Resolving Siemens 840D Error 120202 NC/PLC Connection Failure

Problem Overview

The Siemens SINUMERIK 840D control family uses a distributed architecture where the NCU (Numerical Control Unit), the PCU (Panel Control Unit / PC), the MCP (Machine Control Panel) and a SIMATIC S7-300 PLC communicate over MPI/PROFIBUS. When the HMI/PCU cannot reach the NCK and PLC software on the NCU, the system raises alarm 120202 "Wait for connection to NC/PLC" and the HMI displays NCU: ??? instead of a recognized node.

This alarm is generated by the HMI software on the PCU when the configured NCU address (default 0) does not respond on the MPI/PROFIBUS segment within the timeout. It is therefore a symptom alarm, not a fault of the NCK or PLC user program; the root cause is always in the physical link, the NCU itself, the PCMCIA card, or the loaded software image.

On a system based on the legacy 6FC5247-0AA02-1AA0 PCI/ISA box (used as the PCU 50/70 host platform with HMI-Advanced), the diagnostic LEDs on the NCU front panel and the seven-segment display are the primary indicators during recovery.

Hardware Identification and LED Meaning

Before opening the cabinet, verify which generation of NCU you have. The connector layout, the PROFIBUS connector (X101) and the S3/S4 rotary switches differ between the NCU 561.x, NCU 571.x, NCU 572.x, NCU 573.x and the later NX lines. The 840D PCI/ISA box discussed here is normally paired with an NCU 561.4 / 571 / 572 in the 6FC5 2xx series.

The reported state "both green LEDs on top are ON, S3/S4 at 0, 6 on display, orange flashing LED missing, all MCP LEDs blinking" matches the boot state where the NCU has finished its self-test (display 6 = boot loader finished, ready to load from PCMCIA or flash) but the HMI is still trying to handshake.

Root Cause Tree for Alarm 120202

Alarm 120202 can be triggered by any of the following layers. Walk the list top-down; do not skip steps because the same symptom appears for every layer.

1. PROFIBUS / MPI physical layer - broken cable, missing terminator, wrong connector pinout, address conflict, EMC damage.
2. NCU power and thermal health - 24 V supply sag, NCU fan failed, dust-clogged heatsink causing the NCK to throttle or restart silently.
3. PCMCIA card / boot media - corrupted flash file system, ejected card while writing, aged CompactFlash card, card not fully seated.
4. NCK / PLC software image - partial or incompatible archive, version mismatch between NCK software and PLC hardware (S7-300 CPU type), corrupt .arc file.
5. HMI/PCU side - PCU 50/70 software crash, registry corruption, service HMI watchdog disabled, IP/MPI gateway configuration on the PCU.
6. MCP / operator panel bus - although not a direct cause of 120202, a shorted MCP can drag the PROFIBUS segment down on the NCU side.

Diagnostic Procedure

Perform the checks in the order shown. Each step verifies one layer and isolates the next.

1. Verify PROFIBUS cabling. Inspect X101 on the NCU and X4 (MPI/DP) on the PCU 50/70. With the system powered off, measure each conductor for continuity and check for shorts to shield. Confirm PROFIBUS connectors are wired 1-to-1 (pin 3 to pin 3, pin 8 to pin 8) and that termination is ON only at the two physical ends of the segment. The PCU side and the NCU side are typically both ends of a short segment and must both be terminated.
2. Confirm NCU addresses. The NCU must be at MPI address 0 (default) and baud rate 187.5 kbps (or 1.5 Mbps for NCU 573 and newer). The PCU HMI-Advanced must reference address 0 in mmc.ini / regie.ini. Mismatched baud rate is a common cause of "120202 Wait for connection" without any other alarm.
3. Check NCU fan and ventilation. The 6FC5247-0AA02-1AA0 PCI/ISA box is convection-cooled inside the NCU. Withdraw the NCU from its mounting and inspect the fan. If the fan is sluggish, the heatsink is dust-loaded, or the unit has been in service more than 5 years, thermal shutdown is a probable cause. The clue in the field report is erratic axis enable and tool-change dropouts before the link failure - the NCK was likely restarting silently.
4. Inspect the PCMCIA card. Power the NCU OFF, remove the PCMCIA card (FEPROM/CompactFlash), inspect the contacts, reseat the card firmly, and wait 10-15 minutes with the card out to allow any inrush protection on the cardholder to discharge. Reinsert, power up, and observe the 7-segment display: it should briefly show 6 and then the run state.
5. Re-load from a known-good archive. If a PG/PC with the matching cardware and a backup of the NC/PLC archive (*.arc) is available, copy the archive to the PCMCIA card with the correct cardware tools and reload. Always reload NC first, then PLC, then drive data.
6. Use the rotary switches S3 / S4 for a controlled reset. See the dedicated procedure below.

Standard Reset Procedure with S3 and S4

The NCU front panel has two rotary switches used to force a general reset of the NCK and PLC areas. Use this only when an archive cannot be loaded or when the boot is stuck.

1. Set S3 to position 1 and S4 to position 0 to stop the PLC (CPU 314 / 315 / 317 in the S7-300 rack).
2. Toggle S4 between position 2 and 3 at least three times within 3 seconds to perform an NCK general reset. The 7-segment display will count down and end at 6.
3. Return S4 to position 0 (NCK normal) and S3 to position 0 (PLC run).
4. Power-cycle the NCU, then the PCU, in that order. The orange MPI LED on the NCU should resume flashing once the PCU has finished booting HMI-Advanced.

Layer-Specific Solutions

2.1 PROFIBUS / MPI Recovery

• Replace the X101 to X4 cable if any conductor is above 0.3 ohm or if the shield is broken.
• Verify the two-segment termination: the ON switch is present on the connectors at the NCU and at the PCU; all intermediate MCP / drive connectors must have termination OFF.
• Check for address duplication. Standard address plan: NCU = 0, PCU = 1, MCP = 6 (or 7 for additional panel), each drive = 6 + axis index. A duplicate 0 will produce exactly the 120202 symptom without other alarms.

2.2 NCU Cooling and Power

• Replace the NCU internal fan (6FC5247-0AA02-1AA0 box). The OEM Siemens fan kit is recommended; aftermarket equivalents often have a lower MTBF and produce EMC noise that disturbs the MPI transceiver.
• Clean the heatsink fins and the front air intake. Avoid blowing dust into the PCMCIA slot.
• Measure the 24 V supply at the NCU terminals under load. Sag below 21 V during disk activity on the PCU will reboot the NCK and bring up 120202.

2.3 PCMCIA Card and Software Image

• Use only Siemens-approved CompactFlash or FEPROM cards. Industrial-grade cards with at least 100,000 write cycles are recommended for the 840D's persistent log area.
• When reloading from a backup, the card must be formatted with the correct cardware version matching the NCU type. A 561.x card will not boot a 573.x NCU and vice versa; the symptom is the 7-segment staying on 6 indefinitely.
• Verify CRC of the archive by comparing the file size and timestamp with the last successful backup before reloading.

2.4 PCU 50/70 HMI Software

• If the PCU shows signs of Windows corruption (BSOD, slow boot, missing shortcuts in the HMI shell), the HMI may not start the Regie service that handshakes with the NCU. Restore the HMI image with the Siemens backup/restore tool and re-apply the most recent service pack for HMI-Advanced matching the NCK version.
• Verify the registry entry HKEY_LOCAL_MACHINE\SOFTWARE\Siemens\Sinumerik\HMI-Advanced\connection lists the NCU at the correct MPI node and baud rate.

Verification

After applying the fix, run the following checks before returning the machine to production:

1. On the PCU HMI, open the Diagnostics area and confirm the NCU is shown as active (not ???). Alarm 120202 should clear within 30 seconds of a clean boot.
2. Toggle the NCK mode from AUTO to JOG and back. The PLC should not stop. The 7-segment display on the NCU should be blank (or show 0 / current mode) and the orange MPI LED should flash at a regular rate.
3. Run a reference-point approach on each axis to verify there are no residual Wait: Axis enable missing messages. If the original symptom returns, the underlying cause is almost always thermal; replace the NCU fan even if it appears to spin.
4. Cycle a tool change through the magazine to confirm the tool-change interlock is healthy. The pre-failure symptom of "tool change stops without alarm" is a strong indicator that the NCK was rebooting during the macro.
5. Monitor the NCK temperature and CPU load from the HMI service menu for at least 30 minutes. A steady rise above 65 degC points to a heatsink or fan issue.

Related Alarms and Cross-Reference

Spare Parts and Catalog Numbers

Field-Proven Checklist

• Power supply at NCU measured at > 22 V DC under load.
• NCU fan runs free, no bearing noise, no dust clog.
• PCMCIA card fully seated, contacts clean, archive CRC matches last known-good backup.
• S3 = 0, S4 = 0, no NCK or PLC held in stop.
• PROFIBUS cable X101 to PCU X4: continuity OK, shield grounded at one end, termination only at NCU and PCU.
• MPI addresses unique across the segment (0 / 1 / 6 / 6+n).
• HMI-Advanced version matches NCK version (compatibility list in the Commissioning Manual).
• No active 130001 / 8080 / 120201 alarms at the same time.

Safety and Commissioning Caveats

• Always reload NC, then PLC, then drive configuration in that order. Reversing the order leaves the PLC looking for symbols the NCK has not yet declared.
• Keep at least two generations of archive on separate PCMCIA cards. Card wear is the most common cause of intermittent boot failures on this generation.
• Document the S3/S4 procedure on a laminated card inside the cabinet; during a night shift, the only available operator is often the one who has never seen a 120202 before.

When to Escalate

If the steps above do not clear 120202 within 30 minutes, escalate to Siemens Service with the following minimum information: NCU MLFB and FW version, PCU 50/70 image version, S7-300 CPU MLFB and FW, last successful archive date and CRC, PROFIBUS segment topology, and a photograph of the NCU 7-segment display. The most likely unresolved causes at that point are NCU mainboard failure, PLC CPU hardware fault, or a corrupted NCK option set on the PCMCIA card - all of which require bench-level repair or replacement.

FAQ