Resolving SIMADYN D 'C' Fault on PM6 CPU2 Excitation Racks

Resolving SIMADYN D 'C' Fault on PM6 CPU2 Excitation Racks

The Siemens SIMADYN D control platform is widely deployed in Chinese power-generation plants as the digital regulator for static excitation systems. Field experience across dozens of hydro and thermal units shows a recurring, often winter-clustered 'C' fault on the PM6 CPU2 of the secondary channel, frequently cascading to the SS4 communication modules and the SS52 Profibus gateway on the companion CS7 communication card. The fault may clear on power-cycle and dust cleaning, or it may persist until the affected module is replaced. This reference consolidates the diagnostic procedure—D7-SYS / CFC online error capture, HEX/Debug Monitor extraction, and the offline fallback when CPU serial access is dead—together with the environmental and hardware mitigations that drive fault frequency down to near zero in properly conditioned cabinets.

1. SIMADYN D Architecture in Dual-Channel Excitation

SIMADYN D is a multiprocessor, modular real-time controller from the Siemens SIMADYN family. In excitation service it is configured as a hot-standby, dual-channel regulator: each channel runs independently, exchanges status over a dedicated SS4 cross-link, and shares a process database with an OP operator panel and an external Profibus monitor.

1.1 Typical Rack Population (24-Slot Subrack)

1.2 Communication Paths That the 'C' Fault Touches

• PM6 X01 (local serial) — direct PG/PC port for DUST1 cable and online CFC. This is the only path that survives the 'C' fault when CS7/SS4 stacks drop.
• SS4 ↔ SS4 cross-link — inter-channel synchronization and warm-standby arbitration. Loss of this link is the most common visible symptom of a PM6 'C' fault.
• SS4 → OP — operator panel and alarm annunciator. Goes blank or freezes.
• SS52 → Profibus DP — external monitor, plant DCS, and event/SOE recorder. Drops off-line; DCS sees a station-level communication failure.

The full system base manual with parameter letter codes (H/L for read-write, c/d for read-only) and a/b/c/d parameter-number ranges is published by Siemens as the SIMADYN D System Manual (dyn_system_e.pdf) on the Siemens Industry Online Support portal; refer to its parameter-coding section whenever you need to interpret a tag returned by D7-SYS in the error panel.

2. Defining the 'C' Fault

The letter 'C' is the PM6 firmware class code for Communication-class exceptions raised by the run-time system when a configured mailbox, sampling, or inter-processor handshake fails. In the SIMADYN D exception taxonomy:

A 'C' fault is therefore not a single root cause but a class of symptoms, all of which surface in the PM6 error fields and exception buffer. The diagnostic task is to read the underlying error code (commonly a 16-bit hex value such as 0x6014) and trace it back to a specific FB, signal, or physical interface.

3. Root Cause Analysis

Treat every 'C' fault as a four-axis investigation. The cause is almost always a combination of two or more axes.

3.1 Hardware Axis — Contact and Contamination

• Backplane connector oxidation. PM6 and CS7 use gold-plated DIN-41612 connectors; dust combined with low humidity creates a non-conductive film that breaks high-speed mailbox strobes.
• SS4 / SS52 socket seating. Daughter cards can walk out of their CS7 host by 0.5–1 mm during thermal cycling, producing intermittent contact loss that the PM6 reports as a 'C' exception.
• PM6 EPROM / NV-RAM retention. A weak NV-RAM battery corrupts the configuration signature, leading to a B-class on boot but a C-class on mailbox re-initialization when the CPU reaches run state.

3.2 Environmental Axis — Temperature and Humidity

• Winter clustering. Indoor static-charge spikes (often <30% RH) and cabinet heater cycling cause both ESD events and connector-film buildup.
• Summer mitigation. Higher humidity (40–60% RH) maintains a thin conductive film on connectors, paradoxically reducing open-circuit events even though it increases leakage risk in adjacent analog inputs.
• Dust loading. Coal-handling and hydro-turbine halls deposit conductive particulates (carbon, graphite) that short adjacent backplane pins during high-humidity days and break contact during dry days.

3.3 Configuration Axis — Sampling and Mailbox

• Mismatched sampling time (T0 / TA). If the receiving task runs faster than the producer, the mailbox queue under-runs and PM6 raises a C-class timeout.
• Cross-channel task skew. The two PM6 CPUs must be resynchronized; if the SS4 link is slow, the master keeps logging C-exceptions on the slave mailbox.
• Profibus slave address collision. Adding a new SS52 slave without re-addressing can produce C-class on the SS52 transmit side only.

3.4 Firmware Axis — IBS and D7-SYS Versions

Older IBS releases (pre-5.x) and earlier D7-SYS builds do not expose the full error-panel field set; upgrading the engineering tool is often the difference between an ambiguous 'C' and a precise 0x6xxx fault code with FB reference.

4. Diagnostic Procedure A — D7-SYS / CFC Online Capture

This is the primary procedure and is applicable when the PG/PC can still reach PM6 X01 after the 'C' fault is latched.

4.1 Prerequisites

• D7-SYS (or Step7 Classic with the SIMADYN D option pack) installed on the PG/PC.
• DUST1 programming cable (RS-232 to SIMADYN D local service port) or USB-to-DUST1 adapter for newer PGs.
• Direct connection to PM6 X01 — do not chain through CS7 / SS4 when a 'C' fault is active; the higher-level stacks may already be down.
• CPU2 chart file (*.sdf or *.dat depending on the Struc G/L vs CFC toolchain) opened locally.

4.2 Step-by-Step

1. Power the rack normally; do not reset the 'C' fault — the error fields are volatile and are cleared on the next CPU restart.
2. Connect the DUST1 cable between the PG/PC COM port (or USB-COM bridge) and PM6 X01.
3. Start D7-SYS, open the CPU2 program, and double-click into any chart to put the editor into online mode.
4. From the menu, choose CPU → Module Information (in older CFC builds: Target System → Module Information).
5. In the dialog that opens, switch to the Error fields tab. Record every entry. Typical content:

   Error field 0 :  0x6014   FB=AV44  Sig=QV_FIELD
   Error field 1 :  0x0000
   Error field 2 :  0x0000

6. Switch to the Exception buffer tab. Scroll through the last 16 exceptions; note the time stamps and the FB chain that triggered the cascade.
7. Open CFC online Help → Contents → Index; type the 4-hex-digit code (e.g., 6014) to retrieve the fault description and the recommended remedy.
8. Document the chain in your maintenance log: error code → FB name → I/O tag → physical channel.

5. Diagnostic Procedure B — HEX/Debug Monitor via IBS

When D7-SYS is unavailable, or when the chart file is in the legacy Struc G/L format and cannot be opened by D7-SYS, fall back to the HEX/Debug Monitor utility inside IBS.

5.1 Launch Sequence

1. Start IBS and verify COM-port settings (baud, parity, handshaking) match the rack; 19200 8E1 is the most common SIMADYN D default.
2. From the menu bar: Activities → IBS Program.
3. From the menu bar: Diagnostics → Start Debug Monitor; when prompted, enter the CPU number (typically 2 for the secondary channel where the 'C' fault is latched).
4. Close the IBS program cleanly: Service → Exit IBS Program. The Debug Monitor keeps running in the background.
5. Return to the top-level menu: Activities → HEX/Debug Monitor.

5.2 Reading the Error Panels

1. At the HEX/Debug Monitor prompt, type ? or dh and press Enter. IBS writes a fresh log file to the IBS working directory as HEX.txt.
2. Open HEX.txt in any text editor and search for the literal string Diagnostics of the error panels. The line directly below it contains the suggested command, e.g.:

   Diagnostics of the error panels: jm 0x80000F20

3. Type the suggested command at the HEX/Debug Monitor prompt:

   jm 0x80000F20

   The monitor prints the contents of the error panels. Capture the FB references and the 16-bit error codes, exactly as in Procedure A.
4. Cross-reference the codes against the SIMADYN D System Manual error-code appendix. The same 0x6xxx codes apply.

6. Diagnostic Procedure C — When CPU Communication Is Dead

If the 'C' fault is so severe that PM6 X01 also stops responding, the engineer cannot read error fields in-circuit. The following sequence is the field-proven fallback used by Siemens commissioning engineers in China.

6.1 Physical Checks First

1. Verify 24 VDC on the rack power supply; an under-voltage of >5% on a heavily loaded rack can cause mailbox strobes to fail before CPU is fully down.
2. Remove and reseat every SS4, SS52, and the PM6 in question. Inspect the DIN-41612 pins for oxidation, bending, or fiber contamination. Clean with a Siemens-approved contact cleaner (e.g., Cramolin, Kontakt 60) and a lint-free swab; do not use eraser rubbers on gold-plated contacts.
3. Inspect the backplane under oblique light; even a hairline fiber across two pins can hold a C-class communication line low.

6.2 Card-Swap Triage

6.3 Off-Line Error Capture

When the 'C' fault has latched but PM6 will still boot to its loader, the loader menu exposes the exception buffer over the same X01 port. Cycle rack power, hold the loader break, and read the exception buffer with the DUMP command from a terminal program (HyperTerminal, PuTTY in serial mode, or Tera Term at 19200 8E1).

6.4 Struc G/L vs CFC Toolchain

Many Chinese excitation installations were originally programmed in Struc G/L, the predecessor of CFC. Struc G/L programs (*.sdf) cannot be directly opened in modern D7-SYS, but the error codes produced are identical because they are generated by the PM6 firmware, not by the engineering tool. The HEX/Debug Monitor route (Procedure B) is therefore the recommended path for any Struc G/L site, with the caveat that the online FB reference name will read as a Struc G/L block name (e.g., AI8, AV44, BIC) rather than a CFC FB name.

7. Error Code Quick Reference

This table is derived from field reports on the China excitation fleet; the authoritative description of any code is in the CFC online help of the matching D7-SYS version and in the error-code appendix of the SIMADYN D System Manual.

8. Hardware Mitigation and Environmental Controls

The single largest reduction in 'C' fault frequency is achieved by attacking the environmental axis while the diagnostic axis is being run.

8.1 Cabinet Environmental Targets

8.2 Mechanical Hardening

• Add card locks to PM6 and CS7. The standard plastic extractors loosen with thermal cycling; Siemens card-lock kits or third-party latches are inexpensive and eliminate 80% of reseat-induced 'C' faults.
• Replace all SS4 and SS52 retaining screws. Torque to 0.4–0.5 Nm; over-torque warps the host CS7 PCB.
• Apply a thin bead of dielectric grease on the Profibus D-sub connector at the SS52 to prevent oxidation in humid plant environments. Do not apply grease on the backplane DIN-41612 connectors.

8.3 PM6 NV-RAM and Firmware

• Replace the PM6 NV-RAM battery every 3 years, regardless of measured voltage. A weak battery at the moment of a 'C' fault will mask the true cause by also corrupting configuration.
• Maintain a firmware register: record the PM6 firmware version (visible in CPU → Module Information) and the IBS / D7-SYS version that last successfully connected. Mismatches are a leading source of undecodable 0x6F00-class exceptions.

8.4 Preventive Maintenance Schedule

9. Verification After Repair

After clearing the 'C' fault, whether by reseat, environmental correction, or card replacement, run a four-step verification before returning the unit to service.

1. Watchdog test. From D7-SYS, force a CPU stop / run transition. Both PM6 channels must come back to RUN within 5 s; any longer indicates a lingering C-class.
2. Cross-link check. Confirm that the SS4 cross-link LED is steady green on both racks and that the partner CPU is visible in the CFC online browser.
3. Profibus check. With the SS52, read the slave diagnostic buffer; all slaves must report 0x00 (no diag) and the cyclic exchange must be at the configured baud rate (typically 1.5 Mbps).
4. OP check. Force a process variable change in the chart and verify the OP reflects it within one OP scan. This proves the SS4 OP link is fully alive, not merely electrically present.

10. Frequently Asked Questions