Simatic 545 Fatal Error 0107: Troubleshooting CPU Hardware Failure

The Simatic 545 (a member of the legacy TI Series 505 PLC family originally manufactured by Texas Instruments and continued by Siemens) reports Fatal Error 0107 when its on-board diagnostic firmware detects a non-recoverable CPU subsystem fault. In the field, the most common symptom is a controller that boots, runs, and enters Fatal state after approximately 20-40 minutes of operation, with both the CPU GOOD and RUN LEDs extinguishing simultaneously. Because the fault is hardware-resident, power-cycling or reloading the program does not eliminate the underlying defect; the controller re-enters fatal state on the next warm-up cycle. This article documents how to confirm the root cause, interpret the operational status log, evaluate replacement options, and migrate the program to a third-party compatible CPU such as the CTI 2500-C200 from Control Technology Inc.

1. Problem Description

Field reports across multiple Simatic 545 installations describe a consistent failure signature:

• The controller powers up normally, completes self-test, and begins executing the user program.
• Both the CPU GOOD LED (indicating successful self-test) and the RUN LED (indicating active program execution) remain illuminated for a period ranging from 15 to 45 minutes.
• After this interval, both LEDs extinguish and the controller drops into the Fatal state. Any connected I/O is de-energized.
• A manual reset, a Mode-to-Program-to-Run transition, or a full program re-download allows the controller to restart, but the failure recurs after a similar warm-up period.
• The Operational Status page in the programming software records Fatal Error code 0107 (hex) immediately before the controller halts.

This behavior is distinct from a software bug, a corrupt ladder program, or a scan-time overflow: those conditions typically fail immediately or on the first scan that triggers the bad logic. A time-dependent fault that requires thermal warm-up of the silicon is characteristic of a hardware defect on the CPU module.

2. Affected Hardware and Error Code Definitions

The Simatic 545 product line includes several CPU catalog numbers. The Fatal Error 0107 condition has been observed on the entire legacy catalog, including:

The TI505/Simatic 545 firmware encodes fatal errors as 16-bit hexadecimal values stored in the Operational Status log. The most common hardware-related codes are summarized below. Always verify against the official Siemens Industry Online Support documentation for the firmware revision loaded in the controller.

Fatal Error 0107 is treated by Siemens as a hardware-class fault. There is no published software workaround; the firmware deliberately halts execution to prevent uncontrolled outputs.

3. LED Status Reference

Before opening the programming software, observe the front-panel LEDs on the CPU. The combination of LED states is itself a diagnostic input:

The reported field condition (both CPU GOOD and RUN extinguishing) followed by recovery after a reset, combined with the Operational Status entry FATAL 0107, is the canonical signature of a CPU module that must be replaced.

4. Root Cause Analysis

The TI505/Simatic 545 family is built around discrete CMOS memory devices, a proprietary CPU ASIC, and an EPROM-based firmware socket. After three decades of continuous service, several age-related failure modes converge on the 0107 code:

1. Memory device degradation. The CMOS static-RAM used for user program and data storage exhibits single-bit and multi-bit failures once cell retention degrades. The 545 firmware periodically exercises a memory-diagnostic routine; a warm silicon die produces an intermittent parity or read error that trips fatal 0107.
2. EPROM / firmware socket contact loss. Pin-1 of legacy PLCC or DIP firmware sockets oxidizes, producing CRC errors on read. The firmware loads successfully when cold but fails after thermal expansion.
3. CPU ASIC timing drift. The CPU gate array relies on an external crystal or oscillator. Aging of the crystal, or drift in the support capacitors, pushes the clock outside the ASIC's tolerance window. The watchdog timer then flags a fatal.
4. Failed backup battery / capacitor on the memory-retention circuit. A weak lithium cell or tantalum cap causes the RAM refresh logic to flag a non-volatile memory error during the post-power-on test.
5. Solder-joint fatigue on through-hole pins. Backplane-edge connector pins or memory socket tails develop micro-cracks that open under thermal cycling.

None of the above are field-repairable. The component vendors that supplied the proprietary ASIC and firmware mask sets have long since discontinued the parts. Siemens no longer offers repair service on Simatic 545 CPUs, and most authorized repair shops have removed the platform from their supported list. The only practical remediation is module replacement.

5. Diagnostic Procedure

Before declaring the CPU failed, complete the following checks to rule out environmental causes that mimic hardware faults.

5.1 Capture the Operational Status Log

Connect the programming PC to the 545's serial port (RS-232C, typically COM1 on the CPU front panel, 9600 baud, even parity, 8 data bits, 1 stop bit in the default configuration). Open TISOFT, SoftShop, or the 545 Workstation Software and go online:

1. From the menu, select Controller > Status or navigate to the Operational Status page.
2. Record every entry in the log, including timestamp, fatal code, and scan counter.
3. Export or print-screen the Operational Status display for the maintenance record.

Even after the CPU halts, the log persists across power cycles. The most recent fatal entry should read FATAL 0107.

5.2 Verify Power Supply Rails

Measure the backplane voltages at the CPU socket with the chassis energized:

A +5 VDC rail sagging below 4.85 VDC under load is a known cause of RAM read errors that surface as fatal 0107. Replace or rebuild the power supply before condemning the CPU.

5.3 Inspect the Chassis Environment

• Verify ambient temperature inside the enclosure is below 40 °C (104 °F). The 545 is rated to 60 °C operating, but derating of the CPU ASIC begins above 55 °C.
• Confirm that the cooling fan (if equipped) is operating.
• Inspect for accumulated dust or conductive debris on the CPU PCB; clean with ESD-safe vacuum and isopropyl alcohol if needed.
• Tighten the CPU retaining screws; a partially seated module can produce intermittent backplane contact.

5.4 Attempt Cold-Start Verification

1. Power down the chassis.
2. Remove all I/O base modules, leaving only the CPU and its terminating base.
3. Power up and observe the time-to-fatal. If the controller still drops to fatal 0107 within 30 minutes with no I/O attached, the fault is isolated to the CPU module.

This isolation test is the definitive confirmation that the CPU itself is defective.

6. Replacement Options

Because the Simatic 545 is end-of-life and unrepairable, the maintenance strategy is to replace the CPU with a compatible unit. Two paths are available.

6.1 New-Obsolete / Refurbished Original

Surplus resellers and used-equipment brokers occasionally stock tested Simatic 545 CPUs. When sourcing used hardware:

• Demand a 30-day operational warranty.
• Request a recent burn-in test report (minimum 4 hours continuous run).
• Avoid modules that show capacitor bulging or PCB rework.

6.2 Third-Party Compatible CPU (Recommended)

Control Technology Inc. (CTI) manufactures the 2500-C200 as a drop-in replacement for the Simatic 545 CPU. The 2500-C200 is designed for the same TI505 backplane, supports the same instruction set, and accepts the same program image exported from TISOFT or SoftShop. Key specifications:

The 2500-C200 provides more memory and faster scan times than the original 545, while remaining pin- and program-compatible. Most existing programs load without translation.

7. Migration Procedure (545 to CTI 2500-C200)

Follow this sequence when migrating from a failed Simatic 545 CPU to a CTI 2500-C200.

7.1 Prerequisites

• Verified good copy of the program archive (.PC1, .TXT, or .LDR) on the programming PC.
• CTI 2500-C200 module with matching I/O base footprint.
• TISOFT version 6.x or later, or CTI Workshop, installed and licensed.
• RS-232C serial cable (DB-9 female to DB-9 female, straight-through).
• Backed-up copies of all V-memory and K-memory constants.

7.2 Program Export and Conversion

1. Connect to the (still-functional, if possible) 545 and go online.
2. Select File > Upload Program > To File. Save the archive to the PC.
3. Note any special-function or drum instructions; verify the 2500-C200 supports the entire instruction set used. CTI documents a small subset of obsolete 545 opcodes that may require manual rewriting.
4. Re-compile the program in TISOFT or CTI Workshop targeting the 2500-C200.

7.3 Physical Replacement

1. Power down the chassis. Lock out / tag out the disconnect.
2. Record the DIP-switch and jumper settings of the failed CPU (mode, port configuration, base address).
3. Remove the failed CPU by loosening the retaining screws and lifting straight up.
4. Inspect the backplane connector pins for bent or recessed contacts; straighten with a needle probe if necessary.
5. Set the DIP switches on the 2500-C200 to match the removed CPU's configuration.
6. Insert the 2500-C200 into the same slot, tighten the retaining screws to the specified torque (typically 0.5 N·m).

7.4 Download and Commission

1. Power up the chassis and connect the programming PC to the new CPU's serial port.
2. Go online and clear any residual faults. Cycle the CPU to PROGRAM mode.
3. Download the program archive to the 2500-C200.
4. Restore V-memory and K-memory constants from the backup.
5. Switch the CPU to RUN mode and verify that the CPU GOOD and RUN LEDs remain steady.
6. Walk the I/O: force each output briefly, confirm each input updates in the live status display.
7. Run the machine through one full production cycle before releasing the line to operations.

8. Verification

After replacement, log the following to confirm the fault has been resolved:

• Continuous RUN time without fatal entry: minimum 8 hours, target 72 hours.
• Operational Status log shows no fatal errors since commissioning.
• CPU GOOD and RUN LEDs remain illuminated throughout the verification window.
• Scan time is within the historical baseline documented before the failure.
• All I/O responds correctly to stimulation.

If a new fatal 0107 appears on the 2500-C200, the root cause is environmental (power supply, temperature, backplane) rather than the CPU module itself, and the diagnostic loop in Section 5 must be revisited.

9. Preventive Measures for Remaining TI505 / Simatic 545 Fleet

Once a single CPU has failed in a fleet, the remaining units are likely on a similar aging trajectory. Recommended preventive actions:

1. Battery replacement program. Replace the on-board lithium memory-backup battery on every 545 CPU every 5 years, regardless of measured voltage.
2. Thermal monitoring. Install a panel-mounted temperature indicator and alarm at 50 °C inside the enclosure.
3. Power-supply rebuild. Capacitors in the 545-era PSUs have a 15-20 year service life; rebuild the PSU on a 15-year cycle.
4. Spare CPU stock. Maintain at least one spare 2500-C200 per site. Lead time on surplus parts is unpredictable.
5. Program archive discipline. Store the latest approved program archive on a managed file server with date-stamped revisions.

10. Field-Proven Caveats

• Do not attempt to repair the proprietary CPU ASIC. The component is not available on the open market, and substitution attempts destroy the module.
• Fatal 0107 is not cleared by firmware reload. The diagnostic routine runs in mask ROM; reloading user memory does not affect it.
• A small population of 545 CPUs with fatal 0107 can be temporarily revived by re-seating the firmware EPROM and resetting the memory-retention battery. This is a stop-gap only; failure will recur.
• CTI 2500-C200 modules ship with default Ethernet settings disabled; if the application relied on a special-function module (e.g., 2572-A or NIM), verify the replacement module is configured before commissioning.
• Some 545 programs use the DRUM, BIT_PACK, or MOVE_LONG instructions that have minor semantic differences in the 2500-C200. Validate with a controlled test before going live.

11. Related Siemens Documentation

For verification of error codes, replacement part status, and migration paths, consult the official Siemens Industry Online Support portal:

• Siemens Industry Online Support - product lifecycle, manuals, firmware downloads for SIMATIC legacy controllers.
• SIMATIC TI505 / Series 505 Legacy Information - product discontinuation notices and recommended replacement modules.
• TI Series 505 Programming Reference - instruction set reference for migrating to CTI 2500 series.
• Control Technology Inc. (CTI) - manufacturer of the 2500-C200 replacement CPU.

FAQ