Qualification answers one question with evidence: does this bench give the right verdict on known parts, repeatedly, in its real environment?<\/strong><\/p>\n Our own process runs from specification, architecture and design review to manufacturing, FAT, then delivery with SAT and training, and finally maintenance. The steps below follow that order.<\/p>\n Qualification starts on paper. Before building, agree with the customer:<\/p>\n Without written criteria, FAT becomes a demonstration instead of a test.<\/p>\n FAT takes place at the bench builder’s site, before shipping. The customer attends or receives the evidence. Typical content:<\/p>\n Every deviation is recorded, corrected and re-tested before shipping.<\/p>\n SAT repeats the key checks once the bench is installed in its real environment: local mains, the production network, the real operators and often the real products. Transport can loosen connections; the site may have different grounding or noise. SAT typically includes:<\/p>\n SAT ends with a signed acceptance report. From that point the bench can be used to judge products.<\/p>\n Every channel from the instrument terminal to the contact touching the product must be verified against the wiring list:<\/p>\n A practical method is a verification adapter<\/strong>: a dummy connector that loops or loads each pin in a known pattern, so the bench can check its own wiring automatically.<\/p>\n Example, illustrative values.<\/em> The bench measures a 1.00 \u03a9 shunt in the product. The bench harness adds 2 \u00d7 0.15 \u03a9 of wire and contact resistance.<\/p>\n FAT should include at least one low-resistance reference to prove that the measurement method, not only the wiring, is correct.<\/p>\n Example.<\/em> The bench supplies a unit drawing 5 A through 3 m of wire (6 m loop) with 0.0133 \u03a9\/m (about 1.3 mm\u00b2 copper, example). Drop = 5 \u00d7 6 \u00d7 0.0133 \u2248 0.40 V. A 12.0 V setting becomes about 11.6 V at the product. Either sense at the product or account for the drop in the specification.<\/p>\n Known good products, characterised independently. The bench must pass them, repeatedly.<\/p>\n Parts or simulators with a known defect for each critical test: an open, a short, a value just outside the limit. The bench must fail each one on the right step, with the right message. Without bad parts, you prove only that the bench can say PASS.<\/p>\n Run the same reference units several times, and by several operators if the process allows it. A gauge repeatability and reproducibility study (gauge R&R) is a common way to quantify this; the acceptance threshold is set in the bench specification. A widely used reference (AIAG MSA manual): below 10 % of the tolerance or process variation, acceptable; 10 to 30 %, may be acceptable depending on the application; above 30 %, not acceptable. The number of distinct categories (ndc \u2265 5) is often checked as well.<\/p>\n Qualification links directly to the acceptance logic explained in ATP vs QTP<\/a> and our ATP methodology<\/a>. Two parts of a bench deserve their own checks: harness continuity and insulation testing<\/a> and acquisition timing<\/a>. Software traceability is covered in test automation: from script to sequencer<\/a>.<\/p>\n External reference:<\/strong> a general overview of factory acceptance testing<\/a>.<\/p>\n RAYNOV delivers custom validation systems<\/a> through FAT, SAT and training, and maintains benches after delivery.<\/p>\n2. The sequence: design review, FAT, SAT<\/h2>\n
![\"Test](\"https:\/\/raynovtechnology.com\/wp-content\/uploads\/2026\/10\/raynov-fat-sat-qualification-flow.png\" "\\"Qualifying")
2.1 Before FAT: the test specification of the bench<\/h3>\n
\n
2.2 FAT: Factory Acceptance Test, at the builder<\/h3>\n
\n
2.3 SAT: Site Acceptance Test, on site<\/h3>\n
\n
3. Channel verification<\/h2>\n
3.1 Pinout and continuity of the bench harness<\/h3>\n
\n
3.2 Worked example: harness resistance in a 4-wire versus 2-wire measurement<\/h3>\n
\n
3.3 Worked example: voltage drop on a supply line<\/h3>\n
4. Instruments and calibration<\/h2>\n
\n
5. Reference units: golden and bad parts<\/h2>\n
5.1 Golden units<\/h3>\n
5.2 Known bad units (limit samples)<\/h3>\n
5.3 Repeatability<\/h3>\n
6. Documentation delivered with the bench<\/h2>\n
\n
7. Common mistakes<\/h2>\n
\n
8. Qualification checklist<\/h2>\n
\n
9. Going further<\/h2>\n