S7-1200 SCADA Loss Detection: Watchdog Tag and Auto WiFi Reset

1. Problem Overview

Distributed remote sites that rely on Wi-Fi backhaul for SCADA telemetry are exposed to a recurring failure mode: the S7-1200 station controller remains fully operational, but the wireless link to the central SCADA server drops. Because the PLC has no way to know that the higher-level system is no longer requesting data, it never raises a station-level alarm, and field crews must travel to each site to power-cycle the Wi-Fi antenna.

This reference implements a one-way watchdog handshake between WinCC V7.5 SP2 and the SIMATIC S7-1200 station. The SCADA server cyclically toggles a heartbeat tag; the PLC monitors the tag for change. If the tag fails to toggle inside the configured time window, the PLC raises a digital output that drives a relay coil to remove and re-apply power to the Wi-Fi antenna. After a cooldown period the watchdog arms itself again, providing a self-healing remote station.

2. Prerequisites

3. Architecture: One-Way Watchdog Handshake

The handshake is intentionally one-way: only the SCADA writes, only the PLC reads. This makes the design tolerant of asymmetric failures (e.g., PLC can still see the antenna, antenna cannot reach the server).

4. WinCC V7.5 SP2 - Cyclic Heartbeat Action

Create a 16-bit unsigned integer tag named SCADA_HB mapped to a writable DB word on the S7-1200 (in this example, DB100.DBW0). In the WinCC Tag Management, confirm the tag is configured with the correct S7-1200 connection and address; WinCC V7.5 SP2 uses the SIMATIC S7-1200/1500 channel unit.

In WinCC Explorer, open Global Scripts → C-Actions (or VB-Actions), right-click and select New → Action. Configure the trigger as a cyclic trigger with a 20-second interval. Paste the following VB-Action into the editor:

' WinCC V7.5 SP2 - SCADA Heartbeat Toggle
' Trigger: cyclic, 20 s
Dim objTag
Set objTag = HMIRuntime.Tags("SCADA_HB")
objTag.Read
Dim curVal, newVal
curVal = CLng(objTag.Value)
newVal = curVal + 1
If newVal > 65535 Then
    newVal = 0
End If
objTag.Write newVal
Set objTag = Nothing

Save the action and verify it appears under the cyclic trigger list. Confirm the action executes by reading the tag in the WinCC variable diagnostics; the value must increment once per 20 seconds. Reference the WinCC V7.5 scripting entry at Siemens Support entry 109792622 for the full HMIRuntime object model.

4.1 Optional: VB Action for Alarm Logging

Add a second cyclic VB action (1 s) that monitors a second tag (PLC_CommOK) sent from the PLC back to WinCC. When that bit clears, the action can fire an alarm using HMIRuntime.Trace for diagnostic logging:

' Alarm raise when PLC reports comm NOT OK for 30 s
Dim objAlarm
Set objAlarm = HMIRuntime.Alarms
Dim objTag
Set objTag = HMIRuntime.Tags("PLC_CommOK")
objTag.Read
If CLng(objTag.Value) = 0 Then
    objAlarm.CreateAlarm 1, 1, "Field station SCADA link lost", 0, 0, 0
End If
Set objTag = Nothing
Set objAlarm = Nothing

5. S7-1200 - TIA Portal Watchdog Implementation

The PLC side is built from three blocks: a data block, a function block containing the watchdog logic, and an OB1 rung that calls the FB.

5.1 Data Block DB100

5.2 Function Block FB_CommWatchdog (SCL)

Create FB_CommWatchdog using SCL for clarity. This block must be called from OB1 (or OB30 if you prefer a fixed 100 ms cyclic task).

FUNCTION_BLOCK "FB_CommWatchdog"
{ S7_Optimized_Access := 'TRUE' }
VERSION : 0.1
VAR_INPUT
    i_Heartbeat : UINT;          // From DB100.DBW0 (WinCC toggle)
    i_Timeout   : TIME;          // Default T#25s
    i_Cycle     : TIME;          // OB1 call period, e.g. T#100ms
END_VAR
VAR_OUTPUT
    o_CommOK      : BOOL;
    o_AlarmActive : BOOL;
    o_ResetPulse  : BOOL;
    o_ResetLatched: BOOL;
END_VAR
VAR
    s_LastValue    : UINT;
    s_Age          : TIME;
    s_Init         : BOOL;
    s_PulseTON     : TON;        // 5 s reset pulse
    s_CooldownTON  : TON;        // 5 min re-arm
    s_AgeTON       : TON;        // re-trigger for age counter
END_VAR
BEGIN
    // Initialisation
    IF NOT s_Init THEN
        s_LastValue := i_Heartbeat;
        s_Age       := T#0s;
        s_Init      := TRUE;
    END_IF;

    // Edge detection on heartbeat
    IF i_Heartbeat <> s_LastValue THEN
        s_LastValue := i_Heartbeat;
        s_Age       := T#0s;
        o_CommOK    := TRUE;
        o_AlarmActive := FALSE;
    ELSE
        s_Age := s_Age + i_Cycle;
    END_IF;

    // Alarm condition
    IF s_Age >= i_Timeout THEN
        o_AlarmActive := TRUE;
        o_CommOK      := FALSE;
    END_IF;

    // 5 s reset pulse, with 5 min cooldown to prevent thrashing
    s_PulseTON(IN := o_AlarmActive, PT := T#5s);
    s_CooldownTON(IN := s_PulseTON.Q, PT := T#5m);

    o_ResetPulse := s_PulseTON.Q AND NOT s_CooldownTON.Q;

    // Latch for SCADA visibility
    IF o_ResetPulse THEN
        o_ResetLatched := TRUE;
    END_IF;
END_FUNCTION_BLOCK

5.3 OB1 Call

// OB1 - Main
"DB_Watchdog".i_Heartbeat := "HB_FromSCADA";
"DB_Watchdog".i_Timeout   := T#25s;
"DB_Watchdog".i_Cycle     := T#100ms;

"FB_CommWatchdog_DB"(
    i_Heartbeat  := "DB_Watchdog".i_Heartbeat,
    i_Timeout    := "DB_Watchdog".i_Timeout,
    i_Cycle      := "DB_Watchdog".i_Cycle,
    o_CommOK     => "CommOK",
    o_AlarmActive=> "AlarmActive",
    o_ResetPulse => "ResetPulse",
    o_ResetLatched=> "ResetLatched"
);

// Drive the antenna relay on Q0.0
"AntennaReset" := "ResetPulse";

// Mirror status back to WinCC
"HB_ToSCADA".CommOK      := "CommOK";
"HB_ToSCADA".AlarmActive := "AlarmActive";
"HB_ToSCADA".ResetLatched:= "ResetLatched";

6. Hardware Output - Cycling the Wi-Fi Antenna

The ResetPulse signal is wired to a digital output module of the S7-1200 (for a 1214C the onboard %Q0.0 is sufficient). The output drives a 24 V DC relay whose contact breaks the L+ supply to the antenna. A second contact may be used to light a panel indicator.

6.1 Wiring Schematic

6.2 S7-1200 Output Rating Reference

Per the SIMATIC S7-1200 Programmable Controller System Manual, the digital outputs on a DC/DC/DC CPU source up to 0.5 A per channel at 24 V DC, derated to 0.3 A when grouped. The relay coil must stay below this limit; otherwise insert an interposing relay (Phoenix Contact PLC-OSC or equivalent) capable of switching the antenna supply.

7. Timing Parameters and Tuning

Select the four timing constants as a function of the application. The values below are starting points and must be tuned for the actual link and SCADA load.

8. Commissioning and Verification

1. Compile and download. In TIA Portal, compile the project and download the blocks to the S7-1200. Confirm the CPU transitions to RUN with no SF/BF LED activity.
2. Verify the heartbeat writes. In WinCC, open the variable diagnostics (Tag Management → right-click the S7-1200 connection → Variable Diagnostics). Force an update on SCADA_HB and confirm the value increments every 20 s. Use the WinCC Online tag display to monitor.
3. Verify the watchdog. In TIA Portal, place the PLC in Monitor & Modify on DB100. Confirm HB_Age_ms resets to 0 each toggle and CommOK stays TRUE.
4. Simulate a SCADA outage. Stop the WinCC runtime. Within (timeout + 1 × cycle), confirm AlarmActive asserts and ResetPulse goes TRUE for 5 s.
5. Confirm antenna cycle. With a multimeter on the antenna supply rail, verify 24 V → 0 V → 24 V transition during the reset pulse.
6. Confirm recovery. Restart WinCC runtime. After one full toggle period, CommOK returns TRUE and ResetLatched remains TRUE until operator-acknowledged.
7. Acceptance sign-off. Document the ResetLatched counter increments in the station logbook so field crews can audit auto-reset events.

9. Edge Cases and Field-Proven Caveats

9.1 Loss of WinCC Runtime vs. Loss of Network

The handshake cannot distinguish between the WinCC runtime crashing and the Wi-Fi link dropping. Both appear as a frozen heartbeat. This is intentional - the antenna reset cures the Wi-Fi case, and the WinCC case is harmless because the SCADA simply reconnects on its next polling cycle. Document this behaviour so operators do not interpret an antenna reset as a SCADA fault.

9.2 S7-1200 CPU Stop During SCADA Outage

If the CPU stops while SCADA is also offline, the relay coil de-energises and the antenna falls back to whatever state the relay uses as fail-safe (normally closed contact = continuous power). This is acceptable because no further automation can be performed until the CPU is recovered.

9.3 OB1 Cycle-Time Drift

The age counter increments by i_Cycle on each call. If OB1 is interrupted by long tasks (e.g., recipe block calls), the counter can drift. For deterministic behaviour, place the FB call in a cyclic OB30 with a fixed 100 ms period. Configure OB30 in Device configuration → Properties → System and clock memory → Cyclic interrupts.

9.4 Multiple Stations, One Heartbeat

Each of the eight water wells must have a unique SCADA_HB tag and a unique DB word on its respective PLC. Never share one heartbeat across stations - a single antenna failure would otherwise reset all eight sites, including healthy ones.

9.5 Stuck Value at 65535

The VB action wraps from 65535 to 0, which is still a transition and is correctly detected by the <> comparison. Do not use a comparison such as curVal + 1 = newVal on the PLC side; rely on <> which is robust against wrapping.

9.6 GPS-Synchronised Time Stamp

For diagnostic purposes, the PLC can stamp the last good heartbeat time into a DTL variable and mirror it to WinCC. This makes alarm correlation easier in the WinCC alarm log. Use the RD_SYS_T system function from the S7-1200 base library.

10. Troubleshooting Matrix

11. Operator HMI Display

Mirror the three PLC status bits back to WinCC and display them on the station faceplate:

• CommOK - green indicator, steady on while link is healthy
• AlarmActive - amber indicator, flashing during the cooldown
• ResetLatched - red indicator, steady on after the first auto-reset, operator must acknowledge

This gives the operator immediate visibility of auto-recovery actions and forces acknowledgement so that the station logbook captures every intervention.

12. Variants and Extensions

12.1 Two-Way Handshake

If the SCADA needs to know that the PLC is in RUN, mirror a PLC RUN bit back to a PLC_Running tag and let the SCADA raise an alarm when it clears. This catches CPU STOP events that the one-way heartbeat cannot.

12.2 Multiple Reset Attempts with Escalation

Replace the single 5-minute cooldown with a counter that fires up to three reset attempts and escalates to a permanent alarm if the link does not recover. Use a CTU counter on ResetPulse rising edges; when the counter reaches 3, latch a HardFault bit and notify dispatch.

12.3 OPC UA Variant

For new WinCC projects using OPC UA, expose the heartbeat tag as an OPC UA writable variable on the S7-1200 OPC UA server (firmware V4.4+). The VB action remains identical except that HMIRuntime.Tags is replaced by the OPC UA channel name. Refer to the Siemens Industry Online Support S7-1200 OPC UA configuration guide for server-side setup.

FAQ