Splitting development benches from EOL stations
Development benches are built for measurement accuracy, operating-point coverage and complete data capture; a single run can last tens of hours while mapping efficiency, validating thermal strategy or completing endurance schedules. EOL stations are built for cycle time and consistent judgement; their job is to catch assembly defects, not to re-characterise the product. The two specifications should be written separately.
The measurement chain behind an efficiency map
Efficiency uncertainty comes from the torque transducer, speed measurement, power analyser, temperature compensation and sampling synchronisation acting together. A ±0.1% FS transducer with poor shaft alignment, or sampled out of phase with the electrical measurement, can give an error several times larger than the sensor rating suggests. At acceptance, ask for a system-level uncertainty budget and verify repeatability with a reference motor of known efficiency.
NVH judgement on the line
Order analysis is the core of e-drive NVH judgement. Collect order-amplitude distributions from a few hundred known-good units early in production, set initial limits statistically, then refine them against rework and field-complaint data. Fixed thresholds carried over from a previous programme usually stop working after a gear supplier or batch change.
The data platform
Raw waveforms, judgements, equipment status and calibration validity should all be tied to the serial number and archived long-term. The LANDI LDEV-X platform logs up to 1000 channels at 1 kHz with automatic report generation, and together with the LD-LMS laboratory management system covers the path from execution to archiving and retrieval.