TECHNIQUE: Redraw the Circuit

JEE-Main has a favourite piece of theatre: take a circuit worth ninety seconds, and draw it wrong. Resistors slanted across diagonals, the battery hiding in a corner, wires looping over each other like headphone cables from a pocket. Students who know every rule in Current Electricity freeze — not because the physics is hard, but because the picture is hostile. The examiner is betting you will negotiate with the drawing instead of replacing it. Here is the secret the drawing is hiding: the picture is not the circuit. A circuit is nothing but a list of connections — which element ends touch which junctions. Wires have zero resistance, so a wire's two ends are the same electrical point, however long and tangled the artist made it. Label every junction and the scary diagram collapses into a short list — "6 Ω from A to B, 3 Ω from A to B, battery from A to B…" — which you may redraw any way you please, preferably as two clean rails with elements hanging between them. This module trains that replacement reflex as a five-step technique: label the nodes, walk the wires, list the elements, redraw between the terminals, then reduce with the ordinary rules. You already own the machinery — the reduction rules, Kirchhoff's discipline, the symmetry folds. What this technique buys is the entry ticket: sixty seconds of relabelling that turns nine out of ten "hard" circuit MCQs back into the easy ones they secretly are.

Key takeaways
  • The drawing is not the circuit: a circuit is a list of connections between labelled junctions, and any two drawings with the same list are the same circuit.
  • An ideal wire has zero resistance, so its two ends are one node — trace every wire and give everything it touches the same letter.
  • A resistor whose two ends carry the same label is short-circuited: it carries no current and drops out of the equivalent resistance.
  • Ideal meters redraw too: an ammeter is a plain wire (merge its ends), a voltmeter is an open gap (erase it) — then read the answer off the clean circuit.
  • Redraw between the terminals that matter: put the battery's two nodes as left and right rails and hang every element between the rails it actually touches.
  • Key idea: label the nodes first and the topology cannot lie to you — same connection list, same circuit, however different the picture looks.
The core insight

Because an ideal wire enforces zero potential difference between its ends, every point joined by plain wire is a single node; a circuit is therefore fully specified by which elements bridge which nodes, and redrawing that node list in clean form changes nothing physical while making series, parallel and dead branches visible at a glance.

Where this lives today

Every circuit-design tool on Earth runs this technique as its native language. A SPICE simulator never sees the schematic you drew — the drawing is compiled into a netlist, literally the Step-3 list ("R1 node2 node5 4.7k"), and all analysis happens on the labelled graph. Printed-circuit-board designers live the distinction daily: the clean schematic and the genuinely tangled physical layout are verified against each other by software comparing connection lists, because the pictures may look nothing alike as long as the lists match — the redraw license, industrialised.

The physical picture

Watch a topper receive one of these hostile diagrams. She does not squint at the wire crossings. She takes a pencil and starts colouring: every point connected to the battery's positive terminal by plain wire gets the letter A — through junctions, around corners, across the figure. Everything wired to the negative terminal gets B. Ninety seconds later the drawing has two letters on it and she writes a list: battery A–B, 6 Ω A–B, 3 Ω A–B, 2 Ω A–B. Three resistors between the same two nodes — a parallel trio wearing a costume.

Figure
Before and after — the tangle and the two-rail redraw are the same circuit
BEFORE: the exam drawing + 6 Ω 3 Ω 2 Ω three resistors, three different angles — panic bait AFTER: label nodes, redraw node A node B 6 Ω 3 Ω 2 Ω parallel trio: 1/R = 1/6 + 1/3 + 1/2 = 1 → R = 1 Ω same connection list = same circuit — the redraw changed nothing physical battery A–B · 6 Ω A–B · 3 Ω A–B · 2 Ω A–B the whole technique: replace the picture with the list, then redraw the list cleanly

Why is she allowed to do this? Because the physics never sees the artwork. Current and voltage are governed by which element ends share a junction — nothing else. A wire has zero resistance, so it supports no potential difference; electrically, its whole length is one point. Melt every plain wire down to a labelled dot and the circuit's true skeleton appears. That theory is worth making law-grade before we drill it.

The license: why redrawing never changes the physics

Every technique needs its warrant. Here the warrant is one line of Ohm's law plus one observation about what determines a circuit's behaviour. Start with the wire. Ohm's Law & Resistance prices any element at V=IRV = IR. An ideal connecting wire has zero resistance, so whatever current II it happens to carry,

Vwire=I×0=0(1)V_{\text{wire}} = I \times 0 = 0 \qquad (1)

where: VwireV_{\text{wire}} — the potential difference between the wire's ends; II — its current, which can be anything at all. Zero times anything is zero: an ideal wire's two ends sit at exactly the same potential, always. A wire is not a road between two places — electrically it is one place, stretched out for the artist's convenience.

Derivation AThe one-node license and the short-circuit corollary

We prove the two facts the whole technique stands on: first, that every set of points joined by plain wire may be treated as a single node without changing any current or voltage anywhere; second, the corollary that a resistor whose two ends are wired together carries no current at all. The strategy is to ask what actually determines the circuit's behaviour — node potentials — and show the redraw leaves every one of them untouched. Watch how little physics is needed: equation (1) does all the work, and the rest is bookkeeping.

Step 1. A circuit's state is fixed by the potential at each junction and the current through each element. Every element law — Ohm's law for resistors, the emf relation for batteries — connects the element's current to the potentials at its two end junctions. No element law mentions positions, lengths, angles, or how the connecting wires bend.

Step 2. By equation (1), all points joined by plain wire share one potential. Give them one label. Every element law now reads exactly as before — each element still sees the same pair of end potentials it always saw. Merging the wired points into one named node therefore changes no equation, hence no current and no voltage, anywhere in the circuit. Why the drawing is now irrelevant: the full set of circuit equations mentions only node labels and element values. Two drawings producing the same labelled list produce identical equations — they are the same circuit in different handwriting.

Step 3. The corollary. Suppose a resistor RR has both ends on the same node — a plain wire bridges it. The potential difference across it is the difference of one potential with itself: zero. By Ohm's law its current is I=0/R=0I = 0/R = 0.

Step 4. Conclude: a short-circuited resistor is electrically absent. Every microampere prefers the zero-resistance wire; for equivalent-resistance purposes, erase the resistor from the list. (It still sits in the physical circuit — idle, dissipating 02R=00^2R = 0.) Verified: solving the tangle of the hero figure numerically as drawn — three resistors 6, 3, 2 Ω between the same two nodes — gives 1.000 Ω, identical to the clean parallel redraw; adding a 4 Ω resistor bridged by a wire changes the answer by exactly nothing (its computed current is zero to machine precision).

One more license, because exams love meters. An ideal ammeter has zero resistance — electrically a wire, so its two ends merge into one node. An ideal voltmeter has infinite resistance — no current enters it, so erase it, leaving a gap, and evaluate its reading afterwards as the potential difference between the two nodes it watched. (Non-ideal meters are a different game — Real Meters: Loading Effects & Reading Errors owns that story; here every meter is ideal unless the stem says otherwise.)

The five-step redraw

Step 1 — plant the flags. Find the battery (or the two terminals the question asks between). Label its positive-side junction A and its negative-side junction B. These are your rails.

Step 2 — walk the wires. From A, follow every plain wire through every bend and junction; everything reachable without crossing an element is also A. Same from B. Then pick the next unlabelled junction, call it C, and walk again, until every element end has a letter. Most exam networks finish with three or four.

Step 3 — write the list. One line per element: "6 Ω: A–C", "battery: A–B", and so on. Two ends with the same letter? That element is short-circuited — strike it out (Step 3 of the derivation). This list is the circuit; the original picture is now scrap paper.

Step 4 — redraw between the rails. Draw A as a left rail, B as a right rail, intermediate letters as bars between them, and hang each surviving element between its two letters. Elements sharing both letters are parallel; chains through an intermediate letter that touches nothing else are series.

Step 5 — reduce and solve. Now Resistors in Series & Parallel applies by inspection; for anything still knotted, Kirchhoff's Rules or the symmetry folds of Equivalent-Resistance Puzzles take over — on a diagram that finally tells the truth.

Figure
The labelling walk on the wire-diagonal square — merge, then spot the dead branch
square ABCD, 4 Ω per side, plain wire A–C — find R between A and B A B C = A D 4 Ω 4 Ω 4 Ω 4 Ω plain wire: A and C are ONE node A (=C) B AB · 4 Ω CB · 4 Ω A → D → A loop via 4 Ω + 4 Ω dead branch — never reaches B: strike it survivors: two 4 Ω resistors, both A–B → R = 4 ∥ 4 = 2 Ω the dashed loop hangs off one node — no potential difference drives it, so it carries nothing between A and B

The judgment calls live in Step 2, and there are exactly two worth naming. Crossings without a dot are not junctions — standard convention draws connected crossings with a filled dot; no dot, the wires pass over each other like an overbridge. And a wire straight across a diamond is not a bridge resistor — it merges the diamond's two middle corners into one node, which is a different (and easier) circuit than the five-resistor Wheatstone shape it visually resembles. The balanced-bridge erase-or-short craft from Equivalent-Resistance Puzzles applies when a resistor sits in the middle and the balance condition holds; a plain wire in the middle needs no balance at all — merging is always legal.

what you seewhat it isredraw move
long tangled wire, no elements on itone nodesame letter everywhere it touches
resistor with a plain wire across itshort-circuited elementstrike it from the list
crossing with no connection dotoverpass, not a junctionkeep the labels separate
ideal ammeter in a branchzero resistancereplace by wire — merge its two ends
ideal voltmeter across two nodesinfinite resistanceerase; read VV between the nodes at the end
wire (not resistor) across a diamond's middlenode merge, not a Wheatstone bridgeone letter for both middle corners
element ends on nodes A and B in any orientationparallel with everything else on A–Bhang it between the rails

Worked examples — before and after

Each drill is fought twice: once negotiating with the hostile drawing, once with the redraw. Feel the seconds being bought.

Board level — the tangle that was a parallel trio

Three resistors of 6 Ω, 3 Ω and 2 Ω are connected between points A and B as shown (a deliberately tangled figure), with a 6 V battery across A–B. Find the equivalent resistance and the battery current.

Before. Hunt for series chains in the artwork; find none; reach for Kirchhoff — three unknowns, three minutes gone.

After (the redraw). Step 1. Label the battery's terminals A and B. Step 2. Walk the wires: every end of all three resistors lands on A or B — no third letter appears. Step 3. The list: 6 Ω A–B, 3 Ω A–B, 2 Ω A–B. Parallel trio. Step 4. Reduce: 1R=16+13+12=1\frac{1}{R} = \frac{1}{6} + \frac{1}{3} + \frac{1}{2} = 1, so R=1 ΩR = \boxed{1\ \Omega}. Step 5. Current: I=V/R=6/1=6 AI = V/R = 6/1 = \boxed{6\ \mathrm{A}}. Timing note: 60–75 seconds, most of it the labelling walk. The planted distractor is the series sum 11 Ω — the tangle read left-to-right as a chain. Boards award the sentence "all three resistors connect the same two points, hence parallel"; write it.

Board level — the square with a wire across the diagonal

Four resistors of 4 Ω each form a square ABCD (one resistor per side). A plain connecting wire joins A to C across the diagonal. Find the resistance between A and B.

Before. The figure looks like a bridge network begging for Kirchhoff — or for a Wheatstone balance test that does not apply (the diagonal is a wire, not a fifth resistor).

After. Step 1. The wire makes A and C one node: relabel C as A. Step 2. The list: side AB = 4 Ω between A–B; side BC = 4 Ω between B–A; side CD = 4 Ω between A–D; side DA = 4 Ω between D–A. Step 3. Read it: two resistors join A and B directly — parallel pair, 4∥4=2 Ω4 \parallel 4 = 2\ \Omega. The other two (CD and DA) run from A out to D and straight back to A — a loop hanging off one node that never touches B: a dead branch, struck from the list. Step 4. RAB=2 ΩR_{AB} = \boxed{2\ \Omega}. Timing note: 60 seconds. Two kills in one drill: the wire-merge and the dead branch. Solving the full four-node network numerically confirms 2.000 Ω — the merge is exact.

Competitive level — the fake Wheatstone (JEE-Main pattern, ~90 seconds)

In the diamond network, A is the left corner and C the right; AB = 2 Ω (upper left), BC = 4 Ω (upper right), AD = 4 Ω (lower left), DC = 2 Ω (lower right). A plain wire joins the middle corners B and D. An 8 V battery of negligible internal resistance is connected across A and C. Find the equivalent resistance and the battery current.

Before. The shape screams Wheatstone. Reflex check: P/Q=2/4≠S/T=4/2P/Q = 2/4 \ne S/T = 4/2 — not balanced, so the erase-or-short craft is illegal; panic suggests full Kirchhoff.

After. The middle branch is a wire, not a resistor — balance is irrelevant; merging is always legal. Step 1. Merge B and D into one node M. Step 2. The list: A–M carries 2 Ω and 4 Ω in parallel; M–C carries 4 Ω and 2 Ω in parallel. Step 3. Reduce: each pair is 2×42+4=43 Ω\frac{2 \times 4}{2 + 4} = \frac{4}{3}\ \Omega; the two pairs sit in series through M: R=43+43=83≈2.67 ΩR = \frac{4}{3} + \frac{4}{3} = \frac{8}{3} \approx \boxed{2.67\ \Omega}. Step 4. Current: I=88/3=3.0 AI = \frac{8}{8/3} = \boxed{3.0\ \mathrm{A}}. Timing note: 45 seconds after the merge. The examiners' two harvests here: students who "erase the middle branch like a balanced bridge" get (2+4)∥(4+2)=3 Ω(2{+}4) \parallel (4{+}2) = 3\ \Omega — wrong, because the network is not balanced and the middle branch is not a resistor; and students who burn three minutes on loop equations get the right answer late. Node analysis on the full network confirms 8/3 Ω exactly.

Competitive level — what does the ideal ammeter read? (JEE-Main/NEET pattern, ~2 minutes)

A 10 V battery (negligible internal resistance) is connected between A and D. From A, a 6 Ω resistor leads to B and a 3 Ω resistor leads to C. From B, a 3 Ω resistor leads to D; from C, a 6 Ω resistor leads to D. An ideal ammeter is connected between B and C. Find its reading and direction.

Before. Five unknown currents, junction equations at B and C, two loops — doable, slow, sign-error-prone.

After. Step 1. Ideal ammeter = wire: merge B and C into node M. Step 2. The list: A–M has 6∥3=2 Ω6 \parallel 3 = 2\ \Omega; M–D has 3∥6=2 Ω3 \parallel 6 = 2\ \Omega; series total 4 Ω4\ \Omega. Step 3. Battery current I=10/4=2.5 AI = 10/4 = 2.5\ \mathrm{A}; potential at M is 10−2.5×2=5 V10 - 2.5 \times 2 = 5\ \mathrm{V} (taking VD=0V_D = 0). Step 4. Branch currents by Ohm's law: into the merged node, IAB=(10−5)/6=5/6 AI_{AB} = (10-5)/6 = 5/6\ \mathrm{A} arrives on the B side and IAC=5/3 AI_{AC} = 5/3\ \mathrm{A} on the C side; out of it, IBD=5/3 AI_{BD} = 5/3\ \mathrm{A} leaves from B and ICD=5/6 AI_{CD} = 5/6\ \mathrm{A} from C. Step 5. Audit node B alone: in 5/65/6, out 5/35/3 — the deficit, 5/3−5/6=5/6 A5/3 - 5/6 = 5/6\ \mathrm{A}, must arrive through the ammeter. Reading 5/6≈0.83 A\boxed{5/6 \approx 0.83\ \mathrm{A}}, flowing from C to B. Timing note: ~2 minutes, and the ammeter branch was never given a variable — the merge solved the network, and one junction audit recovered the meter's current. This "solve merged, audit one junction" pattern retires a whole family of meter MCQs. Full numerical solution confirms 0.833 A from C to B.

Exam traps

Treating the drawing as data. Angles, lengths, and which resistor comes "first" carry zero information. The only data is the connection list. If your solution mentions where something is drawn, it is already wrong.

Missing the no-dot crossing. Two wires crossing without a junction dot do not touch. Merging them manufactures a different circuit — usually an easier one, which should itself raise suspicion.

Wheatstone reflex on a wire bridge. Erase-or-short belongs to balanced bridges with a resistor in the middle (Equivalent-Resistance Puzzles owns the license). A plain wire in the middle merges nodes whether or not anything balances. Check what the middle branch is first.

Erasing the ammeter instead of shorting it. Ideal ammeter → wire (merge); ideal voltmeter → gap (erase). Swapping the replacements produces confidently wrong circuits. Mnemonic: the ammeter sits in the lane and must not block it; the voltmeter watches across it and must not open a new one.

Forgetting the shorted resistor still exists physically. For equivalent resistance it drops out — but a question may ask its power (zero) or the current in the shorting wire (everything the branch carries). Strike it from the reduction, not from your awareness.

Re-deriving instead of relabelling. Symmetry folds and loop equations have their place — after the redraw. Spending them on a network a sixty-second relabel would flatten is the time-loss this technique exists to stop.

Origins · 1 minLeonhard Euler and how this was discovered
LE
Leonhard Euler
1707–1783 · Basel and St Petersburg

In 1736 Euler settled a puzzle the citizens of Königsberg argued over on Sunday walks: can one stroll cross each of the city's seven bridges exactly once? Euler's move was to declare the map irrelevant — he shrank each land mass to a dot, each bridge to a line between dots, and showed the walk is impossible by counting lines at each dot. Distances, shapes, the curve of the river: all discarded; only connections mattered. That act founded graph theory, and it is precisely the act you perform on every hostile circuit diagram — melt the artwork down to nodes and links, then reason on the skeleton. A century later Kirchhoff (a son of the same Königsberg) built circuit analysis on Euler's graphs, and today every simulator encodes circuits exactly this way. Euler produced roughly half his life's output after going blind — a man for whom the picture was never the point.

History
1736
— Euler abstracts Königsberg's bridges into dots and lines: topology is born, and with it the idea that only connections matter.
1827
— Ohm publishes the law that makes the zero-resistance wire a precise statement rather than a slogan.
1845
— Kirchhoff states the junction and loop rules on circuit graphs, importing Euler's abstraction into electricity.
1873
— Maxwell's Treatise analyses resistor networks node-by-node, redrawing them as connection tables.
1973
— SPICE at Berkeley encodes every circuit as a netlist — a labelled node list, the redraw technique made executable.

Going further

Texts. NCERT Physics XII, Ch. 3 exercises (several figures are deliberate tangles) · HC Verma Vol. 2, Ch. 32 objective problems · Irodov §3.1 for networks that stay hard after a perfect redraw.

Provenance. Fresh module (exam-technique class); drill networks chosen to complement, not repeat, the archetype gym. A verifier should recompute (all confirmed by numerical node analysis): the parallel trio 1 Ω1\ \Omega, 6 A at 6 V; the wire-diagonal square 2.000 Ω2.000\ \Omega with the CD–DA pair dead; the fake Wheatstone 83 Ω\tfrac83\ \Omega, 3.0 A at 8 V; the ammeter drill R=4 ΩR = 4\ \Omega, I=2.5I = 2.5 A, node potential 5 V, reading 5/6=0.8335/6 = 0.833 A from C to B.

Prerequisites. Resistors in Series & Parallel · Kirchhoff's Rules · ARCHETYPE: Equivalent-Resistance Puzzles. Leads to. TECHNIQUE: P = V²/R Ratio Craft — the redraw finds series and parallel; the ratio craft prices the power. Non-ideal meters live in ARCHETYPE: Real Meters — Loading Effects & Reading Errors; the full speed protocol returns in Circuit Solving at Speed (coming up in the E8 exam-craft band).

How do I actually do the labelling walk without missing anything?

Use a pencil tip and one rule: keep moving along plain wire, through every junction dot and bend, stopping only at an element (resistor, battery, meter). Everything touched gets the current letter. Lift, find an unlabelled element end, assign the next letter, walk again. When every end has a letter: same letter twice on one element ⇒ shorted; both letters shared by two elements ⇒ parallel. The walk feels slow the first five times and takes twenty seconds by the tenth.

Why does the current not prefer the shorter or straighter wire?

Because "short" and "straight" are properties of the drawing, not the circuit. What divides current is resistance — the current-division rule — and ideal wires all have exactly zero of it, whatever their drawn length. Real copper has a few milliohms, which matters in power engineering, but exam wires are ideal unless the stem prices them. Weighing geometric distance means you have started solving the picture instead of the circuit.

When a wire shorts a resistor, where does that branch's current actually go?

All of it flows through the shorting wire. The pair "resistor plus wire across it" is RR in parallel with 00, which is 00 — the pair as a whole still conducts, often heavily; it is the resistor that idles. A question can therefore ask "find the current in the shorting wire" and expect a large answer while the resistor's own current is zero. Strike the resistor from the reduction, keep the wire as part of its node.

How can merging two points possibly be legal when current flows between them through the wire?

Merging never claims no current flows through the wire — only that its ends share one potential, which equation (1) guarantees at any current. Circuit equations care only about potentials at element ends, so the merge changes no equation. Need the wire's own current (the ammeter drill)? Audit one junction: the mismatch between its other inflows and outflows is what the wire carries.

Is a wire across the middle of a bridge the same as a balanced bridge?

No — and the difference is the most common mark-loser in this family. A balanced bridge has a fifth resistor in the middle that carries no current because the ratio condition holds; only then may you erase or short it, per Equivalent-Resistance Puzzles. A wire in the middle carries whatever current it likes and merges its ends unconditionally — no ratio needs checking.

What if the network is still a mess after a correct redraw?

Then it is genuinely hard, and you escalate deliberately: symmetry folding and the balanced-bridge license from Equivalent-Resistance Puzzles first, Kirchhoff's Rules as the guaranteed fallback. The redraw was still worth it — loop equations on a clean three-letter diagram are half the length and a quarter of the sign errors. The technique's promise is not that every circuit becomes trivial; it is that no circuit stays artificially hard.

Before this

Where this leads