Explainer
F1 fuel and energy rules: flow limits, ERS and the sample
How F1 fuel and energy rules work: the fuel flow ceiling, the race allowance, the post-race sample, per-lap ERS limits and the sustainable fuel switch.
By CricketTaken EditorialPublished Explainer18 min read
Two cars cross the line, and one of them stops on the slowing-down lap while the other drives back to parc fermé. Nothing has broken. The car that stopped was running out of something it was required to have, and the F1 fuel and energy rules are the reason a driver will park a perfectly healthy machine on a kerb rather than complete another kilometre. Everything about how a Formula 1 car is allowed to make and spend energy comes out of a small set of limits, and the limits are stranger and more interesting than the horsepower figures they produce.
Here is the whole system in one paragraph. Fuel may be delivered to the engine only at a capped rate, measured continuously by a meter the governing body controls. A car may carry only a set mass of fuel for a race, and may not be refuelled during it. After the session a sample must still be drawable from the car, so the fuel that finished can be checked against the fuel that was declared. Electrical energy may move into and out of the battery only in capped amounts each lap, with the battery itself constrained in mass and position. And the fuel's chemistry is specified, with the sport now committed to a fully sustainable specification. Every one of those is a separate rule with its own enforcement, and the racing you see is the shape they make together.
What the fuel flow limit actually says
Take the flow rule first, because it is the load-bearing one.
Through the turbo hybrid era that ran from 2014, the ceiling was written as a mass flow: no more than 100 kilograms of fuel per hour may pass to the engine. That is an instantaneous rate rather than a total, so it applies at every moment the engine is running, whether the car is at full throttle down a straight or half throttle through a fast corner.
Below a defined engine speed the ceiling is lower, and it rises with revs on a straight line. The published relationship is a flow of 0.009 multiplied by engine speed in revolutions per minute, plus 5.5, expressed in kilograms per hour. Feed 10,500 rpm into that and the answer is exactly 100, which is where the taper meets the ceiling. Above that speed the flat 100 applies and the line stops rising.
Calculated directly from the published relationship in the FIA Formula 1 Technical Regulations, a permitted flow of 0.009 multiplied by engine speed in rpm, plus 5.5, in kilograms per hour, capped at 100 kg/h at and above 10,500 rpm. This is the legal ceiling, not any engine's actual consumption.
Show the numbers
| Item | Maximum permitted fuel flow |
|---|---|
| 5,000 rpm | 50.5kg/h |
| 7,000 rpm | 68.5kg/h |
| 9,000 rpm | 86.5kg/h |
| 10,500 rpm | 100kg/h |
| 12,000 rpm | 100kg/h |
That sloping section is doing deliberate work. Without it, a manufacturer could take the full 100 kilograms per hour at any engine speed it liked, and the optimum would be to run the engine as slowly as the rest of the design allowed, because a slow engine loses less to friction and pumping. The taper removes that option by making low engine speeds expensive in flow. Equally, it stops the reverse trick: once you are at 10,500 rpm you have all the fuel you are ever going to get, so spinning to 15,000 buys nothing except wear. The rule quietly pins the useful operating band without ever naming it.
Why a flow limit binds harder than a tank limit
The instinctive way to control a racing engine is to limit its size, and for most of motor racing's history that worked. Capacity sets how much air an engine can draw per revolution, air sets how much fuel can be burned, and fuel sets power. Cap the capacity and you cap the output.
Forced induction breaks that chain. A compressor raises the density of the air entering the cylinders, so the same swept volume can be filled with far more oxygen. A 1.6 litre engine with enough boost swallows the air of something several times its size. Capacity limits stop being power limits and become packaging limits.
So the rulebook moved the constraint to where it still binds. Fuel is where the energy is, and energy per second is power. Cap the fuel arriving per second and you have capped the chemical energy arriving per second, which is a hard ceiling on what the engine can possibly produce no matter what boost, compression ratio or valve timing is chosen.
Now compare that with a limit on tank size, which is the alternative people usually assume must exist. A tank limit constrains the total energy available for the race but says nothing about the rate. A team facing only a tank limit would build the most powerful engine it could, run it flat out for as long as the fuel lasted, and then coast. What you would get is not efficient racing but a sequence of sprints separated by cruising, and a large part of every grand prix would be run at a pace nobody chose for sporting reasons.
The flow limit does the opposite. It makes the peak the thing you cannot exceed and leaves the total as a separate, softer problem. Both limits exist in the current rules, and they bite at different moments: flow decides how fast the car can be at any instant, while the race allowance decides how many instants at that pace the driver can afford.
The consequence for engineering is the interesting part. With energy arriving at a fixed maximum rate, every manufacturer receives an identical budget per second. The only remaining variable is the fraction of that budget converted into work at the crankshaft. Formula 1 engine competition since 2014 has therefore been a thermal efficiency contest rather than a power contest, which is why the numbers manufacturers quote internally are percentages rather than horsepower, and why the design priorities of a modern power unit look so unlike a racing engine of the previous era.
How the flow is measured, and why the meter became the story
A limit that cannot be measured is a suggestion, and this one is measured continuously.
A fuel flow meter sits in the line between the tank and the injection system, sampling at high frequency and logging what it sees. The device is a homologated part: teams do not choose it, cannot modify it, and the governing body holds the calibration. The reason for that severity is obvious once stated. A team that could influence the instrument could exceed the limit at will, and the breach would be undetectable by any other means, because nothing else on the car reveals the instantaneous flow.
The measurement is not trivial. Fuel arrives in pulses rather than a smooth stream, the pressure varies, and the pump is doing violent things upstream. A meter has to resolve an average rate across a window short enough that a team cannot exceed the limit inside it. That window, and what counts as an excursion within it, has been the subject of some of the most technical arguments the sport has produced, including a long-running dispute in which a team's own instrument and the governing body's disagreed and each side had a defensible reading.
The 2026 answer to that history was to remove one of the two instruments. A single standardised meter now feeds identical data to the team and to the governing body, which does not make measurement easier but does make disagreement impossible. There is only one number, and both parties are looking at it.
The race fuel allowance, and what refuelling being banned actually did
Separately from the rate, there is a limit on the amount.
The regulations set a maximum mass of fuel that may be used for the race. In the first seasons of the turbo hybrid rules that figure was 100 kilograms, and it rose across the era to 110 kilograms as races and circuits demanded. The 2026 rules cut it sharply, to somewhere around 70 kilograms, because a much larger share of the lap is now supplied electrically and the combustion side simply needs less.
None of that would matter much if a car could be topped up during the race, and it cannot. Refuelling has been prohibited since 2010, and its absence shapes the sport more than any single technical rule. A car starts heavy and finishes light. Its lap time improves steadily as the load burns off, which is why a stint gets faster even as the tyres get worse and why comparing a lap on lap five with a lap on lap forty tells you almost nothing.
The removal of refuelling also changed what a pit stop is. With fuel in play, a stop was a variable-length event and strategy was largely a fuel calculation, with the tyres following along. Without it, a stop is a fixed few seconds of tyre change and strategy becomes an argument about rubber, which is the reason the modern race is decided by tyre life and track position rather than by fuel windows.
Fuel saving is a driving technique, not an emergency
Because the allowance is fixed and the flow ceiling is fixed, a team can calculate before the start how much fuel a flat-out race would need. Frequently the answer is more than the rules permit, and the team starts the car underfuelled on purpose.
That is a deliberate trade. Carrying less fuel makes the car lighter and therefore faster for every lap of the race, at the cost of requiring the driver to save a certain amount along the way. Teams run that sum properly: a kilogram of fuel costs a measurable amount of lap time in weight and buys a measurable amount of pace at the end, and the optimum is rarely to start with enough for full attack.
The saving itself is done in ways a viewer can spot. Lifting off before a braking point and coasting is the cheapest in lap time for the fuel it saves, because the car is decelerating anyway. Short-shifting on a straight keeps the engine below the point where the flow ceiling is highest. Running a leaner mixture reduces consumption at a direct cost in power. Drivers dislike all of it, and the radio traffic about targets during a race is not theatre, it is a driver being told how much of the remaining race has already been spent.
The important part for a spectator is that fuel saving is not a sign of a problem. It is the plan, executed. A car being told to lift and coast in the middle stint is usually doing exactly what the pre-race calculation said it would.
The one-litre sample, and what it is for
At the end of a session the car has to be able to give something back.
The regulations require that a sample of fuel, one litre in volume, can still be drawn from the car after it has finished. The purpose is verification of chemistry rather than of quantity. Every team submits its fuel specification for approval before it is used, and the sample lets the scrutineers confirm that what was in the car matches what was declared, using a chemical fingerprint of the blend rather than a simple list of ingredients.
The consequence teams care about is that the requirement is absolute. It is not enough to have been running legal fuel; the car must physically be able to surrender the sample when asked. A car that has run itself dry cannot, and the failure to provide is itself the offence, without any need to allege that anything improper was burned.
This is why cars sometimes stop on the slowing-down lap and are recovered rather than driven back. Bringing the car home under its own power costs fuel, and if the pre-race sum was aggressive, that last kilometre is the difference between a legal car and an exclusion. Parking it is not a failure. It is the cheapest way to protect a result, and it happens more often at circuits with a long run back to the pits.
What the energy recovery system may deploy in a lap
The electrical half of the power unit is metered just as carefully, and the metering is done per lap rather than per race.
Under the turbo hybrid rules the kinetic motor generator was capped at 120 kilowatts, and the energy store was permitted to send it up to 4 megajoules on any given lap. Those two numbers produce the whole behaviour of the system. Four megajoules delivered at 120 kilowatts lasts a little over thirty-three seconds, so a driver had roughly half a minute of full electrical assistance per lap, to be distributed across the corners exits where it was worth most.
That distribution was never even. Deployment out of a slow corner onto a long straight is worth a great deal; deployment onto a short squirt between two corners is worth almost nothing. So the software allocated the lap's four megajoules against a map of where the time was, and the driver's job was to work with an allocation that had been decided in advance rather than to press a button when it felt right.
The 2026 rules change the scale of this rather than the principle. The kinetic unit's ceiling rises to 350 kilowatts, and the per-lap allowances rise with it. What that does to the racing is a different question, because 350 kilowatts is enough to change the character of a straight rather than merely to shorten it, and the rules taper deployment as speed rises so that the end of a long straight is not simply a contest of remaining charge. That taper, and the manual override that replaced the movable rear wing, sit alongside the overtaking aid the sport used for the previous fifteen seasons.
What it may harvest, and why the two numbers were never equal
The recovery side had a lower ceiling than the deployment side, and the asymmetry confused a lot of people.
Under the turbo hybrid rules the kinetic unit could put back only 2 megajoules per lap while the store could hand out 4. That looks like an accounting error until you remember that a second generator existed. The heat motor generator, mounted on the turbocharger shaft, recovered exhaust energy that would otherwise have gone out of the wastegate, and its recovery was not capped at all. It could fill the store, or pass energy straight across to the kinetic unit without the store being involved. The 2 megajoule limit constrained one pathway, not the total.
Deleting the heat unit for 2026 removed that second source, so the kinetic recovery ceiling had to rise substantially to keep the electrical side viable. The practical effect is a much larger harvesting task carried out through one route, which means drivers now have to find energy in places that cost lap time: lifting early to convert speed into charge rather than into heat in the brake discs, or running below the car's potential on a straight in order to bank something for the following lap.
That is the part worth understanding as a spectator. In this formula a driver is not always trying to be as fast as possible at every instant. Some of the slow moments are purchases.
- 120Kinetic motor generator power ceiling, kilowatts
- 4Megajoules the store could deploy per lap
- 2Megajoules the kinetic unit could recover per lap
- 33Seconds of full electrical assistance that buys
Limits from the FIA Formula 1 Technical Regulations for the 2014 to 2025 power unit formula. The deployment figure is what the energy store could send to the kinetic motor generator in a lap; the recovery figure is what that unit could send back. The heat motor generator, deleted for 2026, had no equivalent cap.
The energy store: a mass window, a place to live and a state-of-charge rule
The battery is regulated in three directions at once, and each has a distinct purpose.
There is a permitted mass range with a floor as well as a ceiling. The ceiling is obvious: without it a team would carry as much storage as it could package. The floor is the interesting half, and it exists to stop an arms race in exotic cell chemistry, because a manufacturer that could halve the weight of the store would gain twice, once in mass and once in the ballast freed up to move the car's balance around.
There is a location requirement. The store lives within the protected structure of the car rather than being distributed wherever the packaging is convenient, on the straightforward grounds that after the fuel it is the most energetic object aboard and belongs inside the strongest part of the chassis.
And there is a limit on how much its state of charge may change across a lap, which closes a loophole that would otherwise sit under the deployment and recovery rules. Without it, a team could arrive at a circuit with a fully charged store and spend it down across several laps, exceeding the per-lap deployment intent while never breaching a per-lap number. Requiring the store to end a lap near where it started forces the electrical system to be genuinely cyclic, recovering what it spends, which is what makes it a hybrid rather than a battery-assisted engine.
E10, and the road to a fuel with no crude oil in it
The chemistry of the fuel has been controlled for far longer than most people realise, and the direction of travel has been steady.
Formula 1 fuel has never been a special potion. The regulations define an envelope of permitted compounds and proportions, deliberately close to something a road car could burn, and each team's blend must be declared and approved before use. The reason is partly relevance and partly enforcement: a tightly defined envelope makes the post-race chemical fingerprint meaningful, because a blend that drifts outside what was approved is detectable even when the difference is small.
From 2022 the envelope has required at least ten per cent ethanol, and not any ethanol: the regulations specify an advanced sustainable source, which excludes food crops and pushes suppliers towards waste and residue feedstocks. Ten per cent is a modest number and it was chosen to be modest, because the point was to prove the supply chain and the engine compatibility at low risk rather than to make an immediate emissions claim.
The 2026 specification completes the move. The fuel must be fully sustainable, produced from non-food biomass, municipal waste or carbon captured from the atmosphere, with no fossil crude in the blend. The carbon still comes out of the exhaust, which is the honest way to describe it; the claim is about where the carbon came from rather than about where it goes.
For a manufacturer this is a much larger change than the ethanol step. A synthetic fuel is designed rather than refined, which means its properties can be tuned deliberately, and a fuel supplier becomes a development partner in the way an engine component supplier is. It also means that fuels from different suppliers do not necessarily carry the same energy per kilogram, and that fact broke the old limit.
Why 2026 rewrote the limit in megajoules
A mass flow limit assumes that a kilogram of fuel is a kilogram of energy. With fossil-derived petrol inside a tight specification envelope, that assumption held closely enough to be workable. With designed synthetic fuels it does not.
If the rules had kept a 100 kilograms per hour ceiling, the competitive move would have been to formulate the most energy-dense blend physically possible, because more energy in each permitted kilogram means more power for the same legal flow. That would have turned a chunk of the competition into a chemistry race about calorific value, which is not what anybody wanted to reward.
So the 2026 limit is expressed in energy directly: a maximum of 3,000 megajoules per hour, with the same style of taper below a defined engine speed. The unit change removes the incentive entirely. Whatever a supplier's fuel contains, the car is allowed the same joules per second as everybody else's, and the contest returns to the fraction of those joules that becomes forward motion.
The arithmetic is worth doing once, because it makes the whole formula legible. Three thousand megajoules an hour divided by 3,600 seconds is a shade over 0.83 megajoules per second, which is to say roughly 833 kilowatts of chemical energy arriving continuously. Everything the combustion side produces has to come out of that, and the difference between a good engine and an excellent one is a few percentage points of it. Those few points are worth tens of kilowatts, which is worth several tenths of a second per lap, which is worth a championship. That is the entire competitive proposition of the modern formula stated in one division.
- The blend is declared and approvedA team submits its fuel specification before it is used, and the governing body approves it. Everything downstream is checked against this declaration rather than against a general idea of what petrol should be, which is what makes the post-session test enforceable.
- Fuel is loaded, and the mass is cappedThe car is filled with no more than the permitted mass for the race, and it cannot be topped up once the race starts. Teams routinely load less than a flat-out race would need, because the weight saved is worth more than the pace given away.
- Flow is capped at every instantA homologated meter between the tank and the injection system logs the delivery rate continuously. The ceiling applies moment by moment rather than as an average, so there is no window in which a team can overshoot and repay later.
- The engine converts what it is givenWith energy arriving at a fixed maximum rate, output is decided by thermal efficiency alone. Every manufacturer has the same joules per second to work with, which is why the engineering conversation is about percentages rather than horsepower.
- Braking is converted to chargeThe kinetic motor generator resists the crankshaft under deceleration and sends electricity to the energy store, within a per-lap recovery ceiling. Drivers extend this by lifting early, which costs lap time and buys energy.
- The store holds it, within limitsThe battery has a permitted mass range, a required location inside the protected structure and a cap on how far its state of charge may move across a lap. That last rule forces the system to be cyclic rather than allowing a slow drain across a stint.
- Deployment is released, and capped per lapEnergy flows back through the kinetic unit up to a power ceiling and a per-lap total, allocated by software to the corner exits where it is worth the most lap time. The driver works to an allocation rather than deploying on instinct.
- A sample must still be drawableAfter the session the car has to surrender a one-litre sample so the fuel can be fingerprinted against the declaration. Being unable to provide it is a breach in its own right, which is why a car sometimes stops rather than driving back.
The sequence of regulated checkpoints from the moment fuel is declared to the moment a sample is drawn. Each step is a separate rule with its own enforcement, which is why a car can be legal at one checkpoint and in breach at the next.
The cost side of the fuel question
One thing that surprises people is how little of this is constrained by budget. Fuel and lubricant development sits with the power unit manufacturer rather than the chassis team, and the financial rules that cap a team's spending treat the power unit as a purchased item with its own separate arrangements. A fuel supplier's laboratory is not competing for the same pounds as a floor upgrade.
That separation is deliberate and it has a consequence worth noticing. Under a regime where almost every other route to performance is capped in money as well as in rules, fuel chemistry is one of the few remaining places where a partner outside the team can spend heavily on a competitive advantage. It is also why fuel and lubricant branding on a car is not merely sponsorship, and why the technical partnerships between engine manufacturers and energy companies are as old as the sport.
The tank itself, and the rules nobody argues about
Underneath the flow and energy limits sits a set of plumbing regulations that almost never make the news, and they are worth a paragraph because they explain several things a viewer sees.
A Formula 1 car carries one fuel tank, not several, and it is a flexible bladder built to a safety standard rather than a rigid vessel. It sits between the driver's back and the engine, inside the strongest part of the car, which is the same reasoning that puts the energy store there. Ancillary tanks and reservoirs are prohibited beyond the small collector that keeps the pumps fed under cornering loads, so there is nowhere to hide a litre that the sampling rule cannot reach.
There is also a temperature rule, and it is one of the few limits in the book that exists purely to stop a cheap trick. Fuel may not be chilled far below the ambient temperature at the circuit. Cold fuel is denser, so a tank filled with chilled fuel holds more mass in the same volume, and since the flow limit is expressed in mass rather than in litres, a team could gain by refrigerating what it loaded. Capping how far below ambient the fuel may be removes that route without needing to police anybody's freezer, because ambient temperature is measured and published anyway.
The consequence a spectator notices is the pit lane routine on a hot afternoon: fuel being handled with far more ceremony than the volume seems to justify, checked and logged before it goes anywhere near the car. That is a legality process, not a safety one, and it happens at every event.
What to watch and listen for
A handful of signals will let you read the fuel and energy story of a race without any telemetry.
Listen to the engine on the approach to a braking zone. A driver lifting noticeably early and coasting is either saving fuel or harvesting energy, and often both. It sounds like hesitation and it is arithmetic.
Watch the gear the car takes on a long straight. Short-shifting keeps the engine below the point where the flow ceiling is at its most generous, and it is one of the clearest visible signs of a fuel target being managed.
Watch where a car deploys. Under the electrical rules, a lap's allocation is spent where it earns the most, which means the exits of slow corners onto long straights. A car that appears to have nothing left at the end of a straight has not run out of engine; it has reached the end of an allocation or the top of a taper.
Watch the slowing-down lap. A car stopping on the circuit after the flag, with no smoke and no obvious damage, is usually protecting a fuel sample. It looks like a failure and it is a decision.
Watch for the difference between qualifying pace and race pace at circuits with long straights. A single flying lap can spend energy without repaying it. A race lap cannot, and the gap between the two is the clearest available measure of how tightly a car is being managed. More explainers on how the sport's rulebook shapes what you see are in our motorsport section, and the rest of the archive is on the blog.
Common questions
What is the F1 fuel flow limit?
It is a ceiling on how fast fuel may be delivered to the engine, rather than on how much fuel the car carries. Through the turbo hybrid era the ceiling was a mass flow of 100 kilograms per hour, tapering below a defined engine speed so that revving harder bought nothing. From 2026 the same idea is expressed as an energy flow, capped at 3,000 megajoules per hour, which suits a fuel whose energy per kilogram is no longer fixed.
Why does F1 limit fuel flow instead of engine size?
Because engine size stopped being a power limit the moment turbochargers returned. A turbocharger can raise the density of the incoming air almost at will, so a small engine can swallow the air of a much larger one and burn fuel accordingly. Capping the rate at which fuel arrives caps the rate at which energy arrives, which caps power directly, and it leaves manufacturers competing on how much of that energy becomes work.
How much fuel does an F1 car carry in a race?
Through the 2019 to 2025 seasons the maximum permitted for a race was 110 kilograms, up from 100 kilograms when the turbo hybrid rules began. The 2026 regulations cut it sharply, to roughly 70 kilograms, because the larger electrical contribution means less of the lap has to come from combustion. Refuelling during a race has been banned since 2010, so whatever the car starts with is what it has.
Why does the FIA take a fuel sample after the race?
So that the fuel a car finished on can be compared against the specification the team declared and had approved before the event. The rules require that a sample of one litre can still be drawn from the car after the session, which is why cars sometimes stop on the slowing-down lap rather than driving back to the pits. Failing to provide that sample is treated as a breach in itself, without any need to prove the fuel was illegal.
How much energy can an F1 car deploy per lap?
Under the turbo hybrid rules the energy store could send up to 4 megajoules per lap to the MGU-K, which at that unit's 120 kilowatt ceiling is a little over thirty seconds of full electrical assistance. Recovery through the MGU-K was capped separately and lower, at 2 megajoules per lap, with the difference made up by the heat motor generator that no longer exists. The 2026 rules raise both the power and the per-lap allowances substantially and release the energy in several bursts rather than one.
Is F1 fuel the same as petrol?
It is a tightly specified fuel that must sit inside a defined chemical envelope, declared to the governing body and approved before it is used. Since 2022 it has had to contain at least ten per cent ethanol from advanced sustainable sources, and from 2026 the requirement is a fully sustainable fuel made from non-food biomass, municipal waste or carbon captured from the air rather than from crude oil.
Filed under Motorsport·formula 1 · motorsport · regulations · power unit · sustainability