New Diesel Style Gasoline Engine
 

This is interesting, I think:

Engine Could Boost Fuel Economy by Half - Technology Review

They say it would be about 45% efficient (similar to diesels) but that it would have much less dirty exhaust. One would hope that the displacement could be proportionally smaller, as well.

Sounds kind of like a fork from HCCI?

This is the same HCCI engine GM has been working on for years.
http://www.youtube.com/watch?v=FOKZBAJM180

That makes sense; Delphi is a major supplier to GM, right? I thought it was interesting that the main challenge with using gasoline in a diesel style (compression ignited) engine is to *slow down* the combustion. With higher RPM spark ignited engines, the problem often is flame spread not burning all the fuel; especially in the various nooks and crannies like on the sides of the piston just above the first ring.

I wonder if they tried something similar to the Ducati Supermono, where they found they could prevent predetonation (due to high heat and compression) by having the injector spray a *stream* of fuel (rather than a spray) that then evaporated when it hit the hot crown of the piston? This is also a counterintuitive, as is 3 separate bursts of gasoline, used in this Delphi design.

Delphi Automotive is a spin off of GM. It basically allows GM to sell its in house tech to other OEMs without the stigma of GM. Magnetic shock absorbers is one GM tech that made its way to Ferrari and just about every luxo barge on the road today through Delphi. Delphi became 'fully independent' in 1999 when it went public but GM remains its largest shareholder.

As I remember it one of the main issues with HCCI was that it was very sensitive to the different types of gasoline sold throughout the country (ethanol content, summer/winter fuel ect) so while the technology is very solid, its application in the real world isn't. The other issue is cost. It requires a very high pressure piezo fuel injection system (1000bar +) much like modern diesels and very powerful engine management. Most modern diesels fire more than one burst of fuel per cycle (some up to 10) so this isn't really an issue technically.

This article is good news since HCCI sort of dropped off the radar for GM a few years ago. If Delphi can bring it to market at another manufacturer that would be great.

Imagine how well they could design an engine if it only had to run at one RPM and load regardless of power train demands. Much easier than trying to make an engine run efficiently at all loads.
One size fits all loads will always be a huge stumbling block for engine design. It should be one RPM, one specific load, then you could redesign every component for that load and RPM. Lighter pistons, rods, crankshaft (if it had one), valves (if it had them), bore to stroke ratio could be 1 to 2 or 3 or even more. No throttle restriction, only a fixed amount of fuel delivery.
It will happen when they realize that there are alternatives to connecting the engine-motor directly to the wheels.

regards
Mech

Old Mechanic -- this might interest you, HCCI + hydraulic: Hydraulic Hybrid Research | Clean Automotive Technology | US EPA

Is engine speed really that much of an issue? Basically all normal driving happens under 4000rpm, between 2000 and 4000 what you lose in friction you gain in reduced cooling loss, unavoidable with these engine sizes. Without a very powerful hybrid assist there's no way to avoid the wide range of loads either. A mechanical gearbox and clutch is simple, cheap, lightweight, transmits power efficiently, and with the correct gear ratios allows the engine to run reasonably close to its maximum efficiency.

Considering cost restraints and fueling infrastructure, I think the way things are advancing mostly makes sense. Try to improve low load efficiency as much as possible (by sacrificing high load efficiency if necessary, which is what we're seeing with higher compression ratios + running extra rich and lots of spark retard at high load), more sensible gear ratios for fuel economy, downsizing and forced induction to make up for the power deficit, and slowly adding brake (and exhaust heat) regeneration capability and improving accessory efficiency. Compared to vehicles running near maximum possible tank to wheel efficiency via some sort of complex hybrid system, these steps can probably get us most of the way there at a much lower upfront cost (which is important as cars get replaced quite frequently).

From what I was reading on Mazda's web site, the next iteration of the Skyactiv engine is going to be HCCI.

To me, there's too much overcomplexity ... too many computer parts to go wrong.

Gas explodes, diesel burns. They really have to do some neat things to switch back and fourth I bet.

I went to Detroit in 2006 to meet with some people at Ricardo and Next Energy. I had just built my demonstration model for my IVT. Brian Waddington, from Next Energy went with me to Ricardo where they basically told me they were not prepared to consider a in wheel hydraulic IVT. Mr. Waddington said my presentation was one of the best he had ever seen, then two weeks later he went to a new job at an investment bank in New England and I lost my best contact in Detroit.

I have some of the 2006 EPA documents and one of them was basically a plea for a new "clean sheet of paper design" for a hydraulic pump to replace the bent axes pump they were using in their vehicles at that time.

The model I carried with me to Detroit in June 2006 was 3 months old when I went there in 06. I had just reversed the cylinders and pistons and realized the potential of the design. The photo attached is one of the graphs from their 06 documents. I still have it 6 years later. It projects 58 MPG EPA combined for a class 2b commercial truck.

regards
Mech