Ready to Get Your VFD Solution?
Whether you need a single drive or a complete system, we have the VFD products and engineering support to ensure your project's success.
VFD PID Control: How to Tune Built-In PID for Pumps and Fans
VFD PID control is a closed-loop method that lets a variable frequency drive adjust motor speed automatically based on feedback from a process sensor. A pressure transmitter, flow meter, level sensor, or temperature signal tells the drive the actual process value. The drive compares that value to a target setpoint and changes the motor frequency to close the gap.
Luis found out how powerful this can be during a commissioning job at a Mexico City hotel. The booster pumps for the guest floors had been cycling on and off every 45 seconds, hammering the pipes and driving the maintenance team crazy. The problem was not the pumps. It was that the system was using simple on/off pressure switches instead of a continuous VFD PID controller. Luis wired a 4-20 mA pressure transducer to the drive, enabled the built-in PID, and spent twenty minutes tuning the gains. The pumps began to modulate smoothly between 35 Hz and 48 Hz, pressure held within 0.1 bar of the setpoint, and the pipe hammer disappeared.
That is the difference between a drive that simply runs a motor and one that controls a process. This guide explains how VFD PID control works, when to use the drive’s built-in controller, how to wire and scale the feedback, and how to tune the loop for stable pressure, flow, level, and HVAC static pressure.
Key Takeaways
- VFD PID control uses sensor feedback to adjust motor speed and maintain a process setpoint without an external controller.
- Built-in VFD PID is fastest and simplest for single loops; PLC PID is better for multi-pump systems and operator visibility.
- Most pump and fan applications need only PI control, with derivative left at zero.
- Proper sensor scaling, minimum frequency limits, and sleep/wake settings protect equipment and prevent cycling.
- Tune P first for response, then add I to remove steady-state offset, and test under real load changes.
What VFD PID Control Actually Means
A VFD PID controller is a closed-loop control algorithm that runs inside the drive’s firmware. It continuously reads a process variable, compares it to a setpoint, and updates the motor speed. The goal is to keep the process variable as close as possible to the setpoint even when demand changes.
The loop has four parts. The sensor measures the process. The VFD calculates the error, which is the difference between the setpoint and the measured value. The PID algorithm converts that error into a speed command. The motor and driven equipment then change the process, and the cycle repeats.
Common process variables in VFD PID control include pressure, flow rate, liquid level, duct static pressure, and temperature. A pump system might use a pressure transducer on the discharge line. An HVAC air handling unit might use a differential pressure sensor in the supply duct. A cooling tower might use a temperature sensor in the return water line. The same control principle applies to all of them.
For the broader commissioning context that comes before enabling PID, see our VFD configuration guide.
When to Use Built-In VFD PID Control
Built-in VFD PID control makes sense when the process loop is local, relatively simple, and fast. It is a good fit when the drive can do the entire job without help from a PLC or external controller.
Typical applications include constant-pressure water supply, booster pump systems, irrigation pumps, HVAC supply and return fans, cooling tower fans, small circulating pumps, and simple level-control loops. In these cases, the drive receives a single analog feedback signal, runs the PID internally, and outputs a speed command directly to the motor.
Built-in PID is less appropriate when the control strategy is complex. Multi-pump staging, cascade control, coordinated speed matching between multiple drives, and integration with a plant-wide SCADA system usually belong in a PLC or DCS. The VFD can still handle the fast inner speed loop, but the higher-level logic should live elsewhere.
When built-in PID fits
- Single pump or fan maintaining pressure, flow, level, or temperature
- Local setpoint and limited need for operator interaction
- Fast response time is important
- Cost and wiring simplicity matter
When to move PID to a PLC
- Multiple pumps or fans must stage on and off
- The operator needs frequent setpoint changes and trend data
- The loop is part of a larger plant control strategy
- Recipe-based or sequence-driven control is required
Built-In VFD PID vs PLC PID vs External Controller
Choosing where to run the PID loop affects response speed, wiring complexity, cost, and flexibility. The right choice depends on the size of the system and how much visibility the operator needs.
| Factor | Built-In VFD PID | PLC PID | External PID Controller |
|---|---|---|---|
| Response time | 5-10 ms | 10-50+ ms | 100-500 ms typical |
| Wiring | Simple: sensor to drive | More complex: AI/AO modules | Dedicated loop wiring |
| Hardware cost | Lowest | Higher | Highest |
| Operator interface | Keypad or small HMI | Full SCADA/HMI capability | Local panel display |
| Multi-loop/multi-pump | Difficult | Easy | Moderate |
| Data logging | Limited | Easy | Limited |
| Best for | Single, fast, local loops | Complex plant control | Precision standalone loops |
Built-in VFD PID control wins on speed and simplicity. The sensor connects directly to the drive’s analog input. The loop executes inside the drive’s processor, so there is no scan-time or communication delay.
PLC-based PID wins on flexibility. The PLC can read multiple sensors, run several loops, coordinate pump staging, log data, and present everything on an HMI. The trade-off is a slower loop because the PID must wait for the PLC scan and the fieldbus communication to the VFD.
External PID controllers are less common today but still useful for specialized standalone loops where the drive lacks adequate PID functionality.
For a deeper look at programming the drive to accept commands from a PLC or fieldbus, see our VFD programming guide.
Components and Wiring for VFD PID Control
A working VFD PID control system needs three things: a sensor that measures the process, a signal path back to the drive, and a way for the operator to set the target value.
Sensor Selection
The most common industrial feedback signal is 4-20 mA. A 4 mA lower limit makes broken-wire detection possible, and current signals are less sensitive to electrical noise than voltage signals. Some sensors use 0-10 V, which is acceptable for short cable runs in low-noise environments.
Pressure transmitters are the most common sensor in pump systems. Flow meters are used for flow control. Level sensors are used for tank level control. Temperature sensors with a transmitter output are used for cooling and heating loops.
Analog Input Wiring
Run a shielded twisted-pair cable between the sensor and the VFD analog input. Connect the shield at the drive end only, unless the sensor manufacturer specifies otherwise. Keep the analog cable away from power cables and motor leads to avoid induced noise. If the feedback value jumps around on the display, the wiring is the first thing to check.
Setpoint Sources
The setpoint can come from the drive keypad, a potentiometer connected to a second analog input, or a communication register from a PLC. Many drives allow the operator to enter a digital setpoint in engineering units such as bar, psi, or percent of sensor range. Choose the source that matches how the operator will actually use the system.
Feedback Scaling
This is where many VFD PID settings go wrong. The drive must know what 4 mA and 20 mA mean in real engineering units. If the pressure transmitter is rated 0-10 bar, then 4 mA equals 0 bar and 20 mA equals 10 bar. If the setpoint is 5 bar, enter 50% of the scaled range, not 50% of the raw current. Incorrect scaling is a common reason a pump runs at the wrong speed even though the PID appears to be working.
How to Configure VFD PID Parameters Step by Step
Every manufacturer uses different parameter numbers, but the configuration sequence is nearly universal. Use this workflow as a brand-agnostic checklist.
Step 1: Enable the Built-In PID Function
Find the parameter that enables PID or selects a pump/fan control macro. Some drives call it PID enable. Others call it constant pressure mode or application macro. Enable it before you configure anything else.
Step 2: Choose the Feedback Source
Select the analog input that receives the sensor signal. Configure the input type to match the sensor, either 4-20 mA current or 0-10 V voltage. Verify that the drive displays a live feedback value that changes when you pressurize or vent the sensor.
Step 3: Scale the Feedback
Enter the engineering-unit values that correspond to the minimum and maximum sensor outputs. For a 4-20 mA pressure transmitter rated 0-6 bar, set the minimum process value to 0 bar and the maximum to 6 bar. Confirm the displayed value matches a handheld pressure gauge.
Step 4: Set the Setpoint Source and Value
Choose where the setpoint comes from and enter the target value. A keypad setpoint is fine for systems that rarely change. A second analog input or a communication register is better if the operator changes the setpoint frequently.
Step 5: Configure the PID Output
Tell the drive to use the PID output as the speed reference. Set minimum and maximum output limits to protect the pump or fan. For a centrifugal pump, a minimum frequency of 20-30 Hz is often needed to avoid dead-head and overheating. The maximum frequency should match the mechanical limits of the pump or fan.
Step 6: Protect the Equipment
Set the motor overload, overcurrent, and overvoltage thresholds. Configure stall prevention so the drive pauses acceleration if current climbs. These settings are part of good VFD parameter settings practice, not just PID tuning.
Step 7: Enable Sleep and Wake Functions
Sleep mode stops the motor when demand is low and the pressure is above the setpoint. Wake pressure restarts the motor when pressure drops. Use a small deadband between sleep and wake pressures to prevent rapid cycling. Without this, the motor can chatter on and off at low flow.
Step 8: Save and Test
Save the parameters and run the drive in manual mode first if the drive supports it. Confirm that the feedback value is correct, the motor direction is correct, and the speed changes smoothly with a setpoint change. Then switch to automatic PID mode and tune.
Tuning the VFD PID Loop
Tuning is the step that turns a working loop into a stable loop. Most VFD PID tuning problems come from using someone else’s gain values instead of tuning for the actual system.
What P, I, and D Do
The proportional term reacts to the current error. A higher proportional gain makes the drive respond faster but can cause oscillation.
The integral term accumulates error over time and eliminates steady-state offset. A shorter integral time means a stronger integral action. Too much integral action causes slow hunting.
The derivative term reacts to the rate of change of the error. It can dampen overshoot, but in pump and fan systems it usually amplifies noise and makes the output jumpy. Most VFD PID settings for pressure and flow leave derivative at zero.
Manual Tuning Method for VFDs
This is the safest approach for field commissioning.
- Set integral and derivative to zero.
- Start with a low proportional gain, such as 0.5 or 1.0.
- Increase the gain in small steps while opening and closing a valve or changing the setpoint slightly.
- Find the gain where the system begins to oscillate with constant amplitude, then reduce the gain by roughly half.
- Add integral action by decreasing the integral time until steady-state error disappears without causing oscillation.
- Leave derivative at zero unless overshoot is a clear problem and the feedback signal is clean.
For a detailed explanation of PID methods, the team at Canroon covers the theory behind PID control methods for VFD applications.
Ziegler-Nichols Shortcut
The Ziegler-Nichols method is useful for getting a starting point quickly. Set integral and derivative to zero, increase proportional gain until the system oscillates steadily, and record the ultimate gain and oscillation period. Then use the standard formulas to calculate starting P, I, and D values. Treat these values as a starting point only, because they often produce an aggressive response that needs to be softened in the field.
Practical Starting Values by Application
These values are starting points, not final answers. Every system has different piping, inertia, and sensor response.
| Application | Proportional starting point | Integral starting point | Derivative |
|---|---|---|---|
| Pump pressure control | Kp = 0.5-2.0 | Ki = 0.1-0.5 or Ti = 10-30 s | 0 |
| HVAC static pressure | Kp = 0.5-1.0 | Ti = 0.5-2.0 s | 0-0.5 s |
| Flow control | Kp = 0.3-1.0 | Ti = 5-20 s | 0 |
| Level control | Kp = 0.2-0.8 | Ti = 30-120 s | 0 |
Maria, a commissioning engineer in Sao Paulo, learned that these starting values matter. She was tuning a level-control loop on a wastewater equalization tank and started with the same gains she had used on a pressure booster. The pump speed oscillated wildly because liquid level changes much more slowly than pressure. She reduced the proportional gain by half and lengthened the integral time to 90 seconds. The loop stabilized within two tank volumes.
Application-Specific Tuning Guidance
Different processes behave differently. The tuning strategy that works for a booster pump may be completely wrong for an HVAC fan or a cooling tower.
Constant Pressure Pump Systems
Pressure loops are fast and responsive, which makes them relatively easy to tune. Install the pressure transmitter on the discharge line at a point that sees representative pressure, not right at the pump outlet where pulsations are highest.
Set a minimum frequency that keeps the pump operating safely. For many centrifugal pumps, 20-30 Hz is the practical minimum. Below that, the pump can overheat, cavitate, or run dead-head.
Use sleep and wake pressures with a deadband. A common setup is to sleep when pressure exceeds the setpoint by 5% and wake when it falls below the setpoint by 5%. This prevents the motor from cycling at low demand.
For multi-pump systems, the lead pump is usually controlled by the VFD while additional fixed-speed pumps are staged by a PLC based on the VFD output frequency or system pressure. The PLC handles the staging logic, and the VFD handles the fine speed control. GTAKE explains how this architecture works in constant pressure water supply systems.
HVAC Static Pressure and Fan Control
Duct static pressure control keeps the supply duct pressure constant as VAV boxes open and close. The pressure sensor is typically located two-thirds to three-quarters of the way down the main duct, where the reading best represents the pressure available to all branches.
Static pressure loops can be noisy because dampers move suddenly. Use a feedback filter if the pressure signal jumps. Setpoint values are usually expressed in inches of water column or Pascals. A typical commercial building might target 1.0 to 1.5 inches w.c.
Fan systems often use higher proportional gains than pump systems because the duct pressure responds quickly to speed changes. Start with a proportional gain of 0.5 to 1.0 and an integral time of 0.5 to 2.0 seconds, then adjust based on the response.
Flow, Level, and Temperature Control
Flow loops are faster than level loops but slower than pressure loops. Flow meters can produce noisy signals, so filtering is often necessary. Level loops are slow because tanks have large volumes. Use low proportional gains and long integral times to avoid overcorrecting. Temperature loops are the slowest of all because thermal mass delays the response. Integral times of several minutes are common.
For energy-saving background on why slower speed is worth the tuning effort, see our VFD for pumps and fans guide.
Common VFD PID Problems and Fixes
Even a correctly wired loop can misbehave if the tuning or scaling is wrong. Here are the symptoms we see most often in the field.
| Symptom | Likely cause | Fix |
|---|---|---|
| Oscillation at setpoint | Proportional gain too high or integral time too short | Reduce Kp or increase Ti |
| Slow response with steady-state offset | Proportional gain too low or integral time too long | Increase Kp or decrease Ti |
| Pump cycling on and off | No deadband, integral too aggressive | Add 2-5% deadband, reduce Ki |
| Pressure never reaches setpoint | Output limits or pump capacity too low | Check maximum frequency, pump curve, and sensor scaling |
| Erratic speed jumps | Noisy signal with derivative enabled | Set derivative to zero, add feedback filter |
| Large overshoot on startup | Integral windup | Use anti-windup, start in manual, or limit integral action |
| Drive ignores setpoint changes | Wrong command source selected | Verify that PID output is the active speed reference |
Chen, a maintenance supervisor in a Shanghai electronics plant, saw several of these symptoms at once. A new chiller pump loop was cycling every two minutes and never holding the setpoint. The proportional gain was copied from a fan drive and was far too high for the slower pump loop. The minimum frequency was set to 5 Hz, which was below the pump’s safe operating range. Chen lowered the gain, lengthened the integral time, raised the minimum frequency to 25 Hz, and added a 3% deadband around the sleep pressure. The loop stabilized within an hour.
Advanced VFD PID Features
Modern drives offer features that make PID control more reliable and easier to live with.
Anti-Windup and Bumpless Transfer
Integral windup happens when the PID output saturates at the maximum or minimum speed limit while the error remains large. The integral term keeps accumulating, and when the error finally reverses, it takes time for the integral to unwind. Anti-windup limits this accumulation. Bumpless transfer smooths the switch between manual and automatic mode so the output does not jump.
Feed-Forward Control
Feed-forward adds a known disturbance signal to the PID output before the error appears. In a multi-zone HVAC system, for example, the feed-forward signal might anticipate that several dampers are about to open. The drive starts accelerating before the duct pressure drops.
Multi-Pump Staging and Pump Alternation
Some pump-specific drives include multi-pump control macros. They can stage auxiliary pumps on and off based on the speed of the variable-speed pump, alternate lead duty between pumps for even wear, and run a backup pump if the lead pump fails. Complex staging logic usually still requires a PLC, but simple two- or three-pump systems can often be handled entirely by the drive.
Fire Override and Safety Bypass
In HVAC applications, a fire override signal can force the drive to full speed regardless of the PID output. This ensures smoke extraction or stairwell pressurization during an emergency. Configure these safety functions carefully and test them during commissioning.
VFD PID Control: FAQ
What is VFD PID control?
VFD PID control is a closed-loop control method in which the drive reads a process sensor, compares the measured value to a setpoint, and automatically adjusts motor speed to minimize the difference.
When should I use built-in PID instead of a PLC?
Use built-in VFD PID control for simple, single loops where speed and simplicity matter. Use a PLC for multi-pump staging, operator visibility, data logging, and complex sequencing.
How do I scale a 4-20 mA pressure transducer in a VFD?
Enter the engineering units that correspond to 4 mA and 20 mA. For a 0-10 bar transmitter, set the minimum process value to 0 bar and the maximum to 10 bar. Verify the displayed value against a calibrated gauge.
Why does my pump pressure oscillate?
Oscillation usually means the proportional gain is too high or the integral time is too short. Reduce the gain or lengthen the integral time until the response settles.
What are good starting PID values for a pump?
A common starting point is proportional gain of 0.5 to 2.0, integral gain of 0.1 to 0.5 or an integral time of 10 to 30 seconds, and derivative at zero.
How does sleep/wake work in a VFD PID pump system?
Sleep mode stops the motor when pressure is above the setpoint by a set margin and demand is low. Wake pressure restarts the motor when pressure falls. The deadband between sleep and wake pressures prevents rapid cycling.
Can I use VFD PID for HVAC static pressure?
Yes. Install a duct static pressure sensor two-thirds to three-quarters of the way down the main duct, set the pressure setpoint in inches w.c. or Pascals, and tune the loop with conservative starting gains.
Do I need derivative control in a VFD PID loop?
Usually no. Derivative is rarely needed in pump, fan, and pressure loops and often amplifies noise. Start with PI control and add derivative only if overshoot is a real problem and the signal is clean.
VFD PID Control: Tune the Loop, Then Trust It
VFD PID control turns a variable frequency drive into a local process controller. When the loop is wired, scaled, and tuned correctly, the motor speed becomes invisible to the operator. Pressure stays where it belongs. Flow matches demand. Duct static pressure remains stable as dampers move. The system simply works.
The key is to start with the basics before chasing advanced features. Make sure the sensor is wired and scaled correctly. Enable the PID function and route the output to the speed reference. Set minimum and maximum frequencies that protect the equipment. Tune P first for response, then add I to remove offset. Leave derivative alone unless you have a clear reason to use it.
If you are commissioning a pump, fan, or process-control system and need drives with reliable built-in PID, the Shandong Electric VFD range includes low-voltage and high-voltage solutions with keypad, software, and fieldbus support backed by our engineering team.
