By Faults Are Information
A Fuji Electric IPM reports its faults through a single alarm output, but that one signal can mean overcurrent, short circuit, undervoltage or overtemperature. The temptation is to treat the alarm as a nuisance and restart the drive, yet the alarm is telling you something specific and useful. This article presents a systematic method for diagnosing the common Fuji Electric IPM faults in inverter, HVAC and servo applications, with a practical fix for each.
Step 1: Identify the Fault Type
The first step is to determine which protection tripped. Many controllers log only that an alarm occurred, which is not enough. Record the alarm type and its hold time, and note the load and operating condition at the moment of the trip. A trip during acceleration points to overcurrent; a trip at high ambient temperature points to overtemperature; a trip with no obvious load fault points to noise or a supply problem. Matching the symptom to the protection saves a great deal of time.
Reading the Alarm Hold Time
Each fault asserts the alarm for a characteristic hold time, and the differences can help identify the cause even when the controller does not decode the type directly. Measure the alarm pulse width and compare it with the datasheet values to narrow the possibilities.
Overcurrent Trips
Overcurrent is the most common alarm. It has three typical causes: a genuine load fault such as a short or a locked rotor, a trip level that is set too low by the sense resistor, or a gate-drive or shoot-through problem inside the inverter. Start by checking the sense-resistor value against the intended trip current. Then measure the phase current with a clamped probe during the trip to see whether the current is genuinely high or the trip is premature. Inspect the motor and wiring for a short, and check the DC-link capacitance and power-loop layout for the ringing that can produce a spurious trip.
Shoot-Through and Dead Time
In a discrete design, insufficient dead time causes shoot-through and a fast overcurrent trip. An IPM fixes its own dead time, so shoot-through from that cause is unlikely, but a damaged module or a wiring fault between phases can still produce the same effect. Inspect for a phase-to-phase short and replace any module that has been stressed by a prior fault.
Undervoltage Faults
An undervoltage fault means the control supply fell below the lockout threshold, which weakens the gate drive and risks linear-mode operation of the switches. Check the 15 V control supply for ripple and sag, especially during switching, and verify that its decoupling is close to the module. A supply that measures 15 V with a multimeter can still collapse for microseconds during a switching edge, so inspect it with an oscilloscope near the module pins.
Overtemperature Faults
An overtemperature fault means the module temperature exceeded its threshold, typically around 175 C for the X series. This is usually a thermal-design problem: an insufficient heatsink, a thick or uneven thermal interface, an incorrectly mounted module or an ambient temperature higher than assumed. Inspect the mounting for flatness and torque, measure case temperature at rated load, and confirm the heatsink against the loss estimate. Repeated overtemperature events stress the module and shorten its life, so treat them as a design defect rather than a transient.
Thermal Interface and Mounting
The thermal interface is a frequent culprit. A thick, uneven or contaminated interface raises the thermal resistance enough to cause overtemperature at rated load. Use a thin, uniform layer and the specified mounting torque, and re-measure case temperature after any rework.
False Faults from Noise
Sometimes the fault is real to the module but not to the load: electrical noise couples into the control supply or the sense path and trips the protection. Symptoms include faults at specific switching transitions, faults that correlate with motor cable length, and faults that disappear when the load is disconnected. Fixes include better control-supply decoupling, a shorter sense loop, shielding and a tighter power loop.
Clear the Fault Correctly
Once the cause is identified and fixed, clear the fault and re-enable the drive only when the load is safe. Do not rely on fast auto-restart, because a persistent fault will trip again and may damage the hardware. Logging the alarm type and the operating condition at every trip turns the fault output from an annoyance into a diagnostic tool.
Conclusion
A Fuji Electric IPM fault is a precise, useful signal once you decode it. Identify the protection that tripped, match the symptom to the cause, fix the design or the load, and clear the fault properly. With that discipline, IPM faults become a guide to better design rather than a recurring interruption.