Walk through almost any plant that has been running for twenty years and a substantial share of the motors will be considerably larger than the machines they drive require. Nobody decided this. It accumulated, one safe choice at a time, and it costs money every hour.
How it happens
The mechanisms are all reasonable in isolation.
A design engineer calculates the required power, adds a margin for uncertainty, and rounds up to the next standard frame. A second margin gets added by the machine builder. A third gets added at installation because a slightly larger motor was available.
A motor fails and the replacement selected is the next size up, because the failure was blamed on the motor being too small even though the real cause was ventilation or unbalance.
A process changes and the load falls, but nobody revisits the motor. A line that once ran at full capacity now runs at 60 percent, and the drives were sized for the original throughput.
A spare from the store is fitted temporarily and becomes permanent.
What it actually costs
Motor efficiency stays reasonably flat from about 100 percent down to 50 percent of rated load. Below 50 percent it falls, and below about 35 percent it falls sharply, because the fixed losses, iron losses and friction and windage, remain constant while the useful output shrinks.
A motor running at 25 percent load may be operating ten or more percentage points below its rated efficiency. That is a far larger penalty than the difference between two efficiency classes, which puts the whole class upgrade discussion in perspective.
Power factor degrades much faster than efficiency. Magnetising current is essentially constant regardless of load, so at light load it dominates the total current. A motor with a rated power factor of 0.85 can fall to 0.5 or below at 25 percent load.
Poor power factor means higher current for the same useful power, which means higher losses in every cable and transformer feeding it, and on many industrial tariffs a direct financial penalty.
The costs that are not electricity
An oversized motor costs more to buy, and the difference compounds through the whole installation. A larger motor needs a larger frame, a larger starter, larger cable, larger protection and more space.
Starting current scales with motor size, so an oversized machine imposes a larger disturbance on the supply every time it starts, which can be the deciding factor in whether soft starting is needed at all.
There is a mechanical cost too. An oversized motor can deliver far more torque than the driven machine was designed for, so a jam produces a broken shaft, a sheared key or a destroyed gearbox rather than a stalled motor and a tripped overload.
Finding the oversized motors
The measurement is simple and needs only a clamp meter. Measure the running current under normal production conditions and compare it with the nameplate current.
The complication is that current does not fall proportionally with load, because magnetising current continues to flow. A motor at 50 percent load typically draws around 65 to 70 percent of rated current, not 50 percent. So a motor drawing 60 percent of rated current is running at roughly 40 percent load, and a motor drawing 50 percent of rated current is very lightly loaded indeed.
For a more accurate picture, measure actual power with a power meter rather than inferring from current, since that removes the power factor effect. Many frequency converters already display output power and torque as a percentage of rated, which makes the assessment free on any drive that has one.
The slip method works on direct online motors and needs no electrical measurement at all. Measure actual shaft speed with a tachometer, compare it with synchronous speed and rated speed from the plate, and the ratio of actual slip to rated slip approximates the load factor.
Whichever method is used, measure over a representative period rather than at one moment. A batch process may look lightly loaded for most of a cycle and fully loaded for a short peak, and the peak is what determines the required rating.
What to do about it
Do not replace a lightly loaded motor that is running reliably unless the numbers justify it. Replacement costs money and creates risk, and the saving from correcting a motor running at 45 percent load is modest.
Below about 35 percent load on a motor running long hours, the case becomes clear. Replacing it with a correctly sized machine of a current efficiency class delivers two improvements at once, correct loading and a better class, and the combined saving is usually well worth the work.
At failure, always reassess the size rather than ordering the same rating automatically. This is the cheapest possible moment to correct the problem, because the labour is being spent anyway and only the incremental price of a different rating is at stake.
Where the load genuinely varies widely, a frequency converter is a better answer than a smaller motor, because it reduces the magnetising current at light load as well as matching the output.
Where oversizing is deliberate and correct
Some margin is justified and should not be stripped out.
Loads with high starting torque or high inertia need a motor sized for the start rather than the run, and that motor will appear lightly loaded in normal operation.
Applications with genuine future expansion plans may reasonably carry capacity.
Motors operating above 40 degrees ambient or above 1000 metres altitude must be derated, so an apparently oversized machine may be correctly sized for its conditions.
Critical drives where a failure stops production sometimes justify margin as insurance rather than as engineering.
The distinction worth making is between margin that someone chose for a reason and margin that accumulated because nobody checked. The second kind is where the savings are.
A practical starting point
List every motor above about 7.5 kilowatts that runs more than 4000 hours a year. Measure the load on each. Anything below 40 percent goes on a list for replacement at next failure, and anything below 30 percent running continuously is worth evaluating for immediate replacement.
That exercise typically takes a day for a medium sized plant and identifies more saving than most formal energy audits.



