Capacitor combinations

Almost no real circuit uses just one capacitor. A camera flash, a smoothing filter, or a tuning circuit wires several capacitors together — in series, in parallel, or in mixed networks — to get exactly the total capacitance a design needs. Learning to collapse any such network into one equivalent capacitance is one of the most exam-tested skills in ISC Class 12 electrostatics, because it turns a tangle of wires into a single number you can plug back into C=Q/VC=Q/V. This section sits right after potential energy of a charge and just before energy stored in a capacitor, and it leans directly on the definitions of capacitance C=Q/VC=Q/V and the parallel-plate formulas C=ε0A/dC=\varepsilon_0A/d (air) and C′=Kε0A/dC'=K\varepsilon_0A/d (with a dielectric). Two combination rules — one for series, one for parallel — cover every network you will meet, however tangled it looks, once you learn to spot the innermost series or parallel group and reduce it first.

The core insight

Capacitors in series share the same charge and add their capacitances as reciprocals (so CeqC_{\text{eq}} falls below the smallest one), while capacitors in parallel share the same voltage and add their capacitances directly (so CeqC_{\text{eq}} exceeds the largest one) — exactly the opposite of how resistors combine.

The core idea

Capacitance recap. A capacitor's capacitance is C=Q/VC=Q/V, the charge stored per unit potential difference, measured in farads (1 F=1 C/V1\ \text{F}=1\ \text{C/V}). For a parallel-plate capacitor of plate area AA and separation dd, C=ε0A/dC=\varepsilon_0A/d in air and C′=Kε0A/dC'=K\varepsilon_0A/d with a dielectric of dielectric constant KK filling the gap. Everything below is about combining several such capacitors into one equivalent device.

Series combination. When capacitors are joined end to end, with no branch point between them, exactly the same charge QQ flows onto every capacitor in the chain (charge cannot appear or vanish at an isolated junction). The applied voltage splits across the capacitors, V=V1+V2+V3+⋯V=V_1+V_2+V_3+\cdots, and since each Vi=Q/CiV_i=Q/C_i uses the same QQ,

1Ceq=1C1+1C2+1C3+⋯\frac{1}{C_{\text{eq}}} = \frac{1}{C_1}+\frac{1}{C_2}+\frac{1}{C_3}+\cdots

where: CeqC_{\text{eq}} is the single capacitance that replaces the whole series chain, and C1,C2,C3,…C_1,C_2,C_3,\ldots are the individual capacitances. Reciprocals adding means CeqC_{\text{eq}} always comes out smaller than the smallest capacitor in the chain. For exactly two capacitors in series there is a fast shortcut, Ceq=C1C2C1+C2C_{\text{eq}}=\dfrac{C_1C_2}{C_1+C_2} — useful, but valid only for two capacitors at a time.

Parallel combination. When capacitors are joined so all their positive plates share one node and all their negative plates share another, every capacitor feels the same voltage VV. The total charge drawn is the sum of the individual charges, Q=Q1+Q2+Q3+⋯Q=Q_1+Q_2+Q_3+\cdots, and since each Qi=CiVQ_i=C_iV uses the same VV,

Ceq=C1+C2+C3+⋯C_{\text{eq}} = C_1+C_2+C_3+\cdots

where: the symbols mean the same as above. Capacitances simply add, so the parallel equivalent is always larger than the largest individual capacitor. Series and parallel rules for capacitors are mirror images of the rules for resistors: capacitors in series behave like resistors in parallel (reciprocals add), and capacitors in parallel behave like resistors in series (values add directly).

Mixed networks. A network combining both patterns is reduced from the inside out: find the smallest recognisable series or parallel sub-group, replace it with its single equivalent capacitance, redraw the (now simpler) circuit, and repeat until only one equivalent capacitance remains between the two terminals.

Figure
Two capacitor networks side by side, one showing three capacitors chained in series carrying equal charge Q and one showing three capacitors joined in parallel sharing equal voltage V
Series: same charge Q on each C1 C2 C3 Q flows onto every capacitor equally V splits: V = V1 + V2 + V3 Parallel: same voltage V on each C1 C2 C3 V same across all Q splits: Q = Q1 + Q2 + Q3
Derivation AEquivalent capacitance of capacitors in series

For nn capacitors connected end to end across a battery of voltage VV, with no branch point between successive plates so that the induced charge on every plate pair is forced to be the same value QQ, show that the reciprocal of the single equivalent capacitance CeqC_{\text{eq}} equals the sum of the reciprocals of the individual capacitances, 1/Ceq=1/C1+1/C2+⋯+1/Cn1/C_{\text{eq}}=1/C_1+1/C_2+\cdots+1/C_n, and confirm that this makes CeqC_{\text{eq}} smaller than every individual CiC_i.

Step 1. Consider capacitors C1,C2,…,CnC_1, C_2, \ldots, C_n joined end to end between the terminals of a battery of voltage VV. Because there is no branch point between one capacitor's negative plate and the next one's positive plate, that isolated conductor can only redistribute the charge it already has — it cannot create or lose net charge. So charging the outer plates with +Q+Q and −Q-Q induces exactly ±Q\pm Q on every intermediate plate pair too: the charge QQ is common to all nn capacitors.

Step 2. Each capacitor therefore carries its own voltage Vi=Q/CiV_i=Q/C_i, and these voltages must add up to the total applied voltage, since VV is the sum of the potential drops across each capacitor in the chain: V=V1+V2+⋯+Vn=QC1+QC2+⋯+QCnV=V_1+V_2+\cdots+V_n=\dfrac{Q}{C_1}+\dfrac{Q}{C_2}+\cdots+\dfrac{Q}{C_n}.

Step 3. The equivalent capacitance CeqC_{\text{eq}} is defined by the same relation the whole chain obeys from the outside, V=Q/CeqV=Q/C_{\text{eq}}, using the same common charge QQ. Substituting this into Step 2 and dividing through by QQ (which is nonzero and common to every term) gives 1Ceq=1C1+1C2+⋯+1Cn\dfrac{1}{C_{\text{eq}}}=\dfrac{1}{C_1}+\dfrac{1}{C_2}+\cdots+\dfrac{1}{C_n}, the series formula.

Step 4. Because every term on the right is positive, 1/Ceq1/C_{\text{eq}} is greater than any single 1/Ci1/C_i, so CeqC_{\text{eq}} is smaller than every individual CiC_i — adding a capacitor in series can only reduce the net capacitance, never increase it. For the special case n=2n=2, clearing denominators in 1/Ceq=1/C1+1/C21/C_{\text{eq}}=1/C_1+1/C_2 gives the shortcut Ceq=C1C2/(C1+C2)C_{\text{eq}}=C_1C_2/(C_1+C_2).

Worked example

Three capacitors of 6 μF6\ \mu\text{F}, 3 μF3\ \mu\text{F}, and 2 μF2\ \mu\text{F} are connected in series across a battery of 22 V22\ \text{V}. Find (a) the equivalent capacitance, (b) the charge on each capacitor, and (c) the potential difference across the 2 μF2\ \mu\text{F} capacitor.

Step 1. Apply the series formula to combine all three capacitances:

1Ceq=16+13+12=16+26+36=66=1 (μF−1)\frac{1}{C_{\text{eq}}} = \frac{1}{6}+\frac{1}{3}+\frac{1}{2} = \frac{1}{6}+\frac{2}{6}+\frac{3}{6} = \frac{6}{6} = 1\ (\mu\text{F}^{-1})

so Ceq=1 μFC_{\text{eq}} = 1\ \mu\text{F}.

Step 2. In series the charge is the same on every capacitor and equals the charge that would sit on the equivalent capacitor at the full applied voltage:

Q=CeqV=(1×10−6)×22=22×10−6 C=22 μCQ = C_{\text{eq}}V = (1\times10^{-6})\times22 = 22\times10^{-6}\ \text{C} = 22\ \mu\text{C}

So each of the three capacitors carries Q=22 μCQ=22\ \mu\text{C}.

Step 3. The voltage across the 2 μF2\ \mu\text{F} capacitor uses the same common charge:

V3=QC3=22×10−62×10−6=11 VV_3 = \frac{Q}{C_3} = \frac{22\times10^{-6}}{2\times10^{-6}} = 11\ \text{V}

The equivalent capacitance is 1 μF1\ \mu\text{F}, each capacitor carries 22 μC22\ \mu\text{C}, and the 2 μF2\ \mu\text{F} capacitor has 11 V11\ \text{V} across it — exactly half the 22 V22\ \text{V} supply, because it is the smallest capacitance and therefore takes the largest voltage share.

Competitive stretch

Take the same 4 μF4\ \mu\text{F} and 12 μF12\ \mu\text{F} pair from the parallel/series comparison below. If instead they were connected in parallel first and you wanted the same 160 μC160\ \mu\text{C} total charge from a series connection instead, what voltage would be needed? Since series Ceq=3 μFC_{\text{eq}}=3\ \mu\text{F} (worked out below) and charge is common in series, V=Q/Ceq=160 μC/3 μF≈53.3 VV=Q/C_{\text{eq}}=160\ \mu\text{C}/3\ \mu\text{F}\approx53.3\ \text{V} — over five times the original 10 V10\ \text{V}, showing how much a series connection "throttles" achievable charge for a given voltage compared to parallel.

Common traps

  • Swapping the series and parallel rules: capacitors in series combine like resistors in parallel (reciprocals add), and capacitors in parallel combine like resistors in series (values add directly). This reversal is the single most common ISC exam slip.
  • Assuming charge is equal on all capacitors in a parallel group — it is the voltage that is equal in parallel; equal charge is a series property.
  • Missing that in series, the smaller capacitance grabs the larger share of voltage (since Vi=Q/CiV_i=Q/C_i with QQ fixed), which feels backwards at first.
  • Using the two-capacitor shortcut Ceq=C1C2/(C1+C2)C_{\text{eq}}=C_1C_2/(C_1+C_2) on three or more series capacitors — it only holds for exactly two; for more, sum the reciprocals directly.

Ask the tutor: Stuck reducing a mixed series-parallel capacitor network, or want to check your own voltage-divider-style capacitor problem step by step?

Why is charge the same for series capacitors but voltage the same for parallel ones?

It comes down to what is physically shared. In a series chain, the plates between adjacent capacitors form isolated conductors with no other connection, so any charge appearing on one side must be matched by induced charge of the same size next to it — the charge is forced to be identical all the way along. In parallel, every capacitor's plates connect directly to the same two circuit nodes, so every capacitor simply sees the same node-to-node voltage; nothing forces the charges to match, and they differ according to each Ci=Qi/VC_i=Q_i/V.

Can I use the two-capacitor series shortcut for three capacitors by applying it twice?

Yes. Combine any two of the three using Ceq=C1C2/(C1+C2)C_{\text{eq}}=C_1C_2/(C_1+C_2) to get an intermediate equivalent, then combine that result with the third capacitor using the same shortcut again. This gives the identical answer as adding all three reciprocals directly — it is just a longer route to the same number, useful if you are more comfortable with the product-over-sum form than with fractions.

In a mixed series-parallel network, does it matter which group I simplify first?

No, as long as each group you simplify is a genuine series or parallel sub-group (a set of capacitors sharing only the two nodes between which you are combining them, with no other branch tapping into the middle). Reducing the innermost, most obviously recognisable group first usually keeps the redrawn circuit simplest to read, but the final equivalent capacitance between the two terminals comes out the same regardless of the order you eliminate groups in.

Before this

Where this leads