Door position contacts test plan: A buyer-ready test plan for verifying door position contacts across normal, fault, offline, and handover scenarios.

Đăng lúc: Th01 24, 2025

Door Position Contacts: The Test Plan a Buyer Should Demand

Door Position Contacts: The Test Plan a Buyer Should Demand

A door position contacts test plan should prove more than the arrival of an open or close event. It should establish which device observed the door, when it observed it, how the controller interpreted it, what remains available during an outage, and who must correct a discrepancy. That distinction matters because a door position contact, or DPS, reports an open and closed state; it does not report why the door moved. Before commissioning, the integrator should also use an Office Smart Lock Site Survey or Smart Lock Retrofit Site Survey to record the physical conditions that the electrical test must reproduce.

Direct answer. The one-line answer — door position contacts test plan is the working record for door position contacts test plan. door position contacts test plan is worthless until door-position-contacts is attached to it as evidence. A door position contacts test plan is an integration view. Lock, door, reader and system have to behave together. door position contacts test plan is where the first interface behind door position contacts test plan fails.

  • Map the interface — write down what hands over between lock, door and system at door position contacts test plan
  • Test the handover — record the sequence that proves door-position-contacts works end to end
  • Keep the record — store the commissioning evidence with the file for testing plan charter
  • For door position contacts test plan, the order below is fixed and door position contacts is the reference point.

This handbook turns those expectations into testable clauses. Each clause identifies the evidence to demand, the result that permits acceptance, and the boundary that remains unresolved. It is designed for a system integrator preparing an access-control package, a client approver signing the result, and a supplier demonstrating the proposed configuration rather than merely describing it.

1. Test Plan Charter and Acceptance Boundary

The test plan must identify the system under test, the doors included in each test group, the controller and server versions, the access-control software configuration, and the evidence repository. A screenshot of a dashboard is not an acceptance test. A signed record must connect a physical door action to the event received at the controller and, where applicable, to the audit record visible through the local API or cloud API. The catalogue side of this is the smart door locks.

The charter should also state what the test does not claim. A DPS event can establish that a monitored state was detected. It cannot, by itself, establish forced entry, tailgating, an unauthorized person, a fire event, a request to exit, or compliance with an egress requirement. Those interpretations require other evidence or an explicitly configured policy.

flowchart TD
    A[Charter and acceptance boundary] --> B[Define state before event]
    B --> C[Prove source identity and clock ownership]
    C --> D[Exercise installed door transitions]
    D --> E[Inject faults and verify degradation]
    E --> F[Separate position cause and exit requests]
    F --> G[Test contact integrity and power recovery]
    G --> H[Verify persistence reconciliation and records]
    H --> I[Assemble handover evidence]
    I --> J[Draw ownership boundary]
    J --> K[Run final acceptance gate]
    K --> L[Accept tested configuration]
    K --> M[Assign owner and schedule retest]
    M --> K
Decision to closeBuyer-required evidenceMeaning after acceptance
Correct stateA physical action, raw contact transition, controller state, and corresponding event recordThe tested path reliably distinguishes the defined open and closed states
Correct timingSource and receiving timestamps measured against the approved test methodThe event can be placed in the door sequence with an explained tolerance
Correct degradationA deliberate outage, observed local behaviour, and documented recoveryThe installed configuration has a defined offline case rather than an improvised response
Clear ownershipNamed party, defect reference, agreed correction, and successful retestNo unresolved discrepancy passes under an informal handover note

The buyer should define acceptance thresholds before the demonstration. Use named values such as T_open, T_alarm, T_reconcile, and T_stale in the test script. Record the approved value and its source rather than copying a generic number that may not fit the door, controller, or operational requirement.

2. Clause 1: Define the State Before Defining the Event

The parties must agree on the raw state, normalized state, and business meaning of every door position transition. The raw state is what the switch presents through the approved interface. The normalized state is what the controller stores, such as open, closed, or unavailable. The business meaning is what the access-control policy does with that state, such as raising a notification or initiating a defined response.

This separation prevents a configuration label from becoming an unsupported conclusion. A notification called Door Forced Open, for example, may be permissible only if the system has separate evidence that distinguishes a forced condition from a normal open. If the only input is a DPS, the buyer should expect the system to report a state and leave the cause unresolved.

The required test record should contain the door identifier, contact endpoint identifier, raw transition, normalized transition, event type, source identifier, source timestamp, receipt timestamp, and resulting policy action. The test must also show what appears when a transition cannot be normalized. A missing value, unknown state, or stale indication must not silently become a closed-door claim.

DPS testing begins by proving the simple transition, then checking ambiguity around repeated and irregular changes. At minimum, the script covers a deliberate open, a deliberate close, an open held through several reporting intervals, a close after that hold, and a rapid movement that produces repeated transitions. The tester records every raw transition so a later reviewer can distinguish contact bounce from genuine movement.

The project definition should also identify which states are permitted. Use the Door Position Contacts for Smart Lock Projects: What Buyers Must Define First as a companion to this clause. It helps prevent the test from beginning with an assumed state model and forces the buyer to settle terminology before suppliers demonstrate behaviour.

3. Clause 2: Prove Source Identity and Time Ownership

Every event must remain attributable to a known door and a known sensing path. The controller, not an unlabeled dashboard field, should hold the door state and make the access decision. The record must identify whether the state came from a particular contact, a controller input, a downstream edge device, or another approved source. If the chain contains more than one hop, the test must expose each hop rather than hiding it behind the final event name.

The buyer should require the source identifier to survive normal operation, controller restart, server disconnection, event buffering, and reconnection. During each case, the tester must be able to answer: which physical door was moved, which endpoint detected it, which controller accepted it, and where the resulting record appears. A generic device label shared by several doors is insufficient.

The clock clause is equally strict. Name one authoritative clock for event ordering and state the tolerance method used to compare it with every other clock. The record should distinguish the time observed at the source, the time received by the controller, and the time written to the server record. If those clocks differ, the test evidence must show the offset or the approved correction method.

A sensible sequence uses the source timestamp for physical transition order, the controller timestamp for decision order, and the server timestamp for platform receipt. The buyer must not treat a server receipt time as proof that the door changed at that moment. Where the controller is the authoritative event source, its clock must be monitored and its ownership documented. Where a downstream platform owns timestamping, that boundary must be stated explicitly.

The five questions an integration must answer are visible here: each side sends a state or event, one side owns the clock, the record has a defined home and reader, a missing delivery has a defined result, and the supplier or integrator owns the correction. A test that proves only one of those five is incomplete.

4. Clause 3: Exercise Normal Transitions at the Installed Door

The normal test must occur on the installed door with its approved hardware, frame, locking arrangement, wiring, controller, and configuration. A bench demonstration does not establish how the installed contact behaves. The installer and tester should first confirm the door identifier against the physical label and the access-control record, then execute the test from both sides where safe and permitted.

Test actionExpected observationRequired evidenceAcceptance question
Open the door from the normal approach sideRaw and normalized state change according to the agreed modelSource, controller, and event records with timestampsIs the change attributable to the correct door?
Close the door and allow it to settleThe final state reflects the approved closed conditionController state and resulting event historyDoes contact bounce settle without a false persistent state?
Open and close repeatedlyRepeated legitimate transitions remain distinguishableRaw transition sequence and normalized event sequenceAre duplicates or reordered records visible?
Hold the door open beyond a policy intervalThe contact continues to represent open while policy may act on durationState history, policy event, and notification recordDoes the system distinguish duration from cause?
Operate the exit side without using the request-to-exit inputPosition evidence follows only from observed contact changeDoor-state and exit-device records viewed togetherHas position been confused with an exit request?
Restore normal operation after each caseNo stale test state remainsBefore-and-after controller and platform viewsDoes the next test begin from a known baseline?

The approved test method should define how the tester observes the physical door. Use a synchronized visual witness, a tester input, or another approved reference. Store the reference with the system output. This allows an investigator to compare the door’s physical condition with the contact transition without treating the platform display as evidence of the physical condition by itself.

Timing thresholds belong in the script. T_open covers the permitted interval between the physical action and the controller state; T_event covers the interval before the record is available to the intended consumer. The buyer should state which timing boundary is being judged and why. A delay that is acceptable for a daily occupancy report may be unacceptable for a local door alert.

5. Clause 4: Inject Faults and Prove the Degraded Case

Fault injection testing is the part of the plan that separates an observable contact from a dependable integration. The buyer should not ask whether the system continues normally during an outage. It should ask what state the controller holds, which events remain available, how staleness is identified, and what happens when service returns.

The script must cover at least the controller restart, controller-to-server network interruption, server unavailability, local service interruption, event-buffer pressure, and repeated reconnection. Before execution, the supplier must confirm which behaviours apply to the proposed configuration. The demonstration must show the installed behaviour rather than rely on a statement that the product is capable of it.

Injected conditionEvidence to captureRequired resultEscalation if unmet
Controller loses its network pathLocal door state, local event, stale indication, and later reconciliationThe local case is documented and distinguishable from a live stateController or integration owner must specify and implement the missing behaviour
Server becomes unavailableController acceptance, event memory, and queued or retained recordsAccepted events are not silently lost and their availability is visibleApplication owner must correct buffering or delivery handling
Local service is interruptedDoor decision path, controller state, and recovery sequenceThe approved local case continues or escalates without inventing a releaseAccess-control owner must approve the operational response
Controller restartsReconstructed state, retained events, and duplicate handlingThe system distinguishes reconstruction from a new physical transitionController owner must correct persistence or deduplication
Connection repeatedly interruptsEvent order, gaps, duplicates, and final reconciliationThe resulting record can be explained without erasing evidenceIntegration owner must identify the faulty boundary
Event storage reaches its configured limitRetention state, operator warning, and subsequent behaviourThe condition is visible and handled according to the approved procedureApplication owner must resolve capacity or policy handling

The lock reports the state; the system decides what the state means. During an outage, that separation must remain visible. A stale open indication should not be represented as a new open event, while a stale closed indication should not be represented as a fresh confirmation that the door is closed. The accepted design must define how the controller exposes the difference.

Centralised and standalone operation require separate treatment. Confirm on the dated product page for the specific SKU which state-holding arrangement applies, then test that arrangement. The offline case is the design, not the exception. Likewise, fail-safe or fail-secure behaviour is decided by the door’s role and the applicable approval; it is not selected by the test script or assumed from a product family.

6. Clause 5: Separate Position, Cause, and Exit Requests

Door position, request to exit, and release are three different signals with three different owners. A DPS reports the open and closed state of the monitored door. A request to exit, or RTE, is the signal from an exit device indicating operation from the escape side. It is not a release command. Release on the escape side is the default behaviour of the exit device, while the RTE signal tells the system that the operation occurred.

The test must show the three records side by side. Operate the exit device without changing the door position, operate it while the door moves, and move the door without operating the exit device. The acceptance question is not whether the names match. It is whether each source retains its own meaning and whether downstream policy combines them only through an approved rule.

A related trap is the conversion of an open state into a cause. The test script should include an event label that claims a forced or unauthorized condition. If the system has no independent evidence for that cause, the buyer should require the record to remain a position event or apply a clearly defined policy label. Naming a notification aggressively does not improve the underlying evidence.

CONFIRM-WITH-LOCAL-EGRESS-APPROVAL: Confirm the locally approved exit-device, door-force, release sequence, and inspection requirements with the authority responsible for the installation.

CONFIRM-WITH-LOCAL-FIRE-APPROVAL: Confirm any fire-alarm interface, approved release sequence, and handover responsibility with the authority responsible for the installation.

The evidence pack must identify which system owns each decision. The exit device owns its mechanical release function. The controller owns the access decision and state. The fire interface, where applicable, has a separately approved role. The test proves the messages exchanged; it does not create a regulatory approval.

7. Clause 6: Test Contact Integrity, Wiring, and Power Recovery

The contact circuit must be tested against the approved wiring design, including the intended response to an open circuit, short circuit, intermittent connection, and controller input fault. A qualified technician should use the measurement method and thresholds stated in the approved design or manufacturer instructions. The test record should capture the measured result without modifying the design during the demonstration.

The distinction between a wiring fault and a legitimate door movement is essential. If the circuit faults while the door remains still, the system should not manufacture a normal physical transition. If the door moves but the circuit remains stable because of a mechanical or installation issue, the event stream may remain quiet. Both results require an owner and corrective action; neither should be dismissed as a dashboard problem.

Power-recovery testing should cover the controller, the approved local power arrangement, and any upstream device that influences state reporting. The script must show the state before interruption, the state during interruption, the first event after restoration, and whether the system identifies reconstructed or delayed data. The tester should not cycle power repeatedly until an acceptable result appears. The recovery sequence and repeatability criteria should be agreed before testing begins.

Contact integrity also requires a physical-to-logical cross-check. Verify the door label, wiring endpoint, controller input, logical door identifier, notification destination, and access-control report. A correct event on the wrong door is a critical configuration defect even if the signal path itself works. Any remapping, replacement, or retermination should invalidate the affected result and trigger a defined retest.

8. Clause 7: Prove Persistence, Reconciliation, and Readable

The record clause must identify two different stores. The audit log is the retained event record with time and source. Event memory is the buffer used when the device cannot reach the server. Confusing the two makes it possible to claim that no event was lost when it was merely waiting in memory, or to claim that an event was delivered when it only existed locally.

Each event should carry enough information to reconstruct the decision path: logical door, event type, source, source time, controller receipt time where available, platform receipt time, sequence or reconciliation marker, and a quality indicator for delayed or restored delivery. The buyer should decide which fields are mandatory for the installed system and record that requirement in the test script.

The recovery test must compare three ledgers: events accepted locally, events later delivered to the server, and events visible in the audit log. The reviewer should be able to account for every transition without unexplained disappearance or duplication. Where a record is rejected, the reason should be visible to an authorised operator or retained in an approved diagnostic view.

CONFIRM-WITH-LOCAL-PRIVACY-APPROVAL: Confirm applicable retention, access, disclosure, and employee-monitoring rules for door-position records with the client’s legal and privacy owners.

A door-position record may not identify a person by itself, but a timestamp, door identifier, credential event, video reference, or occupancy context can make it personal information. The test should therefore use representative but non-production identities where possible. The handover pack must state who may read operational records, who may export them, and who may approve retention changes without turning that statement into an unsupported legal conclusion.

9. Clause 8: Build an Evidence Package That Survives Handover

Acceptance evidence must allow another integrator to repeat the test without asking the original tester to reconstruct it from memory. Use the Door Position Contacts Requirements Document Template to capture the approved state model, thresholds, fault cases, door groups, and responsible parties before the demonstration.

The evidence index should include the approved test script, configuration baseline, door schedule, controller and application versions, network diagram, clock-source record, completed test sheets, raw event exports, controller snapshots, exception list, defect records, retest results, and signed acceptance. Each artifact should carry the same project, door, and test identifiers so that a transition can be followed from physical action to retained record.

Use the Documenting Door Position Contacts for Handover and Audit to turn the package into an operating record. The handover must explain how to identify a stale state, how to distinguish a delivery delay from a physical transition, how to retrieve buffered records after restoration, and who may reset a controller without erasing evidence.

The evidence review should include a deliberate negative case. Select one failed or intentionally degraded test and trace it through the package. If the reader cannot identify the expected result, observed result, owner, corrective action, and retest, the package is incomplete. A dashboard export alone will fail this review because it omits the physical baseline and the decision path.

10. Clause 9: Draw the Ownership Boundary Before Testing

The buyer should name the party accountable for each result, but accountability must follow the boundary. The supplier may own the demonstrated device or component behaviour. The access-control integrator owns the controller, event mapping, local-to-platform flow, and configuration under its scope. The electrical or door installer owns the physical contact, wiring terminations, and door hardware adjustment. The client approver owns the accepted thresholds and operational rules.

Failure symptomFirst owner to investigateRequired escalation evidence
No transition appears after a physical door actionAccess-control integrator, with the installer checking the sensing pathInput state, endpoint mapping, and measured circuit result
A transition appears against the wrong doorConfiguration ownerDoor schedule, endpoint map, and correction record
A false transition appears without door movementInstaller and integrator, depending on the measured evidenceRaw transition history and contact integrity result
Event is delayed after reconnectionApplication and network ownersLocal acceptance, delivery timing, and reconciliation record
Duplicate records appear after recoveryController and application ownersSource sequence, deduplication rule, and retained audit record
State cannot be classifiedController owner and client rule approverRaw value, normalized result, and approved meaning

Ownership must extend through correction and retest. Assigning a defect is not closure. The named owner must provide a correction, the approver must accept the associated risk where relevant, and an independent tester must repeat the affected case. A configuration change that alters event names, thresholds, state mapping, clock handling, or offline behaviour triggers impact analysis and targeted regression.

For a multi-door portfolio, repeatability matters more than a one-off demonstration. The Door Position Contacts for a Multifamily or Office Portfolio can help group doors by expected behaviour and reduce the chance that a successful laboratory example is mistaken for portfolio-wide acceptance. The buyer should still retain evidence for each tested configuration.

11. Clause 10: Pass the Final door position contacts test plan

The final acceptance gate should use a controlled review rather than a general impression. The supplier demonstrates one complete normal transition, one deliberately injected fault, one recovery and reconciliation, and one previously documented exception. The tester then opens the retained records independently and confirms that they support the stated result.

Acceptance outcomeRequired decisionCommercial consequence
PassAll mandatory clauses have linked evidence, no open critical defect, and retests are acceptedThe tested configuration may proceed within its documented scope
Conditional passA bounded non-critical issue has an approved owner, action, and review dateScope and expiry are recorded; unverified functions remain excluded
FailA mandatory state, timing, degraded-mode, ownership, or evidence clause is unmetThe affected scope is not accepted and the related test remains open
IncompleteEvidence is missing, unattributed, or not repeatableAcceptance is deferred until the evidence gap is closed

Use a controlled test record for each case. It should contain the project and door identifiers, test version, participants and roles, preconditions, fault introduced, physical action, expected result, approved threshold, source and receipt timestamps, observed result, evidence links, outcome, defect reference, owner, and retest reference. Keep the record even when the result is a pass, because it establishes the actual tested boundary.

The acceptance decision should identify the exact configuration that passed: door type, hardware arrangement, controller and application versions, firmware or software baseline where relevant, and applicable project settings. Any later change that could alter the sensing path, event mapping, timing, retention, or degraded operation should receive a documented impact review. The test plan then becomes a controlled baseline rather than a one-time demonstration.

Before approval, ask the bidder to walk the client through one completed test record, one deliberate fault, one recovery, and one unresolved exception. A promise that a dashboard will show door activity is not an acceptance package. Request the door position contacts test plan as a contractual deliverable with named owners, objective evidence, and retest obligations. The buyer retains the right to reject a result supported only by a screenshot, a verbal assurance, or evidence from a configuration that differs from the one proposed for deployment.

It is worth pairing this with the smart lock durability testing. That is the same question door position contacts test plan has to answer here.The test plan is complete when another qualified party can repeat it, challenge its conclusions, and trace every reported state to its source, clock, controller decision, retained record, and responsible owner. That standard turns door contact monitoring from a feature description into evidence that the building integration can be accepted, maintained, and audited without relying on undocumented assumptions.

The next step is to request a quote.

Giỏ hàng của bạn
Trải nghiệm của bạn trên trang web này sẽ được cải thiện bằng cách cho phép cookie Chính sách cookie
Chat on WhatsApp