Author: baiyuzhan

Posted on by

Troubleshooting a KNC-400 Online Film Thickness Gauge That Returns to Zero but Does Not Start Continuous Scanning

1. Background of the Fault

Online film thickness gauges are widely used in blown film, cast film, composite film, packaging film, and other plastic film production lines. Their main function is to continuously monitor the thickness of the film during production and provide real-time data to operators or to an automatic control system. In many film extrusion lines, the measuring head is mounted on a circular scanning carriage. During normal operation, the carriage moves around the film bubble or across the measuring path, allowing the gauge to build a complete thickness profile.

A KNC-400 type film thickness measuring system is not just a single sensor. It is a complete measuring and motion-control system. It usually includes the measuring head, circular scanning carriage, drive motor, guide rail or belt mechanism, pneumatic air-bearing or air-gap control, proximity switch or reference-position sensor, data processor, industrial I/O module, communication interface, and upper-level display software.

In this case, the customer reported the following symptoms. After power-on, the small local display showed “Warm Up / waiting” for more than twenty minutes without any obvious change. When F1 was pressed to start measurement, the measuring carriage only shook slightly and did not start continuous scanning. However, after the carriage was manually pushed away from the zero position, pressing F1 caused the carriage to move. It passed two mechanical reference blocks and then returned to the zero position, where it stopped. After pressing F1 again to stop and restart, the same symptom returned: the carriage only moved slightly and stopped.

The upper computer displayed the Kdesign software interface, including pages such as Trend, Polar diagram, Linear diagram, Alarm list, and production data. The alarm history included messages such as “No communication to the kundig measuring device – Check Power Supply and data link” and repeated “Valve error compare special ALARM-page appeared/disappeared” records.

Several field observations are especially important. The measuring head had continuous compressed air blowing. The pressure display fluctuated around 290–300 mBar. When the pressure was manually increased, it later returned to the original range, which suggests that the system may have automatic pressure regulation. The ME and SE values changed synchronously when the pressure was adjusted. The small display on the measuring head showed changes before and after pressing F1. The measuring head status lamp remained green. The control cabinet contained Phoenix Contact I/O modules, power supply modules, a data processor board, and an RS-485 communication board, all with indicator lights. However, when the carriage passed the two reference blocks, the customer did not observe any obvious change in the Phoenix Contact input module indicators or the LEDs on the small control board.

At first glance, this fault can easily be mistaken for a motor stall, mechanical jam, or drive failure. However, the later tests show that the carriage can move when pushed away from the zero position and can return to zero under F1 command. Therefore, the failure cannot simply be attributed to a bad motor or a completely jammed mechanical system. The more likely fault area is the transition between “homing completed” and “continuous measuring scan allowed.”

This case is technically valuable because it shows a common problem in online measuring equipment repair: the apparent symptom is “the machine does not move,” but the root cause may be in position feedback, pressure control, communication, measurement permission logic, or software status, rather than in the motor itself.

KNC-400 circular online film thickness gauge showing the circular rail, scanning carriage, measuring head, proximity switch, 0-degree home position, and 38-degree and 87-degree reference markers around a blown film line.

2. Basic Operating Logic of a KNC-400 Online Thickness Gauge

To diagnose this kind of fault correctly, it is necessary to understand the normal operating sequence of a circular scanning film thickness gauge.

After power-on, the data processor and the measuring head usually enter an initialization stage. During this stage, the system checks the power supply, internal communication, measuring head status, temperature condition, pneumatic pressure condition, position sensor condition, and external interlocks. If the measuring head or measuring environment requires stabilization, the local display may show messages such as “Warm Up,” “waiting,” or similar status indications.

When the operator presses F1 to start measurement, the system does not necessarily begin continuous scanning immediately. In many circular scanning systems, the carriage first performs a homing or reference-position search. The controller must know where the carriage is before it can start a complete measuring cycle. This reference position is normally detected by a proximity switch, optical sensor, Hall sensor, or another position feedback device.

Only after the controller confirms the correct reference position, and after all measuring conditions are satisfied, does the carriage enter continuous scanning mode. During continuous scanning, the measuring head collects film thickness data and sends it to the data processor and upper computer. The software then displays actual profiles, basic centering profiles, linear diagrams, trend diagrams, maximum and minimum thickness values, average thickness, and other measurement data.

In this case, the customer confirmed an important point: in the past, the KNC-400 could start and run even when no film was being produced. It simply had no valid thickness data. This means the current problem is not caused merely by the absence of film. The system should be able to perform an empty scan. Therefore, the current failure is more likely caused by a missing internal permission signal, abnormal position feedback, pressure control problem, communication issue, or measurement state error.

The fact that the carriage can move after being manually pushed away from zero means that F1 is accepted by the system and the motion control loop still has basic functionality. The motor, transmission, belt, and guide rail cannot be considered completely failed. The key issue is that after the carriage returns to the zero position, the system does not continue into normal scanning.

Fault sequence diagram of a KNC-400 circular scanner showing the carriage moving from the measuring command through the 87-degree and 38-degree reference positions, returning to 0-degree home, and stopping instead of continuing the scan.

3. Analysis of the Main Fault Phenomena

3.1 The “Warm Up / waiting” Message Does Not Disappear

The “Warm Up / waiting” indication does not automatically mean a hardware fault. Many online measuring systems require a warm-up period before measurement is allowed. The system may wait for the measuring head temperature, internal electronics, pneumatic air gap, or communication state to stabilize.

However, if the system remains in this state for more than twenty minutes without any progress, it usually means that one of the measuring permission conditions has not been met. Possible causes include:

The actual measuring head temperature has not reached the target value.
The air pressure has not reached the required range.
The measuring head communication is abnormal.
The data processor has not received a valid measuring head status.
An external interlock signal is missing.
The carriage position or zero reference is not confirmed.
The measuring head remains in a stopped, waiting, or initialization state.
The system parameter or internal status is abnormal.

In this case, the local display showed a target temperature of 32.0°C. At first, only the target temperature was available, while actual temperature values such as ME, SE, Actual, or Current were not clearly identified. Later, the customer reported that ME and SE changed when the air pressure was adjusted. This proves that the measuring head is not completely dead; at least part of its sensing and display functions are active.

Therefore, “Warm Up / waiting” should be treated as a general waiting status, not as a single fault code. It may be caused by temperature, pressure, communication, position feedback, or external interlock conditions.

3.2 Pressing F1 Causes Only a Slight Shake

When F1 is pressed and the carriage only shakes slightly, it is tempting to suspect a blocked motor, jammed carriage, damaged belt, or failed drive output. But the later field test does not support this conclusion.

After the carriage was manually pushed away from the zero position, pressing F1 caused it to move and return to the zero point. This proves that the motor and transmission can produce effective motion. If the motor were completely stalled, or if the mechanism were seriously jammed, the carriage would not be able to perform this movement.

A more reasonable explanation is that when the carriage is already near the reference position, the controller only makes a short positioning or confirmation movement. Because the next permission condition is not satisfied, the controller does not start continuous scanning. As a result, the customer sees only a small shake.

This kind of symptom is common in automated equipment. The machine appears not to run, but in reality it is waiting for the next logical condition. The difference between “cannot move” and “not allowed to continue moving” is very important in fault diagnosis.

3.3 The Carriage Can Return to Zero After Being Manually Moved

This is the most important observation in the whole case.

It proves that the F1 command is recognized.
It proves that the motion system has at least partial functionality.
It proves that the motor and transmission are not completely defective.
It proves that the system can perform a homing-related action.
It suggests that the failure occurs after the homing action is completed.

In many industrial systems, the machine can return to home but cannot enter automatic operation. This usually means the problem is not the basic motion hardware, but the automatic-cycle enable condition. Examples include missing reference confirmation, missing safety input, missing process-ready signal, abnormal pressure, communication timeout, or incorrect process state.

For the KNC-400 in this case, the most likely point of failure is the logic between “home position found” and “continuous measuring scan started.”

3.4 Pressure Fluctuation Around 290–300 mBar

The pressure display fluctuates around 290–300 mBar. When F1 is pressed, the pressure changes. The customer also reported that manual pressure adjustment affects ME and SE values, but the pressure later returns toward the original value. This suggests that the pneumatic system may be closed-loop controlled, rather than purely manually regulated.

In an air-bearing or air-gap measuring head, stable pressure is critical. If the pressure is too low, the measuring head cannot maintain a stable air cushion or measuring distance. If the pressure is too high, it may disturb the film or shift the measuring geometry. If the controller compares target pressure and actual pressure, a deviation may trigger a valve or pressure comparison alarm.

The alarm history contains repeated “Valve error compare” messages. This may indicate that the valve control system, pressure feedback, or pressure comparison logic has detected an inconsistency.

However, the presence of 290–300 mBar pressure means the pneumatic system is not completely inactive. The ME and SE values respond to pressure changes, which indicates that the measuring head and air system have dynamic response. Therefore, the pneumatic system may be abnormal, but it should not be assumed to be the only fault without further confirmation.

The key question is not simply “is there pressure?” but rather:

What is the target pressure?
What is the actual pressure?
What is the allowable tolerance?
Does the actual pressure reach the target pressure during F1 startup?
Is the “Valve error compare” alarm active at the moment of failure, or only historical?
Does the pressure deviation prevent the system from entering measuring mode?

If the target pressure is 300 mBar and the actual pressure is stable around 290–300 mBar, pressure may not be the main cause. If the target pressure is higher and the actual value cannot reach it, then the pressure control loop must be investigated.

3.5 The Proximity Switch and Reference Block Signals Are Unclear

The field inspection originally suggested that there were three mechanical reference blocks on the circular rail. Later, the customer confirmed that there were only two blocks and only one proximity switch, which was partly hidden inside the carriage.

The customer tried touching the proximity switch with a copper sheet but did not observe an indicator light flashing on the measuring head. This test is not reliable. Many industrial proximity switches are inductive sensors, and they respond best to ferrous metal such as steel or iron. Copper and aluminum greatly reduce the sensing distance. A copper sheet may not trigger the switch even if the switch is good.

The correct test is to use a steel screwdriver, steel screw, or iron plate near the sensing face, while measuring the output voltage with a multimeter. The LED on the sensor may be hidden, dirty, damaged, or not visible from the current viewing angle. Therefore, the electrical output must be measured.

The customer also reported that when the carriage passed the two reference blocks, the Phoenix Contact input module indicators did not appear to change. This may be important, but it must be interpreted carefully. The visible indicator may not correspond to the proximity switch input. It may belong to another input, output, status, or communication signal. The correct terminal must be identified by tracing the sensor cable.

If the reference switch signal is abnormal, the system may behave in several ways:

The carriage may return to zero but the controller may not confirm homing completion.
The controller may believe the carriage is always at zero.
The controller may believe a limit condition is permanently active.
The controller may complete homing but fail to switch into scan mode.
The controller may produce only a short movement when F1 is pressed.
The system may remain in waiting or stopped state.

For this reason, the reference-position sensor and its wiring must be treated as a top-priority inspection item.

Diagnostic illustration of a KNC-400 scanner carriage with a technician checking the proximity switch, 24 VDC supply, signal output, reference target, and I/O input module using a multimeter.

4. Why This Is Unlikely to Be a Simple Motor Stall or Mechanical Jam

The customer asked whether the fault could be caused by motor stall or mechanical jamming. Based on the available evidence, this is not the most likely diagnosis.

A true motor stall usually has typical features: high motor current, abnormal motor heating, drive alarm, inability to move regardless of position, obvious mechanical resistance, belt slipping, gear jumping, or repeated failed movement attempts. A severely jammed carriage would also be difficult to move manually and would not be able to return to zero over a longer distance.

In this case, after the carriage was pushed away from zero, it moved under F1 command and returned to zero. This means the motor, drive, belt, guide rail, and carriage are capable of movement. The fault is more consistent with a control sequence problem than a basic motion hardware failure.

This does not mean the mechanical system should be ignored. The circular guide rail, rollers, belt tension, reference blocks, carriage bearings, and cable chain should still be inspected. Dirt, wear, local friction, misaligned blocks, or loose mechanical parts can cause unstable movement. But based on the available symptoms, mechanical blockage is not the first suspect.

The key difference is this: the carriage does not fail to move because it lacks mechanical capability; it stops because the control logic does not allow it to enter continuous scanning.

KNC-400 circular scanner troubleshooting guide showing position feedback, pneumatic pressure, data processor communication, target status versus actual status, and the diagnostic flow from F1 start to home position and scan interruption.

5. Correct Method for Testing the Proximity Switch

The proximity switch is one of the most important parts to verify. In a circular scanning thickness gauge, the controller must know the reference position. If the reference signal is wrong, the entire measuring cycle can be blocked.

A common three-wire proximity switch uses the following wiring convention:

Brown wire: +24 VDC
Blue wire: 0 VDC
Black wire: signal output

This is a common industrial convention, but the actual wiring should still be confirmed from the sensor label or wiring diagram.

The correct test procedure is as follows.

First, measure the supply voltage between brown and blue. It should normally be approximately 24 VDC. If there is no 24 VDC, the sensor has no power. The fault may be in the power supply, terminal block, fuse, cable, connector, or common line.

Second, measure the output voltage between black and blue. Move a steel object toward and away from the sensing face. The voltage should change clearly. For a PNP sensor, the output may change from 0 V to 24 V when activated. For an NPN sensor, the output may change from 24 V to 0 V when activated. The exact direction is less important than the fact that it must change reliably.

Third, trace the signal to the input module. A sensor output change at the sensor itself does not prove that the controller receives the signal. The same signal must be checked at the terminal block, connector, cable chain, Phoenix Contact input module, and data processor input.

Fourth, check whether the signal is stable. A proximity switch can be partially faulty. It may switch only at a very short distance, flicker because of contamination, or fail when the carriage moves. Long-term vibration, metal dust, cable fatigue, and connector oxidation can all cause intermittent switching.

Fifth, test the sensor with the actual mechanical reference block. A handheld steel tool is useful for initial testing, but the final test must verify that the real reference block triggers the sensor at the correct position and distance.

Using copper for this test is not recommended. Copper may not trigger an inductive proximity sensor reliably, so a “no response” result with copper does not prove the sensor is defective.

6. Pressure Control and Valve Error Diagnosis

The repeated alarm history related to “Valve error compare” suggests that the pressure control loop must be checked. In an air-gap measuring system, the controller may compare the target air pressure with the measured actual pressure. If the difference exceeds a threshold, it may block measurement or generate an alarm.

The field pressure reading of approximately 290–300 mBar may be normal, but this cannot be confirmed unless the target pressure is known. The display showed “Pressure 300 mBar” in one screen, which may be either a target or actual value depending on the menu. The temperature target was 32.0°C. The pressure target and pressure actual must be distinguished clearly.

The following checks are recommended.

Record the pressure before pressing F1.
Record the pressure during F1 startup.
Record the pressure after the carriage returns to zero.
Find the pressure target or pressure setpoint in the local menu.
Check whether the actual pressure reaches the target.
Check whether the valve error appears as an active alarm during the failure.
Check the air filter, regulator, tubing, solenoid valve, proportional valve, and measuring head nozzle.
Check whether the pressure sensor output is stable.

The fact that manual adjustment is followed by automatic return may indicate a closed-loop pressure controller. Therefore, the operator should not randomly change the pressure setting. Incorrect pressure may affect measurement calibration and cause additional error.

If the valve error is active during F1 startup, the pneumatic control loop may be preventing continuous scanning. If the valve error is only historical and does not reappear after clearing alarms, it may not be the immediate cause.

7. Meaning of Target Status and Actual Status

The local data processor menu showed status information such as Target status and Actual status. This distinction is important.

Target status refers to the state requested by the operator or upper-level system. For example, after pressing F1, the target status may become “measuring.” This only means that the system has been commanded to measure.

Actual status refers to the real state reported by the measuring system. If the actual status remains “waiting” or returns to “stopped,” the equipment did not truly enter measuring mode, even if the target status says “measuring.”

In this case, the customer observed that after pressing F1, Target status changed to measuring. After pressing F1 again to stop, Actual status changed to stopped. This means the command path is not completely broken. The data processor receives the operator command and changes the requested state. But the system may not be able to maintain actual measuring operation.

This difference is critical. Repeatedly pressing F1 will not solve the problem if the actual measuring permission is missing. The correct direction is to identify why the actual status does not remain in measuring after homing.

Possible reasons include:

Reference position not confirmed.
Pressure condition not satisfied.
Measuring head not ready.
Temperature condition not satisfied.
Communication abnormal.
External interlock missing.
Data processor parameter or status abnormal.
Input module signal missing.
Scan enable logic not satisfied.

8. Communication Alarm Analysis

The alarm history included “No communication to the kundig measuring device – Check Power Supply and data link.” This message should not be ignored. It indicates that, at least at some point, the data processor or upper computer lost communication with the measuring device.

Possible causes include unstable power supply, loose RS-485 wiring, poor connector contact, broken cable in the cable chain, communication board fault, measuring head power fault, shielding problem, or intermittent data link failure.

However, the later field evidence shows that the measuring head display works, the local processor menu is accessible, and the status values change. Therefore, the communication problem may be intermittent or historical rather than a complete current failure.

Still, the communication path should be checked carefully, especially because the measuring carriage moves. Cable-chain wiring is a common failure point in moving measuring systems. A cable may appear normal when stationary, but lose contact when the carriage reaches a certain position. This can cause intermittent communication alarms, sensor signal loss, or missing measurement data.

Recommended checks include:

Inspect all RS-485 terminal screws.
Check the shielding and grounding.
Check the cable chain for bending damage.
Move the carriage slowly while observing communication indicators.
Gently shake the cable at different carriage positions.
Clear historical alarms and check whether communication alarms reappear during F1 startup.
Measure data processor power supply stability.
Check the RS-485 board and connectors for oxidation or contamination.

If a communication alarm reappears exactly when the carriage moves or reaches a certain position, the cable chain or connector should be strongly suspected.

9. Most Probable Fault Chain in This Case

Based on all the available information, the most probable fault chain is as follows.

The KNC-400 powers on and enters a waiting state. When F1 is pressed, the upper computer or local processor issues a measuring command. If the carriage is away from zero, the system first performs a homing movement. The carriage moves past the reference blocks and returns to the zero position. After reaching zero, the system should transition into continuous scanning. However, one or more measuring permission conditions are not satisfied, so the scan does not start. When F1 is pressed again while the carriage is already near zero, the system only performs a short confirmation movement, which appears as a slight shake.

The most likely causes are:

Abnormal zero/reference proximity switch signal.
Incorrect or unstable signal transmission from the proximity switch to the input module.
The controller incorrectly believes the carriage is already at a limit or zero position.
The pressure control comparison condition is not satisfied.
The measuring head remains in waiting status.
The data processor does not receive a valid ready signal from the measuring head.
The external measuring enable or line-run interlock is missing.
The cable chain has an intermittent connection fault.
Historical or active valve/communication alarms are blocking the measuring cycle.

Among these, the proximity switch and its input signal should be checked first, because the movement behavior is strongly related to the zero/reference position.

10. Recommended On-Site Troubleshooting Sequence

A systematic troubleshooting sequence is necessary. Randomly replacing the motor, sensor, data processor, or measuring head may waste time and cost.

Step 1: Clear alarms and reproduce the fault

Record all current alarms first. Then clear the alarm list if the system allows it. Press F1 and reproduce the fault. The new alarms that appear during the fault are more important than old historical alarms.

Step 2: Record Target Status and Actual Status

Before pressing F1, record Target status and Actual status.
After pressing F1, record them again.
After the carriage returns to zero, record them again.
After pressing F1 to stop, record them again.

If Target status becomes measuring but Actual status does not remain measuring, the system command is received but the machine is not allowed to enter measuring mode.

Step 3: Confirm pressure target and actual pressure

Find the pressure-related menu in the local processor or upper software. Record the pressure target, actual pressure, and any pressure or valve alarms. Do not rely only on the 290–300 mBar display unless it is clear whether it is a target or actual value.

Step 4: Test the proximity switch electrically

Use a steel object, not copper. Measure the sensor power supply and output with a multimeter. Confirm that the output changes reliably when the reference block passes.

Step 5: Trace the proximity switch signal to the input module

Find the exact input channel receiving the proximity switch signal. Confirm that the signal changes at the Phoenix Contact module or data processor input. Do not judge from unrelated LEDs.

Step 6: Check the cable chain and moving cables

Move the carriage by hand or during homing while watching the signal and communication indicators. Intermittent cable faults are common in moving measuring devices.

Step 7: Check the pneumatic control loop

Inspect the air filter, regulator, proportional valve, tubing, fittings, and measuring head air outlet. Confirm that the actual pressure reaches the required target and does not oscillate beyond the allowed range.

Step 8: Check communication and power supply

Measure the DC power supply stability. Inspect RS-485 wiring and connectors. Check whether the communication alarm reappears during movement.

Step 9: Check external interlock signals

If the machine previously could run without film but now cannot, there may still be a missing external enable signal, changed parameter, disabled control mode, or lost line-run signal. Check the production line interface and input module signals.

Step 10: Consider board-level faults only after signal checks

Only after the sensor, pressure, communication, and interlock signals are confirmed should the data processor board, RS-485 board, input module, or measuring head electronics be suspected.

11. Practical Diagnostic Principles

Several principles are important in this type of repair.

Do not assume the motor is bad just because the carriage does not scan. If the carriage can return to zero, the motion hardware is at least partly functional.

Do not assume the pneumatic system is normal just because air is blowing. The target pressure and actual pressure must be compared.

Do not test an inductive proximity switch with copper and draw a conclusion. Use steel or iron and verify the output voltage.

Do not rely only on indicator lights. Measure the actual signal with a multimeter.

Do not confuse Target status with Actual status. A command to measure is not the same as actual measuring operation.

Do not ignore historical alarms, but do not let old alarms mislead the diagnosis. The most important alarm is the one that appears during the current fault.

Do not randomly adjust pressure, temperature, or calibration parameters. These settings may affect measurement accuracy.

Always suspect cable-chain wiring in moving systems. A cable can be normal when stationary and fail only during movement.

12. Conclusion

The KNC-400 fault in this case is unlikely to be a simple motor failure or a severe mechanical jam. The carriage can move after being manually pushed away from zero and can return to the zero position under F1 command. This proves that the basic movement system still works.

The real problem is that after the carriage returns to zero, the system does not enter continuous scanning. This points to a missing measurement permission condition, abnormal reference position feedback, pressure control comparison fault, communication problem, or external interlock issue.

The most important checks are the zero/reference proximity switch, its wiring to the input module, the pressure target versus actual pressure, the valve comparison alarm, Target status versus Actual status, and the moving cable chain. The proximity switch should be tested with a steel object and a multimeter, not with a copper sheet or by visual observation alone.

A correct diagnosis should follow the complete control sequence: power supply, communication, temperature, pressure, reference position, external enable, data processor status, and upper-computer command. Only by checking these conditions one by one can the actual reason for the KNC-400 failing to start continuous scanning be found.

For this type of online thickness gauge, the most effective repair strategy is not to replace parts blindly, but to determine exactly which condition blocks the transition from homing to measuring. Once that missing condition is identified, the repair path becomes clear: adjust or replace the proximity switch, repair wiring, restore pressure control, correct communication, or fix the relevant input or processor board.

Posted on by

In-Depth Technical Analysis of Parker 590P DC Drive AUX POWER Fault: From Auxiliary Power Supply to CODING and Three-Phase Synchronization Detection

1. Fault Background and Initial Symptom

The Parker 590P / Eurotherm 590 series DC drive is widely used in industrial DC motor speed control systems. Unlike a simple DC power controller, the 590P has a relatively complex internal power and detection structure. It has an independent auxiliary control supply, a three-phase mains input, a thyristor firing circuit, phase synchronization detection, phase-loss detection, power-board coding identification, and multiple status feedback signals sent to the CPU control board.

In this case, the observed fault was very typical but also easy to misjudge:

The drive powered up normally. The keypad display was normal. However, once a run/start command was issued, the drive immediately tripped with an AUX POWER alarm.

At first glance, this alarm seems to point directly to an auxiliary power supply problem. A common assumption would be that the internal switching power supply is unstable, or that one of the secondary outputs such as +5 V, +24 V, +15 V, or -15 V collapses when the drive is enabled. The CPU control board was replaced first, but the fault remained unchanged. This proved that the fault was not caused by the CPU board itself. The diagnosis therefore had to move toward the power/drive board, auxiliary supply detection circuit, three-phase mains detection circuit, CODING signal, PHASE signal, and the signals exchanged between the power board and the CPU board.

The important lesson from this case is that AUX POWER on a 590P must not be interpreted only as “the low-voltage switching power supply has failed.” In the Parker 590 series, this alarm can also be triggered by abnormal three-phase mains detection or abnormal coding/synchronization signals.

Parker 590P DC drive showing an AUX POWER alarm on the keypad with the internal power drive board exposed for troubleshooting.

2. Two Different Power Concepts in the 590P

To understand this fault correctly, two power systems inside the drive must be clearly separated.

The first is the single-phase auxiliary supply. The 590P normally has a separate auxiliary supply input, often 110 V or 220 V AC depending on configuration. This auxiliary input powers the internal switching power supply. The switching supply then generates the low-voltage rails used by the electronics, such as:

  • +5 V for logic and CPU-related circuits;
  • +24 V for relays, I/O, fan, and auxiliary control functions;
  • +15 V for analog circuits;
  • -15 V for analog circuits.

These voltages can be measured on the power board test points:

  • TP7: +5 V
  • TP6: +24 V
  • TP4: +15 V
  • TP5: -15 V
  • TP8: 0 V reference

The second is the three-phase mains input, typically L1/L2/L3. This is not only the main power source for the thyristor bridge and DC armature output. It is also used by the control system to generate synchronization information. A DC thyristor drive must know the phase position of the AC supply in order to fire the SCRs at the correct angle. If the phase detection is wrong, missing, unstable, or inconsistent with the expected coding signal, the drive cannot safely run.

Therefore, the three-phase input participates in:

  • phase synchronization;
  • phase-loss detection;
  • phase sequence tracking;
  • mains voltage range recognition;
  • SCR firing reference generation;
  • power-board / stack coding validation.

This is why an AUX POWER fault can still occur even when +5 V, +24 V, +15 V, and -15 V are all present and stable.

Parker 590P power drive board with labeled diagnostic test points including TP1 CODING, TP2 PHASE, +15V, -15V, +24V, +5V, 0V and T15 switching transformer.

3. Why the Low-Voltage Switching Power Supply Was Not the Main Fault in This Case

The initial suspicion was reasonable: if the drive powers up normally but trips immediately after the start command, the auxiliary switching power supply could be weak under load. On older industrial drives, this is common. Aging electrolytic capacitors, a weak UC2844 supply capacitor, poor secondary rectifiers, high ESR output capacitors, or bad solder joints around the switching transformer can all cause a supply to look normal at no load but collapse when the drive is enabled.

The 590P board in this case used a UC2844 PWM controller and a switching transformer, marked T15. Its secondary side generated the low-voltage rails. If T15’s secondary output were weak, one would expect to see one or more of the following:

  • +5 V dipping below about 4.7 V during start;
  • +24 V falling significantly under load;
  • +15 V or -15 V becoming unstable;
  • UC2844 entering undervoltage lockout or hiccup mode;
  • all secondary voltages pulsing or dropping simultaneously;
  • excessive ripple on the electrolytic capacitors near T15.

However, measurements were made at the test points for +15 V, -15 V, +24 V, and +5 V before and after the start command. No obvious voltage change was observed with a multimeter. Although a multimeter may miss very narrow transient dips, the later comparison with a known good power/drive board strongly shifted the diagnosis away from the switching supply itself.

The conclusion was that the basic auxiliary low-voltage supply was probably healthy. The original AUX POWER alarm was more likely caused by the detection and coding section associated with three-phase mains recognition and synchronization.

Technician measuring the TP1 CODING signal on a Parker 590P power drive board, comparing a stable 2.3V good board reading with abnormal faulty board coding voltages.

4. The Real Importance of the CODING Circuit

The 590 series manual describes the coding circuit as being located on the power board. It is not merely a simple fixed resistor identification circuit. It is associated with the generation of synchronization signals for the main processor and the thyristor stack. It also participates in phase-loss detection and automatic phase-sequence tracking.

This is the key point in this case.

The CODING circuit performs several possible functions:

  1. Hardware identification
    The CPU board must know what type of power board, voltage range, stack configuration, and hardware version it is connected to.
  2. Power stack / thyristor synchronization support
    The CPU requires correct timing information before it can fire the SCRs.
  3. Three-phase mains supervision
    If the three-phase input is missing, if one phase is lost, or if the phase detection chain is abnormal, the CPU may receive an invalid coding or phase signal.
  4. Fault classification
    Depending on how the signal fails, the drive may report different alarms, such as SEQ PRE READY, coding-related faults, or AUX POWER.

The 590C documentation also lists two important fault codes:

  • 0xF003: pre-ready fault / coding not present;
  • 0xFF03: auxiliary power fault, with the recommended action to check the auxiliary supply or the three-phase mains input.

This directly matches the field behavior in this case. When the CODING line was manually grounded, the drive displayed SEQ PRE READY, proving that the CPU actively reads this coding signal. But the original fault was AUX POWER, which indicates that the CODING line was not simply absent. Instead, the CPU was likely receiving an abnormal or unstable combination of coding, phase, or mains-status information during the start sequence.

Parker 590P AUX POWER fault diagnostic flow diagram showing three-phase mains input, phase coding detection, LM324 transistor network, TP1 CODING, TP2 PHASE and CPU board signal path.

5. Key Test Result: Good Board vs Faulty Board

The most important diagnostic breakthrough came from comparing the faulty power/drive board with a known good board.

On the good board:

  • Without three-phase 380 V mains applied, TP1 CODING was about 2.3 V DC.
  • With three-phase 380 V mains applied, TP1 CODING remained about 2.3 V DC, with only a slight change.

On the faulty board:

  • Without three-phase mains, TP1 CODING was about 1.4 V DC.
  • With three-phase mains applied but before starting, TP1 CODING rose to about 2.7 V DC.

This comparison is extremely important.

It shows that on a healthy board, TP1 CODING should be a relatively stable identification voltage. It may be related to the coding/synchronization system, but it should not be strongly pulled up or down by the presence of three-phase mains.

On the faulty board, the CODING voltage was already abnormal without three-phase input. It was too low at 1.4 V. When the three-phase supply was applied, it shifted too high to 2.7 V. This means the CODING node was being incorrectly affected by the three-phase detection circuit, PHASE detection circuit, transistor network, op-amp circuit, leakage path, or board contamination.

The fault was therefore not simply “no coding.” If CODING were completely missing, the drive would more likely report a 0xF003 / SEQ PRE READY type fault. Instead, the faulty board produced a wrong or unstable coding condition, which the CPU interpreted as an auxiliary power / mains input abnormality.

6. Why Grounding CODING Caused SEQ PRE READY

During testing, the CODING signal was grounded. The drive then displayed SEQ PRE READY.

This result is logical.

Grounding CODING forces the CPU to see an invalid hardware/coding state. The CPU no longer sees a valid power board identity or coding supply. As a result, it stops at the pre-ready stage and reports a coding-related fault.

This proves several things:

  • CODING is definitely read by the CPU board.
  • CODING is not an ordinary digital alarm line.
  • CODING cannot be grounded, shorted, or bypassed as a repair method.
  • The normal CODING voltage range is meaningful to the CPU.
  • An incorrect CODING level can change the alarm category.

In simple terms:

  • CODING completely invalid or missing → SEQ PRE READY / 0xF003 type fault.
  • CODING present but abnormal in relation to mains/phase detection → AUX POWER / 0xFF03 type fault.

This explains why forcing CODING low did not reproduce the original AUX POWER alarm. It created a different, more fundamental pre-ready fault.

7. The Three Transistors Connected to CODING

Another important observation was that the CODING line was connected to three transistors on the power/drive board. When two of them were removed, the CODING voltage rose to around 4.5 V. When all were removed, the voltage became around 0.5 V.

This proves that these transistors are not unrelated components. They are part of the CODING voltage-generation network.

Such a transistor network may be used for:

  • weighted analog coding;
  • hardware version identification;
  • power stack identification;
  • voltage class coding;
  • phase/mains status gating;
  • fault-state encoding;
  • switching resistor branches into or out of the CODING node.

If one transistor develops leakage, if a base resistor drifts, if a collector-emitter path becomes partially conductive, or if contamination creates a leakage path across the board, the CODING voltage can shift significantly. Because the normal voltage is only around 2.3 V, even a few hundred millivolts of offset may be enough to confuse the CPU.

In this case, the faulty board’s CODING voltage changed from 1.4 V to 2.7 V depending on three-phase mains presence. That is too large to be considered normal. The three-transistor CODING network is therefore one of the first areas to inspect and repair.

The correct repair approach is not to remove transistors and test whether the drive runs. Instead, restore the original circuit and compare each transistor’s base, collector, and emitter voltages against the good board.

8. The Role of LM324 Near the CODING Circuit

The board also has an LM324 near the CODING and PHASE test points. LM324 is a quad operational amplifier commonly used in industrial analog circuits. In this kind of drive board, it may be used for:

  • buffering analog coding voltage;
  • filtering phase detection signals;
  • generating weighted voltage levels;
  • conditioning mains detection signals;
  • summing or comparing several status inputs;
  • driving transistor networks.

If the LM324 has input leakage, output offset, damaged output stage, poor supply, or defective feedback components, it can easily shift the CODING voltage.

The LM324 should be checked carefully by comparing the good board and the faulty board. Important pins include:

  • Pin 4: positive supply;
  • Pin 11: negative supply or ground, depending on circuit design;
  • Pins 1, 7, 8, and 14: op-amp outputs.

The practical method is to measure these pins on both boards under the same conditions:

  1. auxiliary supply only;
  2. auxiliary supply plus three-phase mains;
  3. start command applied;
  4. alarm present.

If one LM324 output on the faulty board changes abnormally with the three-phase mains while the corresponding output on the good board remains stable, that op-amp channel or its surrounding resistor/capacitor network should be investigated.

9. PHASE Signal Must Be Checked Together With CODING

TP2 PHASE should not be ignored. Unlike CODING, which appears as a DC identification voltage, PHASE may be a shaped synchronization signal or a logic signal related to three-phase mains detection. A multimeter may not reveal much about it. An oscilloscope is the correct instrument.

A healthy PHASE signal should be stable when the three-phase mains is present. It should not disappear, jitter heavily, or collapse during the start command.

If TP1 CODING is abnormal and TP2 PHASE is also abnormal, the fault may lie upstream in the three-phase detection chain rather than in the CODING transistor network alone.

The three-phase detection chain may include:

  • L1/L2/L3 mains input;
  • contactor input and output;
  • sampling wires;
  • burnt or oxidized connectors;
  • high-value power resistors;
  • 47 nF Y2 capacitors;
  • optocouplers or isolation modules such as Schurter IF-0321-G;
  • LM393 comparator;
  • LM324 signal conditioning circuit;
  • transistor coding network;
  • TP1 CODING and TP2 PHASE;
  • CPU board input circuits.

Because the drive is a thyristor DC drive, phase synchronization is essential. If the CPU cannot trust the phase signal, it will not allow normal running.

10. The Burnt Connector and Contamination Problem

Several photos showed burnt or darkened connectors and wiring near the three-phase sampling/coding area. This is not a cosmetic issue.

A carbonized connector can cause:

  • high resistance contact;
  • intermittent signal loss;
  • leakage between adjacent pins;
  • unstable three-phase sampling;
  • abnormal analog coding voltage;
  • false phase-loss detection;
  • false AUX POWER alarm.

This is especially serious around high-impedance analog nodes such as CODING. A +24 V relay circuit may tolerate some dirt or contact resistance, but a 2.3 V analog coding node may be disturbed by very small leakage currents.

Any burnt connector in this part of the board should be replaced, not merely cleaned. The PCB surface should be thoroughly cleaned. If the board material is carbonized, the carbonized area should be scraped away and insulated. The terminals and wire crimps should also be replaced or re-crimped if they show heat damage.

11. Why Phase Sequence Alone Is Not the Main Suspect

It is correct that a DC thyristor drive must consider phase sequence and phase synchronization. However, the manual indicates that the coding circuit provides automatic phase-sequence tracking. This means that a simple L1/L2/L3 sequence reversal may not necessarily cause this exact alarm.

More likely causes include:

  • one phase not being detected;
  • one sampling resistor open or drifting;
  • one optocoupler channel weak;
  • one isolation module output abnormal;
  • PHASE signal missing;
  • CODING signal being pulled by the phase detection circuit;
  • contactor output unstable;
  • sampling connector burnt or intermittent;
  • board contamination causing leakage;
  • CPU receiving invalid coding voltage.

Swapping two phases can be used as a diagnostic comparison, but if the fault remains unchanged, the focus should return to phase detection and coding signal conditioning, not merely phase order.

12. Recommended Diagnostic Procedure

For a Parker 590P that powers up normally but trips with AUX POWER when started, the following sequence is recommended.

Step 1: Verify the low-voltage auxiliary rails

Use TP8 as the 0 V reference and measure:

  • TP7 +5 V;
  • TP6 +24 V;
  • TP4 +15 V;
  • TP5 -15 V.

Check these values:

  • with auxiliary supply only;
  • with three-phase mains applied;
  • during start command;
  • after the alarm.

If these voltages remain stable, the low-voltage switching supply is not the main suspect.

Step 2: Measure TP1 CODING

Compare the value with a known good board if possible.

In this case:

  • good board: about 2.3 V with or without three-phase mains;
  • faulty board: 1.4 V without three-phase mains and 2.7 V with three-phase mains.

This confirms an abnormal CODING circuit.

Step 3: Measure TP2 PHASE with an oscilloscope

A multimeter may not be enough. Confirm whether the PHASE signal is present, stable, and consistent when three-phase mains is applied and during the start command.

Step 4: Compare the CODING transistor network

With power off and capacitors discharged, compare the good and faulty boards:

  • TP1 to 0 V resistance;
  • TP1 to +5 V resistance;
  • TP1 to +15 V resistance;
  • TP1 to -15 V resistance;
  • TP1 to TP2 PHASE resistance;
  • TP1 to each transistor pin.

Any major deviation points to leakage or incorrect loading.

Step 5: Replace suspect CODING transistors and inspect resistors

If the CODING voltage is abnormal, the three transistors connected to CODING should be tested or replaced. Their base resistors, collector resistors, emitter resistors, small signal diodes, and filter capacitors should also be inspected.

Step 6: Check LM324 and surrounding components

Compare LM324 output pins on the good and faulty boards. Replace LM324 if one channel output is offset or reacts abnormally to three-phase input.

Step 7: Inspect three-phase sampling and isolation components

Check:

  • high-value sampling resistors;
  • 47 nF Y2 capacitors;
  • Schurter IF-0321-G modules;
  • optocouplers;
  • LM393 comparator;
  • solder joints;
  • burnt plugs;
  • wiring harness.

Step 8: Repair all burnt connectors and contamination

Do not leave carbonized connectors in the circuit. Replace damaged plugs, clean the PCB, repair solder joints, and ensure there is no leakage between signal traces.

13. Final Technical Conclusion

This case shows that the AUX POWER alarm on a Parker 590P DC drive can be misleading if interpreted too narrowly. Although the term suggests an auxiliary power supply fault, the actual detection logic also involves three-phase mains input, coding signal, phase synchronization, and power-board identification.

In this case, the low-voltage auxiliary outputs +5 V, +24 V, +15 V, and -15 V were stable before and after the start command. Therefore, the UC2844 switching power supply and T15 transformer section were not the main fault.

The decisive clue was TP1 CODING. On a good board, TP1 CODING remained approximately 2.3 V whether the three-phase 380 V supply was applied or not. On the faulty board, TP1 was approximately 1.4 V without three-phase supply and 2.7 V with three-phase supply. This proves that the faulty board’s CODING node was being abnormally pulled by the three-phase detection or coding network.

The most probable fault area is therefore:

  • CODING transistor network;
  • LM324 signal conditioning circuit;
  • TP1 surrounding resistors and capacitors;
  • PHASE / three-phase detection coupling path;
  • burnt sampling connectors;
  • isolation components such as IF-0321-G or optocouplers;
  • LM393 phase/mains comparator circuit;
  • PCB contamination or leakage.

The correct repair strategy is to restore the CODING voltage to a stable value close to the good board’s 2.3 V and ensure that TP2 PHASE remains valid during start. Once the CPU receives a valid coding voltage and reliable phase/mains detection signals, the AUX POWER alarm should no longer appear.

The key diagnostic principle is simple:

Do not treat AUX POWER only as a low-voltage power supply fault. On a 590P DC drive, always check the CODING and PHASE detection chain together with the auxiliary power rails.

Posted on by

Technical Analysis and Diagnostic Methods for ABB PSTX Soft Starter F0613 Shunt Fault

ABB PSTX series soft starters are widely used for starting three-phase asynchronous motors in applications such as pumps, fans, compressors, conveyors, crushers, mixers, cooling towers, and other industrial drive systems. Compared with direct-on-line starting by contactor, a soft starter gradually increases the motor terminal voltage through thyristor phase-angle control. This reduces starting current, limits mechanical shock, minimizes voltage dip on the supply network, and extends the service life of both the motor and the driven mechanical equipment.

The ABB PSTX series belongs to the more advanced category of soft starters. It integrates the control unit, thyristor power modules, current measurement, voltage measurement, protection logic, and an internal bypass mechanism. Because of this integrated design, the PSTX is able to monitor the main power circuit and detect abnormal operating states more precisely than a simple starter.

In practical repair work, F0613 Shunt Fault is a relatively common but often misunderstood fault on ABB PSTX soft starters. When the HMI displays “Shunt Fault,” many field technicians may misinterpret it as a problem related to multiple motors connected in parallel, communication parallel operation, or some kind of external parallel control. However, from the perspective of the soft starter main power circuit, this fault usually refers to an abnormal bypass path, an unintended short-circuit path, uncontrolled conduction, or a low-resistance path detected in the main circuit.

Therefore, F0613 should not be treated as a normal parameter alarm or a simple resettable warning. In many cases, it points to a real problem in the main power circuit, internal bypass contactor, thyristor module, external wiring, or load-side circuit. A correct diagnosis must follow the logic of power electronics and three-phase motor control rather than simply clearing the fault code.

ABB PSTX105 soft starter installed in an electrical control cabinet, displaying F0613 Shunt Fault and Chinese text for parallel fault on the HMI screen.

1. Basic Working Principle of a PSTX Soft Starter

To understand the F0613 shunt fault, it is necessary to understand the basic structure of a soft starter. In a three-phase soft starter, each phase is controlled by an anti-parallel thyristor pair. During motor starting, the soft starter controls the firing angle of the thyristors and gradually increases the RMS voltage applied to the motor. At the beginning of the start ramp, the motor receives a reduced voltage. As the ramp progresses, the voltage rises until it reaches nearly full line voltage.

This controlled voltage ramp reduces inrush current and avoids the mechanical shock associated with direct-on-line starting. It is especially useful for centrifugal pumps, large fans, belt conveyors, and other loads where sudden torque can damage couplings, belts, bearings, shafts, or hydraulic systems.

However, if the motor current continues to flow through the thyristors during long-term operation, the thyristors generate heat and reduce overall efficiency. For this reason, many medium- and high-end soft starters include a bypass contactor. During the start ramp, the thyristors control the voltage. Once the motor reaches full speed, the bypass contactor closes and carries the main current. The thyristors are then largely removed from the main current path, reducing heat generation and improving efficiency.

The ABB PSTX series has an internal bypass design. This is beneficial for energy efficiency and thermal management, but it also creates an important diagnostic point. If the internal bypass contactor becomes welded, stuck, mechanically jammed, or if a thyristor becomes short-circuited, the soft starter may detect an abnormal shunt path and report a fault such as F0613.

2. Meaning of F0613 Shunt Fault

The essential meaning of F0613 can be understood as follows: the soft starter has detected a main power circuit condition that does not match the expected control state. This may involve an abnormal bypass path, low-resistance path, or uncontrolled conduction between the input and output sides of the soft starter.

In a normal stopped condition, the soft starter should not provide an effective output voltage to the motor. During starting, the thyristors should conduct only according to the firing commands from the control board. During full-speed operation, the bypass contactor may close. During stopping or soft stopping, the bypass contactor and thyristors should return to their proper non-conducting or controlled states.

If the soft starter detects that one or more phases are already conducting when they should not be, or if it detects an abnormal low-resistance path between the input and output terminals, it may interpret this as a shunt fault.

This fault is different from common faults such as phase loss, overcurrent, overload, or overtemperature. A phase loss fault usually indicates missing or unbalanced supply or load phases. An overcurrent fault indicates excessive motor current. An overload fault indicates excessive thermal load. An overtemperature fault indicates insufficient cooling or excessive heat. A shunt fault is more structural in nature. It is usually related to the integrity of the main power circuit and the correct isolation or conduction of the thyristors and bypass path.

3. Common Causes of F0613 Shunt Fault

3.1 Internal Bypass Contactor Contact Welding

One of the most common causes is welded or stuck contacts in the internal bypass contactor. In a PSTX soft starter, the internal bypass contactor closes after the motor has completed its start ramp. Under normal conditions, the bypass contacts carry the motor current during continuous running.

If the motor is started frequently, operates under heavy load, experiences repeated overloads, or suffers from mechanical jamming, the bypass contacts may be exposed to repeated electrical and thermal stress. Over time, the contact surface may become burned, pitted, oxidized, or even welded together.

Once the bypass contactor contacts are welded, they may remain closed even after the control board commands them to open. This creates a direct low-resistance path between the input and output of the soft starter. The control system then detects that the actual main circuit state does not match the expected state and generates a shunt fault.

This condition is potentially dangerous because the soft starter may lose normal control over the motor. In some situations, the motor may remain electrically connected when it should be isolated by the starter. Therefore, repeated reset and restart attempts are not recommended before the main circuit has been checked.

3.2 Short-Circuited SCR Thyristor

Each phase of the soft starter contains an anti-parallel thyristor pair. These thyristors withstand high current, voltage transients, thermal stress, and switching stress during motor starting. If the motor load is too heavy, the start time is too long, the cooling condition is poor, the supply has severe surge voltage, or the output side experiences a short circuit, the thyristors may fail.

A failed thyristor may become short-circuited. Once this happens, the affected phase may show a low-resistance path between the line input terminal and the motor output terminal. A normal thyristor should block voltage when it is not triggered. A shorted thyristor loses this blocking ability and creates an uncontrolled conduction path.

From the soft starter’s monitoring logic, this is similar to an abnormal shunt path. The controller detects that a phase is conducting when it should not be and may report F0613.

A shorted thyristor and a welded bypass contactor can produce similar external measurement results. Both may cause low resistance between L and T terminals. Therefore, further internal inspection is needed to determine whether the fault is in the bypass contactor or in the thyristor power module.

3.3 External Bypass Contactor Wiring Error

Some control panels use an external bypass contactor in addition to or instead of the internal bypass function. In many retrofit projects, old contactors, star-delta starters, or previous bypass circuits may remain inside the cabinet. If the external bypass contactor is wired incorrectly, or if its control logic is wrong, it may connect the input and output sides of the soft starter at the wrong time.

The ABB PSTX already has an internal bypass mechanism. If an external bypass contactor is added, the wiring and control sequence must be carefully designed. The external bypass contactor should not close before the soft starter has completed its starting sequence, and it must open correctly during stop or fault conditions.

If the external bypass contactor closes too early, remains closed after stop, or has welded contacts, the soft starter will detect an unexpected shunt path. This can cause F0613 even if the soft starter itself is not internally damaged.

This type of problem is common in old control cabinets that have been modified. For example, a panel originally designed for direct-on-line starting, star-delta starting, or autotransformer starting may later be converted to soft starter control. If old contactors and wiring are not completely removed or correctly interlocked, an unintended parallel path may remain.

3.4 Mismatch Between Wiring Method and Parameter Setting

Soft starters may support different wiring configurations, such as standard in-line connection and inside-delta connection. These configurations produce different voltage, current, and phase relationships. If the actual main circuit wiring does not match the parameter setting inside the soft starter, the controller may misinterpret the measured signals.

For example, if the unit is physically wired in a standard in-line configuration but the parameter is set for inside-delta operation, the internal measurement logic may not match the actual current path. Conversely, if the motor is wired in inside-delta but the soft starter is configured as in-line, abnormal current and voltage relationships may be detected.

However, a parameter mismatch is usually not the highest-probability cause if the equipment has been operating normally for a long time and suddenly starts reporting F0613. In such cases, hardware faults in the main power circuit should be suspected first. Parameter and wiring mode verification is especially important after new installation, cabinet modification, soft starter replacement, or parameter reset.

3.5 Abnormal Short Path on the Motor or Load Side

The motor, motor cable, terminal box, and downstream contactors may also cause abnormal electrical conditions. Motor winding short circuits, cable insulation breakdown, water ingress in the terminal box, incorrect motor connection, or additional load-side contactors can all create abnormal low-resistance paths.

Although F0613 often points to the soft starter’s internal power circuit or bypass logic, external load-side faults must not be ignored. Industrial sites such as pump rooms, cooling towers, mining plants, chemical plants, and outdoor fan systems often have harsh environments. Moisture, conductive dust, oil mist, vibration, and cable aging can all contribute to insulation failure.

For this reason, the motor output cables should be disconnected during diagnosis to separate the soft starter from the external load. If the fault condition disappears after disconnecting the motor cables, the external circuit must be inspected before condemning the soft starter.

3.6 Current or Voltage Detection Circuit Fault

The PSTX soft starter uses internal current and voltage feedback to determine the operating state of the main circuit. If the current sensor, voltage sampling circuit, connector, ribbon cable, sampling resistor, comparator, or control board input circuit becomes faulty, the control unit may misjudge the actual state of the main circuit.

This cause is less common than a welded bypass contactor or shorted thyristor, but it does occur in real repair work. Surge voltage, poor control power quality, board contamination, moisture, corrosion, or cracked solder joints may affect the measurement circuit.

If the L-to-T resistance measurements are normal, the bypass contactor is not welded, the thyristors are not shorted, and the external wiring is correct, yet F0613 still appears repeatedly, the internal detection circuit or control board should be considered.

Technician diagnosing an ABB PSTX soft starter F0613 shunt fault with a multimeter, showing L-T terminal checks, bypass contactor, and SCR module inspection.

4. Initial Field Diagnostic Procedure

4.1 Record the Fault Condition Before Resetting

The first step is not to repeatedly press reset. F0613 is related to the main circuit, so repeated forced resets can expand the damage or create safety risks. The technician should first record when the fault appears:

Does it appear immediately after power-on?
Does it appear only after pressing start?
Does it occur during acceleration?
Does it occur after the motor reaches full speed?
Does it appear during stop or soft stop?
Does it appear intermittently after a period of operation?

The timing of the fault provides important diagnostic information. If the fault appears immediately after power-on before any start command, the most likely causes are a welded bypass contactor, shorted thyristor, or external shunt path. If it appears during acceleration, wiring method, motor load, thyristor firing, and parameter settings should also be checked. If it appears during running, the internal bypass contactor and its feedback logic should be inspected carefully.

4.2 Measure Resistance Between Input and Output Terminals

After completely isolating the three-phase supply and verifying absence of voltage, measure the resistance between the input and output of each phase:

1L1 to 2T1
3L2 to 4T2
5L3 to 6T3

Use a multimeter in resistance or continuity mode. In a normal stopped and de-energized state, the input and output of each phase should not show a direct low-resistance short.

If one or more phases show nearly 0 ohms or continuity, there is likely an abnormal conduction path. If all three phases are low resistance, the internal bypass contactor may be welded closed or an external bypass path may still be connected. If only one phase is low resistance, a shorted thyristor or one welded bypass contact is more likely.

This measurement is one of the most direct and useful checks for F0613.

4.3 Disconnect the Motor Cables and Measure Again

To determine whether the fault is inside the soft starter or outside in the load circuit, disconnect the motor output cables from the soft starter and repeat the L-to-T resistance measurements.

If the soft starter itself still shows low resistance after the motor cables are removed, the fault is almost certainly inside the soft starter or its connected bypass circuit.

If the low-resistance condition disappears after the motor cables are removed, the motor, cable, terminal box, downstream contactors, or load-side wiring must be checked. In this situation, replacing or repairing the soft starter alone may not solve the problem.

4.4 Inspect for External Bypass or Residual Contactors

In many industrial cabinets, the soft starter is not the original starting device. The cabinet may have been modified from a direct-on-line, star-delta, or autotransformer starter system. Old contactors and wiring may remain inside the panel.

The technician should trace the main power cables and inspect whether any external contactor is connected across the soft starter input and output. Check whether the contactor is mechanically stuck, whether its auxiliary contacts are wired correctly, and whether its control logic is properly interlocked with the soft starter.

A wrongly wired or stuck external bypass contactor can produce the same symptom as an internal soft starter fault.

4.5 Verify Wiring Mode and Parameters

The wiring mode parameter must match the actual main circuit. If the soft starter is installed in the standard in-line configuration, the corresponding parameter must be set accordingly. If inside-delta connection is used, the motor wiring, current setting, and configuration parameters must all match that method.

It is not enough to rely on drawings. The actual cabinet wiring must be inspected, because drawings are often outdated after field modifications.

5. Key Components to Inspect During Repair

5.1 Internal Bypass Contactor

After disassembling the soft starter, the bypass contactor should be inspected first. Look for burned contacts, pitting, melted contact surfaces, mechanical jamming, coil damage, loose connections, and overheated copper bars.

With the contactor in the open state, check whether the main contacts are still conductive. If the contacts remain closed or show very low resistance when they should be open, the contactor is welded or mechanically stuck.

For high-current soft starters, minor contact wear may not immediately cause failure. However, if the contact surface is badly burned or welded, the contactor must be repaired or replaced. Simply polishing the contacts may only provide a temporary solution and is not suitable for reliable long-term operation.

5.2 Thyristor Power Module

The thyristors should be checked with a multimeter and, when necessary, with more advanced test equipment. Compare the forward and reverse resistance of the three phases. If one phase shows a much lower resistance than the others, or if it is nearly shorted in both directions, the thyristor is likely damaged.

However, some thyristor faults are not obvious under cold multimeter testing. A thyristor may have increased leakage current, reduced blocking voltage, or thermal failure that appears only under voltage or load. In such cases, insulation and withstand-voltage testing may be required.

A repair technician should not rely only on a simple continuity test to conclude that all thyristors are good. The measurement results must be compared phase by phase and interpreted with the actual fault behavior.

5.3 Gate Trigger Circuit

If the thyristors are not shorted but the fault occurs during starting, the gate trigger circuit should also be inspected. A missing trigger pulse, incorrect phase synchronization, weak gate drive, or failed isolation component may cause abnormal conduction or serious phase imbalance.

The trigger circuit may include optocouplers, pulse transformers, gate resistors, isolated drive power supplies, and synchronization circuits. Failure in this section may cause the soft starter to start abnormally or report power circuit faults.

Although F0613 is more directly related to shunt or bypass detection, trigger problems can sometimes cause secondary fault symptoms during the start sequence.

5.4 Current Sensors and Sampling Circuit

The current feedback system is essential for the PSTX protection logic. A loose connector, damaged current transformer, cracked solder joint, burned sampling resistor, or failed signal conditioning component can lead to incorrect current feedback.

During repair, inspect current sensor wiring, connectors, ribbon cables, solder joints, and nearby components. Pay special attention to signs of overheating, corrosion, conductive dust, or mechanical stress.

5.5 Control Board and Detection Board

If the power circuit measurements are normal, the bypass contactor is not welded, the thyristors are not shorted, the motor circuit is healthy, and the wiring mode is correct, but the F0613 fault still appears repeatedly, the control board or detection board may be misjudging the main circuit state.

Possible causes include unstable control power, excessive ripple in the internal power supply, drifted reference voltage, failed comparator circuit, damaged input protection components, optocoupler aging, moisture leakage on the PCB, or contamination by conductive dust.

Board-level diagnosis requires careful inspection and signal measurement. In harsh industrial environments, board contamination and moisture-related leakage are common causes of intermittent and misleading faults.

6. Fault Symptom Patterns and Diagnostic Direction

If F0613 appears immediately after power-on before any start command, the most likely causes are internal bypass contact welding, shorted thyristor, or an external bypass path that is already closed.

If the fault appears immediately after pressing start, the technician should check not only the internal power circuit but also wiring mode, motor connection, parameter configuration, and thyristor triggering.

If the fault appears after the motor reaches full speed, the internal bypass contactor should be checked carefully. It may be failing to close correctly, chattering, overheating, or producing abnormal feedback.

If the fault appears during stopping, inspect whether the bypass contactor releases correctly and whether there is any residual external shunt path. Also check whether the load has regenerative or backfeed effects that may interfere with detection.

If the fault is intermittent and can be reset temporarily, suspect unstable mechanical contacts, loose terminals, thermal-related component failure, control board signal drift, or intermittent insulation problems.

7. Safety Precautions During Diagnosis

A soft starter is a high-power three-phase electrical device. Repair work must be performed only after complete isolation, lockout, voltage verification, and appropriate safety procedures.

The input terminals may remain live even when the motor is stopped. Some cabinets may also contain separate control voltage, external bypass supply, capacitor circuits, or backfeed sources. Before measuring resistance, all power sources must be isolated. A multimeter resistance range must never be used on a live circuit.

When removing motor cables, mark the phase sequence clearly to avoid incorrect reconnection. After repair, all main circuit terminals and copper bar connections must be tightened to the proper torque. Loose power terminals can cause overheating, arcing, and repeated failure.

After replacing parts or repairing the main circuit, the soft starter should not be immediately returned to full-load operation. It should first be tested under safe conditions, then with the motor connected, and finally under normal load. During testing, monitor the three-phase current balance, start ramp behavior, bypass contactor action, fault history, and thermal condition.

8. Practical Repair Logic for F0613

The key to diagnosing ABB PSTX F0613 is to focus on one question: is there an unintended conduction path in the main power circuit?

If the L-to-T resistance is low when the soft starter is stopped and de-energized, the diagnosis should focus on the bypass contacts, thyristor modules, and external bypass wiring.

If the L-to-T resistance is normal but the fault occurs during starting, check the wiring mode, parameter setting, motor condition, current feedback, voltage feedback, and trigger circuit.

If the main power circuit is normal and the external circuit is confirmed healthy, but the fault still appears, the detection circuit or control board may be causing a false shunt fault.

The diagnosis should proceed in the following order:

First, record the fault timing and operating condition.
Second, isolate power and measure the main circuit resistance.
Third, disconnect the motor cables and repeat the measurement.
Fourth, inspect external bypass contactors and residual old wiring.
Fifth, check the internal bypass contactor and thyristor modules.
Sixth, verify wiring mode and parameter configuration.
Seventh, inspect current and voltage detection circuits.
Finally, evaluate the control board if all power circuit checks are normal.

This sequence prevents unnecessary replacement of expensive components and reduces the risk of misdiagnosis.

9. Conclusion

ABB PSTX F0613 Shunt Fault is a main-circuit-related fault. It usually means that the soft starter has detected an abnormal shunt path, bypass path, or uncontrolled low-resistance conduction state. It should not be treated as a simple parameter warning or a normal resettable alarm.

The most common causes include internal bypass contactor welding, SCR thyristor short circuit, external bypass wiring error, mismatch between wiring method and parameter setting, motor or cable short path, and internal detection circuit malfunction.

In real repair work, the most effective starting point is to isolate power and measure the resistance between the line input and motor output terminals of each phase. If a low-resistance path exists between L and T, the fault direction becomes clear: bypass contactor, thyristor module, or external shunt wiring. If the main circuit resistance is normal, the technician should then investigate wiring configuration, current and voltage feedback, trigger logic, and control board detection.

For long-running equipment used in heavy-load, high-temperature, humid, dusty, or frequently started applications, F0613 has a strong hardware fault implication. A careful step-by-step diagnosis of the main circuit, bypass mechanism, thyristor modules, load circuit, and detection electronics is essential for accurate repair.

The core principle is simple: a soft starter should only conduct when its control logic commands it to conduct. If the unit detects conduction when it should be off, or detects a bypass path that does not match the expected state, it will report a shunt fault. Understanding this principle makes the diagnosis of ABB PSTX F0613 much clearer and prevents unnecessary guesswork during repair.

Posted on by

In-Depth Diagnosis of Schneider ATV340 bUF Braking Unit Short-Circuit Fault: A Repair Case Involving the DESAT Detection Circuit

1. Fault Overview

Schneider ATV340 series variable frequency drives are widely used in industrial applications such as conveyors, hoisting systems, centrifuges, winding machines, fans, pumps, and other equipment with medium or high inertia loads. In applications requiring fast deceleration, the drive often needs an external braking resistor. During deceleration, the motor may enter a regenerative state and feed energy back into the DC bus. If this energy is not dissipated in time, the DC bus voltage rises and may eventually cause overvoltage faults or damage to the power stage.

A Schneider ATV340D37N4E drive was received for repair after the customer reported a bUF braking unit short-circuit fault at the site. After the drive was brought back for bench testing, normal start and stop operation appeared to be fine. The motor could run, and no obvious abnormality was observed during ordinary operation. However, when the deceleration time was set very short, the drive triggered the braking unit short-circuit fault.

Further testing showed that this model has a parameter related to braking unit detection or braking function enablement. When this function was enabled, the drive reported a braking unit short-circuit fault during fast deceleration. When the braking function was disabled, the drive no longer generated the fault even under fast stop conditions.

At first glance, this appeared to be a typical external braking resistor fault, braking transistor fault, or braking IGBT short circuit. However, the actual repair process proved that the braking IGBT power device itself was not the root cause. The real fault was located in the DESAT detection circuit of the intelligent IGBT driver optocoupler used for the braking IGBT.

The final diagnosis confirmed that resistor R704, connected to the DESAT pin of the TLP5214A intelligent IGBT driver optocoupler, had become abnormally high resistance or open circuit. Its correct value should be 681, meaning 680Ω, but the faulty board measured more than 10MΩ, effectively open circuit. After replacing R704 with a 680Ω resistor, the braking function returned to normal, fast stop testing passed, and the bUF braking unit short-circuit fault disappeared.

This case is highly representative. It shows that when troubleshooting a braking unit short-circuit fault, technicians should not only check whether the braking IGBT is shorted with a multimeter, nor should they only focus on the external braking resistor value. In circuits using intelligent gate drivers and DESAT protection, an open, leaky, contaminated, or drifted detection circuit can also cause the drive to falsely report a braking unit short circuit.

Schneider ATV340 drive displaying bUF braking unit short-circuit fault, with external braking resistor connected between P/+ and PB terminals and highlighted braking driver section on the power board.

2. Basic Working Principle of the Braking Unit

In an inverter drive such as the ATV340, when the motor decelerates a high-inertia load, the motor may act as a generator. Mechanical energy is converted into electrical energy and fed back into the DC bus through the inverter bridge. This causes the DC bus voltage to rise. If the voltage rises beyond the protection threshold, the drive will trip on DC bus overvoltage. In severe cases, power components may be damaged.

The braking unit is used to dissipate this regenerative energy. When the DC bus voltage reaches the braking threshold, the drive turns on the braking IGBT. Current flows through the external braking resistor, converting the excess electrical energy into heat.

A typical braking current path is:

P/+ DC bus positive → braking resistor → PB terminal → braking IGBT → N/- DC bus negative

When the braking IGBT is off, the PB terminal is not pulled toward N/-, and almost no braking current flows through the resistor. When the braking IGBT turns on, PB is pulled down toward N/-, current flows through the braking resistor, and the regenerative energy is dissipated.

Therefore, whether the braking unit works correctly depends on several factors:

  1. The resistance and power rating of the external braking resistor;
  2. The reliability of the P/+, PB, and N/- power connections;
  3. Whether the braking IGBT turns on and off correctly;
  4. Whether the braking IGBT gate drive is normal;
  5. Whether the overcurrent, short-circuit, and DESAT detection circuits are normal;
  6. Whether the control board correctly receives the braking unit fault feedback signal.

In practical repairs, the first two items are relatively easy to check. The third can also be roughly checked with a multimeter. However, the fourth, fifth, and sixth items require deeper understanding of the gate driver optocoupler, DESAT detection, FAULT feedback, and dynamic waveforms.

Close-up of Schneider ATV340 braking IGBT driver circuit showing TLP5214A gate driver, DESAT pin 14, D705, C708, R703, and failed R704 680 ohm resistor in the DESAT detection path.

3. Common Causes of bUF Braking Unit Short-Circuit Fault

When a drive reports a braking unit short-circuit fault, the common causes can be grouped into the following categories.

3.1 External Braking Resistor Value Too Low

If the braking resistor value is lower than the minimum value allowed by the drive, the current through the braking IGBT becomes excessive as soon as the IGBT turns on. The driver or protection circuit may then immediately report a braking transistor short circuit or braking unit fault.

For a 400V-class drive, the DC bus voltage is commonly around 540V to more than 700V, especially during deceleration. If the braking resistor value is too low, the instantaneous braking current can become very high, placing excessive electrical and thermal stress on the braking IGBT.

3.2 Braking Resistor Wiring Short Circuit or Ground Leakage

Incorrect wiring between P/+ and PB, damaged braking resistor cables, carbonized terminals, loose connections, or ground leakage in the braking resistor box can all cause abnormal braking circuit behavior. In environments with moisture, dust, oil mist, or conductive contamination, insulation failure at the resistor terminals and cables is especially common.

3.3 Braking IGBT Collector-Emitter Short Circuit

This is the most direct cause. If the braking IGBT collector and emitter are shorted, PB is effectively pulled toward N/- continuously. Once the braking resistor is connected, an abnormal current path may exist from P/+ to N/- through the resistor. This type of fault can often be detected with a multimeter in diode or resistance mode.

3.4 Braking IGBT Gate Leakage or Abnormal Gate Drive

Some IGBTs do not fail as a direct C-E short. Instead, the gate insulation may degrade, the G-E path may leak, or the gate resistor, gate clamp, or turn-off circuit may become abnormal. In such cases, the IGBT may partially turn on when it should remain off, or it may fail to saturate properly when it should conduct.

These faults are not always easy to detect with a normal multimeter.

3.5 Driver Optocoupler Failure or Driver Power Supply Abnormality

Medium and high-power drives usually use isolated driver optocouplers or intelligent gate driver chips to drive IGBTs. The braking IGBT is no exception. If the driver output voltage is too low, negative turn-off voltage is abnormal, or the driver supply decoupling capacitor has failed, the IGBT may not turn on fully or may turn off incorrectly.

3.6 DESAT Detection Circuit Abnormality

This is the core issue in this case.

Many intelligent IGBT driver optocouplers include DESAT protection. DESAT detection is used to determine whether an IGBT has entered normal saturation when it is commanded to turn on. If the IGBT receives a gate drive command but the C-E voltage remains too high, it may indicate short circuit, overcurrent, insufficient gate drive, module failure, or load abnormality. The driver chip then quickly shuts down the IGBT and outputs a fault signal.

However, if the DESAT detection circuit itself is open, leaky, contaminated, drifted, or has a cracked solder joint, the driver chip may falsely detect desaturation even when the IGBT is actually normal. The result is a false braking unit short-circuit fault.

Technical infographic explaining ATV340 deceleration braking fault repair, showing motor regeneration energy flow through the braking resistor and replacement of open R704 680 ohm resistor to restore normal DESAT detection.

4. Diagnostic Process in This Case

4.1 Fault Condition Confirmation

The ATV340D37N4E could run normally under ordinary conditions. Normal start and stop operation did not produce any abnormal alarm. The bUF braking unit short-circuit fault appeared only when the deceleration time was set very short and the braking unit participated in energy dissipation.

This indicated that the main inverter bridge, current detection circuit, control logic, and auxiliary power supply were unlikely to be the primary fault areas. If the main inverter bridge or the main DC power stage had a serious defect, the drive would likely report overcurrent, short circuit, undervoltage, phase loss, or drive faults even during ordinary operation.

Since the fault was strongly associated with fast deceleration and braking unit activation, the diagnostic focus was shifted to the braking circuit.

4.2 External Braking Resistor Check

During bench testing, a temporary resistance wire of approximately 10Ω was connected between P/+ and PB to simulate the braking resistor, and the same fault could be reproduced. Although a temporary resistance wire is not equivalent to a standard braking resistor in every aspect, the customer’s site also reported the same fault with a standard braking resistor. Therefore, the external resistor itself was not considered the main suspect.

In real repair practice, the following items should still be checked:

  • Actual braking resistor resistance;
  • Whether the resistance is below the minimum value allowed by the drive;
  • Braking resistor power rating;
  • Wiring reliability between P/+ and PB;
  • Insulation resistance of the braking resistor to ground;
  • Cable damage, loose terminals, overheating, carbonization, or arcing marks.

In this case, because the same fault occurred with the customer’s standard braking resistor and the temporary resistor was only used to reproduce the fault, the investigation continued inside the drive, focusing on the braking IGBT driver circuit.

4.3 Braking IGBT Inspection

The braking IGBT module was checked using conventional methods. No obvious C-E short circuit, G-E short circuit, or severe leakage was found. Based on standard repair experience, the braking IGBT appeared normal in static testing.

However, this point must be emphasized:

A normal static IGBT test does not prove that the IGBT and its protection circuit will behave normally under dynamic braking conditions.

DESAT faults usually occur at the instant when the IGBT is driven on. Only when the braking IGBT is under high DC bus voltage, carrying braking current, and controlled by the gate driver will dynamic problems such as desaturation, insufficient drive, or detection circuit failure appear.

Therefore, relying only on a diode-mode multimeter test of the IGBT can easily lead to an incorrect conclusion.

4.4 Misleading Comparison with an ATV610 Control Board

During troubleshooting, an ATV610 control board was used for comparison. Since some ATV610 and ATV340 power boards, driver boards, and modules may look similar or share similar hardware structures, it was tempting to conclude that if the ATV610 board did not trigger the fault, the power board must be normal.

Further analysis showed that this comparison was not decisive. The ATV610 control board did not have the same braking detection or braking function logic as the ATV340. It may not have actually triggered the braking IGBT in the same way, or it may not have monitored the braking unit fault feedback in the same manner.

Therefore, the fact that the ATV610 board did not report the same fault could not be used as proof that the braking driver circuit was healthy.

This is an important lesson. Even if two drive series share similar hardware platforms, their firmware logic, alarm judgment, drive enable conditions, and fault feedback processing may be different. A drive not reporting a fault does not necessarily mean that the tested power board is fully normal.

4.5 Locking the Fault Area to the TLP5214A and DESAT Circuit

The braking driver section of the board used a TLP5214A intelligent IGBT driver optocoupler. This device is not an ordinary optocoupler. It integrates IGBT gate drive, undervoltage protection, soft shutdown, fault feedback, and DESAT detection.

Pin 14 of the TLP5214A is the DESAT detection pin. When the IGBT is turned on, the DESAT pin monitors the IGBT C-E voltage through an external diode, resistor, and capacitor network. If the detected voltage exceeds the internal threshold, the driver interprets this as IGBT desaturation, shuts down the output, and sends a fault signal through the FAULT pin.

Around the TLP5214A DESAT pin, components such as R703, R704, D705, and C708 were identified. R704 was found to be related to the DESAT path. Its measured resistance was more than 10MΩ, clearly abnormal.

To confirm the expected value, a 55kW drive driver board was used for comparison. The corresponding positions on the comparison board showed the following resistor markings:

  • R704: 681, meaning 680Ω;
  • R703: 472, meaning 4.7kΩ.

On the faulty board, R703 measured around 5kΩ, consistent with 4.7kΩ. However, R704 measured more than 10MΩ, completely inconsistent with the expected 680Ω. This strongly indicated that R704 was open circuit or had failed to a very high resistance.

5. Why an Open R704 Causes a Braking Unit Short-Circuit Fault

When technicians see the alarm description “braking unit short circuit,” the first reaction is often to suspect a shorted braking IGBT or shorted braking resistor. However, in a circuit using DESAT detection, the alarm name does not always mean that there is a physical short circuit. It may be the result of the intelligent driver detecting an abnormal protection condition.

Under normal operation, when the braking IGBT is turned on:

  1. The control board sends a braking IGBT drive command;
  2. The TLP5214A outputs the gate drive voltage;
  3. The braking IGBT turns on normally;
  4. PB is pulled toward N/-;
  5. The IGBT C-E voltage drops to a low value;
  6. The DESAT detection circuit confirms that the IGBT has entered saturation;
  7. The driver optocoupler does not output a fault signal, and braking proceeds normally.

When R704 is open:

  1. The control board sends a braking IGBT drive command;
  2. The TLP5214A outputs the gate drive;
  3. The braking IGBT may actually turn on normally;
  4. But the DESAT detection path loses its normal sampling or clamping function because R704 is open;
  5. The internal DESAT charging current inside the TLP5214A causes the DESAT pin voltage to rise abnormally;
  6. The driver falsely determines that the IGBT has not entered saturation;
  7. The TLP5214A shuts down the output and sends a FAULT signal;
  8. The control board receives the braking unit fault feedback and displays bUF / braking unit short circuit.

Therefore, the real problem in this case was not an actual shorted braking IGBT. It was a false braking unit short-circuit fault caused by an open DESAT detection resistor.

Typical characteristics of this type of fault include:

  • Normal ordinary running;
  • Normal ordinary stopping;
  • Fault appears only when the braking unit is activated;
  • Disabling the braking function makes the fault disappear;
  • Braking IGBT passes static testing;
  • External braking resistor is normal;
  • DESAT circuit components show open circuit, drift, leakage, cracked solder joints, or contamination.

6. Repair Procedure

The final repair procedure in this case was as follows:

  1. Lift or remove one side of R704 and confirm its abnormal resistance;
  2. Compare with a similar driver board and confirm the correct value of R704 as 681, meaning 680Ω;
  3. Replace R704 with a 680Ω SMD resistor;
  4. Clean the area around TLP5214A, R703, R704, D705, and C708;
  5. Reflow or resolder relevant TLP5214A pins, especially DESAT, VOUT, VCC2, VE, and VEE pins;
  6. Check connector S23 and its solder joints to ensure reliable connection to the braking IGBT circuit;
  7. Reassemble and test the drive;
  8. Enable the braking function, perform fast deceleration testing, and verify that the bUF fault no longer appears.

After R704 was replaced, the drive passed fast stop testing. The braking unit worked normally, and the fault was eliminated.

7. Key Measurement Points for Similar Faults

For similar braking unit short-circuit faults, the following diagnostic sequence is recommended.

7.1 External Braking Resistor

Measure the resistance between P/+ and PB. Confirm that the resistor value is not below the minimum allowed value for the drive. Also inspect the resistor box, cable, terminals, and insulation to ground.

7.2 Static Test of the Braking IGBT

After power is removed and the DC bus capacitors are fully discharged, check the braking IGBT C-E, G-E, and G-C paths for short circuit or leakage. If an obvious short circuit is present, the power device must be handled first.

7.3 Gate Drive Voltage

Under safe test conditions, observe the braking IGBT G-E voltage at the instant of braking. During conduction, a gate drive voltage of around +15V is typically expected. During turn-off, the voltage may be 0V or negative depending on the driver design.

7.4 TLP5214A FAULT Pin

Observe whether the FAULT pin of the TLP5214A changes state when the fault occurs. If the FAULT pin is pulled low, the driver itself has detected an abnormal condition. If the FAULT pin does not change but the control board still reports a braking unit fault, then the control board’s fault feedback input circuit should be checked.

7.5 DESAT Pin and Peripheral Circuit

Focus on the DESAT-related components connected to pin 14 of the TLP5214A, including the series resistor, sampling diode, blanking capacitor, clamping components, and solder joints. In this case, R704 was the key component.

7.6 Connectors and Solder Joints

The braking IGBT driver signal is often transmitted through a small connector. Loose connectors, cracked solder joints, oxidation, poor contact, or damaged harnesses may cause abnormal gate drive or abnormal detection signals.

8. Why a Small Resistor Can Cause a Major Fault

R704 is only a small SMD resistor with a value of 680Ω. However, because it is located in the DESAT detection path of the braking IGBT driver, it has a critical protection role.

A drive protection system does not only determine whether a large power device is physically shorted. It depends on many small signal detection circuits to judge whether the power stage is operating safely.

In the high-voltage, high-current, and high-dv/dt environment of a variable frequency drive, the intelligent driver optocoupler must quickly determine whether the IGBT is healthy during turn-on. If the DESAT circuit becomes abnormal, the driver will prioritize protection and shut down the IGBT, even if the result is a false alarm.

When a 680Ω resistor becomes open circuit, the braking IGBT may still be good, and the external braking resistor may also be normal. However, because the driver cannot receive correct DESAT information, the system reports a braking unit short circuit.

If the technician only follows the literal meaning of the alarm and repeatedly replaces the IGBT module or suspects the external resistor, the repair will go in the wrong direction.

9. Diagnostic Logic for Similar Braking Faults

When handling braking-related faults on Schneider ATV340, ATV630, ATV930, ATV610, or similar drives, the following logic is useful.

9.1 Is the Fault Strongly Related to Braking Action?

If the fault appears only during fast stop, regenerative operation, DC bus voltage rise, or braking resistor operation, the braking unit should be the first diagnostic target.

9.2 Does the Fault Disappear When Braking Is Disabled?

If disabling the braking function makes the fault disappear, the problem is related to braking IGBT drive or detection. However, this does not mean the drive can be safely returned to the customer with the braking function disabled. The customer’s load may require braking resistor operation to prevent DC bus overvoltage.

9.3 Is a Normal Static IGBT Test Sufficient?

No. A normal static test only rules out obvious breakdown. It does not rule out dynamic desaturation, insufficient drive, false detection, or DESAT circuit open faults.

9.4 Is There a Similar Board for Comparison?

If a similar power driver board is available, compare DESAT circuit resistor values, diode direction, capacitor placement, and component markings. In this case, comparison with a 55kW driver board helped confirm that R704 should be 681 rather than a high-resistance value.

9.5 Is There Contamination, Moisture, or Solder Cracking?

DESAT detection is a high-speed protection signal circuit. Board contamination, flux residue, moisture, carbonization, and cracked solder joints can all cause false triggering. Cleaning, drying, and resoldering are often necessary.

10. Suggested Technical Repair Report

The repair conclusion for this case can be written as follows:

The Schneider ATV340D37N4E drive reported a bUF braking unit short-circuit fault during fast deceleration. Inspection confirmed that the external braking resistor wiring method was correct, and the braking IGBT module showed no obvious C-E short circuit or G-E short circuit. Further inspection of the braking IGBT driver circuit found an abnormality in the DESAT detection circuit of the TLP5214A intelligent IGBT driver optocoupler. The resistor R704 connected to the DESAT circuit of pin 14 had drifted to an abnormally high resistance, measuring more than 10MΩ. On a similar driver board, the corresponding component value was 681, meaning 680Ω. The open R704 caused the DESAT detection signal to become abnormal when the braking IGBT was triggered. As a result, the driver optocoupler falsely detected IGBT desaturation or short circuit and sent a FAULT signal to the control board, triggering the bUF braking unit short-circuit alarm. After replacing R704 with a 680Ω resistor and cleaning/resoldering the related driver detection circuit, the braking function and fast deceleration operation returned to normal.

11. Conclusion

The bUF braking unit short-circuit fault on a Schneider ATV340 drive does not always mean that the braking IGBT is physically shorted. In circuits using intelligent IGBT driver optocouplers such as the TLP5214A, an abnormal DESAT detection circuit can also trigger the same alarm.

The key features of this case were: the fault appeared only when braking was enabled and fast deceleration was performed; the braking IGBT passed static testing; the external braking resistor condition could not explain the fault; and comparison with a similar driver board showed that R704 should be 680Ω, while the faulty board measured more than 10MΩ. After replacing R704, the drive returned to normal.

This case reminds repair technicians that VFD power-stage fault diagnosis should not focus only on large power components. In many cases, the component that actually triggers the alarm is a small part of the drive, protection, feedback, or detection circuit. DESAT resistors, sampling diodes, blanking capacitors, driver optocouplers, FAULT feedback circuits, and connector solder joints can all determine whether the braking unit operates correctly.

A correct repair approach should begin by confirming the fault trigger condition, then distinguishing between a real power-stage fault and a false detection fault, and finally verifying the driver optocoupler and protection circuit point by point. Only by understanding the braking unit operating principle and DESAT protection mechanism can technicians avoid unnecessary module replacement and improve repair accuracy.

Posted on by

Agilent VacIon Plus 20 Ion Pump User Guide — Technical Interpretation & Operational Essentials Based on the Official Manual

I. Product Positioning & Technical Background

The Agilent VacIon Plus 20 is a mid-range ion pump with a nitrogen pumping speed of 20 L/s, making it a core model in the VacIon Plus series (which covers a full range from 0.4 to 1,000 L/s). It is designed for ultra-high vacuum (UHV) and extreme high vacuum (XHV) systems, with key applications including: academic research, high-energy physics (HEP) experiments, particle accelerators and synchrotron rings, scanning electron microscopy (SEM), imaging equipment, radiation therapy devices, and surface analysis instruments.

The official model numbers include 919-1114 (Diode, no magnet), 919-1115 (Diode, with ferrite magnet), 919-1144 (StarCell, no magnet), 919-1145 (StarCell, with ferrite magnet), and other variants. Users can select different pumping unit types, magnet configurations, and high-voltage feedthrough orientations based on actual pumping requirements.

StarCell 9191146 Vacion Plus 20

II. Core Technical Principles

2.1 Diode Ion Pump Operating Mechanism

The diode ion pump core consists of a positively charged anode ring and a negatively charged titanium (Ti) cathode, both immersed in a magnetic field. When powered on, electrons collide with gas molecules in the plasma, generating positive ions. Light ions (e.g., H₂, He) accelerate toward the cathode, penetrate the Ti layer, and are buried and absorbed. Heavy ions strike the cathode, causing Ti sputtering. The sputtered Ti coats the inner surface of the anode ring, forming a fresh chemically active film that continuously traps active gases (H₂, N₂, O₂, etc.).

Key Limitation: The two-electrode structure cannot effectively trap noble gases (e.g., Ar), because noble gases do not chemically react with Ti. They must rely on physical sputtering for transport to the anode for pumping, which is far less efficient.

2.2 Triode / StarCell Ion Pump Operating Mechanism

Agilent’s proprietary StarCell pumping unit is a fundamental improvement over the traditional two-electrode design. It uses a star-shaped cathode geometry, which greatly increases the probability of noble gases being transported to the anode as energetic neutral particles, while the titanium cathode ensures high pumping speed for H₂. According to official technical documentation, StarCell is the only ion pump capable of handling large quantities of noble gases and hydrogen simultaneously, offering the highest pumping speed and capacity for methane (CH₄), argon (Ar), and helium (He).

Bottom line: if your system contains a significant proportion of noble gases or hydrogen, StarCell is the clearly superior choice over Diode.

2.3 Ion Pump as Vacuum Gauge

Because the ion current generated by an ion pump is proportional to pressure, in many applications (especially SEM), the VacIon Plus 20 can directly double as an ionization vacuum gauge — a capability that mechanical pumps like turbomolecular pumps do not have. However, this requires extremely low leakage current. Agilent achieves this through a patented anode design (reducing void volume, sharp edges, and metal “whiskers”), and the SEM version is further optimized for this parameter.

III. Key Performance Parameters

ParameterValue
Nitrogen pumping speed20 L/s (series range: 20–75 L/s)
Inlet flange2¾” ConFlat (NW 35 / CFF)
Maximum starting pressure≤ 5 × 10⁻² mbar
Ultimate pressure< 1 × 10⁻¹¹ mbar
Maximum bake-out temperature350 °C (vacuum processing up to >400 °C)
Heater voltage100–120 V / 200–240 V, 140 W
Service life80,000 hours
Pump weight (no magnet)7 kg (15 lbs)
High-voltage feedthrough optionsFischer, King, DESY, Varian, SHV 10kV (Safeconn)
Magnet optionsFerrite magnet, rare-earth magnet (NdFeB)

Parameter Interpretation: The maximum starting pressure of ≤ 5 × 10⁻² mbar means this pump cannot be started from atmospheric pressure. A backing pump (e.g., scroll pump, diaphragm pump) must first reduce the system to the 10⁻² mbar range before the ion pump can be turned on. The ultimate pressure of < 10⁻¹¹ mbar is already in the XHV regime, sufficient for most surface analysis and particle physics experiments.

StarCell 9191146 Vacion Plus 20

IV. Pre-Installation Preparation & System Integration

4.1 Cleanliness & Vacuum Integrity

The manual explicitly requires: the pump be vacuum-processed at >400 °C and clamped off under vacuum to ensure cleanliness and vacuum integrity before installation. This step is not a formality — any surface contamination becomes an outgassing source in UHV environments and will severely degrade ultimate pressure.

Practical Tips:

  • Wipe the ConFlat flange face with anhydrous ethanol before installation; ensure no scratches or particles.
  • The copper gasket (OFHC copper) must be new or annealed if reused; never use gaskets with deformation marks.
  • Tighten bolts in a diagonal cross pattern in three passes, with torque per manual recommendations (typically ~20–25 N·m for a 2.75″ ConFlat flange).

4.2 Magnet Configuration Selection

ConfigurationApplicable Scenarios
No magnet (919-1114 / 919-1144)System already has external magnets, or scenarios extremely sensitive to magnetic interference
Ferrite magnet (919-1115 / 919-1145)General-purpose use, lower cost, moderate field strength
Rare-earth NdFeB magnet (919-1146 series)Scenarios requiring stronger magnetic field for higher pumping speed; be aware of interference with nearby electronics

4.3 High-Voltage Feedthrough Orientation

The ConFlat flange is rotatable, and the high-voltage feedthrough can be oriented in different directions (Fischer, SHV, etc.). Consider cable routing space, interference with other components, and whether optical baffles or other accessories are needed. The manual supports custom pump geometries and additional ports — specify these when ordering.

V. Startup Procedure

Step 1: Backing Pump Evacuation

Use a scroll pump or diaphragm pump to reduce system pressure to ≤ 5 × 10⁻² mbar. This is a hard requirement for ion pump startup — do not skip it.

Step 2: Power On the Ion Pump

Connect the heater power supply (100–120 V or 200–240 V, match to model), wait for the heater to stabilize (~5–10 minutes), then turn on the high-voltage supply.

Note: The manual does not specify an exact HV turn-on sequence, but per Agilent’s general ion pump operating guidelines, the filament (if equipped) should be turned on first, and the ion current should be allowed to stabilize before ramping to operating voltage. The Diode/StarCell versions of the VacIon Plus 20 typically do not require a filament — apply operating voltage directly.

Step 3: Monitor Ion Current & Pressure

After startup, the ion current should start high (corresponding to poorer vacuum) and gradually decrease, eventually stabilizing in the nA range. The vacuum gauge reading should continue to drop, ultimately reaching < 10⁻¹⁰ mbar.

Abnormal Condition Diagnosis:

  • Ion current stays persistently high → possible leak or backing pump did not evacuate sufficiently.
  • Ion current fluctuates abnormally → check HV power supply stability or whether the magnet is affected by external interference.
  • Pressure cannot drop below 10⁻⁹ mbar → consider bake-out (350 °C, hours to days).

Step 4: Bake-Out (If Required)

If the system requires ultimate pressure better than 10⁻¹⁰ mbar, bake-out is necessary. The VacIon Plus 20 can withstand up to 350 °C (vacuum processing above 400 °C), but note:

  • The ion pump should remain on during bake-out to continuously pump desorbed gases.
  • Recommended ramp rate: ≤ 5 °C/min to avoid thermal shock causing flange leaks.
  • After bake-out, cool naturally to room temperature before shutting down the ion pump.

VI. Daily Operations & Troubleshooting

6.1 Leakage Current Monitoring

Low leakage current is the foundation of stable VacIon Plus 20 operation. The manual notes that the SEM version features low leakage current and high stability, minimizing electronic interference. In daily operations, periodically record the ion pump’s ion current in the closed-valve state (i.e., leakage current). If leakage current rises significantly (exceeding several hundred nA), the anode or insulator may be contaminated — schedule maintenance.

6.2 Magnet Demagnetization Risk

Ferrite magnets may demagnetize at high temperatures. If the system requires frequent bake-outs above 300 °C, consider high-temperature-rated rare-earth magnets (NdFeB), but evaluate their magnetic interference with surrounding equipment. The manual offers shielded magnet options for use in strong-field environments.

6.3 Pumping Unit Replacement

Per Agilent’s official maintenance guide:

  • Ion pumps ≥ 150 L/s can have their pumping units replaced individually (StarCell / Diode / Noble Diode).
  • The VacIon Plus 20 is a smaller model; typically the entire pump body is replaced while retaining the magnet, to reduce maintenance cost.

Pumping Unit Comparison:

Unit TypeStrong GasesWeak GasesTypical Application
DiodeN₂, H₂, O₂Ar, HeGeneral UHV, electron microscopy
Noble DiodeMixed gases, H₂Pure noble gasesParticle accelerators, synchrotron rings
StarCellAr, He, CH₄, H₂—Systems with high noble gas loads

6.4 Common Faults & Solutions

SymptomPossible CauseSolution
Cannot start (HV will not establish)Backing pressure > 5×10⁻² mbarRe-evacuate with backing pump
Abnormally high ion currentSystem leak / outgassing sourceHelium leak check, inspect flange seals
Ultimate pressure not reaching specInsufficient bake-out / aging pumping unitExtend bake-out time, consider pump replacement
Unstable pressure readingExcessive leakage current / external EMICheck grounding, evaluate need for shielded magnet

VII. Shutdown & Long-Term Storage

7.1 Normal Shutdown

  1. Turn off the ion pump high-voltage supply.
  2. Keep the heater on for an additional 10–15 minutes (to purge residual gas from the pump body).
  3. Turn off the heater power supply.
  4. Turn off the backing pump.
  5. Backfill with dry nitrogen or argon to atmospheric pressure (to prevent moisture condensation inside the pump).

7.2 Long-Term Storage (Exceeding 1 Week)

The manual recommends filling the pump with methanol or inert gas before long-term storage to prevent seal drying. Procedure: flush the system with methanol for 30 minutes, reduce flow to zero before shutdown, and keep the pump filled with methanol. To restart, first use the backing pump to remove methanol vapor, then follow the normal ion pump startup procedure.

VIII. Model Selection Decision Tree

Faced with model numbers like 919-1114, 919-1115, 919-1144, 919-1145, follow this logic:

  1. Does the system contain large amounts of noble gases (Ar, He)?
    • Yes → Choose StarCell (919-1144 / 919-1145)
    • No → Next step
  2. Is mixed-gas pumping speed also needed?
    • Yes → Choose Noble Diode (if available)
    • No → Choose Diode (919-1114 / 919-1115)
  3. Does the system already have external magnets?
    • Yes → Choose no-magnet version (save cost)
    • No → Choose ferrite or rare-earth magnet based on magnetic interference tolerance
  4. Is there a special requirement for HV feedthrough orientation?
    • Yes → Choose the corresponding custom model (e.g., 919-1145M021 = SHV feedthrough, 919-1145M022 = 90° feedthrough)
    • No → Standard Fischer feedthrough is sufficient

IX. Summary

The VacIon Plus 20 is not a “plug-and-play” device. It is a complete vacuum solution that must work in concert with a backing pump, vacuum gauge, and bake-out system. Every parameter in the manual — from the 5 × 10⁻² mbar maximum starting pressure to the 350 °C bake-out limit, from StarCell’s noble gas handling capability to the SEM version’s low-leakage-current design — represents an engineering constraint forged over decades of UHV applications.

Understanding these constraints matters more than memorizing the steps. Because when something goes wrong, what actually helps you diagnose the problem is not what page of the manual says what — it’s whether you truly understand the ion pump’s physical behavior at every pressure range and in every gas environment.

Posted on by

Troubleshooting Schneider ATV660 SCF1 “Motor Short Circuit” and Power Brick 1 Diagnostic Failure

1. Equipment Background and Fault Overview

The Schneider Electric Altivar Process ATV660 is a cabinet-type variable frequency drive designed for medium- and high-power industrial motor control applications. It is commonly used on large fans, pumps, compressors, cooling systems, process machinery, and other heavy-duty industrial equipment. The case discussed in this article involves a Schneider Electric ATV660C31Q4X10-EP drive system.

According to the drive nameplate, the unit is rated:

  • Model: ATV660C31Q4X10-EP
  • Input voltage: 3-phase 380–415 V
  • Power rating: ND 315 kW / HD 250 kW
  • Output current: ND 590 A / HD 477 A
  • Protection class: IP23

The connected motor is a WEG three-phase induction motor. The motor nameplate indicates approximately:

  • Voltage: 400 V
  • Frequency: 50 Hz
  • Power: 260 kW
  • Rated current: 435 A
  • Speed: 2984 rpm
  • Power factor: 0.90

From a capacity-matching point of view, the motor current of 435 A is lower than the drive’s HD current rating of 477 A. Therefore, under normal conditions, this motor and drive combination should be basically suitable.

However, after startup, the drive reports:

SCF1 – Motor short circuit

The diagnostic screen also shows:

IGBT Diag w motor
Power Brick 1 Diag: Not OK

These two pieces of information are very important. This is not a simple communication setting issue, and it should not be treated as an ordinary keypad configuration problem. The fault must be analyzed from the perspectives of motor insulation, motor cable condition, output wiring, IGBT power module condition, gate driver circuit, current detection, and cabinet environment.

For a 250–315 kW class drive, this type of fault has a high repair risk. Repeatedly resetting and restarting the drive without diagnosis may damage the IGBT module, gate driver board, DC bus components, fuses, or even the entire power section. Therefore, the correct approach is to stop repeated startup attempts and perform a structured electrical diagnosis.

Technician inspecting a Schneider Electric ATV660 drive cabinet showing SCF1 Motor Short Circuit alarm, with callouts for checking U/V/W output terminals, motor cable, and power stage.

2. Meaning of SCF1 “Motor Short Circuit”

The keypad fault SCF1 Motor short circuit literally means that the drive has detected a short-circuit condition related to the motor output. However, in actual VFD diagnosis, this fault does not only mean that the motor winding is physically shorted.

SCF1 may be triggered by several conditions:

  1. Phase-to-phase short circuit inside the motor winding.
  2. Motor winding insulation breakdown to ground.
  3. Phase-to-phase short circuit in the motor cable.
  4. Motor cable insulation leakage to ground.
  5. Incorrect wiring at the output terminals U/V/W.
  6. Capacitor bank or power factor correction device connected at the drive output.
  7. Old star-delta starter circuit not fully removed.
  8. Output contactor contact failure or incorrect switching sequence.
  9. IGBT power module internal short circuit.
  10. Gate driver board failure causing abnormal IGBT switching.
  11. Current sensor or current detection circuit abnormality.
  12. Severe parameter mismatch between motor and drive.
  13. Mechanical load locked or jammed, causing abnormal starting current.

Therefore, SCF1 should not be interpreted too narrowly. It does not automatically prove that the motor is bad, and it does not automatically prove that the VFD is bad. The correct diagnostic strategy is to determine whether the problem is located outside the drive, such as motor, cable, or output wiring, or inside the drive power stage, such as IGBT, gate driver, current sensor, or power brick.

Schneider ATV660 SCF1 troubleshooting workflow showing motor cable disconnection, insulation resistance testing with a megger, and Power Brick 1 IGBT module diagnostic check.

3. Importance of “Power Brick 1 Diag: Not OK”

The second fault indication is more serious:

IGBT Diag w motor
Power Brick 1 Diag: Not OK

In a high-power cabinet drive such as the ATV660, the power section is typically composed of IGBT modules, gate driver boards, current sensors, DC bus capacitors, copper busbars, cooling fans, control boards, and internal diagnostic circuits. “Power Brick” can be understood as a power module group or power unit inside the drive.

When the diagnostic page shows Power Brick 1 Diag: Not OK, it means the drive has detected an abnormal condition related to the first power brick or power module group.

This diagnostic result may be caused by two different situations:

First, the motor or motor cable is connected and has a short circuit or leakage problem. The external fault causes the drive to report an abnormal power brick diagnostic result.

Second, the power brick itself is defective. In this case, the diagnostic result may remain Not OK even after the motor cables are disconnected.

This distinction is very important. The next diagnostic action should not be to keep changing communication parameters or command source settings. The first critical test is to disconnect the motor cables from the drive output and check whether the Power Brick 1 diagnostic result changes.

4. Electrical Mechanism Behind the Fault

A VFD output is not a normal sine-wave power supply. It is generated by high-speed switching of IGBTs using PWM modulation. The drive control board monitors output current, DC bus voltage, IGBT feedback, current balance, and protection signals from the gate driver circuit.

When there is a short circuit or severe insulation failure at the output side, the following conditions may occur during startup:

  1. One output phase current rises abnormally.
  2. Three-phase output current becomes seriously unbalanced.
  3. IGBT desaturation protection is triggered.
  4. DC bus current rises sharply.
  5. The gate driver board detects an unsafe switching condition.
  6. The control board judges the output circuit as shorted.
  7. The drive stops output and reports SCF1.

If the IGBT module itself is already damaged, a similar fault can occur even when the motor is normal. Examples include shorted IGBT chips, damaged gate resistors, abnormal gate drive signals, faulty desaturation detection, damaged driver optocouplers, or defective current feedback circuits.

This is why repeated startup is dangerous. Every restart applies another high-current stress to the IGBT power stage. If there is a real short circuit, the damage may become much worse.

5. First Rule: Do Not Repeatedly Reset and Restart

When an ATV660 displays SCF1 together with Power Brick 1 diagnostic failure, the first rule is:

Do not repeatedly press RUN or reset the fault again and again.

For some minor alarms, such as temporary undervoltage or external interlock faults, reset and restart may sometimes be acceptable. But SCF1 is a short-circuit-related fault. Repeated startup may cause serious damage.

Possible consequences include:

  1. IGBT module explosion.
  2. Gate driver board failure.
  3. DC bus fuse failure.
  4. Copper busbar arcing.
  5. Rectifier section stress.
  6. Additional internal faults.
  7. Higher repair cost.

The correct procedure is:

Stop the drive, isolate the power supply, wait for DC bus discharge, disconnect the motor output, perform insulation tests, and diagnose section by section.

6. Key Diagnostic Step: Disconnect the Motor and Test the Drive Alone

The most important step is to separate the drive from the motor and cable.

Recommended procedure:

  1. Switch off the main power supply.
  2. Wait until the DC bus is fully discharged.
  3. Confirm with a multimeter that the voltage between DC+ and DC- is at a safe level.
  4. Disconnect the motor cables from the drive output terminals U/T1, V/T2, and W/T3.
  5. Leave the drive output terminals open, with no motor connected.
  6. Power on the drive.
  7. Enter the diagnostic menu.
  8. Check whether Power Brick 1 Diag changes from Not OK to OK.

The interpretation is as follows:

If Power Brick 1 Diag becomes OK after disconnecting the motor, the drive power section is probably not internally shorted. The fault is more likely related to the motor, motor cable, output wiring, output contactor, capacitor, or load.

If Power Brick 1 Diag remains Not OK after disconnecting the motor, the fault is very likely inside the drive. The main suspects are IGBT Power Brick 1, gate driver board, current sensor, busbar insulation, power module connection, or detection circuit.

If the drive still reports SCF1 with the motor disconnected, this strongly suggests an internal power stage fault or a short/leakage near the output terminals.

If the drive is normal without the motor but immediately faults when the motor is connected, then the focus should shift to the motor, cable, terminal box, and output circuit.

This single test is critical because it divides the fault into two major categories: external circuit fault or internal drive fault.

7. Motor and Cable Insulation Testing

For a 400 V, 260 kW motor, insulation testing must not be done only with a standard multimeter. A normal multimeter may detect a dead short, but it cannot reliably identify moisture, insulation aging, partial breakdown, or leakage that only appears under higher test voltage.

A 1000 V insulation resistance tester, commonly called a megger, should be used.

Before insulation testing, the motor cables must be disconnected from the drive output terminals. This is essential because megger voltage can damage the VFD output circuit if applied while the drive is still connected.

Recommended insulation tests:

  1. U phase to earth.
  2. V phase to earth.
  3. W phase to earth.
  4. U to V.
  5. V to W.
  6. U to W.

For this class of motor, the insulation resistance should ideally be in the tens or hundreds of megaohms. If the value is low, the motor should not be connected back to the drive until the cause is found.

A practical interpretation:

  • Above 100 MΩ: generally good.
  • 10–100 MΩ: suspicious, especially in a humid site.
  • Below 10 MΩ: not recommended for VFD operation without further investigation.
  • Below 1 MΩ: serious insulation problem.

The cable should also be tested separately if possible. Many SCF1 faults are caused not by the motor winding itself, but by the motor cable.

Common cable-related causes include:

  1. Damaged cable insulation.
  2. Moisture in cable joints.
  3. Cable crushed inside conduit.
  4. Shielding layer touching a phase conductor.
  5. Carbonized terminals.
  6. Loose cable lugs.
  7. Water inside the motor terminal box.
  8. Phase conductor touching the motor frame.

For large drives, cable insulation problems are very common, especially after equipment relocation, long shutdown, humid storage, or poor cabinet maintenance.

8. Checking the Output Circuit

The drive output terminals U/V/W should normally be connected directly and correctly to the motor. Any device inserted between the drive and motor must be checked carefully.

The following components can cause SCF1 if incorrectly connected at the VFD output:

  1. Power factor correction capacitor.
  2. Capacitor bank.
  3. Old star-delta starter circuit.
  4. Output contactor with poor contact.
  5. Output contactor switching during drive operation.
  6. Incorrectly connected thermal overload relay.
  7. Incorrect output filter.
  8. Incorrectly placed reactor.
  9. Multi-motor connection without proper configuration.
  10. Carbonized or loose output terminals.

A capacitor at the output of a VFD is especially dangerous. Since the VFD output is PWM, a capacitor can produce large high-frequency charging currents. This may be detected as a short circuit and may also damage the IGBT module.

Output contactors also require special attention. If a contactor opens or closes while the drive is producing output voltage, it can generate severe electrical stress and trigger short-circuit or overcurrent protection. If an output contactor must be used, it should be properly interlocked so that it never switches while the drive is actively running.

9. Can Parameter Errors Cause SCF1?

Parameter errors usually cause overcurrent, overload, unstable speed, motor overheating, or poor starting torque. However, in severe cases, incorrect parameters may contribute to SCF1 or short-circuit-like protection.

Possible parameter-related causes include:

  1. Motor rated current set too high.
  2. Motor rated voltage or frequency set incorrectly.
  3. Wrong motor control law.
  4. Acceleration time too short.
  5. Excessive torque boost.
  6. Incorrect starting frequency.
  7. Incorrect auto-tuning result.
  8. Motor power rating mismatch.
  9. Multi-motor system configured as a single motor.
  10. Heavy mechanical load with aggressive acceleration.

For the motor in this case, the basic motor parameters should be entered according to the nameplate:

  • Motor rated voltage: 400 V
  • Motor rated frequency: 50 Hz
  • Motor rated power: 260 kW
  • Motor rated current: 435 A
  • Motor rated speed: 2984 rpm
  • Motor power factor: 0.90

Initial acceleration and deceleration times should not be too short. For a high-power motor, an initial acceleration time of 30–60 seconds is safer. Heavy-load applications may require even longer ramp times.

However, if the diagnostic menu already shows Power Brick 1 Diag: Not OK, parameter adjustment alone is not enough. Parameters should be checked, but they cannot replace hardware diagnosis.

10. Mechanical Load Considerations

Although SCF1 is mainly related to electrical short-circuit protection, the mechanical side should not be ignored. If the motor is mechanically locked or the driven equipment is jammed, the starting current may become extremely high and trigger protection.

The following items should be checked:

  1. Whether the motor shaft can rotate freely.
  2. Whether the pump or fan is jammed.
  3. Whether the bearing is seized.
  4. Whether the coupling is locked.
  5. Whether the fan impeller touches the casing.
  6. Whether the belt or mechanical transmission is too tight.
  7. Whether the pump has foreign material inside.
  8. Whether the valve position is correct.
  9. Whether the process line is blocked.
  10. Whether reverse pressure or backflow exists.

If it is safe to disconnect the motor from the load, an unloaded motor test can help identify whether the fault is electrical or mechanical. If the motor runs normally without load but faults immediately under load, the mechanical system must be investigated.

11. Diagnosing Internal Drive Hardware Faults

If Power Brick 1 Diag remains Not OK even after disconnecting the motor, the fault is likely inside the drive.

The main parts to inspect are:

  1. Power Brick 1 IGBT module.
  2. IGBT gate driver board.
  3. Gate driver power supply.
  4. Gate resistors.
  5. Driver optocouplers or isolation devices.
  6. DC busbar.
  7. Output copper busbar.
  8. Current sensors.
  9. Power module connection cables.
  10. Cooling system.
  11. Control board to power board connection.
  12. Dust, moisture, oil contamination, or metal particles inside the cabinet.

IGBT module faults can appear in different forms. Sometimes the module has an obvious collector-emitter short circuit that can be found with a multimeter in diode mode. Sometimes static testing looks normal, but the IGBT fails under voltage or switching conditions. Sometimes the IGBT itself is good, but the gate driver board is defective and causes abnormal triggering or protection.

Therefore, a simple multimeter test is only a preliminary check. It cannot fully prove that a high-power IGBT module is good. A proper repair workshop should also check gate drive signals, driver power supply, desaturation feedback, current feedback, and insulation conditions.

12. Current Sensor and Detection Circuit Problems

The SCF1 judgment depends heavily on current feedback. If the current sensor or its detection circuit is faulty, the drive may incorrectly detect a short-circuit condition.

Possible symptoms include:

  1. Output current displayed when the drive is not running.
  2. Unbalanced phase current display.
  3. Overcurrent or short-circuit fault with no motor connected.
  4. Power brick diagnostic failure.
  5. Fault condition changing when cabinet wiring is touched or vibrated.
  6. Intermittent SCF1 after the drive warms up.

The current monitoring values should be checked from the keypad. When the drive is stopped, output current should be close to zero. If the displayed current is abnormal in the stopped state, the current sensor, sensor power supply, signal cable, or control board input circuit should be checked.

Loose connectors, oxidized plugs, moisture, dust, and damaged shielding can all affect current detection accuracy.

13. Cabinet Environment and Maintenance Factors

The ATV660 is a cabinet drive. Its reliability depends strongly on the cabinet environment and cooling condition. Dust, moisture, oil mist, metal powder, blocked filters, and poor ventilation can all cause electrical and thermal problems.

Environmental problems may cause:

  1. IGBT overheating.
  2. Gate driver board leakage.
  3. Conductive dust between busbars.
  4. Oxidized connectors.
  5. Cooling fan failure.
  6. Heat sensor abnormality.
  7. Capacitor aging.
  8. Condensation inside the cabinet.
  9. Terminal overheating.
  10. Control board misdiagnosis.

The cabinet should be inspected carefully. Air filters should be cleaned or replaced. Cooling fans should be checked. The power section should be inspected for black marks, smell of burning, carbon tracking, loose screws, and foreign objects.

For high-power drives, poor cooling can gradually weaken the power module and finally cause power brick diagnostic failure.

14. Recommended Troubleshooting Procedure

A structured troubleshooting process for this fault should be as follows:

  1. Record the fault code, diagnostic screen, running frequency, current, status word, and command word.
  2. Stop repeated reset and restart attempts.
  3. Switch off the main power and wait for full DC bus discharge.
  4. Check output terminals U/V/W for looseness, burn marks, wrong wiring, or foreign objects.
  5. Disconnect the motor cables from the drive output.
  6. Power on the drive without the motor connected.
  7. Check whether Power Brick 1 Diag changes to OK.
  8. If the drive is normal without the motor, test the motor and cable insulation with a 1000 V megger.
  9. Check the motor terminal box for water, loose terminals, carbon marks, and winding imbalance.
  10. Check whether there are contactors, capacitors, star-delta circuits, output filters, or other devices between the drive and motor.
  11. If all external circuits are normal, perform a cautious low-frequency local test.
  12. If Power Brick 1 Diag remains Not OK without the motor, inspect the internal power section.
  13. Test the IGBT module, gate driver board, current sensor, busbar, and power module connections.
  14. After repair, test the drive without load first.
  15. Reconnect the motor only after the drive and external circuit pass inspection.
  16. Start with low frequency, observe current balance, then gradually increase speed.
  17. Restore remote UCP or PLC control only after the hardware fault is cleared.

The key principle is to separate the system into sections and test each section independently.

15. Commissioning After Repair

After the fault is repaired, the drive should not be returned to full operation immediately. A gradual commissioning process is necessary.

Recommended steps:

  1. Confirm all power terminals are tightened.
  2. Confirm motor insulation is acceptable.
  3. Confirm no tools, screws, or metal particles remain inside the cabinet.
  4. Confirm all fans and cooling paths are working.
  5. Confirm motor nameplate data is correctly entered.
  6. Select Local mode from the keypad.
  7. Start at low frequency, such as 5 Hz.
  8. Observe motor direction, current, sound, and vibration.
  9. Increase to 10 Hz, 20 Hz, and 30 Hz step by step.
  10. Check three-phase current balance.
  11. Check motor temperature and mechanical load condition.
  12. Only after stable local running should Remote or UCP control be restored.
  13. Confirm the command source and speed reference source before automatic operation.

For UCP or PLC control, the command source and reference source must be correctly configured. Start/stop may come from terminals, communication, or keypad. Speed reference may come from analog input, Ethernet communication, Modbus RTU, or another fieldbus. Incorrect communication setup usually does not directly cause SCF1, but it may cause unintended run commands, wrong speed reference, or confusion during troubleshooting. Therefore, communication configuration should be handled after the SCF1 fault source has been identified and cleared.

16. Conclusion

When a Schneider ATV660 drive reports SCF1 Motor short circuit and the diagnostic page shows IGBT Diag w motor / Power Brick 1 Diag: Not OK, the fault must be treated as a serious output short-circuit or power-stage diagnostic failure. It should not be considered a simple keypad setting issue, and it should not be handled by repeatedly resetting and restarting the drive.

The correct diagnostic logic is to first protect the equipment, then isolate the motor from the drive, and then determine whether the fault is external or internal. If Power Brick 1 Diag becomes OK after disconnecting the motor, the focus should be on motor insulation, motor cable, output wiring, contactors, capacitors, old starter circuits, and mechanical load. If Power Brick 1 Diag remains Not OK after disconnecting the motor, the drive itself probably has an internal hardware problem, such as IGBT module failure, gate driver board fault, current sensor abnormality, busbar insulation issue, or power brick detection failure.

For a 250–315 kW cabinet drive, the cost of misdiagnosis can be high. A systematic approach using isolation testing, megger testing, output circuit inspection, power module checking, and controlled low-frequency commissioning is essential. Only after the motor, cable, output circuit, and drive power section are confirmed normal should the system be returned to UCP, PLC, or remote automatic control.

The SCF1 fault is not merely a single alarm code. It is a protection result generated by the interaction of the drive, motor, cable, load, and control system. A professional repair approach must follow the principle of safety first, isolation second, measurement third, judgment fourth, and commissioning last. This is the only reliable way to avoid secondary damage and restore the drive system safely.

Posted on by

Troubleshooting Schneider ATV660 Drives: Internal Error 29 Caused by X22 Communication Cable Issues

Abstract

Industrial variable frequency drives (VFDs) play a critical role in modern automation systems, providing precise control of motor speed and torque. The Schneider Electric Altivar Process ATV660 series is widely used in heavy industrial applications, where high reliability and operational continuity are paramount. Despite robust design, certain internal errors can occur, potentially disrupting production. This article examines Internal Error 29, a common fault code reported on the ATV660, and analyzes a case where the root cause was identified as a poorly connected X22 communication cable, rather than a control board failure. The discussion covers system architecture, error diagnostics, maintenance best practices, and preventive strategies for engineers and technicians.

A clean, technical illustration of a Schneider Electric Altivar Process ATV660C31Q4X10-EP drive. The front HMI panel displays "Internal Error 29" with detailed status lines. The drive is shown in a semi-cutaway view, revealing the internal control board and power module. A red dashed outline and arrow highlight the X22 communication cable connecting the control board to the power module. Callouts on the right label the HMI Panel, Control Board, Power Module, and X22 Communication Cable. A note at the bottom-right warns that a loose connection may cause Internal Error 29. The background is white, layout is article-ready, with clear, professional labels and diagram style suitable for technical documentation.

1. Introduction

Variable frequency drives (VFDs) regulate AC motor operation by adjusting frequency and voltage supplied to the motor. In high-power industrial environments, failures in VFDs can result in substantial downtime and production losses. Schneider Electric’s ATV660 series is designed for demanding applications, offering integrated process control, energy efficiency, and communication with higher-level automation systems via standardized protocols.

Despite their robustness, internal errors such as Internal Error 29 occasionally occur. Traditionally, this error has been associated with control board malfunctions, EEPROM issues, or firmware anomalies. However, real-world case studies demonstrate that internal errors can sometimes arise from external hardware connections, particularly communication cables. This article documents a case study, explores root cause analysis, and provides guidelines for troubleshooting similar faults.

2. ATV660 System Architecture Overview

The ATV660 drive consists of several key modules:

  1. Power Module (PM): Handles the conversion from AC input to controlled AC output for the motor. Includes IGBT bridges, DC bus capacitors, and output filters.
  2. Control Board (CB): Implements drive logic, motor control algorithms, and communication with other modules and supervisory systems.
  3. Interface and Communication Modules: Facilitate connectivity to process automation networks and higher-level SCADA systems. The X22 cable is one of the internal communication links connecting the control board to the power module.
  4. Input/Output Terminals: Accepts field signals for start/stop, speed commands, and feedback.
  5. HMI Panel: Provides status, parameter configuration, and fault reporting to operators.

The X22 cable specifically transmits critical synchronization and monitoring signals between the control board and power module. Any disruption in this link can cause the drive’s self-diagnostic routine to report Internal Error 29, as the control board detects a lack of expected communication.

A troubleshooting workflow infographic for the Schneider Altivar Process ATV660C31Q4X10-EP showing Internal Error 29 resolution. Four sequential panels illustrate: 1) Drive front panel displays "Internal Error 29"; 2) Technician inspects internal communication wiring, highlighting the X22 cable not properly fixed; 3) Technician reseats and secures the X22 connector; 4) Drive returns to normal operation with the display showing "RUN" and parameters like frequency, motor current, and output voltage. Each step has numbered green badges, arrows indicate progression, and captions provide concise instructions. The layout is clean, technical, with a light background, suitable for an engineering article.

3. Internal Error 29: Typical Causes

The drive’s self-diagnostics categorize Internal Error 29 as a critical internal communication or control failure. Common triggers include:

  • Faulty control board components (logic IC, FPGA, or EEPROM)
  • Power module faults affecting control signals
  • Firmware corruption or mismatch
  • Poor connections in internal communication cables (e.g., X22, X21)
  • Intermittent or loose wiring between modules
  • Environmental factors (vibration, dust, or moisture)

Historically, many engineers default to assuming the error indicates a board failure, leading to unnecessary control board replacements. While hardware faults can cause this error, as demonstrated in this case study, cable issues may produce identical symptoms without damaging the main boards.

4. Case Study: X22 Communication Cable Issue

4.1 Fault Description

A client in Nigeria reported an ATV660C31Q4X10-EP drive displaying Internal Error 29. Initial assessment assumed a control board malfunction, given the error’s reputation. The drive was rated at 315 kW / 250 kW, connected to a 3-phase 380–415V input, and used in a heavy industrial process.

4.2 Initial Diagnostic Approach

  1. Power cycle and reset via the HMI panel
  2. Verified input and output voltages
  3. Inspected motor connections for short circuits
  4. Checked drive parameters and firmware version using SoMove

Despite these steps, the error persisted. No visible damage was found on the control board or power module.

4.3 Root Cause Identification

A detailed visual inspection revealed the X22 communication cable was not properly fixed. The connector had a slight displacement, leading to intermittent loss of signal between the control board and power module. This misalignment caused the drive’s internal logic to detect a failure in communication, thus triggering Internal Error 29.

After securely reconnecting the X22 cable:

  • Internal Error 29 cleared immediately
  • Drive returned to normal operation without replacing any hardware
  • No additional faults or error codes appeared

This case highlights the importance of inspecting all internal communication links, especially following maintenance or transport.

5. Diagnostic Strategy for Internal Error 29

To troubleshoot Internal Error 29 effectively, engineers should follow a structured approach:

  1. Document Drive Status
    • Note error codes, operating conditions, and drive parameters.
  2. Perform Controlled Restart
    • Power down the drive for 5–10 minutes.
    • Power it back on and attempt a reset.
  3. Verify External Connections
    • Ensure motor cables, field wiring, and grounding are correct.
    • Inspect X22 and other internal communication cables for proper seating.
  4. Examine Internal Modules
    • Check the control board, power module, and interface connections.
    • Look for loose connectors, dust, or mechanical stress on boards.
  5. Software and Firmware Verification
    • Backup parameters via SoMove or Drive Composer.
    • Confirm firmware versions and compatibility.
  6. Controlled Test Operation
    • Run the drive without a load, monitor for error recurrence.
    • If errors persist after cable inspection, consider module replacement.

By following this approach, unnecessary control board replacements can be avoided, reducing downtime and repair costs.

6. Maintenance Best Practices

  1. Cable Management
    • Ensure all internal connectors are securely fastened.
    • Label cables and document connections for future reference.
  2. Periodic Inspection
    • Schedule visual inspections of connectors, especially after maintenance or shipping.
    • Use retention clips or cable ties to prevent loosening due to vibration.
  3. Environmental Control
    • Keep drive compartments clean and dry.
    • Limit exposure to dust, moisture, and extreme temperatures.
  4. Operator Training
    • Train personnel on proper handling of internal connectors.
    • Emphasize the importance of checking communication cables when errors occur.
  5. Parameter Backup
    • Regularly backup drive parameters and firmware.
    • Maintain logs of firmware updates and maintenance activities.

7. Preventing Recurrence

Internal Error 29 caused by communication cable issues is entirely preventable:

  • Ensure proper mechanical fixation of X22 and other critical cables.
  • Verify connectors after any maintenance, transport, or vibration exposure.
  • Document all maintenance and inspections to provide traceability.

Implementing these preventive measures ensures higher operational reliability, minimizes unnecessary hardware replacements, and maintains consistent production uptime.

8. Conclusion

The Schneider ATV660 series is robust, but like all industrial drives, it is vulnerable to internal errors triggered by minor issues such as improperly fixed communication cables. The case study presented here demonstrates that:

  • Internal Error 29 is not always indicative of control board failure.
  • Loose or poorly connected X22 cables can produce identical error conditions.
  • Systematic diagnostics, careful inspection, and preventive maintenance can resolve errors efficiently without hardware replacement.

By adopting structured troubleshooting and preventive strategies, industrial engineers can enhance drive reliability, reduce repair costs, and prevent unnecessary downtime.

9. References

  1. Schneider Electric, Altivar Process ATV660 User Guide, 2021.
  2. SoMove / Drive Composer Software Manuals, Schneider Electric.
  3. Industrial VFD Maintenance Best Practices, ISA (International Society of Automation), 2020.

Posted on by

AC Servo Drive MG-KAS20AA “A.03” Alarm Fault Analysis and Handling Guide

1. Overview of the Fault

In industrial automation, servo drives are essential for precise control of position, speed, and torque. Based on user reports and inspection images (Attachments 1 and 2), the MG-KAS20AA K-series AC servo drive exhibits the “A.03” alarm code. This code appears on the front panel digital display (Image 2), and the drive fails to rotate, halting the connected mechanical load.

According to the K Series AC Servo Drive User Manual (2017 Engineer Edition V3.0), Chapter 7 and Appendix C, the A.03 alarm falls under overload/torque anomaly faults, primarily associated with:

  1. Servo drive circuit board faults
  2. Motor wiring issues
  3. Encoder signal errors
  4. Load torque exceeding the drive’s limits

Images show U/V/W motor terminals correctly connected, CN1/CN2 encoder interfaces installed, and PE properly grounded, yet the A.03 alarm persists. This indicates the fault is likely related to the drive board or signal compatibility rather than simple wiring issues.

MG-KAS20AA front panel and wiring terminals.

2. Fault Trigger Conditions

Based on the manual, A.03 may be triggered in the following scenarios:

ScenarioTrigger ConditionPossible CauseRecommended Action
Servo ONMotor does not rotateMotor wiring abnormality, encoder wiring issueInspect and correct motor and encoder wiring
Command inputServo motor unresponsiveStart-up torque exceeds maximumAdjust load conditions or re-evaluate motor capacity
Normal operationDrive reports A.03High internal temperature of servo driveReduce drive temperature below 55℃, verify cooling system
Any operationDrive board faultDrive power module or control board malfunctionReplace servo drive or repair circuit board

Given the images, the drive reports A.03 under normal power and command input. Hence, drive board or power module failure is the primary suspected cause.

Front panel display showing A.03 alarm.

3. Detailed Fault Diagnosis Steps

3.1 Visual Inspection and Wiring Verification

  1. Power Check
    • Verify L1/L2/L3 terminals receive 220V three-phase within ±15%.
    • Confirm L1C/L2C control voltage is stable.
  2. Motor Wiring
    • U/V/W terminals correspond to the drive terminals.
    • Measure resistance across phases; check for open or short circuits.
  3. Grounding and Shielding
    • PE terminals connected to drive, motor, and cabinet.
    • Encoder shield connected to the chassis.

3.2 Encoder Signal Check

Per Manual Section 3.4:

  • CN1: Axis A encoder
  • CN2: Axis B encoder

Procedure:

  1. Measure A/B phase signals with an oscilloscope.
  2. Verify PG pulse output matches user parameter settings.
  3. Ensure IN1~IN8 input allocation (P□509~P□512) is correct.
  4. Confirm wiring length and shield integrity (max 3m for command input, max 20m for feedback).

Faulty encoder signals may mislead the drive’s load detection and trigger A.03.

3.3 Drive Board and Power Module Inspection

Manual 7.2.3 highlights common failure points:

  1. Power Modules (IGBTs/MOSFETs)
    • Shorted or open MOSFETs can trigger overcurrent protection (A.03).
    • Measure U/V/W terminal resistances offline; check MOSFETs.
  2. Drive Temperature
    • Overheating or sensor failure can cause A.03.
    • Use infrared thermometer to monitor PCB temperature (<55℃).
  3. Control Board
    • MCU or logic faults may prevent overload signal processing.
    • Check for burnt components or swollen capacitors; replace control board if needed.

3.4 Load Evaluation

A.03 may also result from excessive load torque:

  • Load inertia exceeding 5× motor inertia.
  • Mechanical resistance or over-torque beyond rated motor torque.
  • Aggressive start/stop conditions causing current peaks.

Mitigation:

  1. Inspect load bearings and couplings for jamming.
  2. Measure mechanical torque against motor rating.
  3. Adjust dynamic braking or P-OT / N-OT limit parameters.

3.5 Software and Parameter Verification

  • Check user parameters P□□□: torque limit, load inertia, and travel limits.
  • Confirm control mode (position/speed/torque) matches mechanical load.
  • For absolute encoders, ensure F□009/F□010 settings are correct.
Appendix C fault table excerpt (overload/circuit board/wiring fault).

4. Fault Handling and Recovery

  1. Immediate Measures
    • Power down for at least 15 minutes to discharge capacitors.
    • Inspect cooling and airflow.
  2. Wiring and Encoder Verification
    • Cross-check terminals per manual 3.1–3.4.
    • Confirm encoder signals via oscilloscope.
  3. Circuit Board or Module Maintenance
    • If wiring and encoder are correct, replace power modules or control board.
    • Alternatively, send to manufacturer for repair.
  4. Parameter and Load Adjustment
    • Ensure user parameters are within safe limits.
    • Adjust load or enable torque compensation to reduce peak currents.
  5. Long-term Protection
    • Maintain drive environment below 45℃.
    • Avoid high humidity, dust, or corrosive gases.
    • Ensure proper PE grounding.
    • Inspect encoder and motor connections periodically.

5. Conclusion

The A.03 alarm on K-series AC servo drives indicates overload/torque anomaly, caused by:

  1. Drive board or power module failure
  2. Motor wiring or encoder signal issues
  3. Load exceeding motor capacity
  4. Overheating or insufficient cooling

Resolution Principle:

  • Inspect wiring and connections first.
  • Verify encoder signals and input/output parameter allocation.
  • Evaluate load and mechanical conditions.
  • Replace drive board or power module if necessary.

Following this systematic approach ensures reliable operation of MG-KAS20AA drives, minimizing downtime and safeguarding industrial automation processes.

Posted on by

Systematic Analysis of SIMCO-ION CM20-P Electrostatic High-Voltage Generator Tripping and OVERLOAD Faults

1. Equipment Background and Fault Description

The SIMCO-ION Chargemaster CM20-P is an industrial electrostatic high-voltage generator. It is commonly used in electrostatic charging, electrostatic adhesion, film processing, printing and packaging, plastic sheet handling, lamination systems, and automated production lines. Its main function is to convert a standard AC input supply into a high-voltage DC output, which is then supplied to static bars, electrodes, charging heads, or electrostatic holding devices.

According to the nameplate, the main specifications of this unit are:

ItemSpecification
BrandSIMCO-ION
ModelChargemaster CM20-P
Input Power230VAC, 50/60Hz
Input Current0.3A
Output Voltage+20kV
Output Current0.5mA
Fuse630mA, 5×20mm, Time Lag

Although the input power of this type of equipment is not high, the output voltage reaches +20kV. It is a typical low-current, high-voltage device. Such equipment is very sensitive to grounding, insulation condition, humidity, contamination, high-voltage cable condition, static bar cleanliness, and electrode-to-metal distance.

Once leakage, discharge, short circuit, or insulation breakdown occurs at the high-voltage output side, the unit may show symptoms such as OVERLOAD alarm, blown fuse, leakage breaker tripping, no output, abnormal display, or unstable operation.

In this case, the fault development was very typical:

At the beginning, the OVERLOAD red indicator on the front panel was lit. This usually means that the high-voltage output was abnormal or the load was excessive. Later, inspection found that an internal fuse had blown. After replacing the fuse, the main display powered on, and the OVERLOAD red light was no longer lit. However, a new fault appeared: the breaker tripped when the unit was powered on.

The customer also reported that when the equipment was used inside the factory, the breaker tripped as soon as the earth wire was connected. The site power system was described as three-phase four-wire.

These pieces of information indicate that the fault is not simply a blown fuse or a grounding question. It involves several possible fault directions, including high-voltage output overload, internal power circuit short circuit, leakage to ground, and improper site grounding system. The troubleshooting must be carried out systematically from four aspects: input power, protective earth, high-voltage output, and internal components.

Technician troubleshooting a SIMCO-ION Chargemaster CM20 electrostatic high-voltage generator on a repair bench, using a digital multimeter to check the internal power board, fuse area, and input circuit after an overload fault.

2. Meaning of the OVERLOAD Indicator

The OVERLOAD indicator on an electrostatic high-voltage generator is usually not a general power alarm. It is closely related to the high-voltage output condition. It normally means that the unit cannot establish the required output voltage, or the output current has exceeded the permitted range.

Common causes include:

  1. Short circuit at the high-voltage output;
  2. Internal leakage in the static bar;
  3. Damaged or aged high-voltage cable insulation;
  4. Electrode installed too close to a metal frame;
  5. Dust, oil, moisture, or contamination causing surface creepage;
  6. Internal breakdown of the high-voltage module;
  7. Abnormal high-voltage feedback detection circuit;
  8. Load exceeding the design capacity of the generator.

The CM20-P output is +20kV and 0.5mA. Although the current is very small, in a high-voltage electric field, even slight moisture, dust, oil contamination, burrs, or carbonized tracks can form a leakage path. When the generator detects abnormal output current, it lights the OVERLOAD indicator to warn of output overload or insulation failure.

Therefore, the initial OVERLOAD alarm already indicated that the equipment or its external load had a real fault. The later blown fuse was only a result of the fault progressing further. Replacing the fuse does not mean the equipment has recovered. A blown fuse usually means that the power circuit experienced overcurrent, short circuit, or an abnormal surge.

Factory electrical diagnosis of a SIMCO-ION Chargemaster CM20-P high-voltage generator connected to L, N, and PE wiring, showing earth leakage troubleshooting, a tripped leakage breaker, and grounding inspection.

3. A Blown Fuse Is Usually a Result, Not the Root Cause

When field technicians see a blown fuse, the first reaction is often to replace it. However, in industrial electronic equipment, the fuse itself is rarely the root cause. Its function is protection. If it blows, it means the downstream circuit has drawn abnormal current.

The CM20-P nameplate specifies:

630mA, 5×20mm, Time Lag

This means:

T630mA / 250V / 5×20mm slow-blow fuse

There are two important points here.

First, the fuse rating must not be increased randomly.
If the original fuse is 630mA slow-blow, it must not be replaced with a 1A, 2A, or larger fuse simply to prevent it from blowing again. It is even more dangerous to use copper wire instead of a fuse. This may keep the circuit powered temporarily, but it can cause more serious damage to components, PCB traces, transformers, high-voltage modules, or even create a fire and electric shock hazard.

Second, fast-blow and slow-blow fuses must not be used interchangeably without analysis.
When the equipment starts, there may be short-duration inrush current from filter capacitor charging, transformer magnetizing current, or high-voltage circuit startup. A slow-blow fuse is designed to tolerate this short inrush. If a fast-blow fuse is installed incorrectly, it may blow even during normal startup.

However, if the correct slow-blow fuse is installed and the breaker still trips or the fuse blows again, the downstream circuit must be checked. Repeated power-on testing is not acceptable.

In this case, after the fuse was replaced, the display could light up, proving that part of the low-voltage supply had recovered. But the subsequent breaker tripping proves that the root fault remained. Replacing the fuse merely restored the current path. The real short circuit, leakage, or high-voltage insulation fault was still present.

4. Breaker Tripping: MCB Trip or RCD/RCBO Trip?

When a customer says “the breaker trips,” the first task is to identify what type of protection device is tripping. Different protection devices indicate different fault directions.

If an ordinary MCB trips, the main concern is overcurrent or short circuit. Typical causes include L-N short circuit, shorted rectifier bridge, shorted main electrolytic capacitor, shorted power transistor, or shorted transformer primary winding.

If an RCD or RCBO trips, the main concern is earth leakage. Typical causes include leakage in the input filter, abnormal Y capacitor, high-voltage circuit breakdown to chassis, moisture or carbonization at the output socket, damaged high-voltage cable, or incorrect N/PE connection on site.

In this case, the customer’s photo showed a CHINT NXBLE-63 C63 breaker. The NXBLE-63 is a leakage-protection circuit breaker, commonly known as an RCBO. It can trip because of earth leakage, but it can also trip because of overload or short circuit. Therefore, its tripping does not automatically prove that the fault is purely earth leakage. However, it does prove that the equipment caused an abnormal condition at power-up.

The fault must be further divided into the following situations.

4.1 The Breaker Trips Immediately When the Plug Is Inserted with the Power Switch OFF

If the front power switch of the unit is OFF, but the breaker trips immediately when the power plug is inserted, the fault is usually located before the main power switch.

The key areas to check are:

  • Power cord;
  • IEC power inlet;
  • Fuse holder;
  • Input EMI filter;
  • MOV surge suppressor;
  • NTC inrush limiter;
  • Insulation between L/N/PE;
  • Damaged input wiring touching the chassis.

In this state, the main high-voltage circuit may not have started yet.

4.2 The Plug Can Be Inserted Normally, But the Breaker Trips When the Unit Is Switched ON

If the plug is inserted without tripping, but the breaker trips when the front switch is turned ON, the fault is more likely in the downstream power supply or high-voltage generation circuit.

Key areas to check include:

  • Rectifier bridge;
  • Main filter capacitor;
  • Switching transistor;
  • Driver circuit;
  • Step-up transformer;
  • Voltage multiplier circuit;
  • High-voltage output socket;
  • High-voltage feedback circuit.

4.3 It Does Not Trip Without Ground, But Trips as Soon as PE Is Connected

If the equipment seems to power on when the earth wire is not connected, but trips immediately when protective earth is connected, there are usually two possibilities.

The first possibility is that the equipment itself has leakage to chassis or PE.
The second possibility is that the site grounding system is incorrect, with neutral and protective earth mixed, or the so-called “earth wire” is not a real PE conductor.

This kind of symptom is common in high-voltage electrostatic equipment because the high-voltage output, input EMI filter, and metal chassis have complex insulation relationships. Once an internal leakage path exists, connecting PE allows leakage current to return through the grounding system. The RCBO detects imbalance between line and neutral current and trips.

5. Grounding Risk in a Three-Phase Four-Wire Factory Supply

The customer mentioned that the site supply is “three-phase four-wire.” This is a critical detail.

In industrial sites, two common systems are three-phase four-wire and three-phase five-wire.

A three-phase four-wire system usually consists of:

L1, L2, L3, N

A three-phase five-wire system consists of:

L1, L2, L3, N, PE

The CM20-P is a 230VAC single-phase input device. Its correct wiring is:

L + N + PE

Here, PE is protective earth. It must not be replaced by N. In many three-phase four-wire sites, there may be no independent PE conductor. Some users may incorrectly use neutral as earth, or connect the equipment chassis to steel structures, water pipes, cable trays, or machine frames. These practices can cause RCBO tripping and also create electric shock hazards.

An RCBO works by comparing the current flowing through the line conductor and the current returning through the neutral conductor. Under normal conditions, the current going out through L should return through N. If part of the current returns through PE, chassis, steel structure, or another path, the RCBO detects an imbalance and trips.

Therefore, if a unit trips as soon as the earth wire is connected in a three-phase four-wire site, the following possibilities must be considered:

  1. There is no real PE at the site;
  2. Neutral is being used incorrectly as protective earth;
  3. N and PE are mixed on the load side of the leakage breaker;
  4. The equipment PE is connected to the wrong neutral bar;
  5. The factory steel structure has a different potential from the power supply earth;
  6. Other equipment on the same grounding network has leakage;
  7. The CM20-P itself has internal leakage to chassis, and grounding exposes the fault.

The correct solution is to have a qualified electrician verify the grounding system. It is not acceptable to randomly select a wire and call it “earth.” For a +20kV electrostatic generator, protective earth is not optional. It is a necessary safety condition.

6. Correct On-Site Verification Method

When a unit shows breaker tripping, earth-wire tripping, previous fuse blowing, and OVERLOAD history, it should not be tested directly with the external static bar connected. The verification must be performed in stages.

Stage 1: Verify Power Supply and Grounding

Use a multimeter to measure the supply point:

Measuring PointNormal Result
L-NApproximately 220V / 230V
L-PEApproximately 220V / 230V
N-PEClose to 0V, usually within a few volts

If L-N is normal but L-PE is abnormal, PE is unreliable.
If N-PE has a significant voltage, the site neutral-earth system may be faulty.
If the site has only three-phase four-wire without independent PE, N must not be used directly as protective earth.

Stage 2: Disconnect All High-Voltage Loads

Disconnect the external static bar, high-voltage cable, electrode, and charging head. Only the main generator should remain connected to the input power.

Then test:

  • Does it trip when the plug is inserted with the power switch OFF?
  • Does it trip only when the switch is turned ON?
  • Does it trip only when PE is connected?
  • Does it trip without external high-voltage cables?

If it no longer trips after the external load is disconnected, the fault is likely in the static bar, high-voltage cable, or installation environment.

If it still trips with no external load, the fault is inside the generator.

Stage 3: Separate Input-Side Faults from High-Voltage-Side Faults

If it trips with the switch OFF, check the input side.
If it trips only after switching ON, check the main power and high-voltage generation stage.
If it trips only when PE is connected, check insulation to PE, input filter leakage, and high-voltage module leakage to chassis.
If it trips only after connecting the static bar, check the external high-voltage cable and static bar.

This staged method is much more reliable than blind component replacement.

7. Key Internal Circuit Areas to Inspect

The internal structure of the CM20-P generally includes input protection, rectification and filtering, control circuit, power drive, step-up circuit, and high-voltage output stage. In this case, the following areas should be inspected carefully.

7.1 Input Fuse and Fuse Holder

Confirm that the installed fuse is:

T630mA / 250V / 5×20mm slow-blow

Check whether the fuse holder is burnt, loose, oxidized, or deformed. Poor contact in the fuse holder can cause local heating, arcing, or intermittent faults.

If the fuse blows again, stop replacing it and proceed with short-circuit testing.

7.2 Input MOV Surge Suppressor

The internal photos show a black disc-shaped component near the input area. This type of component is commonly used as an MOV or surge suppressor. It absorbs lightning surges, overvoltage spikes, and switching transients.

When an MOV fails, it may become low-resistance or fully shorted. This can cause the fuse to blow or the breaker to trip immediately at power-up.

For accurate testing, one leg should be lifted from the circuit or the component should be isolated before measuring. If the MOV measures low resistance, it is faulty.

7.3 Input EMI Filter and Y Capacitors

Industrial equipment with a metal chassis often uses an EMI filter. The filter normally includes X capacitors, Y capacitors, and common-mode inductors. Y capacitors are connected between L/N and PE. Under normal conditions, they produce a very small leakage current. If a Y capacitor ages, absorbs moisture, or breaks down, leakage to PE may increase and trip the RCBO.

When the symptom is “the breaker trips as soon as the earth wire is connected,” the EMI filter and Y capacitors must be checked carefully.

7.4 Rectifier Diodes or Rectifier Bridge

The internal board shows several diode positions such as D2, D6, and D7. If any rectifier diode breaks down short, the fuse may blow and the breaker may trip. Use the diode test mode of a multimeter to measure forward and reverse voltage drops. If both directions read nearly zero, the diode is shorted.

7.5 Main Electrolytic Capacitors

High-voltage generators usually include main filter capacitors in the power supply section. If an electrolytic capacitor is shorted, severely leaky, swollen, or leaking electrolyte, it can cause abnormal input current.

After power is disconnected and capacitors are safely discharged, measure the resistance across the capacitor terminals. It should not remain near zero ohms. If a low resistance is present, isolate the capacitor or downstream DC bus to confirm whether the capacitor or the circuit is shorted.

7.6 Power Switching Transistor

Electrostatic high-voltage generators often use a switching power supply topology to drive a high-voltage step-up transformer. If the power switching transistor fails short between D-S or C-E, it can short the DC bus and cause fuse blowing, breaker tripping, or failure to start.

The main terminals of the switching device must be checked for short circuit. The gate/base drive circuit, gate resistors, snubber components, and fast recovery diodes should also be inspected because they are often damaged together.

7.7 Step-Up Transformer and High-Voltage Module

The initial OVERLOAD alarm strongly suggests a high-voltage output-side abnormality. If the step-up transformer, voltage multiplier capacitors, high-voltage diodes, or encapsulated HV module breaks down, it may cause leakage to chassis, internal discharge, or output short circuit.

A normal multimeter may not always detect high-voltage insulation failure. Megger testing, sectional isolation, unloaded power testing, and visual inspection for discharge marks may be required.

7.8 High-Voltage Output Socket and HV Cable

The output sockets and high-voltage terminals must be inspected carefully. At +20kV, even light dust, moisture, oil contamination, or carbonized marks can form a creepage path. If the high-voltage cable insulation is damaged or routed too close to the metal chassis or frame, leakage and OVERLOAD alarms can occur.

A typical feature of this type of fault is that low-voltage resistance checks may appear normal, but the unit alarms or trips once high voltage is generated.

8. Influence of External Static Bars and Installation Environment

In an electrostatic system, the generator is only the high-voltage source. The real fault is often located in the external load. Static bars, electrodes, and high-voltage cables work under strong electric fields for long periods. They are easily affected by dust, oil, moisture, aging, mechanical stress, and improper installation.

Common external problems include:

  1. Damaged high-voltage cable jacket;
  2. Reduced insulation inside the static bar;
  3. Contaminated electrode needles;
  4. Static bar installed too close to a metal roller, frame, or guard;
  5. High ambient humidity;
  6. Oil mist or dust causing surface creepage;
  7. Carbonization inside the high-voltage plug;
  8. Breakdown at cable bending points.

Therefore, to determine whether the CM20-P generator itself is faulty, an unloaded test is mandatory. If the generator no longer trips or alarms after all external loads are disconnected, the generator may not have a serious internal short circuit. The troubleshooting focus should then move to the high-voltage cable, static bar, and installation environment.

If the generator still trips with no external load, an internal fault is confirmed.

9. Why the Unit Must Not Be Tested Without Protective Earth

The customer once tested the equipment in another room without an earth wire. This is unsafe and unsuitable for diagnosing high-voltage electrostatic equipment.

The warning label on the CM20-P clearly requires that the earthing wire be connected before operation. The reasons include:

  1. The metal chassis must be protectively earthed to prevent electric shock if internal leakage occurs;
  2. The high-voltage output system requires a stable reference potential;
  3. The EMI filter requires a PE path;
  4. High-voltage discharge energy must be safely released;
  5. Without earth, the chassis may float to an unsafe potential.

When the unit is not grounded, some leakage faults may not immediately show because the leakage current has no clear return path. This does not mean the equipment is normal. Once PE is connected, the leakage path becomes complete, and the RCBO may trip immediately. In that case, grounding has not caused the fault; it has exposed the fault.

For high-voltage electrostatic equipment, the correct interpretation is:

After the equipment is grounded, the leakage fault inside the unit or the site wiring system is detected by the protection device.

10. Recommended Troubleshooting Procedure

For this case, the following procedure is recommended.

Step 1: Confirm Fuse Specification

Confirm that the replacement fuse is:

T630mA / 250V / 5×20mm slow-blow

Do not increase the rating. Do not replace the fuse with copper wire.

Step 2: Disconnect All High-Voltage Outputs

Remove the static bar, high-voltage cable, and external electrode. Test only the main generator.

Step 3: Confirm Power Supply

Use single-phase 230VAC with:

L, N, PE

PE must be a real protective earth. It must not be replaced by neutral.

Step 4: Test in Different Power States

Insert the plug with the power switch OFF and observe whether the breaker trips.
Then turn the power switch ON and observe whether it trips.
Record exactly when the trip occurs.

Step 5: Measure Input-Side Insulation and Short Circuit

After disconnecting power and discharging capacitors, measure:

  • Resistance between L and N;
  • Resistance between L and PE;
  • Resistance between N and PE;
  • Fuse downstream side to N;
  • DC bus resistance;
  • PCB insulation to chassis.

If L-PE or N-PE reads low resistance, there is leakage or short circuit to chassis.

Step 6: Check Input Protection Components

Inspect:

  • MOV surge suppressor;
  • EMI filter;
  • X/Y capacitors;
  • NTC inrush limiter;
  • Fuse holder;
  • Power switch;
  • Input wiring.

Step 7: Check Rectifier and Power Stage

Inspect:

  • Rectifier diodes;
  • Rectifier bridge;
  • Main electrolytic capacitors;
  • Power switching transistor;
  • Snubber circuit;
  • Driver circuit;
  • Primary winding of the step-up transformer.

Step 8: Check High-Voltage Output and Insulation

Inspect:

  • High-voltage module;
  • Voltage multiplier capacitors;
  • High-voltage diodes;
  • Output sockets;
  • High-voltage cable;
  • Static bar;
  • Insulation between output and chassis.

Step 9: Clean and Dry the HV Area

If dust, oil, moisture, or discharge marks are present around the high-voltage output, clean and dry the area before retesting. Severely carbonized insulation parts must be replaced. Wiping the surface is not enough.

Step 10: Power-On Verification

After repair, test the generator without load first. Then connect the high-voltage load. Before connecting the load, confirm that the static bar and high-voltage cable have no short circuit, leakage, or insulation damage.

11. Probable Fault Conclusion

Based on all the symptoms in this case, the following conclusion is reasonable:

First, the initial OVERLOAD red light indicates a high-voltage output overload, leakage, or internal high-voltage fault.
Second, the internal fuse blowing indicates that the abnormal condition developed into an overcurrent condition on the power side.
Third, after the fuse was replaced, the display recovered, but the breaker still tripped, proving that the root fault had not been removed.
Fourth, the report that the unit trips as soon as the earth wire is connected indicates possible internal leakage to earth, while the three-phase four-wire factory supply may also have an improper PE/N grounding arrangement.
Fifth, the correct next step is to verify whether the site has a real PE conductor, disconnect all external high-voltage loads, and perform an unloaded test to determine whether the fault is inside the generator or in the external static bar system.

In order of probability, the most likely fault areas are:

  1. Leakage in the high-voltage output socket, static bar, or high-voltage cable;
  2. Insulation breakdown in the high-voltage module or step-up section;
  3. Earth leakage from the input EMI filter or Y capacitors;
  4. Damaged MOV, rectifier diode, or rectifier bridge;
  5. Shorted power transistor or main electrolytic capacitor;
  6. Improper factory grounding system, especially neutral-earth confusion in a three-phase four-wire supply.

12. Repair and Operation Precautions

Electrostatic high-voltage generators must not be repaired in the same way as ordinary low-voltage power supplies. Their input power may be small, but the output voltage is very high. Incorrect operation can cause electric shock, arcing, equipment damage, or fire.

The following precautions are essential:

  1. Do not repeatedly power the unit on after it trips.
    Repeated tripping can enlarge the damage to power components and high-voltage modules.
  2. Do not increase the fuse rating.
    A blown fuse means that the downstream circuit has an abnormal condition.
  3. Do not operate the unit without protective earth.
    Grounding is a safety requirement, not an optional connection.
  4. Do not use neutral as earth.
    Three-phase four-wire does not automatically mean that PE is available. The CM20-P requires a real protective earth.
  5. Do not test the unit directly with the static bar connected.
    First disconnect external loads and confirm the condition of the generator itself.
  6. Do not ignore high-voltage output cleanliness.
    Dust, moisture, oil, and carbonized tracks are common causes of HV leakage.
  7. Do not replace the high-voltage cable with ordinary low-voltage wire.
    A +20kV output requires a suitable HV-rated cable and connector.
  8. Do not judge the equipment only by whether the display lights up.
    A working display only proves that part of the low-voltage supply is operating. It does not prove that the high-voltage section is healthy.

13. Summary

The SIMCO-ION Chargemaster CM20-P fault involving OVERLOAD indication, blown internal fuse, breaker tripping after fuse replacement, and tripping when the earth wire is connected is a typical combined fault scenario in electrostatic high-voltage equipment. The root cause may be external high-voltage cable or static bar leakage, internal high-voltage module breakdown, input filter leakage, rectifier or power-stage short circuit, or an improper factory grounding system.

The correct approach is not to repeatedly replace the fuse, repeatedly power on the unit, or remove the earth wire. Instead, the troubleshooting should follow a clear sequence: confirm the correct fuse rating, verify the L/N/PE supply, disconnect all high-voltage loads, distinguish whether the trip occurs with switch OFF, switch ON, PE connected, or external load connected, and then inspect the input protection circuit, rectifier stage, power switching stage, high-voltage module, output socket, and site grounding system.

For a +20kV electrostatic generator, grounding, insulation, and cleanliness are the foundation of safe operation. Once OVERLOAD, fuse blowing, or RCBO tripping occurs, the equipment should be taken out of service and tested systematically. Only by separating the input power, protective earth, internal power circuit, high-voltage output stage, and external static bar system can the fault be accurately located and repaired safely.

Posted on by

Designing and Debugging an M221 PLC Modbus RTU System for Multiple VFDs: ATV310, ATV31P and PowerFlex 525 Mixed-Network Application

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.

Chinese engineer commissioning a Schneider M221 PLC and multiple ATV310 VFDs in an industrial Modbus RTU control cabinet.

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.

Close-up of Schneider ATV310 Modbus wiring and R1 relay feedback terminals being checked with a test probe.

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.

Engineer programming Modbus VFD start-stop and speed control logic in Schneider Machine Expert Basic with HMI monitoring screen.

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:

DeviceSuggested Modbus Address
PowerFlex 5251
ATV310 7.5 kW2
ATV310 5.5 kW #13
ATV310 5.5 kW #24
ATV31P 5.5 kW5

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.

Safe VFD power supply setup showing a three-phase step-up transformer feeding PowerFlex 525 and ATV310 drives under engineer inspection.

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:

  1. Use twisted shielded cable for RS485 communication.
  2. Do not run communication cable parallel to motor power cables for long distances.
  3. Keep the bus structure clear and avoid random star wiring.
  4. Use termination resistors at the ends of the bus when required.
  5. Handle shield grounding correctly to avoid ground loops.
  6. Do not reverse D0 and D1.
  7. Do not treat RJ45 Modbus ports as normal Ethernet ports.
  8. Each slave address must be unique.
  9. All serial settings must be the same.
  10. 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:

  1. Confirm PLC model, HMI model, VFD models and communication interfaces.
  2. Confirm the function and wiring of the LU9GC3.
  3. Connect only one ATV310 first.
  4. Set its Modbus address, baud rate, parity and stop bit.
  5. Set its command source and speed reference source to Modbus.
  6. Configure the M221 serial line as Modbus RTU master.
  7. Read one status register from the ATV310.
  8. Write a low frequency reference, such as 10 Hz.
  9. Write the start command.
  10. Confirm that the drive starts.
  11. Write the stop command.
  12. Confirm that the drive stops.
  13. Read fault and running status.
  14. Copy the logic to the other ATV310 drives.
  15. Commission the ATV31P separately.
  16. Commission the PowerFlex 525 separately.
  17. Integrate the HMI operation screen and alarm screen.
  18. Test emergency stop, communication loss, relay feedback and fault reset.
  19. 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:

  1. Schneider ATV310 and Allen-Bradley PowerFlex 525 use different Modbus register maps.
  2. The command word bit definitions may be different.
  3. The speed reference scaling may be different.
  4. The status word decoding may be different.
  5. The fault reset command may be different.
  6. The communication parameter menus are different.
  7. Some drives default to terminal control, while others may default to keypad control.
  8. Communication timeout behavior must be configured.
  9. The HMI should be unified, but the low-level Modbus control cannot be identical.
  10. 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.