ARCHETYPE: The Ballistic Pendulum

In 1742, a British engineer named Benjamin Robins needed to answer a question nobody could answer: how fast does a musket ball actually fly? No clock on Earth could time something that crosses a room in a hundredth of a second. His solution was to stop measuring speed and start measuring consequences: fire the ball into a heavy wooden pendulum, let it embed, and watch how high the pendulum swings. The swing is slow, visible, measurable with a ruler — and from that one height, the bullet's speed follows by pure physics. The ballistic pendulum measured supersonic speeds a full two centuries before electronics could. The reason this setup rules exam papers is the reasoning shape it forces on you. It is two problems stitched together, and each stage obeys a different conservation law. The embedding is a perfectly inelastic collision: momentum is conserved, kinetic energy emphatically is not (over 99% of it burns into heat and splintered wood). The swing is smooth pendulum motion: mechanical energy is conserved, momentum emphatically is not (the string and gravity drain it continuously). Use the right law in the right stage and the problem is four lines. Swap them — the single most common error — and your answer is wrong by a factor of hundreds. JEE, NEET, AP, IB and A-Level all pose this archetype, dressed as bullets, darts, clay balls and railway buffers. Learn the two-stage discipline here and you have also learned the master template for every multi-stage conservation problem in mechanics.

Key takeaways

The ballistic pendulum is a two-stage problem: collide, then swing — and each stage gets its own conservation law.
Stage 1 (bullet embeds, perfectly inelastic): momentum conserved, $mv = (m+M)V$ ; kinetic energy is not — the fraction $\frac{m}{m+M}$ survives, the rest becomes heat.
Stage 2 (pendulum swings up, smooth): mechanical energy conserved, $\tfrac12(m+M)V^2 = (m+M)gh$ ; momentum is not — string tension and gravity act throughout.
The master result: bullet speed $v = \dfrac{m+M}{m}\sqrt{2gh}$ , with $h = L(1-\cos\theta)$ if the swing angle is given.
Never write one conservation law across both stages: energy across the embedding (or momentum across the swing) is the classic mark-killer.
Key idea: momentum bridges the collision; energy bridges the swing — two stages, two laws, never swapped.

The core insight

Split the event at the instant the bullet finishes embedding: across the collision use momentum conservation ( $mv = (m+M)V$ , energy is lost to heat), and across the swing use energy conservation ( $\tfrac12 V^2 = gh$ , momentum is drained by gravity and tension) — giving $v = \frac{m+M}{m}\sqrt{2gh}$ .

Where this lives today

The pendulum's two-stage logic outlived the instrument. The Charpy impact test — a standard in every materials lab and quality-control floor — swings a heavy hammer through a notched steel specimen and reads the height lost on the follow-through: energy absorbed by fracture, measured by a swing, exactly Robins's bookkeeping run in reverse. Crash-test engineering does the same accounting between a car's momentum (conserved during the crumple) and the energy the structure must dissipate (not conserved — by design). And forensic ballistics still reconstructs bullet speeds from what they push and how far it moves — ballistic gel blocks are the pendulum without the strings. Meanwhile, in every undergraduate lab the original apparatus survives as the cleanest demonstration ever built of the difference between the two conservation laws: one number in ( $h$ ), one number out ( $v$ ), and 99.5% of the energy honestly declared missing.

The setup

A block of mass $M$ hangs at rest from light strings of length $L$ . A bullet of mass $m$ arrives horizontally at unknown speed $v$ , embeds in the block, and the combined body $(m+M)$ swings upward, rising a maximum height $h$ (equivalently, swinging through angle $\theta$ , with $h = L(1-\cos\theta)$ ). Given any one of $\{v, V, h, \theta\}$ plus the masses, find the others.

Figure

The ballistic pendulum — collide with momentum, swing with energy

Why can't one law cover the whole story? Look at what each stage does to each quantity. During the millisecond of embedding, enormous internal forces tear wood and deform metal — kinetic energy pours into heat, so any energy equation written across the impact is fiction; but those violent forces are internal to the bullet-block system, and the external ones (gravity, string tension) are far too small to matter in a millisecond, so momentum sails through conserved. During the half-second swing, the situation exactly reverses: nothing burns energy (smooth pivot, no drag), so mechanical energy is conserved; but gravity and string tension now have ample time to push, and they drain the system's momentum all the way to zero at the top of the swing. Two stages, two conservation laws, no overlap — that is the entire archetype.

The master solution pattern

Stage 1 — the collision (momentum only). Just before impact: bullet $mv$ , block at rest. Just after (bullet embedded, block barely moved): combined mass moving at $V$ .

mv = (m + M)\,V \qquad (1)

where: $V$ — the common speed the instant embedding completes. This is the perfectly inelastic collision of the previous module with $e = 0$ : one momentum equation, one unknown. Note what is not written: any energy statement. The kinetic energy surviving the impact is the capture fraction

\frac{KE_{\text{after}}}{KE_{\text{before}}} = \frac{m}{m+M} \qquad (2)

where: for a 10 g bullet in a 2 kg block this is $0.010/2.010 \approx 0.5\%$ — over 99% of the bullet's energy is gone before the swing even begins. That number is the quantitative reason the "energy all the way" shortcut fails so spectacularly.

Stage 2 — the swing (energy only). From the bottom of the arc to the top of the swing, only gravity does work (tension stays perpendicular to the motion):

\tfrac12 (m+M)V^2 = (m+M)\,g\,h \quad\Longrightarrow\quad V = \sqrt{2gh} \qquad (3)

where: $h$ — the rise of the centre of the block; if the problem gives the swing angle, $h = L(1-\cos\theta)$ . The masses cancel — the swing is pure kinematics of a rising body.

Chain the stages. Substitute (3) into (1):

v = \frac{m+M}{m}\,\sqrt{2gh} = \frac{m+M}{m}\,\sqrt{2gL(1-\cos\theta)} \qquad (4)

where: the amplification factor $(m+M)/m$ is huge — 201 for a 10 g bullet in a 2 kg block — which is exactly the instrument's genius: a bullet too fast to see produces a slow, measurable swing, and the factor scales the measurement back up.

Figure

Two ledgers across the two stages — what survives, what does not

given	find	route
$h$ (or $\theta$ , $L$ )	bullet speed $v$	eq. (3) for $V$ , then eq. (1)
$v$	rise $h$ or angle $\theta$	eq. (1) for $V$ , then $h = V^2/2g$
$v$ , $h$	unknown mass	chain eq. (4), solve for $m$ or $M$
any	heat generated	$Q = \tfrac12 mv^2 - \tfrac12(m+M)V^2$

Derivation AThe ballistic pendulum formula — and why each law stops where it does

We derive the master result (4), paying attention to the legality of each conservation law — because the derivation's real content is the two-stage discipline, not the algebra. Strategy: split time at the instant embedding completes; justify momentum conservation across stage 1 by comparing impulses; justify energy conservation across stage 2 by auditing the work done by each force. Watch how short the collision time is — that is what makes the external impulses negligible.

Step 1. Legality of momentum in stage 1. During the embedding time $\Delta t \sim 10^{-3}$ s, the external forces on the bullet+block system are gravity ( $\approx (m+M)g \approx 20$ N for a 2 kg block) and string tension (similar size). Their impulse is at most $\sim 20 \times 10^{-3} = 0.02\ \mathrm{kg\,m/s}$ — while the bullet delivers $mv \approx 0.010 \times 200 = 2\ \mathrm{kg\,m/s}$ , a hundred times more. External impulses are negligible: momentum is conserved across the impact to about 1%. Why the strings add no trouble: the impact is horizontal and the strings hang vertical, so tension is perpendicular to the motion — it has no horizontal component to steal momentum with.

Step 2. Momentum across the collision: $mv = (m+M)V$ , equation (1). One line, one unknown $V$ . Energy is NOT invoked — the embedding destroys $\tfrac12 mv^2 \times \frac{M}{m+M}$ of kinetic energy (heat, sound, deformation), which is nearly all of it.

Step 3. Legality of energy in stage 2. From the bottom of the swing to the top: tension acts perpendicular to the velocity at every instant (circular arc), so it does zero work; gravity is the only working force, and it is conservative. Mechanical energy is exactly conserved. Momentum is NOT: gravity and tension now act for $\sim 0.5$ s — thousands of times longer than the collision — and bring the momentum from $(m+M)V$ to zero at the top.

Step 4. Energy across the swing: $\tfrac12(m+M)V^2 = (m+M)gh$ , giving $V = \sqrt{2gh}$ , equation (3). Geometry converts an angle reading to height: the pendulum bob at angle $\theta$ has risen $h = L - L\cos\theta = L(1-\cos\theta)$ .

Step 5. Chain: $v = \dfrac{m+M}{m}\sqrt{2gh}$ , equation (4). The amplification $(m+M)/m$ is the design: heavy block, slow visible swing, huge lever back to the bullet's speed. Verified: limits — $M \to 0$ gives $v = \sqrt{2gh}$ (a lone bullet rising ballistically ✓); $M \gg m$ gives $v \approx (M/m)\sqrt{2gh}$ , the practical instrument. Numbers: $m = 10$ g, $M = 2.00$ kg, $h = 5.0$ cm, $g = 9.8$ : $V = \sqrt{2 \times 9.8 \times 0.05} = 0.99\ \mathrm{m/s}$ , $v = 201 \times 0.99 = 199\ \mathrm{m/s}$ — matching the standard textbook result (Kleppner, HC Verma).

Limiting cases

case	result	meaning
$M \to 0$	$v = \sqrt{2gh}$	no block: the bullet itself rises $h$ — plain projectile physics
$M \gg m$	$v \approx \dfrac{M}{m}\sqrt{2gh}$ , tiny $h$	the real instrument: huge amplification, gentle swing
$\theta \to 0$	$h \approx L\theta^2/2$	small swings: height grows with the square of the angle
$\theta = 90^\circ$	$h = L$	horizontal swing: the largest $h$ a taut-string swing can read
energy used across stage 1 (wrong)	$v_{\text{wrong}} = \sqrt{2gh}\,\sqrt{\tfrac{m+M}{m}}$	too small by $\sqrt{(m+M)/m}$ — the factor-of-14 error for a 200:1 mass ratio
bullet passes through	$mv = mv' + MV$	stage 1 gains one term; stage 2 unchanged

Common trap“Kinetic energy of the bullet = potential energy at the top of the swing”

This single line — $\tfrac12 mv^2 = (m+M)gh$ — is the most reliably wrong equation in the whole momentum chapter, and examiners bait it deliberately. It assumes kinetic energy survives the embedding; in fact the capture destroys the fraction $\frac{M}{m+M}$ of it — over 99% for a real bullet and block — as heat and splintered wood. The error is not small: for a 10 g bullet in a 2 kg block it underestimates the bullet's speed by a factor of $\sqrt{(m+M)/m} \approx 14$ . The two-stage discipline is not optional style; it is the physics. Momentum crosses the collision (external impulses negligible in a millisecond); energy crosses the swing (tension workless, gravity conservative). If you remember one sentence from this article: never write an energy equation across a perfectly inelastic event.

Worked examples

1 · Board level — the classic instrument (AP/A-Level pattern)

A $10.0\ \mathrm{g}$ bullet is fired horizontally into a $2.00\ \mathrm{kg}$ wooden block hanging from light strings. The block (with bullet embedded) rises $5.0\ \mathrm{cm}$ . Take $g = 9.8\ \mathrm{m/s^2}$ . Find (a) the speed of the block just after impact, (b) the bullet's speed, (c) the fraction of the bullet's kinetic energy that became heat.

Step 1. Stage 2 first — the swing hands us $V$ . Energy conservation from bottom to top: $V = \sqrt{2gh} = \sqrt{2 \times 9.8 \times 0.050} = \sqrt{0.98} = \boxed{0.99\ \mathrm{m/s}}$ .

Step 2. Stage 1 — momentum across the embedding: $mv = (m+M)V$ , so $v = \dfrac{2.010}{0.010} \times 0.99 = 201 \times 0.99 = \boxed{199\ \mathrm{m/s}}$ .

Step 3. Energy audit. Before: $\tfrac12 (0.010)(199)^2 = 198\ \mathrm{J}$ . After embedding: $\tfrac12(2.010)(0.99)^2 = 0.99\ \mathrm{J}$ . Heat: $198 - 0.99 \approx 197\ \mathrm{J}$ — a fraction $\dfrac{197}{198} = \boxed{99.5\%}$ . Sanity: the surviving fraction is $m/(m+M) = 0.010/2.010 = 0.5\%$ ✓ — the shortcut of equation (2) reproduces the long audit instantly.

2 · Board level — swing angle instead of height (IB/A-Level pattern)

A $10\ \mathrm{g}$ dart embeds in a $0.99\ \mathrm{kg}$ block hanging on a string of length $1.0\ \mathrm{m}$ . The pendulum swings through $60^\circ$ . Find the dart's speed. ( $g = 9.8\ \mathrm{m/s^2}$ .)

Step 1. Convert angle to rise: $h = L(1 - \cos 60^\circ) = 1.0 \times (1 - 0.5) = 0.50\ \mathrm{m}$ .

Step 2. Swing (energy): $V = \sqrt{2 \times 9.8 \times 0.50} = \sqrt{9.8} = 3.13\ \mathrm{m/s}$ .

Step 3. Collision (momentum): $v = \dfrac{m+M}{m}V = \dfrac{1.00}{0.010} \times 3.13 = \boxed{313\ \mathrm{m/s}}$ — a supersonic dart, measured with a protractor. Note the clean geometry gift: $\cos 60^\circ = \tfrac12$ makes $h$ exactly half the string length. Examiners reuse $60^\circ$ for this reason; memorize $1-\cos 60^\circ = 0.5$ and $1-\cos 37^\circ = 0.2$ .

3 · Competitive level — bullet passes through (JEE-Main pattern, ~3 minutes)

A $20\ \mathrm{g}$ bullet moving horizontally at $600\ \mathrm{m/s}$ passes through a $2.0\ \mathrm{kg}$ block suspended by long strings and exits at $400\ \mathrm{m/s}$ . Take $g = 10\ \mathrm{m/s^2}$ . Find the height to which the block rises.

Step 1. Stage 1, momentum with three terms now: $mv = mv' + MV$ , so $V = \dfrac{m(v - v')}{M} = \dfrac{0.020 \times (600-400)}{2.0} = \dfrac{4.0}{2.0} = 2.0\ \mathrm{m/s}$ .

Step 2. Stage 2, energy for the block alone (the bullet has left): $h = \dfrac{V^2}{2g} = \dfrac{4.0}{20} = \boxed{0.20\ \mathrm{m}}$ .

Step 3. Sense check: the bullet delivered only the momentum it lost, $m\Delta v = 4.0\ \mathrm{kg\,m/s}$ — the block's momentum, exactly. Timing note: ~2.5–3 min. The pass-through variant is one extra term in stage 1 and nothing else. Distractor watch: using the bullet's full momentum $mv$ (gives $h = 0.72$ m) or its energy loss (nonsense — most of that energy became heat inside the block).

4 · Competitive level — JEE-Advanced stretch: swing a full circle (~4 minutes)

A $50\ \mathrm{g}$ bullet embeds in a $0.95\ \mathrm{kg}$ block hanging from a light rigid ROD of length $1.0\ \mathrm{m}$ . Take $g = 10\ \mathrm{m/s^2}$ . Find the minimum bullet speed for the block to complete a full vertical circle. How does the answer change for a string?

Step 1. Full-circle condition for a rod: the combined body may barely crawl over the top with $V_{\text{top}} \approx 0$ (the rigid rod can support it). Energy bottom-to-top: $\tfrac12 V^2 = g(2L)$ , so $V = \sqrt{4gL} = \sqrt{40} = 6.32\ \mathrm{m/s}$ .

Step 2. Momentum across the embedding: $v = \dfrac{m+M}{m}V = \dfrac{1.00}{0.050} \times 6.32 = 20 \times 6.32 = \boxed{126\ \mathrm{m/s}}$ .

Step 3. For a string, gravity must not exceed the needed centripetal force at the top: $V_{\text{top}} = \sqrt{gL}$ , giving $V_{\text{bottom}} = \sqrt{5gL} = \sqrt{50} = 7.07\ \mathrm{m/s}$ and $v = 20 \times 7.07 = \boxed{141\ \mathrm{m/s}}$ .

Step 4. The architecture of the problem: three stages now — momentum (embed), energy (climb), circular-motion dynamics (the top condition from the vertical-circle archetype). Multi-concept chaining is precisely what JEE-Advanced is testing. Timing note: ~4 min. The entire decision is rod versus string — $\sqrt{4gL}$ versus $\sqrt{5gL}$ at the bottom. Tag the pivot type before writing anything; the two answers differ by 12% and both appear among the options.

Exam traps

Energy across the collision. The headline trap — see the pitfall box. Momentum for the embed, energy for the swing. The tell that you've sinned: your bullet speed comes out an order of magnitude too small.

Momentum across the swing. The mirror error. "Momentum at the bottom = momentum at the top" is false — gravity and string tension act throughout the half-second swing and eat all of it. Momentum conservation needs either no external force or an interval too short for external forces to matter. The swing is neither.

Forgetting the bullet rides along. After embedding, the moving mass is $m + M$ everywhere — in the stage-1 right side, in the swing energy, in a full-circle condition. In pass-through problems (Example 3), the bullet does not ride along: stage 2 belongs to the block alone.

$h$ versus $L$ . The rise is $h = L(1-\cos\theta)$ , not $L\sin\theta$ and not $L\theta$ . Draw the vertical drop from pivot to bob: the bob's height deficit below the pivot shrinks from $L$ to $L\cos\theta$ . For $\theta = 90^\circ$ , $h = L$ exactly.

The heat question. "Energy lost" means $\tfrac12 mv^2 - \tfrac12(m+M)V^2$ , computed across stage 1 only. Do not subtract the potential energy at the top — that energy is not lost; it came from the surviving kinetic energy and is stored, retrievable, in gravity.

Which $g$ ? AP allows $10\ \mathrm{m/s^2}$ for estimation; IB uses $9.8$ ; A-Level (CIE) uses $9.81$ ; JEE/NEET papers almost always declare $g = 10$ . Use what the paper states, and never mix values between the two stages.

Origins · 1 minBenjamin Robins and how this was discovered

Benjamin Robins

1707–1751 · London

Benjamin Robins, a self-taught Quaker mathematician, invented the ballistic pendulum and with it the science of ballistics. His 1742 book New Principles of Gunnery reported the first true measurements of musket-ball speeds — around 500 m/s, far faster than anyone had guessed — obtained by firing into a suspended pendulum and reading the swing, exactly the two-stage logic of this module. He went further: by measuring the same ball at different ranges he demonstrated that air drag at supersonic speeds was enormously larger than Newton's theory predicted, effectively discovering the sound barrier two centuries early. Euler himself translated Robins's book into German. Robins died young in Madras, engineering fortifications — but every chronograph, crash-test rig and impact pendulum since is his instrument, refined.

History

1742

— Benjamin Robins publishes New Principles of Gunnery: the ballistic pendulum turns an unmeasurable bullet speed into a measurable swing.

1745

— Euler translates Robins into German, adding commentary twice the length of the book; ballistics becomes mathematics.

1783

— Charles Hutton at Woolwich scales the method up to cannon, measuring shot above 600 m/s.

1900

s — Charpy's pendulum impact test adopts the same collide-then-swing logic to grade the toughness of steels.

1920

s — Electric chronographs finally time bullets directly; the pendulum retires from ballistics and moves into teaching labs.

Going further

Texts. HC Verma, Concepts of Physics Vol. 1, Ch. 9 (collision problems) · NCERT Physics Class 11, Ch. 6 · Kleppner & Kolenkow, An Introduction to Mechanics, Ch. 4 (the ballistic pendulum worked in full) · Robins, New Principles of Gunnery (1742), for the origin story.

Provenance. All results are standard. A verifier should recompute: $V = \sqrt{2\times9.8\times0.05} = 0.99$ m/s and $v = 201\times0.99 = 199$ m/s (Example 1); the surviving fraction $m/(m+M) = 0.5\%$ against the explicit audit $0.99/198$ ; Example 2's $h = 0.5$ m and $v = 313$ m/s; Example 3's $V = 2.0$ m/s, $h = 0.20$ m; Example 4's rod/string pair $126$ vs $141$ m/s from $\sqrt{4gL}$ and $\sqrt{5gL}$ with the factor 20.

Prerequisites. Inelastic Collisions & Coefficient of Restitution · Conservation of Linear Momentum · Conservation of Mechanical Energy & Efficiency. Leads to. Bullet & Block (archetype) · Which Conservation Law? (technique).

Why can't I just use energy conservation for the whole problem? Energy is always conserved!

Total energy is always conserved — but the swing equation uses mechanical energy (kinetic + potential), and the embedding converts kinetic energy into heat, sound and torn wood fibres, which the mechanical books don't track. Across the collision, mechanical energy genuinely decreases (by 99.5% for a real bullet and block). Momentum has no such leak: the violent forces are internal, and internal forces can never change a system's total momentum. So each stage keeps the law that survives it.

How do I know the collision is "fast enough" for momentum conservation?

Compare impulses. The bullet delivers momentum $mv \approx 2\ \mathrm{kg\,m/s}$ in about a millisecond. Gravity's impulse on the system in that millisecond is $(m+M)g\,\Delta t \approx 20 \times 0.001 = 0.02\ \mathrm{kg\,m/s}$ — one percent of the exchange. Whenever the interaction time is tiny, external forces simply don't get enough time to matter. This "impulsive approximation" is the licence behind every collision problem you solve; the ballistic pendulum just makes it explicit.

Doesn't the string tension mess up the momentum conservation during impact?

No, for two stacked reasons. First, the impact is horizontal and the string hangs vertical at that moment, so tension has no horizontal component at all — it literally cannot touch the horizontal momentum. Second, even its vertical effect over a millisecond is negligible (previous FAQ's impulse estimate). This is also why problems say the strings are long: a long pendulum stays nearly vertical during the brief embedding.

What changes if the bullet bounces back instead of embedding?

The block swings higher. Momentum conservation with the bullet recoiling at $v'$ (negative direction): $mv = -mv' + MV$ , so $MV = m(v + v')$ — the block receives more momentum than in the embedding case ( $mv$ versus $m(v+v')$ ). A perfectly elastic bounce transfers the most. This is the impulse-doubling idea from the momentum module: bounce-backs always deliver bigger kicks than captures. Exam versions of this usually give you $v'$ and ask for $h$ ; the two-stage recipe is unchanged.

Does it matter where the bullet hits the block?

For this module's analysis — no, as long as we treat the block as a particle on strings, which the standard problem does. In reality a bullet striking off-centre also sets the block rotating, stealing some momentum-transfer into spin, and the pendulum swings slightly less. That refinement needs angular momentum (the rotation chapter's collision archetype). Exam problems signal the simple case with "hits centrally" or just by silence; take the gift.

Why does the pendulum stop momentarily at the top but the momentum isn't zero "lost" — where did it go?

Into the Earth, via the string and gravity. Momentum is only conserved for systems free of external forces; during the swing, tension (anchored to the ceiling) and gravity (anchored to the planet) both pull on the block for half a second, transferring its momentum to the Earth — whose response is immeasurably small. Nothing is violated; the system for which momentum was conserved (bullet + block, during the millisecond of impact) simply isn't the system in play during the swing.

For the full-circle version, why √(5gL) for a string but √(4gL) for a rod?

The difference is what happens at the top. A rod can push — so the body may crawl over the top at essentially zero speed, and energy alone sets the requirement: $\tfrac12 V^2 = g(2L)$ gives $V = \sqrt{4gL}$ . A string can only pull — at the top, gravity must not exceed the required centripetal force, demanding $V_{\text{top}}^2 \ge gL$ ; feeding that back through energy gives $V = \sqrt{5gL}$ at the bottom. It is the vertical-circle archetype's signature result, imported here as stage 3 of the hardest pendulum variants.