Sinumerik 840D Powerline PLC STOP on Power Cycle: Causes & Fixes

A Sinumerik 840D powerline turning machine boots into PLC STOP after every complete mains power-down, even when a fresh PLC backup has just been loaded. The 7-segment status display reads 6, the SF (System Fault) LED and BF (Bus Fault) LED are both red, and the rotary selector switches on the NCU are confirmed at position 0. A new backup battery has been installed. The standard "delete PLC and re-upload backup" procedure recovers operation temporarily, but the next complete power cycle reproduces the fault. This reference covers the complete diagnostic matrix, root cause analysis, and a verified recovery sequence for the SIMATIC S7-300 PLC integrated inside the Sinumerik 840D powerline NCU.

Problem Description

The 840D powerline was released as the predecessor of the Sinumerik 840D sl (Solution Line) and uses an integrated SIMATIC S7-300 CPU as its PLC. Unlike the sl generation, the powerline NCU combines the NC kernel, the PLC, and the operator-panel interface on a single hardware module with a battery-buffered SRAM and an optional Flash EPROM card for retentive program storage.

The reported symptom chain is:

1. Machine operates normally for an extended period.
2. Without warning, the HMI reports connection to NC/PLC lost or PLC not responding.
3. Operator performs an emergency procedure: delete the PLC project, re-upload the most recent backup, and confirm.
4. Control restarts and resumes production.
5. On the next complete mains power-down (not just an NC reset), the PLC returns in STOP state with the 7-segment display reading 6.
6. The upload procedure must be repeated; otherwise the NC cannot be commissioned.

Because the PLC is the only subsystem failing to retain its user program, the root cause is local to the PLC side of the NCU — not the NC kernel, the drive bus, or the HMI. The combination of a 6 on the 7-segment display, a red SF LED, and a red BF LED narrows the diagnostic search considerably.

Affected Hardware

Identify the exact NCU model before troubleshooting. The classic 840D powerline ships with the following variants:

All variants share a common battery slot on the NCU front, a single multi-pin connector for the operator panel (MCP/HT6), and the rotary selector switches. The X130 port carries MPI/DP, the X120 port carries the PROFIBUS DP external master, and the X122/X132 ports carry the drive bus to the SIMODRIVE 611D or Sinamics S120 modules.

Diagnostic Indicators: 7-Segment, SF, BF

Seven-Segment Status Display

The integrated PLC CPU reports its current state on the 7-segment display located on the NCU front. The codes relevant to the reported fault are:

Display 6 is the most significant indicator. The CPU has either received an MRES command (memory reset) or has detected a non-retentive memory condition at the moment OB1 start was requested. Combined with the operator performing a delete-and-upload each time, the PLC is reporting that its work memory is empty when the start request is evaluated.

SF LED (System Fault)

The SF LED is lit solid red whenever the CPU detects an internal diagnostic event. On the S7-300 CPUs used in 840D powerline the most common SF triggers are:

• Memory card missing or unreadable.
• Flash EPROM write error (write-protect switch active, card full, or card failing).
• Hardware fault in the CPU module.
• Firmware checksum mismatch after a partial update.
• Battery missing while a battery-backed area is configured.
• Time-of-day or bit-memory retentive area larger than the battery can hold.

SF red on cold-start with display 6 is a strong pointer to a memory-retention problem and not a transient bus or slave failure.

BF LED (Bus Fault)

The BF LED on the integrated PLC illuminates red whenever the PROFIBUS DP master (the PLC) cannot communicate with one or more configured slaves. The BF LED can also illuminate when the DP master is not yet parameterised, which is normal during PLC STOP. Distinguish the two cases:

• BF red while PLC is in RUN: active bus fault — investigate cabling, terminators, slave addresses, and slave power.
• BF red while PLC is in STOP: expected because the DP master has not yet started token passing; informational and clears once RUN is established.

In the reported case, the BF is red because the PLC is in STOP and the DP master has not started. Once the PLC is brought to RUN, the BF may clear automatically if the PROFIBUS network is healthy. If it does not clear, the network must be inspected separately.

Rotary Selector Switches S2 / S3 / S4

The 840D powerline NCU module carries two or three rotary selector switches on the front. The exact designation varies by hardware revision. The original post references "S2 & S3" while the operator reply references "S3 and S4 both are in 0" — the ambiguity reflects the NCU variant in use. Treat the following as the canonical mapping:

All PLC-related switches must be in position 0 for normal operation. Position 2 on the PLC switch (MRES) issues a memory reset on the next power-up; if a switch is left in 2, every boot deletes the PLC program from work memory. The operator has confirmed the switches are at 0, which rules this out as the cause. The PLC STOP and display 6 must therefore have another origin.

PLC Memory Architecture and Battery Backup

The integrated S7-300 PLC of the 840D powerline stores its user program and data across three physical areas:

1. Load memory on the Flash EPROM card (PC card or Memory Card). Non-volatile; persists without battery.
2. Work memory (RAM) on the CPU module. Volatile; battery-backed.
3. Retentive data (bit memory, timers, counters, instance DBs marked retentive). A subset of work memory; backed by battery.

The normal boot sequence is:

1. Power ON.
2. CPU performs self-test and reports any diagnostic events on SF LED.
3. If load memory contains a project, the CPU copies it to work memory (or runs in place from the Flash card, depending on CPU generation).
4. CPU enters RUN, executes OB100 (restart), then OB1 (cyclic).
5. If a battery is present and healthy, work memory and the configured retentive data survive the power-down.

When the PLC enters STOP with display 6 after a power-down, one of the following must be true:

• The Flash EPROM card is not seated, has lost contact, or has been write-protected while a write was pending.
• The load memory is empty — the project lives only in work memory, which is volatile.
• The battery is new but the polarity is reversed, the connector is loose, or the contact resistance is too high to keep the RAM refreshed during power-down.
• The retentive area has been mis-configured in STEP 7 to include more data than the battery can hold (rare; normally triggers a different code).
• The CPU module has a hardware fault that prevents reading the Flash card.

Battery and Memory Card Verification

Battery Test

A "new battery" is the most common misleading data point. A lithium thionyl chloride 3.6 V cell that has been on the shelf for several years can measure 3.6 V open-circuit but fail under the 50–200 µA load required to back up a 256 KB SRAM. Always verify the battery under load.

If any of these checks fail, replace the battery with a fresh cell from a sealed package and rerun the boot test. Genuine Siemens part numbers for the 840D powerline backup battery include 6FC5247-0AA18-0AA0 (single cell) and 6FC5247-0AA19-0AA0 (double cell for NCU 580).

Flash EPROM / Memory Card Verification

Once the battery is confirmed good, the next suspect is the Flash EPROM card (also called the PC card, Memory Card, or S7 Memory Card). The card holds the load memory for the PLC project. If the project has been deleted from the card, every power cycle returns the PLC to an empty state.

Verification procedure:

1. Power down the NCU completely; wait 30 s for the power supply to discharge.
2. Open the NCU front cover and remove the memory card.
3. Inspect the card edge connector for oxidation, bent pins, or contamination. Clean with isopropyl alcohol and a non-abrasive swab.
4. Insert the card into a SIMATIC Field PG or an external prommer.
5. Open SIMATIC Manager and select PLC > Memory Card Object > Properties to read the project ID and timestamp.
6. Confirm the project timestamp matches the most recent PLC backup. If the card is empty, the original upload procedure has only written to work memory, not to the card.

Root Cause Matrix

The following table consolidates the most common causes of "PLC does not retain program after power cycle" in a 840D powerline, ranked by frequency in field service reports.

Step-by-Step Resolution

Follow the procedure in order. Do not skip a step; each diagnostic test produces a result that determines the next action.

Prerequisites

• SIMATIC Field PG (or any PG with STEP 7 V5.5 or TIA Portal with the appropriate compatibility package) connected to the NCU X130 (MPI/DP) port.
• Most recent PLC project backup on a USB stick or PG hard drive.
• Most recent NC archive (series start-up file) available.
• Replacement lithium battery from sealed packaging.
• Spare Flash EPROM card formatted to the matching part number.

Step 1 — Capture the Current State

1. Power up the machine, leave the PLC in STOP.
2. Connect the PG to X130, set PG/PC interface to PC Adapter (MPI) or CP 5611 (PROFIBUS).
3. Open SIMATIC Manager and select PLC > Diagnose/Set > Module Information.
4. Record the diagnostic buffer entries, the CPU type, the firmware version, and the order number.
5. Take screenshots of the Diagnostic Buffer tab, the Memory tab, and the Performance Data tab.

The diagnostic buffer is the most informative single tool. Look for entries such as Memory card removed/inserted, Battery voltage low/failed, STOP because of memory reset, and Power off/on. The timestamps on these entries identify whether the failure happened during normal operation or during the operator's upload procedure.

Step 2 — Verify the Battery Under Load

1. Power down the NCU.
2. Remove the battery and measure open-circuit voltage.
3. Apply a 100 µA load through a 36 kΩ resistor and measure voltage after 60 s.
4. If voltage drops below 3.4 V, replace the cell with a sealed fresh unit.
5. Reinstall the battery, confirm the holder detent is engaged, and reconnect.

Step 3 — Verify the Flash EPROM Card

1. Power down the NCU.
2. Remove the memory card and inspect contacts.
3. Insert the card into the PG prommer slot.
4. In SIMATIC Manager, select File > Memory Card > Open and read the project ID.
5. If the card is empty or shows a stale project, the load memory is the problem.

Step 4 — Reload the PLC Project

1. Connect the PG and open the matching STEP 7 project (or open the most recent backup).
2. Set the PG/PC interface to MPI or PROFIBUS and reach the target CPU.
3. Select PLC > Download to Target System. This writes to work memory and brings the CPU to RUN if the switch is in position 0.
4. Once the CPU is in RUN, select PLC > Save to Memory Card (or the toolbar icon). This writes the project to the Flash EPROM card.
5. Confirm the operation completes without error. The diagnostic buffer should show Save to Memory Card completed.

Step 5 — Verify Retention by Power-Cycle Test

1. Power down the machine completely (mains OFF, not just NC reset). Wait 60 s.
2. Power up. Observe the 7-segment display sequence: it should read 0 after approximately 30–60 s and the SF and BF LEDs should go off.
3. From the HMI, open the PLC service menu and confirm the project timestamp matches the latest upload.
4. Repeat the power cycle three times consecutively. Each time, the PLC must come up in RUN with display 0.

Step 6 — Clear the BF LED if Persistent

If the BF LED remains red after the PLC reaches RUN, the PROFIBUS network has a separate fault. Follow the network diagnostic procedure in the next section.

PROFIBUS Network and NCK–PLC Handshake

PROFIBUS Verification

If the BF LED stays red after RUN is established, walk the bus segment by segment:

1. Confirm the bus terminator is enabled at both physical ends of the segment and disabled at every node in between.
2. Measure the bus voltage at each drop: 5 V at the connector with respect to the shield; idle A–B differential near 0 V; alternating during traffic.
3. Confirm the slave addresses match the HW Config assignment.
4. Check the slave diagnostic buffer for missing configuration or parameter errors.
5. Look for duplicate slave addresses; Siemens DP allows only one master per segment, but slaves can have address collisions if not set correctly.

Common slaves on a 840D powerline turning machine include the spindle drive (611D or Sinamics S120), the feed drives, the I/O stations (ET 200S or ET 200M), and the hand-held terminal. Any of these becoming unreachable will light the BF LED.

NCK–PLC Handshake

The 840D NCK periodically reads a status word from the PLC to confirm that the PLC is alive. If the PLC is in STOP at the moment the NCK queries, the NC raises an alarm and refuses to continue. The relevant NC alarms are in the 2000–2999 range and typically read:

• Alarm 2000: PLC sign-of-life monitoring
• Alarm 2100: NCK sign-of-life monitoring
• Alarm 2200: PLC not in RUN

These alarms clear automatically once the PLC is in RUN and the handshake is restored. Do not attempt to clear them with the NCK RESET key; the root cause is the PLC, and the NCK will re-trigger the alarm on the next cycle. Refer to the Sinumerik 840D powerline diagnostics manual for the full alarm list.

Firmware and Sub-Image Consistency

The 840D powerline stores its NC firmware and the PLC runtime on the NCU Compact Flash Card (CF card) or on the hard disk of the PCU 50. The PLC sub-image version is reported in the HMI under Startup > System Info > PLC Firmware. The NC sub-image version is reported under Startup > System Info > NC Firmware. A mismatch in which the PLC sub-image is older than the version expected by the NC can cause a cold-start failure where the PLC never leaves STOP.

Match the PLC sub-image version to the NC bundle version as documented in the Sinumerik 840D powerline commissioning manual. The combination is encoded in the boot order number, for example NCK 75.10 + PLC 75.10. If the sub-images are not paired, reflash both to the matching pair using the SINUMERIK boot disk and the procedure described in the Siemens Industry Online Support documentation library.

Verification Matrix and Related Cases

Verification Matrix

After every corrective action, run the following matrix. Each test has a pass criterion; if any test fails, return to the corresponding root cause.

Related Fault Scenarios

The same symptom (PLC STOP after power cycle) can be produced by several other conditions. Review them before closing the case.

Defective Memory Card. Flash EPROM cards have a finite write-cycle endurance (typically 100 000 cycles). A card that has been written to excessively can develop unreadable sectors. The CPU detects this and refuses to load the project, falling back to STOP. Confirm by cloning the card to a known-good unit and repeating the power-cycle test.

NCK Retentive Data Corruption. If the NCK has marked its own non-volatile area as corrupt, the NC may request a cold restart that includes a PLC reset. This is controlled by the machine data family MD20150 $MC_G_CODE_DEFAULT_VALUE and MD11240 $MN_PROFIBUS_SDB_NUMBER. Inspect the machine data for non-default values that may have been changed accidentally.

Improper Shutdown Sequence. Powering down the machine while the PLC is in the middle of a Flash card write can corrupt the card. The CPU detects the corruption on next boot and refuses to load. Always shut down the NC properly via the HMI Shutdown sequence before opening the mains breaker. A UPS on the NCU line is recommended for production environments.

EMI or Ground Loop. An unstable 24 V supply to the NCU or a ground loop on the PROFIBUS shield can cause the CPU to detect a memory parity error and enter STOP. Check the 24 V rail for ripple and the shield for single-point grounding.

Preventive Measures and Field Notes

Preventive Measures

• Replace the backup battery on a fixed schedule: every 3 years for production machines, every 5 years for low-utilisation machines. Mark the replacement date on the cell.
• Always use the HMI Save to Memory Card command, never the simple Download, when deploying a new PLC project.
• Maintain a 6-monthly backup of the PLC project, the NC archive, and the drive parameters in a versioned repository.
• Install a UPS on the NCU 24 V supply to avoid corruption from sudden mains loss.
• Verify the rotary switch detents during every major service.
• Train operators to perform the proper upload procedure from the documented service manual, not from improvised memory.

Field Notes

Two field patterns are common in 840D powerline service. The first is the "intermittent battery" — a battery that passes a quick multimeter check but fails to back the RAM because of a cracked internal tab or a holder that has lost spring tension. Always buy cells from a sealed package and prefer the original Siemens part number. The second is the "stale project on the card" — operators often keep an old backup on the card because it is easier than flashing a new one, and a power-down event reveals the staleness when the NC asks for the live project. The 7-segment 6, the red SF, and the red BF together are unambiguous: the PLC has nothing to run with at the moment it is asked to start. Reading the diagnostic buffer and confirming whether the Flash EPROM card holds a current project is the single highest-value diagnostic step.