7.1 Process Control Modes

Introduction to Process Control Modes

In industrial automation, maintaining a process variable (PV) at a desired setpoint (SP) requires a controller to compute an appropriate corrective action. The mathematical mechanism a controller uses to determine its output based on the error (the difference between SP and PV) is known as the control mode. Understanding controller modes helps a Level I technician trace loop behavior and recognize abnormal operation. The current 2025 ISA task list does not mark PID tuning task 0112 for Level I; it is a Level II/III task, so Level I work must remain within the approved troubleshooting procedure and authorization boundary. The control modes range from simple two-position algorithms to sophisticated three-term PID (Proportional-Integral-Derivative) algorithms.

On-Off (Two-Position) Control

The simplest form of automatic control is On-Off or two-position control. In this mode, the controller's output is either 0% or 100%. A common example is a home thermostat. When the temperature drops below the setpoint, the heater turns on at 100% capacity; when it rises above the setpoint, the heater turns completely off.

While inexpensive and simple to implement, On-Off control introduces constant cycling of the process variable around the setpoint. To prevent the final control element from rapidly switching on and off—a phenomenon known as "chattering," which can cause premature mechanical wear—a differential gap or deadband is introduced. The deadband creates a range around the setpoint where no change in output occurs. For instance, a heater might turn on when the temperature drops to 68°F but won't turn off until it reaches 72°F. The 4°F difference is the differential gap. This mode is generally unsuitable for processes requiring precise control, such as chemical reactors or sensitive pressure vessels, but it remains widely used in level control and simple temperature applications.

Proportional (P) Mode

For tighter control, we use modulating control modes. The fundamental building block is the Proportional (P) mode, which produces an output directly proportional to the magnitude of the error. The mathematical relationship is governed by the controller gain ($K_c$).

Controller Gain ($K_c$) determines how much the output changes for a given change in error. A higher gain means a more aggressive response. In many industrial controllers, proportional action is expressed as Proportional Band (PB), which is the inverse of gain, expressed as a percentage: $PB = \frac{100\%}{K_c}$ A narrow proportional band (e.g., 20%) equates to a high gain (5), meaning a small error produces a large output change. Conversely, a wide proportional band (e.g., 200%) equates to a low gain (0.5), resulting in a sluggish response.

The primary limitation of purely proportional control is offset (or steady-state error). Because the proportional controller requires an error to generate an output, it cannot maintain the process variable exactly at the setpoint if the load changes. The controller will reach a new equilibrium where the output matches the new load, but a residual error—the offset—will persist.

Integral (I / Reset) Mode

To eliminate offset, the Integral (I) or Reset mode is added. Integral action continues to change the controller output as long as an error exists over time. It looks at the history of the error and integrates it, steadily driving the output until the PV equals the SP and the error is zero.

Integral action is typically measured in one of two ways, depending on the controller manufacturer:

1. Repeats per Minute (RPM): Represents how many times the proportional action is repeated each minute. A higher number means more aggressive integral action.
2. Minutes per Repeat (MPR) or Time Constant ($T_i$): Represents the time required for the integral action to duplicate the proportional action. This is the reciprocal of RPM. A lower time value means more aggressive integral action.

While integral action perfectly eliminates steady-state offset, excessive integral action can cause the loop to overshoot the setpoint and oscillate, a condition known as reset windup.

Derivative (D / Rate) Mode

The Derivative (D) or Rate mode provides an anticipatory action based on the rate of change of the error. If the error is changing rapidly, the derivative mode produces a large output to counteract the disturbance before it pushes the PV too far from the SP.

Derivative action is measured in Rate Time (minutes). It effectively looks into the future to predict where the error is heading. However, derivative control has a significant drawback: it is extremely sensitive to process noise. If the PV signal is noisy (e.g., turbulence in a flow measurement), the derivative mode will interpret the high-frequency noise as rapid changes in error and cause the controller output to fluctuate wildly. For this reason, derivative action is rarely used in noisy processes like flow or level control, but it is highly beneficial in slow-moving processes like temperature control, where it can counteract thermal lag.

Controller Action: Direct vs. Reverse Acting

Another critical configuration in a PID controller is its control action direction.

• Direct Acting: As the process variable (PV) increases, the controller output increases.
• Reverse Acting: As the process variable (PV) increases, the controller output decreases.

Determining the correct controller action is inextricably linked to the fail-safe state of the final control element, typically a control valve.

Valve Pairing and Fail-Safe States

Control valves are designed to fail in a specific position upon loss of instrument air or power. The two primary types are:

• Air-to-Open (Fail Closed): The valve requires air pressure to open. If air is lost, a spring forces the valve closed.
• Air-to-Close (Fail Open): The valve requires air pressure to close. If air is lost, a spring forces the valve open.

To select the correct controller action, the technician must analyze the process. Consider a tank level control system where the valve controls the outflow. If the valve is Air-to-Open and the tank level (PV) goes too high, the controller must increase its output to open the valve more and drain the tank. Since a PV increase requires an output increase, the controller must be Direct Acting.

Conversely, if the same outflow valve was Air-to-Close, a high tank level would require the valve to open, which means the air pressure to the valve must be decreased. Since a PV increase requires an output decrease, the controller must be Reverse Acting. Careful attention to these relationships ensures process stability and safe failure modes.