Transfer Function Poles Zeros Interactive Calculator

General Transfer Function (Zero-Pole-Gain Form)

H(s) = K · (s - z₁)(s - z₂)...(s - z_m) / (s - p₁)(s - p₂)...(s - p_n)

K = system gain (dimensionless)

z_i = zero locations in s-plane (rad/s)

p_i = pole locations in s-plane (rad/s)

First-Order System

H(s) = K / (s - p) = K / (s + 1/τ)

p = pole location (rad/s, negative for stability)

τ = time constant = -1/p (seconds)

t_s = settling time (2%) = 4τ (seconds)

Second-Order System (Standard Form)

H(s) = Kω_n² / (s² + 2ζω_ns + ω_n²)

ω_n = natural frequency (rad/s)

ζ = damping ratio (dimensionless)

Poles = -ζω_n ± jω_n√(1-ζ²) for 0 < ζ < 1

ω_d = damped frequency = ω_n√(1-ζ²) (rad/s)

Frequency Response

|H(jω)| = K / |jω - p|

∠H(jω) = -∠(jω - p)

|H(jω)| = magnitude response (absolute value)

∠H(jω) = phase response (degrees or radians)

Magnitude (dB) = 20·log₁₀(|H(jω)|)

Stability Criteria & Transient Parameters

Percent Overshoot = exp(-πζ/√(1-ζ²)) × 100%

Peak Time t_p = π/ω_d

Settling Time t_s = 4/(ζω_n) [for 2% criterion]

Stability: All poles must have negative real parts (left half-plane)

Scenario: Tuning a Servo Motor Controller

Marcus, a robotics engineer at an industrial automation company, struggles with oscillation in a new pick-and-place robot arm. The servo controller exhibits 18% overshoot and 2.3 Hz ringing when moving to target positions, causing positioning errors and part damage. He measures the step response and estimates the dominant poles lie at -7.2 ± j14.5 rad/s. Using this calculator's second-order mode, he enters ω_n = 16.2 rad/s and ζ = 0.44. The calculator reveals percent overshoot of 20.8% and settling time of 560 ms, matching his observations. He then explores increasing ζ to 0.65 by adjusting the velocity feedback gain, which the calculator predicts will reduce overshoot to 6.2% with negligible increase in settling time. After implementing the change, the arm achieves smooth, accurate positioning—the calculator provided the exact pole placement targets he needed to tune the PID gains without trial-and-error iteration.

Scenario: Designing an Audio Crossover Network

Elena, a loudspeaker designer for a high-end audio manufacturer, develops a three-way crossover separating bass (below 300 Hz), midrange (300-3500 Hz), and tweeter (above 3500 Hz) signals. She needs fourth-order Linkwitz-Riley alignment, which requires two cascaded second-order Butterworth sections (ζ = 0.707) at each crossover frequency. For the 300 Hz low-pass feeding the woofer, she uses this calculator to analyze poles at -1333 ± j1333 rad/s (ω_n = 1885 rad/s, f = 300 Hz). The frequency response mode confirms -3 dB at exactly 300 Hz and -24 dB/octave roll-off. She repeats the analysis for the 3500 Hz high-pass protecting the tweeter, verifying pole locations at -15558 ± j15558 rad/s. The zero-pole-gain mode helps her design the midrange bandpass by combining high-pass zeros from the low section with low-pass poles from the high section, ensuring flat summed response and phase coherence at crossover points. The calculator eliminated hours of SPICE simulation, providing instant verification that her pole placements meet the Linkwitz-Riley criteria.

Scenario: Stabilizing a Switching Power Supply

Rajesh, a power electronics engineer, debugs instability in a 500 kHz buck converter powering a telecom base station. Under heavy load transients, the output voltage rings at 15 kHz before settling, violating the 50 mV ripple specification. He models the LC output filter as a second-order system with poles at -4712 ± j94247 rad/s (from measured L = 4.7 μH, C = 100 μF). Using the calculator's frequency response mode, he evaluates the uncompensated gain and phase at the crossover frequency (40 kHz), finding phase margin of only 12°—explaining the instability. He designs a Type III compensator adding two zeros at 15 kHz (to cancel the LC poles) and a pole at 200 kHz (for high-frequency noise attenuation). The zero-pole-gain mode confirms the compensated phase margin improves to 58° while maintaining fast transient response. After implementing the compensator with resistor-capacitor networks around the error amplifier, the converter achieves stable operation with 25 μV ripple and 80 μs load transient recovery time. The pole-zero analysis transformed an unstable design into a production-ready power supply.

What's the difference between poles and zeros in practical terms? +

Poles represent the natural resonant frequencies and damping of a system—they determine where energy storage elements (capacitors, inductors, masses, springs) create characteristic oscillations or exponential responses. Zeros, by contrast, block or attenuate specific frequencies; they shape the relative gain between different frequency components without adding energy storage. In practical filter design, poles create the fundamental roll-off slope (each pole adds -20 dB/decade), while zeros can create notches, flatten passband ripple, or steepen transition bands. For example, an LC low-pass filter has two poles at the resonant frequency but no finite zeros (zeros at infinity), giving -40 dB/decade roll-off. Adding a series RC branch creates a zero that can cancel one pole, reducing the roll-off to -20 dB/decade but improving phase margin. In control systems, pole placement determines stability and transient response speed, while zeros affect the control effort required and can introduce overshoot or non-minimum phase behavior if placed in the right half-plane.

Why do complex conjugate poles always appear in pairs? +

Physical systems described by real-valued differential equations must have transfer functions with real coefficients, which mathematically requires complex poles and zeros to occur in conjugate pairs. If a system has a pole at s = σ + jω, it must also have one at s = σ - jω to ensure that when the denominator polynomial is expanded, all coefficients remain real. This pairing represents physically observable sinusoidal oscillations: the complex exponential e^(σt)·e^(jωt) and its conjugate combine via Euler's formula to produce real-valued damped sinusoids of the form e^(σt)·cos(ωt). A single complex pole without its conjugate would require imaginary coefficients in the differential equation, violating physical realizability. This constraint extends to zeros as well—any complex zero must have a conjugate partner. The only exception occurs in discrete-time systems sampled from continuous systems, where the z-transform can yield unpaired complex values under certain aliasing conditions, though even these derive from underlying conjugate pairs in the s-domain before sampling.

How does pole location relate to bandwidth and rise time? +

For a first-order system with pole at s = -σ, the -3 dB bandwidth equals exactly σ rad/s, and rise time (10% to 90%) approximates 2.2/σ seconds. This direct relationship makes pole location a powerful design parameter: doubling σ (moving the pole farther left) doubles bandwidth and halves rise time. Second-order systems with complex poles at σ ± j��_d have bandwidth approximately equal to ω_n for ζ ≈ 0.7 (Butterworth response), but the relationship becomes nonlinear for other damping ratios—underdamped systems (ζ < 0.4) exhibit resonant peaking that extends bandwidth beyond ω_n, while overdamped systems (ζ > 1) have reduced bandwidth. Rise time for second-order systems depends on both ζ and ω_n, roughly following t_r ≈ (1.8/ω_n) for 0.3 < ζ < 0.8. The fundamental trade-off emerges clearly: faster time response (larger |σ| or ω_n) requires wider bandwidth, which admits more high-frequency noise. Control system designers balance this by placing poles far enough left for adequate speed while adding compensator zeros to boost phase margin without excessive bandwidth expansion. The calculator's settling time output (4/|σ| for dominant real part) provides the complementary time-domain metric—systems must wait approximately four time constants for transients to decay to 2% of final value.

What causes right half-plane zeros and why are they problematic? +

Right half-plane (RHP) zeros arise in non-minimum phase systems where the initial response to an input moves opposite to the eventual steady-state direction, creating a magnitude boost with phase lag instead of the lead provided by left half-plane zeros. Classic examples include boost and buck-boost DC-DC converters operating in current-mode control, where increasing the duty cycle initially decreases output voltage due to inductor current dynamics before the voltage rises to the new setpoint. Aerospace systems exhibit RHP zeros when actuators positioned behind the center of mass create initial pitch-down before pitch-up. The fundamental problem: RHP zeros contribute +20 dB/decade magnitude slope (beneficial for compensator gain) but -90° phase lag (detrimental to stability), limiting achievable crossover frequency and bandwidth. Controllers attempting to overcome RHP zero phase lag by adding aggressive lead compensation risk amplifying high-frequency noise and creating multiple crossover points with unstable regions. Practical design strategies include accepting reduced bandwidth (crossover frequency typically limited to 1/5 to 1/3 of the RHP zero frequency), using multi-loop control architectures that move the RHP zero outside the critical frequency range, or switching to control modes that eliminate the RHP zero entirely (such as voltage-mode control in converters, though this trades the RHP zero problem for other stability challenges).

How accurate is the dominant pole approximation for higher-order systems? +

The dominant pole approximation remains accurate when non-dominant poles have real parts at least five times larger in magnitude than the dominant poles, ensuring their contributions to transient response decay five times faster and contribute negligibly to overall settling time and overshoot. For a fourth-order system with poles at -2 ± j3, -12, and -18 rad/s, the complex pair at -2 ± j3 dominates (|σ| = 2 rad/s), and the approximation predicts settling time of 4/2 = 2 seconds. The non-dominant real poles at -12 and -18 contribute transients dying out in 0.33 and 0.22 seconds respectively, adding minor fast-decaying components but not significantly altering the dominant 2-second settling. However, the approximation breaks down when poles cluster closer together or when non-dominant poles interact with system zeros. A common failure mode occurs in cascade filter designs where multiple second-order sections produce pole groups at -10 ± j8, -12 ± j9, and -14 ± j10 rad/s—no single pair truly dominates, and the composite response shows beating effects and unexpected overshoot characteristics. The approximation also fails for zeros near dominant poles (pole-zero dipoles), which can effectively cancel the pole's influence despite proximity to the imaginary axis. Engineers verify dominant pole assumptions by calculating the residue (contribution magnitude) of each pole to the step response: the dominant pole should have residue magnitude at least three times larger than any non-dominant contribution for reliable second-order approximation.

Can pole-zero analysis predict real-world component tolerances effects? +

Pole-zero sensitivity analysis quantifies how component variations shift pole locations and degrade performance, critical for robust design with real components having 1-20% tolerances. For an LC filter with pole at ω_n = 1/√(LC), a ±10% capacitor tolerance shifts ω_n by approximately ±5% (since ω_n ∝ 1/√C), moving poles and altering passband edge frequency. Damping ratio ζ shows even higher sensitivity in active filters where ζ = (1/2)√(C/L)/R, making ζ dependent on three component tolerances simultaneously—Monte Carlo analysis reveals that 5% resistor and capacitor tolerances can cause ζ variation from 0.6 to 0.9, dramatically changing overshoot from 10% to 0.4%. Pole-zero migration plots show statistical clouds of probable pole locations rather than single points, with quality control requiring sufficient distance from the jω-axis stability boundary to accommodate worst-case tolerance stacking. Temperature coefficients introduce additional time-varying pole migration: ceramic capacitors can lose 50% capacitance from 25°C to 85°C, shifting poles by 30% and potentially destabilizing control loops designed with room-temperature models. Robust design practices include: specifying 1% resistors and NP0/C0G capacitors for critical frequency-determining networks, adding tunable components for production alignment, and designing with ±20% pole location margins to accommodate component drift over product lifetime.

Free Engineering Calculators

Explore our complete library of free engineering and physics calculators.

🔗 Explore More Free Engineering Calculators

• Bjt Transistor Bias Interactive Calculator
• Op Amp Comparator Interactive Calculator
• 555 Timer Astable Monostable Interactive Calculator

Browse All Engineering Calculators →

About the Author

Robbie Dickson — Chief Engineer & Founder, FIRGELLI Automations

Robbie Dickson brings over two decades of engineering expertise to FIRGELLI Automations. With a distinguished career at Rolls-Royce, BMW, and Ford, he has deep expertise in mechanical systems, actuator technology, and precision engineering.

Wikipedia · Full Bio