Skip to content
Tutorials Center
My learning

Topics

Lesson 8 of 37 ยท Modelling and Transfer Functions

Block Diagram Reduction

Reduce complex control block diagrams with series, parallel and feedback rules and by moving junctions, then check the result with polynomial code in Python.

Real systems have several loops and branches, so a block diagram may look tangled. Block diagram reduction simplifies it, one rule at a time, into a single transfer function from input to output. The rules are just algebra on G(s), and they are easy to mechanise in code, so we do both.

The three basic rules

Series (cascade). Blocks in a chain multiply: G = G1 G2.

Parallel. Blocks that receive the same input and are summed add: G = G1 + G2 (or G1 - G2 if one is subtracted).

Feedback. A forward block G with feedback block H:

  • negative feedback: G / (1 + G H)
  • positive feedback: G / (1 - G H)
flowchart LR
    A["G1"] --> B["G2"]
    B --> C["G3 in parallel with G4"]
    C --> D["negative feedback H"]

A typical reduction order: combine series, then parallel, then close loops from the innermost outwards.

Moving blocks past junctions

Sometimes no rule applies directly because loops overlap. You then move a summing junction or take-off point past a block, keeping the equations unchanged.

Move What you must do
Take-off point forward past block G Add a block G in the branch
Take-off point backward past block G Add a block 1/G in the branch
Summing junction forward past block G Add G in the other branch
Summing junction backward past block G Add 1/G in the other branch

The goal of each move is to separate a nested loop so that the standard feedback rule applies. Always check that the overall relationship is unchanged.

Reduction toolbox

Series Multiply the blocks.
Parallel Add the blocks.
Feedback G over one plus G H.
Move junctions Add G or 1/G to keep the equations.

Work from the innermost loop outwards.

1 / 2

Automating the rules

A transfer function is a pair (num, den). The rules become short functions on polynomials.

Python Block-diagram algebra as functions

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

The printed series result has numerator [2] and denominator s^2 + 4s + 3; the parallel result has numerator 3s + 5; the closed loop of 1/(s+1) is 1/(s+2), so the printed denominator is [1, 2].

A worked example

Consider a system with an inner loop around G2, in series with G1, and an outer loop with feedback H:

flowchart LR
    R["R"] --> S1(("+ -"))
    S1 --> G1["G1"]
    G1 --> S2(("+ -"))
    S2 --> G2["G2"]
    G2 --> Y["Y"]
    Y -- "H2" --> S2
    Y -- "H1" --> S1

Two nested loops: an inner loop around G2 (feedback H2) and an outer loop around everything (feedback H1).

Reduce in three steps. (1) Close the inner loop: Gin = G2 / (1 + G2 H2). (2) Put it in series with G1: G1 Gin. (3) Close the outer loop: T = G1 Gin / (1 + G1 Gin H1). With code:

Python Reducing nested loops

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

The inner loop gives 1/(s + 5), the forward path is 4/(s (s+5)), and the overall T = 4 / (s^2 + 5 s + 4), with poles at -1 and -4 and DC gain 1. The integrator in the forward path is why the DC gain is exactly 1 (see Steady-State Error and System Type).

Multiple inputs and superposition

If a system has several inputs, such as a reference and a disturbance, find each transfer function to the output with the other inputs set to zero, then add the responses: Y = T_r R + T_d D. This works because the system is linear.

Check your understanding

Quick check ยท 4 questions

of 4 answered You scored / 4 ยท perfect!

  1. 1. Blocks G1 and G2 are in series. What is the combined transfer function?

    Series blocks multiply because the output of the first is the input of the second.
  2. 2. Negative feedback with forward block G and feedback block H gives

    Y = G (R - H Y), so Y (1 + G H) = G R.
  3. 3. To move a take-off point backward past a block G, what do you add to the branch?

    The branch must still carry the signal it had before, so it needs to undo the block it jumped over: 1/G.
  4. 4. With two inputs R and D, how do you find the effect of D on Y?

    By superposition in a linear system, set the other input to zero and reduce the diagram for the input of interest.

Summary

  • Reduce diagrams with series (multiply), parallel (add) and feedback (G/(1 + G H)) rules.
  • Move summing junctions and take-off points by adding G or 1/G to keep the equations equal.
  • Work from the innermost loop outwards.
  • Code can automate the polynomial algebra and check a result numerically.
  • Use superposition for several inputs.

Next: Signal Flow Graphs and Mason's Rule