The terms multiply, they do not simply add
Mechanical power is torque times angular velocity and efficiency is mechanical over electrical power, so the relative error of each term propagates into the result. In the high-efficiency region — say 96% — the difference between input and output is small, so measurement error is heavily amplified: a 1% power error can shift the loss estimate by more than 25%.
The three most overlooked factors
First, alignment: a misaligned coupling puts bending moments into the torque transducer and produces a speed-dependent systematic error. Second, synchronisation: mechanical and electrical quantities must be sampled in the same time window, otherwise the pairs are meaningless under dynamic conditions. Third, temperature: both torque transducers and current transducers drift, and ambient change over a long run shows up as slow drift in the data.
What to do about it
Ask for a system-level uncertainty budget rather than a list of sensor specifications; verify repeatability with a reference motor of known efficiency; keep torque transducers and power analysers on a calibration cycle; and record ambient temperature and coolant conditions in the report. These do more for the result than buying a higher sensor class.
Need a configuration and quotation for your specific unit? Our test engineers reply within one working day.
Talk to an Engineer