A friend pointed out that a GDI engine could run w/o a throttle plate like a diesel and be governed by the fuel/ignition controller. This would reduce backwork and improve fuel economy except at WOT.
Running without a throttle plate would increase pumping losses because the engine would always be running at 100pct air flow.
If it ran, it wouldn't be running an Otto cycle. I'm not sure it would run at all with that much excess air.
If the ECU were not set up to compensate somehow (and I'm not convinced it could be), the engine would most likely IMHO detonate, overheat, and self-destruct, while running like crap.
I stand corrected. It can be done, and all it takes is roughly twice the high tech stuff it takes to run an electronic Diesel, plus the extra mass for some fancy valve gear.
I'm still fuzzy on how they can get a stoich or rich mixture without throttling the intake. ... unless they're throttling by diddling the valve timing or lift.
Ah. I think I understand. Because of the high injection pressure, they can shoot in all of the fuel for a cycle within a very short time just before the spark, so the charge can be locally stoich or rich, and the remainder super-lean.
Thanks all for the education.
I think I'll wait a while to do it to my lawnmower...
BMW's HP-DI with Valvetronic does not have a conventional butterfly throttle. Valve timing meters the amount of air that goes into the engine. This system runs lambda 1 to be compatible with a 3-way catalyst (TWC).
VW/Audi (T-)FSI has a throttle and also runs at lambda one with a TWC. Ditto Mercedes-Benz and GDI systems from GM, Ford, Hyundai, etc.
All the above implementations could also run in lean/stratified mode, but the ineffectiveness of a TWC (and the high tailpipe NOx that would result) in that regime and the need for a lean-NOx trap catalyst (more complex, more expensive) results that nobody is currently implementing it where Euro 5 / Tier 2 Bin 5 regs need to be met.
There's plenty of disagreement even in engine circles regarding the distinction between the Miller and Atkinson cycles but they ultimately achieve much the same thing and some even (mistakenly) use the terms interchangeably. The thermodynamic cycles when plotted on P-v and T-S diagrams and the true implementations of each cycle are very distinct and much more complex than simply varying valve timing in an otherwise conventional engine.
For the purposes of most discussions I use the convention that the Miller cycle refers to late intake valve closing causing some of the cylinder charge to flow back into the intake ports, while Atkinson denotes very early intake valve closing before BDC resulting in a pre-expansion before the compression stroke proper.
TDI - I am afraid I must disagree with your description of the Atkinson Cycle. If you go back to the original idea of an Atkinson Cycle it was with a engine that had a physically longer expansion stroke than compression stroke - thus having a higher TE.
The modern equivalent of this I would have thought would have to be the high geometrical CR/LIVC approach.
The Miller Cycle is never as clear to me but I think to be classed as "Miller" it must involve a supercharger plus an element of Atkinson Cycle. Just why the supercharger is a better method of cylinder-filling I have never quite been able to picture.
The EIVC approach I don't think has a cycle name and is much less effective than Atkinson (especially) or Miller.
In the above reference, the authors use EIVC as Miller and LIVC as Atkinson. Mazda uses the term Miller to denote LIVC with supercharging; Toyota uses Atkinson to denote LIVC in the naturally aspirated Prius engine. You can see that there's no real agreement on the nomenclature convention. A survey of literature shows that Miller and Atkinson can denote both EIVC and LIVC, but Miller is usually combined with some form of forced aspiration, but Atkinson is (usually) normally aspirated.
I never paid much attention to that concept, but I always figured that the supercharger was mainly there as a checkvalve to prevent the airflow sensor from getting an irregular signal.
TDI - you originally wrote above: "Atkinson denotes very early intake closing before BDC resulting in a pre-expansion etc."
I don't see how this can be construed as Atkinson by anbody's definition.
Draw a P-v diagram and you will see how expansion is longer than compression (the latter of which is to include the pre-expansion), which is the intent of the Atkinson -- and Miller -- cycle. Better, find the MTZ article I cited; it will even illustrate the P-v diagram for you.
Commercial engines using Miller cycle use high pressure ratio turbocharging in order to realize the power density needed to have a cost-effective product.