Material Insights · Understanding the Material World

Force, motion and energy: the laws every machine obeys: Efficiency and losses: putting the laws to work

All courses / Understanding Engineering / Module 3: Force, motion and energy: the laws every machine obeys Part 5 of 5

Efficiency and losses: putting the laws to work

Every machine is a chain of conversions, and each link keeps only a fraction of the energy it receives. This part multiplies efficiencies along an electric locomotive, lists where real machines lose energy and which of those losses can be cut, and finishes with a capstone estimate that uses force, power, energy and friction together to ask how long a heavy train takes to reach speed.

5.1 Efficiency multiplies along a chain, so every link counts and the weakest costs most

Efficiency is useful output divided by input, for the same kind of quantity and the same time. A value of 0.9, or 90 per cent, means a tenth of the input ends up somewhere other than where you wanted it, almost always as heat. Machines are rarely a single conversion. Fuel or electricity passes through several links, and the useful output of one link is the input of the next. The overall efficiency is the product of the link efficiencies.

An electric locomotive gives a clean example. Electricity from the overhead wire passes through a transformer and power converter, then the traction motors, then the gearing to the axles. Suppose these are 97, 93 and 98 per cent. The product is about 88 per cent, so of each 100 kWh drawn from the wire, about 88 kWh reaches the wheel rim, and 12 kWh warms the equipment. Those values are typical orders of magnitude rather than the data of a given locomotive, and the point is the method. If a link falls from 98 to 90 per cent, the chain drops from 88 to 81 per cent. The weakest link has the most room to improve, so that is usually where effort pays.

The figure also says what the losses are. They show up as heat, in windings, in switching devices, in gear teeth, in oil. Removing that heat forces the designer to add cooling fans, radiators and bulk, so the cost of a loss is both wasted energy and the equipment to get rid of it.

Process flowchart From the overhead wire to the wheel of an electric locomotive (typical values)
Inputs Electricity from the overhead line100 kWh Process steps 1 Transformer and converter

about 97 per cent efficient; about 3 kWh lost as heat

2 Traction motors

about 93 per cent efficient; about 7 kWh lost in windings and iron

3 Gearing to the axles

about 98 per cent efficient; about 2 kWh lost in gear teeth and oil

Outputs Work at the wheel rimabout 88 kWh, spent on pulling against rolling and air drag, hills and acceleration

Product of the three links: 0.97 x 0.93 x 0.98 = 0.884 (calculated); losses 3 + 6.8 + 1.8 kWh. These are assumed round values for illustration, not measurements of a particular locomotive or figures from a cited source.

  • Bloomfield, How Things Work (Wiley, 2015) · ch. 2, 'The Laws of Motion, Part 2'

5.2 Losses come from a short list of causes, and each one points to a design remedy

A handful of causes cover nearly all losses. Friction in sliding contacts turns work into heat; the remedy is rolling instead of sliding, smooth surfaces and lubricant. Rolling resistance from the flexing of tyres is reduced by harder tyres and smoother roads, at a cost in comfort and grip. Air drag grows with the square of speed, so it dominates at motorway speeds and can be cut by shape. Electrical resistance in wires and windings heats them in proportion to the current squared, which module 8 explains. Finally there is the heat that thermal engines must reject by their nature, set by physics rather than design.

Two patterns recur. The first is that losses scale differently with speed. Rolling resistance is nearly constant in force, so its power rises in proportion to speed; air drag force rises with the speed squared, so its power rises as the cube. With the numbers of part 2, a bicycle at 10 km/h spends about four fifths of its effort on rolling resistance, and at 30 km/h about seven tenths on air. The second is that stop-start use wastes energy in a way steady cruising does not. Every time a vehicle brakes, the energy that took fuel to build up is turned into brake heat.

The remedy is to recover a part of the motion energy: regenerative braking, or a flywheel or battery store, as in the Regenerative braking and motion recovery and the Flywheels and small mechanical stores entries. Each recovers some, never all, because every conversion has its own loss, and the chain of braking into storage and then back to traction multiplies three efficiencies together.

Improving efficiency always has a price: a better bearing costs more, a smoother tyre grips less in rain, a lighter vehicle is less robust. The sensible target is usually the largest loss on the audit.

Data Typical efficiencies of links in everyday machines (approximate, round values)
LinkTypical efficiencyWhere the rest goes
Bicycle chain drive (laboratory best case about 98.6 per cent)95 to 98 per centHeat in the chain and sprockets
Gear stage in a gearbox97 to 99 per centHeat in the teeth and oil
Large electric motor90 to 96 per centHeat in the windings and iron core
Small household motor, such as a fan40 to 70 per centHeat in the windings; air-moving design matters more
Human muscle, converting food to workabout 25 per centBody heat
Petrol car engine, best case30 to 40 per centHot exhaust and cooling water
Diesel engine, large and well loaded35 to 45 per centHot exhaust and cooling water
Hand pump, lever and piston60 to 80 per centFriction in the pivot, leakage past the piston

The ranges are engineering rules of thumb and vary with load, size and age. Only the bicycle chain figure comes from a cited measurement. Use them for a first audit and replace them with data sheets when you need accuracy.

Source: Johns Hopkins University news release of 20 August 1999 on chain drive tests by J. B. Spicer and colleagues (ScienceDaily copy), chain efficiency 81 to 98.6 per cent; larger sprockets and higher tension gave higher efficiency; lubricant made little difference; https://www.sciencedaily.com/releases/1999/08/990819180058.htm
  • Bloomfield, How Things Work (Wiley, 2015) · ch. 2, 'The Laws of Motion, Part 2'
  • Johns Hopkins University news release of 20 August 1999 on chain drive tests by J. B. Spicer and colleagues (ScienceDaily copy) · chain efficiency 81 to 98.6 per cent; larger sprockets and higher tension gave higher efficiency; lubricant made little difference; https://www.sciencedaily.com/releases/1999/08/990819180058.htm

5.3 Capstone: how long does a heavy express take to reach speed, and what does that say about the design?

How long does the 1,000-tonne express of part 3 take to get from rest to 130 km/h behind its 123-tonne locomotive?

First, the steady-speed cost. On level track at 130 km/h, the train must overcome rolling and mechanical drag of about 3 N per kN of weight, near 30 kN, and air drag along a long train, which a rough estimate puts near 20 kN. Together that is about 50 kN, and times the 36 m/s speed it takes about 1.8 MW at the wheel. The locomotive's continuous rating of about 4.5 MW, after drivetrain losses, gives about 4.0 MW at the wheel, which leaves about 2.2 MW spare at top speed and more at lower speeds.

Second, the start. At low speed the limit is not power but friction on the rail. The maker's starting tractive effort of 322 kN, minus about 30 kN of drag, gives a net force of about 290 kN and an acceleration of 0.29 m/s2 on 1,000 tonnes. This holds until the power limit takes over, where force = power / speed, at about 12.5 m/s, 45 km/h. Reaching that speed takes about 43 s.

Third, the rest of the climb is a matter of energy. Going from 12.5 to 36 m/s needs about 570 MJ. With a surplus averaging about 2.8 MW, that takes about 200 s. The total is around 4 minutes, and the train will have covered nearly 6 km.

The lesson is that mass and power set the timetable. A line with stations 10 km apart suits a heavy train; one with stops every 3 km does not, since the train would spend its time accelerating and braking. High-speed rail spreads power along the whole train for this reason.

Data Worked estimate 13: a 1,000-tonne express from rest to 130 km/h behind one 123-tonne locomotive
StepWorkingResult
Drag at 130 km/h on level trackrolling and mechanical 30 kN + air about 20 kN (rough assumption)about 50 kN
Power to hold that speed50 kN x 36 m/sabout 1.8 MW at the wheel
Wheel power available4.5 MW x 0.88 (assumes the rating is measured before the drivetrain losses of this part)about 4.0 MW
Net force at start322 kN starting effort - 30 kN dragabout 290 kN
Acceleration at start290,000 N / 1,000,000 kg0.29 m/s2
Speed where power limit takes over4.0 MW / 322 kNabout 12.5 m/s (45 km/h)
Time and distance to that speed12.5 / 0.29; half x 0.29 x 43 x 4343 s; about 270 m
Energy to go from 12.5 to 36 m/shalf x 1,000,000 x (36 x 36 - 12.5 x 12.5)about 570 MJ
Average surplus power over this stageabout 4.0 MW - average drag power of about 1.2 MWabout 2.8 MW
Time for this stage570 MJ / 2.8 MWabout 200 s
Total time and distance43 + 200 s; roughly 270 m + 200 s x 27 m/sabout 4 minutes; about 5.7 km

Check: this is an order-of-magnitude estimate that ignores gradients, speed limits, the energy in spinning wheels and the locomotive's real power curve. A step-by-step calculation with the same assumptions gives 244 s and 5.7 km (calculated). No timetable was used to calibrate it. If the real figure for a stretch you know is within a factor of two of 4 minutes, the method is working.

Source: RDSO (Ministry of Railways), Specification RDSO/2007/EL/SPEC/0048 Rev. 0, IGBT based three-phase drive electrics for WAG-9 and WAP-7 locomotives (January 2007), WAP-7: starting tractive effort not less than 340 kN on dry rail; continuous power not less than 6,000 hp; maximum design speed 140 km/h; https://rdso.indianrailways.gov.in/uploads/files/48.pdf
  • Bloomfield, How Things Work (Wiley, 2015) · ch. 1, 'The Laws of Motion, Part 1'
  • Bloomfield, How Things Work (Wiley, 2015) · ch. 2, 'The Laws of Motion, Part 2'
  • RDSO (Ministry of Railways), Specification RDSO/2007/EL/SPEC/0048 Rev. 0, IGBT based three-phase drive electrics for WAG-9 and WAP-7 locomotives (January 2007) · WAP-7: starting tractive effort not less than 340 kN on dry rail; continuous power not less than 6,000 hp; maximum design speed 140 km/h; https://rdso.indianrailways.gov.in/uploads/files/48.pdf
  • Wikipedia, 'Indian locomotive class WAP-7' (read November 2026, citing Chittaranjan Locomotive Works and Indian Railways data), and loco-info.com, WAP-7 entry (read November 2026) · secondary sources for weight 123 t and starting tractive effort about 322 kN; continuous power given as 4,563 kW (Wikipedia) or 4,740 kW (loco-info)

See it in the table

Fly Flywheels and small mechanical storesRgn Regenerative braking and motion recoveryEl Electricity

Check your understanding

A chain has three links with efficiencies of 0.97, 0.93 and 0.98. What is the efficiency of the whole chain?

Efficiencies multiply along a chain, because each link works on what the one before delivers. 0.97 x 0.93 x 0.98 is about 0.88. The weakest link pulls the product down more than the average suggests.

A claim is made for a device that takes in 1 kWh of electricity and gives out 1.3 kWh of useful mechanical work. What does a first audit say?

Conservation of energy rules it out. Heat pumps appear to beat 100 per cent only because they move heat that is already there, which module 6 explains; a device making only mechanical work from electricity cannot.

A heavy train holds a steady 130 km/h on level track needing about 2 MW at the wheel, while the locomotive can supply about 4 MW there. What is the large reserve of power mainly for?

At steady speed on level track only the drag must be overcome. Extra power turns into acceleration, so the reserve sets how quickly the train reaches speed and how well it climbs.
← Part 4: Levers, pulleys, gears and wheels: trading force for distanceMark read · Wrap-up →