Ultimate FI Setup 2

[NcBlueTJ]

I know that modern formula one cars use fuel injectors mounted above the individual throttles that spray fuel directly down into the intake trumpets (I beleive they are called standoff injectors). This seems like it would be ideal from a fuel atomization/vaporization standpoint seeing as the fuel has the most time and space with which to mix with the air charge. Although direct gasoline injection is banned in F1, would they use it if they could?
I understand the ability of GDI to produce stratified A/F ratios and run lean without producing dangerous cylinder temperatures, therefore acheiving better fuel economy (like the audi R8 race cars for example). But does this ability come into play at WOT? I also understand the concept of acheiving greater VE due to the air charge entering the engine being pure air as opposed to an Air/Fuel mixture. Thus the engine is able to ingest more air for a given head flow rate, since the fuel is added later. However, during a WOT scenario given a high performance engine such as an F1 car, do these advantages outweigh the presumed loss of air/fuel mixing that a standoff injector would provide?
Sorry for the long-winded post. My basic question is this: Would F1 use GDI if they could and why? Thanks

 

[ivymike]

I would guess that would depend a bit on what technology the sponsors are trying to market...

 

[Bribyk]

I don't know if any current model-year sportbikes use this setup but I know at least Hondas used to use two injectors per cylinder. One injector is close to the intake port for the lower load and speed ranges (for drivability), and the second was positioned farther upstream (like the F1 injectors) for higher engine speeds and loads to allow the fuel more time to atomize and mix. A combination of GDI and F1-type injection might give the best of both worlds...

 

[SomptingGuy]

As you have stated:

- Lower consumption
- Higher torque & power
- Better driveability

And of course, lower emissions to pander to the greenies (although the whole F1 circus is never going to be green).

So my guess is yes, F1 would probably use GDI if they could, otherwise it wouldn't be banned. I feel this is a real shame, because it prevents the tricking down of F1 technology into road car technology.

- Steve

 

[NcBlueTJ]

Thanks for the responses. Bribyk, yes I believe you are right about the high mount injectors and a lack of low-rpm driveablility and the solution: engines with 2 sets of injectors. I believe the BMW S70 V12 (Mclaren F1) uses this setup. My main question however, relates to F1 engines that really do not have to deal with low RPM driveability issues (5000RPM+ idle speed).
SomptingGuy, if I understand you correctly, the VE advantage of not having to flow fuel through the intake/head with a GDI setup outweighs the improved fuel mixing attributes of a standoff injector setup. If so, this answers my question.
I completely agree with you that the ban on GDI is a mistake that limits technology trickle-down. IMO, the FIA is going the complete wrong direction with technology regulation. Freezing engine development and forcing manufacterers to use an energy recovery system (next year) and a common ecu (this year)is lunacy. Engine development (including ecu development) is IMO one of the sport's most valuable technological contributions to the real world.

 

[SomptingGuy]

NcBlueTJ:

I agree. KERS _might_ have technology spin-offs (ever driven a Honda Civic IMA?), but banning ECU development is stupid. Our Controls & Electronics group is probably the fastest growing group in the company. The next big thing will be control strategies based on individual in-cylinder conditions (measured or predicted). We'll be way ahead of F1 by the time the freeze thaws.

F1 used to have technology. Now it has politics, intrigue and scandals (all still fun though).


- Steve

 

[dgallup]

I'm not aware of any GDI FIE that is capable of operating at the ~18,000 RPM redline of current F1 engines. That doesn't mean it couldn't be done given the F1 cubic dollar budgets. But there isn't much time available to inject & mix at that kind of engine speed. Hence the "showerhead" or "standoff" injectors currently used.

 

[TDIMeister]

I don't think we will see GDI in F1 given anywhere near the current engine spec. Reason is that at 19000 RPM, the amount of time available to inject fuel during the intake and compression strokes combined, and mixture formation, is about 280 degrees crank angle, or about 2.5ms. Not a heck of a lot of time.

Second, the biggest motivations for using GDI -- lean-burn operation and charge stratification -- will not be exploited in a race engine. The Audi FSI system used in the now-retired LeMans R8 car got its fuel economy benefit at full-load by reducing the degree of fuel-enrichment for knock-control. It was by no means lean-burn nor stratified-charge. That engine's fuel economy benefit came from the being likely able to run much closer to a stoichiometric AFR than other PFI race engines that would have to be enrichened to lambda 0.8-0.85.

Third, port injection basically allows the full 720 degrees of a complete 4-stroke cycle for fuel injection in the intake ports and mixture preparation (much of the injection and mixture preparation actually occur in the ports while the intake valves are closed).

The less than half the available time is a big challenge given the large amount of fuel needed to be injected within the limited window of time.

First challenge: *IF* the FIA would actually allow GDI in the rules (and I hope they do, because it may lead to some further interesting innovation in GDI technology)

Second challenge: Major advancement in injection technology will be necessary, but I think it is a challenge that can be overcome (borrowing some knowledge from common-rail DI Diesel technology, having to operate at an order of magnitude higher pressures and even shorter time window for the injection event and mixture preparation).

 

[2stroketech]

Look at the aprillia ditech engine. It is a direct inject 2 stroke that is high preasure air over fuel. this is how I understand it to work:

It has a small compressor that works off of a lobed crankshaft. As the piston clears the exhaust port a blast of compressed air pushes the exhaust out the port. As the piston closes the port then the gas is injected under the stream of air. This makes the engine more efficient and cleaner than even the best 4 strokes out there. I got to see proto type engine that adopted this technology over. The 72cc engine was developing a little under 13HP and VE was through the roof. I hope to pick up one of these engines to play with a little later this summer.

 

[patprimmer]

Fuel economy might become an issue for F1 if they ever severly restrict tank size, filler neck size and pressure in the refuelling system. They kind of did that toward the end of the turbo era.

Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[GregLocock]

Don't the current cars run massively rich in order to cool the exhaust valves ?

Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[BrianPetersen]

No one (that can talk about such things) really knows. But, given limitations on fuel tank size, running "massively" rich is neither in the interest of fuel economy (think: number of laps between pit stops) nor power (too rich means less power).

I would expect lambda to be in the same range at full load as for "normal" engines. If exhaust valve durability is an issue, just substitute really expensive materials ...

 

[NcBlueTJ]

Wow, thanks for the great info.
TDIMeister, I had not considered the fact that standoff injectors utilized a full 720 degrees of stroke to inject and mix the fuel. I can see why the severe drop in injector cycle duration required for a GDI type system could make for some substantial engineering challenges at 19,000rpm.
Nevertheless, it seems as if GDI is theoretically possible for a formula one type engine given sufficient development time, money and innovation. If they can do it on an Audi R8 that turns what? 9 or 10 thousand RPM, it seems feasable to make it work at double that speed. Also, I understand that lean-burn and charge stratification are not terribly advantagous for a race motor, but couldn't an F1 car benefit from the same advantages the R8 does by "reducing the degree of fuel-enrichment for knock-control."
2stroketech, I'll definitely look into that aprillia engine; it sounds interesting. I've always wondered if a 2-stroke engine could be created using GDI combined with air to push the burned charge out the exhaust port (perhaps with a turbocharger?).
Thanks to all for the great responses; it looks like we have a pretty good discussion going here. I'd love to hear some more about this subject.

 

[2stroketech]

NcBlueTJ, The turbo charger is not needed. The system I seen was using 826 pounds of compressed air. The compressor generates all the air needed and is basically a super charged system.

 

[Lou Scannon]

TDIM, good job articulating the logical but less than obvious.
Question, you bring up "mixture preparation" relating to common-rail DI diesel technology, in a discussion of F1 fuel injection. Would you care to elucidate the similarities and differences in the mixture preparation objectives for common-rail DI diesel and PFI gasoline, in an F1 context (e.g. with reference to the physical & temporal constraints)?

 

[TDIMeister]

Goodness, the differences are quite huge in a general context, but the basic fuel injection technologies can be considered similar. In both DISI (direct-injected spark ignition, analogous to GDI) and DICI (direct-injected compression ignition), fuel is pressurized by a mechanical pump and held in a rail accumulator before solenoid- or piezo-actuated injectors spray fuel into the combustion chamber.

Detailing all the differences could take up a conference attended by a whole bunch of specialist engineers and academics from OEMs, fuel injection Tier-1 suppliers and engineering schools.

For the present purposes, it would be constructive to limit discussion to a motorsport (possible F1) context, and a highly simplified one at that. There are many subclassifications of DISI systems (stratified or homogeneous; stoichiometric or lean burn; spray-, wall- and air guided; piezo- or solenoid actuation; servo- or direct- actuation; type of injectors used; etc.) in like manner that there are quite wide degrees of freedom in the design of DICI systems and operating strategies.

In the motorsport context, we are mostly interested in homogeneous stoichiometric charge mixture preparation. The goal here is, like PFI systems, to have the most evenly-distributed, stoichiometric and fully-vapourised air-fuel mixture at the time of spark ignition. This is not an easy task because of the in-cylinder charge motion; highly complex and unsteady physical-, chemical- and thermal processes; and relatively short time scales for all these to happen.

DICI takes this to yet another level, because as already stated, the fuel injection and mixture preparation must occur at even short time scales -- the entire fuel injection event could begin and end inside of about 50° crank angle, and within this time you could have up to 7 (yes, seven in the current state-of-the-art) separate injection pulses! In spite of this, DICI systems operate at injection pressures about 10X more than DISI (current series production state-of-the-art: 2000 bar vs. 200 bar); peak cylinder pressure as much as 3 times more (200 bar vs, 70 bar); and a much more complex and active in-cylinder flow-field.

DISI as described before has the relative simplicity of just having the task of preparing the mixture as homogeneously as possible before start of combustion, but it's already difficult enough. In DICI engines, the paradigm is regarding "rate shaping", that is, using the injection event to control the shape and quantity of the injected fuel with respect to crank angle, in order to shape the rate of heat release (RoHR) curve. This is done primary for emissions and noise purposes.

However, I believe that more application of direct injection technology in motorsports, and the inevitable cross-pollination of SI- and CI technology (since for one thing EVERY gasoline fuel injection supplier like Bosch, Siemens, Delphi and Denso, are also in the Diesel fuel injection business) will lead to significant innovations in much the same way that, in the racing leagues that permit it, PFI has completely eclipsed carburetors, even when the latter was considered to be already at a very high level of development and sophistication.

 

[TDIMeister]

By the way, I have a technical paper by Audi Motorsports' engine head Ulrich Baretzky, which supports my statement above that the R8 Le Mans race car got part of its fuel economy advantage by being able to run "less-rich" than other race cars. In the development test bench, they were running at a lambda of 0.93 at rated power, and there were even lean maps with lambda > 1 for race periods with lower power requirements (e.g. pace car and yellow flags). The other big advantage lay in being able to run a compression ratio "above" 12:1. This is a turbocharged engine, folks!!

 

[murpia]

F1 engines won't run any richer than peak torque AFR's unless they're in an 'engine save' mode used in free practice. Fuel wasted is mass wasted and pitstop time wasted. Both slow the cars too much.

I understood the engine limit to be pistons not exhaust valves. Retarded ignition (less peak pressure / higher exhaust temps) was seen as an engine saver. This was before the 19,000rpm limit though.

Regards, Ian

 

[TDIMeister]

Yes, and best power AFRs for NA PFI engines are in the area of lambda 0.85 plus-or-minus. Forced-induction engines typically require even richer mixtures to prevent detonation.

 

[mattsooty]

"Forced-induction engines typically require even richer mixtures to prevent detonation."

Enrichment is not used to prevent detonation, full load enrichment is used to reduce temeratures for component protection and produce max torque. Infact detonation is most severe just rich of stoich.

MS