ARCHETYPE: Connected Blocks, Strings & Constraints

Watch a freight locomotive ease sixty loaded wagons out of a yard. The engine pulls only the first coupling — yet sixty wagons move as one. Somewhere inside that train, each coupling carries a different force: the first hauls fifty-nine wagons, the last hauls one. Railway engineers can tell you every coupler's load without measuring it, because the whole train obeys a two-line calculation you are about to master. The connected-bodies problem — blocks in a row pushed across a floor, wagons on a string, masses chained over pulleys — is the single most-asked multi-body setup in mechanics examinations. It looks like it needs simultaneous equations for every block. It never does. The master pattern has exactly two moves: treat the whole train as one body to find the acceleration (internal forces cancel in pairs — Newton's third law working for you), then cut the train at the one interface you care about and apply Newton's second law to everything beyond the cut. Two lines: one for $a$ , one for the internal force. Every board poses this family — CBSE and ISC as two-and-three-block numericals, AP and A-Level as "connected particles," JEE with pushes from alternating ends and movable pulleys that double speeds and halve forces. The physics content is small and deep: internal forces come in third-law pairs, contact pushes and string pulls transmit motion differently, and an inextensible string is an equation. This article drills the setup as the concept: the geometry, the master solution, the constraint relations, and the four variants examiners rotate through year after year.

Key takeaways

Connected bodies (blocks touching, or joined by strings/rods) move with one acceleration as long as the connections hold — find it from the whole system: $a = F_{\text{ext,net}}/M_{\text{total}}$ . Internal forces cancel in third-law pairs.
To find any internal force, cut at that interface and apply $F = ma$ to everything on one side (choose the side with fewer forces): internal force $=$ (mass beyond the cut) $\times a$ (± its external forces).
Contact forces push; strings pull. Same bookkeeping, opposite personalities — a push from the left loads the interfaces differently than a pull from the right.
In a block train pushed by $F$ , the interface force drops interface by interface: each junction carries only the mass beyond it — the tension-profile idea in block form.
Movable pulleys change the constraint: the free end moves twice as far (and fast) as the load, so $a_{\text{end}} = 2a_{\text{load}}$ and the load feels $2T$ .
Key idea: solve the system for one $a$ , then cut at the asked interface — internal force = mass beyond the cut × a. Two lines, any train.

The core insight

Bodies joined by contacts or taut inextensible strings share one acceleration set by the external forces on the whole system; every internal force is then found by cutting at that connection and applying Newton's second law to the piece beyond the cut.

Where this lives today

The tension staircase is a daily engineering constraint. Heavy-haul railways run computations of exactly Example 2 for every consist: the front coupler of a 20,000-tonne ore train sees forces near its rated limit, which is why the world's longest trains put extra locomotives in the middle — distributed power flattens the staircase by shortening each hauled "piece beyond the cut." Truckers live the same equation at highway speed: the kingpin load on a tractor–trailer during braking is a connected-bodies cut, and anti-jackknife control is the art of keeping that interface force pointing the survivable way. In rock climbing, a hauling system built from movable pulleys ("3:1 Z-drag") trades rope speed for force exactly by equation (3) — rescuers count supporting strands the way you did in Example 4. And every container crane on every port skyline is a movable-pulley stack whose blocks multiply tension eight- or ten-fold, hoisting forty tonnes with a motor that could never lift it on a single line.

The setup

Geometry first, because in an archetype the geometry is the physics. The family portrait:

Figure

The connected-body family — block trains, string trains, and the cut that solves them

Two species of connection, one bookkeeping rule.

Contact (blocks touching). Each interface transmits a normal force — a push, never a pull. The blocks share an acceleration only while they press together; if the physics ever demands a pull, the blocks simply separate ( $N \ge 0$ is the reality check).

Strings and rods. A taut string transmits a tension — a pull, never a push (the whole personality profile lives in Tension & Ideal Strings). The inextensible string is also an equation: connected ends share one speed and one acceleration magnitude along the string, even around ideal pulleys, where the direction turns but the magnitude survives. And the deep simplification, the reason trains are easy: at every internal connection, the forces come in Newton's-third-law pairs — $m_1$ pushes $m_2$ exactly as hard as $m_2$ pushes back on $m_1$ . Sum Newton's second law over the whole train and every internal pair cancels. What survives is external forces only, driving the total mass. One equation, one $a$ , however long the train.

The master solution pattern

Move 1 — the system line. Add up external forces along the motion, divide by total mass:

a = \frac{F_{\text{ext, net}}}{m_1 + m_2 + \cdots + m_n} \qquad (1)

where: $F_{\text{ext,net}}$ — applied pulls/pushes plus gravity components, minus friction on the system; the denominator — every mass that moves. For three blocks totalling 9 kg pushed by 18 N on a smooth floor: $a = 2\ \mathrm{m/s^2}$ , before you have drawn a single FBD. Internal forces never appear — they cancelled in pairs, which is exactly why this line is legal.

Move 2 — the cut line. To find the force at any one connection, cut there and apply Newton's second law to everything on one side of the cut — normally the side with fewer external forces:

F_{\text{interface}} = (\text{mass beyond the cut}) \times a \;\; (\pm\ \text{external forces on that piece}) \qquad (2)

where: "beyond" means the piece not carrying the applied force. On a smooth floor the correction term vanishes and the interface force is purely (mass beyond) × $a$ — the block-train version of the dragged-rope tension profile. Each junction hauls only what is behind it: interface forces decrease toward the free end, coupler by coupler, exactly like the locomotive's train.

Derivation AThe three-block push — acceleration and both contact forces

We solve the canonical block train — masses $m_1, m_2, m_3$ in a row on a smooth floor, pushed by $F$ on $m_1$ — for the acceleration and the two interface forces, by the master pattern, then verify with a per-block audit. The strategy: system line for $a$ , one cut per interface. Watch which piece each cut isolates: always the piece the applied force does NOT touch.

Step 1. System line. External horizontal force: $F$ alone (smooth floor). Total mass $m_1 + m_2 + m_3$ :

a = \frac{F}{m_1 + m_2 + m_3}

Step 2. Cut between $m_1$ and $m_2$ . The piece beyond the cut is $(m_2 + m_3)$ ; the only horizontal force on it is the contact push $N_{12}$ :

N_{12} = (m_2 + m_3)\,a = \frac{(m_2 + m_3)F}{m_1 + m_2 + m_3}

Why: cutting on the far side of the applied force keeps the FBD one-force clean. Cutting the other way — the piece containing $m_1$ — also works: $F - N_{12} = m_1 a$ gives the same $N_{12}$ , one extra term.

Step 3. Cut between $m_2$ and $m_3$ . Beyond: just $m_3$ :

N_{23} = m_3\,a = \frac{m_3 F}{m_1 + m_2 + m_3}

Step 4. Audit the middle block, the piece both cuts touch: forward $N_{12}$ , backward $N_{23}$ (third-law partner of pushing $m_3$ ):

N_{12} - N_{23} = (m_2 + m_3)a - m_3 a = m_2\, a \;\checkmark

Newton's second law holds for $m_2$ — the pattern is self-consistent. Verified: limits — $m_3 = 0$ kills $N_{23}$ ; equal masses give $N_{12} = \tfrac{2}{3}F$ , $N_{23} = \tfrac{1}{3}F$ , the clean staircase. Numbers: $F = 18$ N, masses 4, 3, 2 kg: $a = 2\ \mathrm{m/s^2}$ , $N_{12} = 10$ N, $N_{23} = 4$ N, and the middle block's net $10 - 4 = 6 = 3 \times 2$ ✓.

Go deeper · Derivation B · 1 minThe string train — why the last coupling is the laziest

Derivation BThe string train — why the last coupling is the laziest

We derive the tensions in a train of blocks pulled by a force $F$ through connecting strings — the wagon problem. Same master pattern; the new physics is that strings pull, so the "mass beyond the cut" is now behind the puller, and tensions decrease toward the rear. Watch the direction of each tension on each block: strings pull inward along themselves from both ends.

Step 1. System line for blocks $m_1$ (front, where $F$ is applied), $m_2$ (middle), $m_3$ (rear), smooth floor:

a = \frac{F}{m_1 + m_2 + m_3}

Step 2. Cut the rear string (between $m_2$ and $m_3$ ). Beyond the cut: $m_3$ , pulled only by $T_2$ :

T_2 = m_3\, a

Step 3. Cut the front string (between $m_1$ and $m_2$ ). Beyond: $m_2 + m_3$ , pulled only by $T_1$ :

T_1 = (m_2 + m_3)\, a

Step 4. Read the structure. $T_1 > T_2$ : each coupling hauls only the wagons behind it, so the front coupler works hardest and the last is laziest. For equal masses the tensions run $\tfrac{2}{3}F, \tfrac{1}{3}F$ — the same staircase as the contact train, because the bookkeeping is identical whether interfaces push or pull. Verified: middle block audit: $T_1 - T_2 = (m_2 + m_3)a - m_3 a = m_2 a$ ✓. Limit $m_3 \to 0$ : rear string goes slack-equivalent ( $T_2 \to 0$ ), as an unloaded coupling should. Numbers: $F = 60$ N, equal 5 kg wagons: $a = 4\ \mathrm{m/s^2}$ , $T_1 = 40$ N, $T_2 = 20$ N.

Constraint relations — when connected bodies do NOT share one acceleration

The single- $a$ rule holds for straight trains and single fixed pulleys. A movable pulley breaks it in a precise, examinable way. Hang a load from a pulley that itself rides the string: two string segments support the pulley, so when the free end is hauled in by a length $\ell$ , the two supporting segments must shorten by $\ell$ between them — the load rises only $\ell/2$ .

v_{\text{end}} = 2\,v_{\text{load}}, \qquad a_{\text{end}} = 2\,a_{\text{load}}, \qquad F_{\text{on load}} = 2T \qquad (3)

where: the factor 2 appears in reciprocal pairs — the free end moves twice as fast, and the load enjoys twice the string's tension. That reciprocity is no accident (it is energy conservation wearing a constraint: force × speed must match through an ideal machine), and it generalizes: count the segments supporting the movable pulley, and that count is your factor. This is the "mechanical advantage" your gym's cable machine and every construction crane exploit.

How to generate any constraint relation: write the total string length as a sum of segments, then use "length is constant" — if one segment grows, others must shrink to match. At exam speed: tug the free end mentally by 1 cm and track where the centimetres come from.

Figure

Movable pulley bookkeeping — two segments shorten together, so the free end moves twice as fast as the load

Which cut answers which question

asked	move
acceleration of the train	system line (1) — never draw internal forces
contact force / coupling tension at one interface	cut there; Newton on the piece beyond: eq. (2)
force between blocks when the push swaps ends	re-run the cut — the answer changes (see Example 1)
movable pulley speeds/forces	constraint (3): factor 2, reciprocal pair
"does the train stay together?"	check $N \ge 0$ at contacts, $T \ge 0$ on strings

Limiting cases

case	result	meaning
equal masses, push $F$ on a train of $n$	$k$ -th interface carries $F(n-k)/n$	clean staircase down to the free end
push applied to the heaviest end	smallest interface forces	the interfaces haul less mass
connector mass $\to 0$	interface force continuous	massless strings transmit tension unchanged
$N$ at a contact $\to 0$	blocks separate	contacts cannot pull; re-solve as separate bodies
movable pulley, load held still	end force $= T$ , load feels $2T$	half the effort, twice the rope — no free lunch
rigid rod connector	can push AND pull	rod tension may be negative (compression) legally

Common trap"The contact force between the blocks equals the applied force"

Feels plausible — "the push is transmitted through" — and it is wrong at every interface except the trivial one. The applied force $F$ accelerates the entire train; by the time the push crosses the first interface, part of $F$ has already been spent accelerating the first block, and only the remainder — exactly $(m_{\text{beyond}}/m_{\text{total}})F$ — crosses over. Transmission with commission. The five-second audit: the interface force must vanish if the mass beyond it vanishes, and must approach $F$ only when the block being pushed directly is massless. Corollary examiners adore: push the same two blocks from the other side and the interface force changes — from $(m_2/(m_1+m_2))F$ to $(m_1/(m_1+m_2))F$ — because the mass beyond the cut changed. If your method gives the same interface force from both ends for unequal blocks, you are transmitting without commission, and the marking scheme will charge you for it.

Worked examples

1 · Board level — two blocks, push from each end (CBSE/AP pattern, 3 marks)

Blocks of $4.0$ kg and $2.0$ kg touch on a smooth floor. A horizontal force $F = 18$ N pushes the 4 kg block from the left. Find the acceleration and the contact force. Then the same force pushes the 2 kg block from the right — find the new contact force.

Step 1. System line: $a = \dfrac{18}{6.0} = \boxed{3.0\ \mathrm{m/s^2}}$ (both cases — total mass and external force unchanged).

Step 2. Push on 4 kg: cut the interface; beyond it lies the 2 kg block alone: $N = 2.0 \times 3.0 = \boxed{6.0\ \mathrm{N}}$ .

Step 3. Push on 2 kg (from the right): now the piece beyond the cut is the 4 kg block: $N' = 4.0 \times 3.0 = \boxed{12\ \mathrm{N}}$ .

Step 4. Same blocks, same $F$ , different interface force — the asymmetry is the concept. The interface carries whichever mass it must drag, and dragging the heavy block costs more. Audit: pushed block's FBD in case 1: $18 - 6 = 12 = 4 \times 3$ ✓.

2 · Board level — three wagons and their couplings (ISC/A-Level connected-particles pattern)

An engine pulls three wagons of $2000$ kg each along a level track with a net forward force of $12{,}000$ N (resistances already subtracted). Find the acceleration and the tension in each coupling.

Step 1. System: $a = \dfrac{12000}{6000} = \boxed{2.0\ \mathrm{m/s^2}}$ .

Step 2. Rear coupling (engine–side cut after wagon 2): beyond it, one wagon: $T_{\text{rear}} = 2000 \times 2.0 = \boxed{4000\ \mathrm{N}}$ .

Step 3. Middle coupling: beyond it, two wagons: $T_{\text{mid}} = 4000 \times 2.0 = \boxed{8000\ \mathrm{N}}$ .

Step 4. Front coupling (engine to wagon 1): all three wagons: $T_{\text{front}} = 6000 \times 2.0 = \boxed{12000\ \mathrm{N}}$ . The staircase $12000/8000/4000$ N — each coupler hauls only what trails it. This is why couplers fail at the front of heavy trains, and why the system line plus cuts beats six simultaneous equations.

3 · Competitive level — table block pulled both ways (JEE-Main pattern, ~3 minutes)

A block $M = 4.0$ kg on a smooth table is connected by light strings over ideal pulleys at opposite edges to hanging masses $m_1 = 1.0$ kg and $m_2 = 3.0$ kg. Take $g = 10\ \mathrm{m/s^2}$ . Find the acceleration and both tensions.

Step 1. One string each side, all taut: single acceleration magnitude $a$ , direction toward the heavier hanger ( $m_2$ falls). System line along the string path — drives: $m_2 g$ forward, $m_1 g$ backward:

a = \frac{(m_2 - m_1)g}{m_1 + M + m_2} = \frac{(3.0 - 1.0)(10)}{8.0} = \boxed{2.5\ \mathrm{m/s^2}}

Step 2. Tension on the light side — cut above $m_1$ (it accelerates upward): $T_1 = m_1(g + a) = 1.0 \times 12.5 = \boxed{12.5\ \mathrm{N}}$ .

Step 3. Tension on the heavy side — cut above $m_2$ (accelerating downward): $T_2 = m_2(g - a) = 3.0 \times 7.5 = \boxed{22.5\ \mathrm{N}}$ .

Step 4. Audit with the table block: $T_2 - T_1 = 22.5 - 12.5 = 10 = 4.0 \times 2.5$ ✓. Note $T_1 \ne T_2$ — two different strings, each with its own tension; the one-tension rule applies per string, never across separate strings. Timing note: the system line "unrolls" the whole path into a straight line — weight of the heavy hanger drives, light hanger opposes, total mass resists. Spot that and $a$ takes 30 seconds; the two cuts take another minute. The classic error is forcing one tension on both strings — the table block would then feel zero net force, contradiction. Budget 3 min.

4 · Competitive level — JEE-Advanced stretch: the movable pulley (~4 minutes)

Block $A$ ( $m_A = 2.0$ kg) on a smooth table is tied to a string that runs over a fixed pulley at the table edge, down and under a light movable pulley carrying hanging block $B$ ( $m_B = 4.0$ kg), then up to a fixed anchor. Take $g = 10\ \mathrm{m/s^2}$ . Find the accelerations of both blocks and the string tension.

Step 1. Constraint first. Two segments support the movable pulley; if $A$ moves a distance $x$ toward the edge, the supporting segments share the slack: $B$ descends $x/2$ . Hence $a_A = 2a_B$ . Call $a_B = a$ , so $a_A = 2a$ .

Step 2. One string throughout (ideal pulleys): single tension $T$ . FBD of $A$ along the table: $T = m_A(2a) = 4.0a$ .

Step 3. FBD of $B$ + movable pulley (massless): supported by two segments, so upward force $2T$ : $m_B g - 2T = m_B a \Rightarrow 40 - 2T = 4.0a$ .

Step 4. Substitute $T = 4.0a$ : $40 - 8.0a = 4.0a \Rightarrow a = \dfrac{40}{12} = 3.33\ \mathrm{m/s^2}$ . So $a_B = \boxed{3.3\ \mathrm{m/s^2}}$ down, $a_A = \boxed{6.7\ \mathrm{m/s^2}}$ along the table, $T = \boxed{13.3\ \mathrm{N}}$ . Timing note: movable-pulley problems die instantly to the two-line ritual — "count supporting segments" (factor 2 in forces on the load) and "tug 1 cm mentally" (factor 2 in accelerations, reciprocal side). Write $a_A = 2a_B$ BEFORE any FBD; students who FBD first tend to invent $a_A = a_B$ midway. Check the reciprocity: load feels $2T$ but moves at half the acceleration — the ideal-machine trade. Budget 4 min.

Exam traps

One tension across different strings. The uniform-tension rule covers one continuous string. Separate strings (Example 3's two sides) each carry their own $T$ — label them $T_1, T_2$ immediately, or the middle body's equation collapses into contradiction.

Transmitting the applied force undiminished. The interface force is never $F$ (unless nothing lies before the cut) — see the pitfall box. It is (mass beyond) × $a$ on a smooth floor.

Wrong piece after the cut. Cut and take the side without the applied force: one-force FBD. Taking the pushed side works but demands you keep $F$ and the interface force — dropping one is the standard slip.

Contacts that pull. If a block train decelerates (force from the front, or friction-heavy trailing blocks), a contact interface may demand a pull — impossible: the blocks separate and must be re-solved independently. Watch for $N < 0$ in your algebra; it is the physics announcing separation, not a sign error.

Movable-pulley factor on the wrong side. The load gets $2T$ and half the acceleration — never $2T$ and double acceleration (that machine would mint energy). If your numbers violate $F_{\text{end}} \cdot v_{\text{end}} = F_{\text{load}} \cdot v_{\text{load}}$ , the factor 2 is misplaced.

Forgetting gravity is external. In mixed table-and-hanging systems, hanging weights drive the system line; only the string routing changes direction. "Unroll the path" and sum drives minus opposers over total mass.

Which $g$ . CBSE/ISC 9.8; JEE/NEET as stated (usually 10); AP 10 permitted; A-Level 9.81. Coupling-force staircases scale linearly with the applied force, so a wrong $g$ shifts every rung.

Origins · 1 minJoseph-Louis Lagrange and how this was discovered

Joseph-Louis Lagrange

1736–1813 · Turin, Berlin & Paris

Joseph-Louis Lagrange looked at mechanics' growing zoo of pulleys, rods, and constrained blocks and asked the ruthless question: why solve for forces you never wanted to know? His Mécanique analytique (1788) — a book he boasted contained not a single diagram — rebuilt mechanics so that constraints (the fixed string lengths and rigid contacts of this article) are handled by choosing coordinates that respect them from the start, making the constraint forces vanish from the equations entirely. The "system line" you used today is the elementary shadow of that idea: internal forces cancel, so don't compute them unless asked. Lagrange's machinery — generalized coordinates, the Lagrangian — became the native language of advanced physics, from robotics to quantum field theory. Napoleon called him "the lofty pyramid of the mathematical sciences"; Lagrange, characteristically, preferred quiet work to the compliment.

History

1608

— Stevin's clootcrans (wreath of spheres) settles equilibrium on connected inclines by pure thought.

1687

— Newton's third law makes internal-force cancellation exact: trains can be summed as systems.

1736

— Euler's Mechanica systematizes the cut principle for bodies and cords.

1788

— Lagrange's Mécanique analytique eliminates constraint forces wholesale — mechanics without diagrams.

1901

— railway knuckle couplers are standardized after decades of coupler failures — the tension staircase, learned in blood and steel.

Going further

Texts. NCERT Physics Class 11, Ch. 4 (Laws of Motion) · HC Verma, Concepts of Physics Vol. 1, Ch. 5–6 · Irodov §1.2 (constraint-rich problems, for the ambitious).

Provenance. All results are standard. A verifier should recompute: Example 1's $a = 3.0$ , $N = 6.0$ vs $N' = 12$ N; Example 2's staircase $12000/8000/4000$ N; Example 3's $a = 2.5$ , $T_1 = 12.5$ , $T_2 = 22.5$ N with the table-block audit; Example 4's $a_B = 10/3$ , $a_A = 20/3\ \mathrm{m/s^2}$ , $T = 40/3$ N.

Prerequisites. Newton's Second Law · Newton's Third Law · Tension & Ideal Strings · The Atwood Machine & Variants. Leads to. The Block-on-Incline Family · Stacked Blocks & Friction Between · Block on an Accelerating Wedge. Deep dive (UG). Constraints & Constraint Forces (UG).

Why am I allowed to treat several blocks as one system — isn't that cheating?

It is Newton's third law doing the work. Sum the second-law equations of every block: each internal force appears twice, once on each partner, equal and opposite — the pairs cancel identically, leaving (external forces) = (total mass)(common acceleration). Nothing is neglected; the internal forces are exactly eliminated, not approximated away. The only requirement is that the bodies genuinely share one acceleration — taut connections, no slipping, no separation. The moment they don't (slack string, blocks parting, movable pulleys), the system shrinks to whichever pieces still move together.

How do I choose where to cut and which side to analyze?

Cut at the interface whose force is asked — no other cut ever helps. Then analyze the side with fewer forces, which is almost always the side away from the applied force: on a smooth floor that piece feels only the interface force, so the answer is one multiplication, (mass beyond) × a. The other side is legal but carries both $F$ and the interface force — one extra term, one extra chance to drop a sign. With friction or gravity components in play, "fewer forces" is still the criterion; just include that piece's own external forces in equation (2).

When the push swaps ends, why does the contact force change but the acceleration doesn't?

The acceleration is a system property: same external force, same total mass, same $a$ — the system line is blind to where the force lands. The interface force is a cut property: it equals the mass beyond the cut times $a$ , and swapping ends swaps which block is "beyond." Pushing the heavy block means the interface drags only the light one (small $N$ ); pushing the light block makes the interface drag the heavy one (large $N$ ). Two different questions, two different answers, one consistent $a$ .

What if the connector is a rod instead of a string?

A rod can push as well as pull, so its internal force may come out negative (compression) without any physical alarm. This matters in decelerating trains: braking from the front puts the couplings into compression — a string would go slack and the wagons would pile onto each other's buffers, but a rod (or a railway buffer-and-coupler) carries the compression and the train stays a single system. Procedure: solve exactly as with a string, but accept negative tension as "the rod is being squeezed" rather than re-solving for separation.

In the two-pulley table problem, why do the two strings have different tensions? I thought pulleys don't change tension.

An ideal pulley preserves tension along one continuous string. The table setup has two separate strings — one to each hanging mass — and separate strings owe each other nothing. Each has its own uniform tension ( $T_1$ throughout the left string, $T_2$ throughout the right), and the difference $T_2 - T_1$ is precisely the net force that accelerates the table block. Quick audit for any setup: count the strings; that is the number of distinct tensions (then pulleys with mass, if any, split tensions further — but that is the rotation chapter's business).

Is there a fast way to get constraint relations for weird pulley systems?

Two rituals cover everything at this level. Ritual one: count the segments supporting each movable pulley — that count multiplies the tension delivered to it and divides its speed relative to the hauled end. Ritual two: tug 1 cm — mentally pull the free end by 1 cm and track, segment by segment, where the length comes from; whatever each body must do to conserve total string length is its displacement ratio, and the same ratios hold for velocities and accelerations. For genuinely tangled systems, write the string length as a sum of segment lengths and differentiate mentally ("length constant ⇒ growths cancel shrinkages"). The UG formalization of all this — constraints as equations, forces that enforce them — is the Constraints & Constraint Forces deep dive.