Topic 8 of 9
RC charging and discharging
Trace the switch connection first. Then keep one sign convention as the capacitor charges or discharges, so the current graph agrees with the change in its plate charge.
Use a resistor R between switch terminal S and capacitor plate B. The other plate connects to common return node C. The source positive terminal is A, at Es above C. Here the letter C labels a node in the drawing; the italic quantity C in an equation is capacitance.
Charge: connect the resistor to the source
The current reference stays S to B in both experiments. B/C names the capacitor terminals; the value 10 microfarad is its capacitance. Each experiment starts its own clock.
Discharge: isolate the positive source terminal
The current reference stays S to B in both experiments. B/C names the capacitor terminals; the value 10 microfarad is its capacitance. Each experiment starts its own clock.
Define Q as the charge on plate B and VC = VB - Vnode C = Q/C. Define positive I through the resistor from S towards B. Then I = dQ/dt: positive current increases Q and negative current reduces it. In the examples Q stays positive while falling during discharge.
A switch that simply disconnects every conducting path does not discharge an ideal capacitor. The discharge state shown provides a closed resistor-capacitor loop without connecting the source across a short circuit.
Charging from zero
Take an initially uncharged capacitor, a constant ideal source e.m.f. Es, fixed R and C, and negligible leakage and meter loading. At each instant, Es = IR + Q/C. Initially the capacitor p.d. is zero, so current is Es/R. As Q and its p.d. grow, less voltage remains across R and current decreases.
Q = CEs(1 - e-t/τ)
VC = Es(1 - e-t/τ)
I = (Es/R)e-t/τ
Charge and capacitor p.d. rise towards their final values; charging current decreases towards zero. The source voltage stays constant in this model. A finite resistor prevents capacitor voltage from jumping instantaneously, though the current can change abruptly when the switch changes a connection.
Discharging from a specified initial voltage
Disconnect the source and connect R across a capacitor initially at V0, with Q0 = CV0. The stored charge now drives current through R. Keep the same positive current reference:
VC = V0e-t/τ
I = -(V0/R)e-t/τ
The current is negative and approaches zero from below. Its magnitude decreases. Calling it negative does not mean that the capacitor has reversed plate polarity. If a question chooses the opposite current reference, reverse the current sign while keeping the physical behaviour unchanged.
In the general rising form x = x0(1 - e-t/τ), x0 is the final limit when the initial value is zero. In the decreasing form x = x0e-t/τ, x0 is the initial value. Check which quantity the symbol describes.
Compare matched charge and discharge models
Let Es = V0 = 10 V, R = 200 kΩ and C = 10 µF. These are two separate runs, each with its own t = 0: charging starts uncharged, while discharging starts fully at 10 V.
CEs = CV0 = 100 µC
Es/R = V0/R = 50 µA
Charge on the B-side plate
Solid: charging. Dashed: discharging. These are separate experiments with separate time zeros; neither exponential reaches its final limit at a finite time.
Capacitor p.d. between B and C
Solid: charging. Dashed: discharging. These are separate experiments with separate time zeros; neither exponential reaches its final limit at a finite time.
Signed current with one reference direction
Solid: charging, I > 0. Dashed: discharging, I < 0. Both use the same S-to-B reference; these are separate experiments.
| t / s | VC / V | Q / µC | I / µA |
|---|---|---|---|
| 0 | 0 | 0 | +50 |
| 2.0 | 6.321 | 63.21 | +18.394 |
| 4.0 | 8.647 | 86.47 | +6.767 |
| t / s | VC / V | Q / µC | I / µA |
|---|---|---|---|
| 0 | 10 | 100 | -50 |
| 2.0 | 3.679 | 36.79 | -18.394 |
| 4.0 | 1.353 | 13.53 | -6.767 |
Extra digits show the model comparison; final experimental precision would depend on the actual component values and readings. Use micro = 10-6 consistently for both µF and µA.
What the time constant does and does not mean
At t = τ, a decaying quantity is e-1 = 0.368 of its initial value, while a rising quantity is 1 - e-1 = 0.632 of its final value. RC has units of time:
The time constant is not a half-life and is not the time for mathematically complete charging. At 5τ, charging reaches about 99.33% of its final value. To find a different fraction, solve the exponential:
t = τ ln 2 = 1.386 s
Charge to 90%: 0.10 = e-t/τ
t = -τ ln(0.10) = 4.605 s
Doubling R alone makes τ = 4.0 s and halves the initial charging current to 25 µA, while final charge stays 100 µC. Doubling C alone also makes τ = 4.0 s, but now final charge is 200 µC and initial current remains 50 µA. An equal change in time constant does not imply identical changes in every quantity.
Energy and the actual resistor path
For charging from zero in this ideal constant-voltage model, the source supplies energy EsQfinal = CEs2. The capacitor stores ½CEs2; the rest is transferred to internal energy in the resistance. Here the source supplies 1.00 mJ, the capacitor stores 0.500 mJ and the resistor receives 0.500 mJ. In a complete discharge through R, the stored energy is transferred through that resistance.
Choose R from the path present in the relevant state. Source internal resistance can add to a charging path while being absent from the separate discharge loop. A finite-resistance voltmeter adds a parallel path; leakage can also matter. An arbitrary product of a whole network's named resistance and capacitance is not automatically its time constant.