and the $\sin\phi = 1/\sqrt5$ cross-check; Example 4's $36-12 = 24$ N·m and the couple check $40\times0.60 = 24$ ; Derivation B's steering-wheel pivot-independence ( $25\times0.38 = 9.5$ N·m about centre and rim alike).

Prerequisites. Angular Displacement, Velocity & Acceleration · Dot & Cross Products · Free-Body Diagrams & Problem-Solving Strategy. Leads to. Moment of Inertia: the Concept · Newton's Second Law for Rotation · Torque & Fixed-Axis Dynamics (UG deep dive).

Why does the angle matter? A newton is a newton wherever it points.

Because rotation only "feels" the part of the force that moves the application point around the axis. Split any force into a radial piece (pointing at or away from the axis) and a tangential piece: the radial piece would drag the point straight through the axis line — no turning, just stress on the pivot — while the tangential piece $F\sin\phi$ sweeps it around the circle. The $\sin\phi$ in the definition is exactly this split. At $\phi = 90^\circ$ all of the force turns; at $\phi = 0$ none of it does, which you can feel by pulling a door handle toward its hinges.

Torque about what? Do I always need an axis?

Yes — torque is meaningless without a stated reference (axis or point), and the same force has different torques about different axes. That dependence is a tool: in balance problems you choose the pivot to make unknown forces (hinge reactions, normal forces at a support) pass through it and vanish from the equation. The one exception to axis-dependence is the couple: its torque $Fd$ is the same about every point, which is why "the torque of a couple" may be quoted without naming an axis.

If torque and energy share the unit N·m, why can't I convert one into the other?

Because the metre in each measures a different geometry. Work's metre is displacement along the force ( $W = Fd\cos\theta$ — a projection); torque's metre is offset perpendicular to the force ( $\tau = Fd\sin\theta$ — a rejection). Same dimensions, orthogonal meanings. They do meet, but only through rotation: when a torque $\tau$ actually turns its body through angle $\theta$ (in radians), it does work $W = \tau\theta$ — and that is in joules. A torque on a non-rotating bolt does no work at all, however long you strain.

How can something turn when the net force on it is zero?

Through a couple. Two equal and opposite forces cancel as a push (the centre of mass stays put) but, acting along different lines, they gang up as a turn: $\tau = Fd$ about any point. Spin a pen between two fingers moving opposite ways and you are applying one. The deeper lesson: force balance and torque balance are independent conditions — a rigid body needs both $\sum F = 0$ and $\sum\tau = 0$ to be in equilibrium, which is the whole business of the equilibrium module ahead.

Is torque the same thing as "moment" in my A-Level/IGCSE syllabus?

Same physics, different dialect. "Moment of a force" = force × perpendicular distance from the pivot to its line of action — exactly $\tau = r_\perp F$ . Your syllabus works with perpendicular distances directly (no $\sin\phi$ resolution needed in most questions), examines the principle of moments (clockwise = anticlockwise for balance), and treats couples as $F\times d$ . The vector cross product and component formula are beyond the CIE specs — physicists say "torque," engineers and UK boards say "moment," and the nut being loosened cannot tell the difference.

Which way does the torque vector point, and why should I care at this level?

Along the rotation axis, by the right-hand rule: curl your right hand's fingers in the turning sense, and your thumb points along $\vec\tau$ (out of the page for anticlockwise). For fixed-axis board problems, "+ anticlockwise" honestly does the same job. You care for two reasons: JEE asks the direction of $\vec r \times \vec F$ outright (Example 3), and the whole next tier of the chapter — angular momentum, gyroscopes, why a spinning top precesses instead of falling — runs on torque and angular momentum being genuine axis-vectors. The rule costs ten seconds to learn now and pays for months.

Why does a longer spanner work better — where does the "extra" effect come from? It feels like cheating.

No cheating — you pay in distance what you save in force, exactly as with every lever. To turn the nut through the same angle, your hand at double the radius sweeps double the arc: same work ( $W = \tau\theta$ either way), delivered as half the force through twice the path. The nut experiences the torque, and torque is what shears rusted threads loose; your muscles experience the force, and force is what they are bad at. A spanner is a machine for re-pricing effort — the physics bill is identical, but paid in a currency your body finds cheaper.