Two small devices on the same secured door, answering two completely different questions — and a system that can't tell a legitimate exit from a break-in without both of them.
Walk up to almost any electrified access-controlled door and you'll find two low-voltage devices that get lumped together as "door hardware" and treated as roughly interchangeable. They aren't. A door position switch (DPS) — usually a magnetic contact or reed switch split across the frame and the door leaf — reports one thing and one thing only: is this door physically open, or is it physically closed. A request-to-exit (REX) device— usually a ceiling-mounted PIR motion sensor or a push-button on the secure interior side — has a completely different job: telling the access control system that someone inside intends to leave, so the lock can release and the alarm logic can stand down, without ever asking that person for a credential. Confuse the two, or assume either one is optional because "the other one basically covers it," and the door either can't distinguish a legitimate exit from a forced entry, or loses the ability to know its own physical state at all.
The door position switch doesn't know or care why the door opened. It's two halves of a magnetic circuit — one embedded in the frame, one in the door leaf — that report a binary physical state to the access control or alarm panel: contact made means the door is closed, contact broken means the door is open. A valid badge, a REX trigger, a fire-alarm unlock relay, someone propping it with a wedge, or a crowbar — the DPS reports the identical "door is open" state every single time, because it has no concept of authorization at all. It is purely a physical-state sensor. The REX device exists for an entirely different reason. Life-safety and building egress codes generally require that people be able to leave a secured space freely, without presenting a credential — fire and life-safety egress must never depend on card access. So a PIR motion sensor (or a push-button) is mounted on the secure interior side near the door, watching for someone approaching to leave. When it fires, it tells the access control panel two things at once: release the electric lock momentarily, and — critically — suppress the forced-door/propped-door alarm logic for that unlock window, because this opening is an expected, authorized exit, not a breach.
The reason both devices exist at the same opening is that the access control system needs to distinguish between genuinely different real-world events, and neither sensor alone can do it. The DPS state (open/closed) combined with whether REX fired and whether a valid credential was presented produces four distinct, mutually exclusive system responses.
The DPS answers "is the door physically open or closed right now" — a pure physical-state question with no concept of authorization built in at all. The REX answers a different question entirely: "was this specific opening an authorized, expected exit, so the alarm logic should stand down for it." Without a REX device, every legitimate exit produces the exact same DPS signature — door open, no credential presented at the exit side, since egress must be free — as an unauthorized forced entry from outside. The system would either alarm on every single legitimate exit (unworkable) or have to disable forced-door alarming altogether (unsafe). Without a DPS, the system has no independent way to confirm the door's actual physical state at all — whether it's open, closed, or sitting propped open long after a valid unlock — regardless of how or why it was unlocked. Put together, the two signals let the panel distinguish all four scenarios; either one alone leaves a real gap the other was specifically built to close.
False. A door position switch is a pure physical-state sensor — open or closed, full stop — with no concept of authorization built into it whatsoever. A REX device specifically exists to distinguish an authorized, intentional exit from an unauthorized breach, so that a person legitimately leaving a secured space doesn't trip a false forced-door alarm. Removing either device breaks a genuinely different function, not a redundant one: pull the DPS, and the system loses the ability to know the door's actual physical state at all — open, closed, propped — no matter how it was unlocked. Pull the REX, and every legitimate exit becomes indistinguishable from a forced-entry event to the alarm logic, since the DPS alone can never explain whythe door opened. They aren't two sensors checking the same thing twice — they're two sensors answering two different questions that the access control system needs answered independently, at the same door, for every single opening.
Explains why a request-to-exit (REX) device and a door position switch (DPS) are not redundant sensors, but two devices answering two different questions at the same secured door: the DPS reports the door's physical open/closed state with no concept of authorization, while the REX signals the access control system that a specific opening is an authorized, intentional exit so the forced-door and propped-door alarm logic can stand down for it.
A door position switch (DPS) is typically a magnetic reed contact split into two halves — one mounted in the door frame, one in the door leaf. When the door is closed, the magnetic circuit is complete and the switch reports "closed"; when the door opens, the circuit breaks and the switch reports "open." That is the entirety of its function. It has no wiring back into the credential/authorization logic and no way to know whether the opening was produced by a valid badge, a REX trigger, a fire-alarm unlock relay, a maintenance override, or someone forcing the door with a pry bar. Every one of those events produces the identical "door open" signal from the DPS. It is, by design, a pure physical-state sensor.
A request-to-exit (REX) device — most commonly a ceiling-mounted PIR motion sensor covering the area just inside a secured door, sometimes a mechanical push-button instead — detects that an occupant on the secure side intends to leave. When it fires, it does two things simultaneously at the door controller: it releases the electric lock for a momentary unlock window, and it tells the alarm logic to suppress the forced-door and propped-door alarms for that same window. This exists because life-safety and building egress codes generally require free egress from a secured space — people must be able to exit without presenting a credential, since fire/life-safety egress can never depend on card access. Without a mechanism to flag "this opening is an expected exit," every legitimate departure would look identical, from the alarm logic's point of view, to someone forcing the door from outside.
The access control system has to distinguish four genuinely different events at the same opening: an authorized exit (REX fires, lock releases, DPS shows open, no alarm), an unauthorized forced entry (DPS shows open with no REX trigger and no valid credential, so the forced-door alarm sounds), a door propped open too long (DPS shows open beyond the allowed time even after a legitimate unlock, so a separate propped-door alarm sounds), and the normal secure/armed state (DPS shows closed, system armed, no alarm). The DPS alone can only ever report the first bit — open or closed — and has no way to explain why. The REX alone only fires at the moment of an intentional exit; it says nothing about the door's ongoing physical state afterward, including whether it swung shut again or was left propped. Each device supplies exactly the piece of information the other cannot.
Remove the REX device and every legitimate exit becomes indistinguishable from a forced-entry event to the alarm logic — the system either has to alarm on every normal exit (unworkable for occupants and monitoring staff alike) or disable forced-door alarming altogether, which defeats the purpose of the alarm. Remove the DPS and the system loses all visibility into the door's actual physical state — it has no way to detect a door that has been forced open, propped open after a legitimate unlock, or simply left ajar, regardless of how the unlock happened. The two failure modes are unrelated because the two devices were never doing redundant work in the first place.
No. PIR motion sensors are the most common form because they require no action from the person leaving, but mechanical or electronic push-buttons mounted near the door are also widely used as REX devices, particularly where a motion sensor's detection zone would be hard to control (e.g., adjacent to a busy corridor). Both accomplish the same function: telling the access control panel that this specific opening is an authorized exit.
Yes — a DPS reporting "open" with no valid credential presented and no REX trigger is exactly the condition that produces a forced-door alarm. And a DPS reporting "open" for longer than a configured time limit, even after a fully legitimate unlock, is what produces a propped-door alarm. The DPS is the sensor doing the reporting in both cases; it just has no authorization context of its own, so the panel's logic (informed by REX and credential state) decides what the open state means.
Life-safety and building egress codes generally prohibit that. Free egress from a secured space is typically a code requirement — occupants must be able to leave without a credential, because emergency egress can never be allowed to depend on an access control system staying powered, online, or correctly provisioned with the person's credential. A REX device satisfies that free-egress requirement while still letting the door's forced-entry alarm logic stay armed against unauthorized entry from the outside.
No — only for a brief, configured unlock window tied to that specific triggered exit. Once that window elapses, forced-door and propped-door alarm supervision returns to normal. A REX trigger is not a mode change; it's a momentary, event-scoped suppression tied to one detected exit.
That is a legitimate, distinguishable condition in a well-designed system: REX fired (so the unlock window and alarm suppression were granted) but the DPS never reports "open," meaning the door was never actually used. Many access control systems log this as a no-activity or unused-REX event rather than an alarm condition, since nothing insecure occurred — it simply confirms the two devices are reporting independently, exactly as intended, rather than one substituting for the other.
Not universally — requirements depend on the applicable fire/life-safety code, the AHJ, and the door's function (an interior office door has different requirements than a fire-rated egress door on an electrified lock). But wherever free-egress and forced-door alarm supervision both apply to the same electrified opening, both devices are typically specified together precisely because each covers a function the other cannot.
Try our Physical Security Studio
More calculators, simulators, and guides for this discipline.