New 2015 fuel flow Directive

[toraabe]

[...]If you can get that that to function in a stable and controlled way and show me a controllable increase of mass flow at the end point of a system I'll be highly impressed.

And even if you did manage that, the injectors like a bit of pressure to atomize the fuel properly so that it is distributed in an even way in the cylinder and therefor burn more stocheometrically (I probably spelled that word wrong). It is not so important with throttle body or intake injection because the fuel has more time to vaporize than in a direct injected engine. If you had so low fuel pressure that you'd vaporize parts of it you'd also just put a blurb of fuel on top of the piston that wouldn't burn right and it'd combust in the exhaust instead (and you'd have wonderful rally anti-lag sounds *dreaming*).

I think there is a misunderstanding here.

There is no possible "cheating" wrt the fuel mass flow over a certain prolonged time, let's say over one lap (indeed, it would be even a shorter time...) with the idea Gruntguru proposed.

But, it is well possible to use the elasticity of the fuel transport system and/or the "elasticity of fuel in fluid vs. gaseous phase.

And, it has nothing to do with the rail pressure, i.e. the pressure the injectors are seeing, which is between 250 and max. 500 bar as far as info has been released.

All this must happen before the High Pressure Pump (HPP) which produces thede 250-500 bar and after a feeding pump. This pump could feed constantly 100kg/h, and be pressure limited to xx bar.

- In case of higher demand >100kg/h of the injectors/HPP, it would feed in the volume limited mode. The pressure in the fuel transport system falls, and elasticity/bubbles are produced. Mass flow sensor reads 100kg/h.

- When a certain low limit of the feeding pressure is reached, the engine has to reduce mass flow to 100kg/h.
- As soon as there is less than 100 kg/h used (braking, cornering), the feeding pump, still delivering 100 kg/h can repressure the system, until its pressure limit is reached, compressing the fuel bubble to liquid and/or expanding the fuel lines or whatever.

This can be made with very few electronic controls, one of the most important things being the fuel pump feeding constant volume independently from the pressure, but limiting this flow when a certain pressure level is reached.

Another thing is that you can have constant flow to the injectors, but the injectors can accumulate fuel so they can inject pulses up to 500 bar. Actually they have to accumulate fuel because they are delivering fuel in pulses. Not the common rail fuel pump in which is constantly delivering 100kg/h max.

[Abarth]

[...]FIA are not complete idiots. [...]

I agree.
And mandating fuel pressure measurements on several points means nothing else that they have a strong suspect but no evidence that something is going on.
Why would they mandating it otherwise?

[Cold Fussion]

Why is the engine thread locked in favour of this thread?

[itix]

But I still have an issue with your original statement, because the measurement is actually an estimate of flow velocity, and corrected for fuel type, pressure & temperature. I guess RBR thought they had a better measure of mass flow and chose to ignore the FIA sensor (they had the right to do that, if the sensor was in error). The FIA proved that were not complete idiots by declaring their sensor was the standard, and everybody else was wrong. Incidentally, I believe that output of the device was digital, not analogue.

Sound travel trough various density fluids with various speed which is what I assume they use for measuring the density. If you have density and fluid velocity you get mass flow W = D^2 x rho x pi x v (bernouli equation). We use the speed of sound in my industry (offshore sesimic) to measure varied density in materials with sound.

Another thing is that you can have constant flow to the injectors, but the injectors can accumulate fuel so they can inject pulses up to 500 bar. Actually they have to accumulate fuel because they are delivering fuel in pulses. Not the common rail fuel pump in which is constantly delivering 100kg/h max.

I think there is a misunderstanding here.

There is no possible "cheating" wrt the fuel mass flow over a certain prolonged time, let's say over one lap (indeed, it would be even a shorter time...) with the idea Gruntguru proposed.

But, it is well possible to use the elasticity of the fuel transport system and/or the "elasticity of fuel in fluid vs. gaseous phase.

And, it has nothing to do with the rail pressure, i.e. the pressure the injectors are seeing, which is between 250 and max. 500 bar as far as info has been released.

All this must happen before the High Pressure Pump (HPP) which produces thede 250-500 bar and after a feeding pump. This pump could feed constantly 100kg/h, and be pressure limited to xx bar.

- In case of higher demand >100kg/h of the injectors/HPP, it would feed in the volume limited mode. The pressure in the fuel transport system falls, and elasticity/bubbles are produced. Mass flow sensor reads 100kg/h.

- When a certain low limit of the feeding pressure is reached, the engine has to reduce mass flow to 100kg/h.
- As soon as there is less than 100 kg/h used (braking, cornering), the feeding pump, still delivering 100 kg/h can repressure the system, until its pressure limit is reached, compressing the fuel bubble to liquid and/or expanding the fuel lines or whatever.

This can be made with very few electronic controls, one of the most important things being the fuel pump feeding constant volume independently from the pressure, but limiting this flow when a certain pressure level is reached.

I'm going to answer you two in compound.

Sure I get the proposed layout of the system, but creating such a low pressure on the suction side of the pump and getting the fuel to evaporate at a constant rate would be immensely difficult in my eyes. First you'd require some form of pressure/flow reducer valve to keep a constant flow before the HPP (from the low pressure supply pump and through the sensor), then you'd need to heat the fuel, then you'd need to make sure the gas bubble doesn't reach the HPP. Also it would be very difficult to control where the cavitation bubble would appear. This seems like such whichcraft to me that I highly doubt it. Less alone hiding all this stuff from the prying eyes of the FIA.

I actually assumed there was some kind of high pressure pump in the fuel system. I also assume there is a pressure regulating return valve like there is on diesel common rail fuel systems, but I could be wrong... I believe it is difficult to build pressure like that without a plunger type pump or gear pump and that would require a return valve that keeps the pressure steady.

I agree.
And mandating fuel pressure measurements on several points means nothing else that they have a strong suspect but no evidence that something is going on.
Why would they mandating it otherwise?

I'd agree there... I still can't see anything proposed so far that you would be able to hide though. The accumulator would be the one that make the most amount of sense but I still fail to see that one could hide it well enough.

[Arco Cigomático]

Hello everyone. First of all I apologize for my English from "google translator".

It seems clear that some teams are getting inject more than 100 Kg / h of fuel into the cylinders at certain times without being detected by the sensor beside the fuel tank.

I would like to know your opinion on my theory of how they do it.

For me the key to everything is at a point of technical regulation that I have not seen mentioned in the thread:

5.1.5 Below 10500rpm the fuel mass flow must not Exceed Q (kg / h) = 0.009 N (rpm) + 5.5.

We always talk about the maximum fuel flow is 100 kg / h, but it is not. That is the limit, but below 10,500 rpm, peak flow has lower limits.

For example: If the motor is 8000 rpm, applying the above formula:

Q = (0.009 x 8000) +5.5 = 77.5 Kg / h

My guess is that those teams who are skipping the norm, manage to keep revolutionized the car above 10,500 rpm at certain times for extra fuel in the circuit and when it exceeds 10,500 rpm adds that extra accumulated 100 Kg / h.

In the above example, we would have a surplus of 22.5 Kg / h at a time if the car had to go to 8000 rpm but will actually 10,500 rpm.

The flow sensor, in such cases would not detecting any excess.

I'm not an engineer and surely be wrong, so I'd like your opinion

[Richard]

It seems clear that some teams are getting inject more than 100 Kg / h of fuel into the cylinders at certain times without being detected by the sensor beside the fuel tank.

Is it clear? What evidence?

All we know is that the FIA have developed more thorough checks. That doesn't mean someone has been caught cheating, it could just be the natural evolution of the rules.

[mrluke]

I think its a knee jerk reaction to Red Bull wanting something to be banned / legislated.

Bearing in mind the only team at any time to be caught abusing the fuel flow regulation was red bull.

The proposed abuse of the fuel flow rate does not tally with Mercedes's consistently lowest fuel use nor with their strong race pace.

If they are constantly flowing 100kg/hr past the meter, even when the engine cannot use it then effectively the fuel system is operating under a full throttle condition for more of the lap than their competitors, therefore their fuel use should be higher.

[Arco Cigomático]

It seems clear that some teams are getting inject more than 100 Kg / h of fuel into the cylinders at certain times without being detected by the sensor beside the fuel tank.

Is it clear? What evidence?

All we know is that the FIA have developed more thorough checks. That doesn't mean someone has been caught cheating, it could just be the natural evolution of the rules.

No one has been caught cheating because the FIA can not prove that it was done.

According to my theory, the sensor would have never encountered a more fuel flow permitted, but if more sensors installed elsewhere is because the FIA is clear that they are bypassing the rule or may do so in the future.

By such acts the FIA, I think not act in prevention but rather knowing that it is already happening.

(Sorry for my english, again)

[gruntguru]

I think there is a misunderstanding here.
There is no possible "cheating" wrt the fuel mass flow over a certain prolonged time, let's say over one lap (indeed, it would be even a shorter time...) with the idea Gruntguru proposed.
But, it is well possible to use the elasticity of the fuel transport system and/or the "elasticity of fuel in fluid vs. gaseous phase.
And, it has nothing to do with the rail pressure, i.e. the pressure the injectors are seeing, which is between 250 and max. 500 bar as far as info has been released.
All this must happen before the High Pressure Pump (HPP) which produces thede 250-500 bar and after a feeding pump. This pump could feed constantly 100kg/h, and be pressure limited to xx bar.
- In case of higher demand >100kg/h of the injectors/HPP, it would feed in the volume limited mode. The pressure in the fuel transport system falls, and elasticity/bubbles are produced. Mass flow sensor reads 100kg/h.
- When a certain low limit of the feeding pressure is reached, the engine has to reduce mass flow to 100kg/h.
- As soon as there is less than 100 kg/h used (braking, cornering), the feeding pump, still delivering 100 kg/h can repressure the system, until its pressure limit is reached, compressing the fuel bubble to liquid and/or expanding the fuel lines or whatever.
This can be made with very few electronic controls, one of the most important things being the fuel pump feeding constant volume independently from the pressure, but limiting this flow when a certain pressure level is reached.

Sure I get the proposed layout of the system, but creating such a low pressure on the suction side of the pump and getting the fuel to evaporate at a constant rate would be immensely difficult in my eyes. First you'd require some form of pressure/flow reducer valve to keep a constant flow before the HPP (from the low pressure supply pump and through the sensor), then you'd need to heat the fuel, then you'd need to make sure the gas bubble doesn't reach the HPP. Also it would be very difficult to control where the cavitation bubble would appear. This seems like such whichcraft to me that I highly doubt it. Less alone hiding all this stuff from the prying eyes of the FIA.

1. With heated fuel, say 70*C like the Honda RA168E, the vapor pressure is not that low, possibly even higher than atmospheric. Easy to modify by adding small amounts of a light fraction BTW.
2. The evaporation rate would be equal to the drawdown rate which is controlled by the injectors/engine management.
3. The vapor bubble would appear in the hottest part of the fuel. If the fuel heater was a slender vertical canister with flow from top to bottom, the HPP (or the pre-HPP booster pump?) would only draw liquid fuel.

All components required would have a legitimate function and arouse no suspicion whatsoever. Mapping would be the only change required to start "cheating".

[itix]

1. With heated fuel, say 70*C like the Honda RA168E, the vapor pressure is not that low, possibly even higher than atmospheric. Easy to modify by adding small amounts of a light fraction BTW.
2. The evaporation rate would be equal to the drawdown rate which is controlled by the injectors/engine management.
3. The vapor bubble would appear in the hottest part of the fuel. If the fuel heater was a slender vertical canister with flow from top to bottom, the HPP (or the pre-HPP booster pump?) would only draw liquid fuel.

All components required would have a legitimate function and arouse no suspicion whatsoever. Mapping would be the only change required to start "cheating".

I must have been sooo sleepy when I wrote above message. So many language mistakes and tautologies -_-

Anyway I don't believe in that model. I still don't believe you could make that work in a stable and controllable way. Until I see it working you are not going to convince me and I am not going to convince you, so let's agree to disagree (and I'll be so ashamed if in the end it turned out that you were right all along, hahahaha).

[Abarth]

It seems clear that some teams are getting inject more than 100 Kg / h of fuel into the cylinders at certain times without being detected by the sensor beside the fuel tank.

Is it clear? What evidence?

All we know is that the FIA have developed more thorough checks. That doesn't mean someone has been caught cheating, it could just be the natural evolution of the rules.

Well, that would be a veeery special case.....FIA requesting more control sensors without some suspicion...

[Sevach]

All we know is that the FIA have developed more thorough checks. That doesn't mean someone has been caught cheating, it could just be the natural evolution of the rules.

Weird timing honestly.

If they did it in planned way before the season i would buy that, as it is i believe someone pushed for it.

[PhilS13]

1. With heated fuel, say 70*C like the Honda RA168E, the vapor pressure is not that low, possibly even higher than atmospheric. Easy to modify by adding small amounts of a light fraction BTW.
2. The evaporation rate would be equal to the drawdown rate which is controlled by the injectors/engine management.
3. The vapor bubble would appear in the hottest part of the fuel. If the fuel heater was a slender vertical canister with flow from top to bottom, the HPP (or the pre-HPP booster pump?) would only draw liquid fuel.

All components required would have a legitimate function and arouse no suspicion whatsoever. Mapping would be the only change required to start "cheating".

I must have been sooo sleepy when I wrote above message. So many language mistakes and tautologies -_-

Anyway I don't believe in that model. I still don't believe you could make that work in a stable and controllable way. Until I see it working you are not going to convince me and I am not going to convince you, so let's agree to disagree (and I'll be so ashamed if in the end it turned out that you were right all along, hahahaha).

Making it work every corner cycling a vapor bubble up and down in volume and managing all that seems complicated.

Could they be filling up the chamber before the race and slowly grow the vapor bubble all race long only with good fuel temperature management? If you can "store" 1 kg of fuel on the formation lap and that kg is gone by the end of the race it has the same impact as accumulating a tiny amount whenever off throttle and spending it right away.

[slugmeister]

This my first post, so apologies for jumping straight in, but this topic touched on my F1 and fuel injection interests and experience.

I was at the Melbourne GP and a few things were of note. Working backwards to my point, the Renaults had horrible driveability and there was talk from drivers etc of engine mapping issues. Renault apparently made some late changes to their engine mapping. You could actually hear the problems at the track, on corner exit for example.

To my ears too, and some others, it did sound like the Merc and Ferrari had more revs at some parts of the track this year, but that's not been technically confirmed.

Im going to hypothesise that Merc and probably Ferrrari are indeed (somehow) using more fuel flow than is legal for certain parts of the rev range. The fuel flow through the meter remains legal at all times, and ECU log analysis looks legitimate too. However, through some trick at certain times more fuel can be injected than is legal based on some accumulation downstream of the meter. However, the trick is injection mapping related, or to utilise it the injection mapping must use legitimate sensor input (that is logged) but sort of ignore that and inject on what is really happening ( higher fuel pressure or more air mass or something).

Where is this coming from? Well I think Renault know how to get the increased fuel pressure, or that at certain times there will be higher fuel pressure, but they cant get an injection map that works reliably to use it. In short, Renault aren't programmatically able to exploit the trick yet, probably because injection mapping relies on using sensor input and religiously believing it - that is, its difficult programmatically. I think at the Melbourne GP they had some injection maps that were "experimental" (and probably they know aren't legitimate) and decided to go with them because they realised how far behind they were without this trick. However, these maps weren't quite working and Ricciardo et al struggled with "spiky power delivery" (sic). I even think that is possibly what lunched Ricciardos engine.

Post race Horner starts whining that Renault haven't delivered on their promises (to "cheat" as well?) and that theyre needs to be equalisation. Marko starts threatening to leave F1. Then, someone must have let the cat out of the bag, probably RBR or Renault to the FIA about some fuel flow trick, and bang a new directive comes out - pseudo equalisation? Any thoughts?

[maccafan]

Very interesting post. And a possible theory on why FIA imposed these new flowmeters