AP Physics C: E&M · Unit 11 lecture · Part 10 of 11 · 10 min

11.8 RC circuits: discharging, networks and dataDischarging, bigger circuits, and measuring τ

Mr. Thompkins
You've done the hard derivation. Discharging is the same idea run backward, and the two-limits trick handles capacitors buried in bigger circuits. Then we'll do the lab analysis the exam asks for. You've got this.

Disconnect the battery and let a charged capacitor push current through a resistor. Walking around the loop: $\frac{q}{C} - IR = 0$. Now the capacitor's charge is draining, so the current is $I = -\frac{dq}{dt}$. That gives $\frac{dq}{dt} = -\frac{q}{RC}$, whose solution is $q = q_0e^{-t/RC}$. Voltage and current follow the same curve: $V_C = V_0e^{-t/RC}$ and $I = \frac{V_0}{R}e^{-t/RC}$.

?Wait, why does dq/dt = −q/RC give an exponential?
Separate: $\frac{dq}{q} = -\frac{dt}{RC}$. Integrate from $q_0$ at t = 0: $\ln\frac{q}{q_0} = -\frac{t}{RC}$, so $q = q_0e^{-t/RC}$. Any quantity whose rate of change is proportional to itself, with a minus sign, decays exponentially.
04812162002468101214t (s)$V_C$ (V)half: 3.3 s$t = RC$: 4.4 V
A 470 μF capacitor discharging from 12 V through 10 kΩ (RC = 4.7 s). Half the charge is gone at RC ln 2 = 3.3 s; at t = RC it is down to 37%, 4.4 V.
A discharge
A 470 μF capacitor charged to 12 V is connected across a 10 kΩ resistor at t = 0. Find (a) the time constant and the initial current, (b) the time for the charge to fall to half, (c) the time for the stored energy to fall to half, and (d) the total heat produced in the resistor.
  1. (a) $\tau = RC = (10 \times 10^3)(470 \times 10^{-6}) = 4.7$ s. $I_0 = \frac{V_0}{R} = \frac{12}{10 \times 10^3} = 1.2$ mA.
  2. (b) $\frac{1}{2} = e^{-t/RC}$, so $t = RC\ln 2 = (4.7)(0.693) = 3.3$ s.
  3. (c) Energy goes as $q^2$: $U = U_0e^{-2t/RC}$. Setting that to half: $t = \frac{RC\ln 2}{2} = 1.6$ s.
    Why?
    The energy halves sooner than the charge, because when the charge is at 71% the energy (charge squared) is already at 50%.
  4. (d) All the stored energy ends up as heat: $U_0 = \frac{1}{2}CV_0^2 = \frac{1}{2}(470 \times 10^{-6})(12)^2 = 0.034$ J.
    Why?
    You could also integrate $\int_0^\infty I^2R\,dt = \frac{V_0^2}{R}\cdot\frac{RC}{2} = \frac{1}{2}CV_0^2$. Same answer.

Answer: (a) 4.7 s and 1.2 mA (b) 3.3 s (c) 1.6 s (d) 0.034 J.

Stuck? Get a hint
  1. τ is R times C, in ohms and farads.
  2. For half the charge, solve $e^{-t/RC} = 0.5$ with a natural log.
  3. Energy is proportional to q², so it decays as $e^{-2t/RC}$.

Capacitors inside bigger circuits

Often the capacitor sits in one branch of a network. Use the two limits. Just after a switch closes, replace each empty capacitor with a wire and solve the resistor circuit. Long after, remove each capacitor (a break) and solve again; the capacitor's voltage then equals the voltage across whatever it's in parallel with. The time constant uses the resistance the capacitor "sees": look out from its two terminals with each ideal battery replaced by a wire.

18 VS$R_1 = 3.0$ kΩ$R_2 = 6.0$ kΩ20 μFX
Close S at t = 0 with the capacitor empty. The capacitor is in parallel with R₂, so it always has R₂'s voltage.
The two limits
In the circuit shown, the capacitor is empty when switch S closes at t = 0. Find (a) the current through $R_1$ just after closing, (b) the current through $R_1$ long after, (c) the final charge on the capacitor, and (d) the time constant.
  1. (a) Just after: the empty capacitor acts like a wire, which shorts $R_2$. Only $R_1$ limits the current: $I = \frac{18}{3.0 \times 10^3} = 6.0$ mA, and all of it flows into the capacitor.
    Why?
    The short is the key insight: a wire in parallel with $R_2$ means $R_2$ has no voltage and no current at that instant.
  2. (b) Long after: the capacitor is a break. $R_1$ and $R_2$ are in series: $I = \frac{18}{9.0 \times 10^3} = 2.0$ mA.
  3. (c) $V_C = V_{R_2} = (2.0\ \text{mA})(6.0\ \text{k}\Omega) = 12$ V, so $Q = CV = (20\ \mu\text{F})(12\ \text{V}) = 240\ \mu$C.
  4. (d) Looking out from C with the battery as a wire, $R_1$ and $R_2$ are in parallel: $\frac{(3.0)(6.0)}{9.0} = 2.0$ kΩ. So $\tau = (2.0 \times 10^3)(20 \times 10^{-6}) = 0.040$ s.
    Why?
    Not $R_1C$ and not $(R_1 + R_2)C$. The capacitor charges through both resistors at once.

Answer: (a) 6.0 mA (b) 2.0 mA (c) 240 μC at 12 V (d) 40 ms.

Stuck? Get a hint
  1. Just after: replace the capacitor with a wire. What happens to R₂?
  2. Long after: remove the capacitor. Now R₁ and R₂ are simply in series.
  3. The capacitor's final voltage equals R₂'s voltage, because they're in parallel.
Pause and think
Long after, someone opens S, cutting off the battery. Through which resistance does the capacitor discharge, and what is the new time constant?
I thought about it. Show me
Only through $R_2$: with S open, $R_1$ leads to a dead end. So $\tau = R_2C = (6.0 \times 10^3)(20 \times 10^{-6}) = 0.12$ s, three times longer than the charging τ. A discharge path can differ from the charging path.
ElementJust after a switchLong afterNever jumps
Resistor$\Delta V = IR$ at every instant$\Delta V = IR$Nothing: it reacts at once
Empty capacitorActs like a wire (V = 0)Acts like a break (I = 0)Its charge and voltage $q/C$
Charged capacitorKeeps its voltageDischarged, if it has a pathIts charge and voltage $q/C$
The rules for the instant after a switch and for a long time later.

Capacitors also combine, the opposite way from resistors. In parallel they share ΔV and their charges add, so $C_p = C_1 + C_2 + \dots$. In series they carry the same charge (the inner plates form an isolated conductor) and their voltages add, so $\frac{1}{C_s} = \frac{1}{C_1} + \frac{1}{C_2} + \dots$. For example, 6.0 μF and 3.0 μF in series make 2.0 μF. Across 12 V each holds 24 μC, so the 6.0 μF has 4.0 V and the 3.0 μF has 8.0 V: the smaller capacitor takes the bigger share of the voltage.

Measuring a time constant

On an Experimental Design and Analysis question, you'll be handed decay data and asked for R or C. The move is to linearize. Take the natural log of $V_C = V_0e^{-t/RC}$: $\ln V_C = \ln V_0 - \frac{1}{RC}t$. That's a straight line in t with slope $-\frac{1}{RC}$ and intercept $\ln V_0$. A student discharges a capacitor through 100 kΩ and reads 10.0, 4.8, 2.3, 1.1, 0.53 and 0.25 V at t = 0, 5, 10, 15, 20 and 25 s.

?Wait, why does taking ln straighten an exponential?
The natural log undoes $e^x$: $\ln(e^x) = x$, and $\ln(ab) = \ln a + \ln b$. So $\ln(V_0e^{-t/RC}) = \ln V_0 - \frac{t}{RC}$, which is linear in t. A straight ln V graph is how you show the data really are exponential.
0510152025-1.6-0.800.81.62.43.2t (s)$\ln(V_C/1\ \text{V})$slope $-0.147\ \text{s}^{-1}$
ln V against t falls on a straight line, which confirms the exponential model. The best-fit slope is −0.147 per second.
Analyze the data
Use the data above (R = 100 kΩ) to find the capacitance.
  1. Take logs: $\ln V$ = 2.303, 1.569, 0.833, 0.095, −0.635, −1.386.
    Why?
    Plot $\ln V_C$ on the vertical axis and t on the horizontal axis, with labeled axes and units. That's a rubric point.
  2. Draw a best-fit line and read its slope from two points on the line (not data points): from (0, 2.31) to (25 s, −1.38), slope $= \frac{-1.38 - 2.31}{25} = -0.147\ \text{s}^{-1}$.
  3. Slope $= -\frac{1}{RC}$, so $RC = \frac{1}{0.147} = 6.8$ s and $C = \frac{6.8}{100 \times 10^3} = 6.8 \times 10^{-5}$ F.

Answer: $C \approx 68\ \mu\text{F}$. The intercept, $e^{2.31} \approx 10$ V, matches the starting voltage.

Check yourself
A capacitor discharges through a resistor. A graph of $\ln V_C$ against t has slope $-0.25\ \text{s}^{-1}$. If R = 40 kΩ, what is C?
  1. A100 μF
  2. B10 μF
  3. C160 μF
  4. D0.25 μF
Show the answer

A. $\frac{1}{RC} = 0.25\ \text{s}^{-1}$, so $RC = 4.0$ s and $C = \frac{4.0}{40 \times 10^3} = 1.0 \times 10^{-4}$ F = 100 μF. The trap is 10 μF, a slip of a factor of 10 in the kilo-ohm conversion.

Exam tip from Mr. T

When a graph of $V_C$ or I against t is given, find τ from it first: the time to fall to 37%, the time to rise to 63%, or the half-time divided by 0.69. Mark the point you read on the graph. Then use τ = RC.

  • Discharging: $q = q_0e^{-t/RC}$, with $I = -\frac{dq}{dt}$. Charge halves every $0.69RC$; energy halves every $0.35RC$.
  • Networks: just after, empty capacitors are wires; long after, they are breaks with the voltage of their parallel partner.
  • τ uses the resistance the capacitor sees, with batteries replaced by wires. Discharge paths can differ.
  • Capacitors: parallel add, $\frac{1}{C_s}$ adds in series. Linearize decay data with $\ln V$ against t: slope $-1/RC$.