Explain feedback, error and simple on/off control
A feedback controller compares a measured process value with a setpoint and changes an actuator to reduce the error. Define error consistently: here e = setpoint − measurement. A sensor measures the process, the controller decides an action, and the actuator changes it. Open-loop operation does not use the measured result to correct the command. A heater can add heat but cannot actively cool, which limits the actions available to the controller.
Worked practice example
A training chamber has a setpoint of 60 °C. To reduce rapid switching, a controller uses hysteresis: turn the heater ON at or below 58 °C; turn it OFF at or above 62 °C; between those limits retain the previous state. Starting with the heater OFF, determine its state at measurements 57, 59, 61, 63 and 60 °C.
| Step | Working |
|---|---|
| 57 °C | At or below 58: ON. Error = 60 − 57 = +3 °C. |
| 59 °C | Inside the band: remain ON. |
| 61 °C | Inside the band: remain ON, even though error is now negative. |
| 63 °C | At or above 62: OFF. |
| 60 °C | Inside the band: remain OFF. |
Check your method
Hysteresis deliberately makes the command depend on the previous state as well as the current reading. It prevents noisy readings near one threshold causing continual switching. It does not guarantee zero temperature error: inertia, delay and disturbance still affect the process. Real equipment also needs independent protection appropriate to the application; this example is a control-logic exercise.
Try it yourself
With the same thresholds, the heater is ON at 61 °C. The next reading is 60.5 °C. Does it turn OFF? What happens at 62 °C?
Show the answer and method
At 60.5 °C it remains ON because the reading is within the band. At 62 °C it turns OFF. A single comparison against the 60 °C setpoint would implement a different controller.
This written practice example was prepared for Jemshah revision. It is additional practice, not an official examination question or a transcript of the video.