Alarm 32 on a Mitsubishi MDS drive means power module overcurrent — the drive's intelligent power module (IPM/IGBT stage) has detected current beyond its protective limit and shut down instantly to save itself. It appears across the family — MDS-A/B, MDS-C1, MDS-D/DH, MDS-D2/DH2, MDS-E/EH — as ‘32’ on the unit LED and servo alarm 0032 on the M60/M64/M70/M80 screen. Of the three classic MDS power alarms (10 — undervoltage, 33 — overvoltage, 32 — overcurrent), this is the one where careless resetting does real damage: every reset into a hard short stresses the IPM further. Diagnose first, reset second.
What is happening
The IPM switches the DC bus onto the motor windings. Overcurrent trips when current rises faster or higher than the module allows — caused by a short (in the module itself, the cable, or the motor), a ground fault, or a control/feedback problem that commands impossible current. The trip is instantaneous and latched.
Step 1 — Isolate: drive, cable or motor?
This is the whole game with alarm 32. Power down, lock out, wait 5+ minutes for the bus to discharge, then:
-
Disconnect the motor power cable at the drive end (U/V/W). Power up and enable.
- Alarm 32 still trips with the cable off → the fault is inside the drive — shorted IPM/IGBT. Go to Step 2.
- No alarm with the cable off → the fault is downstream: cable or motor. Continue.
- Megger / insulation-test the cable and motor (500 V DC, drive DISCONNECTED at both ends): each phase to earth, and phase to phase. Anything below ~10 MΩ to earth is suspect; below 1 MΩ is your fault. Coolant-soaked motor connectors and crushed cables in the drag chain are the two classic finds.
- Split cable from motor: test the motor alone at its terminal box, then the cable alone. Winding resistance across U-V, V-W, W-U should be equal within a few percent — an imbalanced or open phase means motor rewind/replacement.
Step 2 — If the drive itself is shorted
- A shorted IPM is confirmed on the bench with a diode test across the module's phase legs — and the repair is component-level: replace the module, check gate-drive circuits (a failed gate driver kills the new module in seconds if missed), verify current-sensing, then load-test. This is exactly the repair we do daily; exchange units are the fast option when production is waiting.
- Do not just fit a new module and ship it — the cause of the original failure (gate drive, dried capacitors raising ripple current, a motor with weak insulation) must be found, or the replacement fails the same way.
Step 3 — If cable and motor test fine and the drive runs on the bench
Then something is commanding or permitting excessive current in operation:
- Mechanical bind: a seizing ballscrew, crashed way cover jamming the slide, or a failed Z counterbalance forces the drive to max current. Check the load meter on the diagnosis screen while jogging slowly — abnormally high load at low speed is mechanical.
- Wrong motor/drive parameters after a repair or swap: motor type parameter mismatch makes commutation wrong and current spikes at the first move. Verify the motor code parameters against the nameplate.
- Encoder/commutation fault: a damaged encoder feeding wrong rotor position makes the drive push current at the wrong angle — often trips 32 together with feedback alarms. See the encoder checks in our zero point guide if position alarms accompany it.
- Phase-to-phase wiring error (U/V/W swapped) after motor or cable replacement — the motor kicks and trips immediately at servo-on.
Diagnosis checklist (in order)
- When does it trip — at servo-on, at first motion, under load, or randomly? Note it before touching anything.
- Cable-off test at the drive → splits drive vs downstream in ten minutes.
- Megger cable and motor separately; check winding balance.
- Inspect the drag chain, motor connector and terminal box for coolant, crush damage and carbonized tracking.
- If all external checks pass → bench-test the drive under load; check mechanics with the load meter; verify motor parameters.
Can you keep running?
No — not with a repeating alarm 32. Unlike a mains-dip alarm 10, overcurrent means something is genuinely wrong in the power path, and repeated resets convert a repairable fault into a destroyed module and sometimes a damaged motor. One clean trip after coolant cleanup and a passing megger test is acceptable; anything more, stop and isolate.