EV Charger EOL Test Jig and Production Tester

Test objectives

Coverage is defined against what can go wrong in assembly, not against what is easy to measure.

• Safety relevant checks including earth continuity and insulation where specified
• Functional checks covering the charging state machine and control pilot behaviour
• Communications verification for the fitted modules
• Metering verification against the required accuracy class
• Identity and configuration confirmation, including firmware version

Fixture interfaces and safe architecture

The fixture presents the unit under test with the electrical and signal conditions it needs, while keeping the operator away from anything hazardous. Any station handling mains potential needs interlocked access, defined isolation and a documented safe state on abort.

Design the abort path first. A tester that cannot fail safely will eventually be bypassed by someone under schedule pressure.

Programming and provisioning integration

Firmware, configuration and identity are applied and then verified in the same controlled sequence. Verification matters: applying a value and confirming it took are different operations, and only the second one is evidence.

Example test sequence

1. 1

   Serial scan

   Unit identity captured and the record opened.

2. 2

   Firmware and version check

   Confirm the intended build is present.

3. 3

   Configuration

   Apply and read back the required parameter set.

4. 4

   Pilot state simulation

   Exercise the control pilot states and confirm correct transitions.

5. 5

   Relay and contactor verification

   Confirm switching behaviour and, where designed for it, welding detection.

6. 6

   Metering verification

   Compare against a reference within the defined tolerance.

7. 7

   Communications check

   Confirm the fitted modules connect and report.

8. 8

   Record and disposition

   Store the full result against the serial number and route the unit to pass or fail.

The order is deliberate. Identity first means every subsequent measurement is attributable, including the ones taken on units that fail.

Limits, calibration and golden units

Test limits come from product characterisation, not from convenience. Limits that are too tight generate false failures and pressure to widen them informally. Limits that are too loose pass marginal units.

A golden unit, verified independently, is run periodically to prove the station itself is still measuring correctly. Without it, station drift and product failure are indistinguishable.

Result storage

Every result is stored against serial number, date and time, operator or station identity, firmware version and configuration applied. Retention period is defined in the quality plan.

This record is what lets you scope a field problem to a date range, a station or a firmware build rather than recalling everything.

Retest and failure codes

Retest rules are defined in advance: what may be retested, how many times, and what must be routed to rework or quarantine instead. Unlimited retesting converts a failing product into a passing record.

A failure code taxonomy turns test data into a defect Pareto. Without codes you know your yield but not your problem.

Fixture maintenance and version compatibility

Fixtures wear. Contacts degrade, cables fatigue, references drift. Maintenance intervals, calibration schedule and spare part holdings are defined at handover.

Test software is versioned against product revisions, so a fixture cannot silently test a new variant with old limits.

What the jig does

Every unit is exercised rather than inspected, and the results are recorded against its serial number.

• Functional test of the charging sequence, using a charger simulator so behaviour is verified without a vehicle
• Electrical and protection checks
• Metering verification
• Connectivity and protocol checks
• Every result written to the product lifecycle management system

In our own operation the PLM system is the S3 suite. What matters is not which tool, it is that the record exists somewhere it can be retrieved from years later.

Why the record is the point

A test that passes and is not recorded proves nothing when an auditor asks.

When a charger fails in the field, its production record is what turns an argument into an analysis. What was tested, what the readings were, which firmware it left with. We can look at that and say whether the unit shipped correctly, which is a different conversation from a replacement decision.

For a company buying chargers rather than building them, the same record is what lets you go back to your supplier with evidence rather than with a complaint.

Built to your product

Jigs are designed to the product and the line rather than supplied from a catalogue. What gets tested comes from what your product does and what your process needs, and our engineers build it with your production and quality teams and stay through the first runs.

Test limits get revised once a line starts producing real data, which is why handing over a sealed box does not work.