Here's a cool idea that you may or may not have in the back of your mind—why not a petrol-fueled diesel-combustion-cycle engine? Advantages would include cleaner burning of the gasoline fuel combined with greater efficiency.
The folk at the U.S. Argonne National Laboratory are working on the technology, as well as industry groups investigating such technology as HCCI (homogeneous charge compression engine).
But rather than me attempting to fill you in, here is a blog just posted by Steve Ciatti, a mechanical engineer at the Laboratory, about their research.
The only comment I would offer is that I disagree with his statement, "There’s really no significant improvement that can be achieved with either one [gasoline or diesel combustion alone]." It seems to me that with each industry meeting, such as Convergence coming up in Detroit in October, there is always some news of improvements to these internal combustion engines that design engineers, thanks to microelectronics-enabled controls, can continue to wring out of these high power-density systems.
Why not add a comment here and let us know your opinion?
For combustion engine efficiency:
1.) Process fuel droplet diameters below 6 microns. Any combustion still in progress when the exhaust valve opens is called "waste" by definition. You don't need turbo boosting if you have enough cubic inches for the drive train and vehicle weight so extract as much energy from the fuel as possible through the crankshaft.
2.) Use asymmetric valve timing to permit differential compression / expansion cycles. The mechanical conversion efficiency is determined by: peak temp - EGT / peak temp. The larger the differential, the higher the efficiency... by definition.
3.) Water injection with 5 micron droplets after warm up under medium or high demand conditions. Water injection must be shut off at low throttle settings (idle) if the droplet diameters exceed 5 microns or the plugs get wet and won't fire... NOT good.
4.) Use fractured zircon coatings on piston crowns and the under the heads to reduce energy losses.
5.) Set spark plug firing position in the compression cycle as close to the Lambda position as possible for the operating RPM and work load.
6.) Use roller cams for friction reduction.
7.) Use roller crank shafts for friction reduction.
8.) Use diester based tribology for better shear stability, temperature tolerance and friction reduction.
9.) If you want engine longevity, use larger crankshaft journals, ultra fine oil filtering, bearing steel cylinder sleeves and stellite rings. Bronze banded rings also reduce some friction.
10.) Use HydrOx fuel augmentation... especially with heavier alkane series fuels.
Since you don't see these engine design features on any commercially available platforms we have to assume that the upper limit of combustion efficiency is not yet known... well... at the very least it has not yet been released for public use. After 150 years of IC engine experience do we still not yet know how to engineer these systems properly? How come?
Here in Australia we have large gas reserves too. All of our taxi's run on what we call LPG or Liquefied Petroleum Gas and is derived from natural gas and/or petroleum. New Zealand uses CNG otherwise known as Compressed Natural Gas. Yep it burns cleaner but as you say less calories per litre. Down here it's about half the price of gasoline.
I do remember using water injection in the 1970's, strange they were able to patent it in 1987?? I guess you can patent anything if you make it sound complicated enough.
What I will say with water injection, you need to wait until the engine is up to temperature before using it else engine wear will increase. It's also a good idea to turn it of around 15-30secs before stopping the engine which means you would need variable compression or take the NOX hit.
I remember Mercedes with a similar study, the Diesotto. But AFAIK they have both an injector and a spark plug. In Italian cities you may find natural-gas-fed street buses, which have a diesel engine and uses a small diesel fuel injection to ignite the natural gas mixture.
Using liquid fuel with good evaporation enthalpy allows to do the same as water injection: evaporating gasoline steals compression heat. That's already used in GDI engines. I agree with Wketel in the fact that better efficiency means more NOx.
Reading the linked blog, once again I see the usual problem: a good idea for the average power demand has issues when dealing with peak power requests. Man, don't you see everywhere a blinking sign "go hybrid go hybrid"?
Sticking an advertising message in here gets you invited to not come back. The challenge with internal combustion engines is that what improves efficiency also increases nitrogen oxides formation. Adding water can slow things down, but it makes it much more complex. The urea addition function for diesels is a thing that has to constantly be replenished and it will be a nightmare when it is fully implemented. And nobody knows what urea emissions will do. The challenge is to have slower burning, since when the mixture explodes, we have destructive knock, also known more correctly as detonation. One good choice would be to use natural gas as a fuel, and add a turbocharger to improve the performance and efficiency. The emissions are nicer to deal with, unfortunately the specific energy of natural gas is not as high. Fuel injection that continued after ignition is an interesting possibility, but it would be a technical challenge to make it last. Things that work well fr a quarter-mile car may not all be practical for other applications. But, the huge advantage with natural gas is that we have lots of it right here in the USA!
So theres a 'DenoX' solution applied already to diesel engines. This is catalytic reaction supported by injecting Urea to decompose the NOx. Some automakers are trying now to port the solution to petrol engines. Potentially huge petrol saving by allowing lean combustion and also with turbocharged units.
I'm not an expert on this topic, but I seem to recall that there are also emissions issues with high-compression gasoline engines. In particular, I vaguely recall that on some of the early '70s small American cars, they jacked up the compression ratio in order to get mid-30s MPG (extraordinary mileage at the time), but they ended up producing a lot of nitrogen oxides.
David Patterson, known for his pioneering research that led to RAID, clusters and more, is part of a team at UC Berkeley that recently made its RISC-V processor architecture an open source hardware offering. We talk with Patterson and one of his colleagues behind the effort about the opportunities they see, what new kinds of designs they hope to enable and what it means for today’s commercial processor giants such as Intel, ARM and Imagination Technologies.