Sinumerik 840D sl PLC Stopped: Recovering the S7-300 317F CPU With the S4 IBN Master Reset

Problem Overview

On a Sinumerik 840D sl system equipped with an S7-300 317F (fail-safe) PLC, the controller occasionally drops into a hard STOP condition where:

• The HMI (Sinumerik Operate / HMI Pro) loses every connection to the NCK and PLC and shows only the connection fault overlay.
• The NCU 7-segment display shows the Ready plus Stop LED combination (Stop LED flashing, no application cycle running).
• The PLC is unreachable from STEP 7 / TIA Portal over MPI, PROFIBUS, PROFINET, or the NCU routing channel.
• Even a direct laptop connection to the NCU service interface (X127) or to the PLC's PROFINET port (X150) does not bring the CPU back online.
• The PLC does respond to ping 192.168.214.1 (the default NCU/PLC PROFINET address on the 840D sl), confirming that the network interface and the embedded Ethernet switch are still alive, but the CPU's user program is not executing.
• A full power off/on of the NCU line-up (NCU, PLC, HMI, drive line-up) does not clear the condition.
• The standard PLC reset / restart pushbutton path from the HMI (HMI softkey PLC → Reset PLC) is acknowledged but has no effect.

When the controller reaches this state, the only field-proven recovery documented in this guide is a master reset of the 317F CPU using the IBN (commissioning) switch S4 on the front of the PLC, followed by download of the last good PLC.arc backup. The procedure below is taken from the Siemens 840D sl commissioning manual and the SIMATIC S7-300 installation manuals and is reproduced here in field-engineer order.

Affected Hardware and Firmware

The 317F CPU is the standard integrated PLC on 840D sl. It is a true SIMATIC controller with its own firmware (e.g., V3.3 for the 2FK14 / 2EK14 hardware), its own retain/load memory, and a separate IBN switch S4 on the front door. The PLC and NCK are tightly coupled over PROFINET but they are independent CPUs and can fail independently.

System Topology and Default IP Addressing

The 840D sl system bus on a 317F-based line-up is a single PROFINET segment with fixed default addresses. Knowing these is required to determine whether a "ping works but no STEP 7 connection" symptom is network or CPU related.

Network reachability test: from a service laptop on the same subnet, run ping 192.168.214.1 (NCU) and ping 192.168.214.241 (PLC port X1). If only the NCU replies, the PLC's PROFINET stack is up but the CPU is not running user code. That is the signature of a CPU in STOP with a reachable PROFINET interface — a very common state on the 317F when the F-I/O watchdog trips or after a cold restart with a corrupted project.

Diagnostic Decision Tree

Before reaching for the IBN switch, walk through the following checks. They isolate the failure to a specific layer and prevent unnecessary master resets.

1. NCU 7-segment display — record the number and the dot. A continuous 6 with a dot means the NCU is in boot; a steady 0 or blank means ready; a flashing value indicates a fault. Reference the NCU diagnostics manual to interpret the number.
2. PLC LED pattern — on a 317F the front panel shows SF (red, group fault), BF (red, bus fault), DC5V (green), FRCE (yellow, force active), RUN (green, CPU running), STOP (yellow, CPU in stop). In the failure described above, STOP is solid yellow, RUN is off, and SF or BF may be lit.
3. ICR / SIMATIC diagnostics buffer — if you can get online, the diagnostic buffer is the source of truth. If you cannot get online, this is exactly the case where the master reset applies.
4. Ping test — from a service laptop on the system bus, ping the NCU and the PLC PN ports. The fact that ping works but STEP 7 cannot open a connection is the textbook symptom of a CPU in STOP with the PROFINET stack still alive.
5. NCU service interface (X127) and DHCP — connect a laptop to NCU port X127 with DHCP enabled. The NCU typically serves 169.254.200.200 as a default. Use the Sinumerik sc command set or VNC to access the NCU shell and run show to read the PLC state.
6. VNC / WinSCP into the NCU — on the NCU side, the plc_arc directory under /card or /oem holds the last PLC.arc archive. If the file is timestamped and intact, the master reset procedure below is safe.
7. Try online over X150 (PN port 1) — with a direct crossover cable from the laptop to the PLC's PROFINET port, assign the laptop a static IP in the 192.168.214.x range (e.g. 192.168.214.99 mask 255.255.255.0) and try to go online with STEP 7 over TCP/IP. If the CPU is truly in STOP and not crashed, this still fails — which is the trigger to escalate to the IBN switch reset.

Root Cause Analysis

The 840D sl PLC is a stock SIMATIC S7-300 CPU; it follows standard S7-300 stop behaviour. The most common root causes that produce the "ping works, no online, restart does not help" signature are:

A standard HMI-side Reset PLC and even a power-cycle will not recover a corrupted project or a password-locked CPU. The master reset clears the load memory and brings the CPU to a known-factory condition so that a fresh PLC.arc download can be performed.

Step-by-Step Master Reset Using IBN Switch S4

This is the field-proven recovery used on the 317F-2 PN/DP when the CPU is unreachable from STEP 7, no project will download, and the HMI cannot restart the PLC. The procedure is taken from the SIMATIC S7-300 CPU 31xC and CPU 31x installation manual and is reproduced in operator order.

1. Confirm that a PLC.arc backup exists. On the HMI, go to Start-up → System data → Series start-up or on the NCU file system check /card/user/system/ for a recent PLC.arc (or older PLC_xx.arc series-archive set). Without this file, do not proceed with the master reset.
   Tip: copy the entire /card tree to a USB stick before continuing, so the current state of the NCU can be examined later if the recovery is unsuccessful.
2. Open the PLC front door on the 317F. Locate the four-position IBN switch S4. The positions are 0 (RUN), 1 (RUN-P), 2 (STOP), and 3 (MRES — master reset). The same switch on the 317F is also sometimes labelled with the F-variant's mode selector symbols, but the positions are identical.
3. Set S4 to position "2" (STOP). The yellow PS (Power Supply / Status) LED on the PLC will light up. Wait for at least one second; the CPU is now definitively in STOP regardless of the previous state.
4. Set S4 to position "3" (MRES) and hold for 3 seconds. The PS LED will go out briefly and then come back on. The 3-second hold arms the master-reset state machine. Releasing before 3 seconds cancels the procedure.
5. Within 3 seconds of the PS LED coming back on, perform the toggle sequence:
   1. Move S4 from "3" to "2".
   2. Move S4 from "2" back to "3".
   3. Move S4 from "3" back to "2".
   Each transition must happen within 3 seconds of the previous one; the entire triple-toggle must happen within 3 seconds of the PS LED coming back on. The PS LED will begin blinking at 2 Hz, indicating that the master reset is in progress, and will then change to a steady on state when the reset has completed.
6. Wait for the PS LED to be steady on (not flashing). The CPU is now in factory condition. The load memory is empty and the diagnostic buffer is reset.
7. Set S4 to position "0" (RUN). At this point the CPU will start in a clean state and will sit in STOP because there is no project loaded.

Post-Reset: Reconnect and Download the Project

1. Confirm laptop connectivity. Connect a service laptop directly to the PLC's PROFINET port X1 (X150) with a static IP in the 192.168.214.x range, for example 192.168.214.99 mask 255.255.255.0. Verify that ping 192.168.214.241 now replies promptly. On a freshly-master-reset CPU, the PN port will still respond to ICMP because the PROFINET stack is up; STEP 7 will now be able to open a TCP connection because the CPU is no longer in a partial-startup deadlock.
2. Open STEP 7 (Classic) or TIA Portal with the project that matches the machine, or open the saved PLC.arc from the HMI. In STEP 7, the project file is typically *.s7p; in TIA Portal it is a .ap15 / .ap16 project archive.
3. Go online via TCP/IP to the PLC's IP 192.168.214.241. The master-reset CPU has default empty access protection, so you should be prompted for no password. If the CPU does still prompt, enter the project password that matches the PLC.arc archive — do not skip the prompt by entering a wrong password repeatedly or the CPU may temporarily lock.
4. Perform a full download (not delta). In STEP 7 this is PLC → Download to Target System → Hardware and Software (entire project). The download will reset the CPU, write load memory, and restart the CPU.
5. Wait for the PLC to reach RUN. The RUN LED will come on steady green; STOP LED will be off; SF and BF should be off. If SF is lit, read the diagnostic buffer — typically a PROFIsafe address mismatch that must be re-acknowledged in the Safety Administration Editor.
6. Re-acknowledge the F-signature (F-CPU only). In STEP 7 open Safety → F-Configuration → F-Signature and accept the compiled F-signature. The F-CPU will then enter productive F-mode. If the machine is a non-F variant (317F-2 PN/DP non-F) this step is not required.
7. Verify NCK-PLC hand-shake on the HMI. On the HMI, open Diagnostics → Bus Diagnostics and confirm that the NCK, PLC, and drive objects all show active. The HMI overlay connection fault should clear on the next polling cycle.

Verification Checklist

Troubleshooting Matrix

Preventive Practices

• Series commissioning archive (Reihen-Inbetriebnahme). On 840D sl, configure a series start-up in the HMI's start-up area so that a numbered set of PLC_xx.arc files is kept automatically with each NCU backup. This is the single most valuable preventive measure for any PLC-related recovery on this platform.
• Mandatory F-I/O acknowledgement procedure. Document the F-signature workflow for the specific machine. A master reset always invalidates the F-signature, and the recovery should be rehearsed in cold-state so the on-call engineer does not need to debug Safety Administration under production pressure.
• Diagnostic buffer export to USB. After every PLC STOP event, export the diagnostic buffer to USB before resetting. The buffer is the only way to determine whether the root cause is a F-peripheral fault, OB85, or corrupted project — all of which require different corrective actions after the master reset.
• UPS on the 24 V supply. Many PLC STOP events on 317F-based 840D sl systems are caused by brown-out on the 24 V supply feeding remote F-I/O stations. A small UPS on the 24 V distribution prevents the watchdog from triggering.

Related Siemens Documentation

• SIMATIC S7-300 CPU 31xC and CPU 31x: Installation manual — contains the canonical S4 MRES sequence and timing.
• SINUMERIK 840D sl Operator Components and Networking Manual — default IP plan and PROFINET device names for the NCU/PLC/TCU.
• SINUMERIK 840D sl Commissioning Manual, NC part — NCU boot / 7-segment diagnostics.
• SINUMERIK 840D sl Commissioning Manual, PLC part — PLC.arc backup and series start-up procedure.
• SIMATIC S7 F-CPU Programming and Operating Manual — F-signature workflow after a master reset.

FAQ