Red Bull exceed fuel flow limit, Ricciardo DSQ at Australian GP

[Pup]

The ECU only tells the engine components what to do. It doesn't measure, other than to accept feedback from other components, like, for example, the fuel flow meter.

[djos]

The ECU only tells the engine components what to do. It doesn't measure, other than to accept feedback from other components, like, for example, the fuel flow meter.

Measurements ARE calculations based on known data. The ECU knows fuel pressures and injector open and close times as it controls the fuel system in minute detail, it is therefore able to calculate the fuel flow rates better than a low res ultrasonic sensor!

[beelsebob]

The ECU only tells the engine components what to do. It doesn't measure, other than to accept feedback from other components, like, for example, the fuel flow meter.

Measurements ARE calculations based on known data. The ECU knows fuel pressures and injector open and close times as it controls the fuel system in minute detail, it is therefore able to calculate the fuel flow rates better than a low res ultrasonic sensor!

And what makes you think that these "known" values are more accurate than the values that the FFM is measuring? The ECU attempts to let x ml of fuel into the combustion chamber by opening the valve for y ms at a pressure of z bar. The values x, y and z in theory are nice and accurate, in practice though, these will all have minute variance on them, partly because of physical reality, and partly because of electronic timing issues. There is absolutely no reason why you should trust the computation carried out here to be more accurate than the measurement taken by the FFM.

[djos]

As others have pointed out, ultrasonic measurement of fluids under pressure is subject to all sorts of issues from pressure waves, fuel density variations and air bubble content.

Ultrasonics simply are not able to be as accurate as measuring the physical goings on in the engine and fuel system under the extreme conditions of an F1 car IMO.

[beelsebob]

As others have pointed out, ultrasonic measurement of fluids under pressure is subject to all sorts of issues from pressure waves, fuel density variations and air bubble content.

Ultrasonics simply are not able to be as accurate as measuring the physical goings on in the engine and fuel system under the extreme conditions of an F1 car.

Ultrasonics certainly will have an error rate. The error rate is well specified and controlled. What makes you think that the un-specified and un-controlled error rates of the electro-mechanical system that is the ECU and the injectors are lower?

[ChrisM40]

Exactly, RB are assuming that the injectors inject the exact same volume every time when pressure is X and opening time is Y, in reality thats just not the case.

Unless every sensor in the Renault engine is calibrated (with certificated proof), then they cannot rely on the ECU for anything. The FIAs system relies on the accuracy of one sensor, and therefore has only 1 error to contend with, RBs relies on any number of sensors, calculations and therefore a multitude of errors, however small, compounding each other. It may well be accurate most of the time, but is it accurate enough to the degree that they can guarantee the engine NEVER goes over the 100Kg/h limit?

[djos]

Simple BB, F1 has been using electronic fuel injection since the 80's, the industry experience behind EFI system is simply mammoth and fuel metering has been brought to a very precise science.

[ChrisM40]

Simple BB, F1 has been using electronic fuel injection since the 80's, the industry experience behind EFI system is simply mammoth and fuel metering has been brought to a very precise science.

Maybe, but there are 2 things going on here. How close were RB pushing it? Were they running at exactly 100Kg/h (by their measure)? 99.9, 99.5? If so can they be sure the engine is taking every variable into account? Fuel varies in viscosity naturally, and by temperature, this will effect how much is injected at a fixed pressure and temperature. Do they take that into account? If they do is their temperature monitoring system calibrated (with documentation to back it up). No? Then they cant rely on it either practically or legally, its that simple.

[djos]

Good points Chris, if we are lucky we might find out in April.

[motobaleno]

I'm little bit astonished: 36 pages of 3d and apparently nobody linked the flowgate with the fact that one of the major issue with renault engines is knocking: the simplest way to avoid knocking is to throw more fuel into your cilinders...

[jz11]

As others have pointed out, ultrasonic measurement of fluids under pressure is subject to all sorts of issues from pressure waves, fuel density variations and air bubble content.

Ultrasonics simply are not able to be as accurate as measuring the physical goings on in the engine and fuel system under the extreme conditions of an F1 car.

Ultrasonics certainly will have an error rate. The error rate is well specified and controlled. What makes you think that the un-specified and un-controlled error rates of the electro-mechanical system that is the ECU and the injectors are lower?

I am 1000000000% confident that they ran all sorts of sims and tests for those high pressure injectors and weighted amount of fuel that went through them at any and all imaginable duty cycles, now this info together with the actual duty cycles and fuel rail pressure during the race IS THE ACCURATE INFO that is available to the team, the info from FFM, as I remember, IS NOT!

basically the team does this homework before the race, that is how they get to that 100kg/h limit in the first place, and then FIA steps in, introduces it's sort-of-working solution to monitor fuel flow and says - based on our data you run your engine too high, tune it down by this or that amount! that is it! now you have 1 car that got screwed up by following this directive, would you screw the other up as well?

and ECU doesn't give a rats ass about some ultrasonic flow meter, it doesn't care nor should it (because that FFM is there for the FIA, not a source of info for the ECU), it assumes the pressure is constant and works the injectors according to preset maps, that is it!

[ChrisM40]

The problem for RB is that their system is blind and relies on assumptions. It assumes the injectors are all identical and behave with absolutely perfect consistency, it assumes the fuel is consistent, it assumes any sensors they do use are accurate, it assumes the fuel pressure is constant and measured in the same time frame as the injectors are operating in. RBs system doesnt really measure flow at all.

[beelsebob]

As others have pointed out, ultrasonic measurement of fluids under pressure is subject to all sorts of issues from pressure waves, fuel density variations and air bubble content.

Ultrasonics simply are not able to be as accurate as measuring the physical goings on in the engine and fuel system under the extreme conditions of an F1 car.

Ultrasonics certainly will have an error rate. The error rate is well specified and controlled. What makes you think that the un-specified and un-controlled error rates of the electro-mechanical system that is the ECU and the injectors are lower?

I am 1000000000% confident that they ran all sorts of sims and tests for those high pressure injectors and weighted amount of fuel that went through them at any and all imaginable duty cycles

Are you? Why would you be confident of that? You're talking about a team that did only a very few real world laps before the season started. Why would they waste those laps measuring how much fuel was going into the engine, rather than trying to get it running reliably, or trying to get some aero data down.

now this info together with the actual duty cycles and fuel rail pressure during the race IS THE ACCURATE INFO that is available to the team, the info from FFM, as I remember, IS NOT!

Again, no it's not accurate info. What happens when a tiny bit of debris starts to block one of the injectors? What happens when a pressure wave increases the amount going into the engine. What happens when one of the solenoids is slightly out of calibration and opens the valve slightly further/for longer? What on earth makes you think that this is in any way accurate data?

basically the team does this homework before the race

Except again, we're talking about a team that infamously could not do their homework.

[motobaleno]

besides, all this questioning about RB being sure or not to comply with the flow limit is completely pointless.
let's assume that RB will positively prove this ( a posteriori!)on 14 april. If FIA accepts this, then every team will have the holy right to screw up the flow meter and measure the flow by its own in the following gp..fantastic...I guess that laptimes and most of all overtakings will improve dramatically

[ChrisM40]

As others have pointed out, ultrasonic measurement of fluids under pressure is subject to all sorts of issues from pressure waves, fuel density variations and air bubble content.

Ultrasonics simply are not able to be as accurate as measuring the physical goings on in the engine and fuel system under the extreme conditions of an F1 car.

Ultrasonics certainly will have an error rate. The error rate is well specified and controlled. What makes you think that the un-specified and un-controlled error rates of the electro-mechanical system that is the ECU and the injectors are lower?

I am 1000000000% confident that they ran all sorts of sims and tests for those high pressure injectors and weighted amount of fuel that went through them at any and all imaginable duty cycles, now this info together with the actual duty cycles and fuel rail pressure during the race IS THE ACCURATE INFO that is available to the team, the info from FFM, as I remember, IS NOT!

basically the team does this homework before the race, that is how they get to that 100kg/h limit in the first place, and then FIA steps in, introduces it's sort-of-working solution to monitor fuel flow and says - based on our data you run your engine too high, tune it down by this or that amount! that is it! now you have 1 car that got screwed up by following this directive, would you screw the other up as well?

and ECU doesn't give a rats ass about some ultrasonic flow meter, it doesn't care nor should it (because that FFM is there for the FIA, not a source of info for the ECU), it assumes the pressure is constant and works the injectors according to preset maps, that is it!

Are you saying that every single time the injector opens it injects the exactly correct amount of fuel? To within 0.25%? Sorry im not buying it. The error is usually a lot more than that, but does tend to average out over time. Car engines can operator seemly fine with even significantly more error than that, especially under load.

Lets also remember that RB dont make the engine, Renault do, a company that couldn't even get the Alternator right for number of years.