281 replies to this topic

[Wuzak]

Over on the RC forum Whiteblue made this statement:

This is completely made up. The cutting edge in ICE design is where the efficiency technologies are applied. We are talking power to weight ratio of engines, direct fuel injection with spray guided combustion, throttle less engine control, ultra high injection pressures, fast injection times of 500 ns, super high exhaust turbine temperatures, bore/stroke ratios with minimized heat loss, stratified injection charge, variable valve lift, electric turbo compounding, CVT turbo compounding, valves that work with extremely high air fuel ratios and laser ignition. The list can go on and on. McLaren, Porsche, Mercedes and Ferrari are packing those technologies into their road cars and F1 sits there and twiddles its thumbs.

I am wondering what will be allowed, and what the manufacturers will choose to do.

With lower power of the ICE compared to today it is quite likely that the cars will run full throttle for a larger percentage of the lap. Considering hat at some tracks the cars already hold full throttle for as much as 70% of the time, will there be any need for variable valve timing and lift, and particularly to use it for throttle-less engine control?

Is it likely that injection pressures will be limited?

What is a bore/stroke ratio which minimises heat loss? The maximum allowed bore is to be 88mm, which would give a stroke of about 65.8mm, and a bore:stroke ratio of 1.34. Surely the manufacturers will still look to maximise piston area and valve area?

To contain costs some of the materials used in the engines will be specified. Will that dictate how hot various parts of the engine will run?

From the articles I have read it is unlikley that turbocompounding will be allowed for 2013, but may appear in 2014. How would it be achieved, and will it be worthwhile downstream of the turbo?

Apart from some Diesel truck engines, is anybody developing a turbo-compound for road use?

Also, some technologies will be proprietry, and thus may not be allowed unless it is freely available to all manufacturers. Could this be a problem?

With efficiency being the goal, will some teams/engine manufacturers try for a transverse installation?
Lastly, with the engine freeze system, and the likelihood of the FIA imposing limitations on engines used per season, how much development will occur?

[Tony Matthews]

McLaren, Porsche, Mercedes and Ferrari are packing those technologies into their road cars and F1 sits there and twiddles its thumbs.

I wonder how long that's been going on for.

[MatsNorway]

With lower power of the ICE compared to today it is quite likely that the cars will run full throttle for a larger percentage of the lap. Considering hat at some tracks the cars already hold full throttle for as much as 70% of the time, will there be any need for variable valve timing and lift, and particularly to use it for throttle-less engine control?

Is it likely that injection pressures will be limited?

Yes but it might be bumped up a fair bit. 100bars now.

With efficiency being the goal, will some teams/engine manufacturers try for a transverse installation?

Efficiency is not a primary goal for the engine builders. Power is. so if a lower efficiency konfig gives more power than a high efficiency konfig they will allways go for max power.

Now a transverse layout would be super cool to see. you get a lot more space longitudally. and easier packaging. Problem is aero because of the with and aero is king. They could lay it down transversely?
would not be at all supriced if some teams ditches the KERS and go all in aero to win at fast and twisty tracks. Like turkey.

Lastly, with the engine freeze system, and the likelihood of the FIA imposing limitations on engines used per season, how much development will occur?

Alot! i bet no one wants to get out teched. This is a chance for the manufacturers to jump the opposition and show their technological superiority. Its a big commercial opportunity.

Edited by MatsNorway, 09 December 2010 - 08:13.

[cheapracer]

Efficiency is not a primary goal for the engine builders.

Actually it is - you can either carry less fuel (weight) or go faster for longer.

I never want to see the 80's running out of fuel on the last lap crap ever again though.

[MatsNorway]

Actually it is - you can either carry less fuel (weight) or go faster for longer.

I never want to see the 80's running out of fuel on the last lap crap ever again though.

I guess its a mix. Something you calculate based upon predicted efficiency and the weight saved and what that does to cornering speed. and blabla

But with a fuel flow limit and that low initial power figures i think power becomes more important again.

[cheapracer]

Example - one year for the Bathurst 1000 race Dick Johnson's Falcon was 30hp down, 2 seconds per lap slower optimised for fuel consumption as they went for a 'one less pitstop' than the others strategy - as with many races in the modern era the safety car ruined it for them but they were on target.

[WhiteBlue]

With lower power of the ICE compared to today it is quite likely that the cars will run full throttle for a larger percentage of the lap. Considering hat at some tracks the cars already hold full throttle for as much as 70% of the time, will there be any need for variable valve timing and lift, and particularly to use it for throttle-less engine control?

I estimate average power at 75% of peak power today. In 2013 that could change to 79%. It will not affect the question of throttle less power modulation via variable valve stroke and timing. The variable valve technology aims at reducing internal aerodynamic losses (pumping losses) which are created by throttling the air flow. Instead of a throttle plate variable valve mapping in combination with variable injection and ignition mapping will modulate the power and reduce the parasitic losses. It simply means that the air compressor will have slightly less load and the turbine power can be used for other purposes.

Is it likely that injection pressures will be limited?

No, the current injection pressure of standard DI systems with solenoid actuators is 150 bar. Piezo systems use 200 bar. This is in contrast to diesel pumps going up to 2000 bar and higher. The problem with petrol is the lack of lubrication by the medium. The history of the development (mainly by Bosch) shows that they will probably increase the petrol injection pressure in the future. The point is that rpm are not likely to increase a great deal beyond 10,000 rpm thus much faster injection times would not bring much better stratification. The ceiling is probably somewhere around 300 bar but you would need different actuator designs for that as well.

What is a bore/stroke ratio which minimises heat loss? The maximum allowed bore is to be 88mm, which would give a stroke of about 65.8mm, and a bore:stroke ratio of 1.34. Surely the manufacturers will still look to maximise piston area and valve area?

Bore stroke ratio is fixed for the new turbo engines. The point is it will be considerably better for avoiding heat losses compared to the extremely short stroke of the current F1 engines.

To contain costs some of the materials used in the engines will be specified. Will that dictate how hot various parts of the engine will run?

The critical parts for the new engines are valves and exhaust gas turbines. Valves get burned when the stratification produces excessive AFRs with too much residual oxygene in the exhaust gas. It would be counter productive if material development for valves would be choked by narrow regulations. I think that the engine manufacturers will look at advanced surface treatment methods as they are used by in the aircraft engine industry to protect compressor blades. Turbine temps are dictated by the back pressure and expansion ratio of the turbine and the thermal insolation towards the bearing system. A lot of progress is being made by the industry leaders like Garrett. They will continue to push temps beyond 1000°C.

From the articles I have read it is unlikley that turbocompounding will be allowed for 2013, but may appear in 2014. How would it be achieved, and will it be worthwhile downstream of the turbo? Apart from some Diesel truck engines, is anybody developing a turbo-compound for road use? Also, some technologies will be proprietry, and thus may not be allowed unless it is freely available to all manufacturers. Could this be a problem?

Torotrak CVT proposal for turbo compounding This is one of the proposed systems.



The other one is essentially a conventional turbo charger with an electric servo motor that boosts the compressor at low rpm and generates electric power at high rpms. You see that both methods are using an over sized turbine compared to the compressor to extract more energy from the exhaust gas. Both solutions will work without a waste gate and will require elaborate electronic management.

The question of intellectual property does not apply to F1. In order to violate a protected right you have to exploit it commercially. Building racing engines and race cars for F1 isn't considered a commercial operation. The FiA rules have excluded the applicability of intellectual property rights for a long time.

With efficiency being the goal, will some teams/engine manufacturers try for a transverse installation?
Lastly, with the engine freeze system, and the likelihood of the FIA imposing limitations on engines used per season, how much development will occur?

A transversal engine position is pretty unlikely due to aerodynamic packaging aspects. The more likely configuration will be a 45° lay flat like this Ferrari solution.



The concept of long life engines will definitely be maintained because it is fixed in the RRA which is a legally binding document between the teams and the FiA. I think that the target number of engines will probably be reduced from eight to five. That number has been mentioned in the past as a reasonable target. Five year homologation is planned for the core engine parameters like the L4 configuration, the displacement, CoG, weight targets and the bore diameter. Basically the bottom of the engine will be pretty much fixed for five years. The top is going to be homologated for one year intervals according to an interview given by Tim Routsis of Cosworth in October 2010.

I wonder how long that's been going on for.

Pretty much since the end of the nineties when Porsche, Audi, Saab and Mercedes all made significant development efforts with their fuel injection and turbo charger suppliers.

Yes but it might be bumped up a fair bit. 100bars now.

NO, they do between 150 and 200 bar now.

Efficiency is not a primary goal for the engine builders. Power is. so if a lower efficiency konfig gives more power than a high efficiency konfig they will allways go for max power.

No, thermal power is limited by the fuel flow restriction to 1279 kW absolute. The only way to get more mechanical power is to increase efficiency. Today efficiency of the V8 is 29% which would translate to 371 kW. Four % more efficiency would translate into 422 kW or 14% more power.

[MatsNorway]

NO, they do between 150 and 200 bar now.

5.8.1 The pressure of the fuel supplied to the injectors may not exceed 100 bar. Sensors must be fitted which
directly measure the pressure of the fuel supplied to the injectors, these signals must be supplied to the
FIA data logger.

[cheapracer]

I wonder how long that's been going on for.

I think flight 473 to New York just flew over.

[WhiteBlue]

5.8.1 The pressure of the fuel supplied to the injectors may not exceed 100 bar. Sensors must be fitted which
directly measure the pressure of the fuel supplied to the injectors, these signals must be supplied to theFIA data logger.

You got the question wrong. Wuzak asked what would be allowed for the new direct injected engines. Since the technology is supported by positive comments from Ferrari and Mercedes one has to assume that state of the art pumps and injectors will be used as for the road cars of those companies or even more advanced.

The current regulations apply to port injection which will not be applicable to the 2013 engines. Thus your quotation of the current regulations is misleading readers about the expected spec of the 2013 direct injection. A limit on injection pressure for direct injection makes no sense. None of the engine manufacturers develop proprietary systems in that field and there are only very few cutting edge suppliers like Bosch, Siemens and Delphi. The manufacturers would not block them to develop special racing systems. It would potentially negatively impact on efficiency and would not help with cost. If one supplier goes for pressures beyond 200 bar he will never agree to make that a proprietary system for one manufacturer. It did not happen that way in the past and is not likely to happen in the future.

Edited by WhiteBlue, 10 December 2010 - 02:51.

[cheapracer]

5.8.1 The pressure of the fuel supplied to the injectors may not exceed 100 bar. Sensors must be fitted which
directly measure the pressure of the fuel supplied to the injectors, these signals must be supplied to the
FIA data logger.

I doubt that will change much.

Port injection will be used still (not direct injection) because with the total sum of a racing car, including the most important aero packaging, port injection offers denser (cooler) intake charge which means smaller intercoolers (more aero efficiency) than you could use with direct injection.

Although I mentioned that economy is a priority which it is, it's clearly ranked behind aero and power.

[WhiteBlue]

I doubt that will change much. Port injection will be used still (not direct injection) because with the total sum of a racing car, including the most important aero packaging, port injection offers denser (cooler) intake charge which means smaller intercoolers (more aero efficiency) than you could use with direct injection. Although I mentioned that economy is a priority which it is, it's clearly ranked behind aero and power.

I call bullshit on that one. The 2013 engines will definitely have direct injection. This has been agreed since May 2010 between all parties. Your information on the thermodynamics btw is also incorrect. The best DI systems inject the fuel late in the compression stroke versus the intake stroke being used by port injection. Piezo driven outward opening spray guided injectors atomize the fuel by a factor of 100 finer than old port injectors. That way the vaporization occurs at maximum speed and the fuel air mixture is generated in fractions of a millisecond before ignition occurs. The process generates more power with leaner AFRs than the best port injection can generate.

[cheapracer]

I call bullshit on that one. The 2013 engines will definitely have direct injection. This has been agreed since May 2010 between all parties.

Sorry, not sure what your calling bullshit? If the DI gets mandated (which it hasn't yet) then it merely makes my post moot not bullshit unless you can verify why it's bullshit ........

Piezo driven outward opening spray guided injectors atomize the fuel by a factor of 100 finer than old port injectors. That way the vaporization occurs at maximum speed and the fuel air mixture is generated in fractions of a millisecond before ignition occurs. The process generates more power with leaner AFRs than the best port injection can generate.

Ahhh thought you were an Amway salesman with all your spiels but apparently you work for Piezo.

Anyway, thanks for the DI explaination, saved people simply 'Googling' it, but I would be more interested if you explain to us why DI will improve an F1 car and I mean overall as in the whole F1 car package - lets imagine that DI doesn't get mandated and why it's a better choice over port injection for an F1 car?

[MatsNorway]

Is it likely that injection pressures will be limited?

Yes but it might be bumped up a fair bit. 100bars now.

NO, they do between 150 and 200 bar now.

5.8.1 The pressure of the fuel supplied to the injectors may not exceed 100 bar. Sensors must be fitted which
directly measure the pressure of the fuel supplied to the injectors, these signals must be supplied to the
FIA data logger.

You got the question wrong.

Off course they are going to limit it. Its FIA they are going to do so and say its for cost savings.

And about DI being so effective. remember the injectors take away space from the valves. power is first priority. So if they don`t gain on DI vs bigger valves they will not do it. Unless its speced ofc.

Hydraulic oil compresses 3% i believe at 500bar. so if fuel is less dense they will be able to compress it more and the more they are able to compress it the more fuel you get into the engine/out of injector. in addition you will release the compression energy and add cooling.

Edited by MatsNorway, 10 December 2010 - 16:13.

[kikiturbo2]

Anyway, thanks for the DI explaination, saved people simply 'Googling' it, but I would be more interested if you explain to us why DI will improve an F1 car and I mean overall as in the whole F1 car package - lets imagine that DI doesn't get mandated and why it's a better choice over port injection for an F1 car?

I'd think that there is not enough time to get a good air/fuel mixture at WOT and max RPM with DI... Usually high RPM engines have their injectors far away from the head because of this reason..

[J. Edlund]

If you look at the conditions unique to F1 and other racing series, you can conclude that some of the mentioned technologies aren't really that interresting.

F1 engines operate at mostly full throttle or zero throttle during braking, which means technology aimed at reducing part load pumping losses such as using variable valve lift and duration instead of a throttle plate will be unnecessary. The same probably goes for spray guided fuel injection, which also is a technology aimed at increasing part load efficiency by charge stratification.

I suspect they will use the same kind of direct injection used by other racing series; that is homogeneous charge direct injection with the fuel injectors placed in the cylinder head below the intake ports. That way, the injectors will not obstruct the valve or port sizes. But direct injection will be used, let me be clear about that. Direct fuel injection offer a slightly higher volumetric efficiency as the inlet air is free of fuel vapor (yes, the effect of fuel vapor is bigger than the effect of cooling, so getting rid of the vapor is an advantage), it provides a better in cylinder cooling effect which allows the use of a higher compression ratio, and it prevent fuel from escaping with the exhaust. I also don't think the fuel injection pressures will be that extreme; fuel is injected during the intake stroke after the exhaust valve have closed. Ilmor ran direct injection in F1 engines up to about 17k rpm or something like that in the past without using extreme injection pressures, although, the did only have a power benefit up to about 15k rpm.

Production engines, they either operate with homogeneous charge direct injection, or they switch between homogeneous charge and stratified charge where the latter is used at part load. In case they use NOx traps, they are also forced to run rich to regenerate the trap on occation.

Normally, a high efficiency means that the engine will have a low exhaust temperature. However, since spark ignition engines operate rich at high engine loads, saving fuel means using a less rich mixture, and that result in high exhaust temperatures. In the eighties the exhaust temperatures could exceed 1100 degC. As turbines made of Inconel 713 only would survive in those conditions for about half an hour, Garrett made the turbines of IDM 5322 and the turbine housings of HK30, a cast stainless steel, and employed bleed air cooling. Today the directionally solified alloy Mar-M 247 would be a possebility (commonly used in for instance WRC applications), or if the demands are higher, one of the many single crystal alloys used in modern gas turbines. Si3N4 would also be a possebility, it was used already in the eighties. But this will all depend upon what materials are allowed. I know Garrett have made prototypes for turbine housings in composite material too - I suspect some sort of carbon/carbon.

For exhaust valves the trick is more to keep the exhaust valve temperature down rather than increase the temperature resistance of the valve material. A hot valve is not only problematic from a mechanical standpoint, but it can also cause surface ignition or knocking. A hollow Nimonic 80A design would probably be good enough, but if the choice was free, I think many would go the titanium aluminide route.

Turbocompounding will probably be able to boost the output of the engines with a few percent. The reason why they only are found in commercial diesel engines these days is because those are the only engines which operate at a load that is high enough for the turbocompound to function. Passenger car engines are used at so low average loads, say 2 bar BMEP vs. 9 bar BMEP for a truck engine, that turbocompound isn't really useful. As for the drive, I would suspect that a fixed mechanical drive will be the choice. A mechanical drive will be lightweight, efficient and need little space and cooling. Yes, when the jet-to-blade velocity go outside the optimun it will cost some efficiency, but so will a CVT, and the usable engine speed range of a F1 engine is quite narrow in either case. The turbocompound turbine is typically downstream of the turbocharger turbine.

Can't see anyone using a transverse engine installation. As for using an angled engine installation, I'm not so sure that is the way to go.

[Tony Matthews]

If you look at the conditions unique to F1 and other racing series, you can conclude that some of the mentioned technologies aren't really that interresting.
.
.
.Can't see anyone using a transverse engine installation. As for using an angled engine installation, I'm not so sure that is the way to go.

Very interesting JE, thanks.

[Magoo]

If you look at the conditions unique to F1 and other racing series, you can conclude that some of the mentioned technologies aren't really that interresting.

F1 engines operate at mostly full throttle or zero throttle during braking, which means technology aimed at reducing part load pumping losses such as using variable valve lift and duration instead of a throttle plate will be unnecessary. The same probably goes for spray guided fuel injection, which also is a technology aimed at increasing part load efficiency by charge stratification.

I suspect they will use the same kind of direct injection used by other racing series; that is homogeneous charge direct injection with the fuel injectors placed in the cylinder head below the intake ports. That way, the injectors will not obstruct the valve or port sizes. But direct injection will be used, let me be clear about that. Direct fuel injection offer a slightly higher volumetric efficiency as the inlet air is free of fuel vapor (yes, the effect of fuel vapor is bigger than the effect of cooling, so getting rid of the vapor is an advantage), it provides a better in cylinder cooling effect which allows the use of a higher compression ratio, and it prevent fuel from escaping with the exhaust. I also don't think the fuel injection pressures will be that extreme; fuel is injected during the intake stroke after the exhaust valve have closed. Ilmor ran direct injection in F1 engines up to about 17k rpm or something like that in the past without using extreme injection pressures, although, the did only have a power benefit up to about 15k rpm.

Production engines, they either operate with homogeneous charge direct injection, or they switch between homogeneous charge and stratified charge where the latter is used at part load. In case they use NOx traps, they are also forced to run rich to regenerate the trap on occation.

Normally, a high efficiency means that the engine will have a low exhaust temperature. However, since spark ignition engines operate rich at high engine loads, saving fuel means using a less rich mixture, and that result in high exhaust temperatures. In the eighties the exhaust temperatures could exceed 1100 degC. As turbines made of Inconel 713 only would survive in those conditions for about half an hour, Garrett made the turbines of IDM 5322 and the turbine housings of HK30, a cast stainless steel, and employed bleed air cooling. Today the directionally solified alloy Mar-M 247 would be a possebility (commonly used in for instance WRC applications), or if the demands are higher, one of the many single crystal alloys used in modern gas turbines. Si3N4 would also be a possebility, it was used already in the eighties. But this will all depend upon what materials are allowed. I know Garrett have made prototypes for turbine housings in composite material too - I suspect some sort of carbon/carbon.

For exhaust valves the trick is more to keep the exhaust valve temperature down rather than increase the temperature resistance of the valve material. A hot valve is not only problematic from a mechanical standpoint, but it can also cause surface ignition or knocking. A hollow Nimonic 80A design would probably be good enough, but if the choice was free, I think many would go the titanium aluminide route.

Turbocompounding will probably be able to boost the output of the engines with a few percent. The reason why they only are found in commercial diesel engines these days is because those are the only engines which operate at a load that is high enough for the turbocompound to function. Passenger car engines are used at so low average loads, say 2 bar BMEP vs. 9 bar BMEP for a truck engine, that turbocompound isn't really useful. As for the drive, I would suspect that a fixed mechanical drive will be the choice. A mechanical drive will be lightweight, efficient and need little space and cooling. Yes, when the jet-to-blade velocity go outside the optimun it will cost some efficiency, but so will a CVT, and the usable engine speed range of a F1 engine is quite narrow in either case. The turbocompound turbine is typically downstream of the turbocharger turbine.

Can't see anyone using a transverse engine installation. As for using an angled engine installation, I'm not so sure that is the way to go.

Excellent synopsis. Top notch.

[Wuzak]

Can't see anyone using a transverse engine installation. As for using an angled engine installation, I'm not so sure that is the way to go.

I guess this is all about aerodynamics?

With the transverse engine the chassis would proably be significantly shorter and therefore less attractive for an installation. Also the width may become a problem for any venturi tunnels that may be allowed.

Similarly for the slant 4, it could get in the way of the underbody aerodynamics. Also the FIA may again mandate a CoG height which would require the engine to be stood up more. And an upright 4 will allow more space for the KERS and other stuff.

[Wuzak]

If you look at the conditions unique to F1 and other racing series, you can conclude that some of the mentioned technologies aren't really that interresting.

F1 engines operate at mostly full throttle or zero throttle during braking, which means technology aimed at reducing part load pumping losses such as using variable valve lift and duration instead of a throttle plate will be unnecessary. The same probably goes for spray guided fuel injection, which also is a technology aimed at increasing part load efficiency by charge stratification.

Would that mean it would not be relevant to road car engine development?

[WhiteBlue]

F1 engines operate at mostly full throttle or zero throttle during braking, which means technology aimed at reducing part load pumping losses such as using variable valve lift and duration instead of a throttle plate will be unnecessary. The same probably goes for spray guided fuel injection, which also is a technology aimed at increasing part load efficiency by charge stratification.

I suspect they will use the same kind of direct injection used by other racing series; that is homogeneous charge direct injection with the fuel injectors placed in the cylinder head below the intake ports. That way, the injectors will not obstruct the valve or port sizes. But direct injection will be used, let me be clear about that. Direct fuel injection offer a slightly higher volumetric efficiency as the inlet air is free of fuel vapor (yes, the effect of fuel vapor is bigger than the effect of cooling, so getting rid of the vapor is an advantage), it provides a better in cylinder cooling effect which allows the use of a higher compression ratio, and it prevent fuel from escaping with the exhaust. I also don't think the fuel injection pressures will be that extreme; fuel is injected during the intake stroke after the exhaust valve have closed. Ilmor ran direct injection in F1 engines up to about 17k rpm or something like that in the past without using extreme injection pressures, although, the did only have a power benefit up to about 15k rpm.

I think you are probably starting with some incorrect assumptions. The new engines are supposed to have elaborate energy management according to today's release by the FiA. Last week we got a leak from the working group by Craig Scarborough who said that the mass fuel flow limiter will be fixed at 100 kg/h (27.8 g/s). At 46 MJ/kg specific energy of petrol it means the engines are limited to 1279 kW thermal power. This would mean 422 kW (574 hp) mechanical power at an assumed efficiency of 33%. If we add 120 kW KERS power we arrive at 740 hp. This is pretty much what the FiA is targeting for 2013. According to educated estimates the current engines have only 29% efficiency. The difference of 4% has to come from the improved efficiencies of the direct injected turbo engine.

The significance of the fuel flow limit is a paradigm change for the F1 engine designer. He cannot afford to run anything on those engines below optimum efficiency. Pumping losses, thermal losses, frictional losses or exhaust gas enthalpy losses must all be taken absolutely serious. The 1.6L engines will have 88 mm bore and that will help improve thermal efficiency. Contrary to JE I'm convinced that throttle losses - even if they are small - will be as much eliminated as possible.

The new engines will have 12,000 rpm and 500 bar injection pressure. This will only make sense if spray guided injection in the final compression stroke is used. No direct injection in the intake stroke needs more than 150 bar. It is known that even at homogeneous charge spray guided injection reduces fuel consumption by several percent.

[Wuzak]

The significance of the fuel flow limit is a paradigm change for the F1 engine designer. He cannot afford to run anything on those engines below optimum efficiency. Pumping losses, thermal losses, frictional losses or exhaust gas enthalpy losses must all be taken absolutely serious. The 1.6L engines will have 88 mm bore and that will help improve thermal efficiency. Contrary to JE I'm convinced that throttle losses - even if they are small - will be as much eliminated as possible.

But they'll only ever be near the fuel flow limit around max power, where the throttle will be fully open with minimal losses. When not at max power they will easily be under the fuel flow limit...

[cheapracer]

http://www.fia.com/e...msc-101210.aspx

The saddest day in my mind for F1 looks like it's going to happen, no mention of DI though here unless it's somewhere else.

[WhiteBlue]

But they'll only ever be near the fuel flow limit around max power, where the throttle will be fully open with minimal losses. When not at max power they will easily be under the fuel flow limit...

That is not the point. In the past your design compromises were mainly driven by getting more power with unlimited fuel. Now your energy budget is fixed. More power will only come from elimination of losses. Throttle losses from imperfect gas exchange do matter even if they are small at full power setting. A throttle plate even full open is an unnecessary restriction. Engine power management is done more effectively by mapping of valve functions, injection and ignition than by throttling the air flow. The other point is that drivability will be greatly enhanced in modern turbo engines. Expect the torque curve to go up to the max by 2000 rpm and stay there all the way to 12,000 rpm. It means that drivers will be able to put variable power down from the apex of the corner much better than they are are able now. Or just think about turn 8 in istanbul where the car needs to be balanced on the edge with the throttle very precisely for several seconds. You also have to consider safety cars where it is essential to profit from good part throttle efficiency to have more power available in the final stage of the race. All good reasons to install variable valves and use stratified injection.

[WhiteBlue]

..no mention of DI though here unless it's somewhere else.

It is implicit by the pressure of he injection.

[Wuzak]

That is not the point. In the past your design compromises were mainly driven by getting more power with unlimited fuel. Now your energy budget is fixed. More power will only come from elimination of losses. Throttle losses from imperfect gas exchange do matter even if they are small at full power setting. A throttle plate even full open is an unnecessary restriction. Engine power management is done more effectively by mapping of valve functions, injection and ignition than by throttling the air flow. The other point is that drivability will be greatly enhanced in modern turbo engines. Expect the torque curve to go up to the max by 2000 rpm and stay there all the way to 12,000 rpm. It means that drivers will be able to put variable power down from the apex of the corner much better than they are are able now. Or just think about turn 8 in istanbul where the car needs to be balanced on the edge with the throttle very precisely for several seconds. You also have to consider safety cars where it is essential to profit from good part throttle efficiency to have more power available in the final stage of the race. All good reasons to install variable valves and use stratified injection.

They will not run much at 2000rpm. With a 7 speed gearbox they will be up near maximum power most of the time, and max torque will be high up in the rev range too.

Efficiency really won't matter that much behind a safety car, as they will be using much less fuel than they have budgeted anyway.

[cheapracer]

It is implicit by the pressure of he injection.

"with high pressure gasoline injection up to 500 bar"

No, all that implies is you're allowed to use up to 500 bar - this opens the door for DI use but doesn't discount port injection until it actually states it.

The FIA have cocked up too many rules in the past to be able to state anything clearly yet and have a history of letting alternatives through loopholes either intentional or not, of course the most recent being the 2009 double diffuser loophole.

[Foyle]

Over on the RC forum Whiteblue made this statement:




I am wondering what will be allowed, and what the manufacturers will choose to do.

With lower power of the ICE compared to today it is quite likely that the cars will run full throttle for a larger percentage of the lap. Considering hat at some tracks the cars already hold full throttle for as much as 70% of the time, will there be any need for variable valve timing and lift, and particularly to use it for throttle-less engine control?

Is it likely that injection pressures will be limited?

What is a bore/stroke ratio which minimises heat loss? The maximum allowed bore is to be 88mm, which would give a stroke of about 65.8mm, and a bore:stroke ratio of 1.34. Surely the manufacturers will still look to maximise piston area and valve area?

To contain costs some of the materials used in the engines will be specified. Will that dictate how hot various parts of the engine will run?

From the articles I have read it is unlikley that turbocompounding will be allowed for 2013, but may appear in 2014. How would it be achieved, and will it be worthwhile downstream of the turbo?

Apart from some Diesel truck engines, is anybody developing a turbo-compound for road use?

Also, some technologies will be proprietry, and thus may not be allowed unless it is freely available to all manufacturers. Could this be a problem?

With efficiency being the goal, will some teams/engine manufacturers try for a transverse installation?
Lastly, with the engine freeze system, and the likelihood of the FIA imposing limitations on engines used per season, how much development will occur?

We now have a max speed of 12000rpm, 500 bar injection pressure. It has been hinted that there will be a 100kg/hr max fuel flow and Ø88 bore.

So what will the engine be like?

With restricted fuel flow it is all about engine efficiency to get max power. This leads to a couple of key things:

1/ Absolutely want to reduce the piston velocity and port flow velocity as much as possible to increase efficiency, and that means maximum piston diameter allowed (just like current engines), Ø88mm seems unnecessarily restrictive to me - should be more like Ø95-100mm

2/ Most turbo engines are heavily overfuelled at max power in order to prevent the turbo from melting/ destroying itself, while max power occurs at about 13:1 A/F, Most high power turbos run at more like 10:1 at max power using fuel to keep the turbines cool (and is a tremendously wasteful approach). This is just not feasible given limited fuel flow so one of the keys will be getting the turbine inlet temps up. TiAl and MarM247 Turbines can handle 1050°C turbine inlet temp and this helps, but it will probably also be necessary to do one of the following:
a/ run a pre-turbine exhaust cooler of some type (water cooled exhaust manifold etc).
b/ run a ceramic turbine (lighter, and can handle higher temps - but can be durability issues)
c/ bypass a little compressor air straight to the turbine (not always possible, but can be done over a reasonable range of engine rpm given no muffler and efficient compressor and turbine - also helps prevent surge at low rpm)
d/ run a Miller cycle engine

From these options the Miller cycle (with late closing inlet valve to reduce the air in the cylinder) is an almost certainty. Miller cycle gives a relatively longer expansion stroke to suck a little more heat (and temp) out of the gas before it goes to the turbine and means that you get a higher efficiency engine (220-230 g/kWh vs 240-250 g/kWh for non-miller cycle that in turn means 5-10% more engine power for a fuel flow limited engine). The engine runs a higher geometric compression ratio (a couple of points above normal turbo, so for this engine might be more like 12:1 or 13:1 compression vs the 10:1 that a DI turbo with 1 bar boost would have. (Incidentally with 100kg/hr fuel flow and 220-250g/kWh we know that the power output will be about 400-450kW = 530-600hp).

I haven't seen any talk of limiting the boost pressure - and shouldn't need to given fuel flow restriction. That would point towards high boost levels and running at much lower speeds to reduce the engine frictional losses and improve efficiency at max power. 3 bar boost is definitely a possibility with max power from 6000rpm dropping off slowly through to 12000rpm. Would probably spend most time at lower end of this speed range - and perhaps voluntary self limited max rpm of maybe 8-9000rpm. There would be far less need to change gear than currently as the engines can deliver basically max power over such a broad rpm range.

This desire to run at lower speeds will run counter to F1 marketing, so I expect that this may push the rules towards either a boost level limit (say 2 bar) or a low rpm fuel flow limit to force engine designers to set max power point at say 9000-10000rpm and above.

Other issues:

Be interesting to see how the compromises on intercooling work out - bigger intercooler with greater aero drag losses and lag but higher engine efficiency, or smaller with efficiency hit?

Compressor wheels should be carbon fiber for reduced inertia and smaller running clearances (giving higher efficiency), though advantage over aluminium isn't huge.

Sodium cooled valves for sure.

VVT would be preferred from a development and performance point of view, but lets see what the rules allow.

It would be disappointing if they did not run brushless electric motors on the turbos for anti-lag (this is a technology that has useful crossover to cars and trucks) - and once that is in place it is very simple to also use it for turbocompounding for extra efficiency.

McClaren's engine supplier for the MP4-12C Riccardo is in the box seat with their experience on that 3.8l turbo, (max power at 7000-8500rpm, though would need a much bigger turbo)
Ton of North American experience from IRL, CART etc.
Cosworth are next door to one of the premier supercar turbo gasoline engine developers - Mahle Powertrain.

[WhiteBlue]

Max torque will be high up in the rev range too. Efficiency really won't matter that much behind a safety car, as they will be using much less fuel than they have budgeted anyway.

Have you ever seen the torque curve of a modern turbo? It is flat as a billiard table! You have to stop the thinking that efficiency wouldn't matter at any time, because it will always be a factor. Teams will always run leaner than possible to lessen the fuel load and anything saved behind the safety car will be a welcome gift to up the power later in the race by inching closer to the max fuel flow than budgeted.

"with high pressure gasoline injection up to 500 bar" No, all that implies is you're allowed to use up to 500 bar - this opens the door for DI use but doesn't discount port injection until it actually states it.

You need to develop more of an engineering perspective to carry that point. Port injection is completely undesirable in an efficiency formula. It has the worst of the three known combustion processes. Next better is wall and air guided combustion and best is spray guided. The efficiency improves with every step you take in process quality but the necessary pressure also increases dramatically towards the more efficient combustion. The high pressure injection is not only a challenge for injector design, you also run into problems with the fuel pumps. Petrol is a very bad lubricant, much worse than diesel oil. At present there is no known solution to 500 bar injection on the market. The best you get is 200 bar from Bosch.

1/ Absolutely want to reduce the piston velocity and port flow velocity as much as possible to increase efficiency, and that means maximum piston diameter allowed (just like current engines), Ø88mm seems unnecessarily restrictive to me - should be more like Ø95-100mm.

Nope, this isn't a one dimensional optimization. Extremely short stroke drops your thermal efficiency by the need of excessive cooling to the cylinder head. It over compensates the influence of the gas exchange particularly with forced induction.

2/ Most turbo engines are heavily overfuelled at max power in order to prevent the turbo from melting/ destroying itself, while max power occurs at about 13:1 A/F, Most high power turbos run at more like 10:1 at max power using fuel to keep the turbines cool (and is a tremendously wasteful approach). This is just not feasible given limited fuel flow so one of the keys will be getting the turbine inlet temps up. TiAl and MarM247 Turbines can handle 1050°C turbine inlet temp and this helps, but it will probably also be necessary to do one of the following:
a/ run a pre-turbine exhaust cooler of some type (water cooled exhaust manifold etc).
b/ run a ceramic turbine (lighter, and can handle higher temps - but can be durability issues)
c/ bypass a little compressor air straight to the turbine (not always possible, but can be done over a reasonable range of engine rpm given no muffler and efficient compressor and turbine - also helps prevent surge at low rpm)
d/ run a Miller cycle engine

You do not want to cool your exhaust gases because the purpose of the turbine is to expand the gas and cool it that way while it sucks the energy out and makes power. It means you absolutely want to maximize the tolerable turbine inlet temps. The correct way to control that is to engineer the turbine pressure differential appropriately to match the possible back pressure of the engine and the temp tolerances.

It would be disappointing if they did not run brushless electric motors on the turbos for anti-lag (this is a technology that has useful crossover to cars and trucks) - and once that is in place it is very simple to also use it for turbocompounding for extra efficiency.

+1
Yep, an electric servo MGU on the turbo shaft is one of the prime options to do turbo compounding and electronically regulate boost pressure. The other one is a mechanical CVT like the Torotrak. as posted in post #7 of this thread. These options are sure to come in 2014 with the turbo liberalization.

Edited by WhiteBlue, 11 December 2010 - 18:30.

[MatsNorway]

why can`t them just allow superchargers as well right of the bat then.

[WhiteBlue]

why can`t them just allow superchargers as well right of the bat then.

I don't think they will prohibit such a thing. The problem with superchargers is they don't contribute to more power efficiency but actually absorb power.

[Canuck]

Teams will always run leaner than possible...

Err...no they won't. Logic fail.

[MatsNorway]

I don't think they will prohibit such a thing. The problem with superchargers is they don't contribute to more power efficiency but actually absorb power.

running a supercharger with a turbo to help once the pressure goes up would been my first choice.

In effect its a turbo configuration once it gets up to speed.

Edited by MatsNorway, 11 December 2010 - 21:44.

[J. Edlund]

I guess this is all about aerodynamics?

With the transverse engine the chassis would proably be significantly shorter and therefore less attractive for an installation. Also the width may become a problem for any venturi tunnels that may be allowed.

Similarly for the slant 4, it could get in the way of the underbody aerodynamics. Also the FIA may again mandate a CoG height which would require the engine to be stood up more. And an upright 4 will allow more space for the KERS and other stuff.

The big questionmark here is how the car regulations, and the detailed engine regulations will look like for 2013.

Having the engine standing up will increase the CoG height, but could on the other hand allow a more narrow rear bodywork.

Would that mean it would not be relevant to road car engine development?

Since a racing engine is built for a high average load, without regard to emissions, in a small series to a high unit cost, I would think that it always will be difficult to develope racing engines that are relevant for production cars more than the image and some small stuff here and there.

I think you are probably starting with some incorrect assumptions. The new engines are supposed to have elaborate energy management according to today's release by the FiA. Last week we got a leak from the working group by Craig Scarborough who said that the mass fuel flow limiter will be fixed at 100 kg/h (27.8 g/s). At 46 MJ/kg specific energy of petrol it means the engines are limited to 1279 kW thermal power. This would mean 422 kW (574 hp) mechanical power at an assumed efficiency of 33%. If we add 120 kW KERS power we arrive at 740 hp. This is pretty much what the FiA is targeting for 2013. According to educated estimates the current engines have only 29% efficiency. The difference of 4% has to come from the improved efficiencies of the direct injected turbo engine.

The significance of the fuel flow limit is a paradigm change for the F1 engine designer. He cannot afford to run anything on those engines below optimum efficiency. Pumping losses, thermal losses, frictional losses or exhaust gas enthalpy losses must all be taken absolutely serious. The 1.6L engines will have 88 mm bore and that will help improve thermal efficiency. Contrary to JE I'm convinced that throttle losses - even if they are small - will be as much eliminated as possible.

The new engines will have 12,000 rpm and 500 bar injection pressure. This will only make sense if spray guided injection in the final compression stroke is used. No direct injection in the intake stroke needs more than 150 bar. It is known that even at homogeneous charge spray guided injection reduces fuel consumption by several percent.

I don't think you actually understand what pumping losses due to throttling is. Pumping losses due to throttling occur only during part load when the partially open throttle decrease the pressure inside the air inlet in order to reduce air mass flow to the engine (as volumetric flow remains fairly constant). When the throttle is fully opened the pressure efter the throttle is equal to the pressure before the throttle and because of that, there isn't any loss due to pumping caused by the throttling. As F1 engines mostly operate at full throttle, throttling losses is a non issue.

As for spray guided direct injection, it does not make sense to propose a high output racing engine running to 12,000 rpm using such a direct injection strategy at full throttle. At part load it could work, but I don't think it is worth it. Ok, at full load it could reduce the combustion and exhaust temperatures by allowing air excess, but injecting fuel for 150 hp per cylinder at 12,000 rpm during a very short window isn't easy and it's going to have a negative impact on mixture preparation. The purpose of spray guided direct injection is to create a zone with a stoichiometric air fuel mixture around the spark plug; commonly refered to as a stratified charge. This allows a engine to operate with air excess during part load conditions which means reduced pumping losses and increased part load efficiency - an important regime for production engine. For a racing engine, it's quite the opposite, it's how the engine performs at full load that is important. To perform at full load the fuel should be injected during the intake stroke to allow for good mixing and vaporisation.

The 500 bar injection pressure is just a cap similar to the current limit at 100 bar. The higher the injection pressure, the smaller the Sauter mean diameter will be and hence there will be a benefit for mixture preparation which can increase engine output and reduce fuel consumption. We saw that development with port injection systems, starting at more conventional injection pressures of 3-4 bar, then increasing to 10, 30 and finally figures around 100 bar until the cap limited development.

46 MJ/kg sounds very high for gasoline. The net energy content for gasoline, which is the only energy a spark ignition engine can make use of, tend be closer to 43 MJ/kg so 100 kg/h would mean 4300 MJ/h or 1200 kWh. So if we want 440 kW on the crankshaft, the engine efficiency have to be 37% - assuming the 100 kg/h figure is correct and something similar to ordinary gasoline is used. Back in 1988 the turbocharged Honda V6 engine was able to reach a specific fuel consumption of 272 g/kWh at 12,000 rpm, using a fuel containing roughly 41 MJ/kg (back then it was the volumetric energy content that mattered). That translates into a fuel efficiency of slightly above 32%. The engine ran quite lean for a gasoline engine at lambda 0.98 and produced 620 hp under those conditions. The charge air temperture was increased by the means of a charge air cooler bypass to improve vaporisation and the fuel was preheated using a coolant/fuel heat exchanger. So already a 1988 F1 engine would be able to reach about 520 hp under the 100 kg/h limit. Illien have said that direct injection reduced fuel consumption by 5%, so that would bring the power up to about 550 hp.

Compared to todays F1 engine these new engines will have a significant advantage with the smaller capacity, fewer cylinders and the much lower engine speed. In the past increasing engine speed by 10% gave an increase of friction losses by about 20%, so reducing the engine speed from 18,000 rpm to 12,000 should be a significant benefit. The engines will also be able to take advantage of modern designs and materials to minimize losses such as the sealed comparment crankcase and DLC coated piston skirts. Technologies such as roller main bearings could possebly also return. But since the engines have to be ready in about two years time, I think the design have to be rather conservative.

[Tony Matthews]

Technologies such as roller main bearings could possebly also return. .

I didn't realise that roller-bearings had any advantage over shell bearings, as they seem not to be used anymore (as far as I know!) Is this because there is less friction? I assumed there was more.

[Wuzak]

I don't think they will prohibit such a thing. The problem with superchargers is they don't contribute to more power efficiency but actually absorb power.

If the engine used a supercharger with variable speed drive and was turbocompounded it would achieve a similar result to a turbo engine, surely?

[WhiteBlue]

running a supercharger with a turbo to help once the pressure goes up would been my first choice. In effect its a turbo configuration once it gets up to speed.

there are more efficient ways to achieve that with a hybrid drive train

I don't think you actually understand what pumping losses due to throttling is. Pumping losses due to throttling occur only during part load when the partially open throttle decrease the pressure inside the air inlet in order to reduce air mass flow to the engine (as volumetric flow remains fairly constant). When the throttle is fully opened the pressure efter the throttle is equal to the pressure before the throttle and because of that, there isn't any loss due to pumping caused by the throttling. As F1 engines mostly operate at full throttle, throttling losses is a non issue.

I understand very well but I simply don't agree with your opinion that part load regime is negligible. It is very handy to be able to save fuel in extended safety car periods or on a parade lap.

As for spray guided direct injection, it does not make sense to propose a high output racing engine running to 12,000 rpm using such a direct injection strategy at full throttle. At part load it could work, but I don't think it is worth it. Ok, at full load it could reduce the combustion and exhaust temperatures by allowing air excess, but injecting fuel for 150 hp per cylinder at 12,000 rpm during a very short window isn't easy and it's going to have a negative impact on mixture preparation. The purpose of spray guided direct injection is to create a zone with a stoichiometric air fuel mixture around the spark plug; commonly refered to as a stratified charge. This allows a engine to operate with air excess during part load conditions which means reduced pumping losses and increased part load efficiency - an important regime for production engine. For a racing engine, it's quite the opposite, it's how the engine performs at full load that is important. To perform at full load the fuel should be injected during the intake stroke to allow for good mixing and vaporisation. The 500 bar injection pressure is just a cap similar to the current limit at 100 bar. The higher the injection pressure, the smaller the Sauter mean diameter will be and hence there will be a benefit for mixture preparation which can increase engine output and reduce fuel consumption. We saw that development with port injection systems, starting at more conventional injection pressures of 3-4 bar, then increasing to 10, 30 and finally figures around 100 bar until the cap limited development.

I guess we will have to disagree on that point as well. Spray guided injection in the compression stroke with proper fuel delivery rates, correct atomization and the right timing delivers better mixing and vaporization that air guided injection. Not only is it possible at elevated rail pressures to meet the timing requirements it will also save fuel in full load and part load regime because the vaporization will focus the cooling effect exactly to the time window that is needed for the fuel to vaporize. You can run higher compression and later ignition that way. As already pointed out running engines with superior fuel efficiency in part load will become increasingly more important. Safety car phases are increasing in length and frequency as F1 is making a massive appearance in Asian monsoon regions. With the refueling ban in place fuel saved will allow the teams to run richer mixtures towards the end of the race when others have already switched to fuel saving.

Back in 1988 the turbocharged Honda V6 engine was able to reach a specific fuel consumption of 272 g/kWh at 12,000 rpm, using a fuel containing roughly 41 MJ/kg (back then it was the volumetric energy content that mattered). That translates into a fuel efficiency of slightly above 32%. The engine ran quite lean for a gasoline engine at lambda 0.98 and produced 620 hp under those conditions.

I remember that I have looked into the efficiency of the Honda engine some time ago but I remember 27% efficiency from in the relevant regime. They claimed much higher efficiency for single working points that were actually irrelevant for a total F1 race regime. Part of the problem at that time was the turbo lag which isn't a problem with modern methods any more. I agree that the new turbos should be much better engines.

I didn't realise that roller-bearings had any advantage over shell bearings, as they seem not to be used anymore (as far as I know!) Is this because there is less friction? I assumed there was more.

Hydrodynamic bearings are safer to use because they do not have a statistic failure mode as roller and ball bearings. They also get safer as faster they go. But at high speeds the large area of the sheared fluid film produces higher friction than roller bearings. The contact area of roller bearings is much smaller.

[J. Edlund]

I didn't realise that roller-bearings had any advantage over shell bearings, as they seem not to be used anymore (as far as I know!) Is this because there is less friction? I assumed there was more.

I know Cosworth had a development program for roller bearings before the engines were frozen, bearing supplier SKF were appearently conviced there was an advantage to be had. Although I suspect the engine speeds used in F1 have made it difficult to realise such advantages. Tests with production engines have showed some promise, altough there have also been cases where the expected benefits have failed to deliver.

[WhiteBlue]

If the engine used a supercharger with variable speed drive and was turbocompounded it would achieve a similar result to a turbo engine, surely?

I don't think so. In total you would have more mechanical conversions with the associated losses. A combined turbo charger/compounder with a servo MGU or the Torotrak design would be more efficient IMO. A realtively large part of the turbine power would be directly absorbed by the compressor without having to go through any conversion of a reduction gear or CVT.

[cheapracer]

I didn't realise that roller-bearings had any advantage over shell bearings, as they seem not to be used anymore (as far as I know!) Is this because there is less friction? I assumed there was more.

There's more friction Tony but you may save on parasitic power losses through requiring less of an oil pumping system and less components overall. In other words with slippers you need the pressure and volume to keep components separated but with rollers obviously you don''t so much.

On the other hand current oil scavenge systems give rise to high vacuums in the crankcase that benefits windage and pumping losses and allows thinner and lower tension oil rings to be used saving some friction there so it's quite a complicated question if rollers are better or not.

It's a question that will be asked as 12,000rpm is within the scope of roller cranks - of course it's not the load problems it's the stopping and starting of the big ends that causes skidding thats the problem and we all know that skidding is bad, especially undies.

Of course if this answer isn't suitable it's because you have asked the wrong question.

You would think the net would have a picture of the most famous F1 roller crankshaft from the '55 Benz W196 which famously also had DI but ......

Edited by cheapracer, 12 December 2010 - 03:08.

[PJGD]

[quote name='WhiteBlue' date='Dec 11 2010, 13:26' post='4756168']
"The high pressure injection is not only a challenge for injector design, you also run into problems with the fuel pumps. Petrol is a very bad lubricant, much worse than diesel oil. At present there is no known solution to 500 bar injection on the market. The best you get is 200 bar from Bosch."


There have of course been many ~500 bar DI petrol injection systems in the past, and development of such systems should be even less of a problem in the future. The WW2-era aero engines that had petrol injection employed a jerk-type pump albeit with an eccentric rather than a high rate cam profile, and outward opening poppet injectors. Certainly in the case of the CAV pump developed for the Ricardo-inspired Rolls-Royce Crecy V12 2-stroke engine, injection was direct into the unthrottled cylinder, although the system ran at ~350 bar peak pressure. The highly loaded components are the cam and roller, and they of course run in an engine oil lubricated cambox. It is the plunger reciprocating in its bore that is exposed to the low lubricity of the petrol and is where the seizure risk lies. Two factors make this challenge easier today than in the past: 1) The plunger today will almost certainly be DLC coated which greatly minimizes the seizure propensity; 2) In the old jerk-pump days, fuel metering was achieved by the helix on the plunger, and exposure of the injection pressure to that helix imposed a significant side load on the plunger which encouraged plunger seizure. Modern pumps have plain plungers which eliminate that issue.

A second issue faced by the old pumps was that the plunger-to-bore clearance was as small as then-current manufacturing techniques permitted. This was good enough for use with the moderately high viscosity of diesel fuel, but marginal for use with petrol which would leak past the plunger and dilute the oil in the cambox. The solution as shown in the cross-section below of the Crecy pump was to machine a groove in the plunger roughly in the middle of its length, which on the pumping stroke would pass across a port connected to the pressure lubrication system, and this oil "dam" would lubricate and seal the plunger at that location (the oil pressure being higher than the leakage at that point down the sealing length).



Essentially the same approach was taken for the in-line diesel injection pumps used for the multi-fuel military engines of the 50's, 60's and 70's, and these really were 500 bar pumps, and they survived without the benefit of DLC and with the side load handicap.

PJGD

[J. Edlund]

I understand very well but I simply don't agree with your opinion that part load regime is negligible. It is very handy to be able to save fuel in extended safety car periods or on a parade lap.

From your last post I think it was quite clear that you didn't fully understand what the throttling losses were.

The time spent during part load is also negligble compared to the time spent during full load so sacrificing fuel consumption at full load in order to save fuel at the little time spent during low loads isn't a good strategy. It's better to optimise the engine for full load running and then accept that the engine will be a little thirstier during low loads than would be possible if the engine was optimised for low loads.

In the case of extended safety car periods you will save fuel compared to racing speed anyway.

I guess we will have to disagree on that point as well. Spray guided injection in the compression stroke with proper fuel delivery rates, correct atomization and the right timing delivers better mixing and vaporization that air guided injection. Not only is it possible at elevated rail pressures to meet the timing requirements it will also save fuel in full load and part load regime because the vaporization will focus the cooling effect exactly to the time window that is needed for the fuel to vaporize. You can run higher compression and later ignition that way. As already pointed out running engines with superior fuel efficiency in part load will become increasingly more important. Safety car phases are increasing in length and frequency as F1 is making a massive appearance in Asian monsoon regions. With the refueling ban in place fuel saved will allow the teams to run richer mixtures towards the end of the race when others have already switched to fuel saving.

Also here it becomes quite clear that you don't really understand the basics of gasoline direct injection. This causes you to draw some strange conclusions.

To clear up the basics for you. All gasoline direct injected engines operate under homogeneous charge mode at high engine speeds and loads during which the fuel injection occur during the intake stroke. However, during low loads and speeds they can operated with a stratified charge instead of a homogeneous charge. A stratified charge can basically be achieved in three different ways: wall guided injection, air guided injection and spray guided injection. In all these cases the injection occur during the compression stroke with the goal of producing a rich combustable zone around the spark plug. Additionally, a gasoline direct injected engine can operate with a lean homogeneous mode. All modes except the homogeneous mode are optional intended to save fuel during part load. Let me be clear about this; at full load all gasoline direct injection engines operate in homogeneous mode by injecting fuel during the intake stroke.

Again, as a F1 engine spends most of the time at high engine speeds and loads they will get the greatest benefit by optimizing the efficiency at those conditions. That would involve picking an injector, injector placement, combustion chamber design and so on that will have the greatest positive effect on the homogeneous charge operating mode. You don't want to compromise those choices just so you can save a few percent fuel on load/speed conditions that you hardly ever use and indeed, direct injected racing engines are generally designed to running only in homogeneous charge mode.

I remember that I have looked into the efficiency of the Honda engine some time ago but I remember 27% efficiency from in the relevant regime. They claimed much higher efficiency for single working points that were actually irrelevant for a total F1 race regime. Part of the problem at that time was the turbo lag which isn't a problem with modern methods any more. I agree that the new turbos should be much better engines.

The Honda engine was basically below 300 g/kWh during the whole full load speed range reaching a minimum value of 272 g/kWh at 12,000 rpm. Using a fuel that contains 11.4 kWh/kg that translates into an efficiency of 29-32% during full load which is very good for a spark ignition engine. But for the fuel flow limit it is the fact that the engine produced 450 kW with a fuel flow of 122 kg/h that is interresting (450*0.272). With a higher energy content gasoline that would be about 116 kg.

As for reducing turbo lag, the Honda engine had about the same technologies to its advantage that a modern racing engine would have. The Honda engine had a turbocharger which used low friction ceramic ball bearings and a low inertia ceramic turbine. Modern racing turbochargers rarely use something more exotic than tool steel ball bearings and a Mar-M 247 superalloy turbine, and using antilag systems would be difficult due to the fuel limitations.

[PJGD]

Rolling element bearings have been used often enough in both regular and racing engines past. Tatra for one [due to Hans Ledwinka] were a strong proponent in both SI and CI engines; V12 CI example below. The Timken example above looks like a conventional crank adapted for roller bearings, whereas the most robust roller crank is the disk-web design because the crank is so much shorter. Of course, at high crank speed, the rollers are going at a really crazy speed.



It can't be long before someone mentions the highly successful supercharged 1.5 litre straight eight Delage of 1926, since that had roller bearing everything; cams, crank, you name it.

PJGD

[cheapracer]

What about the highly successful supercharged 1.5 litre straight eight Delage of 1926, since that had roller bearing everything; cams, crank, you name it.

why can`t them just allow superchargers as well right of the bat then.

Superchargers are RPM specific whereas turbo's are load specific.

Loose explanation...

At 5000 rpm for example a S/C may give 10psi boost and that will be regardless if the car is cruising on the flat or going uphill or downhill.

A T/C however may give 20psi at 5,000rpm going uphill because of the amount of throttle/exhaust gases feeding it and may give zero boost going down the hill off throttle.

Since F1 is nearly all about full throttle and big loads at speed, a T/C is the choice over any advantages that S/C has to offer such as throttle response/control

Of course, at high crank speed, the rollers are going at a really crazy speed.

Theres thousands of 20,000rpm mains roller bearing engines around but they use plain big ends, again it's not the rpm/speed (turbines for example) it's the acceleration and deceleration that gives problems.

Edited by cheapracer, 12 December 2010 - 06:54.

[cheapracer]

A stratified charge can basically be achieved in three different ways: wall guided injection, air guided injection and spray guided injection. In all these cases the injection occur during the compression stroke with the goal of producing a rich combustable zone around the spark plug.

A combo of both?

Leaned off port injection with DI injecting near the spark plug to ignite the otherwise too lean a charge - ala Honda's CVCC system of the 70's.

In Honda's case the used a small valve, combustion chamber and miniscule carby throat that created a rich, ignitable charge that fed into the main chamber's un-ignitable lean charge. Having repaired a few this picture is not indicative of what they actually look like inside but gives the basic theme....

[Tony Matthews]

There's more friction Tony but you may save on parasitic power losses through requiring less of an oil pumping system and less components overall. In other words with slippers you need the pressure and volume to keep components separated but with rollers obviously you don''t so much.

I asked Mario Ilien about the lubrication system on one of his engines - for the life of me I can't be sure which one, but probably Ilmor's first V10 - and everything was done to reduce the power needed to float the crankshaft on oil, the power absorbed by the oil pump was 3 hp. Surely you would also need either a pump for other bearings such as camshafts, or they are going to need to have roller (needle roller?) bearings too.

On the other hand current oil scavenge systems give rise to high vacuums in the crankcase that benefits windage and pumping losses and allows thinner and lower tension oil rings to be used saving some friction there so it's quite a complicated question if rollers are better or not.

That's what I expected - and at this point I've realised that I may have missed something in a later post - I should have looked up the road.

It's a question that will be asked as 12,000rpm is within the scope of roller cranks - of course it's not the load problems it's the stopping and starting of the big ends that causes skidding thats the problem and we all know that skidding is bad, especially undies.

Well, if nothing else that answers the question as to what Australian males wear under their overalls.

Of course if this answer isn't suitable it's because you have asked the wrong question.

You would think the net would have a picture of the most famous F1 roller crankshaft from the '55 Benz W196 which famously also had DI but ......

I think you also get into the choice of individual rollers, like my GN engine, split cages or Hirth (that's probably wrong) crankshafts - more complication.

[Tony Matthews]

What about the highly successful supercharged 1.5 litre straight eight Delage of 1926, since that had roller bearing everything; cams, crank, you name it.

I should have got out of bed earlier!

[MatsNorway]

there are more efficient ways to achieve that with a hybrid drive train

what you mean hybrid? mechanical drive of some sort is surely better than wasting energy from a dyno and losing some over to the electric motor. that in turn also has momentum and reaction times. and is heavy. Its not a big complex task to disconnect the mechanical drive so the turbo is working solo turbo drive.

How much energy goes into a charger? that sort of puts the bar on the electric motor needed right?

As already pointed out running engines with superior fuel efficiency in part load will become increasingly more important. Safety car phases are increasing in length and frequency as F1 is making a massive appearance in Asian monsoon regions. With the refueling ban in place fuel saved will allow the teams to run richer mixtures towards the end of the race when others have already switched to fuel saving.

The fuel flow limit has been added to eliminate the possibility to get race cars stopping before the flag and eliminate eco races because thats gay and no one wants to watch the worlds finest drivers going slow.

So until they sort of figures out that all, yes saving of fuel is important.

Now to the big question.

Will there be a hemi

[cheapracer]

and eliminate eco races because thats gay and no one wants to watch the worlds finest drivers going slow.

Actually although I disagree with eco runs overall it was interesting to watch the drivers try to maximise cornering speeds, they had a tuffer job then and were by no means going slower. But it sucked as a show overall.

[Wuzak]

About turbocompounding - would anyone try a multi stage axial flow turbine or will they just use a radial inflow turbine hooked up to the exhaust from the turbo?