ARCHETYPE: Sphere Potentials & Earthing

Every JEE paper-setter owns a set of Russian dolls: two or three concentric metal shells, a few charges sprinkled on them, and one innocent-looking word — earthed. That word is a trap with a hinge. Most students read "earthed" as "discharged", write zero charge on the earthed shell, and donate four marks to the examiner. The truth is stranger: earthing pins the shell's potential to zero, and the shell then takes on whatever charge the rest of the system forces it to hold. An earthed shell wrapped around $+q$ ends up carrying real, nonzero, negative charge — delivered up the earth wire on demand. You have already mastered half of this machine. In Concentric Shells & Cavity Charges you learned the induced-charge bookkeeping — inner walls mirror the cavity charge, leftovers exile to the outer surface — and that article deliberately parked the quantitative earthing questions here. The missing tool is a potential ledger: a one-line superposition recipe that prices the potential of every shell in the system. With it, "earthed" becomes an equation ( $V = 0$ ), the unknown charge falls out of algebra, and the classic results — an earthed inner shell acquires exactly $-\frac{a}{b}q$ , an earthed outer shell swallows the exterior field whole — become fill-in-the-blanks. This is a deliberately narrow archetype: CBSE and NEET stop at the qualitative story, but JEE-Main and JEE-Advanced return to these shells year after year. Learn the ledger once, and every doll in the set opens the same way.

Key takeaways

The master tool is the potential ledger: for a target radius $R$ , every charge at radius $r \le R$ contributes $kq/R$ , and every charge on a shell of radius $r > R$ contributes $kq/r$ . Sum the column — that is the potential.
Earthed means $V = 0$ , not $Q = 0$ . Write the ledger for the earthed shell, set it to zero, and solve for the unknown charge it carries.
The two classics: earthing the outer shell (radius $b$ ) around inner charge $q$ leaves the outer shell with $-q$ and kills the field outside; earthing the inner shell (radius $a$ ) under an outer charge $q$ at radius $b$ gives it $q' = -\dfrac{a}{b}\,q$ — a fraction, not the full mirror.
Charge flows through the earth wire: compare the shell's charge before and after earthing — the difference travelled to or from the planet, which is an infinite reservoir at $V = 0$ .
Joining concentric shells by a wire sends all the inner charge to the outer shell (the Faraday ice-pail fact), releasing calculable energy.
Key idea: earthing fixes potential, the ledger converts that into an equation, and algebra tells you exactly how much charge the Earth traded.

The core insight

The potential of any shell in a concentric system is the superposition $V = k\sum_i q_i/\max(r_i, R)$ ; earthing a shell imposes $V = 0$ on that one line of the ledger, determining the charge the shell must carry and hence the charge that flowed through the earth wire.

Where this lives today

The earthed outer shell is the working drawing of every screened cable and instrument enclosure around you. A coaxial cable is Example 1 stretched into a line: signal on the inner conductor, earthed braid outside, exterior world dark — which is why your neighbour's Wi-Fi cannot hear your oscilloscope lead. Substation engineers live inside Example 2's logic: any conducting body near high-voltage equipment acquires an induced potential unless deliberately bonded to earth, so ladders, fences and transformer tanks are all wired to ground and happily carry whatever charge that pins them at zero — the "earthed, not neutral" doctrine as a safety code. And electrostatic shielding in chip-scale sensors runs the three-doll problem in silicon: guard rings held at fixed potentials between a sensing node and a noisy neighbour, each ring's charge budget computed by exactly your ledger, just with a solver instead of a pencil.

The setup

Geometry first, because in an archetype the geometry is the physics. Concentric conducting shells of radii $a < b$ (sometimes a third at $c$ ), charges parked on some of them, and one shell wired to the ground symbol. Two facts run the whole show.

Fact one — a shell's potential story. A shell of radius $r$ carrying charge $q$ creates potential $kq/r$ everywhere inside itself (constant — no field inside a shell means no potential change) and $kq/x$ at any exterior distance $x \ge r$ . So when you price the potential at some radius, each charge asks one question: am I outside or inside this radius? Outside charges (larger radius) contribute at their own radius; inside charges contribute at yours.

Fact two — the Earth is an infinite reservoir at zero. Wiring a conductor to the ground pins its potential at $V = 0$ (the same zero as infinity, by convention) and lets charge flow freely either way until that condition holds. The planet is so large that nothing you deposit or withdraw shifts its potential.

Figure

The canonical earthed-shell systems — the ledger prices every shell

Look at the two panels. Earthing the outer shell around $+q$ forces it to $-q$ exactly — the exterior goes field-free, the interior does not care. Earthing the inner shell under an outer $+q$ produces only the fraction $-\frac{a}{b}q$ . Both verdicts come from the same one-line equation, and neither shell ends up neutral. That asymmetry — full mirror versus fraction — is the heart of the family, and guessing it without the ledger is how marks die.

The master ledger

Set up the pricing rule once. For shells of radii $a < b < c$ carrying total charges $q_a, q_b, q_c$ , the potential at the shell of radius $R$ is

V(R) = k\left[\sum_{r_i \le R} \frac{q_i}{R} + \sum_{r_i > R} \frac{q_i}{r_i}\right] \qquad (1)

where: $k = 9.0\times10^9\ \mathrm{N\,m^2/C^2}$ ; charges at or inside radius $R$ are priced at $R$ (their shells look like point charges from outside), while charges on larger shells are priced at their own radius (their potential is flat everywhere inside them). Read it physically: an outer charge cannot push the potential of an inner shell around by geometry — it contributes one constant number wherever you stand inside it. Two immediate specialisations you will use every time. The potential of the innermost shell in a two-shell system:

V(a) = k\left(\frac{q_a}{a} + \frac{q_b}{b}\right), \qquad V(b) = k\left(\frac{q_a + q_b}{b}\right) \qquad (2)

where: the outer shell's potential collapses to "total enclosed over $b$ " because everything is at or inside $b$ . That collapse is why earthing the outer shell always demands total charge zero — the single cleanest verdict in the archetype. The earthing move itself is one sentence: write the ledger line of the earthed shell, set it to zero, solve for that shell's unknown total charge. The charge that flowed through the earth wire is the difference between what the shell held before and what the ledger now says it holds.

\text{earthed shell:} \quad V_{\text{shell}} = 0 \;\;\Rightarrow\;\; q_{\text{shell}} = \text{(algebra)}, \qquad \Delta q_{\text{wire}} = q_{\text{after}} - q_{\text{before}} \qquad (3)

where: $\Delta q_{\text{wire}} > 0$ means positive charge climbed up from the planet (equivalently, electrons drained down). Where the acquired charge sits — inner or outer surface — is the induced-charge bookkeeping of Concentric Shells & Cavity Charges; this article prices the totals.

move	ledger consequence	classic verdict
earth the outer shell ( $b$ )	$q_a + q_b = 0$	outer takes $-q_a$ ; exterior field dies
earth the inner shell ( $a$ )	$\dfrac{q_a}{a} + \dfrac{q_b}{b} = 0$	$q_a = -\dfrac{a}{b}\,q_b$ — a fraction
join inner and outer by a wire	one conductor: all charge to outer surface	$V$ common; energy released
nothing earthed	ledger just prices the potentials	$V(a) - V(b)$ depends on $q_a$ only

Derivation AEarthing the inner shell gives the fractional mirror

We derive the single most-quoted result of the archetype: a small earthed shell of radius $a$ sitting inside a charged shell of radius $b$ carrying $q$ acquires $q' = -(a/b)\,q$ . The strategy is pure ledger: write the inner shell's potential as the superposition of its own charge (priced at $a$ ) and the outer charge (priced at $b$ ), impose $V(a) = 0$ , and solve. Keep one eye on the physics of the answer — the fraction must vanish as the shells separate and approach the full mirror as they close up, and checking those limits is worth a mark's confidence in any exam.

Step 1. Let the earthed inner shell hold unknown total charge $q'$ . Price its potential with equation (1): its own charge contributes at radius $a$ , the outer shell's charge contributes at its own radius $b$ (we sit inside it):

V(a) = k\left(\frac{q'}{a} + \frac{q}{b}\right)

Step 2. Earthed means this line reads zero:

\frac{q'}{a} + \frac{q}{b} = 0 \;\;\Rightarrow\;\; q' = -\frac{a}{b}\,q

Why: the outer charge lifts the inner region to $kq/b$ ; the shell must manufacture exactly enough negative potential, $kq'/a$ , to cancel that lift. Because its own charge is priced at the small radius $a$ , a modest $q'$ has outsized leverage — only the fraction $a/b$ of the full mirror is needed.

Step 3. The charge that flowed: the shell started neutral, so $-\frac{a}{b}q$ arrived through the earth wire (for $q > 0$ : electrons climbed up from the planet).

Step 4. Sanity limits. $b \to \infty$ : $q' \to 0$ — a faraway outer charge barely lifts the inner potential, so earthing has almost nothing to cancel. $b \to a$ : $q' \to -q$ — shells nearly touching recover the full mirror. Both endpoints behave. Verified: with $a = 4.0$ cm, $b = 12$ cm, $q = 6.0\ \mu\mathrm{C}$ : $q' = -(4/12)(6.0) = -2.0\ \mu\mathrm{C}$ , and the ledger check $k(q'/a + q/b) = 9.0\times10^9(-2.0\times10^{-6}/0.04 + 6.0\times10^{-6}/0.12) = 9.0\times10^9(-5.0\times10^{-5} + 5.0\times10^{-5}) = 0$ ✓.

Go deeper · Derivation B · 1 minEarthing the outer shell and what the wire carries

Derivation BEarthing the outer shell and what the wire carries

We derive the other classic — earth the outer shell of radius $b$ around an inner shell at radius $a$ carrying $q$ — and track the charge through the wire, because JEE loves asking for the flow, not just the final state. The strategy: the outer shell's ledger line contains only "total enclosed over $b$ ", so setting it to zero is immediate; the interesting physics is reading off what changed. Expect the exterior universe to go dark while the space between the shells keeps its field untouched.

Step 1. Let the outer shell hold unknown total $q_b$ . Everything sits at or inside radius $b$ , so its ledger line is the collapsed form of equation (2):

V(b) = k\,\frac{q + q_b}{b} = 0 \;\;\Rightarrow\;\; q_b = -q

Step 2. Flow through the wire: if the outer shell began neutral it now holds $-q$ , so a charge $-q$ climbed up from the Earth (electrons, for positive $q$ ). If it began with some endowment $Q$ , the flow is $-q - Q$ — always whatever erases the total.

Step 3. Field census after earthing. Outside ( $r > b$ ): enclosed total $q + (-q) = 0$ , so $E = 0$ — the system is electrically invisible. Between the shells ( $a < r < b$ ): enclosed charge is still $q$ , so $E = kq/r^2$ exactly as before — earthing the boundary never touched the interior story. Why: the induced $-q$ sits entirely on the outer shell's inner surface (the bookkeeping of Concentric Shells & Cavity Charges); its outer surface holds zero, which is the surface-level reading of "the exterior went dark".

Step 4. Potential census: $V(b) = 0$ by construction, and the inner shell sits at $V(a) = k q\left(\frac{1}{a} - \frac{1}{b}\right) > 0$ — the potential difference survives earthing untouched, because it is financed entirely by the enclosed charge $q$ . Verified: $a = 5.0$ cm, $q = +2.0\ \mu\mathrm{C}$ , $b = 10$ cm earthed: $q_b = -2.0\ \mu\mathrm{C}$ ; $V(a) = 9.0\times10^9 \times 2.0\times10^{-6} \times (20 - 10) = 1.8\times10^5$ V; $E(r>b) = 0$ ✓. The difference $V(a) - V(b) = 1.8\times10^5$ V matches $kq(1/a - 1/b)$ with or without the earth wire ✓.

Figure

The wire join — every coulomb exits to the outer surface

The variants examiners actually pose

"How much charge flows?" Always before-minus-after on the earthed shell. Shells with prior endowments make the arithmetic longer, never different.

The three-doll set. Shells at $a < b < c$ with the middle one earthed: its ledger line is $\frac{q_a + q_b}{b} + \frac{q_c}{c} = 0$ — inner charges priced at $b$ , the outer one at its own radius $c$ . One equation, one unknown $q_b$ . This is the standard JEE-Advanced escalation and Example 3 below.

The wire join. Connect inner and outer shells: they become one conductor, and all the charge ends on the outer surface — zero field anywhere inside, both shells at $V = k q_{\text{tot}}/b$ . The released energy is the difference of the stored energies, Example 4.

The invariant difference. For any two-shell system, $V(a) - V(b) = kq_a\left(\frac{1}{a} - \frac{1}{b}\right)$ : only the inner charge finances the gap. Dump extra charge on the outer shell, earth it, shake it — the difference is loyal to $q_a$ alone. (Stored as a seed: this invariance is exactly why a spherical capacitor has a well-defined capacitance.)

Limiting cases

case	result	meaning
earthed inner, $b \to \infty$	$q' \to 0$	distant outer charge lifts nothing; nothing to cancel
earthed inner, $b \to a$	$q' \to -q$	touching shells recover the full mirror
earthed outer, any $a$	$q_b = -q_a$ always	outer ledger sees only the total; geometry drops out
earth wire removed after earthing	charges frozen at earthed values	the wire enforces $V=0$ only while attached
join shells by wire	all charge to radius $b$ , $E = 0$ inside	one conductor, one outer skin — the ice-pail fact
$q_a = 0$ , outer earthed	nothing flows	ledger already reads zero; earthing is a no-op

Worked examples

Board level — earthed outer shell (JEE-Main pattern, single concept)

A conducting sphere of radius $a = 5.0\ \mathrm{cm}$ carrying $+2.0\ \mu\mathrm{C}$ is surrounded by a concentric conducting shell of radius $b = 10\ \mathrm{cm}$ , which is earthed. Find (a) the charge on the outer shell, (b) the potential of the inner sphere, (c) the electric field outside the shell. $k = 9.0\times10^9\ \mathrm{N\,m^2/C^2}$ .

Step 1. (a) The outer shell's ledger line: $V(b) = k(q + q_b)/b = 0$ , so $q_b = \boxed{-2.0\ \mu\mathrm{C}}$ — the full mirror, delivered by the earth wire.

Step 2. (b) Ledger for the inner sphere — own charge at $a$ , shell's charge at $b$ :

V(a) = k\left(\frac{2.0\times10^{-6}}{0.050} - \frac{2.0\times10^{-6}}{0.10}\right) = 9.0\times10^9 \times 2.0\times10^{-6}\,(20 - 10) = \boxed{1.8\times10^5\ \mathrm{V}}

Step 3. (c) Outside, the enclosed total is $2.0 - 2.0 = 0$ : $E = \boxed{0}$ everywhere beyond $b$ . The earthed shell hides the system from the world. Sanity: $V(a) > 0$ (positive charge dominates its own neighbourhood), $V(b) = 0$ ✓, and between the shells the field is untouched $kq/r^2$ .

Board level — earthed inner shell (JEE-Main pattern, the fraction)

A shell of radius $b = 12\ \mathrm{cm}$ carries $+6.0\ \mu\mathrm{C}$ . A concentric conducting shell of radius $a = 4.0\ \mathrm{cm}$ inside it is connected to the earth. Find (a) the charge acquired by the inner shell, (b) the charge that flowed through the earth wire and its direction, (c) the potential of the outer shell.

Step 1. (a) Inner ledger line: $\dfrac{q'}{a} + \dfrac{q}{b} = 0 \Rightarrow q' = -\dfrac{a}{b}q = -\dfrac{4.0}{12}(6.0\ \mu\mathrm{C}) = \boxed{-2.0\ \mu\mathrm{C}}$ — the fraction, not the mirror.

Step 2. (b) The shell began neutral: $-2.0\ \mu\mathrm{C}$ arrived, i.e. $\boxed{2.0\ \mu\mathrm{C}}$ of electrons flowed up from the earth to the shell (equivalently, $+2.0\ \mu\mathrm{C}$ of conventional charge flowed shell → earth).

Step 3. (c) Outer shell's ledger — everything enclosed: $V(b) = k\dfrac{q' + q}{b} = \dfrac{(9.0\times10^9)(4.0\times10^{-6})}{0.12} = \boxed{3.0\times10^5\ \mathrm{V}}$ . Sanity: $V(a) = 0$ by construction (check: $k(-2.0\ \mu\mathrm{C}/0.04 + 6.0\ \mu\mathrm{C}/0.12) = k(-5.0 + 5.0)\times10^{-5} = 0$ ✓); the outer shell stays at a healthy positive potential — earthing the neighbour did not earth it.

Competitive level — the three-doll set (JEE-Advanced pattern, ~4 minutes)

Concentric conducting shells A, B, C have radii $a = 2.0\ \mathrm{cm}$ , $b = 4.0\ \mathrm{cm}$ , $c = 6.0\ \mathrm{cm}$ . A carries $+4.0\ \mu\mathrm{C}$ and C carries $+6.0\ \mu\mathrm{C}$ ; shell B is earthed. Find (a) the charge on B, (b) the potential of A, (c) the field at $r = 5.0\ \mathrm{cm}$ .

Step 1. (a) Ledger line for B — A and B priced at $b$ , C priced at its own radius $c$ :

\frac{q_A + q_B}{b} + \frac{q_C}{c} = 0 \;\Rightarrow\; q_B = -q_A - \frac{b}{c}q_C = -4.0 - \frac{4.0}{6.0}(6.0) = \boxed{-8.0\ \mu\mathrm{C}}

Step 2. (b) Ledger for A — own charge at $a$ , B at $b$ , C at $c$ :

V(A) = 9.0\times10^9\left(\frac{4.0\times10^{-6}}{0.020} - \frac{8.0\times10^{-6}}{0.040} + \frac{6.0\times10^{-6}}{0.060}\right) = 9.0\times10^9(2.0 - 2.0 + 1.0)\times10^{-4} = \boxed{9.0\times10^5\ \mathrm{V}}

Step 3. (c) At $r = 5.0$ cm (between B and C) the enclosed total is $4.0 - 8.0 = -4.0\ \mu\mathrm{C}$ :

E = \frac{k\,|q_{\text{enc}}|}{r^2} = \frac{(9.0\times10^9)(4.0\times10^{-6})}{(0.050)^2} = \boxed{1.4\times10^7\ \mathrm{N/C}, \text{ pointing inward}}

Step 4. Read the surprise: B absorbed more than A's mirror ( $-8$ , not $-4$ ), because it must also cancel C's lift $kq_C/c$ . Between B and C the field now points inward — the earthed shell over-corrected the neighbourhood. Both features are standard JEE-Advanced option-bait. Timing note: one linear equation, two ledger reads — 3.5 to 4 minutes. The only place students die is pricing C at $b$ instead of $c$ in Step 1. Chant the rule: outer charges at their own radius. And keep charges in $\mu$ C and radii in metres separately; mixing them mid-line breeds power-of-ten wrecks.

Competitive level — the wire join and the energy bill (JEE-Advanced flavour, ~3 minutes)

A sphere of radius $a = 2.0\ \mathrm{cm}$ carrying $+3.0\ \mu\mathrm{C}$ sits inside a neutral concentric shell of radius $b = 6.0\ \mathrm{cm}$ . The two are connected by a fine wire. Find (a) the final charge distribution, (b) the common potential, (c) the energy released.

Step 1. (a) Once joined they are one conductor: no field may survive between them, so all $3.0\ \mu\mathrm{C}$ migrates to the outer surface at radius $b$ . Inner sphere: $\boxed{0}$ ; outer surface: $\boxed{+3.0\ \mu\mathrm{C}}$ . (This is Faraday's ice-pail verdict, now with numbers.)

Step 2. (b) The system is a single charged sphere of radius $b$ : $V = \dfrac{kq}{b} = \dfrac{(9.0\times10^9)(3.0\times10^{-6})}{0.060} = \boxed{4.5\times10^5\ \mathrm{V}}$ .

Step 3. (c) Energy before — charge $q$ on radius $a$ inside a neutral shell stores $U_i = \dfrac{kq^2}{2a}$ in total (the neutral shell adds nothing to the stored energy ledger): $U_i = \dfrac{(9.0\times10^9)(3.0\times10^{-6})^2}{2(0.020)} = 2.03\ \mathrm{J}$ . After: $U_f = \dfrac{kq^2}{2b} = 0.675\ \mathrm{J}$ .

Step 4. Released: $\Delta U = \dfrac{kq^2}{2}\left(\dfrac{1}{a} - \dfrac{1}{b}\right) = \boxed{1.35\ \mathrm{J}}$ — dissipated in the joining wire during the microsecond of migration, by exactly the wire-independent logic of Charge Sharing & the Lost Energy. Timing note: parts (a) and (b) are 60 seconds of doctrine — all charge out, one sphere left. The energy part is where the marks hide: remember $U = kq^2/2R$ for a charged sphere and this is subtraction; try to integrate anything and the clock wins.

Exam traps

Earthed = neutral. The headline trap, worth restating as arithmetic: earthed inner shell under $+6.0\ \mu\mathrm{C}$ at three times its radius holds $-2.0\ \mu\mathrm{C}$ , not zero. Setting the earthed shell's charge to zero before writing the ledger is the four-mark donation.

Pricing outer charges at the wrong radius. In the ledger, a charge on a larger shell contributes $kq/r_{\text{its own}}$ , never $kq/R_{\text{target}}$ . The three-doll problem is engineered to catch exactly this (C priced at $b$ gives $q_B = -10\ \mu\mathrm{C}$ — wrong, and among the options).

The full mirror where the fraction belongs. Earthing the outer shell: full mirror $-q$ . Earthing the inner shell: fraction $-\frac{a}{b}q$ . Students memorise one answer and deploy it symmetrically; the ledger never confuses them because the two shells' lines genuinely differ.

Forgetting the flow direction. "Charge flows from shell to earth" versus "earth to shell" is a stated-answer mark in JEE-Main. Negative acquired charge = electrons climbed up = conventional current ran down the wire. Say which.

Removing the earth wire mid-problem. Once disconnected, the shell's charge is frozen at its earthed value — later changes (moving charges, adding shells) alter its potential but not its charge. Multi-step JEE-Advanced problems hinge on spotting which quantity is pinned at each stage: wire attached pins $V$ ; wire removed pins $q$ .

Using $E = 0$ logic on potentials. The field between the shells ignores the outer shell entirely; the potential does not — every shell's charge appears in every interior ledger line. Field census and potential census are different books; the trap is auditing one with the other's rules.

Power-of-ten hygiene. Radii arrive in centimetres, charges in microcoulombs, and $k$ in SI. Convert radii to metres inside the ledger and carry $\mu$ C symbolically to the end. All boards accept $k = 9.0\times10^9\ \mathrm{N\,m^2/C^2}$ ; JEE states it or expects it.

Origins · 1 minWilliam Thomson (Lord Kelvin) and how this was discovered

William Thomson (Lord Kelvin)

1824–1907 · Glasgow

Kelvin treated the Earth itself as a circuit component before anyone knew what a circuit truly was. As scientific commander of the Atlantic telegraph cable — the 1858 attempt that died and the 1866 triumph that stuck — he leaned on a fact this article turns into algebra: the planet is an infinite conducting reservoir at one fixed potential, so a single wire under the ocean plus an earth return completes a working circuit, and anything wired to ground is pinned to the same universal zero. His mirror-galvanometer read the whisper of signals that survived the crossing, and his obsession with clean grounding — bad earths meant drifting zeros and garbled dots — made "earthed" an engineering discipline rather than a metaphor. Every ground symbol in your shell problems, and the convention that it means $V = 0$ with charge on demand, is telegraph-era Kelvin, formalised for the answer sheet.

History

1747

— William Watson runs a telegraph-like circuit through the ground itself, showing the Earth conducts and can close a loop.

1843

— Faraday's ice-pail experiment proves charge lowered into a conductor ends up entirely on the outside — the wire-join verdict, done with a bucket.

1858

— The first Atlantic cable, with Kelvin aboard, uses an earth-return circuit: the planet officially becomes a component at potential zero.

1866

— The permanent Atlantic cable succeeds; clean earthing and Kelvin's mirror galvanometer turn intercontinental telegraphy into routine.

1913

— Millikan's oil-drop programme (built on charged-sphere electrostatics) fixes the elementary charge, closing the loop between shell theory and countable electrons.

Going further

Texts. HC Verma, Concepts of Physics Vol. 2, Ch. 31 (worked shell systems) · NCERT Class 12, Ch. 2 (potential of shells; earthing appears through examples) · JEE-Advanced archives 1990s–present — earthed-shell problems recur in every era.

Provenance. All results are standard electrostatics. A verifier should recompute: Derivation A's check line ( $-2.0\ \mu\mathrm{C}$ zeroes the 4/12 cm ledger); Example 1 ( $-2.0\ \mu\mathrm{C}$ , $1.8\times10^5$ V, exterior 0); Example 2 ( $-2.0\ \mu\mathrm{C}$ , $3.0\times10^5$ V); Example 3 ( $q_B = -8.0\ \mu\mathrm{C}$ , $V_A = 9.0\times10^5$ V, $E = 1.4\times10^7$ N/C inward at 5.0 cm); Example 4 ( $4.5\times10^5$ V, $1.35$ J released).

Prerequisites. Potential of Point Charges & Charge Systems · Concentric Shells & Cavity Charges · Electrostatics of Conductors & Shielding. Leads to. Sign Discipline in V, U & W · Kirchhoff's Rules (coming in E3).

Why does earthing set the potential to zero rather than the charge?

Because a wire enforces potential equality, not charge equality — connected conductors settle at one potential, and the Earth is so vast that its potential is unshakeably the reference zero. Charge then flows through the wire exactly until the shell's ledger line reads zero, and the amount required depends on the whole neighbourhood. Charge is the adjustable payment; potential is the pinned condition. Reading "earthed" as $V = 0$ , unknown $q$ , is the single reflex that unlocks every problem here.

How can the Earth "supply" negative charge without running out?

The planet holds an astronomically large sea of mobile electrons, and the microcoulombs your shell borrows are a rounding error of a rounding error. That is what "infinite reservoir" means operationally: transactions of laboratory size change the Earth's potential by an amount too small to measure, so its zero stays put.

Why does the earthed INNER shell take only a fraction −(a/b)q while the earthed OUTER shell takes the full −q?

Look at the leverage in each ledger. The outer shell's potential is $k(\text{total})/b$ — killing it needs the total to vanish, hence the full mirror. The inner shell must cancel a lift of $kq/b$ using charge priced at its own small radius $a$ ; since $kq'/a$ is very potent per coulomb, only $a/b$ of the mirror is needed. Smaller earthed shell, smaller fraction — and in the limit $b \to a$ the two answers agree at $-q$ , as they must.

Is the potential zero everywhere inside an earthed shell, or only on it?

Only on the shell itself (and through its metal). Inside the cavity the potential varies from point to point — in Example 2 the earthed shell sits at zero while the charged outer shell above it is at $+3.0\times10^5$ V, so the potential climbs as you move outward through the gap. Earthing pins one shell's ledger line; it does not flatten the potential landscape around it. Only a field-free region (like the inside of a single isolated shell) has constant potential.

What actually happens if I earth BOTH shells?

Two ledger lines, two zeros, two unknowns — solve simultaneously. For two shells with no other charges the only solution is $q_a = q_b = 0$ : everything drains away. But add a third charged shell outside and both earthed shells acquire nonzero charges satisfying both zero conditions at once. JEE-Advanced has posed exactly this; the method never changes, only the algebra grows one equation taller.

Does it matter WHERE on the shell the earth wire is attached, or how long it is?

No — the shell is a conductor, so every point of it is at the same potential, and the wire only needs to enforce "shell potential = Earth potential". Length and resistance change how fast the balancing charge flows (an RC-style transient), never the final state. The same wire-independence you met in Charge Sharing & the Lost Energy governs here: endpoints are fixed by potential conditions, not by the plumbing.

In multi-step problems, how do I keep track of what changes when?

One rule sorts everything: wire attached pins potential; wire removed pins charge. While the earth wire is connected, re-solve the ledger after every change — the shell's charge floats. The instant the wire is cut, freeze that shell's charge and let its potential drift with subsequent changes. Annotate each stage of the problem with "pinned: V" or "pinned: q" before touching algebra, and the classic JEE-Advanced sequencing traps lose their teeth.