The most expensive functional test is usually the one that measures the fewest useful things. A line tester with twelve positions and a limit copied from a datasheet costs more per unit than a tester with four positions and limits taken from characterisation data, and it catches less. The cost of the test is not the equipment; it is what the equipment is set up to decide.
Charger functional test: the cost lines behind one line test
Four cost lines: the fixture and its limits, the positions per unit, the data record, and the rework loop.
Only the first is a purchase. The second is a decision about what to measure on every unit, the third decides whether the result can be traced afterwards, and the fourth is what happens when a unit fails and nobody knows which limit it failed.
Reading the four lines separately is what allows two quotations to be compared, because suppliers price the same total from very different assumptions about positions and records.
Cost line 1: the fixture and its limits
The fixture is priced once and the limits are what make it useful.
A tester that measures output voltage and current under load, verifies the negotiation with a reference device and records the result costs more to build than a power-on check, and it costs less per unit over a production run because it replaces manual judgement. The limits loaded into it come from the characterisation data on the design: a limit taken from a component datasheet describes the component, not the finished product.
Cost line 2: the positions per unit
Every port, profile and plug variant can become its own position.
A single-port fixed-output charger has one condition to verify. A three-port product with an adjustable range and a sharing rule has a matrix, and testing the entire matrix on every unit is rarely affordable or necessary. The efficient answer is to test the conditions that a unit can actually differ on from line to line, and to verify the remaining combinations during characterisation and on a defined sample.
Cost line 3: the data and the traceability record
A test that stores nothing protects the shipment and not the brand.
Storing the measured value per unit against its limit, with the serial or lot, costs almost nothing once the tester is connected to a record, and it changes what a later complaint can be answered with. Where the tester flashes pass or fail, the only available response to a field failure is replacement, and the defect pattern cannot be found because the values were discarded at the station.
Cost line 4: the rework loop
The fourth line is the one that grows quietly.
Units that fail a limit are reworked, and rework has a cost per unit plus a risk of latent damage. Where the failure is caused by a drifting process, the rework volume rises without any change to the design, and the line absorbs the cost as normal. Reducing this line means acting on the cause of the failures rather than on the failed units, which is a process decision rather than a test decision.
| Cost line | What it covers | Driven by | What reduces it |
|---|---|---|---|
| Fixture and limits | Measurement capability and the limits loaded | What the test has to decide | Limits taken from characterisation data |
| Positions per unit | The conditions tested on every unit | Ports, profiles and sharing rules | Testing what varies in production |
| Data record | Stored results against the limits | The traceability requirement | Connecting the tester to a record |
| Rework loop | Units that failed and were corrected | A drifting process | Acting on the cause, not the unit |
Where the money is wasted
Four patterns, and they are visible in the test specification.
A limit with no source; positions repeated for conditions that cannot vary between units; a pass or fail result with no stored value; and rework recorded as a quantity rather than as a cause. Each adds cost per unit while reducing what the brand can prove about the shipment, which is the opposite of what a buyer pays a line test to do.
What to ask for in the test specification
Five items, and each one is a sentence rather than a purchase order.
The parameters measured on every unit and the limit for each; the source of each limit; the conditions verified on a sample rather than on every unit; the record produced per unit and how long it is kept; and the disposition of a failed unit. With those five, the test has an owner, a purpose and an audit trail, and the quotation can be compared with another supplier’s line by line.
FAQ
Should every unit be functionally tested?
For a charger, the parameters that decide safety and the delivered behaviour are normally checked on every unit, because the cost per test is small compared with a field failure. Conditions that cannot vary between units belong to design validation and sample verification instead. The parameters that decide safety and the delivered behaviour are normally checked on every unit, because the cost per test is small next to a field failure.
Can a supplier test every port combination on every unit?
They can, and it multiplies the cost without adding information about that unit. The efficient split is to verify the conditions production can vary on every unit, and to check the full matrix during characterisation and on a defined sample per build. Verify the conditions production can vary on every unit and check the full matrix during characterisation and on a defined sample per build.
What is the point of storing results if the unit passed?
A stored value shows the distribution, which is what lets a drift be seen before failures begin. A pass or fail flag removes that information, and it also removes the ability to answer a later complaint about a specific lot. A stored value shows the distribution, which is what lets a drift be seen before failures begin, while a pass or fail flag removes that information.
How do we compare two suppliers’ test costs?
Compare the five specification items rather than the fee: parameters, limits and their source, sampled conditions, the record, and the failure disposition. Two totals that assume different scopes are not comparable, and the difference usually sits in the record and the rework. Compare the parameters, the limits and their source, the sampled conditions, the record and the failure disposition, because two totals that assume different scopes are not comparable.
Next step
Send the port configuration, the claimed profiles and the market requirements to Wecent Group through the contact page. The reply states the parameters tested on every unit, the limits and their source, the record produced and what is verified on a sample. Related reading: our charger factory, charger manufacturing services and the factory floor. External references: USB Power Delivery specification, IEC standards, IEC Webstore, UL standards and certification, TUV testing services and Intertek testing services.
{“@context”:”https://schema.org”,”@type”:”FAQPage”,”mainEntity”:[{“@type”:”Question”,”name”:”Should every unit be functionally tested?”,”acceptedAnswer”:{“@type”:”Answer”,”text”:”For a charger, the parameters that decide safety and the delivered behaviour are normally checked on every unit, because the cost per test is small compared with a field failure. Conditions that cannot vary between units belong to design validation and sample verification instead.”}},{“@type”:”Question”,”name”:”Can a supplier test every port combination on every unit?”,”acceptedAnswer”:{“@type”:”Answer”,”text”:”They can, and it multiplies the cost without adding information about that unit. The efficient split is to verify the conditions production can vary on every unit, and to check the full matrix during characterisation and on a defined sample per build.”}},{“@type”:”Question”,”name”:”What is the point of storing results if the unit passed?”,”acceptedAnswer”:{“@type”:”Answer”,”text”:”A stored value shows the distribution, which is what lets a drift be seen before failures begin. A pass or fail flag removes that information, and it also removes the ability to answer a later complaint about a specific lot.”}},{“@type”:”Question”,”name”:”How do we compare two suppliers’ test costs?”,”acceptedAnswer”:{“@type”:”Answer”,”text”:”Compare the five specification items rather than the fee: parameters, limits and their source, sampled conditions, the record, and the failure disposition. Two totals that assume different scopes are not comparable, and the difference usually sits in the record and the rework.”}}]}