Bench Testing Radar Detectors Before Shipment

Why the bench, not the box

A carton of radar detectors looks finished from the outside. What matters is whether each unit inside does what the specification says: powers up, receives, alerts, displays, and survives handling. The checks that answer this happen on a bench before packing, and they are worth understanding from the buyer's side, because you are the one who requests them.

Power-up check

The first functional gate is power: the unit starts from its intended supply, the display initializes, and startup behavior matches the manual. Confirm whether this check runs on every unit or on a sample, and confirm the power supply used, because a unit tested only on a bench supply may behave differently on a vehicle socket with real-world voltage. Record the check as a pass line per unit or per sample in the test record.

Reception check

The reception check confirms the unit receives in the bands it claims. On a bench this is done with a signal source at a set level, not with roadside signals, which is exactly why it is meaningful: it is repeatable. Ask which bands are tested, at what signal level, and against what reference. Where the supplier quotes a sensitivity figure, this is the test behind it, and a figure without this test is a sentence without a source.

Alert and display check

Confirm the alert path: the display shows the right band indicator, the audio sounds at the right volume, and the mute works. Cycle the modes the manual names and confirm each behaves as described. This check is fast per unit, which is why there is little excuse for skipping it; it catches the firmware or assembly faults that customers would otherwise catch for you, at a higher price.

Controls and mechanicals

Buttons, mounts, and cable connections fail in ways a powered test misses. A mechanical check presses every button, fits the mount, and inserts the plug. Confirm it exists in the plan; it is the cheapest line in the whole inspection and prevents a class of returns that no firmware fix will.

Final inspection

The last gate before packing is a visual and completeness inspection: condition, labels, firmware version, and accessory count against the packing list. Sampling plans and acceptance limits are agreed in writing before production starts. The final inspection report lists sample size, findings by class, and the pass or fail decision. Ask for it with the shipment; it is the document that makes the rest of the paperwork coherent.

The records

Each stage leaves a record: incoming check sheets, the line log, the functional test record, and the final inspection report. Short, dated records that name the SKU are worth more than a thick quality manual nobody opens. Keep the records with the order file; they are what you compare the next run against, and what a dispute is settled with.

What to ask for, in one message

• The inspection plan for your order, before production starts.
• Whether testing is per unit or per sample, for each stage.
• The firmware version recorded per run.
• The final inspection report with the shipment.

None of these requests is unusual, and a factory that resists them is telling you where the risk sits. The bench is where quality is cheap; every stage after shipment multiplies the cost of the same fault.

The golden sample on the bench

Every bench check is a comparison, and comparisons need a reference. The approved sample, kept sealed and dated, is that reference: the inspector powers a production unit and places it beside the golden sample, compares startup behavior, display output and build finish, and records anything that differs. Ask whether the factory retains its own golden sample per run and where it is stored; a factory that cannot produce its reference has told you how its line discipline works. Where a firmware revision is involved, the golden sample should carry the same version as the run, and the version should be stated on the test record.

Statistical honesty in sampling

Sampling plans carry an idea buyers should understand before signing one: a sample pass is a confidence statement, not a guarantee. The plan's level decides how many units are opened and how many findings are tolerated, and tighter levels cost more in labor and in units consumed by destructive checks. Choose the level by consequence: a fault that reaches customers costs far more than the inspection hours that might catch it, so the level should follow the severity, not the budget. Write the level, the classes and the acceptance numbers into the order, and change them only in writing. A sampling plan agreed verbally is a sampling plan renegotiated at the worst moment, which is the day before the vessel sails.

What the buyer does with the results

Test records age into one of two assets. Filed and reviewed, they become the trend line that tells you whether the factory is stable: same checks, same results, run after run. Ignored, they become paper produced for the file. The review habit is one page per order: results versus the previous run, anything drifted, anything improved. Two orders in, the page starts answering questions no quote can: which supplier's records are worth the paper, and which factory's stability deserves the next larger order.

Testing changes between runs

The bench plan agreed for the first order tends to be re-signed without thought for the second, and that is where the gap opens: the second run carries a firmware revision, a swapped component, or a new packing lot that the original plan never tested. Add one line to every repeat order: confirm the test plan against the current bill of materials and firmware, and name anything new the plan must cover. The line takes one message and keeps the inspection as alive as the product it watches. Inspection plans are for products as they are, not as they were when the relationship started.