Temperature Controller Glossary
Glossary of Control Terminology
Hysteresis
ON/OFF control action turns the output ON or OFF based on the set point. The output frequently changes according to minute temperature changes as a result, and this shortens the life of the output relay or unfavorably affects some devices connected to the Temperature Controller. To prevent this from happening, a temperature band called hysteresis is created between the ON and OFF operations.
Hysteresis (Reverse Operation)
Example:
If a Temperature Controller with a temperature range of 0 to 400 °C has a 0.2% hysteresis, D will be 0.8 °C. If the set point is 100 °C, the output will turn OFF at a process value of 100 °C and will turn ON at a process value of 99.2 °C.
Hysteresis (Forward Operation)
Example:
If a Temperature Controller with a temperature range of 0 to 400 °C has a 0.2% hysteresis, D will be 0.8 °C. If the set point is 100 °C, the output will turn OFF at a process value of 100 °C and will turn ON at a process value of 100.8 °C.
Offset
Proportional control action causes an error in the process value due to the heat capacity of the controlled object and the capacity of the heater. The result is a small discrepancy between the process value and the set point in stable operation. This error is called offset. offset is the difference in temperature between the set point and the actual process temperature. It may exist above or below the set point.
Hunting and Overshooting
ON/OFF control action often involves the waveform shown in the following diagram. A temperature rise that exceeds the set pointafter temperature control starts is called overshooting. Temperature oscillation near the set point is called hunting. Improved temperature control is to be expected if the degree of overshooting and hunting are low.
Hunting and Overshooting in ON/OFF Control Action
Control Cycle and Time-Proportioning Control Action
The control output will be turned ON intermittently according to a preset cycle if P action is used with a relay or SSR. This preset cycle is called the control cycle and this method of control is called timeproportioning control action.
Example:
If the control cycle is 10 s with an 80% control output, the ON and OFF periods will be as follows.
Derivative Time
Derivative time is the period required for a ramp-type deviation in derivative control (e.g., the deviation shown in the following graph)that coincides with the control output in proportional control action.The longer the Derivative time is the stronger the derivative control action will be.
PD Action and Derivative Time
Integral Time
Integral time is the period required for a step-type deviation in integral control (e.g., the deviation shown in the following graph) to coincide with the control output in proportional control action. The shorter the Integral time is the stronger the integral action will be. If the Integral time is too short, however, hunting may result.
PI Action and Integral Time
Constant Value Control
For constant value control, control is preformed at specific temperatures.
Program Control
Program control is used to control temperature for a target value that changes at predetermined time intervals.
Auto-tuning
The PID constant values and combinations that are used for temperature control depend on the characteristics of the controlled object. A variety of conventional methods that are used to obtain these PID constants have been suggested and implemented based on actual control temperature waveforms. Auto-tuning methods make it possible to obtain PID constants suitable to a variety of controlling objects. The most common types of Auto-tuning are the step response, marginal sensitivity, and limit cycle methods.
Step Response Method
The value most frequently used must be the set point in this method.Calculate the maximum temperature ramp R and the dead time L from a 100% step-type control output. Then obtain the PID constants from R and L.
Marginal Sensitivity Method
Proportional control action begins from start point A in this method.Narrow the width of the proportional band until the temperature starts to oscillate. Then obtain the PID constants from the value of the proportional band and the oscillation cycle time T at that time.
Limit Cycle Method
ON/OFF control begins from start point A in this method. Then obtain the PID constants from the hunting cycle T and oscillation D.
Readjusting PID Constants
PID constants calculated in auto-tuning operation normally do not cause problems except for some particular applications. In those cases, refer to the following diagrams to readjust the constants.
Response to Change in the Proportional Band
Wider
It is possible to suppress overshooting although a comparatively long startup time and set time will be required.
Narrower
The process value reaches the set point within a comparatively short time and keeps the temperature stable although overshootingand hunting will result until the temperature becomes stable.
Response to Change in Integral Time
Wider
The set point takes longer to reach. It is possible to reduce hunting, overshooting,and undershooting although a comparatively long startup time and set time will be required.
Narrower
The process temperature reaches the set point within a comparatively short time although overshooting,undershooting, and huntingwill result.
Response to Change in Derivative Time
Wider
The process value reaches the set point within a comparatively short time with comparatively small amounts of overshooting and undershooting. Fine-cycle hunting will result due to the change in process value.
Narrower
The process value will take a relatively long time to reach the set point with heavy overshooting and undershooting.
Fuzzy Self-tuning
PID constants must be determined according to the characteristics of the controlled object for proper temperature control. The conventional Temperature Controller incorporates an auto-tuning function to calculate PID constants. In that case, it is necessary to give instructions to the Temperature Controller to trigger the autotuning function. Furthermore, temperature disturbances may result if the limit cycle is adopted. The Temperature Controller in fuzzy selftuning operation determines the start of tuning and ensures smooth tuning without disturbing temperature control. In other words, the fuzzy self-tuning function makes it possible to adjust PID constants according to the characteristics of the controlled object.
Fuzzy Self-tuning in 3 Modes
PID constants are calculated by tuning when the set point changes.
When an external disturbance affects the process value, the PID constants will be adjusted and kept in a specified range.
If hunting results, the PID constants will be adjusted to suppress hunting.