Skip to content
Tutorials Center
My learning

Topics

Lesson 37 of 45 ยท Efficiency, Testing and Applications

Losses and efficiency

Learn the copper, core, mechanical and stray losses of electrical machines, calculate efficiency from a power-flow example, and find the load at which efficiency peaks.

Energy cannot be destroyed, but in a machine some of it ends up as heat that nobody wanted. Those are the losses. They matter for three reasons: they cost money in electricity, they heat the machine and shorten the life of the insulation, and they set the size of the machine, because it has to be big enough to get rid of the heat.

This lesson names each loss, shows where it comes from in the equivalent circuit, and uses it to calculate efficiency. You will also see a classic result: a machine is most efficient when its fixed and variable losses are equal.

What you will learn

  • The four main loss types and how each one depends on load and speed.
  • How to calculate efficiency from the losses.
  • How to follow the power through an induction motor from input to shaft.
  • Where the maximum efficiency occurs and why part-load operation matters.

Key terms

Copper loss: IยฒR heating in the windings; it grows with the square of the current.

Core (iron) loss: hysteresis and eddy-current loss in the magnetic steel; set by voltage and frequency, nearly the same at any load.

Friction and windage: mechanical loss in bearings and from air drag and the cooling fan.

Stray load loss: small extra losses caused by leakage flux, harmonics and eddy currents that appear under load.

Efficiency ฮท: Pout / Pin.

The loss types

Copper loss is 3 Iยฒ R in each winding. Resistance rises with temperature: for copper R2 = R1 (234.5 + T2) / (234.5 + T1), which is why losses are always quoted at a reference temperature. Because the current changes with load, copper loss is a variable loss, proportional to the square of the load fraction.

Core loss has two parts. Hysteresis loss comes from the magnetic domains flipping each cycle and is proportional to the frequency and a power of the peak flux density. Eddy-current loss comes from circulating currents in the steel and goes with fยฒBยฒtยฒ, where t is the lamination thickness. That is why cores are built from thin insulated sheets. Both are set by the flux, which is set by voltage and frequency, so the core loss is nearly constant from no load to full load. In a transformer or an induction motor it is a fixed loss.

Mechanical loss includes bearing friction, brush friction (where present) and windage, the drag on the rotor and the fan. It is nearly constant at constant speed, though windage rises steeply with speed.

Stray load loss is a small catch-all, often assumed to be 0.5 % to 2 % of the output depending on the standard and the machine size.

Efficiency

Text
ฮท = Pout / Pin = Pout / (Pout + losses) = 1 โˆ’ losses / Pin

Large machines are more efficient than small ones because their losses grow more slowly than their rating. A 1 MW motor can exceed 96 %; a 100 W motor may be below 60 %.

Power flow in an induction motor

Revisit the motor of lesson 2 at s = 0.03, and add a core loss of 250 W, a friction and windage loss of 180 W and a stray loss of 0.5 % of the mechanical power. The program in this lesson gives the full waterfall:

Step Power
Input Pin (with core loss) 14 631 W
Stator copper loss 549 W
Core loss 250 W
Air-gap power Pag 13 832 W
Rotor copper loss s Pag 415 W
Mechanical power (1 โˆ’ s) Pag 13 417 W
Friction and windage 180 W
Stray load loss 67 W
Shaft output 13 170 W

The efficiency is 13 170 / 14 631 = 90.0 %, and the shaft torque is 13 170 / (2ฯ€ ร— 1455 / 60) = 86.4 Nยทm, a little less than the 88.1 Nยทm of the air-gap torque. The difference is the friction and stray torque.

Maximum efficiency

Write the load as a fraction x of the full load. The output is x S cos ฯ†, the fixed loss is Pc, and the variable loss is xยฒ Pcu, where Pcu is the copper loss at full load:

Text
ฮท(x) = x S cos ฯ† / ( x S cos ฯ† + Pc + xยฒ Pcu )

Setting the derivative to zero gives the condition for maximum efficiency: the variable loss equals the fixed loss, so

Text
x = โˆš( Pc / Pcu )
Two plots against load: fixed and variable losses cross at 0.71 of full load, where the efficiency curve peaks
Maximum efficiency happens when variable loss equals fixed loss.

A 100 kVA transformer with an iron loss of 0.6 kW and a full-load copper loss of 1.8 kW reaches its maximum efficiency at x = โˆš(0.6 / 1.8) = 0.577, about 58 % of full load. At 0.8 power factor it has an efficiency of 97.5 % there, against 97.1 % at full load and 96.6 % at quarter load. The curve is flat near the top, which is good news: a machine loaded between 50 % and 100 % is within a fraction of a percent of its best.

Utilities design distribution transformers (which are energized all day but lightly loaded much of the time) with low core loss and a peak efficiency near the average load. The same logic applies to all-day efficiency, calculated from energy instead of power over 24 hours.

Motors have efficiency classes IE1 to IE4 in the IEC 60034-30-1 standard, and many countries require premium efficiency (IE3 or better) for new motors. An oversized motor running at 25 % load has poor efficiency and a poor power factor; matching the size to the load is the cheapest efficiency measure.

Common misconceptions

"Efficiency is highest at full load." It usually peaks at 50 % to 80 % of rated load, where variable loss equals fixed loss.

"Core loss rises with the load." It follows the voltage and frequency; the load changes the copper loss.

"A high rated efficiency guarantees low running cost." Hours of operation, load factor and part-load behavior decide the energy bill.

Try it

Compute the full power waterfall for the induction motor and check that every watt is accounted for.

Python Power flow in an induction motor

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

Now the transformer: find the load of maximum efficiency and compare several power factors.

Python Transformer efficiency against load

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

Check your understanding

Quick check ยท 5 questions

of 5 answered You scored / 5 ยท perfect!

  1. 1. Which loss is nearly independent of the load of a transformer?

    Core loss depends on the flux, which is set by the voltage and frequency, so it is nearly constant from no load to full load.
  2. 2. A transformer has an iron loss of 500 W and a full-load copper loss of 2000 W. At what fraction of full load is its efficiency highest?

    x = โˆš(Pc / Pcu) = โˆš(500 / 2000) = 0.5.
  3. 3. A motor takes 10 kW and delivers 9 kW. Its efficiency and total loss are

    ฮท = 9 / 10 = 90 %, and the loss is 10 โˆ’ 9 = 1 kW.
  4. 4. In an induction motor with air-gap power 10 kW and slip 0.04, the rotor copper loss is

    Pcu2 = s ร— Pag = 0.04 ร— 10 kW = 0.4 kW.
  5. 5. Why are transformer cores built from thin insulated laminations?

    Thin laminations break up the paths for eddy currents, which go with the square of the thickness.

Summary

  • Losses: copper (variable, โˆ loadยฒ), core (fixed), mechanical (nearly fixed) and stray load.
  • ฮท = Pout / (Pout + losses); in an induction motor Pcu2 = s ยท Pag and Pmech = (1 โˆ’ s) ยท Pag.
  • Maximum efficiency occurs when variable loss equals fixed loss: x = โˆš(Pc / Pcu).
  • Right-sizing and avoiding long part-load operation save more energy than small differences in rated efficiency.

Next: Testing motors and transformers