Switcher Block (v3.2)

• Full function block name: Lib.Mervis.v3_2.SwitcherBlock
• Inherits from: Lib.Mervis.v2_0.BaseAlarmBlock
• Required project references: Lib.Mervis, Lib.Mervis.Alarms

A universal block for safely switching a single-stage physical device (pump, fan, compressor, valve, heating rod). It bundles the standard functionality needed around such a device: Auto / Manual-On / Manual-Off operating modes, running-feedback supervision, two built-in alarms (Fail-to-start and Failure/Emergency), a periodic exercise “kick”, a run-down overrun, an emergency shutdown that overrides everything, and maintenance counters (operating hours and start count). Because it inherits from BaseAlarmBlock, the block exposes a unified alarm interface that can be chained into the project alarm system.

PROGRAM PUMP_CONTROL_DEMO
    VAR
        Heating_Demand  : BOOL;
        Pressure_Switch : BOOL;
        Thermal_Fault   : BOOL;
 
        PumpSwitcher : Lib.Mervis.v3_2.SwitcherBlock;
        Pump_Relay   : BOOL;
    END_VAR
 
    PumpSwitcher(
        OperatingMode   := OperatingModeType#Automatic,
        Demand          := Heating_Demand,
        Feedback        := BOOL_TO_USINT(Pressure_Switch), // 0/1; use 2 to ignore feedback
        Failure         := Thermal_Fault,
        RunDown         := T#2m,    // 2 min overrun to dissipate residual heat
        FeedbackTimeout := T#10s    // wait 10 s for pressure to build up
    );
 
    Pump_Relay := PumpSwitcher.Command;
END_PROGRAM

Description of Feedback values

Description of ControlFlags values

Being a BaseAlarmBlock descendant, the block also provides the inherited alarm interface (AlarmsInterface / Reset/Acknowledge/GetActiveAlarmsMask…) used to chain its two internal alarms into the project alarm system.

Description of CompositeState output values

The block runs an internal state machine (Off → WaitForFeedback → On → Run-Down, with parallel Failure, Kick and Emergency branches) and manages two alarms:

• Not Run (Fail to Start): raised if Command is issued but Feedback does not become active within FeedbackTimeout. Priority = NotRunAlarmPriority.
• Failure: raised by the external Failure input (and optionally by EmergencyShutdown, see ControlFlags bit 3). Priority = FailureAlarmPriority / EmergencyAlarmPriority.

Core capabilities:

• Operating Modes: Automatic, Manual On, Manual Off. Off is dominant; EmergencyShutdown overrides all modes.
• Feedback Monitoring: verifies the device actually started; if Feedback = 2 the feedback logic is bypassed and the running state mirrors Command.
• Run-Down (Overrun): after Demand is removed in Automatic mode the device keeps running for RunDown (e.g. to dissipate residual heat) before stopping. A device that never reported feedback, or that is dropped on the very first cycle, skips the run-down.
• Periodic Kick (Exercise): in Automatic mode, if the device has been idle for KickFunctionInterval, it is run for KickFunctionDuration to prevent mechanical seizing. The kick has its own feedback supervision.
• Emergency Shutdown: sets Command = FALSE immediately, regardless of mode. If configured (ControlFlags bit 3), a coinciding start request raises a Failure alarm.
• Fault reaction: ControlFlags bits 1/2 decide whether Command is dropped on a fault separately for Automatic and Manual modes (e.g. keep a critical pump running on a soft alarm in Manual).
• Maintenance Counters: accumulated OperatingHours and StartCounter, resettable via OperatingHoursReset.

Execution logic: In Automatic mode a start is requested by Demand; in Manual On it is forced. On a start the block enters WaitForFeedback, asserts Command and waits up to FeedbackTimeout for Feedback. On success it goes On and increments StartCounter; on timeout it raises the Not Run alarm. A Failure input or EmergencyShutdown diverts to the corresponding branch at any time. When Demand clears, the Run-Down branch keeps the device running for RunDown before final stop.

Scenario: Redundant refrigerant pump to the ice-rink floor collector

Two pumps (M4/M5) circulate refrigerant to the floor. Each pump is driven by its own SwitcherBlock:

• The lead/lag demand from the twin-pump logic is wired to Demand; a differential-pressure switch (dP) is mapped to Feedback via BOOL_TO_USINT.
• If the PLC asserts Command but no dP appears within FeedbackTimeout = T#10s, a Not Run alarm flags a broken coupling or seized rotor and the twin-pump logic can switch to the standby pump.
• A motor-protection / inverter fault is wired to Failure; with ControlFlags bit 1 set, the pump's Command is dropped on fault in Automatic mode.
• To avoid seizing during summer standstill, KickFunctionInterval = T#7d and KickFunctionDuration = T#30s exercise the pump weekly.
• A short RunDown keeps refrigerant moving briefly after the demand stops.

Command drives the pump relay; AlarmsInterface is chained into the machine-room alarm aggregator, and OperatingHours / StartCounter feed the maintenance view.