The idea
Beginner explanation — switch to Technical for the engineering detail.
A petrol engine burns a mixture of air and fuel, and the proportions matter. Too much fuel wastes it, fouls spark plugs and makes smoke; too little makes the engine hesitate and run hot. For good economy and low emissions the mixture has to be just right, all the time, from a cold start on a winter morning to full throttle on a motorway.
Fuel injection does this job with electronics. A pump in the tank pushes fuel at pressure to the injectors: small electrically operated valves that spray a fine mist. The engine computer (ECU) reads sensors that tell it how much air the engine is drawing in, how warm it is and how hard the driver is pressing the accelerator. It then decides how long each injector stays open — typically a few thousandths of a second per cycle.
An oxygen sensor in the exhaust reports whether the last mixture was slightly rich or slightly lean, and the ECU nudges the next injection to correct it. This feedback happens many times a second, which is why an injected engine starts easily, idles smoothly and stays clean as it wears.
What you will learn
- Explain why the air-fuel ratio must be controlled and what λ = 1 means
- Name the main parts of an injection system and what each one measures or does
- Compare port fuel injection with direct injection
- Describe how the oxygen sensor closes the loop on mixture control
How it flows
Read the diagram as text
- Air & engine sensors → ECU
- ECU → Injector (pulse)
- Pump & rail → Injector (pressure)
- Injector → Combustion
- Combustion → Oxygen sensor (exhaust)
- Oxygen sensor → ECU (feedback)
- Air & engine sensors: MAF / MAP, temperature, throttle, rpm
- ECU: Fuel mass → pulse width
- Pump & rail: Fuel held at pressure
- Injector: Timed spray
- Combustion: Air + fuel + spark
- Oxygen sensor: Rich or lean?
Step by step
Read each step in order.
Step 1: Measure the air
Sensors tell the computer how much air is flowing into the engine right now.
Step 2: Calculate the fuel
The computer works out how much fuel matches that air, adding a little extra when the engine is cold or working hard.
Step 3: Pressurise the fuel
A pump keeps fuel waiting in a rail at steady pressure so every injector sprays the same way.
Step 4: Open the injector
Each injector opens for a few milliseconds at exactly the right moment, spraying a fine mist.
Step 5: Check the exhaust and correct
An oxygen sensor in the exhaust tells the computer whether the mixture was rich or lean, and it adjusts the next squirt.
Go deeper
Optional. Open a panel for the engineering detail — or switch to Technical above.
The engineering explanation3
Complete combustion of petrol needs about 14.7 kg of air per kilogram of fuel (the stoichiometric ratio). Engineers describe mixture with λ = actual AFR ÷ stoichiometric AFR: λ < 1 is rich, λ > 1 is lean. A three-way catalytic converter can only oxidise CO and HC while reducing NOx when λ is held in a narrow window around 1, so closed-loop control is fundamental to modern emissions compliance.
The ECU determines the air mass per cylinder per cycle either directly from a hot-film mass-air-flow (MAF) sensor or indirectly from manifold absolute pressure, intake air temperature and engine speed (speed-density). Dividing by the target AFR gives the fuel mass required. An injector at a fixed pressure difference flows at a nearly constant rate, so fuel delivered is roughly proportional to the pulse width minus a dead time (the opening delay, which grows as battery voltage falls and is compensated in software).
Port fuel injection (PFI) sprays at roughly 3–5 bar onto the back of the intake valve, giving time for the fuel to evaporate and mix. Gasoline direct injection (GDI) sprays straight into the cylinder at roughly 100–350 bar. Evaporating fuel inside the cylinder cools the charge, allowing a higher compression ratio before knock, and gives precise control of injection timing; the trade-offs are higher particulate emissions (hence gasoline particulate filters) and carbon build-up on intake valves that are no longer washed by fuel. Some engines combine both. The ECU runs open loop when the oxygen sensor cannot be trusted or a different target is needed: cold-start enrichment, full-load enrichment to protect the catalyst and exhaust valves, and fuel cut-off on the overrun.
Key concepts4
- Stoichiometric ratio
- The air-fuel mass ratio at which all fuel and all oxygen can react; about 14.7:1 for petrol.
AFR_stoich ≈ 14.7- Lambda (λ)
- Actual air-fuel ratio divided by the stoichiometric ratio. Below 1 is rich, above 1 is lean.
λ = AFR_actual / AFR_stoich- Injector pulse width
- How long the injector is held open each cycle; fuel delivered is roughly proportional to it once the opening delay is subtracted.
m_fuel ≈ q × (t_pulse − t_dead)- Fuel trim
- The ECU's percentage correction to its calculated fuel quantity, based on oxygen-sensor feedback. Short-term trim reacts quickly; long-term trim learns slowly.
Common misconceptions3
MythAdding more fuel always makes more power.
RealityPower is limited by how much air the engine can burn. Best power occurs slightly rich of stoichiometric; beyond that, extra fuel only cools the charge and wastes fuel.
MythThe oxygen sensor measures how much fuel is in the exhaust.
RealityIt measures leftover oxygen. Little oxygen means the mixture was rich; plenty of oxygen means it was lean.
MythDirect injection is simply a better version of port injection.
RealityIt enables higher efficiency and power, but brings higher particulate emissions and intake-valve deposits, which is why some engines use both systems.
In the real world3
- A trip computer showing zero consumption while coasting downhill in gear is the ECU's overrun fuel cut-off at work.
- A split intake hose after the airflow sensor lets in unmeasured air, so the engine runs lean and often logs a 'system too lean' fault code.
- Cold engines idle faster and use more fuel for the first minutes because the ECU enriches the mixture until fuel evaporates properly.
Common questions
Short answers to the questions people most often ask about this topic.
How does fuel injection work?
Sensors measure how much air the engine is drawing in, and the engine control unit calculates how much fuel will burn cleanly with it. It then opens electrically operated injectors for precisely the right number of milliseconds, spraying finely atomised fuel into the intake or straight into the cylinder.
What is the difference between port and direct injection?
Port injection sprays fuel into the intake port just behind the intake valve, where it mixes with the incoming air. Direct injection sprays it at high pressure straight into the combustion chamber, which cools the charge and allows finer control at the cost of a more complex high-pressure system.
Why did fuel injection replace carburettors?
Electronic injection meters fuel far more precisely in every condition, making engines cleaner, more economical and easier to start. Exact fuel control is also essential for catalytic converters to work, which is why emissions rules made injection universal.
Check your understanding
3 questions. Your best score is saved in this browser.
0 of 3 answered
Question 1 of 3
Roughly how many kilograms of air are needed to burn one kilogram of petrol completely?
The story behind it
Who worked this out, and what they were up against. Optional.
1885–1893
Wilhelm Maybach
Engine designer, partner of Gottlieb Daimler
Taught the petrol engine to mix its own fuel and air.
After both of Maybach's parents died in the mid-1850s, a philanthropic institution in Reutlingen took him in. He followed Gottlieb Daimler to Karlsruhe in 1869; they patented their petrol engine on 16 December 1883, and in November 1885 Maybach rode their engine-powered wooden Reitwagen. His float-fed carburettor dates from 1885 and his spray-nozzle carburettor was patented in 1893. Electronic fuel injection took until 1958 to reach cars: the Chrysler 300D, DeSoto Adventurer, Dodge D-500 and Plymouth Fury offered the Electrojector.
Sources
- Wikipedia: Wilhelm Maybach (opens in a new tab; read 2026-10-25)
- Wikipedia: Fuel injection (opens in a new tab; read 2026-10-25)
Retold in our own words from the sources above.
Almost there
- Overview — Read the introduction to the end.
- See it work — 5 steps left to read
- Go deeper (optional) — Optional: open a deeper panel or switch to Technical.
- Check — Answer the quiz questions.
- Story (optional) — Optional: who worked this out?
Progress is stored only in this browser — there is no account.
Sources & attribution
Further reading
- How a fuel pump worksHow a Car Works · consulted 2026-10-21(external site, opens in a new tab)
Further reading on an external site. Linked only; AutoVerse lesson text is written independently.
- Licence
- AutoVerse original content
This lesson is original AutoVerse writing. The links above are suggestions for further reading, not the source of this text.