Skip to content
ComputeVerse.ai

Control SystemsDesign workbenchPreliminary engineering

Transfer Function Analyzer

Enter H(s) as polynomial coefficients and get poles, zeros, gain, characteristic equation, system order and type, and a stability verdict — with the pole-zero map and a live step-response preview.

About this calculator

The transfer function is the identity card of a linear time-invariant system: everything about its dynamics — how fast it responds, whether it rings, whether it is stable at all — is encoded in the roots of two polynomials. This analyzer takes H(s) = N(s)/D(s) as you would write it on paper and produces the complete record: poles (roots of the characteristic equation D(s) = 0), zeros (roots of N(s)), the leading-coefficient gain, DC gain H(0), system order, and system type (the number of pure integrators).

Stability is read directly from the pole locations: every pole strictly in the left half of the s-plane means every natural mode decays and the system is BIBO stable. A single right-half-plane pole is one growing exponential — no input choice can hide it. Simple poles exactly on the imaginary axis sit on the boundary: a sustained oscillation or a constant offset that never decays, classified here as marginally stable, and repeated poles on the axis grow without bound.

The pole-zero map shows this geometry at a glance, and the step-response preview shows what those root locations mean in time — change any coefficient and both recompute live. When the analysis points somewhere deeper, the related tools take over: the pole-zero analyzer for dominant-pair damping, the time-response tool for performance metrics, the Bode and Nyquist plots for frequency-domain design, and the stability checker for the Routh–Hurwitz table.

Everything here treats the continuous-time Laplace-domain H(s) of a proper system (numerator degree ≤ denominator degree); the entry form enforces properness by construction.

Design notes & common mistakes

  • Poles decide stability; zeros never do. A right-half-plane ZERO is not instability — it is a non-minimum-phase system whose step response starts in the wrong direction and whose achievable bandwidth is fundamentally limited.
  • A near pole-zero cancellation looks clean algebraically but the hidden mode is still there physically — if it is slow or unstable, the practical behavior will not match the reduced model.
  • The classic sign error: writing the characteristic polynomial with a negative coefficient somewhere and not noticing. Any sign change among the denominator coefficients guarantees at least one RHP root.
  • Entering a measured plant? Coefficients spanning many orders of magnitude are poorly conditioned for root finding — rescale time (s → s/ω₀) first so coefficients stay comparable.

Assumptions

  • Continuous-time, linear, time-invariant SISO system in the Laplace s-domain.
  • Proper transfer function (numerator degree ≤ denominator degree) — enforced by the entry form.
  • Coefficients are exact as entered; no uncertainty or parameter tolerance is propagated.
  • Poles and zeros of degree ≥ 3 polynomials are computed numerically (Durand–Kerner iteration) and are accurate to far better than display precision for the supported order range (≤ 4).

When to use this calculator

Appropriate for

  • Analyzing a continuous-time, linear, time-invariant SISO transfer function: poles, zeros, gain, order, type, and stability
  • Confirming a hand-derived transfer function parses to the poles and stability you expect
  • Teaching how pole and zero locations shape response and stability

Not suitable for

  • Nonlinear, time-varying, or multi-input/multi-output systems, which this s-domain model does not represent
  • Discrete-time (z-domain) systems or plants with transport delay, unless the delay is first approximated by a rational term
  • Certifying the stability of a real plant without validating the model against measured behavior

What this calculator does not cover

  • Denominator order is capped at 4 in this tool. Higher-order systems are planned together with the root-locus tool; the underlying engine already computes them but entry and worked-solution display are tuned for n ≤ 4.
  • Coefficient entry only — factored zero-pole-gain entry is not exposed in the UI yet (the ζ/ωn parametric form is available on the second-order calculator).
  • Continuous-time only: no discrete-time (z-domain) analysis, no transport delay e^(−sT).
  • Exact pole-zero cancellations are reported as coincident roots, not cancelled — deciding whether a cancellation is physically meaningful requires knowing where the model came from.
  • As with every calculator on this site: results are preliminary and educational, are not verified for any specific installation, and must be reviewed against the applicable code edition and stamped by a licensed Professional Engineer before real-world use.

Frequently asked questions

What makes a transfer function stable?

BIBO stability requires every pole — every root of the denominator — to lie strictly in the left half of the s-plane, so all natural modes decay. One right-half-plane pole is one growing exponential, and no input shaping can remove it; simple imaginary-axis poles are the marginal boundary case.

What is the difference between system order and system type?

Order is the denominator degree — how many states the system carries. Type counts only the poles at the origin (pure integrators), which is what sets steady-state tracking ability in a unity-feedback loop: type 1 tracks a step exactly, type 2 also tracks a ramp.

Why is my DC gain shown as unbounded?

The denominator has a root at s = 0 (a pure integrator), so H(0) = N(0)/0. Physically, a constant input is integrated forever and the steady output grows without limit — the honest answer is 'unbounded', not a large number.

Do zeros affect stability?

No — stability is decided entirely by the poles. Zeros shape the response: they redistribute mode amplitudes, can add overshoot or undershoot, and a right-half-plane zero marks a non-minimum-phase system with an initial wrong-way response, but a system with all poles in the LHP is stable regardless of its zeros.

References

  • Nise, N. S., Control Systems Engineering, 8th ed., Ch. 2 (transfer functions) and Ch. 6 (stability via pole locations)
  • Ogata, K., Modern Control Engineering, 5th ed., Ch. 2 (mathematical modeling, transfer functions)
  • Franklin, G., Powell, J. & Emami-Naeini, A., Feedback Control of Dynamic Systems, 8th ed., Ch. 3 (dynamic response, poles and zeros)

Related calculators

Get the ComputeVerse.ai quick reference

A free engineering quick-reference PDF, plus a note when new calculators launch across the library — one or two emails a month, nothing else.