In small automation systems, tank control, pump stations, mixing equipment, conveyors and multi-motor applications, it is common to use one PLC to control several variable frequency drives. The control requirements usually include start, stop, speed reference, fault monitoring, running status feedback and centralized operation from an HMI. Compared with traditional terminal control, Modbus RTU control reduces hardwiring, allows the PLC to read more drive data, and makes it easier for the HMI to display frequency, current, status and fault information for each drive.
However, many field problems in these systems are not caused by a single wiring mistake or a missing ladder rung. The real problems often come from unclear system architecture, incorrect Modbus roles, duplicated slave addresses, wrong register addresses, misunderstanding of relay feedback, improper power supply conditions, and an incorrect commissioning sequence. This becomes even more obvious when different brands and series of drives are connected on the same Modbus RTU network, such as Schneider ATV310, Schneider ATV31P and Allen-Bradley PowerFlex 525. If all drives are treated as if they use the same control word, the same speed reference register and the same status word, the system will easily fail.
A typical system may consist of a Schneider M221 PLC as the main controller, an LU9GC3 as the Modbus RTU distribution module, several variable frequency drives as Modbus slaves, and a Schneider Harmony HMI as the operator interface. In this type of project, the key is not only how to draw the ladder logic. The complete engineering work must include communication architecture, Modbus address planning, drive parameter settings, PLC logic structure, relay feedback wiring, HMI design, power supply verification and step-by-step commissioning.

1. Correct System Architecture: PLC as Master, Drives as Slaves, HMI as Operator Interface
In a Modbus RTU system, there must be only one master on the same serial bus. The master sends read and write requests, and the slaves only respond to the master. In this application, the correct structure should be:
Schneider Harmony HMI
↓
Ethernet or HMI communication to PLC
↓
Schneider M221 PLC / TM221CE24R
↓
Modbus RTU Master
↓
Schneider LU9GC3 Modbus distribution module
↓
Multiple VFDs as Modbus RTU slaves
The HMI should not directly control all the drives on the Modbus RTU line. Its function is to send operator commands to the PLC, such as start, stop, reset and speed reference. The PLC then processes the logic and sends Modbus commands to the drives. The PLC also reads the drive status and sends this information back to the HMI for display.
A common misunderstanding in the field is confusing “HMI” with “HDMI.” HDMI is a video interface. HMI means Human Machine Interface, which is the industrial touchscreen used for operation and monitoring. A Schneider Harmony 7-inch touchscreen is an HMI, not an HDMI control device. It should communicate with the PLC, and the PLC should handle the actual drive control.

2. LU9GC3 Is Not an Ethernet Switch
The LU9GC3 module has several RJ45 ports, so it is often mistaken for a normal Ethernet switch. This is a serious misunderstanding.
The LU9GC3 is a Modbus serial distribution module, not a TCP/IP Ethernet switch. It is used to distribute one Modbus serial line to several devices such as VFDs or soft starters. It does not perform Ethernet switching, and it does not automatically convert Modbus TCP to Modbus RTU.
The correct use is:
PLC serial port / Modbus RTU master
↓
LU9GC3
↓
ATV310, ATV31P, PowerFlex 525 and other Modbus RTU slaves
The HMI should not be connected to the LU9GC3 as if it were an Ethernet network switch. If the HMI has an Ethernet port, it should normally connect to the PLC Ethernet port or to a real Ethernet switch in the PLC-HMI network. The LU9GC3 should only be used for the serial Modbus RTU connection between the PLC and the drives.

3. Each Drive Must Have a Unique Modbus Address
A Modbus RTU bus can have multiple slave devices, but each slave must have a unique address. If two drives have the same address, both may respond when the PLC sends a request to that address. This causes communication conflicts, invalid data, timeouts or unstable operation.
For a system with five drives, the address plan can be:
| Device | Suggested Modbus Address |
|---|---|
| PowerFlex 525 | 1 |
| ATV310 7.5 kW | 2 |
| ATV310 5.5 kW #1 | 3 |
| ATV310 5.5 kW #2 | 4 |
| ATV31P 5.5 kW | 5 |
The exact order is not mandatory, but two rules must be followed:
Each address must be unique.
The slave address in the PLC program must match the actual address set in the drive.
If all drives remain at the factory default address, for example address 1, communication may work when only one drive is connected. But once several drives are connected to the bus, the PLC will receive conflicting replies. This is one of the most common reasons why a single-drive Modbus test succeeds but a multi-drive network fails.

4. Communication Format Must Be the Same for All Devices
Modbus RTU communication requires not only correct slave addresses but also identical serial settings. The main settings are:
Baud rate: 9600 or 19200
Data bits: 8
Parity: Even or None
Stop bit: 1
Protocol: Modbus RTU
A practical starting point is:
19200 bps, 8E1
or:
9600 bps, 8E1
The most important point is consistency. The PLC, ATV310 drives, ATV31P drive and PowerFlex 525 must use the same baud rate, parity and stop bit. If one device has a different communication format, that device will not respond correctly to the PLC.
For initial commissioning, it is not recommended to use a very high baud rate. A Modbus RTU network with several VFDs is exposed to electrical noise, long cables, grounding problems and shielding issues. Lower or moderate baud rates are usually more stable during commissioning. After the system is working reliably, the communication cycle and baud rate can be optimized if necessary.
5. Modbus Control Principle for Schneider ATV310
To start and control an ATV310 through Modbus, two conditions must be met.
First, the drive communication parameters must be correct.
Second, the command source and frequency reference source must be set to the communication channel.
If the drive is still configured for terminal start or keypad speed reference, the PLC may write control words and frequency references successfully, but the drive will not run. This often leads technicians to believe that the PLC program is wrong, when the real issue is the drive control channel setting.
The ATV310 normally requires configuration of the following items:
Modbus slave address
Baud rate
Communication format
Command source
Frequency reference source
Control profile
In the PLC program, at least two write operations are normally required:
Write command word: start, stop, reset
Write frequency reference: for example 10 Hz, 20 Hz, 50 Hz
At the same time, the PLC should read status information:
Running status
Fault status
Actual frequency
Output current
Communication status
For the first commissioning stage, the best approach is not to build the full automatic process immediately. A safer and clearer sequence is to test one ATV310 first:
1. PLC reads the ATV310 status register successfully.
2. PLC writes a low speed reference, such as 10 Hz.
3. PLC writes a start command.
4. ATV310 starts from the PLC command.
5. PLC writes a stop command.
6. ATV310 stops from the PLC command.
7. PLC identifies drive fault or communication error.
After one ATV310 works correctly, the same logic can be copied and adjusted for the other ATV310 drives.
6. PowerFlex 525 Cannot Use the Same Register Table as Schneider ATV310
In a mixed system, the Allen-Bradley PowerFlex 525 must be treated separately. It may support Modbus, but its parameter numbers, command source settings, Modbus register addresses, speed reference scaling and status word definitions are not the same as Schneider ATV310.
This is a critical point. All VFDs are not identical just because they all use Modbus. Different manufacturers implement Modbus control in different ways. Even when they all have a command word and a speed reference, the addresses, bit definitions, scaling and command values can be different.
For PowerFlex 525, the RS485 Modbus node address is typically set in:
C124 [RS485 Node Addr]
If the PowerFlex 525 is planned as slave address 1, then:
C124 = 1
Other important parameters include:
P046 [Start Source 1]: start command source
P047 [Speed Reference 1]: speed reference source
C123 [RS485 Data Rate]: RS485 communication speed
C127 [Comm Format]: communication format
C128 [Comm Write Mode]: communication write mode
If the drive must start and receive speed reference through Modbus, the start source and speed reference source must be set to the serial or DSI communication source. Otherwise, the PLC may communicate with the drive but still fail to start it.
In the PLC program, the PowerFlex 525 should have its own control block, for example:
VFD1_PowerFlex525
It should not share the same internal register mapping as the ATV310. The external interface can look the same for all drives, such as Start, Stop, Reset, Speed_Set, Run_Status, Fault_Status and Comm_Error. But the internal Modbus read/write addresses must match the specific drive model.
7. Recommended PLC Program Structure: Common Interface, Separate Drive Blocks
In a multi-drive system, the operator wants a consistent HMI interface, but each drive may require different Modbus registers. The best PLC structure is therefore:
Common HMI interface
Different internal Modbus blocks for each drive
For example, the HMI can provide the same commands for every VFD:
Start
Stop
Reset
Speed reference
Running status
Fault status
Communication status
Actual frequency
Actual current
But inside the PLC, each drive should have its own logic section or function block:
VFD1_PowerFlex525
VFD2_ATV310_7K5
VFD3_ATV310_5K5
VFD4_ATV310_5K5
VFD5_ATV31P_5K5
Each block should handle:
1. Generate internal run command from HMI request.
2. Check permissive conditions.
3. Generate the correct command word for that drive.
4. Write speed reference.
5. Read status word.
6. Decode running and fault states.
7. Detect communication error.
8. Send status back to the HMI.
This structure keeps the main program clear. The HMI remains consistent, while the brand-specific details are isolated inside each drive control block.
8. Ladder Logic Should Not Copy a Direct Motor Starter Diagram
A normal motor starter ladder diagram usually contains:
Start button → latch circuit → contactor output
Stop button → break output
Timer → start next motor
This logic is useful for learning ladder basics, but it cannot be directly used for Modbus drive control. A Modbus-controlled VFD does not start because a PLC output coil turns on. It starts because the PLC writes the correct command word to the correct Modbus register.
The correct logic should be:
Start button or HMI start command
↓
PLC internal run command
↓
Check safety and permissive conditions
↓
Write drive command word
↓
Write frequency reference
↓
Read drive status word
↓
Display running, stopped, fault or communication error on HMI
The PLC physical outputs Q0.0, Q0.1, etc. may not be used for drive start at all. The main control is through Modbus. Physical inputs and outputs are mainly used for emergency stop, interlock, relay feedback and backup hardwired protection.
A basic internal run latch can be designed with the following conditions:
Start condition:
HMI_START = 1
Emergency stop OK = 1
Drive R1 feedback OK = 1
Modbus communication OK = 1
Drive fault = 0
When all conditions are satisfied:
RUN_CMD = 1
Stop conditions include:
HMI_STOP = 1
Emergency stop active
R1 feedback abnormal
Communication fault
Drive fault
If any stop condition is present:
RUN_CMD = 0
Then the Modbus command is generated from RUN_CMD:
RUN_CMD = 1 → write start command
RUN_CMD = 0 → write stop command
The frequency reference is written separately, for example:
HMI_SPEED_SET = 10.0 Hz
PLC converts it to the correct register value
PLC writes it to the drive speed reference register
9. R1 Relay Feedback Is Auxiliary Protection, Not a Replacement for Modbus Status
The ATV310 has relay terminals R1A, R1B and R1C. In most cases:
R1C = common
R1A = normally open contact
R1B = normally closed contact
R1A and R1B are not two separate fault signals. They are opposite contacts of the same relay. If R1C is connected to +24 V, R1A is connected to one PLC input, and R1B is connected to another PLC input, the two inputs will normally show opposite states.
For example:
Relay energized: R1A-R1C closed, R1B-R1C open
Relay de-energized: R1B-R1C closed, R1A-R1C open
In practical engineering, one contact per drive is normally enough for a fault or permissive feedback signal. A normally closed contact is often preferred for protection because a broken wire can also be detected as an abnormal condition.
A typical connection can be:
+24 V → R1C
R1B → PLC input
PLC input common → 0 V
However, the actual meaning of the input depends on how the drive relay function is configured. The R1 relay can be assigned to fault, run, ready, frequency reached or other functions. The PLC program must clearly define what the input means. It should not mix “running feedback,” “fault feedback” and “start permissive” without clear logic.
The best practice is to use both Modbus status and relay feedback:
Modbus status word: running, stopped, fault code, frequency, current
R1 relay: hardwired permissive or fault protection
If only R1 is used, the PLC cannot know the detailed fault. If only Modbus is used, a communication failure may remove an important hardwired protection path. Combining both gives a more robust system.
10. Each VFD Should Have Its Own Fault Feedback Input
For a five-drive system, each drive should ideally have a separate relay feedback input to the PLC:
VFD1 R1 → PLC input
VFD2 R1 → PLC input
VFD3 R1 → PLC input
VFD4 R1 → PLC input
VFD5 R1 → PLC input
This allows the HMI to display:
VFD1 fault
VFD2 fault
VFD3 fault
VFD4 fault
VFD5 fault
If all R1 contacts are combined into one general fault input, the PLC will only know that one drive has a problem, but it will not know which one. This makes troubleshooting slower and less professional.
If PLC inputs are limited, a compromise can be made. Critical drives can have separate hardwired feedback, while less critical devices can rely on Modbus status. But in a new system design, saving a few input points should not be more important than clear fault diagnosis.
11. Modbus Wiring Must Consider Bus Topology, Shielding, Termination and Grounding
RS485 / Modbus RTU wiring is simple in principle, but it is sensitive to poor installation. VFDs are strong sources of electrical noise. PWM motor cables, braking resistors, contactor coils, grounding problems and power harmonics can all affect communication quality.
Important wiring rules include:
- Use twisted shielded cable for RS485 communication.
- Do not run communication cable parallel to motor power cables for long distances.
- Keep the bus structure clear and avoid random star wiring.
- Use termination resistors at the ends of the bus when required.
- Handle shield grounding correctly to avoid ground loops.
- Do not reverse D0 and D1.
- Do not treat RJ45 Modbus ports as normal Ethernet ports.
- Each slave address must be unique.
- All serial settings must be the same.
- Add one device at a time during commissioning.
Many Modbus failures are not caused by the PLC program. They are caused by wiring topology, shielding, grounding, termination or duplicated addresses. When five VFDs are connected to the same bus, it is much better to add and test one device at a time.
12. Commissioning Should Start with One Drive, Not the Entire System
A reliable commissioning sequence is essential. The recommended order is:
- Confirm PLC model, HMI model, VFD models and communication interfaces.
- Confirm the function and wiring of the LU9GC3.
- Connect only one ATV310 first.
- Set its Modbus address, baud rate, parity and stop bit.
- Set its command source and speed reference source to Modbus.
- Configure the M221 serial line as Modbus RTU master.
- Read one status register from the ATV310.
- Write a low frequency reference, such as 10 Hz.
- Write the start command.
- Confirm that the drive starts.
- Write the stop command.
- Confirm that the drive stops.
- Read fault and running status.
- Copy the logic to the other ATV310 drives.
- Commission the ATV31P separately.
- Commission the PowerFlex 525 separately.
- Integrate the HMI operation screen and alarm screen.
- Test emergency stop, communication loss, relay feedback and fault reset.
- Finally run the motors under correct power supply conditions.
The worst approach is to connect all five drives, write all the logic, power everything up and then start troubleshooting. When many problems exist at the same time, it becomes difficult to know whether the fault is caused by addressing, parameters, wiring, registers, power supply or logic.
13. Do Not Run VFDs from an Improvised Single-Phase Capacitor Supply
In some field situations, the available power supply may not match the VFD input voltage. For example, the site may only have 240 V three-phase power, while the drives or equipment require 380 V, 400 V or 480 V three-phase input. Sometimes a technician may try to create a temporary “480 V single-phase” supply and use a capacitor to simulate a missing phase.
This is not a proper power supply for VFD operation.
A VFD input rectifier requires a stable and balanced AC power source. A single-phase high-voltage supply with a capacitor does not become a real three-phase supply. It may cause:
Severe input voltage imbalance
DC bus ripple
Input phase loss fault
Undervoltage or overvoltage fault
Rectifier bridge overheating
Pre-charge circuit damage
Capacitor overheating or explosion
Ground fault or leakage trip
Permanent drive damage
If the site only has 240 V three-phase power and the drives require 380/400/480 V three-phase input, the correct solution is:
240 V three-phase supply
↓
Three-phase step-up transformer
↓
380/400/480 V three-phase supply
↓
VFD input
A properly sized three-phase transformer, correct grounding and correct protection devices are required. PLC, HMI and Modbus communication can be tested first without running the motor. High-voltage drive operation should only be tested when the power supply is correct and safe.
14. HMI Design Should Include Operation, Status and Alarms
In a multi-drive system, the HMI should not only have start and stop buttons. A practical HMI should include operation commands, real-time status and alarm information.
Single-Drive Control Area
Each VFD should have its own control section:
Start
Stop
Reset
Speed reference
Forward / reverse selection
Manual / automatic mode
Single-Drive Status Area
Each VFD should display:
Communication OK / communication fault
Running / stopped
Fault status
Actual frequency
Output current
Output voltage
Fault code
R1 relay input status
Overview Screen
The overview screen should show the status of all drives:
VFD1: running / fault / communication error
VFD2: running / fault / communication error
VFD3: running / fault / communication error
VFD4: running / fault / communication error
VFD5: running / fault / communication error
Alarm History
A good system should record:
Which drive faulted
Fault time
Reset time
Communication loss time
Emergency stop time
If the HMI only has start and stop buttons, maintenance personnel will not have enough information to diagnose problems quickly.
15. Example PLC Variable Planning
Clear variable naming is important for maintenance. For example, for the second drive, an ATV310, the PLC variables can be:
VFD2_Start_HMI
VFD2_Stop_HMI
VFD2_Reset_HMI
VFD2_Speed_Set
VFD2_Run_Cmd
VFD2_Command_Word
VFD2_Frequency_Ref
VFD2_Status_Word
VFD2_Running
VFD2_Fault
VFD2_Comm_Error
VFD2_R1_OK
VFD2_Actual_Freq
VFD2_Actual_Current
For the PowerFlex 525, similar naming can be used:
VFD1_Start_HMI
VFD1_Stop_HMI
VFD1_Reset_HMI
VFD1_Speed_Set
VFD1_Run_Cmd
VFD1_Command_Word
VFD1_Frequency_Ref
VFD1_Status_Word
VFD1_Running
VFD1_Fault
VFD1_Comm_Error
The naming style should be consistent. This makes it easier to copy logic, link HMI objects and troubleshoot the system later. The brand-specific differences should be handled inside the drive block, not scattered throughout the HMI and main program.
16. Run Command and Safety Permissives Must Be Separated
The PLC program should separate the operator request from the actual permission to run. An HMI start button only means the operator wants to start the drive. It does not mean the drive is allowed to start immediately.
Before the PLC sends a start command, it should check:
Emergency stop OK
Safety door OK
Drive not faulted
Communication OK
R1 feedback OK
Motor protection OK
Process conditions OK
Valid speed reference
No interlock conflict
The logic should be layered:
HMI_START → operator request
RUN_PERMISSION → permissive conditions
RUN_CMD → internal PLC run command
MODBUS_COMMAND → actual command written to the VFD
This prevents accidental operation and makes diagnostics easier. If the operator presses Start but the drive does not run, the HMI can show “start condition not satisfied” and then indicate whether the problem is R1 feedback, communication fault, emergency stop, drive fault or another interlock.
17. Stop Command Must Have Higher Priority Than Start
In any motor control system, stopping and fault handling must have higher priority than starting. The PLC logic should follow these rules:
Stop has priority over start.
Fault has priority over run.
Emergency stop has priority over normal stop.
Communication fault should inhibit start.
If HMI_START and HMI_STOP are active at the same time, the PLC should stop the drive.
If RUN_CMD is active but R1 feedback becomes abnormal, the PLC should remove the run command.
If Modbus communication is lost, the PLC should stop sending run commands and display a communication fault.
The VFD’s own communication-loss behavior should also be configured. Some drives can be set to coast stop, ramp stop, hold last command or trip on communication timeout. For most industrial systems, communication loss should cause a stop or a fault response, not continued operation with the last command.
18. Main Challenges in a Mixed-Brand VFD System
The difficulty in this type of project is not only one drive. It is the mixed-brand system. The main challenges include:
- Schneider ATV310 and Allen-Bradley PowerFlex 525 use different Modbus register maps.
- The command word bit definitions may be different.
- The speed reference scaling may be different.
- The status word decoding may be different.
- The fault reset command may be different.
- The communication parameter menus are different.
- Some drives default to terminal control, while others may default to keypad control.
- Communication timeout behavior must be configured.
- The HMI should be unified, but the low-level Modbus control cannot be identical.
- The field power supply may not be suitable for all drives.
Therefore, the project should not be reduced to “how to draw the ladder.” The complete design must include architecture, parameters, addresses, communication, power, safety and commissioning sequence.
19. Recommended Implementation Procedure
For similar projects, the following implementation procedure is recommended.
Step 1: Build a Device List
Record the following information for each device:
Brand
Model
Power rating
Input voltage
Communication interface
Modbus address
Communication format
Command source
Speed reference source
Step 2: Confirm Power Supply Conditions
Check:
Voltage level
Number of phases
Available capacity
Grounding
Circuit breaker
Leakage protection
Transformer capacity
Motor rated data
If the power supply is not suitable, fix the power supply first. Do not force the VFDs to run from an unsafe temporary source.
Step 3: Test One Drive Communication
Connect only one drive and confirm that the PLC can read and write data.
Step 4: Test One Drive Operation
Run at low frequency, preferably no load or light load, and confirm start and stop.
Step 5: Test Status Reading
Read running state, stopped state, fault state, current and frequency.
Step 6: Test R1 Relay Feedback
Create or simulate a status change and confirm that the PLC input changes correctly.
Step 7: Add Drives One by One
Add one drive to the Modbus bus at a time and test after each addition.
Step 8: Commission the HMI
Link HMI buttons, setpoints, status indicators and alarms.
Step 9: Test Safety Interlocks
Test emergency stop, communication interruption, drive fault, broken feedback wire and power recovery.
Step 10: Run with Load
Under correct three-phase power, gradually increase speed and monitor current, temperature and mechanical behavior.
20. Conclusion
Using an M221 PLC to control multiple VFDs through Modbus RTU is a practical and powerful industrial control solution. It reduces hardwiring, improves system integration, and allows the HMI to display drive status, fault information, current and frequency. But such a system cannot be understood as simply “the PLC sends one start command and the drive runs.” A reliable system must handle communication roles, unique slave addresses, serial settings, register differences, command word logic, relay feedback, HMI design, safety interlocks and correct power supply conditions.
In a mixed system using Schneider ATV310, Schneider ATV31P and Allen-Bradley PowerFlex 525, the most important rules are:
Use one clear architecture.
Assign unique slave addresses.
Keep communication settings consistent.
Use different Modbus register maps for different drives.
Build modular PLC logic.
Commission one drive at a time.
The HMI should be only the operator interface. The PLC should be the only Modbus RTU master. The drives should be Modbus slaves. The LU9GC3 should be treated as a Modbus serial distribution module, not as an Ethernet switch. The PowerFlex 525 Modbus address should be set through C124, and its start source, speed reference source and communication format must be configured separately. The ATV310 and ATV31P should be configured according to the Schneider drive communication table and control profile.
Power supply conditions must also be respected. If a drive requires 380 V, 400 V or 480 V three-phase input, but the site only has 240 V three-phase power, the correct solution is a properly sized three-phase step-up transformer. A single-phase high-voltage supply with a capacitor cannot replace a real three-phase supply and should not be used for normal VFD operation.
The safest commissioning strategy is to divide the project into small, verifiable steps. First test one drive, then expand to multiple drives. First read status, then write speed. First run at low speed, then test under load. First make communication stable, then complete the HMI and automatic logic. By following this method, a complex multi-drive Modbus RTU system can be transformed into a clear, maintainable and reliable industrial control project.




























