A new barrier gate lane is not "commissioned" because the arm goes up and down. It is commissioned when the mechanical install, the electrical supply, every detection device, the access-control or payment trigger and every safety function have each been tested, recorded and signed off — before the lane carries paying traffic. Most lanes that fail in month three failed a test in week one that nobody wrote down.
This guide is written for the people who have to accept the lane: the property's facility or operations manager, the installing contractor, the access-control integrator, and the distributor who supplies the kit and then gets the service calls. It covers what to test, in what order, what instruments to bring, and what to collect before the last invoice is paid.
Two related guides go deeper on the layers either side of this one: how the hardware is installed and kept working is covered in the barrier gate installation and maintenance guide, and what happens after handover is scheduled in the service plan guide. This page is only about the acceptance stage itself.
A barrier lane is a chain of five subsystems working in sequence: the drive and mechanism, the power supply and protection, the vehicle detection devices, the access or payment trigger, and the safety devices. A lane can open perfectly on a bench test and still be unsafe or unusable because one link in that chain was never proved under real conditions — a loop that only sees a broom handle pressed on it, a photocell that was aligned in bright sun and drifts in rain, a relay pulse that is 200 ms too short for the controller to accept, or a manual release that nobody has ever operated.
Commissioning exists to convert assumptions into recorded facts. The output is not a working gate — that is the installer's job. The output is a signed test record, a settings record and a documentation pack. Those three things are what let you ask a factory or a service company for help six months later and get an answer instead of a guess.
Three parties normally sign commissioning off, sometimes four:
The sequence matters, because each later test assumes the earlier layers already pass:
Do not let the integrator declare the lane "ready to test" before steps 1 to 4 are recorded. Integration faults are frequently blamed on the gate, and the fastest way to shorten that argument is a recorded mechanical and electrical baseline that everyone has already agreed to.
Print this table and work down it. Each row is a stage that has to be closed before the row below it is worth testing — the order is the point, because integration and safety cannot be signed off on top of an unrecorded mechanical and electrical baseline. The pass criteria come from the specifications we publish for our own models, so they are the numbers to hold an installing contractor to at handover.
| Stage | What to measure on site | Instrument or method | What counts as a pass |
| 1. Mechanical | Cabinet level in both axes; arm on the correct side and covering the lane; spring balance; fasteners torqued; drainage running away from the base | Spirit level on the cabinet’s own reference faces; hand test on the arm — move it to about 45° and push it down toward 20° | The arm rebounds back toward horizontal instead of falling flat: the field reference many manufacturers use is a rebound to around 60°. No creep at the end of travel, and the manual release works from outside the cabinet without tools. |
| 2. Electrical | Supply voltage at the cabinet at rest and during the opening stroke; protective device and isolation; earth continuity; controller configured for the site’s grid | Multimeter at the cabinet terminals with the drive running under load; lock-off test on the isolator | The reading holds in both states — a supply that is correct when idle and sags on start is a cable or circuit fault, not a gate fault. Controller setting matches the delivery: AC 110–240 V, 50/60 Hz. |
| 3. Local control | Opening and closing time measured on the lane; repeatability; limit positions; the controller’s cycle counter | Stopwatch on the lane, then at least 20 consecutive open/close cycles, reading the counter before and after | Measured travel matches the setting the site needs for its peak queue — our adjustable ranges are 1.5–6 s (SC02), 0.9–6 s (intelligent series) and 0.3–1 s (servo high-speed). The arm stops in the same position every cycle. |
| 4. Detection | Each device on its own, then the sequence together: approach loop, safety loop under the arm, photocells, radar | Test vehicle or metal plate per the manufacturer’s method; walk-test across each beam | No device is bypassed, and the arm cannot close while the safety loop or a beam reports a vehicle. |
| 5. Integration | Credential or plate read → release signal; anti-passback; payment or ticket release; remote monitoring | Live test at the lane with every credential type the site will actually issue | Every credential type releases the arm or gate, and the release relay pulse is long enough for the controller to accept — a pulse around 200 ms short is a classic intermittent failure that only appears at peak hour. |
| 6. Safety | Obstacle reaction and force limiting; emergency opening; behaviour on power loss; signage and lane markings | Physical obstruction test at walking speed, repeated at several points along the arm; mains isolation with a live battery test | Force limiting verified against the EN 12453 reference used when a barrier incident is investigated (400 N crushing, 1,400 N impact); the arm reaches open on power loss and by manual release. |
| 7. Handover | Training, keys and release tools, test record, settings record, spares, warranty registration, outstanding snags | Signed checklist per party, filed with the acceptance documents | The three outputs exist in writing — a signed test record, a settings record and a documentation pack — with every open snag listed against an owner and a date. |
Two numbers are worth writing into the acceptance record on the day the lane changes hands, because they are the two nobody can reconstruct later: the as-commissioned opening time measured on the lane, and the controller’s cycle counter at handover. Almost everything a service company is asked about six months later starts from those. The procedure behind the mechanical row is set out in our spring adjustment guide, and the detection chain is covered device by device in the barrier gate safety sensors guide.
Work through the mechanical layer while the drive is still isolated:
Two useful references here are the arm length guide for sizing and the spring adjustment guide for the balance procedure; neither should be improvised on the day of handover.
Measure opening and closing time with a stopwatch — not from the datasheet, from the lane. Models in our range are adjustable rather than fixed: the SC02 opens or closes in 1.5–6 s depending on setting, the intelligent series runs from 0.9 s to 6 s, and a servo high-speed model with a 24 V DC drive and a 480 W motor reaches 0.3–1 s. The number that matters is the one the site actually needs for its peak queue, recorded so that "it got slower" becomes a measurable claim later.
Then run a repeatability test: at least twenty consecutive open/close cycles, and note:
Twenty cycles is not an endurance test; it is a trend test. Surge-protection and overheating problems, arm droop and limit drift all tend to show up in the first twenty movements of a newly installed lane, in quiet conditions, rather than on the first busy Monday.
The electrical layer is where "it worked yesterday" turns into an intermittent fault, so measure it under load:
Write the readings down with the instrument used. A voltage figure without a date, a load condition and a measuring point is not evidence.
Test every detection device separately first, using something that behaves like a vehicle:
Then test the sequence rather than the parts: a vehicle arrives, the trigger fires, the arm opens before the vehicle reaches it, the presence loop holds the arm, the vehicle clears, the loop releases and the arm closes. Repeat it back to back five times with real vehicles in the queue. The loop detector guide covers device-level diagnosis and the safety sensor guide covers device selection; what commissioning adds is the recorded, repeated sequence test.
Finally, record the settings: detector sensitivity, presence or pulse mode, and any DIP switch or menu position changed on site. Settings changed without a record are the single most common reason a lane "goes wrong" after a service visit.
Integration tests fail quietly. Verify each of these with the integrator present and, where possible, with the vendor's own test card or record:
The LPR integration guide covers wiring and protocol detail, and the ANPR network design guide covers the IT layer; commissioning should end with a signed statement that the interfaces, addresses and trigger modes are as designed.
These are not optional and they are not negotiable against a deadline:
If the site operates the lane before these tests are recorded, the acceptance record is a fiction, and liability for the first incident sits with whoever signed.
Pull the supply — deliberately, with the lane clear and the operator watching — and record exactly what happens in each of the two cases:
Ask for the pack before the final payment, because after it, requests slow down:
Commissioning is much faster when the order was written properly, because factory pre-configuration replaces site work. When you buy direct, specify:
Send the site plan, a photo of the entrance and the integrator's interface list with the enquiry — that is enough for a factory to set the unit up for the destination instead of leaving it in a default state that somebody discovers at commissioning.
A new lane is accepted on evidence, not on impressions. The three artefacts that carry that evidence are a measured baseline (travel times, supply readings, detector settings, cycle count at day zero), a completed safety test with the obstacles and the release actually used, and a documentation pack that the site's own technician can act on.
Everything after that — the service interval, the first fault call, the spare-part order — is a comparison against those numbers. Lanes without a baseline are diagnosed by replacement; lanes with one are diagnosed by measurement, which is cheaper for everyone involved.
If you are specifying or importing barrier gates and want the lane delivered pre-configured for your destination grid, arm length and interface list, send the site details and we will come back with a factory-direct configuration for the project.
| Acceptance stage | What is tested | Evidence to keep |
| Mechanical | Cabinet level, foundation, arm length and side, spring balance, fasteners, drainage | Balance check result, bolt torques, photos |
| Electrical | Supply voltage at rest and under load, protective device and isolation, earthing, cable entries | Readings with instrument, date and measuring point |
| Local control | Buttons, remote, manual release, opening and closing time, limit positions, 20-cycle run | Travel times, cycle counter value at day zero |
| Detection | Presence loop, pulse loop, safety loop, photocells and radar, each device then the full sequence | Detector settings plus five repeated sequence runs |
| Integration | Open, close and stop inputs, relay mode, RS-485 or TCP-IP link, LPR trigger, payment release, anti-passback | Interface list signed by the integrator |
| Safety | Obstacle reversal, photocell during closing, manual release powered and unpowered, emergency opening, signage | Signed safety test using a real test obstacle |
| Handover | Training, keys, manual release location, documentation pack, spares, warranty registration | Document pack list, serial numbers, open snag list |
For one lane with a finished power supply and detection already installed, the mechanical, electrical and detection tests are typically about half a day of site work. Add the integrator's time for access-control, payment or camera testing, and expect one further visit if the site is handed over before the electrical work is complete.
A qualified technician for the mechanical, electrical and safety tests, the access-control integrator for the integration tests, and the site operator present for the safety tests. Site staff should not be asked to test entrapment protection or to adjust the balancing spring.
A digital multimeter, a clamp meter for load current, a stopwatch, a tape measure, a torque wrench for anchor and hub bolts, a test obstacle of representative mass, and the factory wiring diagram for the exact controller delivered.
No. Manual release and emergency opening behaviour are the functions the site relies on during a power failure, and they are among the most common findings in a first service visit. Test them with the drive powered and again with the power off before signing.
Recording the settings nowhere. Detector sensitivity, relay mode and controller parameters changed on site without being written down turn every later fault into a guess. Keep a settings sheet together with the test record.
The wiring diagram and terminal assignment for the delivered controller, the controller manual with fault codes and factory defaults, the factory quality record, warranty terms, equipment serial numbers, and a spare parts list that uses the correct part names.
Send your project details — lane width, traffic volume and site photos — and our engineering team will reply with a specification, lead time and factory-direct quotation.
The lane hardware behind this layout
Factory-direct, from the same line
LPR + ANPR
License Plate Recognition System
Pay stations
Self-Service Payment & Change Machine
Pedestrian lanes
Pedestrian Access Gate