Skip to main content
Subject Specific 3% exam weight

Topic 17

Part of the GATE study roadmap. Subject Specific topic subjec-017 of Subject Specific.

Control Systems — Transfer Function and Block Diagrams

🟢 Lite — Quick Review (1h–1d)

Rapid summary for last-minute revision before your exam.

GATE Weightage: ~6–10 marks/year (Electrical/Instrumentation branches); block diagram reduction and Mason’s gain formula are perennial favorites.


Transfer Function G(s) = Laplace of output / Laplace of input (with zero initial conditions)

  • Standard form: G(s) = K·(s+z₁)(s+z₂)… / (s+p₁)(s+p₂)…
  • Poles = roots of denominator; Zeros = roots of numerator
  • Order = degree of denominator (highest power of s)

Block Diagram Algebra

Block in seriesG₁·s G₂ = G₁G₂
Block in parallel (summing point)G₁ ± G₂
Feedback loopG/(1 ± GH) (negative/positive feedback)
Moving a pickoff pointIntroduce/remove equal G in path
Moving a summing pointAccount for new paths created

Mason’s Gain Formula:

M = (1/Δ) · Σ P_k · Δ_k

Where: P_k = kth forward path gain, Δ = 1 – Σ(L₁) + Σ(L₂) – Σ(L₃) + … (determinant), Δ_k = Δ with loops touching P_k removed.

⚡ GATE Tips

  • Always reduce the inner loops first before simplifying the outer structure
  • For positive feedback, sign in denominator is (+GH); for negative feedback it’s (1 – GH) or (1 + GH) depending on convention
  • Mason’s formula is faster than block reduction for complex diagrams — learn it well

🟡 Standard — Regular Study (2d–2mo)

Standard content for students with a few days to months.

Transfer Function — Definition and Properties

The transfer function G(s) of a linear time-invariant (LTI) system is defined as:

G(s) = Y(s) / R(s) with all initial conditions = 0

Where Y(s) = Laplace transform of output y(t), R(s) = Laplace transform of input r(t).

Properties of Transfer Functions

  1. System order = degree of denominator polynomial (highest power of s)
  2. Proper transfer function: degree of numerator ≤ degree of denominator
  3. Strictly proper: degree of numerator < degree of denominator
  4. Poles = values of s that make G(s) → ∞ (denominator roots)
  5. Zeros = values of s that make G(s) → 0 (numerator roots)
  6. dc gain (static gain) = G(s) evaluated at s = 0

Standard Transfer Function Forms

First-order system: G(s) = K / (τs + 1) = ω_n / (s + ω_n) for normalized form

Second-order system (standard form):

G(s) = ω_n² / (s² + 2ζω_n s + ω_n²)

Where ζ = damping ratio, ω_n = natural frequency.

Type number = number of poles at s = 0 (integrators in forward path)

Block Diagram Reduction Techniques

Basic Building Blocks

ElementInputOutput
Summing junctionTake algebraic sum of inputsSingle output
Pickoff point (branch)One inputMultiple outputs (same magnitude)
Forward pathInput → System → Output
Feedback pathOutput → Feedback element H → Summing junction

Step-by-Step Reduction Method

  1. Identify all feedback loops (closed paths)
  2. Combine blocks in series — multiply transfer functions
  3. Combine blocks in parallel — add/subtract at summing points
  4. Reduce inner feedback loops using G/(1 ± GH)
  5. Move pickoff/summing points if needed to expose series/parallel combinations
  6. Repeat until single G(s) remains

Common mistake: Students forget that moving a summing point or pickoff creates new paths that must be accounted for. When moving a pickoff ahead, multiply the moved path by G; when moving a pickoff behind, multiply by 1/G.

Signal Flow Graph vs Block Diagram

Block diagrams show subsystems with arrows; signal flow graphs (SFG) show individual signals as nodes and systems as branches.

  • Node = signal variable
  • Branch = system (direction matters)
  • Forward path = path from input to output node without passing any node twice
  • Loop = closed path returning to starting node

Mason’s Gain Formula

For finding the overall transfer function from SFG directly:

T(s) = (Σ P_k Δ_k) / Δ

Where:

  • P_k = product of branch gains along the kth forward path
  • Δ (system determinant) = 1 – (sum of individual loop gains) + (sum of products of 2 nontouching loops) – (sum of products of 3 nontouching loops) + …
  • Δ_k = cofactor of P_k; value of Δ with loops touching P_k removed

Loop Types

Loop typeDescription
Individual loopL_i = product of branches in one closed loop
Two nontouching loopsProduct of 2 loops with no common node
Three nontouching loopsProduct of 3 pairwise nontouching loops

Key rule: Loops are nontouching only if they share NO common node.

Feedback and Sensitivity

Open-loop: G(s) (no feedback path)

Closed-loop (negative feedback): T(s) = G/(1 + GH)

Sensitivity S_M of parameter M: S_M = (dT/T) / (dM/M) = (d ln T) / (d ln M)

For closed-loop system:

  • Sensitivity to G variations: S_G = 1/(1 + GH) (reduced by feedback!)
  • Sensitivity to H variations: S_H = –GH/(1 + GH) (shows H affects system more)

** Steady-state error:** e_ss = 1/(1 + K_p) for step input, etc. (covered in Topic 18)

Example Problem — Block Diagram Reduction

Reduce to find C(s)/R(s):

R →[G1]→ (+) →[G2]→ C
         ↑        |
         | [H1]  |
         +-------+

Negative feedback with H1.

Solution:

  • Forward path: G1·G2
  • Feedback path: H1
  • Loop gain: G1·G2·H1
  • Closed-loop: C/R = G1·G2 / (1 + G1·G2·H1)

Example Problem — Mason’s Gain Formula

SFG with: input X, output Y. Forward path P₁ = G1·G2·G3. One loop L₁ = G1·G2·H1 (touches P₁). No nontouching loops.

Solution:

  • Δ = 1 – L₁ = 1 – G1·G2·H1
  • Δ₁ = 1 (loop touches P₁, so remove it → just 1)
  • T = P₁·Δ₁/Δ = G1·G2·G3 / (1 – G1·G2·H1)

🔴 Extended — Deep Study (3mo+)

Comprehensive coverage for students on a longer study timeline.

Mason’s Gain Formula — Detailed Derivation

The formula arises from solving simultaneous equations representing all node relationships. For a system with N nodes:

Y_i = Σ_j G_ij · Y_j (sum of all incoming signals)

Solving these using Cramer’s rule and Gaussian elimination yields the Mason’s formula structure.

Multiple Forward Paths

When k forward paths exist:

  • List all P₁, P₂, P₃… (all distinct paths from input to output)
  • Each P_k gets its own Δ_k
  • Total T = (P₁Δ₁ + P₂Δ₂ + …) / Δ

Nontouching Loop Groups

Δ computation rules:

  1. Start with 1
  2. Subtract sum of all individual loop gains (L₁, L₂, L₃…)
  3. Add sum of products of all pairs of nontouching loops
  4. Subtract sum of products of all triples of nontouching loops
  5. Continue alternating signs

GATE Trick: When asked “how many loops are nontouching?” — draw the SFG and physically check which loops share nodes.

Block Diagram to SFG Conversion

Converting a block diagram to SFG:

  1. Replace each block arrow with a branch (gain = block transfer function)
  2. Place a node at each summing junction output
  3. Place a node at each pickoff point
  4. Label all intermediate nodes
  5. Draw branches following the original signal flow

Special Transfer Function Forms

Time Delay Systems

G(s) = e^(–τs) — pure time delay of τ seconds

  • Cannot be expressed as rational polynomial
  • Approximated by Padé: e^(–τs) ≈ (1 – τs/2) / (1 + τs/2) (first-order Padé)

Systems with Time Delay in Feedback

When G(s) has e^(–τs), closed-loop characteristic equation becomes transcendental: 1 + G(s)H(s)e^(–τs) = 0. Requires numerical/approximate methods.

Minimum Phase vs Non-Minimum Phase

  • Minimum phase: All poles and zeros in left-half plane (LHP)
  • Non-minimum phase: Has a zero in RHP or time delay
  • All-pass systems have magnitude = 1 but phase shift

Controller Block Diagrams

In feedback control systems:

G(s) = G_c(s) · G_p(s) — controller × plant

H(s) = sensor/feedback dynamics

Standard form of closed-loop transfer function:

  • T(s) = G_c·G_p / (1 + G_c·G_p·H) for negative feedback
  • Characteristic equation: 1 + G_c·G_p·H = 0

System Type and Error Constants

Type number = number of poles at origin in forward path G(s):

TypePosition error K_pVelocity error K_vAcceleration error K_a
Type 0K_p = lim G(s)00
Type 1K_v = lim s·G(s)0
Type 2K_a = lim s²·G(s)

Higher type → better steady-state tracking, but harder to stabilize.

GATE Exam Strategy — Transfer Functions & Block Diagrams

Expected question types:

  1. Reduce block diagram → find C/R
  2. Draw SFG from given system equations, then apply Mason’s formula
  3. Find transfer function C/R from complex interconnected system
  4. Determine poles/zeros/gain from given G(s)
  5. Find type of system and steady-state error

Common GATE mistakes:

  • Confusing positive vs negative feedback sign in denominator
  • Forgetting that moving summing points changes paths
  • Miscounting nontouching loops in Mason’s Δ
  • Writing characteristic equation with wrong sign for feedback

Content adapted based on your selected roadmap duration. Switch tiers using the selector above.