A1.3.6 Describe the use of the control system components. (HL only)
• The input, process, output, and feedback mechanism (open-loop, closed-loop)
• Controller, sensors, actuators, transducers and control algorithm
The big idea
A control system is an engineered loop that integrates physical processes with computational logic to regulate the behavior of a system.
Sensing and transduction:
A sensor, incorporating a transducer, detects a physical quantity such as temperature, position, or pressure and converts it into a proportional electrical signal. This signal is typically continuous (analog) and represents the measured physical state of the system.
Digitization:
An analog-to-Digital Converter (ADC) samples the analog electrical signal produced by the transducer and converts it into digital data that can be processed by computing hardware.
Processing:
A controller (such as a PLC (Programmable Logic Controller) or microprocessor) receives the digital measurement data and compares it to a desired reference value (set-point) to compute an error signal. Using a control algorithm (for example, PID (Proportional–Integral–Derivative)), the controller determines the appropriate corrective action.
Actuation:
The controller sends a command signal to an actuator, which converts the electrical control signal into physical energy—such as torque, heat, or fluid flow—to influence the system.
Feedback:
In closed-loop systems, the system output is continuously measured and fed back to the controller, enabling automatic correction for disturbances. In open-loop systems, feedback is absent, so the system cannot adjust its behavior if conditions change.
1 Functional flow: input → process → output (+ feedback)
| Stage | Purpose | Typical data / signal |
|---|---|---|
| Input | Provide the reference (set-point) and the current state of the plant being controlled. | Operator-selected temperature, altitude, valve position and sensor readings (voltage, pressure…). |
| Process | Compute the control effort required to make the plant follow the reference. | Digital controller executing a PID, state-space, or model-predictive algorithm. |
| Output | Apply the control effort to the plant through energy-conversion hardware. | PWM duty cycle to a motor driver, current to a solenoid, duty request to a power stage. |
| Feedback | Route a representation of the actual output back to the process stage so that errors are corrected continuously. | ADC samples from sensors, encoder counts, flow-meter pulses. |
In an open-loop system the feedback arrow is absent; the controller assumes the plant obeys commands perfectly.
In a closed-loop (feedback) system the controller continuously adjusts its output based on measured error. Closed-loop control is therefore robust to disturbances and modelling inaccuracies, at the cost of extra complexity and, in digital systems, latency and sampling-rate constraints.
2 Principal physical/logic components
| Component | What it does | Key engineering concerns |
|---|---|---|
| Controller (control unit) | Executes the control algorithm at a defined cycle time (sample period). May be firmware on a microcontroller, a PLC ladder program, or an RTOS task set. | Numerical precision, execution determinism, watchdog coverage, worst-case execution time. |
| Sensors | Generate electrical signals proportional to physical variables (temperature, angular rate, light intensity…). Usually incorporate conditioning (amplification, filtering) and digitisation (ADC). | Resolution, accuracy, bandwidth, noise floor, drift, calibration interval. |
| Transducers | Generic term for devices that convert any energy form to another. A sensor is an input transducer; an actuator is an output transducer. | Linearity, hysteresis, conversion efficiency, environmental ruggedness. |
| Actuators | Convert the controller’s low-power command into mechanical, hydraulic, pneumatic or electromagnetic work. Examples: DC motor, stepper, servo valve, piezo stack. | Stall torque/force, step response, back-EMF isolation, power and thermal limits. |
| Control algorithm | Mathematical rule that maps the error (set-point − measurement) and possibly its history to an actuator command. PID (proportional-integral-derivative), lead–lag, full-state feedback, fuzzy logic and adaptive MPC are common families. | Stability (root-locus, Nyquist), steady-state error, rise time, overshoot, robustness to parameter drift and sampling jitter. |
3 Open-loop versus closed-loop: comparative view
| Criterion | Open-loop | Closed-loop |
|---|---|---|
| Disturbance rejection Disturbance rejection refers to how effectively a control system compensates for external changes in order to keep the output close to the set-point. | None—any load change or sensor drift propagates uncorrected. | High—controller compensates for disturbances within its bandwidth. |
| Stability analysis Stability analysis evaluates whether a control system responds to disturbances in a controlled and predictable manner. | Trivial—no feedback path to oscillate. | Must guarantee phase-margin and gain-margin; requires control-theory design. |
| Complexity & cost Increasing system complexity improves performance and disturbance rejection but increases cost and implementation difficulty. | Minimal: timer or simple lookup table suffices. | Additional sensors, ADCs, real-time firmware, tuning overhead. |
| Example | Microwave oven that runs the magnetron for a fixed time based on weight entered by user. | Domestic thermostat cycling an HVAC compressor based on measured room temperature relative to set-point. |
4 Putting it together: an illustrative cycle (motor-speed control)
- Sensor / transducer – an optical encoder reports 1500 RPM to the controller every 1 ms.
- Input assembly – the operator set-point is 2000 RPM; the error is therefore +500 RPM.
- Controller – a discrete PID computes a voltage command
- Actuator – a MOSFET H-bridge receives a new PWM duty of 78 %.
- Plant (motor) – torque increases; speed rises toward the target.
- Feedback – updated encoder count enters the next control cycle.
This loop executes indefinitely—tens of thousands of times per second in a high-performance servo—maintaining the desired speed despite load torque fluctuations and supply-voltage sag.
Some vocabulary help for you:
PID stands for Proportional–Integral–Derivative. It is a widely used control algorithm that computes how strongly a system should react to an error.
Proportional (P)
Responds to the current error.
Larger error → larger corrective action.Integral (I)
Responds to the accumulated error over time.
Eliminates steady-state error (e.g. motor always stopping short of the target).Derivative (D)
Responds to the rate of change of the error.
Predicts future behavior and damps oscillation.
An H-bridge is an actuator driver circuit that allows a controller to control the direction and speed of a DC motor.
It uses four switching elements (often MOSFETs)
It can:
drive the motor forward
drive it in reverse
brake it
let it coast
The name comes from the circuit layout, which resembles the letter H, with the motor forming the crossbar.
PWM stands for Pulse-Width Modulation.
It is a method of controlling effective power delivered to a device by rapidly switching a digital signal on and off.
The duty cycle is the percentage of time the signal is ON
0% → motor off
50% → half power
100% → full power
PWM allows a digital controller to control an analog effect (motor voltage / speed) efficiently.
Key takeaways
- Input–process–output–feedback formalises how data and energy move through any control loop.
- Controllers, sensors, actuators, and transducers are the physical and computational elements that realise that loop.
- Open-loop systems are simple but fragile; closed-loop systems add sensors and real-time computation to achieve accuracy and disturbance immunity.
- Control algorithms translate error into corrective action; their design dictates stability, responsiveness, and robustness of the entire system.