How Does Installing A Pi Controller Change A Single Tank Pilot System's Dynamic Order? Pi Dynamics Explained

Installing a PI controller does not change the tank’s physical dynamics, but it changes the overall closed-loop dynamic order from first order to second order. The tank itself still stores liquid in one physical state: its level. However, the PI controller introduces a second state through the time integral of the level error. Together, the tank level and accumulated error form a second-order closed-loop system that can exhibit overdamped, critically damped, or underdamped responses.

A single tank is intrinsically first order, while a tank controlled by an ideal PI controller is generally second order in closed loop because integral action adds one independent dynamic state.

What the Uncontrolled Tank Does

The Tank Has One Energy-Storage State

For a single liquid-level tank, the stored liquid volume, or equivalently the liquid height, is the principal dynamic state. A simplified material balance is

[ A\frac{dh}{dt}=q_{in}-q_{out} ]

where (A) is the tank cross-sectional area and (h) is the liquid level.

Because there is one independent accumulation variable, the open-loop tank model is first order.

Its Response Is Normally Non-Oscillatory

After a change in inlet flow, the level approaches a new equilibrium through a single exponential mode. With the usual linearized tank model, this produces a non-oscillatory first-order response.

The tank has no inherent mechanism for producing a pair of complex dynamic modes or sustained oscillation by itself.

How PI Control Adds Dynamic Order

Proportional Action Alone Does Not Add a State

A proportional controller calculates its output from the instantaneous error:

[ c(t)=K_c\epsilon(t)+c_s ]

where (K_c) is the controller gain and (\epsilon(t)) is the level error.

Since proportional action has no memory, it does not introduce an additional independent state. A first-order tank with proportional feedback therefore generally remains a first-order closed-loop system, although its gain and time constant are changed.

Integral Action Introduces a Second State

A PI controller has the form

[ c(t)=K_c\epsilon(t)+\frac{K_c}{\tau_I}\int_0^t \epsilon(t),dt+c_s ]

The integral term continuously accumulates past error. This accumulated value acts as a controller state in addition to the tank level.

The combined system therefore contains:

The tank-level state, representing liquid accumulation.
The integral state, representing accumulated control error.

That additional state increases the overall closed-loop dynamic order from one to two.

What the Second-Order Model Means

The Closed-Loop Characteristic Equation Has Two Modes

For a first-order tank with a PI controller, the closed-loop characteristic equation can be expressed in a form such as

[ \tau_p\tau_I s^2+\tau_I s+K_cK_p=0 ]

where (\tau_p) is the tank process time constant and (K_p) is the process gain.

The exact coefficients depend on the tank model and controller implementation, but the key feature is the (s^2) term. That term confirms that the closed-loop system is second order.

Controller Tuning Determines the Transient Shape

The proportional gain and integral time influence the closed-loop poles and therefore the damping behavior.

Depending on the tuning:

Underdamped: The level overshoots and may oscillate before settling.
Critically damped: The level reaches its target rapidly without oscillation.
Overdamped: The level approaches the target without oscillation but more slowly.

Thus, PI control does more than remove steady-state offset. It changes the kinds of transient behavior the level system can display.

Why Integral Action Is Used

It Eliminates Steady-State Offset

A proportional controller typically leaves a residual error after a load disturbance. The controller needs a nonzero error to maintain the corrective output required by the new operating condition.

Integral action continues adjusting the output while error remains. In normal operation, this drives the steady-state level error to zero.

It Changes the Controller’s Memory

The integral term remembers the history of the error rather than responding only to its current value. This memory is the reason PI control adds dynamic order, but it is also the source of additional tuning and operational considerations.

The controller effectively gives the system another mechanism for storing information over time.

Understanding the Trade-offs

The Tank Itself Is Still First Order

Calling the controlled system “second order” does not mean the physical tank has gained a second liquid-storage compartment or another fluid volume. The equipment remains a one-state process.

The second order belongs to the combined process-controller closed-loop model.

Aggressive Tuning Can Produce Oscillation

Increasing proportional gain or using an overly rapid integral action can reduce sluggishness but also reduce damping. The resulting response may overshoot, oscillate, or become poorly controlled.

A second-order model makes these behaviors possible; tuning determines whether they actually appear.

Integral Windup Can Delay Recovery

If the manipulated flow reaches a physical limit while a large error persists, the integral term may continue accumulating. When the process eventually returns to a controllable range, the stored integral action can cause excessive overshoot.

Pilot systems should therefore use suitable anti-windup or reset-management provisions, especially during startup, large set-point changes, and manual intervention.

Applying the Principle to a Pilot System

The correct interpretation depends on whether the question concerns the equipment alone or the controlled system.

Model the Uncontrolled Equipment Separately

Represent the single tank as a first-order process because its liquid level is the only physical accumulation state. This describes the tank’s intrinsic open-loop dynamics.

Include the PI State in Closed-Loop Analysis

When analyzing the complete feedback system, include the integral of error as an additional state. The tank-plus-PI combination is then modeled as a second-order closed-loop system.

Use the Response to Demonstrate the Difference

A useful pilot-plant experiment is to compare:

Open-loop or proportional-only operation, which generally shows first-order behavior.
PI-controlled operation, which shows second-order behavior and eliminates steady-state offset.

The transient response can then be used to identify whether the selected tuning produces overdamped, critically damped, or underdamped behavior.

Making the Right Choice for Your Goal

The PI controller should be evaluated as part of the complete control loop, not as a modification to the tank’s physical storage dynamics.

If your primary focus is dynamic modeling: Treat the tank as intrinsically first order, but model the tank and ideal PI controller together as a second-order closed-loop system.
If your primary focus is steady-state accuracy: Use integral action to eliminate persistent level offset after disturbances or operating-point changes.
If your primary focus is transient performance: Adjust controller gain and integral time to obtain the desired damping and settling behavior.
If your primary focus is reliable pilot operation: Account for actuator saturation and integral windup when implementing the PI controller.

A PI controller adds a controller state, so it changes the closed-loop dynamic order without changing the tank’s intrinsic first-order physical nature.

Summary Table:

Aspect	Key point
Uncontrolled tank	First-order system with liquid level as the only physical state
Proportional control	Changes gain and time constant without adding a state
Integral control	Adds an accumulated-error state
PI closed loop	Generally becomes a second-order system
Transient response	May be overdamped, critically damped, or underdamped
Main benefit	Eliminates steady-state level offset
Main risks	Overshoot, oscillation, and integral windup
Pilot-plant practice	Use suitable tuning and anti-windup protection

Select the Right Pilot-Plant Control Solution

Need to demonstrate level control, dynamic modeling, or process automation in your laboratory or facility? Contact LABPARK to discuss an Educational or Vocational Unit Operations Pilot Plant tailored to your objectives.

LABPARK provides pilot plants for chemical engineering, bioprocess and biotechnology, and environmental and water-treatment applications. Our systems help universities, research institutes, and enterprises teach first-order process behavior, evaluate PI and other control strategies, study transient responses, and develop practical operator skills using equipment designed for hands-on experimentation and process training.

Tell us your target process, teaching or research goals, scale requirements, and preferred instrumentation. LABPARK can help you identify a suitable pilot-plant configuration for reliable demonstrations, repeatable experiments, and industrially relevant training.