Selecting a safety interlock switch for electrical work is governed by BS EN ISO 14119, with the first decision being whether plain interlocking or guard locking is required: plain interlock suits hazards that stop immediately on guard opening, while guard locking (solenoid-held) is mandatory wherever the machine has run-down time or retains stored energy [S3].
For UK sites, Regulation 11 of the Provision and Use of Work Equipment Regulations 1998 (PUWER) and the Supply of Machinery (Safety) Regulations 2008 place a legal duty on any interlocked guard that is opened regularly, and the choice is constrained by sector environment, hazard nature, and access frequency, not by preference [S2]. Three switching technologies compete here: mechanical contact, coded-magnet, and non-contact (RFID-coded), each with different defeat-resistance, wiring, and wear profiles.
Legal Framework and the Hierarchy of Control
Under PUWER Regulation 11 and the Supply of Machinery (Safety) Regulations 2008, fixed physical guarding remains the first protective measure; interlocked guards are only acceptable where regular access prevents fixed guarding from being practical, and they must be supported by a documented risk assessment and Declaration of Conformity [S2]. The hierarchy of control is set out in BS EN ISO 12100, which PUWER Regulation 11(2) directly mirrors: eliminate the hazard by design first, apply fixed guarding where regular access is not needed, then interlocked guards or light curtains for genuine access points, with administrative controls such as signage and training as the final layer rather than a substitute [S2]. Guard locking is explicitly required wherever machinery has a hazardous run-down time or retains energy after stop, which is the rule that decides between a plain interlock and a solenoid-locked device [S2].
Door Switch or Door Lock: The First Decision
The first technical question is the machine's stopping behaviour, because that single fact determines whether you buy a contact interlock or a guard-locking solenoid switch [S3]. A plain interlock such as the DAIDISIKE DX-D2 or DX-D3 detects the open guard and drops the safety circuit, with forced (positive) contact separation at 60 N minimum over 10 mm of travel so a welded contact cannot fake a closed-door signal [S3]. Where the hazard continues after the stop command (spindles, fans, presses finishing a stroke) or where mid-cycle interruption would damage the process, guard locking is mandatory: the door is physically held shut until the control system confirms a safe state, using devices such as the DX-W2, DX-W3, DX-W5 solenoid locks or the DX-D6 compact guard-locking switch rated to 2000 N holding force [S3]. For full-body access, a DXL safety door bolt (48 mm bolt travel, 1-10 mm door gap, 1,000,000 operation rating) accepts DX-W2/W3/D2/D3 switches directly, combining a handle mechanism with the electrical interlock in one assembly [S3].
Power-to-Release versus Power-to-Lock
Guard-locking switches split into two locking principles that follow from the risk assessment, not from preference: power-to-release (mechanically locked, solenoid released, DAIDISIKE code GC or DX-D6 C-type) holds the door on a spring and releases only when the solenoid is energised, so a power failure keeps the door locked and protects personnel from a coasting machine [S3]. Power-to-lock (code GD or DX-D6 D-type) holds only while energised and unlocks on power loss, which is the right choice where the dominant danger is trapping personnel inside a cell or where the lock exists to protect the process rather than the person [S3]. The DX-W2, W3, and W5 are offered in both GD and GC versions and the DX-D6 in both C and D types, so the mechanical and electrical design carries over whichever way the risk assessment lands [S3]. A power-to-release cell also needs an escape or maintenance release strategy for any person who could be locked inside, and that release must be specified at the same time as the lock, not retrofitted.
Holding Force, Coding Level, and Defeat Resistance
Holding force on a guard-locking switch must be sized against the realistic pull a person can apply to that specific door geometry, since a long lever arm on a tall door multiplies the force a hand can deliver; the DAIDISIKE DX-W2, W3, and W5 are each rated 1300 N, while the DX-D6 raises this to 2000 N, and a determined adult pulling with both hands can exceed 1000 N on a well-levered handle [S3]. The second axis is coding level under BS EN ISO 14119, which defines four coding levels (1 to 4) tied to how easily the actuator can be defeated or substituted; level 1 uses an uncoded mechanical actuator, while level 4 requires a fully coded RFID actuator that the switch validates electronically before allowing start [S2]. Plain mechanical switches are level 1 and can be defeated with a spare key or a bent piece of metal, coded-magnet devices (such as the DAIDISIKE DX-C1) reach higher levels because the switch only responds to a specific magnetic pattern, and RFID-coded non-contact switches (DX-R1) reach level 4 and tolerate misalignment of several millimetres while remaining defeat-resistant [S3]. Selection is therefore not a single number but a three-way trade between force, coding, and environment.
Mechanical, Coded-Magnet, and Non-Contact Compared
Mechanical contact interlocks (DX-D2, DX-D3) are the lowest cost and the easiest to wire, but they need precise actuator alignment, their contacts wear with every cycle, and they are trivially defeated with a substitute actuator [S3]. Coded-magnet switches (DX-C1) tolerate misalignment, are sealed against washdown, and resist defeat at coding level 2 or 3, but they require a dedicated safety relay or safety PLC to validate the coded signal and they cost more per point [S3]. Non-contact RFID-coded switches (DX-R1) reach the highest coding level, tolerate misalignment of several millimetres, and have no mechanical wear face, but they need a safety controller that can interpret the coded protocol and they are the most expensive option per switch [S3]. The comparison across the three types for a typical electrical-cabinet guard door runs: mechanical at lowest cost and shortest spec life, coded-magnet in the middle for washdown or misalignment, RFID non-contact at the top for defeat resistance and long service life.
Contact Configuration, Wiring, and Integration
Safety interlock switches expose either electromechanical contacts (typically 1 NC + 1 NO, or 2 NC for redundancy) or solid-state OSSD outputs on the higher-end non-contact types, and the wiring must reach a safety relay, safety PLC, or a certified contactor expansion module that performs cross-fault monitoring [S1]. A common failure pattern is wiring a 4-wire DC switch to an AC safety relay that expects potential-free contacts, or running a non-coded NO contact as the safety channel rather than the NC, which defeats the positive-opening requirement [S1]. For coded-magnet and RFID devices, the safety controller must be rated for the coding level claimed on the switch data sheet, because a level 4 RFID switch wired into a level 1 controller collapses to level 1 in practice [S3]. Daisy-chaining more than one switch into a single safety input also collapses the coding level, since the controller can no longer tell which guard is closed, and BS EN ISO 14119 flags this as a common defeat path.
Verification, Documentation, and Periodic Test
BS EN ISO 14119 places a verification duty on the installer: every interlocked guard must be tested at commissioning for positive opening, holding force, and coding validation, and the results recorded alongside the Declaration of Conformity and the risk assessment [S2]. Periodic test intervals should be set by the user based on cycle count and environment, not on a fixed calendar, because a washdown or dusty cell will shorten the realistic service life well below the manufacturer's mechanical rating [S2]. For sites that also handle low-voltage distribution work, the same documentation logic applies to LV electrical lockout procedures, where the interlock's NC contact feeds the same safety circuit that drives the electrical automation controller, and a failure of either side invalidates the claim of a coordinated safety function. Two trackable signals to watch over the next quarter are the publication of the next BS EN ISO 14119 amendment and any update to the PUWER Approved Code of Practice that clarifies remote-release requirements for power-to-release cells.
This topic is covered further in FKM fluororubber selection for rail industry: temperature, media, and low-temperature.