Feedback and Control in Integrated Systems

Overview

Feedback is the mechanism by which a system monitors its own output and uses that information to adjust its operation. Control is the use of feedback (or pre-set logic) to keep the system output at a desired value or sequence. Together, feedback and control are what make integrated systems intelligent, reliable, and safe.

KEY TAKEAWAY: Without feedback, a system cannot correct errors or adapt to changing conditions. With feedback (closed-loop control), the system continuously compares its actual output to the desired output and adjusts automatically.

Open-Loop vs. Closed-Loop Control

Open-Loop Control

In an open-loop system, there is no feedback. The controller sends a command to the actuator and assumes the output is correct without checking.

[Input command] → [Controller] → [Actuator] → [Output]

Characteristics:
- Simple and inexpensive to implement
- Cannot correct for disturbances or component variation
- Suitable only when the output does not need to be precisely controlled

Example: A toaster set for 2 minutes — it heats for exactly 2 minutes regardless of whether the bread is already toasted or undercooked.

Example: A traffic light on a fixed timer — lights change on schedule regardless of actual traffic volume.

APPLICATION: Open-loop control is acceptable when the process is predictable and repeatable, and when variations in output are tolerable.

Closed-Loop Control

In a closed-loop system, a sensor measures the actual output and feeds this information back to the controller, which compares it to the desired output (setpoint) and adjusts the actuator accordingly.

[Setpoint] → [Controller] → [Actuator] → [Output]
                  ↑
                  └──── [Sensor/Feedback] ────┘

Key terms:
- Setpoint (reference): The desired output value
- Actual output (process variable): What the sensor measures
- Error: Difference between setpoint and actual output ($\text{error} = \text{setpoint} - \text{actual}$)
- Corrective action: What the controller does to reduce the error

Example — Room thermostat:
- Setpoint: 22°C
- Sensor: Thermistor measures room temperature
- Controller: Compares measured temperature to setpoint
- If actual < 22°C: heater turns ON (error = positive, corrective action = heat)
- If actual ≥ 22°C: heater turns OFF (error = zero or negative, no heating needed)

VCAA FOCUS: Be able to identify, in a described or diagrammed system, whether it is open-loop or closed-loop. Justify your answer by identifying whether a sensor measures the output and whether that measurement is used to adjust the actuator.

The Feedback Loop: Step by Step

1. Sense: Sensor measures the actual system output
2. Compare: Controller compares actual output to setpoint — calculates error
3. Decide: Controller determines the corrective action needed
4. Act: Actuator adjusts to reduce the error
5. Repeat: Cycle continues continuously (or at regular intervals)

Example — Speed-controlled DC motor:

Feedback in Mechanical Systems

Feedback is not limited to electronic systems. Mechanical feedback also exists:

• Centrifugal governor (steam engine): As engine speed increases, rotating flyweights rise and mechanically reduce the steam valve opening, limiting speed
• Pressure relief valve: When pressure exceeds a threshold, the valve opens to release pressure, maintaining safe levels
• Spring-loaded mechanism: Provides proportional restoring force as displacement increases

STUDY HINT: When identifying feedback in any system, look for: (1) a sensor or measuring device, (2) a comparison step, and (3) a corrective action. If any of these is missing, it is open-loop.

Stability and Hunting

A potential problem in closed-loop systems is hunting (oscillation) — where the system overshoots the setpoint, then overcorrects, creating repeated swings above and below the desired value.

Causes of hunting:
- Response delay between sensing and corrective action
- Over-aggressive correction (too much gain)

Engineering solutions:
- Add a dead band: Only correct when error exceeds a threshold (e.g. ±1°C)
- Use proportional control: Correction is proportional to error size (large error → large correction; small error → small correction)

REMEMBER: A well-designed feedback system reaches the setpoint smoothly and maintains it. Poor design causes oscillation (hunting) or sluggish response (slow to reach setpoint).

Summary: Open-Loop vs. Closed-Loop

COMMON MISTAKE: Students sometimes describe a system with a sensor as closed-loop without checking whether the sensor output actually feeds back to control the actuator. A sensor that only triggers an alarm (not a corrective action) does not make the system closed-loop.