Core course · Control Systems

Control Systems

How machines keep themselves on target — from a room thermostat to a solar inverter's MPPT. Transfer functions, time response, stability and PID control, with worked examples, a step-response calculator and a quiz. Useful for B.Tech and GATE preparation.

Join the live course
Lesson 1

Open-loop and closed-loop systems

A control system makes an output follow a desired input (the reference or set point).

Σ +− Controllere.g. PID Plant / processmotor, heater, inverter Sensor R(s)E(s)C(s)Feedback
Closed loop: the sensor measures the output, and the controller acts on the error = reference − feedback.

Open loop

  • No feedback; the output is not measured
  • Simple and cheap, but cannot correct disturbances
  • Examples: a timer-based washing machine, a toaster, traffic lights on a fixed cycle

Closed loop (feedback)

  • Measures the output and corrects the error
  • Accurate, rejects disturbances, less sensitive to parameter changes
  • Can become unstable if designed badly
  • Examples: AC thermostat, cruise control, solar tracker, inverter voltage control
Lesson 2

Transfer function, poles and zeros

  1. Definition

    The transfer function is the Laplace transform of the output divided by the Laplace transform of the input, with all initial conditions zero.

    G(s) = C(s) ÷ R(s)
  2. Poles and zeros

    Zeros are values of s that make G(s) = 0 (roots of the numerator). Poles make G(s) infinite (roots of the denominator). The poles decide stability and the shape of the response.

    Example: G(s) = (s + 2) ÷ (s² + 4s + 3) has a zero at −2 and poles at −1 and −3.

  3. Order and type

    The order is the highest power of s in the denominator. The type is the number of poles at the origin (pure integrators, 1/s) in the open-loop G(s)H(s).

  4. Common Laplace pairs

    unit step → 1/s ramp t → 1/s² e−at → 1/(s + a) sin ωt → ω/(s² + ω²)
Lesson 3

Block diagram reduction

ConnectionEquivalent
Series (cascade)G₁ · G₂
ParallelG₁ ± G₂
Negative feedbackG ÷ (1 + G·H)
Positive feedbackG ÷ (1 − G·H)
Unity negative feedbackG ÷ (1 + G)

The characteristic equation is 1 + G(s)H(s) = 0; its roots are the closed-loop poles. For complex diagrams, Mason's gain formula gives the overall gain from a signal flow graph.

Lesson 4

Time response of first- and second-order systems

First order

G(s) = 1 ÷ (τs + 1) c(t) = 1 − e−t/τ

Reaches 63.2% of the final value in one time constant τ and settles (within 2%) in about 4τ. No overshoot.

Second order

G(s) = ωn² ÷ (s² + 2ζωns + ωn²)

ωn = natural frequency, ζ = damping ratio. ζ decides the shape of the response.

Damping ratio ζTypeStep response
ζ = 0UndampedOscillates forever
0 < ζ < 1UnderdampedOvershoots, then settles with decaying oscillation
ζ = 1Critically dampedFastest response with no overshoot
ζ > 1OverdampedSlow, no overshoot

Underdamped specifications

SpecificationFormula
Damped frequencyωd = ωn√(1 − ζ²)
Rise time (0–100%)tr = (π − θ) ÷ ωd, where θ = cos⁻¹ζ
Peak timetp = π ÷ ωd
Peak overshootMp = e−ζπ/√(1−ζ²) × 100%
Settling timets ≈ 4 ÷ (ζωn) (2% band), 3 ÷ (ζωn) (5% band)

Worked example

Unity feedback with G(s) = 25 ÷ (s(s + 6)). Closed loop: 25 ÷ (s² + 6s + 25), so ωn = 5 rad/s and 2ζωn = 6, giving ζ = 0.6.

ωd = 5 × 0.8 = 4 rad/s · Mp = e−0.6π/0.8 ≈ 9.5% · tp = π ÷ 4 ≈ 0.785 s · tr = (π − 0.927) ÷ 4 ≈ 0.554 s · ts (2%) = 4 ÷ 3 ≈ 1.33 s

Lesson 5

Steady-state error and error constants

The error left after the transient dies out. It depends on the input type and the system type (number of integrators).

InputError constantType 0Type 1Type 2
StepKp = lims→0 GH1 ÷ (1 + Kp)00
RampKv = lims→0 s·GH∞1 ÷ Kv0
ParabolicKa = lims→0 s²·GH∞∞1 ÷ Ka

Example: G(s) = 25 ÷ (s(s + 6)) is type 1, so the step error is 0. Kv = 25 ÷ 6 ≈ 4.17, so the ramp error is 1 ÷ 4.17 ≈ 0.24.

Lesson 6

Stability and the Routh–Hurwitz criterion

What stability means

A system is stable if every bounded input gives a bounded output. For a linear system this means all closed-loop poles lie in the left half of the s-plane (negative real parts). Poles on the imaginary axis give sustained oscillation (marginally stable).

Routh–Hurwitz test

Build the Routh array from the coefficients of the characteristic equation. The number of sign changes in the first column equals the number of poles in the right half-plane. No sign changes means stable.

Example: s³ + 6s² + 11s + 6 = 0

s³111
s²66
s¹(6×11 − 1×6) ÷ 6 = 100
s⁰6

First column 1, 6, 10, 6: all positive, no sign changes, so the system is stable. (The roots are −1, −2 and −3.)

Root locus shows how the closed-loop poles move as the gain K increases from 0 to ∞. The gain at which the locus crosses the imaginary axis is the limit of stability.

Lesson 7

Frequency response, Bode plots and margins

  1. Bode plot

    Two graphs against log frequency: gain in dB (20 log|G(jω)|) and phase in degrees. Each pole adds a −20 dB/decade slope and up to −90° of phase; each zero adds +20 dB/decade and up to +90°.

  2. Gain margin (GM)

    How much the gain can rise before instability, measured at the phase crossover frequency where the phase is −180°.

    GM (dB) = −20 log |G(jωpc)H(jωpc)|
  3. Phase margin (PM)

    How much extra phase lag can be tolerated, measured at the gain crossover frequency where the gain is 0 dB.

    PM = 180° + ∠G(jωgc)H(jωgc)
  4. Rule of thumb

    A closed-loop system is stable when both GM and PM are positive. Good designs aim for GM ≥ 6 dB and PM of about 45–60°. Nyquist plots give the same information for more complex cases.

Lesson 8

PID controllers

The most widely used controller in industry. It combines three actions on the error e(t).

u(t) = Kp·e(t) + Ki∫e(t)dt + Kd·de(t)/dt C(s) = Kp + Ki/s + Kds
ActionEffectDrawback
Proportional (P)Faster response, smaller errorLeaves a steady-state error; too much causes overshoot
Integral (I)Removes steady-state errorSlower, more overshoot; can cause "windup"
Derivative (D)Reduces overshoot, adds dampingAmplifies sensor noise

Tuning in practice

Start with P only and raise Kp until the response is quick with modest overshoot. Add I to remove the remaining error, then a little D if the overshoot is too large. The Ziegler–Nichols method gives starting values from the gain and period at which the loop just oscillates.

Application

Control systems inside a solar plant

MPPT

The inverter continuously adjusts the panel voltage to stay at the maximum power point as sunlight and temperature change. Perturb-and-observe is a simple feedback loop.

Grid synchronisation (PLL)

A phase-locked loop tracks the grid's voltage, frequency and phase so the inverter can inject current in step with the grid.

Current and voltage control

Inner PI loops regulate the inverter output current; outer loops control DC-link voltage and power export limits.

Battery and sun tracking

Charge controllers regulate battery charging stages. Single-axis trackers use closed-loop position control to follow the sun.

Practice

Second-order step response calculator

Change ωn and ζ to see the specifications and the response curve update instantly.

System

rad/s
0 = undamped, 1 = critical
10 time (s)

Results

Response type
Peak overshoot
Damped freq. ωd
Rise time tr
Peak time tp
Settling time (2%)
Study material

Quick formula sheet

Revise before your exam. Use "Print / save notes as PDF" at the top to keep a copy.

TopicFormula / rule
Transfer functionG(s) = C(s) ÷ R(s), zero initial conditions
Negative feedbackT(s) = G ÷ (1 + GH)
Characteristic equation1 + G(s)H(s) = 0
First orderc(t) = 1 − e−t/τ, ts ≈ 4τ
Second orderωn² ÷ (s² + 2ζωns + ωn²)
Damped frequencyωd = ωn√(1 − ζ²)
Peak overshootMp = e−ζπ/√(1−ζ²)
Peak / settling timetp = π/ωd, ts ≈ 4/(ζωn)
Error constantsKp = lim GH, Kv = lim sGH, Ka = lim s²GH (s → 0)
Steady-state errorstep 1/(1 + Kp), ramp 1/Kv, parabola 1/Ka
StabilityAll closed-loop poles in the left half-plane; no sign changes in the Routh first column
MarginsGM at −180° phase, PM at 0 dB gain; aim GM ≥ 6 dB, PM 45–60°
PIDC(s) = Kp + Ki/s + Kds
Track your progress

Control systems topics checklist

Test yourself

Control systems quiz

1. Which of these is a closed-loop system?

The thermostat measures room temperature and switches the compressor to correct the error — that is feedback.

2. The closed-loop transfer function of G with negative feedback H is…

For negative feedback: C/R = G ÷ (1 + GH).

3. A system has s² + 4s + 16 as its denominator. What is ζ?

ωn² = 16, so ωn = 4. 2ζωn = 4 gives ζ = 0.5.

4. A damping ratio of ζ = 1 gives a response that is…

ζ = 1 is the fastest response without overshoot.

5. A type 1 system with a step input has a steady-state error of…

One integrator makes Kp infinite, so the step error 1/(1 + Kp) is zero.

6. In a Routh array, the first column is 1, 3, −2, 5. How many poles are in the right half-plane?

Signs go +, +, −, +: two sign changes (3 → −2 and −2 → 5), so two right-half-plane poles. The system is unstable.

7. Which PID action removes steady-state error?

The integral keeps growing while any error remains, pushing the output until the error is zero.

8. Phase margin is measured at the frequency where…

PM is read at the gain crossover (|GH| = 1, i.e. 0 dB). Gain margin is read at the −180° phase crossover.