DEV Community

Cover image for RC Time Constant: The Simple Timing Formula That Can Hide Inrush and Stored Energy Problems
Evgenii Konkin
Evgenii Konkin

Posted on

RC Time Constant: The Simple Timing Formula That Can Hide Inrush and Stored Energy Problems

RC timing looks easy.

Pick a resistor.

Pick a capacitor.

Multiply them.

τ = R × C
Enter fullscreen mode Exit fullscreen mode

That gives the time constant.

After one time constant, the capacitor has moved about 63.2% toward its final voltage. After five time constants, it is about 99.3% settled.

Simple.

But in real circuits, RC timing is one of those calculations that gets used too casually. The time constant may be correct, while the circuit still has a dangerous inrush current, a capacitor that stores hazardous energy, a slow bleed-down path, or a timing value that changes because the real capacitor is not the value printed on the schematic.

The formula is short.

The engineering consequences are not.

The basic RC time constant formula

For a simple series resistor-capacitor circuit:

τ = R × C
Enter fullscreen mode Exit fullscreen mode

Where:

τ = time constant, seconds
R = effective resistance, ohms
C = capacitance, farads
Enter fullscreen mode Exit fullscreen mode

The word “effective” matters.

The resistance is not always just the resistor shown on the schematic.

It can include:

Series resistor
Source impedance
Switch on-resistance
Capacitor ESR
Trace resistance
Connector resistance
Bleed path, if analyzing discharge
Enter fullscreen mode Exit fullscreen mode

For a rough first pass, the external resistor may dominate.

For power electronics, snubbers, bulk capacitors, and fast transients, the hidden resistance can matter a lot.

Unit shortcuts are useful, but dangerous if misread

The formula uses ohms and farads.

But practical circuits usually use kΩ, MΩ, nF, μF, and sometimes mF.

Some useful shortcuts:

Ω × F = seconds
kΩ × μF = milliseconds
kΩ × nF = microseconds
MΩ × μF = seconds
MΩ × nF = milliseconds
Enter fullscreen mode Exit fullscreen mode

Example:

R = 10 kΩ
C = 1 μF
Enter fullscreen mode Exit fullscreen mode

Using the shortcut:

τ = 10 × 1
τ = 10 ms
Enter fullscreen mode Exit fullscreen mode

Using base units:

R = 10,000 Ω
C = 0.000001 F

τ = 10,000 × 0.000001
τ = 0.01 s
τ = 10 ms
Enter fullscreen mode Exit fullscreen mode

Both are correct.

The danger is entering the wrong unit.

A 1 μF capacitor is:

1 μF = 0.000001 F
Enter fullscreen mode Exit fullscreen mode

A 1 mF capacitor is:

1 mF = 0.001 F
Enter fullscreen mode Exit fullscreen mode

That is 1000× larger.

A timing circuit intended to settle in milliseconds can accidentally become seconds just from a unit mistake.

Charging is exponential, not linear

When a capacitor charges from zero toward a DC supply voltage, the voltage follows:

vC(t) = Vsupply × (1 − e^(−t/τ))
Enter fullscreen mode Exit fullscreen mode

That means the capacitor does not charge at a constant rate.

It rises quickly at first, then slows down as it approaches the final voltage.

Useful checkpoints:

1τ = 63.2% of final voltage
2τ = 86.5%
3τ = 95.0%
4τ = 98.2%
5τ = 99.3%
Enter fullscreen mode Exit fullscreen mode

This is why engineers often use:

t_settle = 5τ
Enter fullscreen mode Exit fullscreen mode

as a practical settling estimate.

It does not mean the capacitor is mathematically “finished” charging. It means the remaining error is small enough for many practical circuits.

Discharge uses the same time constant

When a charged capacitor discharges through a resistor, the voltage follows:

vC(t) = Vinitial × e^(−t/τ)
Enter fullscreen mode Exit fullscreen mode

After one time constant, the capacitor still has about 36.8% of its initial voltage.

After five time constants, it has about 0.7% remaining.

That is why a capacitor can still be energized after the power is off.

The circuit may look dead.

The capacitor may not be dead.

This is especially important in high-voltage DC links, power supplies, motor drives, inverters, test equipment, and capacitor banks.

Worked example: MCU reset delay

Suppose a simple RC delay is used for a microcontroller reset input.

The design uses:

R = 100 kΩ
C = 1 μF
Vsupply = 3.3 V
Enter fullscreen mode Exit fullscreen mode

The time constant is:

τ = R × C
τ = 100,000 × 0.000001
τ = 0.1 s
τ = 100 ms
Enter fullscreen mode Exit fullscreen mode

The 5τ settling time is:

t_settle = 5 × 100 ms
t_settle = 500 ms
Enter fullscreen mode Exit fullscreen mode

At first glance, the circuit gives a nice slow reset ramp.

But many reset inputs do not switch at 99.3% of supply.

They switch at some threshold.

Suppose the reset input releases at about 70% of supply.

The threshold voltage is:

Vthreshold = 0.70 × 3.3
Vthreshold = 2.31 V
Enter fullscreen mode Exit fullscreen mode

The time to reach a threshold while charging from zero is:

t = −τ × ln(1 − Vthreshold / Vsupply)
Enter fullscreen mode Exit fullscreen mode

Since Vthreshold / Vsupply = 0.70:

t = −100 ms × ln(1 − 0.70)
t = −100 ms × ln(0.30)
t ≈ 120 ms
Enter fullscreen mode Exit fullscreen mode

So the reset may release after about 120 ms, not 500 ms.

That is a common interpretation mistake.

means “nearly settled.”

It does not mean “the circuit event happens at 5τ.”

The actual event depends on the threshold.

The engineering mistake: treating supply voltage as part of τ

A common mistake is thinking that a higher supply voltage makes an RC circuit charge “faster.”

It does not change the time constant.

The time constant is:

τ = R × C
Enter fullscreen mode Exit fullscreen mode

There is no voltage term in that formula.

If R and C stay the same, τ stays the same.

For example:

R = 10 kΩ
C = 1 μF
Enter fullscreen mode Exit fullscreen mode

At 5 V:

τ = 10 ms
Enter fullscreen mode Exit fullscreen mode

At 24 V:

τ = 10 ms
Enter fullscreen mode Exit fullscreen mode

At 400 V:

τ = 10 ms
Enter fullscreen mode Exit fullscreen mode

The timing scale is the same.

But supply voltage changes other things:

Final capacitor voltage
Peak inrush current
Stored energy
Shock hazard
Voltage rating margin
Resistor pulse energy
Enter fullscreen mode Exit fullscreen mode

That is the trap.

Voltage may not change the time constant, but it can completely change the safety and stress level of the circuit.

Peak inrush current can be the real problem

When a discharged capacitor is connected to a supply through a resistance, the initial charging current is:

Ipeak = Vsupply / R
Enter fullscreen mode Exit fullscreen mode

This is the maximum current at the instant of switch closure, assuming the capacitor starts at zero volts.

Example:

Vsupply = 24 V
R = 2 Ω
C = 1000 μF
Enter fullscreen mode Exit fullscreen mode

The time constant is:

τ = 2 × 0.001
τ = 0.002 s
τ = 2 ms
Enter fullscreen mode Exit fullscreen mode

That looks fast.

But the peak inrush current is:

Ipeak = 24 / 2
Ipeak = 12 A
Enter fullscreen mode Exit fullscreen mode

The capacitor charges quickly, but the switch, diode, MOSFET, relay contact, connector, or PCB trace may see a large current pulse.

If the engineer only checks τ, the design may look fine.

If the engineer checks Ipeak, the design may need a current limiter, soft-start circuit, NTC thermistor, precharge resistor, or a different switching device.

For many real designs, the question is not only:

How long does it take to charge?
Enter fullscreen mode Exit fullscreen mode

It is also:

What current flows at the first instant?
Enter fullscreen mode Exit fullscreen mode

Stored energy grows with voltage squared

A capacitor stores energy:

WC = 0.5 × C × V²
Enter fullscreen mode Exit fullscreen mode

Voltage is squared.

That makes high-voltage capacitors very different from low-voltage capacitors even when the capacitance looks modest.

Example 1:

C = 1000 μF
V = 12 V
Enter fullscreen mode Exit fullscreen mode

Stored energy:

WC = 0.5 × 0.001 × 12²
WC = 0.072 J
Enter fullscreen mode Exit fullscreen mode

That is 72 mJ.

Now use the same capacitance at 400 V:

C = 1000 μF
V = 400 V
Enter fullscreen mode Exit fullscreen mode

Stored energy:

WC = 0.5 × 0.001 × 400²
WC = 80 J
Enter fullscreen mode Exit fullscreen mode

Same capacitance.

Much higher voltage.

The stored energy increased from 0.072 J to 80 J.

That is more than 1000× higher.

This is why high-voltage bulk capacitors need serious discharge planning. The circuit may be powered off, but the capacitor can still store enough energy to be hazardous.

Bleed resistor design is a tradeoff

A bleed resistor gives the capacitor a controlled discharge path after power-off.

For a bleed resistor:

τbleed = Rbleed × C
Enter fullscreen mode Exit fullscreen mode

The time to discharge from a starting voltage to a target voltage is:

t = τbleed × ln(Vstart / Vtarget)
Enter fullscreen mode Exit fullscreen mode

Suppose a DC link capacitor has:

C = 1000 μF
Vstart = 400 V
Target voltage = 50 V
Enter fullscreen mode Exit fullscreen mode

The design goal is to discharge below 50 V in 60 seconds.

Rearrange the equation:

τbleed = t / ln(Vstart / Vtarget)
Enter fullscreen mode Exit fullscreen mode

Calculate:

τbleed = 60 / ln(400 / 50)
τbleed = 60 / ln(8)
τbleed ≈ 60 / 2.079
τbleed ≈ 28.9 s
Enter fullscreen mode Exit fullscreen mode

Now find the bleed resistor:

Rbleed = τbleed / C
Rbleed = 28.9 / 0.001
Rbleed = 28,900 Ω
Enter fullscreen mode Exit fullscreen mode

So a resistor around 29 kΩ would meet the timing target in a first-pass calculation.

But now check continuous power while the supply is on:

Pbleed = V² / Rbleed
Pbleed = 400² / 28,900
Pbleed ≈ 5.5 W
Enter fullscreen mode Exit fullscreen mode

That is not a tiny resistor.

The engineer may need a higher power rating, multiple resistors in series, voltage-rating checks, thermal spacing, and derating.

The mistake is sizing a bleed resistor only by discharge time.

Faster discharge requires lower resistance.

Lower resistance increases continuous power loss.

There is always a tradeoff.

RC cutoff frequency is related, but not the same question

The first-order RC cutoff frequency is:

fc = 1 / (2π × R × C)
Enter fullscreen mode Exit fullscreen mode

Since:

τ = R × C
Enter fullscreen mode Exit fullscreen mode

then:

fc = 1 / (2π × τ)
Enter fullscreen mode Exit fullscreen mode

This connects time-domain behavior and frequency-domain behavior.

Example:

R = 10 kΩ
C = 1 μF
τ = 10 ms
Enter fullscreen mode Exit fullscreen mode

Cutoff frequency:

fc = 1 / (2π × 0.01)
fc ≈ 15.9 Hz
Enter fullscreen mode Exit fullscreen mode

That number is useful for first-order filter intuition.

But it is not a magic boundary where signals suddenly stop.

At the cutoff frequency, the output is down by about 3 dB. Above that, the signal is increasingly attenuated. Below that, it is not perfectly unchanged either, depending on the exact topology.

The mistake is treating fc as an on/off limit.

A first-order RC filter is a slope, not a brick wall.

Real capacitors are not ideal

The formula assumes a simple capacitor.

Real capacitors bring extra behavior:

ESR
ESL
Leakage current
Voltage rating
Temperature coefficient
Tolerance
DC bias loss
Dielectric absorption
Ripple current limit
Polarity limits for electrolytic and tantalum capacitors
Enter fullscreen mode Exit fullscreen mode

For timing circuits, tolerance can dominate.

A nominal design may use:

R = 100 kΩ ±1%
C = 1 μF ±20%
Enter fullscreen mode Exit fullscreen mode

Nominal time constant:

τ = 100 ms
Enter fullscreen mode Exit fullscreen mode

But the capacitor tolerance alone can push the actual value roughly between:

80 ms and 120 ms
Enter fullscreen mode Exit fullscreen mode

Add resistor tolerance and temperature effects, and the real spread can be wider.

This matters for reset timing, watchdog delays, debounce circuits, one-shot timers, and analog filtering.

For stable timing, capacitor type matters. A Class II ceramic capacitor may lose effective capacitance under DC bias. An electrolytic capacitor may have wide tolerance and leakage. A C0G/NP0 capacitor is stable, but usually available only in much smaller capacitance values.

The formula uses the capacitance value you enter.

The circuit uses the capacitance value the part actually has under voltage, temperature, tolerance, and aging.

The common mistake: ignoring the actual threshold

Many RC timing mistakes come from assuming the circuit switches at 1τ or 5τ.

But real inputs switch at thresholds.

A CMOS input may switch around a logic threshold.

A Schmitt trigger may switch at a defined upper or lower threshold.

A comparator may use a reference voltage.

A 555 timer has internal thresholds.

That means the correct question is not always:

What is τ?
Enter fullscreen mode Exit fullscreen mode

The correct question may be:

When does vC(t) cross the actual threshold?
Enter fullscreen mode Exit fullscreen mode

For charging from zero:

t = −τ × ln(1 − Vthreshold / Vsupply)
Enter fullscreen mode Exit fullscreen mode

For discharge:

t = τ × ln(Vinitial / Vtarget)
Enter fullscreen mode Exit fullscreen mode

This is where engineering judgment enters.

The RC formula gives the time scale.

The receiving circuit defines the event.

Practical design takeaway

The RC time constant is a great first-pass tool.

It helps answer:

How fast does the capacitor charge?
How fast does it discharge?
When is the circuit practically settled?
What is the peak inrush current?
How much energy is stored?
What is the first-order cutoff frequency?
How long does a bleed resistor take to reduce the voltage?
How much continuous power does the bleed resistor dissipate?
Enter fullscreen mode Exit fullscreen mode

But it does not answer everything.

It does not automatically verify capacitor voltage rating.

It does not account for DC bias derating unless you do that separately.

It does not guarantee timing accuracy with wide-tolerance capacitors.

It does not make a high-voltage capacitor safe after power-off.

It does not replace checking switch, diode, MOSFET, fuse, and resistor pulse ratings.

The formula is the starting point, not the whole design.

Final thought

The RC time constant is one of the most useful equations in practical electronics:

τ = R × C
Enter fullscreen mode Exit fullscreen mode

It explains debounce circuits, reset delays, filters, snubbers, bulk capacitor charging, discharge paths, soft-start behavior, and timing networks.

But the same simple formula can hide important problems.

Supply voltage does not change τ, but it changes stored energy and inrush current.

A capacitor may look charged after 5τ, but a logic input may switch much earlier.

A bleed resistor may discharge the capacitor quickly, but it may also burn continuous power while the equipment is running.

And a high-voltage capacitor can remain dangerous long after the circuit is switched off.

For quick RC transient checks, 5τ settling time, peak inrush current, stored capacitor energy, cutoff frequency, and bleed resistor discharge calculations, use the RC Time Constant Calculator on CalcEngineer.

Top comments (0)