2014-2020 Formula One 1.6l V6 turbo engine formula

[gruntguru]

Some Renault information I have seen indicates they use 3.5 bar absolute boost. Back of the envelope calculation on airflow at 10,500 rpm, 3.5 bar boost, 100% VE, intercooled to ambient, gives 2293 kg/hr. With 100kg/hr fuel flow you get AFR = 22.93 Very lean and even leaner at higher revs!

On conventianal engines, best economy (thermal efficiency) has always been about 10% lean of stoichiometric (16.2:1) and I will bet combustion is occuring somewhere in that vicinity. With DI however, the combustion begins in a richer zone, continues and ends in a leaner zone and is surrounded by air - which obviously doesn't burn at all. This enables the engines to be run at significantly leaner overall mixtures although - like a diesel - a significant fraction of the intake plays no part in the combustion at all.

20% higher massflow will increase compressor power by approx 20%. It also increases potential turbine power by 20% so there is no disadvantage. In fact because only a portion of the turbine power is needed to run the compressor, the power available for harvesting by the MGUH will also increase!

[Tommy Cookers]

the engine is charged about half naturally and half by the compressor
so to add another 20% (of these 2 halves combined) we need to raise the overpressure by about 40%
needing 40% more supercharger work and further heating the charge and demanding more charge cooling
and the compressor efficiency may be lower at the higher pressure, so exacerbating this heat and work
(only with an engine synthetically charged entirely by compressor) would a 20% massflow increase only demand 20% more work)

we have seen the Renault 'boost' figure of 3.5 bar abs, afaik some posters say they don't take it seriously and none say they do
and actual boost used is anyway strongly related to mean exhaust pressure

what determines whether DI is overall lean or overall stoichiometric (or overall rich even) is the engine designer
or do people believe in the 5 DI WC-winning years (in 1954 - 64) M-B, Vanwall, and Ferrari used overall lean mixture ? !!

how can 20% higher massflow increase potential turbine power by 20% ?
the compressor is driven by power taken from the turbine (directly or indirectly)
'lean-air' work taken from the turbine will be returning at the combined (in) efficiency of the turbine and compressor ie about 50% ?

normally, lean running improves efficiency only at partial power ie in road use (because it reduces throttling)
F1 potential benefits from stratified charge lean running at WOT should be set against the known disbenefits as above
future adjustments to the rules ie the expected fuel reduction may have an affect on this

[gruntguru]

the engine is charged about half naturally and half by the compressor
so to add another 20% (of these 2 halves combined) we need to raise the overpressure by about 40%
needing 40% more supercharger work and further heating the charge and demanding more charge cooling
and the compressor efficiency may be lower at the higher pressure, so exacerbating this heat and work
(only with an engine synthetically charged entirely by compressor) would a 20% massflow increase only demand 20% more work)

we have seen the Renault 'boost' figure of 3.5 bar abs, afaik some posters say they don't take it seriously and none say they do and actual boost used is anyway strongly related to mean exhaust pressure

what determines whether DI is overall lean or overall stoichiometric (or overall rich even) is the engine designer
or do people believe in the 5 DI WC-winning years (in 1954 - 64) M-B, Vanwall, and Ferrari used overall lean mixture ? !!

how can 20% higher massflow increase potential turbine power by 20% ?
the compressor is driven by power taken from the turbine (directly or indirectly)
'lean-air' work taken from the turbine will be returning at the combined (in) efficiency of the turbine and compressor ie about 50% ?

normally, lean running improves efficiency only at partial power ie in road use (because it reduces throttling)
F1 potential benefits from stratified charge lean running at WOT should be set against the known disbenefits as above
future adjustments to the rules ie the expected fuel reduction may have an affect on this

Agreed. I was considering only increased massflow with no boost increase. To increase mass flow by 20% at a given after-cooler outlet temp requires a 20% increase in PR. Peak compressor efficiency will coincide with peak power (unless the engine design team are idiots).

Well here is one person who believes peak boost pressure is of the order of 3.5 bar.

The WC years you speak of were all run under a displacement formula so the AFR used would have been about 10% rich (lambda = 0.9) for NA engines and 10-20% rich (lambda = 0.8-0.9) for supercharged engines. To run leaner would cost lots of hp. The current formula dictates AFR of 10% lean (lambda=1.1) or leaner. To run richer would cost lots of hp.

20% mass flow increase means 20% increase in compressor power and turbine power regardless of the efficiency of each (assuming PR constant although a PR increase will also affect both compressor and turbine equally). We also need to assume that the compressor and turbine are correctly selected for the changed application.

Your last paragraph is incorrect on all points. The primary efficiency gain from lean mixtures does not arise from reduced throttling but from better utilisation of the fuel. Regardless of load, peak efficiency occurs about 10% lean of stoich. Reasons this is not normally done at WOT are,
. Max power from a given engine coincides with rich AFR as above.
. Max CR requires rich AFR for detonation suppression.
. Max combustion temperatures coincide with stoichiometric mixtures and WOT running at stoich can be problematic.

[Tommy Cookers]

you seem to think that an engine needs to be 10% lean to fully burn its fuel
and then assume that 20% lean must be better than 10% ?
2% lean (AFR 1.02) was/is my view - this is just the sort of thing that 500 bar DI should guarantee

any leaner than this just means in one way or another making the engine 'bigger'
the task is to get the best from 100 kg of fuel
yes in your apochryphal account you take an engine and lean it by your chosen amount
this will reduce the max power (but also reduce the rpm and losses at the new (leaned) max power)
so it might be more 'efficient'
but then would need enlarging (in massflow) to give in lean running the original (non-lean) max power

the argument for lean running at high powers is reduced losses to coolant by having no combustion near cylinder and combustion chamber
this (stratified charge) approach means pumping a lot more air (by compressing it a lot more)

presumably Mercedes is the engine with the highest boost
raising the mean exhaust pressure reduces the rather huge losses between exhaust valve and manifold, enabling increased recovery
so higher boost does not prove higher massflow ie lean running

[Abarth]

[[....]Well here is one person who believes peak boost pressure is of the order of 3.5 bar.
[....]Your last paragraph is incorrect on all points. The primary efficiency gain from lean mixtures does not arise from reduced throttling but from better utilisation of the fuel. Regardless of load, peak efficiency occurs about 10% lean of stoich. [....]

You are throwing together a whole lot of semi-truth.

1. If you say that best efficiency at WOT is 10% lean, I could agree, even if I too believe it is not needed in a DI environment, as opposed to non stratified charge of IDI or carburettor. But then you'll not need to have 3.5 bar....
2. Running really lean (and that doesn't mean lambda 1.1, but way higher if you have a boost of 3.5 bar) will not increase efficiency and therefore (in this fuel limited formula) power. Tommy Cookers explained pretty clearly why.
3. Tommy Cookers is right on the main reason for running lean (in road cars): reduced throttling at part loads. By running lean with stratified charge it is not meant lambda 1.1, but way, way higher. In fact, you want to reduce power for part load operations without throttling losses. This does not apply at WOT.
Running very lean at WOT will reduce somewhat heat losses and will increase pumping losses.

[gruntguru]

I am sure you are all familiar with http://www.k20a.org/upload/HondaRA168EEngine.pdf the wonderfully informative technical paper on the Honda RA168E engine. Please refer to the text and particularly the graph on page 6 (pdf page 9). Although this engine was not capable of smooth running at leaner than stoichiometric, it is clear that efficiency is still improving rapidly as mixture is leaned beyond 1.0. (Note - equivalence ratio = 1/lambda.) With modern DI this engine could have been run significantly leaner for even greater efficiency. Of course power would have suffered since this engine was airflow restricted (capacity and boost limit). In short, best economy occurs around lambda 1.1 including WOT.

For further evidence look at the Wright Turbo Combound brochure often mentioned here. http://www.enginehistory.org/Wright/TC%20Facts.pdf Page 32 (pdf page 34) shows where the mixture is set for economy cruise at 1910 bhp. Although the precise mixture is not specified, the stoichiometric point is shown and the lean cruise mixture is clearly substantially leaner than stoichiometric. With 1.3 bar boost and 1910 bhp the throttle setting is closer to WOT than closed.

Running REALLY lean ie around 1.5 as I believe is happening in current F1 engines is not beneficial in itself. Best combustion efficiency occurs at around 1.1 and that is roughly the target mixture in the "rich" zone of the stratified charge DI combustion chamber. The benefits of the extra 40% air are:
1. Reduced gas temperatures, chamber temperatures, turbine temperatures etc
2. Increased massflow through the turbine. Think of the compound engine (or any turbo engine) as an engine within an engine. The "outer" engine is a gas turbine with a recip ICE as its combustion chamber. The cost of compressing the higher massflow is more than compensated by expanding the additional massflow in the turbine. With high efficiency turbos, the exhaust energy available for harvesting is significantly higher than the energy needed to drive the compressor. The surplus can be recovered by the MGUH and this surplus will increase with massflow up to a limit imposed by the energy (fuel) restrictions.
3. Improved detonation resistance. (Leaner mixture in the endzone and a fresh air blanket wrapped around that.)

The most efficient engines on earth run even leaner mixtures. Marine diesels are also turbocharged recip. They don't need to run as lean as they do to achieve complete combustion - it is for the same reasons as 1 and 2 above.

[ringo]

What is being discussed here about the turbo engines?
I'm lost.

[gruntguru]

We are discussing what mixture they are most likely running. Also what charge temperature exiting the Charge air cooler. My guess is a mixture of 18 - 22:1 and a charge temp of 80 - 150*C.

[Tommy Cookers]

What is being discussed here about the turbo engines?
I'm lost.

2014 F1 is not for turbo engines, it's for turbocharged turbocompounded engines

@gruntguru
the 1988 Honda job was to get the best race performance from 150 litres of (84% Toluene) fuel within 2.5 bar abs 'boost'
the Honda paper surely implies that at many tracks running lean would have lost them the race but only used 135 litres of fuel
ie to use the fuel allowance if running lean they would need to use higher rpm (the only way permtted to increase the air massflow)
so degrading the bsfc below the values you have 'cherry-picked'
IMO, to put it bluntly, the paper does not say what you seem to be implying it says
only Honda knows what % of their viscous high boiling point fuel was burnt when they won the WCs running at 2% rich
but only you seem to think that in 2014 with 500 bar DI F1 will need to run 10% lean for full combustion of the 2014 fuel

Johnson Matthey reminds us that our road cars continuously run 2% rich (for 3-way catalysis)
surely according to your ideas there is a high % of the fuel going unburned as there is no leaning ?
fuel unburned in the engine will be burned in the catalyst
so (according to your ideas) enough to burn it out (given that the 2% rich part is already burning there)

btw
the Wright example is at FAR of about 0.057 (well lean)
ie power is reduced by leaning not throttling
the pilot's 'throttle' lever was not connected to the throttle plates
like modern F1 etc

[Abarth]

[....]The cost of compressing the higher massflow is more than compensated by expanding the additional massflow in the turbine. [...]

Sounds like a perpetual motion to me...if you want to increase charge air mass flow, the compressor needs more power.
This has to come from exhaust mass flow. How can you have more exhaust mass flow than the amount you have previously charged, with constant fuel flow?

[gruntguru]

. . . the 1988 Honda job was to get the best race performance from 150 litres of (84% Toluene) fuel within 2.5 bar abs 'boost'
the Honda paper surely implies that at many tracks running lean would have lost them the race but only used 135 litres of fuel
ie to use the fuel allowance if running lean they would need to use higher rpm (the only way permtted to increase the air massflow)
so degrading the bsfc below the values you have 'cherry-picked'
IMO, to put it bluntly, the paper does not say what you seem to be implying it says
only Honda knows what % of their viscous high boiling point fuel was burnt when they won the WCs running at 2% rich
but only you seem to think that in 2014 with 500 bar DI F1 will need to run 10% lean for full combustion of the 2014 fuel

You have tried to respond to my post without even checking the graph or text I referenced - which clearly show that even at WOT, best economy occurs at significantly leaner than stoic' - contrary to the statements by yourself and Abarth.

[gruntguru]

Johnson Matthey reminds us that our road cars continuously run 2% rich (for 3-way catalysis)
surely according to your ideas there is a high % of the fuel going unburned as there is no leaning ?
fuel unburned in the engine will be burned in the catalyst
so (according to your ideas) enough to burn it out (given that the 2% rich part is already burning there)

If an engine is run at 2% rich, the exhaust will contain 2% unburned products (CO and HC) - even post cat. Engines that are tuned 2% rich for NOx reduction will typically use air injection post 3-way cat to lean out the exhaust gas prior to further processing in an oxidising cat to remove the CO and HC.

[gruntguru]

the Wright example is at FAR of about 0.057 (well lean) ie power is reduced by leaning not throttling
the pilot's 'throttle' lever was not connected to the throttle plates like modern F1 etc

Again - the point you might have seen in the Wright data is that best efficiency occurs well lean of stoich - even under high load operation.

And no - the power was not controlled by leaning the mixture. The 10% power drop was a measure used to establish the correct cruise mixture.

[gruntguru]

[....]The cost of compressing the higher massflow is more than compensated by expanding the additional massflow in the turbine. [...]

Sounds like a perpetual motion to me...if you want to increase charge air mass flow, the compressor needs more power.
This has to come from exhaust mass flow. How can you have more exhaust mass flow than the amount you have previously charged, with constant fuel flow?

The point is that although additional air mass flow increases the compressor power requirement - the same extra mass flow is available in the exhaust and increases turbine power by more than enough to cover the shortfall.

Think gas turbine (Brayton cycle). Simple Brayton cycle efficiency is a function of pressure ratio - the higher the pressure ratio the greater the efficiency. (Check Wikipedia for a formula.)

[Abarth]

[...]You have tried to respond to my post without even checking the graph or text I referenced - which clearly show that even at WOT, best economy occurs at significantly leaner than stoic' - contrary to the statements by yourself and Abarth.

1. The graph in the Honda article doesn't show specific consumption mixtures lower than equvalence ratio = 1 (or higher than lambda =1).
So how can you state "significantly leaner" based on that paper?

2. The Honda engine did not have direct injection, not to speak about very high pressure DI. You can't compare quantitatively carburettor or manifold injection engine BSFC graphs with external mixture formation and therefore more or less homogeneous mixture with DI stratified injection.