The second fuel flow meter operates downstream of the first and its data is encrypted

How the FIA’s new encrypted fuel flow meter targets Ferrari’s suspected ‘aliasing’ trick

F1 technology

Posted on

| Written by and

The new, second fuel flow meter the FIA has added to Formula 1 cars this year is designed to prevent a trick Ferrari is believed to have used last season.

Ferrari insists its car conformed to the rules throughout 2019. However rival teams have criticised the FIA for reaching a settlement with the team without disclosing details of its investigation, part of which is believed to have concerned the fuel flow rules.

Since 2014, F1 teams have been required to fit a fuel flow meter which ensures their power units do not consume fuel at a rate greater than 100kg per hour. However last year some teams began to suspect Ferrari had found a means of exceeding the limit.

One theory held that Ferrari had developed a system known as ‘aliasing’ which could allow them to deliver fuel at a higher rate when needed, such as on qualifying laps.

The fuel flow meter measures the rate of flow very quickly – 2,200 times per second. However it remained theoretically possible for teams to deliver fuel to the meter in a sufficiently precise way that it appeared to be flowing at a slower rate that it was.

One technician consulted by RaceFans estimated the potential power gained from aliasing could be as much as 50bhp.

Last year Red Bull asked the FIA to clarify whether teams were allowed to use such a technique. The FIA responded by issuing a technical directive ahead of the United States Grand Prix stating it would be illegal. During that weekend Ferrari’s previous straight-line speed advantage appeared to have lessened.

The most recent version of the 2020 Formula 1 technical regulations, published 10 days ago, continues to refer to “a single fuel flow sensor” in the cars. However teams have been required to fit a second sensor under a further technical directive.

There are two important differences between the new fuel flow sensor at the original one. The new ‘FIA fuel flow meter’ samples the fuel flow rate in a different way, making it harder for teams to get around the limit, and the data it generates is encrypted and available only to the FIA.

Manufacturer Sentronics says this will make it “impossible” for any team to exceed the fuel flow limit.

Go ad-free for just £1 per month

>> Find out more and sign up

“The new device incorporates anti-aliasing technology and full data encryption,” according to Sentronics. “The anti-aliasing technique randomises when the device makes its measurements, ultimately making it impossible to synchronise any ancillary parts to the measurement frequency. Full encryption ensures the authenticity of the data and privacy to the FIA.

F1 fuel flow sensors
The new blue FIA fuel flow meter is encrypted
“This eliminates the possibility of the data being used as part of a feedback system to gain a competitive advantage.”

The new fuel flow meter was added to the cars at pre-season testing and will continue to operate alongside the existing device. The FIA will retain a pool of the new meters and allocate a sensor to each car at the beginning of each event.

“This new variant of the FlowSonic fuel flow meter is not only one of the most technologically advanced currently available, but is an important step forward in improving the FIA’s policing of the maximum fuel flow regulations in F1,” said Sentronics managing director Neville Meech. “We’re proud to lead the market in solid-state fuel flow meters, and to demonstrate our ability to develop world-class technology in rapid timeframes.”

The FIA announced last month it had reached an settlement with Ferrari following its investigation into their power unit but would keep details of its finding secret. This prompted an angry response from the seven non-Ferrari-powered teams, who have demanded clarifications from the FIA as to whether Ferrari operated within the rules at all times last year.

In a further statement, the FIA admitted it was “not fully satisfied” Ferrari’s power unit operated “within the limits of the FIA regulations at all times” during 2019. The careful wording indicates they suspected not that the power unit was illegal, but that it may have operated outside the rules at certain times.

Ferrari “firmly opposed the suspicions and reiterated that its [power unit] always operated in compliance with the regulations,” the FIA noted. Of course, this did nothing to placate the seven teams who complained.

As those teams already had details of the new fuel flow sensor, the fact it is so clearly aimed at preventing anyone getting around the fuel flow rules shows why their suspicions were so strong.

Advert | Become a RaceFans supporter and go ad-free

The second fuel flow meter operates downstream of the first and its data is encrypted
The second fuel flow meter operates downstream of the first and its data is encrypted

Advert | Become a RaceFans supporter and go ad-free

2020 F1 season

Browse all 2020 F1 season articles

Author information

Keith Collantine
Lifelong motor sport fan Keith set up RaceFans in 2005 - when it was originally called F1 Fanatic. Having previously worked as a motoring...

Got a potential story, tip or enquiry? Find out more about RaceFans and contact us here.

73 comments on “How the FIA’s new encrypted fuel flow meter targets Ferrari’s suspected ‘aliasing’ trick”

  1. Although it seems like Ferrari were clearly (but unprovablely) breaking the rules, you have to admire the imagination and precision engineering it’d must’ve taken to make that work.

    1. The same way Mercedes are doing it with DAS, Brawn with the double diffuser and other teams have done it for quite some time. As all other teams, they found a loophole . Unfortunately, due to a long-lasting bias, when it’s Ferrari, it’s cheating.

      1. Duncan Snowden
        16th March 2020, 14:22

        Er, no. Finding loopholes in the rules is different to circumventing enforcement measures. As I say below, while it’s ingenious, and I’m genuinely impressed, it’s a bit like footballers claiming that if the ref doesn’t see it, it isn’t a foul. Compared to, for example, discovering that under a strict reading of the rules, if you hold your hands above your head while whistling Colonel Bogey, it really can’t be counted as a foul whether it’s seen or not.

        1. Antony Makanza
          17th March 2020, 7:02

          @Duncan Snowden, if it was that obvious that Ferrari were ” circumventing enforcement measures” ….. I am sure they would have caught them by now. The FIA has been very careful in their choice of words. I think we too should be careful

      2. no, it really is cheating when you circumvent the measurements of a rules application. It’s like professional athletes providing fake urine samples to circumvent doping tests.

      3. Ferrari Cheaters
        16th March 2020, 14:53

        This is not a loophole, there is a clear number of how much fuel you can use per hour. When you use more by tricking the fia’s sensor you are clearly cheating.

      4. No. Reverse engineering a sensor to circumvent it is clearly cheating and not a loophole.

      5. As others have said, this is not “finding a loophole”. The rule isn’t “the FIA sensor must only measure 100kg/hr”, it’s “the fuel flow must not exceed 100kg/hr”.

        If you fitted some stealth material to your car so speed cameras wouldn’t detect you exceeding the speed limit, that would not mean you could legally drive at 200mph.

        1. When the wings on the Red Bull were flexing, clearly visible for all to see, it was the same, they past the FIA-tests so they were declared legal.

          1. It’s not EXACTLY the same since the regs can’t say “the wings must be rigid” or “wings cannot deform under aerodynamic load” as that is a meaningless statement (any material will deform, as physics; you need to say how much deformation is allowed). The regs said the front wings must pass a static load test, which the RB wings all did).

          2. The problem there was the fact that, at the time, the rules only specified a set deflection at a set static load applied at a set location on the front wing – there was nothing in the regulations that stated that the front wing had to have a linear load-deflection curve.

      6. A simple detail, DAS is not illegal (it´s banned for 2021 not 2020) and, if it were banned for 2020 season, still would not be cheating since Mercedes has not used the device in race

        1. Plus, the system was developed with the FIA having knowledge of it. They weren’t hiding it. There’s a bit of a difference.

      7. Not a Ferrari fan
        17th March 2020, 5:25

        + 1 if Ferrari make a step up, others spruik that they must be cheating.

      8. The fact that you’re comparing the tricking of a fuel flow sensor vs innovation that takes advantage of regulation loopholes, shows that you really shouldn’t be commenting on this issue at all.

    2. We still don’t know for sure this is what Ferrari did, it’s just one theory. It would funny if Ferrari are telling the truth and had some other trick that had nothing to do with the fuel flow.

  2. Thanks for clearing that point @dieterrencken. I must say, it’s hard not to be impressed by the technical creativity of these teams to find even the smallest gap to inject some boost from towards their cars!
    That the teams were able to replicate what Ferrari have done (but cannot probably be proven beyond enough doubt because it wasn’t measured) in relatively short time also shows some superb analytical thinking and problem solving skills.

    So if I understand it right, the trick with the aliasing would be to inject a tad extra fuel in between the measurement cycle – which would therefore not be exactly measured, compared to at the time when it IS actually measued.

    I am sure this skill at super precision injecting of fuel (or any liquid) will come super handing in real world applications in various engines, brake systems, hydraulics controls, aerospace as well as the chemical industry in the near future!

    I think that now that the season is off for the forseeable future, we should get these engineers to get some work done towards better ventilators, instruments to measure temperature and fysiological changes in people etc, to make top notch diagnostics tools, improve protection of vulnerable people and improve care for those suffering! Since the Ferrari Factory is currently shut down anyhow, I would hope the settlement covers Ferrari to provide these skills to the world at their own cost.

    1. @bascb

      Looks more like Ferrari were monitoring the fuel flow sensor and causing EMI to detail offset the sensor, so that they could flow more. It all happens in frequency domain around 2kHz it seems, hence the sampling / aliasing reference. It’s slightly more tricky to do in a highly dynamic state, probably why FIA finds it hard to prove. It is not just offsetting a simple value. In the end engineering wise not such a great feat. If you make the signal digital (could have already been digital, but could also have been analog) and encrypted the Ferrari seems can’t be done anymore. It won’t be able to measure what it needs to offset. System defeated..

      Let’s be clear it is more in line with VWs diesel gate than with the true spirit of the engineering rules. But that is F1 also.

      Nothing to do with the injection directly that I read, that is on the engine down stream of the sensors. Current day petrol and diesel injection is already far advanced, don’t think it needs F1.

      Just my two cents, but I do work for the worlds biggest supplier of automotive sensors :)

      1. @maxv – ah, so you’re saying that the readings which were taken were being altered to show a lower flow, either before the ADC or after? And not sneaking in more fuel between the pulses. Which is why encryption defeats that. Why would they have also switched to a randomized sampling interval, any thoughts? Just to add another level of robustness?

        Let’s be clear it is more in line with VWs diesel gate than with the true spirit of the engineering rules. But that is F1 also.

        Agreed. But as @dvary says, the engineering ingenuity behind such a thing is in itself impressive, and something that excites engineers.

        1. @phylyp I would think before ECU ADC. Typically the real filtering power is in the ECU’s. The sensor just provides the signal although it can do its own ADC / DAC on input (noisy sensing principle) vs. Output signals. Depends really on what you want. It being F1 I think they would go for fast, unfiltered analog, for the system to use the max detail of signal.

          The aliasing remark on the other sensor they added might mean they have increased its filtering bandwidth.

          Some more explanation to simply it for others:
          The simplest form of sensor EMI (Electro Magnetic Interference) is just offsetting an analog output with a field or coupling in to the leads of the sensors. All every day cars have specs for this. Under a field E.g 0.5V output could be driven to become 0.45V. This means you can now flow more up to 0.5V. In reality the reading should have been 0.55V. Typically 0.05V is like 1% accuracy, so 1% more fuel. Now imagine that in frequency domain and possibly digital. It still does roughly the same, just some more time to engineer it.

          1. @maxv – thank you, very insightful.

          2. Another option is that the FIA sensor was not always sampling and that they could figure out when and how long it was sampling. And then just flow more in the not sampled period. That would be hard to find out as well because the Ferrari defeat device is not going to trigger if you increase sampling. Sounds like a too simple system though, would be a crappy regulatory device..

        2. Patrick Chase
          20th March 2020, 0:41

          My guess as an engineer but not an insider is that they used a piezo actuator or something similar to induce a standing pressure wave at the sampling frequency, such that the fuel was moving through the sensor more slowly (under less pressure) than the average during the sampling period of every cycle.

          IMO the only really tricky part would be to maintain phase alignment between the induced wave and the sensor’s sampling, which is why they would have needed realtime access to the sensor’s output. That’s likely why the new sensor encrypts its results.

          1. Patrick Chase
            20th March 2020, 0:45

            On related note, I hope that the designers of that new sensor were aware of and addressed things like power attacks (using measurements of its power consumption to determine when it’s taking readings).

    2. So if I understand it right, the trick with the aliasing would be to inject a tad extra fuel in between the measurement cycle

      (@bascb) No, that’s not how one would use aliasing to fool the FIA, not that what you suggest couldn’t be used, except that the FIA are supposed to vet the software a team uses in the FIA mandated Engine Management System, and that such trick would be obvious in the software (although it might be if you weren’t looking for it you might not notice it, although if Ferrari had done such a thing they were definitely needing a lot of luck because if it was noticed your goose would be cooked).
      Basically, when sampling, there’s a thing called the Nyquist rate, which is the minimum number of samples required to accurately reproduce a waveform. So, if we consider a sine wave, the minimum number of samples required to reproduce a sine wave is two, one to reproduce the positive peak and one to reproduce the negative peak. So, for example, for an audio system the highest frequency you might want to reproduce could be 20 kHz, so the Nyquist rate says to reproduce the audio signal it would, at a minimum, have to be sampled at 40,000 times per second. If, due to say audio filter bandwidth rolloff, a frequency higher than 20 kHz was noticeable to the Analogue to Digital converter (which converts the analogue samples into digital data), say 20,100 Hz, then aliasing occurs, so ADC would record the frequency as 19,900 Hz.
      According to Dieter (@dieterrencken), the FIA’s Fuel Flow Meter sampled at 2200 times per second. That means the FIA were expecting their highest “frequency” to be 1100 Hertz. If a frequency presented to the fuel flow meter was higher than that, e.g. 1101 Hz, then the fuel flow meter would think it was 1099 Hz.
      Of course, if the FIA suspected “a team” were using an aliasing trick to cheat, then why didn’t they increase the sampling rate, e.g. to 3000 times per second? Okay, that would have needed a lot of things to happen, such as increasing the data storage capacity, upgrading the ADC, improving the CPU, etc.
      As it is, I don’t see why the additional piece would be better unless it uses a higher rate of sampling. If it used the same sampling rate (and the same other bits and pieces) as before then why couldn’t a team just keep doing what they did before?

  3. Thanks for shedding light on this, as well as how the second sensor is different.

    One question – back when these PUs were introduced it was a company called Gill (in the UK, IIRC) that were making these fuel flow sensors. Is this Sentronics the same as Gill, or have they sourced this second sensor from a different company?

    Manufacturer Sentronics says this will make it “impossible” for any team to exceed the fuel flow limit.

    “The anti-aliasing technique randomises when the device makes its measurements, ultimately making it impossible to synchronise any ancillary parts to the measurement frequency.

    Ooh, never say “impossible” guys, particularly not in F1.

    Is the sensor truly a well-isolated and shielded piece of equipment? Can the random pattern be inferred by minuscule power draw differences, pick up coils, or something else? I’m sure Ferrari and the other teams will be pondering such thoughts while waiting for the resumption of racing.

    1. Sentronics replaced Gill as the homologated F1 supplier in 2018. Both still supply flowmeters to other motorsport categories.

    2. Sentronics is a different company. They were suppliers since recently I think.

    3. Ooh, never say “impossible” guys, particularly not in F1.

      Absolutely agree with that.

  4. During that weekend Ferrari’s previous straight-line speed advantage appeared to have lessened.

    Leclerc was using a spare PU and Vettel retired in that race. Ferrari were quickest again on the straights in both Interlagos and Abu Dhabi. It might be one of the tricks used by Ferrari to bypass the FIA sensor for extra boost in qualy but I highly doubt this is the only reason behind the mighty power advantage. As removing this trick would only require the team to reprogram its PU routines. The Ferrari guys have redesigned a completely new PU for the 2020 season.
    Many thanks for this insightful article as usual.

  5. Duncan Snowden
    16th March 2020, 14:13

    Ah, that makes sense. So if I’m reading this right, in simple terms, they increased flow when the sensor wasn’t… sensing, and decreased it when it was? (I’m guessing, since they were unlikely to know precisely when it was and wasn’t, and we’re talking about 2000 samples per second, they got clever with frequencies and Nyquist’s law, or something like that.)

    As so often in F1, you almost have to admire cheating like that. It’s ingenious. And, in fact, you can see the FIA’s dilemma: if a system is designed so that it looks perfectly legit to the official sensor, the sole, impassionate, inhuman, arbiter of this regulation, is it really cheating at all? (Although that’s uncomfortably close to the, “If the ref. doesn’t see it, it isn’t a foul” line.) After all, they couldn’t increase flow arbitrarily; they were still limited by the sensor, just not to the rate it was intended to sense. Well played, Ferrari. But you’ve been caught.

    1. The fuel flow meter is an input to the standard ECU just like all the other sensors relevant to the power unit and the rest of the car – hence the new need to encrypt its readings. Pre-encryption, I don’t believe there would have been anything stopping Ferrari using the readings directly in their fuel control system, so the control wouldn’t necessarily have to use very sophisticated theory to work with the sensor sampling. They could have started with their best guess at the modulation they needed and then twiddled the phase and frequency until the dip in fuel flow meter reading showed they had found the synchronisation needed for maximum effect.

      1. Duncan Snowden
        16th March 2020, 20:59

        Ah, okay. Interesting. Still ingenious, though.

    2. The original fuel flow meter measures the total amount of fuel that passes through in 5 second intervals. So if during that interval you use slightly more than the 100 kg/hr limit at one moment but slightly less at another moment in the same time slice, the total will still be OK.
      This is presumably done because these engines use fuel injection. Each injector releases fuel in a short burst into the cylinder, which causes a choppy fuel flow – but a 5 second interval is wide enough to smear out the readings to a value very close to the average.

      If you can read the sensor data however, then you can adapt the fuel flow so that it precisely matches the maximum amount in each time slot. So when you exit a corner, when the first part of a time slot was not on full power, you can use as much fuel as you like for the remaining few seconds as you would not surpass the total maximum for that slot.
      More power, especially in the first few seconds after a corner.

      Of course that way you also use more fuel than the 105 kilogram you can use for the race. To save fuel you’d use slightly less downforce, so you’re faster on the straights but slower in the corners. That way you spend less time on full throttle. At least, that’s the theory.

      Ferrari was running a low downforce setup but switched to higher downforce levels once the directives were brought out. Pure coincidence of course. It would be impossible to prove, we’d never know unless an insider breaks out.

  6. All past articles on this subject have been generalised but this one goes one step further in explaining the situation. Still, it’s not far enough as the key to just how this worked, IMO, lies in what the 21 questions actually asked by the teams of the FIA actually were. If i read this article properly it appears that flow rates were altered despite the sampling taken 2200 times per second!!! To interfere with those measurements by increasing the fuel flow at precise times undetected would surely be feat of some magnitude. The fact that the FIA could not precisely prove Ferrari interfered with the fuel flow speaks volumes. It’s one thing to make allegations and another thing to prove those allegations. The key is in the teams questions as they would be precise in asking what they could or couldn’t do if they decided to emulate what they thought Ferrari were doing. I am still not convinced that Ferrari operated illegally. If they did then they should be penalised but if not then they should cleared by the FIA. This is too big a matter to brush under the carpet.

  7. I agree. It is still all speculation. I find it hard to imagine that ferrari could have this level of control over the fuel flow to deceive the sensors without there being some form of obvious control software and/or hardware given the amount of scrutiny that the engine has been under. Especially as it is all controlled by a spec ECU.

    Didn’t the FIA install monitoring components (which I am sure at one race or another included a second fuel flow meter) throughout last season to ensure that the engine was running within the specified parameters? Like I said initially, it is all speculation in response to Redbull’s usual rumour mongering and the FIA’s inclusion of a 2nd fuel flow meter.

    An ingenious idea but evidence of nothing.

    1. Actually I am wondering how you can manage such fuel flow precision when the car is actually driving through non-smooth surfaces… Vibrations may just produce such bad effect… the only way I think they can manage it is to check the surface state on Friday and to configure the PU accordingly. However that may require to know in advance the configuration of the car suspension… so the problem is even more complex than just setting the PU.
      On the other hand, perhaps Ferrari might just check the track surface and by measure some quantities they know where they can use this hypothetical system.

      Perhaps this is why they were not fast at COTA anyway…

      1. Vibrations are irrelevant. Electronics doesn’t care for the most part, apart from an assembly disintegration due to bad design.
        E.g Does your engine stutter when you drive on a bad road? Does your radio skip? Your Bluetooth connection disable?

    2. @Asanator, you just said it yourself, the second monitoring device was installed AFTER.

      It’s like you are pretty sure a maid of workman has been stealing from your home so you install (or say you have installed) a security camera system. And no items seem to have gone missing after that.

      You couldn’t prove the thefts in the first place, so how does a lack of thefts afterwards actually prove the initial allegations? :)

  8. Martin Elliott
    16th March 2020, 14:50

    Some, but not enough detail.
    Gill (2014) never published full specification of their meter, but it must have had a range >100kg/hr AND a 0.1% resolution for FIA to disqualify Renault for a 0.1kg/hr infringement!!

    The Sentronics claim only a 500:1 turndown (max to min flow) and resolution of 0.5%, so is FIA going to adjust the fuel flow allowed criteria?

  9. If a team employed a hacker to find a way around the fuel meter, what’s to stop them trying to work out how to bypass this new one? Many companies have discovered, to their cost, that their security systems were not robust enough to defeat determined crooks. If I were on the case, I’d have bought a FlowSonic Elite from Sentronics to see how it works. Sentronics also supply a data sheet! You can be sure all the F1 teams did. It’s likely the new FIA meter can be purchased too.

    1. Encryption is much more difficult to “hack”, and if they do it right it’s practically impossible (not within a human lifetime)

  10. Cheated and still couldn’t win. Awkward.

    1. The Tifosi are still proud.

  11. So just a few thoughts.
    Wasnt the fuel flow increased to 110kg/hr not just 100.
    Wondering how they came to the number of 50 extra hp?
    Lecrec was over by 4.88kg in AD last year right?
    So quick and dirty math, says that’s around 1.5 gallons of extra fuel? Over a 2 hr race?
    So .75 gallons, 2.9 liters, of extra fuel per hour.
    So we are talking what, less than 2oz, or about 60 milliters per minute? Not sure how they could be getting so much more performance from such a tiny increase in fuel.

    Please let me know how I’m wrong as I sire I probably am…lol

    1. 50 HP in bursts on the straights. Once a cheat car leads by 10 seconds, they don’t need the extra HP, it back to saving fuel, like the rest of the field.

  12. There is a problem when running 2 sensors, you can’t know which one is correct, these 2 measure in different ways, nevertheless they ought to be calibrated but it should be hard to know how much they deviate from one another, as with the 737-max running 2 sensors may be pointless, you can only trully trust one dataset, that is why boeing turned the 2nd sensor off but the max debacle is much more complex than that.
    Perhaps the fia can learn something by comparing the data even if the data isn’t exact.
    I still wonder if teams don’t store fuel after the sensor, that yellow injector, I wouldn’t be surprised if that isn’t exploiting something.

    1. That’s why you often have three or four. But it’s not like these cars will fall off the road if there’s a fuel flow sensor disagreement.

      I’m guessing the sensing of the second sensor will be at different frequencies and the numbers should line up within reason. The second sensor frequencies will either be fixed or semi-random, and the data encrypted so that Ferrari can’t see the pattern and try to fool that one too.

    2. Indeed, probably a Boeing engineer who designed this sensor system.

      1. Accounts for the crashing…

  13. Reading this makes it seem that Ferrari were able to fool the original first sensor into thinking that the fuel flow was within limits while it wasn’t. Hence FIA have introduced an encrypted, more secure 2nd sensor.

    This feels more like the Volkswagen diesel-gate situation rather than anything ingenious like the DAS or the F-duct or the double diffuser.

    No wonder Ferrari didn’t want details of this investigation to be released to the public. They would have been the next Volkswagen.

    1. They would have been the next Volkswagen.

      The next Mercedes (Daimler) you mean…

      1. You need to win to be Mercedes.

  14. Lovely article; it makes it a lot clearer to me (I think).
    ‘Aliasing’ is like the wheels turning slower/backwards on a speeding car when shown on video.
    FIA suspects Ferrari to drive too fast but has only the video to (dis)prove it. Ferrari points at the video and says it shows that it is not going too fast; just look at the wheels.

  15. With an abundance of cash flow Ferrari has at their disposal and the absence of a championship in many years, I can certainly acknowledge Ferraris appetite for a win at all costs. A remarkable example of ingenious engineering. Sadly they may have been caught but this remains speculation. Cant wait to see the next saga..

  16. This theory is purely that and belongs in fantasy land. The second flow meter was installed to placate merc and rbr who were accusing ferrari of exceeding the flow. it was fitted with the full co-operation of ferrari as they knew that they were not breaking any rules.
    but even if there was any truth about the theory, then it would still mean that there was no cheating as the car complied with all existing rules.
    DAS by merc is a system that avails of the loopholes in the rules. so does that mean that they are cheating too?

  17. I’m sure since last year Honda, Merc and Renault are looking into ways of replicating the effect somehow. Either to use it for themselves or to expose Ferrari.

    That’s why Ferrari cannot disclose what they’ve done. It would start an arms race.

    Now with the second sensor, data encryption and random sample rate it should not encourage anybody to go on that path anymore.

    1. This was a very ordinary cheat, no brilliant technological invention.

  18. As I understand the aliasing technique used by Ferrari, is something like an exploitation of the fixed polling frequency of the previous sensor, and combining their previous fuel flow history or sum to create a temporarily higher than allowed fuel flow rate for a small moment before the next actual polling, and of course using all previous ingredients to comply the rules at that next polling event too.
    Imo it’s against the spirit of fuel flow rules, and it’s quite much against them. (I’m a Ferrari fan to some extent, although I have other favourites too, and it happened to me to support drivers instead of teams, it’s funnier that way :) ) But anyway I don’t want to see circumventings like this.
    More or less I understand things like antialiasing and splines, or problems with this fixed interval polling. Imo this is just nasty and not that clever, although not that easy to put together also. Probably it’s harder to find a leak on the rules or to design the rules to express intentions and shut the gate for hackers, than putting together a system like this fuel flow manipulator.
    While you have a system like this, you are not guaranteed to have more fuel at a given point if you burnt at high rates previously.
    It’s very strange to me that they simply relied on a fixed rate polling and not taken total consumed fuel from previous moments into account. Especially seeing how rigorous they had been when Renault consumed negligibly more than allowed. (I felt their automated brake balance setter case much worse, nowadays rules was not allowing things like that by their spirit… so it looks like occasionally many teams may get away with being a bit like Dennis the Menace).
    At 2200Hz frequency and maximum allowed flow you can burn 1000000 grams / 3600 seconds / 2200 = 0,0126262 gram of fuel :) The aliasing idea is painfully simple, but not fair and hard to implement at this precision.
    I like FIA’s fix, especially making flow data hidden from teams. But this sensor is just much more complicated, like Ferrari’s system to get around the previous.

    Would not it be better to limit the fuel flow only by the tank capacity, basically for free?
    Would it not be greener or more fair to have an even more strict cost cap with engines included and engines’ R&D costs between factory and supplied teams, and then donate a lot of spared money to charity or more commonplace research?
    They could still have 1000+ bhp, loud, non hybrid (so atleast cheap) engines, they could have the thrill whether an engine lasts one single race, and then it goes into the melting furnace or into the museum if they won something significant while using it. Those engines likely would not cost around 10M$/unit, what is a large part of cars built, and their R&D is also a very large part of the engineering costs.

    DAS is more fair, and at least they not tried to hide it at all, but I don’t like the hackiness of it, because the spirit of the rules are against suspension adjustments at the race, so it had to born this way. It’s a nice feature and can improve tyre lifetime, what is good for safety, but what if it’s so good that Mercedes will beat every other car with 1secons/lap, then it will likely banned before the end of the year.

  19. The other teams should demand that Ferrari has their fuel weights measured every race. It was extremely unfair when the second car wasn’t check last season when one car went out with extra fuel.

    Definitely cheating. The rule specifies a maximum fuel flow rate. That is the rule. Ferrari was exceeding that rate and hiding the fact by gaming the gauge used to measure the rate. It isn’t a technical improvement to make the cars faster, they were just using more fuel, which any competitor could have done to go faster.
    Thinking of the times they “missed” the weigh bridges. Hiding their fuel weights most likely.

  20. Being an F1 layman, sometimes it’s extremely difficult to fully comprehend the complexities of this “fuel flow’ drama. There are just so many questions that need to be answered. One question i have for the experts concerns fuel pump pressures. If i understand correctly the fuel pump must be calibrated to deliver a maximum pressure allowance as set by the FIA to a max of 100KG/PH. How then can the pressure be altered and delivered to support an increase above the former set FIA allowance? It would appear that, prima facie, the single sensor can be fooled given the allegations but how does the fuel pump settings become flexible on demand? Maybe one or more of our experts can throw some light on this for me…Thank you in anticipation.

  21. This would all be solved if they gave each team the same exact amount of fuel for the entire weekend. Let them do whatever they want to get through the weekend as efficiently or strategically as possible.

  22. Let each team have the max fuel qty for the race, then let them race. Who cares if they burn it early in each race, or late in a race, or decide to not take-on board the full allocation.

    Let them race, pls

  23. There are too many rules, set a maximum engine capacity and minimum car/driver weight, allow tyre competition and let the cars race.

    To control costs raise the ride height and mandate a reference wing design, since most money goes on aero and that ruins the racing it could be neutralised with a couple of simple measures.

  24. Why have they installed a fuel flow sensor in the first place? Why not install a fuel flow limiter that limits the flow to a maximum amount then there would not be a need for a flow sensor?

    1. @aliced that’s what DTM/Super GT currently do. Paddock rumours report a variation between fuel flow limiters of +/- 5%….That’s a huge variation open to exploitation! F1 teams will end up buying 1000’s of them and cherry-picking the best ones (that doesn’t make much sense)… Maybe they could make a better F1 version but it must be a fairly dumb bit of technology (i.e a fixed diameter bit of pipework that sonically chokes the flow) that’s probably way easier to cheat than the new flowmeter.

      Looking at a number of different forums/reddit the F1 teams are apparently reporting a spread of +/-0.1% between the sentronics sensors which is quite a technical feat if true. If it can deliver that accuracy under different fuel temperatures/pressures then it sounds like an awesome bit of technology that I should fit to my new VW to prove the fuel economy being displayed on my dash is a load of horse “ahem”!!

      As all the powertrains hopefully reach optimum performance (given the current set of design rules), these fuel flowmeters (and the current flow meter used on the battery systems ) are going to be critical to policing that the engine suppliers/teams are playing by the rules… I just hope that whats results this year is more level racing and not the procession its been of late (apart from an odd couple of races).

      I personally want to remove all the fancy technology and let the racers race, however, I do see the conundrum faced by the FIA that a number of these types of sensors are required to police formula 1 as its currently formulated.

  25. I wonder how much ferrari paid the FIA to keep quiet? They could not win by simple machanical engineering so they had to cheat simple. And mercs dad system got disclosed on first day of testing to the FIA which have deemed it legal for 2020 so no cheating there then. End of discussion.

    1. get over it. there was NO cheating, just inventive exploitation of the rules, IF the theory was in fact fact. !
      as regards the merc DAS, it was not revealed by merc on the opening day of testing….it was spotted on the on-boards and questioned. they disclosed that they had discussed with the fia during development, but that does not mean that it has been deemed legal. that cannot be done until merc use it in a race and then someone protests it. only then can it be adjudged legal or illegal

  26. @ Owen Lindsay. I fully concur. People tend to laud Mercedes about their DAS openness but let’s face it, it was almost impossible to hide. I mean you just don’t move steering wheels like they did at 300+KPH for no reason at all and it was bound to be noticed and questioned. One question would be why is there an arbitrary limit of 100KG/HR set in the first place ? I rather lean towards having a set amount of fuel allowed and letting the teams exploit to their hearts content just how they consume that. To limit fuel use as a predetermined
    KG/HR is always going to lead the teams to looking for an competitive edge.

    1. I read an explanation where they wanted a max fuel flow per second so the cars would not have a large disparity in speed when approaching another car.
      Seems unusual in that these are supposed to be the best drivers in the world why would that be a concern? I agree a set amount for the race and let teams flow whatever they want whenever they want, if they run out of gas too bad.

  27. The nosecone was flexible, that is why it appeared the wing was flexing, and the wing passed the load test. A video of a Red Bull pit stop showed a mechanic pulling on the wing and the nose cone flexed quite a bit(at least 10 degrees) in the longitudinal direction.
    I can no longer find the video but remember it clearly.

Comments are closed.