Two different ways to physically stop tailgating — not just detect it — and why picking the wrong one for a busy lobby or a data center door is a real design mistake.
"Anti-tailgating" sounds like one feature, but it's really a spectrum. An optical security turnstile watches for a second body following a single valid credential and reacts with sensors and alarms — while still letting a real line of people move through quickly. A mantrap (interlocking vestibule) doesn't watch for tailgating at all; it makes simultaneous dual-door passage mechanically impossible in the first place. Confusing the two — or treating the choice as a matter of budget or looks — is how a data center ends up with insufficient assurance at its most sensitive door, or a busy lobby ends up with a bottleneck it can never move enough people through.
An optical (or full-height) security turnstile is a physical lane that's supposed to admit one person per one valid credential. Infrared beams across the lane count bodies as they pass; if two bodies cross on a single valid presentation, the turnstile knows a second, uncredentialed person just tailgated in. Depending on the design, it responds with an audible/visual alarm, a hold on the gate, or — in full-height, high-security turnstile models — a physical obstruction of the second person. Crucially, most optical turnstiles used in lobbies keep their physical barrier low (glass or metal gate arms at roughly waist height) precisely so that authorized traffic keeps moving quickly. A mantrap, by contrast, is a small enclosed vestibule with two doors — an entry door and an exit door — wired so that an electronic or mechanical interlock makes it physically impossible for both doors to be open, or unlocked, at the same time. A person enters through door A, door A closes and locks, an identity check (and sometimes a weight sensor or biometric re-verification confirming exactly one occupant) happens inside the sealed vestibule, and only then does door B unlock. There is no sensor decision being made about whether tailgating happened— the geometry of the interlock simply never allows a second person's path in alongside the first.
A mantrap trades all of that speed for certainty. The interlock isn't a rule enforced by software watching a sensor — it's wired so that door B's unlock circuit is physically incapable of energizing while door A is open or unlocked, and vice versa. That means a second person standing right behind the first at door A has nowhere to go: door A only fits, and only admits, the person it just authenticated, and it locks the instant it closes. There's no race to beat, no beam to slip past faster than the system can react — the second door simply cannot open until the first is fully secured, one person at a time.
A turnstile lane accepts a small, real possibility that a fast or aggressive tailgater beats the sensor and gate reaction, in exchange for moving dozens of authorized people through per minute — a reasonable trade at a busy lobby, where the population is large, mostly trusted, and a rare missed tailgating attempt is caught by other layers (badges, cameras, receptionists). A mantrap accepts a hard one-person-at-a-time cycle time, in exchange for making that same failure mode mechanically impossible— a reasonable trade at a data center or vault door, where the population is small, the consequence of a single successful tailgating event is severe, and there is no acceptable rate of "the system usually catches it." Neither device is more advanced than the other; they sit at different points on the same throughput-vs-assurance tradeoff, and the correct choice depends entirely on the traffic volume and risk level of the specific opening.
False, or at least badly incomplete. A turnstile generally detects and deters tailgating — sensors count bodies, and the system alarms or partially obstructs a second person — while still allowing relatively high throughput. But for many turnstile designs, a determined or fast-moving tailgater can still physically get through before the system fully reacts; detection is not the same guarantee as prevention. A mantrap physically preventssimultaneous dual-door passage entirely, through a mechanical or electronic interlock that never allows both doors to be open or unlocked at once — there is no window of time in which a second person could slip through even if they tried. That difference in assurance is real and unavoidable, and it comes at a real, unavoidable cost in throughput. Choosing a turnstile for a data-center entrance would leave insufficient assurance at a very-high-consequence door; choosing a mantrap for a busy building lobby would create a throughput bottleneck the space can't tolerate. The choice is a genuine security-vs-throughput engineering tradeoff tied to the specific area's risk level and traffic volume — not an interchangeable stylistic one.
Explains the real design difference between optical/full-height security turnstiles, which detect and deter tailgating with sensors while preserving high foot traffic throughput, and mantraps (interlocking vestibules), which physically prevent simultaneous dual-door passage via a mechanical/electronic interlock — at the cost of much lower throughput.
An optical (or full-height) security turnstile is a physical lane fitted with infrared sensor beams that count how many bodies cross it per valid credential presentation. One valid badge should produce exactly one body crossing; if the sensors detect a second body crossing on the same presentation — a tailgater following closely behind an authorized person — the turnstile responds according to its design: an audible/visual alarm, a hold or reversal of the gate mechanism, and in some full-height, high-security models, a physical obstruction of the second person. Many turnstiles used in lobbies deliberately keep their physical barrier low (waist-height glass or gate arms) rather than floor-to-ceiling, specifically so that authorized traffic can keep moving quickly during rush periods.
A mantrap is a small, fully enclosed vestibule with two doors — one to the unsecured side, one to the secured side — wired with an electronic or mechanical interlock that makes it physically impossible for both doors to be open, or unlocked, at the same time. A person enters through door A; door A closes and locks; often an additional identity or credential check happens inside the sealed vestibule, sometimes including weight-sensing or biometric verification to confirm that exactly one person is present; only once door A is confirmed fully closed and secured does door B unlock, admitting the person to the secured side. At no point in the sequence can a second person have a door available to follow through alongside the first.
Turnstiles trade a small, real possibility that a fast or determined tailgater beats the sensor-and-gate reaction time, in exchange for high sustained throughput — appropriate for busy, moderate-security entrances like office lobbies, where detection (not absolute physical prevention) is an acceptable security posture given the population size and available secondary layers (video, receptionists, badge logs). Mantraps trade throughput for near-absolute physical prevention via the interlock, appropriate for low-traffic, extremely-high-security points such as data centers, vaults, and secure labs, where even a rare tailgating event carries severe consequences and must be physically impossible rather than merely detected and alarmed on.
Specifying a turnstile at a data center server room entrance provides insufficient assurance for the consequence level of an unauthorized entry — a fast tailgater can still physically get through before sensors and gates react. Specifying a mantrap at a busy building lobby creates a throughput bottleneck the space cannot tolerate — a one-person-at-a-time interlock cycle simply cannot move a rush-period crowd. Both are real, mismatched engineering decisions, not interchangeable preferences.
For many optical turnstile designs, yes — the sensors detect the second body and trigger an alarm and/or gate reaction, but that reaction takes a finite amount of time, and the barrier itself (often waist-height glass or gate arms, chosen to preserve throughput) is a real but not absolute physical impediment. A fast or aggressive tailgater can sometimes get through before or during the system's response. Full-height, high-security turnstile variants raise that barrier substantially but still generally prioritize higher throughput than a mantrap.
For the specific failure mode of two people passing through together on one credential, yes — the interlock is a mechanical/electronic guarantee that both doors are never open or unlocked simultaneously, so there is no window in which a second person could follow the first through. Other attack vectors (forced entry, credential theft used one person at a time, social engineering of an operator) are separate risks a mantrap doesn't address by itself.
Throughput. A mantrap admits roughly one person per interlock cycle, with a real time delay for the door-close, verify, and unlock sequence. Deploying that at a busy building lobby would create a bottleneck the space cannot absorb during normal traffic, let alone rush periods — the assurance gain isn't worth the throughput loss for a large, moderately-trusted population.
It varies by design, but commonly includes a second credential or biometric presentation once inside the sealed vestibule, and sometimes a weight sensor in the floor confirming that only one person's mass is present. The purpose is to catch the rare case where a very small stowaway (a child, for example) enters alongside an adult through door A before it fully closes — the vestibule provides one more explicit checkpoint before door B is allowed to unlock.
They narrow it, not close it. A full-height (floor-to-ceiling) security turnstile raises the physical barrier substantially compared to a waist-height optical turnstile and can more reliably obstruct a following person, but it is still fundamentally a single rotating/sliding barrier reacting to sensor input, not a two-door interlock that makes simultaneous passage geometrically impossible. It sits between a standard optical turnstile and a mantrap on the throughput-vs-assurance spectrum, rather than matching mantrap-level assurance outright.
Yes, commonly. A facility might use optical turnstiles at its main lobby entrance to handle high daily traffic, while reserving mantraps for a small number of especially sensitive interior doors — a data center floor, a vault, a secure lab — where the traffic volume is low and the consequence of a tailgating failure is severe enough to justify the throughput cost.
Try our Physical Security Studio
More calculators, simulators, and guides for this discipline.