RC Filter Cutoff Frequency: How to Calculate and Design RC Filters
Master the RC filter cutoff frequency formula, learn to select the right resistor and capacitor values, and avoid the design mistakes that catch even experienced engineers off guard.
What Is the RC Filter Cutoff Frequency?
The RC filter cutoff frequency is the frequency at which a passive RC filter transitions from passing signals to attenuating them. At this point, the output power drops to exactly half of the input power, which corresponds to a −3 dB reduction in amplitude. For a first-order RC filter—the simplest filter in electronics, built from just one resistor and one capacitor—the cutoff frequency is determined entirely by the product of R and C:
This single equation is the foundation of every passive RC filter design. Whether you are building a low-pass filter to remove high-frequency noise from a sensor signal, or a high-pass filter to block DC offset from an audio input, the cutoff frequency is always given by this formula. The only difference between the two filter types is which component the output is taken across: the capacitor for low-pass, or the resistor for high-pass.
Understanding this equation is not enough on its own. In practice, component tolerances, source and load impedances, and parasitic effects all shift the actual cutoff frequency away from the calculated value. This guide walks through the theory, the calculation, the practical component selection, and the real-world pitfalls—with worked examples you can verify using the RC Filter Calculator.
Deriving the Cutoff Frequency Formula
The RC filter cutoff frequency formula comes directly from the circuit's transfer function. Consider a low-pass RC filter: the resistor R is in series with the signal path, and the capacitor C is connected from the output node to ground. The output voltage is measured across the capacitor.
In the frequency domain, the resistor has impedance ZR = R, and the capacitor has impedance ZC = 1/(jωC). The circuit forms a frequency-dependent voltage divider:
The magnitude of this transfer function is:
The cutoff frequency is defined as the frequency where the output power is half the input power, meaning |H| = 1/√2 ≈ 0.707. Setting the magnitude equal to 1/√2 and solving:
Converting from angular frequency to ordinary frequency (f = ω/2π):
This derivation confirms that the cutoff frequency depends only on the RC time constant τ = RC. The same formula applies to the high-pass configuration because the magnitude response of both filters shares the same 3 dB point—only the passband and stopband are swapped.
Calculating RC Filter Cutoff Frequency: Worked Examples
Let us work through three practical design problems. You can cross-check every result with the RC Filter Calculator, which also recommends the nearest E24 standard component values.
Example 1: Audio Low-Pass Filter at 1 kHz
Design a low-pass RC filter with a cutoff frequency of 1 kHz for an audio application. We need to find a suitable R and C pair.
Rearranging the cutoff frequency formula to solve for C given R and fc:
Choose R = 10 kΩ (a standard value in the recommended 1 kΩ–100 kΩ range for general-purpose filters):
The nearest standard value is 16 nF (E24 series). With C = 16 nF, the actual cutoff frequency is:
A deviation of just 0.53% from the target—well within typical audio filter tolerances. The time constant is τ = RC = 100 μs, meaning the filter reaches 63.2% of its final value within 100 μs of a step input.
Example 2: ADC Anti-Aliasing Filter at 500 Hz
A 12-bit ADC sampling at 1 kSPS needs an anti-aliasing filter. Per the Nyquist criterion, the filter should attenuate frequencies above 500 Hz (half the sampling rate). Setting fc = 500 Hz and choosing R = 3.3 kΩ:
The nearest standard value is 100 nF, yielding fc = 482.3 Hz—a −3.5% deviation. For anti-aliasing, slightly lowering the cutoff frequency is acceptable because it provides more attenuation at the Nyquist frequency. According to Texas Instruments' ADC application guidelines, filter resistor values should be kept below 1 kΩ for precision ADCs to minimize voltage drop from input bias currents. If your ADC requires low source impedance, consider using an op-amp buffer between the filter and the ADC input (TI ADS101x datasheet, Section 8.2.2.6).
Example 3: High-Pass Filter for AC Coupling at 20 Hz
Design a high-pass RC filter to block DC while passing audio signals above 20 Hz. With R = 47 kΩ:
The nearest standard value is 150 nF (E24), giving fc = 22.6 Hz. The alternative 180 nF yields fc = 18.8 Hz. For audio coupling, either value is acceptable. Note that the −3 dB point at 20 Hz means the signal at 20 Hz is already attenuated by 3 dB; if you need flat response down to 20 Hz, set the cutoff to around 5–10 Hz instead.
RC Filter Cutoff Frequency: Practical Component Selection
The formula fc = 1/(2πRC) implies that any R-C combination satisfying the product RC = 1/(2πfc) will work. In practice, the choice of R and C is constrained by several factors that determine whether your filter works reliably in a real circuit.
Resistance Range
Too low (<100 Ω): The filter draws excessive current from the source. If the source has finite output impedance (e.g., an op-amp with 50–100 Ω output resistance), that impedance becomes part of R, shifting the cutoff frequency unpredictably. Power dissipation also increases: a 100 Ω resistor with 5 V across it dissipates 250 mW.
Too high (>1 MΩ): High-value resistors are noisy (Johnson noise scales with √R) and susceptible to PCB surface leakage currents. In circuits with op-amps, input bias currents flowing through large resistors create significant offset voltages. For a FET-input op-amp with 1 pA bias current, a 10 MΩ resistor produces a 10 mV offset.
Recommended range: 1 kΩ to 100 kΩ for general-purpose filter designs. Use the lower end when driving low-impedance loads, and the upper end for high-impedance sensor interfaces.
Capacitance Range and Dielectric Selection
Too small (<1 nF): Stray PCB capacitance (typically 2–5 pF per cm of trace) adds to C, shifting the cutoff frequency. A 100 pF filter capacitor with 5 pF of stray capacitance is already 5% off before considering component tolerance.
Too large (>10 μF): Large capacitors are physically bulky, expensive, and often use dielectrics with poor stability (Y5V/Z5U ceramics can lose 80% of capacitance over temperature and voltage). Electrolytic capacitors have wide tolerances (±20% or worse) and high equivalent series resistance (ESR).
Dielectric recommendations:
| Dielectric | Stability | Tolerance | Best For |
|---|---|---|---|
| C0G/NP0 | ±30 ppm/°C | ±5% | Precision filters, oscillators |
| X7R | ±15% over temp | ±10% | General-purpose coupling/bypass |
| X5R | ±15% over temp | ±10% | Power supply filtering |
| Y5V | +22%/−82% | ±20%+ | Avoid for filter applications |
For filter applications, C0G/NP0 ceramics or film capacitors are strongly preferred because their capacitance remains stable across temperature and applied voltage. Need help decoding the markings on your capacitor? Use the Capacitor Code Calculator to convert 3-digit EIA codes to actual values.
Component Tolerance and Worst-Case Analysis
A filter built with ±5% resistors and ±10% capacitors can have its cutoff frequency shift by up to ±15% from the nominal value (tolerances add in the worst case). For a target fc of 1 kHz, this means the actual cutoff could be anywhere from 850 Hz to 1,150 Hz. In applications where the cutoff frequency is critical (anti-aliasing, crossover networks), use ±1% resistors and C0G capacitors with ±5% tolerance to reduce the worst-case shift to ±6%.
Common Mistakes When Designing RC Filters
Even experienced engineers make these errors. Each one shifts the actual RC filter cutoff frequency away from the calculated value or degrades filter performance.
Mistake 1: Ignoring Source and Load Impedance
The cutoff frequency formula assumes ideal source impedance (zero) and load impedance (infinite). Real circuits have neither. If the source has output resistance RS, it adds to the filter resistor: Reffective = R + RS. If the load has finite resistance RL, it appears in parallel with the capacitor (low-pass) or the resistor (high-pass), reducing the effective impedance.
Example: A low-pass filter with R = 10 kΩ and C = 16 nF (fc = 995 Hz) is driven by a source with RS = 1 kΩ. The effective resistance is 11 kΩ, shifting fc to 905 Hz—a 9% error. If the load is 100 kΩ in parallel with the capacitor, the effective R increases further and the cutoff drops even more.
Fix: Make the filter resistance at least 10× the source impedance, and ensure the load impedance is at least 10× the filter impedance. When this is not possible, use an op-amp buffer to isolate the filter from source and load.
Mistake 2: Cascading First-Order Stages Without Recalculating
Cascading two identical RC low-pass filters does not produce a second-order filter with the same cutoff frequency. The −3 dB frequency of two cascaded identical stages is:
Two 1 kHz stages yield a −3 dB point at approximately 607 Hz, not 1 kHz. Additionally, the second stage loads the first, further shifting the response. To get a proper second-order response with the desired cutoff, use a Sallen-Key or multiple-feedback active filter topology instead of passive cascading.
Mistake 3: Using an RC Filter as a Power Supply
A voltage divider is not a voltage regulator, and an RC filter is not a power supply. If you need to drop 12 V to 3.3 V for a circuit drawing 20 mA, an RC filter wastes power in the resistor and the output voltage collapses under load. Use a linear regulator (LDO) or a buck converter instead. RC filters are for signal conditioning, not power delivery.
When to Use Second-Order or Active Filters Instead
A first-order RC filter attenuates at only 20 dB/decade (6 dB/octave) above the cutoff frequency. This gentle roll-off is insufficient for many applications. Here is when you should consider higher-order or active alternatives:
- Anti-aliasing for 16-bit ADCs: A first-order filter provides only 6 dB of attenuation one octave above fc. At 16-bit resolution (96 dB dynamic range), you need at least a 4th-order filter to achieve sufficient attenuation at the Nyquist frequency. A Sallen-Key cascade or switched-capacitor filter (e.g., LTC1062) is more appropriate.
- Audio crossover networks: A 6 dB/octave slope produces significant overlap between frequency bands. Second-order Linkwitz-Riley crossovers (12 dB/octave) are the minimum for acceptable separation. Fourth-order (24 dB/octave) is the professional standard for flat summed response.
- EMI suppression per IEC 61000-4: For electromagnetic compatibility compliance, multi-stage LC or Pi-filters provide the steep roll-off needed to attenuate conducted emissions above 150 kHz. IEC 60384-14 specifies requirements for EMI suppression capacitors (X-capacitors and Y-capacitors) used in these applications.
- Low-frequency applications requiring sharp cutoff: If fc is below 100 Hz, passive RC filters with their gentle roll-off may not provide enough attenuation in the stopband. Active filters using op-amps allow higher Q and sharper transitions without requiring impractically large capacitors.
Summary: RC Filter Cutoff Frequency Design Checklist
Use this checklist every time you design an RC filter to avoid common mistakes and ensure your cutoff frequency is accurate:
- Calculate fc using fc = 1/(2πRC). Verify with the RC Filter Calculator.
- Select R in the 1 kΩ–100 kΩ range to balance current draw, noise, and impedance matching.
- Choose a stable capacitor dielectric—C0G/NP0 for precision, X7R for general purpose. Decode markings with the Capacitor Code Calculator.
- Account for source impedance by making R ≥ 10 × Rsource, or add a buffer.
- Account for load impedance by ensuring Rload ≥ 10 × Rfilter, or add a buffer.
- Perform worst-case analysis using component tolerances to verify the acceptable fc range.
- Consider whether first-order is sufficient—if you need more than 20 dB/decade attenuation, upgrade to a second-order or active topology.
The RC filter cutoff frequency is one of the most frequently calculated values in electronics. Mastering its derivation, its practical constraints, and its failure modes will make you a more effective circuit designer. When in doubt, prototype the filter and measure the actual −3 dB point with a frequency sweep—it is the only way to confirm that real-world component behavior matches your calculations.