Here’s the straightforward answer: Process gain (K) is determined by the ratio of the steady-state change in the output variable (liquid level) to the steady-state change in the input variable that caused it (such as inlet flow rate or valve position). It is expressed as K = ΔLevel / ΔInput. You obtain this value experimentally through a step test—introducing a small, deliberate change in the manipulated variable and measuring the new stable level after transients die out.
Process gain is not just a number; it is the built-in sensitivity of your liquid level loop. An accurate K tells you how aggressively the level will respond to a control action—and it is the single most important piece of information you need to set a safe, effective controller gain (Kc) when tuning a PID loop.
How Process Gain K Is Determined in a Liquid Level System
The Fundamental Definition
Process gain describes the static sensitivity of the system. It purely concerns the final, settled relationship between the controller output (manipulated variable) and the process variable.
For a typical pilot-plant liquid level loop, the manipulated variable is often the inlet flow rate—via a control valve or variable-speed pump. The process variable is the tank level. So:
K = Δh / ΔQ_in
where Δh is the change in steady-state level (in mm or %), and ΔQ_in is the change in inlet flow (in L/min or % of valve opening).
How to Measure K with a Step Test
The only reliable way to determine K experimentally is to perform an open-loop step test.
- Run the system at a constant, steady-state condition. Wait until the level stops moving.
- Record the baseline level (h₁) and the current value of the manipulated variable (e.g., valve signal MV₁).
- Make a small step change in the manipulated variable—say, opening the valve by an extra 5%. The change must be large enough to see a clear level response, but small enough to stay within safe operating limits.
- Do not touch anything. Let the level evolve and eventually settle at a new steady state.
- Record the new steady-state level (h₂) and the final manipulated variable value (MV₂).
- Calculate K = (h₂ - h₁) / (MV₂ - MV₁).
If the level increases for a positive step in inlet flow, K is positive—which is typical for a level loop. The units match those you use for controller scaling.
A Concrete Pilot Plant Example
Imagine a 200‑L feed tank where the inlet flow is set by a pump with a 4–20 mA signal.
- You increase the pump command from 12.0 mA to 13.0 mA (a 1.0 mA step).
- The level eventually rises from 55% to 70% and holds steady.
- K = (70% – 55%) / (13.0 mA – 12.0 mA) = 15% level per 1 mA signal change.
That number, 15% per mA, is your process gain. It tells you that every milliampere you add to the pump signal will permanently lift the level by 15% of the tank’s range.
The Critical Role of Process Gain in Controller Tuning
K Directly Informs the Controller Gain (Kc)
In PID control, the overall loop gain is the product of the controller gain (Kc) and the process gain (K). Stability margins and transient response are dictated by Kc × K.
Because K is a fixed property of the pilot-plant equipment and operating point, you can only adjust Kc. If K is large, a small error signal multiplied by Kc will produce a large corrective action, which then gets amplified again by the sensitive process—leading to oscillation or overshoot. The fundamental tuning rule is: the higher the process gain, the lower you must set Kc to keep the loop stable.
Avoiding Instability, Oscillation, and Overflow
Overestimating the acceptable Kc for a high‑gain level loop creates a dangerous feedback cycle.
- Too much controller gain makes the valve overcorrect.
- The level surges past the setpoint, then reverses direction sharply.
- This oscillation can grow in amplitude until the vessel overflows or triggers a high‑level shutdown.
In a pilot plant—where vessels are small and residence times short—these risks are amplified. The process gain becomes your safety coefficient: it tells you how delicate your tuning must be.
The Relationship Between Sensitivity and Tuning Aggression
Process gain is, at its heart, a sensitivity metric. A large K means that even a hesitant valve movement moves the level significantly.
Therefore, PID tuning for a high‑K system must be far more conservative. You might use a Kc that is an order of magnitude smaller than for a low‑K system. Without measuring K first, you are essentially guessing whether the process is “twitchy” or “sluggish”—the most common source of failed tunings.
Understanding the Trade‑offs and Pitfalls
Static vs. Dynamic Nonlinearity
Process gain is not always constant. In many tanks, the gain changes with level because the outlet flow may be gravity‑driven (varying with head), or the vessel geometry is non‑linear.
If you measure K at 50% level and then apply that same Kc when operating at 90%, the loop may behave aggressively. It is wise to determine K at multiple operating points and use the highest gain for conservative tuning, or implement gain scheduling.
On/Off Control vs. PID Control
The supplementary references describe an educational pilot plant where level control uses two limit switches and a solenoid valve—pure on/off (bang‑bang) control.
In that setup, there is no PID loop, no process gain, and no controller tuning in the traditional sense. The “gain” is essentially infinite: the valve is either fully open or fully closed. Process gain only becomes meaningful when you transition to continuously modulating valves and PID controllers. If your pilot plant currently runs simple limit‑based logic, K is not used—but as soon as you upgrade to a modulating pump or control valve for tighter control, measuring K becomes essential.
Pitfall: Deriving an Incorrect K from a Noisy or Disturbed Test
A step test must be carried out in a quiet environment. If a disturbance (like a varying outlet demand) occurs during the test, you will measure a combined effect, not the true process gain. Similarly, if you read the level before it has truly settled, you will underestimate or overestimate K. Always confirm that both the initial and final states are at genuine steady state.
Making the Right Choice for Your Control Objective
- If your primary focus is safe, responsive PID control on a continuous pilot‑plant level loop: Measure K via a clean step test, then use it to calculate a conservative Kc. For PI or PID, start with a controller gain that makes the product Kc × K approximately 0.5–1.0 (for level loops) and tighten gently.
- If your primary focus is simply maintaining level with high/low alarms and a solenoid valve: You are operating in on/off mode. Process gain is irrelevant to tuning, but you may still want to know how quickly the level changes under full flow to size vessels correctly.
- If your primary focus is developing transferable tuning skills in a teaching pilot plant: Use the level loop to teach the relationship between open‑loop step tests and closed‑loop stability. Have students measure K, calculate an initial Kc, and observe how oscillation develops if they double that value.
Measure process gain carefully, and your liquid level loop will move from a source of constant tuning headaches to a predictable, well‑behaved part of your pilot‑plant operation.
Summary Table:
| Metric / Parameter | Definition & Formula | Role in PID Controller Tuning |
|---|---|---|
| Process Gain (K) | $K = \Delta\text{Level} / \Delta\text{Input}$ | Measures system sensitivity. Higher $K$ requires a lower, more conservative controller gain ($K_c$). |
| Controller Gain ($K_c$) | User-adjusted tuning parameter | Determines response aggressiveness. $K_c$ must be balanced with $K$ to prevent loop oscillation. |
| Loop Gain ($K \times K_c$) | Product of process and controller gain | Controls overall loop stability. Target $K \times K_c \approx 0.5\text{ -- }1.0$ for stable level loops. |
Optimize Process Control Training with LABPARK Pilot Plants
Mastering PID loops requires hands-on practice. LABPARK designs and manufactures high-quality Educational and Vocational Unit Operations Pilot Plants tailored for universities, research institutes, and enterprises.
Our specialized systems cover key fields including:
- Chemical Engineering (including flow, temperature, and liquid level control loops)
- Bioprocess & Biotech
- Environmental & Water Treatment
Equip your students and researchers with the tools they need to perform step tests, analyze process variables, and tune controllers safely.
Contact LABPARK today to request a quote and discover how our pilot plants can enhance your engineering curriculum or research capabilities.
Related Products
- Binary System Vapor Liquid Equilibrium Data Determination Educational Unit Operations Pilot Plant
- Ternary Liquid-Liquid Equilibrium Educational Pilot Plant
- Quantitative Dosing and Liquid Flow Control Educational Unit Operations Pilot Plant
- Liquid-Liquid Mass Transfer Coefficient Determination Educational Pilot Plant
- Two-Dimensional Fluidization Hydrodynamics Educational Pilot Plant for Unit Operations Training
People Also Ask
- For VLE Experiments: Fugacity vs. Activity Coefficient Method Selection Criteria
- Why correlate VLE pilot plant data using Wilson, NRTL, and UNIQUAC? Achieve Accurate Column Scale-Up
- How do pilot plants assist in teaching & researching non-ideal VLE? Bridge Theory & Physical Validation
- How to Model VLE for Supercritical & Unstable Components? Pilot Plant Methods
- Why is accurate VLE data critical? Optimize Unit Operations & Process Scale-Up
