Troubleshooting Yaskawa XRC Alarm 1105 (System Error Servo) and the Cascading 4100 / 4101 / 4110 Faults on UP130 Robots

The Yaskawa Motoman XRC controller, when paired with the UP130 manipulator, will latch alarm 1105 SYSTEM ERROR (SERVO) and a cluster of secondary alarms — 4110 SHOCK SENSOR ACTION, 4100 OVERRUN IN ROBOT AXIS, and 4101 OVERRUN IN EXTERNAL AXIS — whenever the controller detects a fault in the servo power chain that prevents the NCM (Motor Contactor) relay from closing. The alarms are not independent; the 1105 is the root fault, and the 4110/4100/4101 messages are downstream symptoms. This guide explains the cause-effect relationship, the components that must be inspected (WRCA card, zero-volt fuse, base cable, shock-sensor wiring), and the field-proven sequence to clear the fault without a parameter reset. For the canonical Motoman description of code 1105 on the successor controller, see the Yaskawa Motoman Knowledge Base article on YRC1000 Alarm 1105; the XRC and YRC1000 share the same error number and the same root-cause tree.

1. Controller Family and Alarm Architecture

The XRC (eXpert Robot Controller) is the generation that preceded the DX200 and the current YRC1000. On the XRC, alarm reporting is split into four parallel classes:

Whenever the 1105 servo-system fault is active, the controller inhibits the NCM coil drive. With NCM open, the contactor cannot pass three-phase bus power to the servo amplifiers, so the absolute encoder back-up, the brake release lines, and the shock-sensor comparator all read abnormal states. That is why 4110, 4100, and 4101 appear simultaneously: the encoder back-up capacitors have discharged, the axes are not actually moving, and the shock sensor is reporting bit 0 set because its reference voltage is missing. Treat 1105 as the parent; treat 4110/4100/4101 as children. Resetting the children without clearing 1105 will latch them again within one second.

2. Alarm 1105 — SYSTEM ERROR (SERVO)

The Motoman knowledge base entry for the YRC1000 lists alarm 1105 as a "SYSTEM ERROR (SERVO)" raised by the YCP01 / YAS2.0 servo-control CPU when a self-diagnostic, encoder, or amplifier handshake fails. The XRC reports the same four-digit code through its JANCD-XCP01 (XCP02 in later builds) backplane. Common sub-data values reported in the alarm detail field are:

In the field report that drives this article the sub-data was (32807) which in hex is 0x8027. The high byte 0x80 places the failure in the backplane / WRCA category, and the low byte 0x27 indicates a servo-control / power-supply handshake problem. That sub-data is the single most important clue: it tells you to look at WRCA, the 24 VDC logic supply, and the absolute-encoder battery, not at the motors themselves.

3. Cascading Alarms 4110 / 4100 / 4101

Once 1105 has latched and the NCM contactor has dropped out, the controller can no longer confirm the manipulator is at a known position. Three downstream alarms are raised almost immediately:

3.1 Alarm 4110 SHOCK SENSOR ACTION (0000_0000_0000_0001)

The 16-bit pattern is a bit-map of the shock-sensor inputs and the collision-detection software trips. The least-significant bit (bit 0) reflects the reference-voltage OK signal from the SRDA board. When the NCM is open, the +15 VDC reference for the shock-sensor comparators is missing, so bit 0 reports "false" and the controller interprets this as a collision. Wiring the shock-sensor cable does not clear the alarm; the reference must be present at the SRDA input. See the manipulator cable group drawing for the SRDA connector (typically CNR20 / CNR21 on the JZCR-XRC05C harness).

3.2 Alarm 4100 OVERRUN IN ROBOT AXIS

The position-error counter (PEC) for one or more of the six robot axes has overflowed because the absolute position was lost when the encoder back-up battery discharged. The controller falls back to incremental counts and cannot guarantee the axis is inside its software stroke limit, so it raises 4100 as a defensive measure. This is a false overrun in 95 % of the cases linked to 1105 — the axis is not actually out of envelope; the controller simply has no absolute reference.

3.3 Alarm 4101 OVERRUN IN EXTERNAL AXIS

Same root cause as 4100, but on the external (base) axis — typically the travel rail or positioner axis declared in parameter AXEXT. If no external axis is configured, the alarm is still raised because the controller leaves the 4101 decoder enabled by default. Disabling the external-axis overrun alarm requires setting AXEXT.OVRCHK = 0 in the controlled-start parameter set, but that change cannot be made while 1105 is latched.

4. The NCM Contactor — Why It Is the Keystone

NCM is the main motor contactor, mounted on the back panel of the XRC cabinet. Its 24 VDC coil is driven by the safety chain: emergency stop, servo-on request, gate, and the WRCA "servo ready" relay. When 1105 is active, the WRCA drops its "servo ready" contact, breaking the NCM coil path. With the contactor open:

• The three-phase bus to the SVM (Servo Voltage Module) is interrupted.
• The DC bus collapses, removing encoder power.
• The absolute-encoder back-up capacitors discharge within ~30 s; the lithium battery then takes over.
• The shock-sensor +15 VDC reference disappears → 4110.
• The PEC counters reset to zero → 4100 / 4101.

Field experience on the UP130 and UP20 platforms shows that you cannot close the NCM manually without a service-port teach pendant, but you can confirm it is being commanded: measure 24 VDC at terminal block TB1-NCM-A1/A2 while pressing the dead-man switch on the programming pendant. If the voltage is present and the contactor does not pull in, the contactor itself is mechanically damaged. If the voltage is absent, the safety chain is open — almost always because WRCA has not released its "servo ready" relay.

5. WRCA Card — Diagnostic Procedure

The WRCA (Wiring Robot Controller Axis) card, Motoman part number HW0170630-A (later revisions HW0170630-B / -C), sits in slot 3 of the JANCD-XCP01 backplane. It performs three functions relevant to this fault:

1. Multiplexes the absolute-encoder serial data from the manipulator cable to the servo CPU.
2. Hosts the "servo ready" relay K1, whose contact is in the NCM coil path.
3. Provides the +5 VDC and +15 VDC references for the shock-sensor comparator on the SRDA card.

To diagnose the WRCA on the bench (controller de-energised, L1/L2/L3 locked out):

1. Remove the card and inspect the 96-pin DIN backplane connector for bent pins or cold solder joints. Re-seat firmly.
2. Check the on-board fuse F1 (5 A, 32 V, blade type) — this is the "zero-volt fuse" referenced in the original field report. It protects the +24 VDC logic rail that feeds the NCM coil driver. An open F1 is the single most common cause of the symptom cluster described here.
3. Measure across F1 with a DMM on continuity. A healthy fuse reads < 0.1 Ω; an open fuse reads OL. Replace only with the same rating — a 7.5 A fuse will allow downstream damage to the SRDA card.
4. Inspect the K1 relay footprint. Cold solder joints at pins 12 / 14 are common on cards that have seen thermal cycling. Reflow if necessary.
5. Reinstall, reapply power, and observe the front-panel LEDs: LED1 (POWER), LED2 (CPU OK), LED3 (SERVO READY) should all illuminate within 8 s of power-on. If LED3 remains dark, the card has failed its self-test.

6. Zero-Volt Reference and the Base Cable

The XRC uses a "zero-volt" floating common (0 V) on the manipulator side, returned to the cabinet through the base cable. A break in the base cable's 0 V conductor has two effects:

• The shock-sensor comparator on the SRDA card loses its return path, drifting its output high and asserting bit 0 of the 4110 pattern.
• The encoder back-up battery return is opened, draining the lithium cell in 24 – 48 h and eventually latching 1105 with sub-code 0x0012 (encoder battery alarm, axis 2 or 3).

To check the base cable:

1. Power down and lock out the main disconnect.
2. Open the connector panel on the back of the manipulator (typically JZCR-XRC05C or JZCR-XRC06C depending on build year).
3. With the cable still attached to the cabinet, ring out pin 1 (0 V) and pin 2 (24 VDC) end-to-end. Resistance should be < 1 Ω.
4. Inspect the strain-relief grommet at the base axis J1 rotation point; this is the most common flex-fatigue location on UP-series robots that have been in service for > 8 years.
5. Check for green-corrosion on the connector pins, especially if the cell has been washed down or operated in a humid environment.

If the cable is suspect, swap with a known-good unit before replacing any card. Replacing the WRCA on a controller that still has a bad base cable will immediately re-latch 1105 with a different sub-code and create the impression of a "bad new card".

7. Step-by-Step Diagnostic Procedure

The following procedure is the field-proven order of operations. Do not skip steps; the cascade nature of the alarms means a mis-ordered fix can mask the true root cause.

7.1 Prerequisites

• Programming pendant (Motoman part HW1170800-A) with the controller in Teach mode.
• Digital multimeter with continuity and diode-test functions.
• Service manual HW0170630 (WRCA) and HW1170200 (XRC main).
• ESD wrist strap.
• Replacement 5 A blade fuse (Motoman HW9170078-A) and a known-good lithium encoder battery (HW0380870-A).

7.2 Procedure

1. From the pendant, navigate to MENU → 4 ALARM → 1 HISTORY. Press F5 DETAIL on alarm 1105 and record the 5-digit sub-code. Power down the controller.
2. Lock out / tag out the main disconnect. Wait 5 minutes for the DC bus capacitors to bleed below 5 VDC (verify with DMM on the SVM test points).
3. Open the cabinet and visually inspect the WRCA in slot 3. Look for burned traces, bulging capacitors, or discoloured connector pins.
4. Pull the WRCA and check F1 with a DMM. If open, replace with an identical 5 A / 32 V blade fuse.
5. Reinstall the WRCA. Apply control power only (do not enable servo). Confirm LED1 and LED2 illuminate; LED3 may stay dark until the safety chain is closed.
6. With the teach pendant in Teach mode and the dead-man switch held at the middle position, measure 24 VDC at TB1-NCM-A1 / A2. If voltage is present and the contactor closes, the safety chain is healthy.
7. Close the E-stop, gate, and dead-man circuits. Power up fully. Observe the pendant — 1105 should be cleared if F1 was the only failure.
8. If 1105 re-latches within 30 s, the sub-code will tell you the next layer to inspect:
   • 0x0010 – 0x001F → check / replace the encoder back-up battery.
   • 0x0020 – 0x002F → check SVM DC bus and amplifier IPM modules.
   • 0x80xx → reseat or replace WRCA.
9. After 1105 is clear, the cascading 4100/4101/4110 alarms can be reset with MENU → 4 ALARM → 3 RESET. If they re-appear, the base cable 0 V conductor is open — return to step 6.

8. Verification

After a successful repair, run the following verification sequence before returning the cell to production:

1. Power the controller on cold. Confirm the pendant shows SERVO ON READY within 10 s and no alarms are active.
2. Enable servo and jog each of the six robot axes through 10 % of its travel at 5 % speed. Watch the position-error counter on the STATUS → SERVO screen; values should remain < 50 counts on every axis.
3. If an external axis is configured, jog it through the same test. Verify the 4101 alarm does not latch.
4. Run the built-in SHOCK SENSOR TEST program (job SHKTEST.JBI on most UP-series cells). The 4110 bit-map should read 0000_0000_0000_0000 at rest and 0000_0000_0000_0001 only when the manipulator is physically contacted.
5. Execute three full motion programs in Play mode at production speed. Monitor the alarm history for 30 minutes — any re-latch of 1105 indicates an intermittent base-cable or battery issue.

9. Common Pitfalls and Field Notes

10. When to Call a Motoman Service Technician

If the WRCA F1 fuse blows a second time within 30 days, the fault is downstream — typically a shorted SRDA card, a failed SVM IGBT, or a crushed base-cable conductor. At that point, parameter changes and card swaps will not help, and the controller should be left de-energised until a certified service technician with the FS100 / XRC service tool can perform insulation testing and backplane diagnostics. The "we can't change or reset anything" symptom in the field report is consistent with the controller's controlled-start protection: parameters that could mask a safety-chain fault are read-only while 1105 is latched. This is by design.

FAQ