EV Charging Gun Lock Board: Motor and Position Feedback Testing
A charging gun lock board has to do two things at once: drive a geared motor between the locked and unlocked positions, and report back that the mechanism actually arrived. A board that powers up, accepts its programme and lights a status indicator has demonstrated none of that. The verification that matters is a loaded test in which the controller works against the real actuator, and it belongs in the assembly flow rather than in the customer’s incoming inspection.
Fixing the Actuator Before Any Measurement
Every current reading depends on the load. Two motors of the same model differ in coil resistance and gearbox friction, and a mechanism that sits slightly misaligned on a fixture draws a different current from one that is bolted down as it will be in the vehicle. Unless the actuator is fixed for the whole batch, the numbers recorded on one board cannot be compared with the numbers recorded on the next, and a fault that develops across the order stays invisible.
That is why the import stage begins with documents rather than with the line. Fabrication data, the bill of materials, the placement coordinates, the assembly drawing, the firmware, the motor parameters, the connector definitions, the feedback logic and the acceptance criteria all have to be present and mutually consistent before a test can be written. Where one board design is used with several lock variants, the batch record has to state which motor, gearbox and harness version this order uses, because the fixture built from the wrong definition produces a batch of results that mean nothing.
The motor drive section is reviewed in the same pass: the driver device, the protection components, the current sense resistor, the connectors and the feedback input. A substitute part is accepted on the basis of its model, its key parameters and the reference designators it may be used for, never on the basis of a matching footprint alone.
The First Power-Up After Assembly
Soldering is verified before any power is applied. The driver, the diodes, the capacitors and the connectors are checked for orientation, and the joints at the motor header are inspected for insufficient solder, disturbed joints and lead height. A current-limited supply then performs the first power-up with the actuator disconnected.
An abnormal quiescent current points to a shorted rail, a reversed device or a wrong part long before the programme is loaded, and finding it at that moment costs almost nothing. Once the rails are stable the firmware is written, with the hardware and software revisions that belong together, and with the checksum or address requirements the customer has specified.
<img src="https://www.gopcba.com/wp-content/uploads/2026/08/HDI-PCB-design.jpg" alt="charging gun lock board with motor drive circuit under test” />
The no-load stage confirms the driver outputs and the feedback inputs, and it cannot replace the run against the mechanism. The actuator is connected only after those outputs have been shown to be correct, so that a wiring or soldering fault is not passed on to a motor that is expensive to replace and slow to obtain.
Reading Stall Current With the Waveform
Current rises as the mechanism loads, and when the lock reaches its mechanical end stop or jams, the rise is abrupt. The stall current threshold, the time it is allowed to persist and the way the firmware responds are all defined by the customer’s logic and test requirement rather than by a generic figure taken from a datasheet.
If the motor trips its protection the instant it starts, the harness orientation, the mechanism state, the motor parameters and the driver output are examined in that order. If the motor turns but draws clearly more than a known-good board, a confirmed actuator is substituted so that the fault can be attributed either to the control board or to the mechanical load. Changing one condition at a time is slower on the bench and much faster across a production run; replacing the programme, the motor and the harness together can make a symptom disappear without identifying its cause, and the next order inherits it.
The driver signal, the current sense waveform and the protection action are observed in sequence rather than as isolated readings. A protection event that occurs before the current has had time to rise, for example, is an electrical or configuration problem rather than a mechanical one, and the waveform says which.
Isolating a Position Feedback Fault
The position feedback path is verified in both directions, because a lock that reports correctly when it opens may report nothing when it closes. A motor that reaches its end of travel while the board reports no change may have a faulty limit switch, an unpowered sensor, a damaged harness or a joint that was never properly formed.
The input is tested with a simulated signal first. If the board responds to the simulation but not to the mechanism, the harness or the sensor is the likely cause; if it does not respond to the simulation, the fault lies on the board or in the configuration. A reversed direction of travel is not automatically a software error either, since the motor phase order, the connector definition and the assembly orientation all have to be checked before the programme is blamed.
Which logic the lock should follow in each state is a matter for the customer’s interface specification. What the assembly operation confirms is that the board reads the states that specification defines and drives the outputs it requires, under the load the product will actually see.
Re-Testing After Rework
Where a driver, a sense resistor or a connector joint has been reworked, the heating is kept local and the board goes back through the entire sequence afterwards: quiescent current, both directions of travel, protection at the end stop, unlock feedback and lock feedback. Confirming that the motor turns again is not a re-test, and the second fault found after a repair is usually the one that was introduced by the repair.
Protection for transport is part of the same discipline. Connectors and tall components take the load when a carton is compressed, and a terminal bent in transit fails at the customer’s line rather than on the bench, where it would have been caught in seconds.
Why the Loaded Test Belongs in the Assembly Flow
A lock board is a small assembly with a disproportionate amount of function behind it. The cost of testing it against a real actuator is measured in seconds of operator time per board; the cost of finding the same fault in a finished charging station, after the unit has been assembled, sealed and shipped, is measured in service calls and in the trust of the customer.
Testing at the end of assembly also keeps the information together. The board revision, the firmware and the measured behaviour are recorded while the fixture is still set up and the operator still has the unit in hand, so a question raised weeks later can be answered from the record instead of from a new investigation.
What the Delivery Record Should Carry
The hardware revision, the programme revision and the test status are recorded against the batch where the order calls for it, and the test conditions are recorded with them: the actuator used, the supply voltage and the limit values that were applied. Without those conditions the numbers cannot be interpreted later, and an assertion about a field failure cannot be checked against the data released with the boards.
Our work on this type of product runs through SMT assembly, the pin-in-hole operations through through-hole assembly, the material through component procurement and the electrical verification through PCBA testing. An order is quoted from the production files, the firmware, the motor and harness definition, the test conditions, the quantity and the required date, because the actuator and the test time are what decide whether the programme can be run exactly as specified.
FAQ
Why is a no-load test not enough? A board can drive a disconnected output correctly and still fail to move a real mechanism, because the load, the end stop and the feedback state are all absent from the no-load condition.
What has to be fixed before testing starts? The actuator, the supply, the firmware revision and the acceptance limits. If any of them changes between boards, the records from those boards cannot be compared.
How is a direction error classified? It is a join investigation of the phase order, the connector definition and the assembly orientation, and it is only a software question once those three have been checked.
