Logo

MonoCalc

/

RC Time Constant Calculator

Physics

The circuit

Every panel below reads this one R–C pair, so the time constant, the transient, the energy figures and the filter corner always describe the same circuit.

The capacitor climbs towards the supply, closing 63.2 % of the remaining gap every time constant. It never quite arrives, which is why 5τ is treated as done.

Leave blank for an empty capacitor when charging, or the starting voltage when discharging.
The classic switch-debounce network. τ = 47 ms, and 30 ms after the switch closes the capacitor sits at 4.246 V with 1.011 mA still flowing.
Between 1τ and 10τ.
Time constant τ
47.000 ms
Voltage at t
4.246 V
Current at t
1.011 mA
Charged
47.181 %

One time constant reaches 63.21 %; the tick most people quote sits just under two-thirds of the way along.

Half-life t½ = τ·ln2
32.578 ms
10–90 % rise time
103.270 ms
Settling time 5τ
235.000 ms
Cutoff frequency f_c
3.386 Hz
Peak current at t = 0
1.915 mA
Peak resistor power
17.234 mW
Stored charge Q
42.463 µC
Stored energy E
90.155 µJ

Charging curve

The solid trace is the capacitor voltage and the dashed one the resistor current, which decays on exactly the same exponential. Drag the cursor to read any point.

1τ63.2%2τ86.5%3τ95.0%4τ98.2%5τ99.3%V0time →
Cursor at
47.000 ms
Voltage there
5.689 V
Current there
704.450 µA
Charged
63.212 %

Multiples of τ

Every percentage here comes from 1 − e^(−n) evaluated at run time, not from a table of remembered constants.

ElapsedTimeCharging (% of V₀)Discharging (% of V₀)Charging voltageDischarging voltage
1τ47.000 ms63.212 %36.788 %5.689 V3.311 V
94.000 ms86.466 %13.534 %7.782 V1.218 V
141.000 ms95.021 %4.979 %8.552 V448.084 mV
188.000 ms98.168 %1.832 %8.835 V164.841 mV
235.000 ms99.326 %0.674 %8.939 V60.642 mV

Step by step

Time constant

τ = R · C = 4.700 kΩ × 10.000 µF = 4700.000 Ω × 1.0000e-5 F = 47.000 ms

Exponent

t / τ = 30.000 ms ÷ 47.000 ms = 0.638

Charging voltage

V(t) = 9.000 V + (0.000 V − 9.000 V)·e^(−0.638) = 4.246 V

Current

I(t) = (V_f − V_i)/R · e^(−t/τ) = 1.011 mA

Charge and energy

Q = C·V(t) = 42.463 µC; E = ½·C·V(t)² = 90.155 µJ

Milestones

t½ = τ·ln2 = 32.578 ms; t(10→90 %) = τ·ln9 = 103.270 ms; 5τ = 235.000 ms

Cutoff frequency

f_c = 1/(2πτ) = 3.386 Hz

τ = 47.000 ms; after 30.000 ms the capacitor sits at 4.246 V with 1.011 mA still flowing.

τ = 47.000 ms in the export units above — the TXT and CSV downloads carry the figures at this precision rather than raw float tails.

About This Tool

RC Time Constant – Charging, Discharging and the Cutoff Frequency

Put a resistor in series with a capacitor and you have built the most useful circuit in electronics. Its behaviour is governed by a single number, the time constant τ = R · C, which with R in ohms and C in farads comes out directly in seconds. A 10 kΩ resistor with a 100 µF capacitor gives τ = 1 s; a 4.7 kΩ resistor with 10 µF gives 47 ms. This RC time constant calculator evaluates that product and then solves the full first-order response around it — voltage, current, elapsed time, charge, energy and the filter corner — rearranging the exponential for whichever quantity you leave unknown.

Where 63.2 % comes from

The capacitor voltage follows V(t) = V_f + (V_i − V_f)·e^(−t/τ), which for the usual case of an empty capacitor charging towards a supply collapses to V(t) = V₀·(1 − e^(−t/τ)), and for a discharge to V(t) = V₀·e^(−t/τ). Substituting t = τ makes the exponent exactly −1, so the exponential becomes 1/e ≈ 0.368. That is the whole story behind the famous figure: one time constant leaves 36.8 % of the journey outstanding, so 63.2 % of it is done. The number belongs to the exponential function, not to any particular resistor, which is why every RC network on earth shares it.

The same reasoning gives the rest of the settling table, each row computed from 1 − e^(−n) rather than looked up: 86.47 % at 2τ, 95.02 % at 3τ, 98.17 % at 4τ and 99.33 % at 5τ. Because the curve is asymptotic the capacitor never actually arrives, so 5τ is a convention, not a fact — it simply leaves an error of 0.67 %, smaller than the tolerance of the parts themselves.

Current runs the other way

While the voltage rises, the current falls: I(t) = (V_f − V_i)/R · e^(−t/τ). The instant the switch closes the capacitor is still empty, so the full supply appears across the resistor and the inrush current is simply I₀ = V₀/R — the largest current the circuit will ever see. As the capacitor fills, less voltage is left for the resistor and the current decays on exactly the same exponential as the voltage rises. Sizing the series resistor is therefore as much about surviving t = 0 as about hitting a timing target.

Worked example

A 4.7 kΩ resistor charging a 10 µF capacitor from 9 V gives τ = 47 ms. After 30 ms the exponent is 30/47 = 0.6383, so V = 9 × (1 − e^(−0.6383)) = 4.246 V and I = (9/4700) × e^(−0.6383) = 1.011 mA. The capacitor holds 42.46 µC and 90.15 µJ at that moment, having covered 47.18 % of its journey.

Reading the same circuit as a filter

Swap to the frequency domain and the identical R and C become a first-order filter with its corner at f_c = 1/(2πRC) = 1/(2πτ). Take the output across the capacitor and you have a low-pass; take it across the resistor and you have a high-pass. A 1 kΩ resistor with 100 nF corners at 1.592 kHz, where the output is 1/√2 of the input — the −3.01 dB that names the point — with 45° of phase shift. Beyond the corner a single pole rolls off at 20 dB per decade, but the asymptote is only accurate well away from the corner, so the calculator uses the exact magnitude 1/√(1 + (f/f_c)²) instead. At 5 kHz, about 3.142 times the corner, that gives −10.362 dB and −72.343°.

Charge, energy and the missing half

A charged capacitor holds Q = C·V and E = ½·C·V². A 1000 µF capacitor at 24 V therefore carries 24 mC and 288 mJ, and through a 10 Ω series resistor it draws a 2.4 A inrush. There is a well-known surprise hiding here: the supply had to push Q coulombs through V volts, delivering CV² joules, yet the capacitor only keeps half of it. The resistor dissipates the other half — always exactly half, whatever the resistance. A small resistor burns it quickly and a large one slowly, but no series resistor beats 50 % efficiency.

Tolerance dominates real timing

τ = RC rises with both factors, so the worst cases are the all-low and all-high corners. A 10 kΩ ±5 % resistor with a 100 µF ±20 % electrolytic gives a nominal 1 s that can legitimately land anywhere between 0.76 s and 1.26 s. Never trust a bare RC network for precision timing — use a crystal, or design for the band.

Designing to a target

Most real work runs the formula backwards. Given a required delay, R = τ/C and C = τ/R size the missing part: a 2.5 s time constant with a 47 µF capacitor calls for 53.19 kΩ, which is not a value anyone stocks. The nearest E24 part is 51 kΩ, delivering 2.397 s — a −4.12 %error, comfortably inside the capacitor’s own tolerance. To hit a specific voltage rather than a specific τ, invert the exponential: t = −τ·ln(1 − V_target/V₀) for charging and t = −τ·ln(V_target/V₀) for discharging, so reaching 90 % of a 5 V rail through 220 kΩ and 1 µF takes 0.507 s. Useful shortcuts fall out of the same logarithm: the half-life is τ·ln2 ≈ 0.693τ and the 10–90 % rise time that oscilloscopes measure is τ·ln9 ≈ 2.197τ.

Frequently Asked Questions

Is the RC Time Constant Calculator free?

Yes, RC Time Constant Calculator is totally free :)

Can I use the RC Time Constant Calculator offline?

Yes, you can install the webapp as PWA.

Is it safe to use RC Time Constant Calculator?

Yes, any data related to RC Time Constant Calculator only stored in your browser (if storage required). You can simply clear browser cache to clear all the stored data. We do not store any data on server.

How does this RC time constant calculator work?

Enter a resistance and a capacitance in whatever units the parts are marked in — Ω through GΩ and pF through F can be mixed freely — and the calculator normalises both to ohms and farads before multiplying them into τ = R × C. That single unrounded time constant then drives everything else: the charging and discharging exponentials, the instantaneous current, the stored charge and energy, the settling milestones and the filter cutoff. Nothing is rebuilt from a rounded intermediate, which matters because an exponent rounded even at the third decimal visibly shifts the voltage it produces.

Why is 63.2 % the magic number for one time constant?

Because 1 − e⁻¹ = 0.632. After one time constant the exponential e^(−t/τ) has fallen to 1/e ≈ 36.8 % of its starting value, so a charging capacitor has closed 63.2 % of the gap to the supply and a discharging one has 36.8 % of its voltage left. The figure is a property of the exponential itself, not of any particular circuit, which is why every RC network in the world shares it. This calculator computes each row of its 1τ–5τ table from 1 − e^(−n) at run time rather than reading 63.2 and 86.5 out of a table of constants.

Why is 5τ treated as fully charged when the maths never gets there?

It genuinely never gets there — the exponential is asymptotic, so the capacitor is always a little short of the supply. At 5τ it is 99.33 % of the way, leaving an error of 0.67 %, which is smaller than the tolerance of the components themselves and usually smaller than the noise on the rail. Engineers therefore call 5τ settled by convention. If you need a tighter figure, 7τ leaves 0.09 % and 10τ leaves 0.0045 %; asking the calculator for exactly 100 % is rejected because ln(0) diverges.

How is the time constant related to the filter cutoff frequency?

They are the same number seen from the time domain and the frequency domain: f_c = 1/(2πτ) = 1/(2πRC). A 1 kΩ resistor with a 100 nF capacitor gives τ = 100 µs and therefore a corner at 1.592 kHz. At the corner the output is 1/√2 of the input, which is −3.01 dB, and the phase has shifted 45°. Above the corner a first-order section rolls off at 20 dB per decade, so at 5 kHz — about 3.14 times the corner — the low-pass output is −10.36 dB with −72.34° of phase shift.

Why does a resistor waste exactly half the energy when charging a capacitor?

The supply pushes charge Q = CV through a potential difference of V, delivering QV = CV² joules, but the capacitor only ends up storing ½CV². The other half is dissipated in the resistor, and the result is independent of the resistance — a 1 Ω resistor and a 1 MΩ resistor both burn exactly half, one quickly and one slowly. That is why switched-mode converters and inductive charging exist: there is no way to beat 50 % with a plain series resistor.

How accurate is this for a real circuit?

The exponential arithmetic is exact for the values you enter, but the components are not. Electrolytic capacitors are routinely ±20 % and drift further with age and temperature, resistors are typically ±1 % to ±5 %, and a capacitor's own equivalent series resistance and leakage add to the network. The tolerance panel shows the worst-case τ envelope those figures produce, and it is often wide enough to matter — a 10 kΩ ±5 % resistor with a 100 µF ±20 % capacitor gives a nominal 1 s that can legitimately land anywhere between 0.76 s and 1.26 s.