Why do F1 cars keep running out of fuel?

F1 Technology

Sebastian Vettel, Red Bull, Yas Marina, 2012When it comes to running a racing car, allowing it to run out of fuel is a failure of the most fundamental kind. Yet F1 teams run into exactly that kind of problem with surprising frequency.

Already this season we’ve seen Mark Webber run out of fuel during qualifying in China. This was the second such failure for Red Bull in six races. And in Malaysia Lewis Hamilton’s race was dominated by a pressing need to save fuel.

How are teams managing to get a seemingly simple part of going racing wrong? And why are the drivers apparently incapable of keeping an eye on their fuel gauges for themselves?

An old problem

F1 teams have long understood the value of not putting a drop more fuel in a car than is necessary. More fuel means more weight which means slower lap times. When huge sums are being spent in the pursuit of fractions of a second, no one wants to undo that work by needlessly sloshing in a few extra kilos of fuel.

That is as true today as it was 50 years ago. Lotus owner Colin Chapman was famed for his mania for saving weight. He took the practice of putting as little fuel in as possible to such lengths that when he wasn’t looking his mechanics would add a few extra litres to ensure the car reached the chequered flag.

They weren’t always successful in their covert endeavours. At Monza in 1967 Jim Clark’s Ford Cosworth DFV ran dry on the final lap after he had fought his way up from the rear of the field. Jochen Rindt and Mario Andretti were also victims of Chapman’s overzealous weight saving which sometimes extended to taking fuel out of the cars while they sat on the grid.

Working out race fuel loads

Start, Sepang, 2013Today that practice is forbidden by the rules on safety grounds, which means teams must work out how much fuel to put in their cars before they send them to the grid. Making that decision is not as simple as working out the rate of fuel consumption per lap and multiplying by the number of race laps.

Variations in climactic conditions can have a strong effect on fuel consumption. In wet conditions cars lap more slowly and therefore use less fuel. And teams will not know in advance of a race exactly how wet it’s going to be.

This gives some insight into the difficulties Mercedes had in Malaysia. As drivers headed to the grid half an hour before the start the track was so wet several of them skated off at turn four.

But in the race it dried up quickly – everyone was on slick tyres by lap nine. Not long after that the first radio messages to Hamilton urging him to save fuel were played on the team radio channel.

Whatever the conditions, the desire to put in as little fuel as possible remains pressing. Reports suggest some cars have gone to the grid this year with as much as 10% less fuel than they need to do the race flat-out. That’s a potential weight saving of 15kg on F1′s most punishing tracks for fuel consumption.

Teams deliberately under-fuel their cars because they expect their drivers won’t be able to go flat out at times during the race. For example, they may get stuck in traffic – this is especially likely for those in the midfield.

Qualifying

While this explains the incentive for under-fuelling cars in the races, it’s in qualifying sessions that we’ve seen the most extreme examples of cars being underfuelled. The ban on in-race refuelling at the end of 2009 means drivers are now running their lowest possible fuel loads in all three parts of qualifying.

To work out how little fuel needs to be in the car before the engine starts to cough, teams will often deliberately run their cars out of fuel during pre-season testing and then measure how much is left in the tank and collector.

Even so on several occasions since we’ve seen drivers run out of fuel during their qualifying runs. It happened to Lewis Hamilton in Canada in 2010 and again in Spain last year. Sebastian Vettel had the same drama at Abu Dhabi last year and, most recently, the same happened to Mark Webber in China.

Varying explanations were given for these failures, sometimes in the hope of avoiding the dreaded exclusion from qualifying (as Hamilton managed in 2010). But they shared the root cause of the team making an error in fuelling the cars during the high-pressure, time-limited modern qualifying format.

Where’s the fuel gauge?

Why can a F1 driver not tell for themselves whether their car has insufficient fuel? After all, every road car is fitted with a gauge which alerts the driver if this is the case.

The shape and construction of an F1 car’s fuel tank makes this impossible. This is due to the severe forces an F1 car experiences which causes the fuel to move around. Engineers need to control this movement – “slosh” – to keep the car’s centre of gravity low and to ensure a consistent supply of fuel to the engine.

“You can’t just put a dipstick in there,” explains F1 technology expert Craig Scarborough. “At Spa the fuel is actually at the top of the fuel tank as you crest the rise coming out of Eau Rouge!”

This video illustrates the forces at work on a fuel load in an F1 car’s tank:

http://youtu.be/PKoOeNUoNHU?t=42s

F1 fuel tanks feature a series of chambers to keep the fuel in a position where the fuel pump can collect it. This network of chambers controls the fuel ‘slosh’ during acceleration, braking and cornering. Trapdoors in the chambers allow the fuel to travel down but not back up.

This image of a 2008 BMW-Sauber F1.08 shows the fuel tank and the horizontal divides within it which help control the position of the fuel:

BMW-Sauber F1 car sliced in half

But it’s not a foolproof system. In Italy last year Jenson Button retired when one of the trapdoors became jammed in his McLaren’s fuel tank.

Degree of risk

Although they are unable to directly measure how much fuel is in their car at any given time, teams can gather readings from other sources. For example, adding fuel to the car should cause a corresponding increase in the load on the suspension.

Once the car is lapping, the team will study the rate at which fuel is being fed to the engine to calculate how much is left – and whether they’re using it too quickly. The driver has some degree of flexibility to alter the rate of fuel consumption by selecting different engine maps and by altering their driving style. Hence radio messages telling them to “lift and coast” as they approach their braking points.

Fuelling an F1 car is a delicate balance of risk versus reward. The penalty of coming to an early stop is high, but in a sport that’s fixated on performance the temptation to shave a few tenths off by under-fuelling the car is great.

There will always be those like Chapman who are more inclined to push the envelope than their rivals are. It’s telling that some teams have had their fingers burnt more than once – Red Bull and McLaren, for example – while others have avoided this kind of trouble.

Like a driver judging a risky overtaking move or weighing up whether to back off for a high-speed corner, it’s another of the high-stakes decisions at the heart of Formula One.

F1 technology

Browse all F1 technology articles

Images ?? Red Bull/Getty

Advert | Go Ad-free

94 comments on Why do F1 cars keep running out of fuel?

1 2 3
  1. MilleniumBug (@milleniumbug) said on 29th April 2013, 12:45

    Great insightful article! Thanks Keith

    • Atticus (@atticus-2) said on 29th April 2013, 12:49

      I second that. The first words coming to my mind were ‘great insight’ as well.

      Thank you very much.

    • HiPn0tIc (@hipn0tic) said on 29th April 2013, 13:20

      +2.
      It was in fact a great insight.

    • celeste (@celeste) said on 29th April 2013, 21:39

      Thank you, I understand a little now

    • Nick (@nick101) said on 30th April 2013, 9:45

      Agreed, great article!

      But I’m afraid I don’t agree with the practice of under fueling cars. As far as I’m concerned it’s a stupid tactic.

      Obviously much smarter people than I have decided it’s not, but I just can’t see the sense in it. How many times do we hear drivers being told to slow down and conserve fuel? And more importantly, how many times do we not hear it, yet it happens anyway.

      I really don’t understand under fueling a car and then getting you’re driver to coast for half the race. In then end, they are circulating sometimes 4-5 seconds per lap slower than they could because the are fuel saving. If the point of under fueling a car is to save 1/10 per lap, I ask you what the point is when they loose 20-30 seconds over the course of a GP because they are costing in fuel saving mode?

      It’s crazy!

      • Nick (@nick101) said on 30th April 2013, 9:46

        *coasting

        • Robbie (@robbie) said on 30th April 2013, 15:43

          @nick101 Agreed. I too acknowledge that they must know what they are doing, but I find it hard to wrap my head around the risk, for the sake of 1/10 per lap, when so much time can be lost having to conserve for the last third of the race. And it seems to me too, that without a small buffer of fuel to rely on, are not the team’s strategies then limited as the race unfolds? Why would they limit themselves to various strategy adaptations for the sake of a couple of extra pounds of fuel? Oh well. They know what they are doing and in general most cars do make it to the finish line fuel-wise at least. It is rare that cars don’t, although the point is well taken that many cars have had to slow down and conserve in order to get past the checkered flag.

          • Those 1/10ths per lap are invaluable near the start of the race where it can mean a pass vs not a pass. The time lost later is less of a problem.

    • Mark (@marlarkey) said on 2nd May 2013, 23:00

      +1 excellent.

  2. liedra (@liedra) said on 29th April 2013, 12:48

    Great writeup, thanks!

  3. GeeMac (@geemac) said on 29th April 2013, 13:11

    That video is fascinating, I remenber seeing it on TFL and thinking “why didn’t I think of this before”. It didn’t even cross my mind that the fuel in the car would slosh around from side to side, forward and backwards under cornering and braking.

    • MahavirShah (@mahavirshah) said on 29th April 2013, 15:52

      I had seen a similar CFD simulation done for a Ferrari road carI think, during my MSc. Its quite fascinating as to how they analyse it taking actual acceleration and force data through live runs and use that to run the simulation. When, I had seen it the first time (the question then had also been about minimum tank size required) I was shocked that the sloshing at times was so severe. I don’t believe fuel estimation is as simple as it looks.

  4. Jonny C (@loomx92) said on 29th April 2013, 13:20

    That’s a really interesting article. I was talking to some friends (non F1 fans) a couple of days ago about how close the teams run their cars to the limit and they asked me what the fuel consumption was on an F1 car. I was stumped and was wondering if anyone knows what it is? And also what the kg – ltr conversion rate is for petrol.

    • DaveD (@daved) said on 29th April 2013, 20:53

      They use about a 2.7kG per lap or about 6lbs on the average F1 circuit. Or you could look at it as as about 1.5km/liter if you’re normal or you could say 3.5mpg if your one of us Americans :-)

      • DaveD (@daved) said on 29th April 2013, 21:05

        By the way, I was serious about that mpg being American as we have a smaller “gallon” than your imperial gallon. Your Imperial gallon =1.2 American gallons so for you it would be 4.2 mpg of petrol :-)

        I wish we’d all just go with metric and make this easy…but my fellow Americans take some kind of strange pride in using antiquated measurement systems. I think we believe we invented them or something LOL

    • DaveD (@daved) said on 29th April 2013, 20:58

      I forgot your other question: 1 liter weighs about 7.11kg or 1kg would be 1.41 liters.

  5. sumedh said on 29th April 2013, 13:28

    Excellent article! Interesting to read how teams actually measure the fuel left.

    One extra question, fuel consumption varies from race-track to race-track although the distance of each race is almost same (305km). What determines the fuel consumption. Speed or downforce or the expected length of the race? Because I read that Monza is the race that needs least fuel. Is it so?

    • From a physics perspective faster cars need more fuel, and acceleration is costly

      To accelerate a car from 0 to 100 kmph requires at least 4x the fuel to accelerate it to 50 kmph (aerodynamic drag is not considered in that calculation).

      At speed the engine must provide enough acceleration balance the aerodynamic drag which is related to velocity^2, so a car at a steady 100 kmph also requires 4x the fuel to maintain that speed for the same length of time as at 50 kmph.

      Of course, 100 kmph it takes half as long, so the same distance at 100 kmph vs 50 kmph still requires twice as much fuel.

    • beneboy (@beneboy) said on 29th April 2013, 15:17

      Increasing downforce increases the drag (someone with a better understanding of physics will probably be able to tell us by how much) and more drag means you need to put in more energy to go the same speed.
      Monza is the track where teams traditionally use the least amount of downforce but there’s also the fact that it has lots of long straights with very few corners which means you’re not having to slow down/speed up as often as at most other tracks; these two factors make it an average to low fuel consumption track.

      A high percentage of the lap is spent at full throttle, which increases fuel consumption. This is however counterbalanced by the very low drag. In other words, more fuel is injected into the engine than at any other circuit but since the car is going so much faster the effect is cancelled out. Monza is therefore counted as an ‘average to low’ fuel consumption on kg/km.

      http://media.renault.com/global/en-gb/renault/Media/PressRelease.aspx?mediaid=33258

    • Bleu (@bleu) said on 29th April 2013, 15:39

      I think the circuits which have lots of accelerating and braking use most fuel.

    • Essentially as has already been said: tracks with lots of acceleration zones and high top speeds require more fuel to be burnt off in reaching those speeds and also with aerodynamics the cars effectively weigh more with the drag, so it takes more fuel to accelerate them which can be related to Newton’s laws of motion and unbalanced forces (essentially, the increased air resistance creates a larger unbalanced force trying to slow the car down, so the engine has to work harder to accelerate the car by applying a greater force to counteract it).

      • Newton’s first law basically, so it’s really a case of knowing the force of air resistance acting upon the car and how that is proportional to downforce levels. The effect of downforce is quite significant though and since F1 cars can go a good deal faster without their aerodynamic components (Honda did a run at Bonnevile and had an average speed of 397km/h) with obviously a minimal drag setting while the highest speed hit during a Grand Prix (I believe, although it may not be correct) was only 370km/h. That’s a 27 km/h loss and that was a speed set at Monza! This is explained from the fact that the wing profiles and such like are run at steeper angles and just I’m general the amount of area for flow attachments is greater (which increases drag), represented by the equation R = ½ρCAv2 (the C being the important bit, drag co-efficient).

        So to link it back to the point, basically it’s just proportional to all the factors in that equation and importantly the drag co-efficient as that is what is mainly influenced by the aerodynamics.

    • Another thing worth mentioning as was pointed out to me by @jonathan-proc is that the teams who are running coanda exhausts purposely burn more fuel in certain ranges to increase the exhaust flow to the diffuser. More efficient systems create less drag and probably require less fuel so I’d imagine teams like McLaren, who have more developed systems than say Williams would use less fuel in that respect.

      Engines will also determine fuel consumption (although that will only vary from team to team, it isn’t really affected much by track characteristics): the Renault engine uses less fuel than the Mercedes etc.

      • @jonathanproc. As an illustrative example, lets say the Red Bull has a co-efficient of 1 compared to the Lotus which has 0.8. Using P = ½ρCAv^3 to find the power (as P = F (or R on this case) x v) and saying we are maintaing a constant speed (v) of 80m/s (roughly 290km/h), with air density (ρ) at 1.225 kg/m^3 and the area in contact with the air at 2m^2 to fit with the equation, we can input all this and get a power value required of 627.2 kW to remain at a constant speed (or roughly 840hp) for the RB9, but only 501.76 (or roughly 670hp) for the E21.

        Lets then say that at max output (for the purposes of this 1000hp) the engine has an efficiency of 2km/l (this is probably way out) which means a fuel usage at 290km/h (sticking with the original value) per second of 0.04l – if we assume it is on a liners scale, that means the RB9 would use 0.034l/s and the E21 0.027l/s! In an hour and a half long race at that speed if it were constant, that is a fuel saving of around 37l!

        Of course that isn’t anywhere near what we actually see, but I hope that helps to illustrate the difference drag can make @beneboy!

        • Really though in a race situation between different cars we’re looking at a much more marginal change as the downforce levels will be more similar and of course you aren’t doing the whole lap at 290km/h! I’d say a draggy car could use up to 5/10l more fuel over a race though.

        • DaveD (@daved) said on 30th April 2013, 3:10

          Thanks for throwing in some real numbers and calculations Max. I’m trying to find a back issue I have of Race Tech Magazine from last year. But if I’m remembering right, they said an 2012 F1 car had a frontal area of about .72 and Cd of ~.95 and if I’m remembering correctly then they’d be using 215kW (288hp) to overcome aerodynamic drag at 290kph.

          I did some quick calculations and at Monza, with and average speed of 237kph, they could save 9.8l of fuel if they could reduce the Cd from .95 to .93 and keep the same downforce. I don’t know if that’s possible, but they’d kill for it LOL

          It seems that RedBull’s (well, Adrian’s) specialty is to get that downforce while reducing drag compared to everyone else.

          I’ll try to find that issue and verify my memory, but I’m pretty sure that was what they said.

          • @daved thanks for the slightly more accurate contribution! Indeed though the main point here is that drag can have a significant affect on fuel consumption and not to mention speed so it is no surprise then that designers try to minimise it as far as possible without losing the downforce!

          • DaveD (@daved) said on 1st May 2013, 15:43

            Yeah, sorry to try and be so geeky. But I find this area fascinating. I bet they could let them use all the KERS they wanted and get rid of the regs requriing a flat floor. If they let them shape the floor for downforce rather than using front and rear wings, and a few other tricks and they’d be able to do even faster lap times next year with the 6 cylinder turbo’s than they do today.
            And they could probably cut the needed fuel level in half with the reduced drag and extra KERS.

          • DaveD (@daved) said on 1st May 2013, 20:09

            I wonder if Bernie will hunt me down and ban me from the track at COTA this year for looking forward to the V6 turbo engines? He wants to go back to V12′s LOL

          • Max Jacobson (@vettel1) said on 2nd May 2013, 20:55

            @daved oh not at all, if anything I was the one trying to be geeky by contributing many paragraphs! ;) Your accuracy is genuinely appreciated, as I wasn’t actually aware of the exact values.

            I’m looking forward to the new engines as well: they are far more sophisticated and interesting than a V12! As much as the sound would be biblical, I’d still rather see technology prevail over “the show”, which is why I also don’t like the constrictive rule book.

  6. Calum (@calum) said on 29th April 2013, 13:40

    Really great picture of Vettel in Abu Dhabi after running out of fuel.

  7. karter22 (@karter22) said on 29th April 2013, 14:42

    Wow! Just wow! Excellent article!

  8. Traverse (@) said on 29th April 2013, 15:02

    Why do F1 cars keep running out of fuel? Because the teams don’t put enough fuel in.

  9. Boomerang said on 29th April 2013, 15:21

    Great article Keith. I enjoyed it indeed. I thought they used ultrasound to measure the amount of fuel. In my opinion it would be the best way to measure it considering the violent G forces moving the fuel all around the tank.

  10. OmarR-Pepper (@omarr-pepper) said on 29th April 2013, 15:29

    I’m not an engineer, but I imagine somebody can invent a valve measuring the flow getting into the engine, not exactly the fuel IN the tank. So by a simple rest (starting fuel – flow measured) they could know how much fuel they still have. Or is some kind of device banned? Probably impossible to make in real life due to physics?

    • DavidS (@davids) said on 29th April 2013, 16:06

      Since the engine has electronic fuel injection, they would have data on how much fuel is actually being injected into the engine. Injectors have to be very precise, because they have to achieve a specific air/fuel mix across the whole RPM range, throttle positions, and even the density of the air. Thus, they can provide accurate data on fuel consumption.

      The teams are extremely good at forecasting exactly when a car will run out of fuel, as we have seen drivers being told during qualifying to stop on the track so they can still provide a sample. They can predict that it will run out to within a couple of kilometres, while the fuel light in road cars tends to be wildly inaccurate.

      The problem is that the teams can’t predict the future, and all the little things that will affect consumption, this wouldn’t be a problem if they factored in a contingency, but they don’t do that.

    • mantresx (@mantresx) said on 29th April 2013, 16:06

      ehmm that’s exactly what they do, although it makes you wonder how they managed to do it 30-40 years ago? my guess is that they took averages from a number of laps.

    • DaveW (@dmw) said on 29th April 2013, 16:13

      I think they know quite accurately the rate of consumption based on the electronic control of the engine. I doubt they even need to directly measure the flow of fuel physically. When the car is consumping mulitple liters per lap, and you want to get the car back to the pits with 1.01 liters of fuel, you have to constantly monitor and minimize consumption. One injudicious stomp on the gas pedal on a cool down lap could send you to the back of the grid with a penalty.

    • The issue isn’t really meaning the amount of fuel left (as they can do exactly that @omarr-pepper) but foreseeing how much fuel they are going to use: it’s very difficult to accurately predict how much fuel you will need during a race due to all the different factors and of course you want to push to the extremes how little fuel you put in the car.

      I don’t think that fully explains the qualifying situation: I think that is simply a case of the teams either making an error or pushing their luck, as we are talking of three laps here, not +50. So are those instances im qualifyig from Red Bull & McLaren simply explained by human error @keithcollantine?

    • Boomeranga said on 30th April 2013, 14:11

      You’re right mate, that’s the easiest way. To establish the volume of fluid, in this case volume of the fuel, is more chalenging. However, if you read the title of Keith’s article it is obvious they are not 100% sure of the amount of fuel spent and the amount left in tank. Oscilation of ambient temperature and some other factors add more variables in the equation. If you’re unsure about that ask Mark ;-)

  11. Optimaximal (@optimaximal) said on 29th April 2013, 15:50

    The McLaren/Red Bull comparison was interesting because in both instances, McLaren admitted it was human error that meant not enough fuel was loaded, whereas in both cases it’s been ‘an issue’ with the fuel delivery bowser – i.e. not the car, because of course they’re not running the car obscenely low on fuel, hoping to wing it.

  12. Scalextric (@scalextric) said on 29th April 2013, 16:04

    In addition, running flat out for 100% of the race may be avoided due to a variety of concerns: Tyres of course. Gearbox and engine wear. Brake wear. Team orders. Clear performance advantage vs. opposition. Weather. Fuel tank capacity. I’m sure there are more.

  13. DaveW (@dmw) said on 29th April 2013, 16:05

    Terrific article.

    I suspect that Mercedes has a too-small tank. The repeat of the lift and coasting going on in China, after the acute situation in Malaysis seemed, a bit strange, and really they should have been keen to avoid a similar debacle.

    In addition to the concern of reducting the mass of the car on the grid with fuel, teams obviously will want to reduce the size of the tank to improve packaging and lower the CG of the structure. I think that there should be a minimum size-tank and a minimum fuel load for each race. Now that there is no refueling, there is absolutely no benefit to the “show” and its not an interesting facet of technical competition, for teams to be invited to run down to fumes at the end. Obviously the total race ET will be a function involving mass and energy consumption and it may well be that the curve reaches maximum with a lower energy use than the car can carry. I dont want to see fuel-racing.

    • DavidS (@davids) said on 29th April 2013, 16:38

      I disagree, fuel consumption is one of the most important technical issues of the sport.
      If minimum fuel loads were introduced to increase the “show” I think it will come at the cost of technical innovation, which is a fundamental part of the sport.

      With a minimum fuel load and a fixed race distance, there is a minimum fuel consumption, which means engine developers will focus entirely on increasing power, as opposed to forcing them to develop for both efficiency and power.

      The other problem is that the calculation for what the minimum fuel load has to be moved by the regulations periodically to keep up with development. It also becomes a political issue if there is a consumption disparity between engines, as teams look to gain an advantage – efficient engine suppliers will want a much lower limit so they can run with less, and more thirsty engine suppliers will want a higher minimum to negate that advantage.
      No minimum fuel load means that it is self policing.

      • JerseyF1 (@jerseyf1) said on 29th April 2013, 17:43

        its not an interesting facet of technical competition

        Not only is at a very interesting part of technical innovation, but it is one of the major drivers behind the 2014 regulations as fuel is severely limited as is maximum fuel flow rate. @davids is correct, there is great relevance in (and teams should rightly benefit from) developing more efficient fuels and engines.

        With a minimum fuel weight for a race we would see development of ridiculous and completely irrelevant technology to burn off excess fuel as quickly as possible (this reminds me of the original 3 stage qualifying with full fuel loads where drivers went out to burn off fuel to earn fuel credits but qualify with minimal weight).

      • DaveW (@dmw) said on 29th April 2013, 21:54

        Yeah I suppose with a minimum fuel load cars would have special flame chimney on top like a refinery to burn off the “extra” gas. Kidding aside, I’m not sure that the engineers would then focus on “power.” With RPM limits (and boost limits), running richer is not going to get you that much more power. I still think a miminum fuel requirement would not take anything away from the technical competition. Indeed, the venue would shift to weight savings in static components of the car structure, which is the good, normal thing. And which does not result in a Q3 lap cancelled the next day or cars creeping around the track to save fuel for 20 laps. “Coasting” is not a word I want to hear in a race broadcast.

  14. DavidS (@davids) said on 29th April 2013, 16:10

    Interestingly, the official F1 website ran an article about fuel systems
    http://www.formula1.com/news/features/2013/4/14512.html

    I went there to look up a technical regulation and saw that.

  15. mantresx (@mantresx) said on 29th April 2013, 16:14

    Another interesting fact, although the teams are not allowed to change the fuel level of the cars once they are on the grid, the drivers often give a couple of laps before the race to shed some weight, cars like the Red Bull which are known to have a small tank don’t do many installation laps.

1 2 3

Add your comment

Your email address will not be published. Required fields are marked *

All comments must abide by the comment policy. Comments may be moderated.
Want to post off-topic? Head to the forum.
See the FAQ for more information.