To control the temperature of a process, we use a method called feedback control. This method involves sensing the actual temperature, figuring out how much it differs from the desired temperature, and then adjusting the process to reach the desired temperature. This same approach works for controlling other things like pressure or flow in a process.
What Is a Temperature Controller?
A temperature controller is a device designed to regulate a heater or other equipment based on a comparison between a sensor signal and a predetermined set point. It calculates the difference between these values and adjusts the equipment accordingly. Controllers capable of handling signals beyond temperature, such as humidity, pressure, and flow rate, are referred to simply as controllers. Those employing electronic technology are often termed digital controllers.
For example, imagine a system where a controller switches a relay on or off based on whether the temperature measured by a sensor is too high or too low compared to the desired temperature set by the controller.
Several factors influence how we design a control system. First, we need a way to measure the temperature or whatever we’re controlling. We choose a sensor based on the temperature it can handle and where it will be used. Second, we need a way to control the process. This could involve directly using a relay or controlling another device like a contactor, or using an analog voltage or current. Finally, we need to set up the controller to apply the right amount of control to the process. Ideally, the process responds perfectly to the controller’s actions, but in reality, it’s more complicated. Sometimes, the process overshoots the desired temperature, then undershoots, leading to a back-and-forth response. Other times, the response is too slow. Setting up the controller involves finding a balance between these issues to get a smooth response to changes in temperature while responding quickly.
Understanding Temperature Controller Response
In the diagram, we see the ideal response of the process to the controller’s commands. However, real-life responses often differ.
Sometimes, the process goes too far beyond the desired temperature before coming back. To correct this, the controller reduces its power, causing the temperature to drop below the desired level. Then, the controller increases power again. If the controller is too sensitive, this back-and-forth motion can lead to oscillations.
Conversely, the process may respond too slowly. This happens when the controller isn’t sensitive enough. In this case, the controller adjusts power too slowly, resulting in a sluggish response in reaching the desired temperature.
Setting up the controller involves finding a balance between these scenarios. We aim for a smooth response to temperature changes while ensuring a quick reaction time. This often means adjusting the controller’s settings to minimize overshooting and undershooting while maintaining a rapid response time.
Types Of Temperature Controller
1. ON/OFF Control:
This is the simplest form of control. When the temperature is below the desired level, the controller applies full heat. When it’s above the set temperature, it applies no heat or full cooling. While this method works fine in some systems, in others it can lead to the response we saw in Curve A earlier.
2. Proportional (P) Control:
With this method, the controller adjusts the power applied to the process. It does this by turning the relay output on and off for different percentages of a fixed cycle time. For instance, if the relay is always closed, 100% power is applied. If it’s closed for only 10% of the time, then 10% power is applied. Proportional control ensures that the power applied to the process is proportional to how far the temperature is from the desired level. We usually measure this as a Proportional Band (PB).
Here’s a simplified version:
3. Reset (I) Control
Reset control addresses a problem that arises with proportional control. In any real system, there are energy losses to or from the environment surrounding the process. The issue with proportional control is that it tends to balance out with these losses before reaching the desired setpoint. This results in needing an offset, traditionally called a reset, to counteract these system losses.
In older controllers, this offset was adjusted using a reset potentiometer. However, with the introduction of microprocessor-based controllers like the RS stock no. 340-083 series, a mathematical function called integration (I) is used. This function automatically applies the right amount of control action needed to correct for the losses.
The graph shows the response curve with proportional control only up to a certain point in time (t1). At time t1, the integral term is activated, and the controller gradually increases power until the process reaches the desired setpoint.
4. Integral (I) Control
The integral term continuously adds up the deviation from the setpoint (e) over time. If the process temperature goes above the setpoint, the value of e is negative, helping to bring the process back to the setpoint. The accumulator then undergoes a gain adjustment known as the integral action time (IT). A larger integral action time means the error accumulator contributes less to the output power, taking longer to take effect. The gain of the proportional term also affects the integral term, meaning the Proportional Band (PB) affects the overall gain of the controller, not just the proportional term.
5. Rate (D) Control
Reset action, while effective, can be slow to respond to sudden changes in the process. Rate action, also known as derivative (D) control, addresses this issue by continuously monitoring how quickly the process value changes. If the process temperature suddenly changes (e.g., a cool object is introduced into a heating zone), the derivative term calculates a large positive increase in the deviation from the setpoint, resulting in a significant increase in output power to counteract it. The derivative action time (DT) adjusts the effect, with a larger derivative action time leading to a greater impact over a longer period. Like with the integral term, the Proportional Band (PB) gain affects the derivative term, providing an overall gain to the controller.
The graph below illustrates how the derivative term leads to a quicker response to disturbances compared to proportional-only control.
By combining proportional, integral, and derivative control actions, known as PID control, we can achieve the best possible control. This approach ensures that the process responds quickly to disturbances, minimizes overshoot, and accurately controls the process temperature.
The Challenge
However, there’s a catch. If we don’t set the integral action time (IT), derivative action time (DT), and proportional band (PB) correctly, the system’s response can actually be worse than if we just used ON/OFF control.
Finding the Right Settings
There are established methods, like the Ziegler-Nichols method, for calculating the correct values of PB, IT, and DT. However, these methods may not work well with systems that are nonlinear or have significant time delays compared to the action times.
Challenges in Practice
Additionally, we need to test the process to see how much it overshoots and how long it takes to oscillate. Unfortunately, this isn’t always possible or safe for some processes.
Temperature Control Configuration Example
The following example describes the basic configuration for temperature control.
Temperature Controller Principle
The temperature controller operates based on a feedback control system principle, as depicted in the accompanying figure. Within the temperature controller, essential components of this feedback system are integrated. By combining the temperature controller with a compatible controller and temperature sensor tailored to the specific object being controlled, a functional feedback control system is established, enabling precise temperature regulation.
Do you know about Cold Junction Compensation.