Kawasaki C42 Faults Need Independent Axis and Pendant Tests

On this 1999 UX150 with a C42F-A001 controller, the blank pendant, intermittent JT5 current alarm, and later JT3 encoder alarms must be traced as separate fault paths; restoring the screen does not prove an axis is healthy.

Baseline and backup on the C42F-A001

The robot first powered its motors and moved individual axes, then developed a dark, unresponsive pendant. Later diagnostics showed intermittent JT5 current-overload reports, JT5 motor-overload and position-envelope errors during motion, and—after the robot was moved and damage was found on X3—JT3 encoder faults. These changes call for a fresh fault log at each stage, not a single assumed root cause.

1. Record the state before clearing anything. Capture the exact active error text, affected joint, time, controller and pendant indicators, whether motor power was requested, and what changed immediately beforehand. The decisive distinction is whether an alarm exists at power-up or appears only during a commanded move.
2. Retrieve the controller history. Connect with an available Kawasaki terminal application, such as KCWin32, KCWinTCP, or KRTerm, and run errlog. Save the full output before repeated resets; the log can show whether an event preceded or followed a cable change, motion attempt, or controller restart.
3. Save the controller configuration to the computer. In the terminal application, use the controller's save syntax, for example SAVE backup to the application's default folder, or SAVE c:backup. Create any destination folder first; confirm the file exists on the computer and can be reopened before considering initialization.
4. Defer a hardware initialization. A complete wipe can remove the current configuration and network settings. Use it only as a later software-isolation test after a verified backup and with a working recovery connection, including RS232 if Ethernet settings may be lost.

Gate: Proceed only when the complete error history and a restorable configuration backup are available. Do not use a cleared alarm or motor-power lamp as the baseline.

Controller access and error-history capture

The teach pendant is not a generic VGA monitor. A VGA-shaped connector does not identify its electrical signals, and this controller setup was described as having no standard VGA output. Use the controller's documented communications ports instead of attaching a monitor to an unknown connector.

1. Try Ethernet when the controller's address is known. Use a crossover cable for a direct computer-to-controller link; use a standard patch cable through a switch or router. Check the CPU-board Ethernet connection and look for a controller IP label. If the address is unknown, a packet capture on the directly connected network can help identify the controller's traffic.
2. Use RS232 as an alternative, not as a guaranteed route. Some controllers have the programming port disabled. A USB-to-RS232 adapter does not necessarily cross transmit and receive lines; use a null-modem connection or a three-wire lead wired 2>3, 3>2, and 5>5 as described for this connection.
3. Separate recurring log entries by meaning. The history included JT5 -1401 amp-over-current events, Interbus -1215 status errors, FTP autosave -2226 failures, and -586 timeouts. A repeated network or autosave entry is not proof of a JT5 hardware fault; correlate it with the active alarm and the time of a motion command.

Gate: Continue when you can retrieve the error history reliably and distinguish the active axis alarm from unrelated communication or autosave entries.

Pendant backlight, inverter, and harness

The pendant's black screen initially prevented reading alarms. In this case, the small inverter-board fuse had no continuity; after it was bridged with fuse wire, the pendant lit. That result identifies the backlight power path as a fault area, but bridging a protective fuse is not a repair and can expose the board or pendant to further damage.

1. Check for pendant power before condemning the display. Listen for the pendant's piezo beep and inspect the lower edge in a dark room for backlight. A beep with no visible backlight points toward the inverter or CCFL backlight; no beep calls for checking the supply and harness to the pendant as well.
2. Inspect the fuse and board without bypassing protection. The inverter fuse was identified as a small 0.5 A SMD fuse. Check continuity with power removed and replace it only with the board-specified type and rating. If the replacement fuse opens again, investigate the inverter circuitry before applying power again.
3. Check the backlight output with suitable equipment. The reported 0.6 VAC DVM reading on the inverter wiring did not establish that the inverter was operating correctly. An oscilloscope and appropriately rated probe are more suitable for evaluating inverter output; the CCFL output can present a shock hazard. A failed tube can also leave the LCD dark even when the inverter is powered.
4. Inspect the complete pendant cable path. Check the pendant connector for pushed-back pins, then inspect and flex the cable while checking continuity. Static continuity at rest does not rule out a fault that changes with cable position. Check the controller-side harness path, including the referenced XHP-CN4 to X1 connection, and the pendant's internal board connections.

Gate: Before returning to axis diagnosis, confirm the pendant has stable logic power and a functioning backlight, or establish a reliable terminal connection that can show alarms. Do not treat a lit LCD alone as proof that the pendant is fully responsive.

Safety circuit and unused Interbus configuration

Do not use a bridged limit-switch wire as a substitute for verifying the safety circuit. The arm limit switches are in series with the motor-power safety circuit; incorrect wiring can prevent motor power and may blow the small F1 or F3 fuses on the 1HP board. The initial cut-wire repair in this installation was based on continuity checks against another switch, so verify the actual circuit against the correct wiring documentation before further movement.

1. Identify each external safety connection. The X15 connection was identified with the field emergency-stop and safety-fence circuit; associated blue wires normally terminate at the controller terminal strip, where jumper wiring may be present. Do not reconnect loose wires or alter jumpers by color or appearance. Confirm the intended external safety configuration from the controller documentation.
2. Keep the Ethernet cable entry distinct from X15. The unused connector opening identified on this controller was X14; it was being used to pass an Ethernet cable. Protect that opening against dust and moisture without disturbing the safety wiring.
3. Resolve Interbus only if the network is unused. The logged Interbus status error and the blinking green RDY/red FAIL indicators were attributed to an Interbus card without an active network. If the installation does not use Interbus, remove the card or disable/unallocate the fieldbus only using the controller's configuration procedure. If the network is used, diagnose its bus and connections instead.

Gate: Continue only after the safety circuit matches the intended wiring and motor-power interlocks operate correctly. Treat an Interbus status alarm separately from the JT5 and JT3 axis faults.

JT5 current-sensing and power-block path

The repeated JT5 -1401 report was described as the power-block current-sensing path detecting current above 144% of the correct specification. It appeared at times before a motion command, sometimes cleared with umbilicals disconnected, and later returned during testing. That pattern requires checking both the current-feedback electronics and the wiring; it does not by itself identify a bad motor.

1. Remove power before reseating board connections. Inspect CN10, its pins, and its retaining lever; clean with an appropriate electronics cleaner if needed, then reconnect it fully. Inspect CN1 and XM3SA as well. Reseating CN10 temporarily changed the behavior in this case, but the alarm later returned, so that was not proof of a lasting repair.
2. Check the controller supplies at documented test points. The reported 1GB measurements included approximately -12.04 VDC, +11.91 VDC, and 4.95 VDC; after small AVR adjustments, the readings were reported as -12.10 VDC, +12.10 VDC, and 5.05 VDC. These are readings from this unit, not universal targets. Use the troubleshooting manual for test-point identity and allowable ranges before adjusting any potentiometer.
3. Check whether the alarm depends on the umbilicals. With the robot stationary and using the documented isolation procedure, disconnect the umbilicals and observe whether the specific JT5 current alarm remains. Other disconnection alarms are expected and are not the comparison. If the JT5 alarm persists with the external cable path removed, focus next on the controller-side sensor, 1GB connection, or power block. If it clears, reconnect and isolate the cable/connector path.

Gate: Do not proceed to motion while -1401 is active or returns immediately. A loose connection can explain intermittency, but recurrence after reseating means the source remains unresolved.

JT5 umbilical and motor isolation

A DVM check that finds no short to ground is not enough to clear an old, flexed umbilical. In this case, pressing the control-box door and moving the cable coincided with the JT5 alarm returning; removing the robot-side umbilical did not consistently clear it. A damaged or compressed section near a connector can fail only under flex or tension.

1. Inspect X5/X5A and cable strain points. With power removed, inspect both connector housings for foreign material, damaged pins or sockets, poor clasp engagement, and signs of cable crushing near the gland. Check that internal power-block connectors are seated and retained. A damaged housing with intact latches is not automatically the fault, but cable movement that changes the alarm is important evidence.
2. Test each disconnected cable independently. Disconnect the umbilical at both ends before insulation testing so test voltage cannot reach controller or robot electronics. Use equipment and a test method suitable for the cable; a low-voltage DVM continuity check may miss leakage between conductors or to the shield. The X5 cable later passed a megger test in this investigation, but the separate encoder-related X3 issue still required its own diagnosis.
3. Compare the motor and axis paths only under a controlled procedure. Inspect JT5 motor stator resistance and insulation against the manual's criteria, and check for damaged robot-side wiring. If a qualified technician uses the documented motor-connector substitution test with a neighboring axis, label every connection, power down before changing it, and do not move the robot during the test. If the fault follows the motor, investigate the motor; if it remains on JT5, investigate the harness, sensor, or power-block channel.

Gate: Resume no motion until cable tests and connector checks pass, the specific current alarm stays cleared, and the motor path meets the manual's electrical criteria. Do not use visual inspection alone to clear an intermittent fault.

JT5 brake and mechanical checks

Separate the startup current alarm from errors that appeared during motion. The later 1500 JT5 motor-overload and 1504 position-envelope errors occurred while the robot was being jogged; reported grumbling or whirring noises make brake release and mechanical resistance important checks. A brake click proves that something actuated, not that the joint released mechanically.

1. Check the brake-supply breaker. Inspect the resettable breaker identified as F5 for the brake supply and follow the controller procedure to verify its state. Do not repeatedly reset a breaker that trips; diagnose the brake circuit and supply first.
2. Test the brake release mechanically. With the robot stable and the relevant area clear, hold the appropriate brake-release control and assess whether the joint can be moved by hand. A click without movement does not confirm release. The JT4–JT6 joints can feel resistant because of gearing, so use the controller's maintenance procedure rather than judging from sound alone.
3. Jog one axis at a time in joint interpolation. Use joint mode rather than base or tool mode to isolate a joint. If the pendant reports “no program,” verify that the control is in manual joint-jog operation rather than trying to run a program. Stop if the joint binds, makes abnormal noise, or throws an overload or position error. Support JT2 and JT3 against gravity before any brake-release test on those axes.

Gate: Do not command further motion on JT5 until the brake releases, movement is mechanically free, and the startup current fault is absent. A recurring 1500 or 1504 during a move calls for stopping and resolving the mechanical or feedback fault, not repeated jogging.

JT3 encoder signal path

After relocation, the controller showed JT3 1517 “cannot read encoder value” and 1553 “encoder response error.” The X3 cable had visible damage and was re-insulated. Swapping JT3 and JT4 encoders at the robot-side 1FG connection did not move the reported fault from JT3, which points toward the JT3 channel path but does not, alone, prove a failed 1GB board.

1. Inspect X3/X3A end to end. Check the damaged area, strain relief, controller and robot connectors, and each pin/socket for displacement or damage. The X3 cable was later shortened and tested, but the JT3 alarms remained; a physical repair or continuity result is not a complete signal-integrity test.
2. Check the controller-side encoder connection. Inspect and reseat CN2 on the 1GB board with power removed. The encoder signals from X3 were identified as passing directly to CN2. Verify the correct wired pins against the controller pinout; the reported continuity check found 28 of 39 CN2 pins soldered to a cable, so do not expect every connector cavity to have a conductor.
3. Use the swap result as a branch decision. If a correctly performed encoder swap moves the alarm to the other joint, investigate the encoder that followed. If the alarm stays on JT3, inspect the X3/X3A conductors, connectors, CN2, and JT3 channel on the board. A DVM continuity pass cannot prove encoder power, signal quality, or response under operation; use the specified test procedure and readings in the troubleshooting manual.
4. Keep encoder faults active until repaired. Do not bypass the encoder alarm to force jogging. The controller needs valid position feedback to control the axis; investigate the signal path or board channel before asking the robot to move.

Gate: Proceed only after the JT3 encoder errors clear and remain absent through a stationary restart and the approved encoder checks. Replace 1GB only after the X3 path, connectors, and board input have been verified.

End-to-end return-to-motion verification

Do not use a cramped, unstable setup for a full robot test. The robot was initially moved only slightly on a pallet, and the later work found additional faults after it was relocated. Confirm the machine's installation, anchoring, and available clearance before commissioning motion across its working envelope.

1. Confirm prerequisites with power off. Verify the intended limit-switch and external safety wiring, secure controller/robot connections, protective covers, and a verified configuration backup. Leave unused fieldbus hardware disabled or removed only if the system design confirms it is unused.
2. Start without commanding motion. Confirm the pendant remains powered and responsive, record the active error list, and verify that JT3 encoder faults and the JT5 current alarm do not return at startup. A backlight that sleeps after inactivity may need the pendant's normal wake action; distinguish that from a complete loss of pendant power.
3. Test each joint individually in joint interpolation. With clearance and safety controls verified, move each axis through a limited range at controlled speed. Confirm the selected axis moves as commanded, the brake releases, no abnormal noise or unexpected resistance occurs, and no new current, overload, position-envelope, or encoder error appears. Stop immediately if any condition fails.
4. Expand the test only after each joint passes. Continue the approved range and operational checks in a clear work area. Recheck the error log after movement and confirm safety stops and limit behavior through their documented test procedures. Save a new configuration backup only after the controller is stable.

Final gate: Return the robot to service only when the pendant, safety chain, fieldbus configuration, and every axis pass their checks with no recurring alarms or abnormal sounds; capture the final error log and confirm the saved configuration is readable.

Kawasaki C42 troubleshooting questions