Embedded Files
Thermal Process Trainer: VTSU
Thermal Process Trainer: VTSU
This website is for the 2025/2026 senior projects class for Vermont State University.
Created by Richard Page.
The client is Jose Luzardo.
This is the Thermal Process Trainer. This product will be used by Vermont State University in their Control System II class, which Mr. Luzardo teaches. The goal of this Trainer is to give the students a hands-on experience creating PID's and tuning them.
This is the Thermal Process Trainer. This product will be used by Vermont State University in their Control System II class, which Mr. Luzardo teaches. The goal of this Trainer is to give the students a hands-on experience creating PID's and tuning them.
What is a Control System?
Control systems are ways to regulate/automate the behavior of a given device or system, often to achieve a desired state using feedback loops to adjust the performance.
What is a PID?
A PID Stands for Proportional, Integral, Derivative.
Proportional:
The Proportional value produces an output value that is proportional to the current error value of the system.
Integral:
The Integral accounts for past values of the error and integrates them over time to eliminate residual steady - state errors.
Derivative:
The derivative measures the rate of change of the error, predicting future errors and improving system stability.
How I'm going about making this product work
Cascade PI:
The airflow subsystem uses a cascade PI controller to regulate airflow through the system.
A cascade controller uses two PI control loops: an outer airflow loop and an inner fan speed loop.
This improves stability and allows the system to reject disturbances more effectively.
The system uses a four-pin Noctua fan, which includes a PWM control pin and a tachometer feedback pin.
The fan is controlled by an STM32F401RE microcontroller, which sends a PWM signal of approximately 25 kHz to control the fan speed.
The fan speed can be adjusted to match the airflow demand.
The outer PI controller compares the desired airflow setpoint to the measured airflow from the mass airflow sensor.
Based on this error, it calculates the required fan RPM. That RPM command is then sent to the inner PI controller.
The inner PI controller compares the commanded RPM to the actual RPM measured from the fan tachometer.
It then adjusts the PWM duty cycle to make the fan speed match the required value.
This control structure allows the fan to automatically ramp up or down depending on airflow conditions.
For example, if a hand or object blocks part of the airflow path, the airflow sensor detects a drop in airflow.
The controller responds by increasing the fan RPM to pull more air through the system and reduce the airflow error.
Both PI controllers contain proportional and integral gains that can be tuned for better performance.
The proportional term responds to the current error, while the integral term corrects long-term steady-state error.
Together, the cascade PI controller makes the airflow subsystem more stable, accurate, and responsive.
Pure Dead Time Delay PID:
The heater subsystem has a noticeable time delay because the heat generated by the coil does not reach the thermocouples instantly.
The heat must travel through the airflow path before the sensors detect a temperature change.
This creates a transport lag in the system, making the heater more difficult to control.
To regulate the temperature, the system uses a PID controller.
The PID controller adjusts the heater output based on the difference between the desired setpoint and the measured temperature.
This helps reduce overshoot and allows the system to maintain a stable temperature.
The heater system includes a heater coil, solid-state relay, STM32F401RE microcontroller, thermocouples, and thermocouple amplifiers. The STM32 sends a low-frequency PWM signal to the solid-state relay, which controls the power delivered to the heater coil.
The PWM signal is set to approximately 6 Hz, with the heater on for about 15 ms and off for about 30 ms.
Temperature data is collected by the thermocouples and sent back to the STM32.
The microcontroller then transmits the live data through UART so the response can be displayed and analyzed in MATLAB.
During testing, the system took approximately 160 seconds to reach the temperature setpoint and about 150 seconds to settle into steady-state operation.
The slow response is caused by the heater’s thermal mass, airflow delay, and sensor location.
The PID controller helps compensate for this delay and improves the overall stability of the heater subsystem.
This is a simulation of the heater coil regulating itself with the PID
Electrical schematic
SolidWorks Drawing
Heater Box
Heater Box
This was 3d printed with PETG
Airflow Pipe
Airflow Pipe
This was 3d printed with PETG
Mass Airflow Housing
Mass Airflow Housing
This was 3d printed with PETG
Box Fork
Box Fork
This was 3d printed with PETG
Pipe Fork
Pipe Fork
This was 3d printed with PETG
Base
Base
This was 3d printed with PLA
Assembly
Assembly
This is a video demo of the PI controller for the airflow system
This is a video demo of the P controller for the heater system
Hardware
Heater Coil:
Heater Coil:
K Type Thermocouple:
K Type Thermocouple:
K Type Amplifier MAX31855:
K Type Amplifier MAX31855:
Noctua NF - A8 PWM Fan:
Noctua NF - A8 PWM Fan:
Mass Air Flow Sensor:
Mass Air Flow Sensor:
FQFER Solid State Relay:
FQFER Solid State Relay:
Click to go to VTeLABS to learn more
Page updated
Google Sites
Report abuse