Capacitors and capacitance

Every phone that survives a call drop, every camera flash that fires in one bright pop, every laptop that keeps its clock ticking for a few seconds after you pull the charger — all of them are leaning on the same small device: the capacitor. It is nothing more than two conductors, close together but never touching, waiting to hold equal and opposite charge on command. Yet the amount of charge a given capacitor can hold per volt applied — its capacitance — turns out to be a fixed number set purely by geometry, as unshakeable a property of the device as its mass. This chapter asks a deceptively simple question: if I connect several capacitors together, in a chain or side by side, what single capacitance behaves just like the whole network? The answer splits cleanly into two rules — series and parallel — mirror images of the resistor rules you already know from circuits, except capacitors flip which quantity stays common. Get that flip wrong under exam pressure and half your marks vanish. We will also ask what it costs, in energy, to charge a capacitor up — because pushing charge onto a plate that already repels further charge takes real work, and that stored energy is exactly what a camera flash releases in a flash of a millisecond. By the end you should be able to look at any battery of capacitors — some in series, some in parallel — and collapse it, step by step, into one number, then use that number to find charge, voltage and stored energy for the whole circuit or any single capacitor inside it.

Key takeaways

Capacitance is defined as $C = Q/V$ — the charge stored per unit potential difference; the SI unit is the farad (F), with practical capacitors measured in microfarads ( $\mu\text{F}$ ) or picofarads (pF).
Capacitance is a fixed geometric property. It depends only on the shape, size and separation of the conductors and the medium between them — never on the charge or voltage actually applied, exactly as resistance does not depend on the current flowing.
In series, charge is common; reciprocals add. $\dfrac{1}{C_s} = \dfrac{1}{C_1}+\dfrac{1}{C_2}+\cdots$ , and $C_s$ is always smaller than the smallest individual capacitor.
In parallel, voltage is common; capacitances add directly. $C_p = C_1+C_2+\cdots$ , and $C_p$ is always larger than the largest individual capacitor.
Energy stored is quoted, not derived (CBSE syllabus explicitly drops the derivation): $U=\tfrac12CV^2=\tfrac12QV=\dfrac{Q^2}{2C}$ — three equivalent forms for three different combinations of known data.
Mixed networks reduce step by step: collapse the obviously-series or obviously-parallel groups first, replace them with a single equivalent capacitor, and repeat until one number remains.
Key idea: a capacitor's capacitance $C=Q/V$ is a fixed geometric property, series combinations shrink the equivalent capacitance below the smallest member while parallel combinations grow it beyond the largest, and the quoted energy formula $U=\tfrac12CV^2=\tfrac12QV=Q^2/2C$ lets you pick whichever form matches the two quantities a problem hands you.

The core insight

Capacitance $C=Q/V$ is a geometry-fixed constant of a conductor pair; series combination (common charge) always shrinks the equivalent capacitance below the smallest member while parallel combination (common voltage) always grows it beyond the largest, and the energy stored, $U=\tfrac12CV^2=\tfrac12QV=Q^2/2C$ , is quoted rather than derived at this level.

Where it fits in the syllabus

This sits inside the CBSE Class 12 Physics chapter "Electrostatic Potential and Capacitance" (Unit II, Electrostatics), coming right after dielectrics and polarisation are introduced qualitatively, and right before the parallel-plate capacitor gives capacitance a concrete geometric formula. The board expects you to define capacitance, know the farad and its sub-multiples, combine capacitors in series and parallel, and quote — never derive — the energy-stored formula; the CBSE 2025-26 syllabus is explicit that the energy derivation is out of scope, so no worked derivation of $U=\tfrac12CV^2$ should appear on your answer sheet or in your notes.

The physical picture

Picture two conductors — call them plates, whatever their real shape — separated by an insulating gap of vacuum, air, or a dielectric. Connect them to a battery (or otherwise transfer charge from one to the other) and one plate ends up with charge $+Q$ , the other with $-Q$ , and a definite potential difference $V$ appears between them because each plate, being a conductor, is an equipotential surface. Experiment and theory both show $Q$ is directly proportional to $V$ for any fixed geometry, so their ratio is a constant of the device:

C = \frac{Q}{V}

where: $Q$ is the magnitude of charge on either plate (coulombs), $V$ is the potential difference between the plates (volts), and $C$ is the capacitance (farads). The physical reading is that $C$ does not care how much you charge the capacitor — double $Q$ and $V$ doubles right along with it, so the ratio never moves. A farad is enormous for a real device (a football-field-sized parallel-plate capacitor barely reaches one farad), which is why everyday capacitors are rated in microfarads ( $1\ \mu\text{F}=10^{-6}\ \text{F}$ ) or picofarads ( $1\ \text{pF}=10^{-12}\ \text{F}$ ).

Figure

Two parallel metal plates with equal and opposite charge symbols, connected to a battery, with an arrow labelling the potential difference V between them and the charge Q marked on each plate

Capacitors in series. When capacitors sit one after another in a chain, the connecting wire between two neighbouring plates carries no net charge of its own — it is charged purely by induction as the battery pushes charge onto the outer plates — so every capacitor in the chain ends up storing the very same charge $Q$ . What differs is the potential drop across each, and those drops must add to the total applied voltage $V=V_1+V_2+V_3+\cdots$ . Substituting $V_i=Q/C_i$ into that sum, with $Q$ common, gives the reciprocal rule:

\frac{1}{C_s} = \frac{1}{C_1}+\frac{1}{C_2}+\frac{1}{C_3}+\cdots

where: $C_s$ is the single equivalent capacitance replacing the whole series chain, and $C_1, C_2, C_3, \ldots$ are the individual capacitances. Because reciprocals add, $C_s$ always ends up smaller than the smallest capacitor in the chain — exactly the way series resistors always give a resistance smaller than... no, larger than any single resistor; capacitors behave the opposite way to resistors here, which is precisely why this rule trips students who pattern-match too fast.

Capacitors in parallel. When capacitors are wired side by side between the same two nodes, every one of them sees the identical potential difference $V$ (they share the same two points), but each stores its own share of charge, and the total charge drawn from the battery is the sum $Q=Q_1+Q_2+Q_3+\cdots$ . Since $Q_i=C_iV$ with $V$ common, the equivalent capacitance is a plain sum:

C_p = C_1+C_2+C_3+\cdots

where: $C_p$ is the equivalent capacitance of the parallel bank. Adding more branches can only add more charge storage for the same voltage, so $C_p$ is always larger than the largest individual capacitor — the mirror image of the series result.

The physical picture, continued: energy stored

Charging a capacitor is not free. Moving the first bit of charge onto an initially uncharged plate costs almost nothing, but every subsequent bit must be pushed against the repulsion of charge already sitting there, so the work done grows as charging proceeds. The CBSE syllabus asks you to quote the resulting energy, not derive it:

U = \frac{1}{2}CV^2 = \frac{1}{2}QV = \frac{Q^2}{2C}

where: $U$ is the electrostatic potential energy stored in the charged capacitor (joules), and $C$ , $Q$ , $V$ carry their usual meanings. All three expressions are the same physical quantity written with different pairs of known variables — use $Q=CV$ to swap between them. Physically, this stored energy lives in the electric field between the plates, and it is exactly this energy that a camera flash capacitor releases in a fraction of a second, or that a defibrillator delivers to restart a heartbeat.

Derivation AReducing a mixed series-parallel capacitor network to one equivalent capacitance

The goal is to take a network combining both series and parallel connections and collapse it, using only the two combination rules already stated, to a single equivalent capacitance — the same reduction skill examiners test most often, since a lone series or lone parallel question is comparatively rare. The method is always the same: find the smallest sub-group that is purely series or purely parallel, replace it with its single equivalent capacitor, redraw the (now simpler) circuit, and repeat until one capacitor remains. No new physics enters beyond the two rules; the only skill is bookkeeping the redraw correctly at each stage.

Step 1. Identify the innermost purely-series or purely-parallel group. Consider three capacitors $C_1=2\ \mu\text{F}$ and $C_2=2\ \mu\text{F}$ in parallel with each other, and this parallel pair in series with $C_3=3\ \mu\text{F}$ . The innermost group here is the parallel pair $C_1$ and $C_2$ , since they share the same two nodes with nothing else attached between them.

Step 2. Collapse that group using its rule. Since $C_1$ and $C_2$ are in parallel, $C_{12}=C_1+C_2=2+2=4\ \mu\text{F}$ . Redraw the network: it is now just $C_{12}=4\ \mu\text{F}$ in series with $C_3=3\ \mu\text{F}$ — a two-element chain.

Step 3. Apply the remaining rule to the simplified chain. $C_{12}$ and $C_3$ are in series, so $\dfrac{1}{C_{\text{eq}}}=\dfrac{1}{C_{12}}+\dfrac{1}{C_3}=\dfrac{1}{4}+\dfrac{1}{3}=\dfrac{3}{12}+\dfrac{4}{12}=\dfrac{7}{12}$ , giving $C_{\text{eq}}=\dfrac{12}{7}\approx1.71\ \mu\text{F}$ .

Step 4. Sanity-check the result against both bounding rules. The equivalent, $1.71\ \mu\text{F}$ , is smaller than the smallest capacitor anywhere in the original network ( $2\ \mu\text{F}$ ) — correct, because the final combining step was a series step, and series results must fall below every member entering that step (here $4\ \mu\text{F}$ and $3\ \mu\text{F}$ , and indeed $1.71<3$ ). If the last step were parallel instead, you would check the opposite bound. This "is it plausible" check is the single fastest way to catch an arithmetic slip before losing marks.

Worked examples

Board level - series and parallel combination of two capacitors

Two capacitors of $6\ \mu\text{F}$ and $3\ \mu\text{F}$ are connected (a) in series, and (b) in parallel. Find the equivalent capacitance in each case.

Step 1. Series: $\dfrac{1}{C_s}=\dfrac{1}{6}+\dfrac{1}{3}=\dfrac{1}{6}+\dfrac{2}{6}=\dfrac{3}{6}=\dfrac{1}{2}$ , so $C_s=2\ \mu\text{F}$ .

Step 2. Parallel: $C_p=6+3=9\ \mu\text{F}$ .

Step 3. Check both against the bounding rules: $2\ \mu\text{F}$ is below the smaller member ( $3\ \mu\text{F}$ ), and $9\ \mu\text{F}$ is above the larger member ( $6\ \mu\text{F}$ ) — both consistent.

Answer: the series combination gives $2\ \mu\text{F}$ , and the parallel combination gives $9\ \mu\text{F}$ .

Competitive level - charge, voltage split and energy in a series pair, then energy comparison with parallel

The same two capacitors, $C_1=6\ \mu\text{F}$ and $C_2=3\ \mu\text{F}$ , are first connected in series across a $90\ \text{V}$ battery. Find (a) the charge on each capacitor, (b) the voltage across each, and (c) the total energy stored. Then compare with the energy stored if the same battery instead charges them in parallel.

Step 1. Series equivalent capacitance (from the board-level example): $C_s=2\ \mu\text{F}$ . In series, the same charge sits on both capacitors, equal to the charge on the equivalent: $Q=C_sV=(2\times10^{-6})(90)=1.8\times10^{-4}\ \text{C}$ , i.e. $180\ \mu\text{C}$ on each of $C_1$ and $C_2$ .

Step 2. Individual voltages from $V_i=Q/C_i$ : $V_1=\dfrac{1.8\times10^{-4}}{6\times10^{-6}}=30\ \text{V}$ , and $V_2=\dfrac{1.8\times10^{-4}}{3\times10^{-6}}=60\ \text{V}$ . Check: $V_1+V_2=30+60=90\ \text{V}$ , matching the battery — the smaller capacitor always takes the larger share of voltage in series.

Step 3. Total energy stored, using the equivalent capacitance directly: $U_s=\tfrac12C_sV^2=\tfrac12(2\times10^{-6})(90)^2=\tfrac12(2\times10^{-6})(8100)=8.1\times10^{-3}\ \text{J}$ .

Step 4. Parallel case for comparison: $C_p=9\ \mu\text{F}$ across the same $90\ \text{V}$ , so $U_p=\tfrac12C_pV^2=\tfrac12(9\times10^{-6})(8100)=3.645\times10^{-2}\ \text{J}$ . The parallel combination, having a larger equivalent capacitance at the same voltage, stores about $4.5$ times more energy than the series combination — a general rule worth remembering, not just this one number.

Answer: series case — $Q=180\ \mu\text{C}$ on each plate, $V_1=30\ \text{V}$ , $V_2=60\ \text{V}$ , $U_s=8.1\times10^{-3}\ \text{J}$ ; parallel case — $U_p=3.645\times10^{-2}\ \text{J}$ , roughly 4.5 times larger. Timing note: this is a two-and-a-half-minute problem if you keep the series equivalent from part (a) rather than recomputing; the fastest route to individual voltages is always $V_i=Q/C_i$ once the common charge is known, never trying to split $90\ \text{V}$ by inspection.

Common traps

Swapping which quantity is "common" — charge is common in series, voltage is common in parallel; write this at the top of your rough work before starting any network problem.
Believing series capacitance behaves like series resistance — it does not. Series resistors add directly and series capacitors add as reciprocals; the two rules are mirror images of each other, not the same rule.
Forgetting the bounding-value sanity check — a series answer larger than any input, or a parallel answer smaller than any input, is an automatic sign of an arithmetic error.
Reaching for the wrong energy formula — using $U=\tfrac12QV$ when a problem gives $C$ and $V$ but not $Q$ ; always match the formula to the two quantities you actually have, or convert with $Q=CV$ first.
Treating capacitance as if it changes with the amount of charge stored — it does not; $C$ is fixed by geometry, and only $Q$ and $V$ move together, keeping their ratio constant.
Attempting a network reduction in the wrong order — trying to combine two capacitors that are neither purely series nor purely parallel with each other before the sub-groups around them have been collapsed first.

Common trapSeries and parallel rules for capacitors are the reverse of the resistor rules — do not carry the resistor pattern over

The single costliest habit in this topic is muscle memory from Current Electricity. For resistors, series resistances add directly ( $R_s=R_1+R_2+\cdots$ ) and parallel resistances add as reciprocals. For capacitors it is exactly backwards: series capacitances add as reciprocals ( $1/C_s=1/C_1+1/C_2+\cdots$ ) and parallel capacitances add directly ( $C_p=C_1+C_2+\cdots$ ). The physical reason is that resistors in series share the same current while capacitors in series share the same charge but split the voltage, and it is that voltage-splitting that forces the reciprocal rule. If you find yourself writing $C_s=C_1+C_2$ for a series pair, stop and re-derive from $V=V_1+V_2$ with $Q$ common — do not trust the pattern from resistors.

You should be able to…

Define capacitance as $C=Q/V$ , state its SI unit (the farad), and use the microfarad/picofarad sub-units correctly.
Derive and apply the series formula $1/C_s=\sum 1/C_i$ and the parallel formula $C_p=\sum C_i$ , explaining in words why one shrinks and the other grows relative to the individual capacitors.
Reduce a mixed series-parallel network to a single equivalent capacitance by collapsing sub-groups step by step, in the correct order.
Find the charge on, and voltage across, any individual capacitor inside a series or parallel network once the equivalent capacitance and total applied voltage are known.
Apply $U=\tfrac12CV^2=\tfrac12QV=Q^2/2C$ correctly, selecting the form that matches the two known quantities in a given problem.

Ask the tutor: want a mixed series-parallel network with three or four capacitors to practise the step-by-step reduction, or curious how these ideas plug into the parallel-plate capacitor's own $C=\varepsilon_0A/d$ formula next?

Why do capacitors add oppositely to resistors when they combine in series and parallel?

It comes down to what stays the same in each connection. In series, the same charge sits on every capacitor (the connecting wires only ever get charged by induction, so no net charge is left behind between capacitors), but the voltage splits across them — and since voltage is in the denominator of $C=Q/V$ , splitting voltage while holding charge fixed forces the reciprocals to add. In parallel, it is voltage that stays common and charge that splits, so the capacitances themselves add directly. Resistors follow the opposite logic because $R=V/I$ has voltage on top, not charge.

If capacitance never depends on Q or V, why do exam questions always give me a charge or a voltage to work with?

Those numbers describe the state the capacitor happens to be in at that moment, not the capacitor itself. $C$ is fixed by the plates' geometry and the medium between them the moment the device is manufactured; $Q$ and $V$ are whatever charging condition the problem describes, and they can be anything, always in the fixed ratio $C=Q/V$ . Giving you $Q$ and $V$ (or $C$ and one of the two) is simply enough information to pin down the third, not evidence that $C$ itself is changing.

Why is the energy stored in a parallel combination bigger than in a series combination of the same two capacitors, at the same battery voltage?

Because the equivalent capacitance itself is bigger. Parallel banks always give a larger $C_p$ than any individual capacitor (more plates side by side means more room to store charge at the same voltage), while series chains always give a smaller $C_s$ . Since $U=\tfrac12CV^2$ and $V$ is fixed by the same battery in both cases, the larger equivalent capacitance directly means more stored energy — exactly the roughly 4.5 times difference seen in the competitive worked example above.

Do I need to know where the energy formula comes from for my board exam?

No — the CBSE 2025-26 syllabus explicitly lists the energy-stored formula as "formula only, no derivation," so you are expected to quote and correctly apply $U=\tfrac12CV^2=\tfrac12QV=Q^2/2C$ , not derive it from first principles (which would need integrating the work done as charge builds up). Spend your revision time on applying the right form of the formula and combining capacitor networks correctly instead.