Talking to a turbo expert

[Tommy Cookers]

If there’s interest, I can start a more appropriate thread

yes please, fire away !!

[WhiteBlue]

When the Porsche MR6 (V8) engine was released it had a port injected fuel system, but was later modified for direct injection.

That explains the discrepancy.
I have to apologise for going a bit off topic here, but using the fastest direct injection systems would obviously contribute to combustion efficiency. 100-120 bar isn't going to give you a real quick injection and no really good atomization either. Perfect spray at very high pressure and short opening times would be an advantage for a fuel flow limited racing engine, I believe. Even if you do not achieve perfect stratification, the combustion should be a lot leaner than in port injection or slow old direct injection systems.

[Tommy Cookers]

I have to apologise for going a bit off topic here, but using the fastest direct injection systems would obviously contribute to combustion efficiency. 100-120 bar isn't going to give you a real quick injection and no really good atomization either. Perfect spray at very high pressure and short opening times would be an advantage for a fuel flow limited racing engine, I believe. Even if you do not achieve perfect stratification, the combustion should be a lot leaner than in port injection or slow old direct injection systems.

so you're saying that pumping more air through the engine than is necessary for combustion (ie leaner) is your preference ?
isn't that reducing efficiency because of the extra pumping work by the compressor in pumping the surplus air ?

[WhiteBlue]

so you're saying that pumping more air through the engine than is necessary for combustion (ie leaner) is your preference ?
isn't that reducing efficiency because of the extra pumping work by the compressor in pumping the surplus air ?

I don't think that pumping a bit more air is an issue. The power required for charge air compression comes out of the exhaust energy reclaim anyway. I was talking about fuel pressure and achieving a hollow cone of fuel spray which is what you do in spray guided combustion.
The difference to PFI is in the timing and duration of the injection pulse. GDI works in the compression stage in an interval close to a millisecond. It allows you to take the compression higher because it helps suppressing engine knocking. More compression and leaner combustion are advantages in fuel flow restricted engines. It helps to get more power from the limited fuel.

[riff_raff]

so you're saying that pumping more air through the engine than is necessary for combustion (ie leaner) is your preference ?
isn't that reducing efficiency because of the extra pumping work by the compressor in pumping the surplus air ?

I don't think that pumping a bit more air is an issue. The power required for charge air compression comes out of the exhaust energy reclaim anyway. I was talking about fuel pressure and achieving a hollow cone of fuel spray which is what you do in spray guided combustion.
The difference to PFI is in the timing and duration of the injection pulse. GDI works in the compression stage in an interval close to a millisecond. It allows you to take the compression higher because it helps suppressing engine knocking. More compression and leaner combustion are advantages in fuel flow restricted engines. It helps to get more power from the limited fuel.

Just what type of engine configuration, combustion cycle, etc. is the optimum solution depends on many factors. The preferred approach for most automotive engines currently seems to a downsized, highly boosted, GDI design. This gives the best combination of cost, fuel consumption, driveability and emissions. If cost and emissions were no concern, a highly boosted DI diesel would give better fuel consumption.

The biggest factors with brake thermal efficiency are peak cycle temperatures and pressures. The reason a turbocharger helps efficiency is because, overall, the piston engine and turbo combination performs the gas compression/expansion work more efficiently than the piston engine can by itself.

One thing that is very interesting to note is the relative compression ratios currently used by turbocharged GDI engines and turbocharged DI diesel auto engines. The compression ratios are now identical, at around 14:1.

http://www.mazda.com/mazdaspirit/skyact ... tiv-g.html

slider

[Tommy Cookers]

I was talking about fuel pressure and achieving a hollow cone of fuel spray which is what you do in spray guided combustion.
The difference to PFI is in the timing and duration of the injection pulse. GDI works in the compression stage in an interval close to a millisecond. It allows you to take the compression higher because it helps suppressing engine knocking. More compression and leaner combustion are advantages in fuel flow restricted engines. It helps to get more power from the limited fuel.

GDI powered aircraft were filling the skies over 70 years ago, its benefits to CR became well known then
what interests me is your apparent advocacy of leaner combustion (leaner than what?)

simply, you are suggesting either running a significantly leaner than stoichoimetric mixture, or running a near stoichiometric mixture ?
which do you envisage would "get more power from the limited fuel" ?

[FW17]

Does GDI have multiple ignitions like in a TDI?

Will the new GDI have a EGR system?

[Tommy Cookers]

Just what type of engine configuration, combustion cycle, etc. is the optimum solution depends on many factors. The preferred approach for most automotive engines currently seems to a downsized, highly boosted, GDI design. This gives the best combination of cost, fuel consumption, driveability and emissions. If cost and emissions were no concern, a highly boosted DI diesel would give better fuel consumption.

The biggest factors with brake thermal efficiency are peak cycle temperatures and pressures. The reason a turbocharger helps efficiency is because, overall, the piston engine and turbo combination performs the gas compression/expansion work more efficiently than the piston engine can by itself.

One thing that is very interesting to note is the relative compression ratios currently used by turbocharged GDI engines and turbocharged DI diesel auto engines. The compression ratios are now identical, at around 14:1.

http://www.mazda.com/mazdaspirit/skyact ... tiv-g.html

slider

heat release wrt time/piston position is important (eg at high rpm), it also (in part) determines what CR is achievable
do people envisage (eventually ) F1 engines multiply-injecting ie including during the combustion phase
if they don't do that the heat release will not be ideally managed relative to piston position
we may guess that the 2014 lower rpm will give better behaved (ign-timed) heat release than in current (18000rpm) engines

interestingly, current F1 is not Octane No critical (eg at such high rpm)
and the fuel rules now allow unlimited max ON
so the main benefit presumed of DI (ie CR gain) might not materialise in F1

surely the (distant) goal should be to optimise heat release rate by (in part) injecting during combustion (diesels do !) ??
only that will give a real efficiency gain to the crankshaft power
otherwise we will go down a slippery slope towards turbinisation,

regarding turbos, what IYO is the optimum boost ? eg 0.2 bar, 1.2 bar, 5 bar, etc (2014 rules force about 1.2 bar)

(aircraft) engines boosted by mechanically-driven compressors to around 0.2 bar showed BTEs over 35% 60-80 years ago

[Edis]

I have to apologise for going a bit off topic here, but using the fastest direct injection systems would obviously contribute to combustion efficiency. 100-120 bar isn't going to give you a real quick injection and no really good atomization either. Perfect spray at very high pressure and short opening times would be an advantage for a fuel flow limited racing engine, I believe. Even if you do not achieve perfect stratification, the combustion should be a lot leaner than in port injection or slow old direct injection systems.

No, a fast direct injection system will contribute to poor mixing and unstable combustion. That's the problem when you need to do a fast injection just before TDC to achieve charge stratification; in other words you need a high pressure to deal with the short injection and mixing time availible.

For high loads you inject the fuel during the intake stroke instead. That way there is sufficient time for injection and mixing, and you will get a good burn without having to use extreme fuel pressures. After all, the pressure in the cylinder during the intake stroke isn't really higher than in the inlet manifold. The air flowing past the intake valves will also aid the mixing and vaporisation, and there are no problems with wetting the walls of the intake ports. Also, no fuel will lost to the exhaust during valve overlap, and the greater charge cooling effect will allow a slightly higher compression ratio. It's these factors that are behind the 5% reduced fuel consumption Illien mentioned.

Just what type of engine configuration, combustion cycle, etc. is the optimum solution depends on many factors. The preferred approach for most automotive engines currently seems to a downsized, highly boosted, GDI design. This gives the best combination of cost, fuel consumption, driveability and emissions. If cost and emissions were no concern, a highly boosted DI diesel would give better fuel consumption.

The biggest factors with brake thermal efficiency are peak cycle temperatures and pressures. The reason a turbocharger helps efficiency is because, overall, the piston engine and turbo combination performs the gas compression/expansion work more efficiently than the piston engine can by itself.

One thing that is very interesting to note is the relative compression ratios currently used by turbocharged GDI engines and turbocharged DI diesel auto engines. The compression ratios are now identical, at around 14:1.

http://www.mazda.com/mazdaspirit/skyact ... tiv-g.html

slider

Direct injection and turbocharging is certainly an advantage for fuel consumption, but hardly for emissions. Infact, they both tend to be troublesome for emissions, particulary direct injected engines using lean mixtures and/or charge stratification. That's why many direct injected engines only use homogeneous charge modes.

When you've got air excess in the exhaust, like a GDI engine during lean and charge stratified modes a regular three way catalyst won't be able to reduce NOx emissions. This is just like the oxidation catalyst on a diesel engine, which only oxidize HC and CO. So with these GDI engines you need a NOx storage catalust, this catalyst is able to store NOx when the engine operate with lean/statified modes, and when it is full the engine switches to homogeneous charge mode to allow the catalyst to reduce the stored NOx to N2 and O2, this is done at lambda 1 or richer. Stratified charge modes also tend to cause heteregeneous combustion since mixing is poor, and this increase formation of particulate matter, PM, just like a diesel. That's why GDI engines have to meet the same PM limit as diesels these days.

The reason you turbocharge a gasoline engine isn't for efficiency, the engine will do the compression and expansion work itself just fine. Instead you use turbocharging to compensate for the powerloss caused by downsizing. A smaller engine will operate at a more efficient load point during "average driving" due to smaller pumping losses (less throttling), less friction and smaller heat losses. Unfortunatly a smaller engine is also less powerful, making the car underpowered when you need to accelerate. By turbocharging the engine you can reach the desired engine output with the smaller more efficient engine. So, a smaller turbocharged V6 isn't more fuel efficient than a V8 because it's turbocharged but because it's a smaller V6 engine. The turbo simply compensates for the loss of two cylinders in terms of power.

In the past diesels tended to use very high compression ratios simply to improve the startability at low temperatures, for efficiency there is probably not much point to increase the compression ratio more than somewhere between 14-18:1. Doing so will improve thermal efficiency very little while frictional losses will increase causing an overall drop in engine efficiency. Lately tougher emissions requirements have demanded lower NOx emissions. Since NOx from a diesel engine is mostly thermal NOx, formed at high temperatures, EGR and lower compression ratios have been used to lower combustion temperature and with it NOx emissions. This does however have an disadvantage, when combustion temperature is decreased oxidation of PM reduces causing higher PM emissions. So to deal with PM higher and higher fuel injection pressures are used, as this will reduce PM emissions.

Do however note that the Mazda GDI engine is naturally aspiranted, not turbocharged. The two stage charging used by the diesel is probably capable to deliver an absolute boost pressure around 3 bar with that 14:1 compression ratio.

[WhiteBlue]

I have to apologise for going a bit off topic here, but using the fastest direct injection systems would obviously contribute to combustion efficiency. 100-120 bar isn't going to give you a real quick injection and no really good atomization either. Perfect spray at very high pressure and short opening times would be an advantage for a fuel flow limited racing engine, I believe. Even if you do not achieve perfect stratification, the combustion should be a lot leaner than in port injection or slow old direct injection systems.

No, a fast direct injection system will contribute to poor mixing and unstable combustion. That's the problem when you need to do a fast injection just before TDC to achieve charge stratification; in other words you need a high pressure to deal with the short injection and mixing time availible.

For high loads you inject the fuel during the intake stroke instead. That way there is sufficient time for injection and mixing, and you will get a good burn without having to use extreme fuel pressures. After all, the pressure in the cylinder during the intake stroke isn't really higher than in the inlet manifold. The air flowing past the intake valves will also aid the mixing and vaporisation, and there are no problems with wetting the walls of the intake ports. Also, no fuel will lost to the exhaust during valve overlap, and the greater charge cooling effect will allow a slightly higher compression ratio. It's these factors that are behind the 5% reduced fuel consumption Illien mentioned.

Edis, your considerations are certainly useful for an air limited formula but the discussion here was meant for fuel flow limited F1 turbo engines. At least a big part of it. If you check the 2014 rules they do not allow a significant portion of the fuel to be port injected. If memory serves me right it is just 20%. So if you are committed to inject the vast majority of the fuel shortly before TDC you are going to use the most efficient combustion method available to you. And that is spray guided AFAIK. Spray guided combustion achieves a certain amount of stratification by design and the rules leave some head room for even faster injection than we know from todays existing 200 bar systems. They are allowed to use 500 bar. The engines will not go much beyond 10,500 rpm either because they will be fuel starved above that limit.
I cannot answer the question what kind of AFR they will be using at different power levels but I'm confident that they plan to burn leaner than the current crop of air limited engines. The logic of the formula enforces that strategy. As I have said before I'm not working in the field but I can read the rules and I believe that your proposal does not fit the new rules from 2014. So I would be interested to learn what you think will be the injection method for a 2014 project according to your best information.

[WhiteBlue]

The reason you turbocharge a gasoline engine isn't for efficiency, the engine will do the compression and expansion work itself just fine. Instead you use turbocharging to compensate for the powerloss caused by downsizing.

If I read you right you deny that the turbocharged engine will be intrinsically more fuel efficient than a NA engine. I find that hard to believe. If you go back to the example of the Porsche Cayenne V8 engine you will find a discrepancy with your statement. Porsche did not downsize the engine but added a turbocharger in order to raise the power. If you look at the figures they suggest that the turbocharged version is considerably more fuel efficient than the naturally aspired version. This is also what the Garrett expert whose words opened this thread suggest for a road going engine. He said that you turbocharge for better driving experience and fuel efficiency.

If we look at the last time we had turbos going against NA engines in F1 it certainly looked like the turbos were more fuel efficient than the NAs.

There is also the indisputable fact that adding a turbocharger will reduce the kinetic and thermal energy level of the exhaust gas at the tail pipe. All other things being equal that necessitates a higher efficiency of the turbocharged engine. The turbo engine can convert that energy difference into useful power that is wasted by the NA engine.

I'm not saying that downsizing profits only from turbo charging. There is also the aspect of the improved mechanical efficiency that you describe. Both effects are contributing to the success of downsized engines. It would be wrong IMO to deny any of the two effects their contribution to the efficiency improvement.

[Tommy Cookers]

If we look at the last time we had turbos going against NA engines in F1 it certainly looked like the turbos were more fuel efficient than the NAs.

There is also the indisputable fact that adding a turbocharger will reduce the kinetic and thermal energy level of the exhaust gas at the tail pipe. All other things being equal that necessitates a higher efficiency of the turbocharged engine. The turbo engine can convert that energy difference into useful power that is wasted by the NA engine.

the NAs were never asked to be fuel efficient
(they were asked to give maximum power on 195 litres of density-unrestricted fuel, ie any % of Toluene)
they ran on the usual rich mixture (ie passing about 20% more fuel than the air could burn)
(their BTE would have improved about 12-15% on a near stoichoimetric mixture, with some power loss)
the Mercedes M119 Group C had a best BTE of 35% and a max-power BTE of 32%
the Honda turbo in 1988 period that you cite has a best BTE (in economy mixture setting) of 30.5%
the Lamborghini F1 had a max power BTE of 27% on 'full-rich' mixture, surely matching the turbo BTE when on leaner mixture

adding a turbocharger will always increase the energy of the exhaust upstream (adding a 'waste' component, to be recovered)
(partly because in a fair comparison the CR (and cylinder thermal efficiency) will be lower), also due to the increased MEP)

certainly with a recovery-sized turbine, raising the turbine load can raise exhaust pressure and degrade piston-to-crankshaft BTE
this can be useful, crankshaftless piston engines have even been made to exploit this
reducing fuel rate eg after 2014 will take us in this direction ?

if a turbo engine is always more efficient, wouldn't this take us to 3 bar, 5 bar, 7 bar etc ?
EDIT (note to self)
the 1988 Honda fuel (84% Toluene) had a calorific value around 41.41 MJ/kg abou 17700 BTU/lb
the best BTE better than above (Edis gave 32.2% in Efficiency thread 27 Oct)

[WhiteBlue]

the NAs were never asked to be fuel efficient
(they were asked to give maximum power on 195 litres of density-unrestricted fuel, ie any % of Toluene)
they ran on the usual rich mixture (ie passing about 20% more fuel than the air could burn)
(their BTE would have improved about 12-15% on a near stoichoimetric mixture, with some power loss)
the Mercedes M119 Group C had a best BTE of 35% and a max-power BTE of 32%

I was angling on a line of thought that racing engines might be different from road going engines because they run 75% on full power compared to road cars running on average on much lower power settings. Nevertheless turbos seem to have been more fuel efficient in the 80ies era and your post does not seem to contradict that impression.

adding a turbocharger will always increase the energy of the exhaust upstream (adding a 'waste' component, to be recovered)
(partly because in a fair comparison the CR (and cylinder thermal efficiency) will be lower), also due to the increased MEP)

certainly with a recovery-sized turbine, raising the turbine load can raise exhaust pressure and degrade piston-to-crankshaft BTE
this can be useful, crankshaftless piston engines have even been made to exploit this
reducing fuel rate eg after 2014 will take us in this direction ?

You lost me there with your language and syntax. I don't understand what "exhaust upstream" means. And I got totally confused by all these brackets. All I'm getting is that a crankshaftless piston engine exists which I don't know anything about. Perhaps you give us a reference for that engine to study?

if a turbo engine is always more efficient, wouldn't this take us to 3 bar, 5 bar, 7 bar etc ?

How do you arrive at that conclusion? The level of boost will depend of other features like legal rev limit, fuel flow rules or allowing turbo compounding. Changes to those aspects will shift the optimum boost level.

[Tommy Cookers]

REVISED

If we look at the last time we had turbos going against NA engines in F1 it certainly looked like the turbos were more fuel efficient than the NAs.

the NAs were never asked to be fuel efficient (not since 1908-1913, when 9.4 mpg was required)
for 1988 they were asked to give maximum power on 195 litres of fuel, which had little or no Toluene
they ran on the traditional rich mixture (ie passing at least 16% more fuel than the air could burn)
(their BTE would have improved about 10-11% on a near-stoichoimetric mixture, with some power loss)
the NA Lamborghini F1 had max power BTE of 27% on 'full-rich' mixture, corresponding on n-s mixture to at least 29.5%
(V8s and/or V10s would surely have reached 30% BTE ?)

the 1988 Honda turbo F1 engine had max power BTE (on a near-stoichiometric 'economy' mixture setting) of 32.2% (Edis post in Efficiency thread) on fuel with 84% Toluene ie low calorific value/kg
the Honda SAE paper shows a 12% gain in BTE going from 18% rich to stoichiometric mixture, they mostly ran 2% rich


certainly with a recovery-sized turbine, raising the turbine load raises exhaust pressure upstream of the tubine, degrading piston-to-crankshaft BTE
this can be useful, also, reduced fuel rate eg after 2014 will take us in this direction ?
piston engines without crankshafts ('free-piston' engines as gas generators with turbines) fully extended this concept in the 1950s
(interestingly, free-piston engines driving linear electrical generators are being seriously researched today)

[riff_raff]

........if a turbo engine is always more efficient, wouldn't this take us to 3 bar, 5 bar, 7 bar etc ?

If BTE is your primary concern, and assuming we can disregard other things like size, weight, throttle response, etc., then using as much boost as possible, within the detonation, thermal loading, and mechanical stress limits of the engine, will usually produce the best efficiency. BTE generally improves when the cycle pressure ratios are increased.

Turbo machinery is very good at performing the air compression and exhaust gas expansion work, up to a certain point. Dynamic compressors begin to suffer from tip leakage when the pressures get too high, and turbines have definite temperature limits. So the best compromise is to do as much compression/expansion work as practical using turbo machinery, with the rest of the cycle work being done by the piston engine which can handle much greater peak cycle pressures and temperatures.