Runge_Kutta

A few comments on GDI/SIDI in regard to GM. From a casual awareness of many SAE
papers and several patents, it appears that they are simultaneously going after
2V, 3V, and 4V applications. The 2V and 3V applications look like they are aimed
at OHV designs while the 4V applications look like they are destined for DOHC
engines. While I may be wrong, I think side injection-set ups are likely using
high-pressure fuel designs (HPDI) but central injection is likely to use a
low pressure direct injection (LPDI) with air assist. Judging by the recent patents,
the 2V
http://www.uspto.gov/web/patents/pat...-20031125.html

and 4V set-ups
http://www.uspto.gov/web/patents/pat...-20040608.html

are likely side injection HPDI but the 3V design is likely air-assisted LPDI by virtue
of the physical space available at the top of the combustion chamber.
http://www.delphi.com/pdf/techpapers/2003-01-0062.pdf (page 9, figure 11)
http://www.sae.org/servlets/productD...D=2003-01-0545

(The pressure is unclear in 2003-01-0062 but the position of the injector is clear.)
This goes back to
http://finance.groups.yahoo.com/grou.../message/11598

I would also guess that the 2V engines are specifically targeted at the truck/SUV line.
I would be surprised if the Corvette saw GDI and it wasn't in a 3V, air-assisted LPDI
context. Regardless of engine, another potential source of nontrivial cost is the
after-treatment needed. This cost should be directly correlated with what levels of
sulfur the systems are designed to cope with.

As to benefit, a generic analysis in:

----------
JSAE Paper Number: 20004396 Sep, 2000 Issued Vol.54 No.9
Journal of Society of Automotive Engineers of Japan
Characteristics and Improving Ways of Thermal Efficiency on DI Diesel Engines
Author(s): Shuji Kimura
----------

gives the thermal efficiency of MPI (PFI) as 32.5% and 36.5% for GDI. A less
generic analysis is given by GM in SAE 2003-01-3101
http://www.sae.org/servlets/productD...D=2003-01-3101

The context is a 5.3L, 2V, V8 using HPDI side injection. The analysis says that
the GDI/SIDI/DISI set up offered a net 15% fuel economy advantage over a similar
PFI engine operating without EGR. They speculate that with refinement, this
could potentially be increased to 26%.

WhiteDiamond

Duke,

The GDI is being eyed by GM and many other manufacturers mostly for benefits in emissions. I agree that a cost will be associated with the pump, injectors and high pressure portion of the fuel line, but this cost will most likely be minimal as the engines go into production. With the comments from Audi saying anywhere from 10% to 15% better effeciency and the information in many of the links Runge is providing, I would be surprised if there isn't a GM GDI engine next year.

TexasRedLT4

As for the cost of sulfur removal and increased gas prices, I am particularly happy with both. I am interested in fuel cell research, so I see all these improvements(GDI, lean burn, etc) as stepping stones, but gas prices need to rise in the US and remain there so more imphasis is placed in alternative research. Our Escalade drinks the finite resource at incredible levels, and I see a ton of SUVs in Colorado, so it is obvious people just don't get it, yet.

Todd

SWCDuke

Having worked for GM I can assure you that no increase in cost is taken lightly and the cost of GDI is considerably higher than current port systems.
It will probably come, but like DoD, the benefits may serve GM's gas guzzling trucks more than the Corvette.

I used to be the first guy on the block to buy new technology, but after getting burned enough times I, nowadays, prefer well developed systems with proven reliability.

GDI is nothing new. Diamler Benz pioneered it on the DB601 engine in WWII that powered the Bf109 and also equipped several early post war auto engines with it including the 300SL. These early GDI systems were basically modified diesel mechanical injection systems. They proved to be not much better than port injection, and the minor improvement did not justify their cost, so it was abandoned.

Modern emission and fuel consumption targets have brought GDI back into the spotlight and electronic control casts them in a new light, but they will not create a quantum leap in performannce - emissions, fuel consumption, or power, though they do provide incremental improvements and in some applications the cost may be justifiable.

Electronically controlled port injection afforded a quantum leap above carburetors, but GDI is a much lower incremental improvement over port injection.

Duke




[Modified by SWCDuke, 8:51 AM 6/9/2004]

TexasRedLT4

WhiteDiamond,

You are correct. It will take continued high gasoline prices for the alternatives to become cost competitive and for continued research emphasis to remain. Part of the problem is that even with our recent increases in gas prices, we have the cheapest gas prices in the world and in inflation adjusted dollars, we are still below the highs seen the past (though with two V-8's and a V-12 in my garage, I certainly feel the pain today). An interesting side note with all the talk about hydrogen as a fuel source, the most promising (i.e., currently most economic) source of hydrogen is from using fossil fuels (i.e., natural gas) as the feed source!


[Modified by TexasRedLT4, 7:13 PM 6/9/2004]

Runge_Kutta

Todd,

I think your comment "The GDI is being eyed by GM and many other manufacturers
mostly for benefits in emissions." is off the mark. The attraction of GDI is the
increased thermal efficiency and the snag is primarily NOx emissions.

Duke,

The generic analysis I saw showed a 12% gain. The GM study showed 15%-26%
gains in an engine that was not wildly different from an LS1. Given the climate
of 2004 and what likely follows, I suspect it will be embraced by all major car
makers. Central injection, air-assisted, LPDI seems like the way to go. I think
part of why much has changed is the advent of big computers and CFD codes like
KIVA from Los Alamos NL. Shrinking oil reserves and air pollution also place
GDI in a new light.

WhiteDiamond

Agreed. I have been looking at the regenerative fuel cell technology mostly and they are currently trying to make up for the lost in effeciency between energy needed to split the hydrogen from, say, water and the energy recovered (typically around 40% now) with solar power. Improved battery and solar cell performance will be the only way auto makers start to look seriously at fuel cells.

Duke,

I know you have a lot of information, but I did not know you worked for GM at one time. I may IM you at some point. I have been interested in the big auto makers versus small funded research lately.

GDI does not offer the huge performance gain, I agree, but it seems that the makers are very concentrated in pulling all the little gains they can and emissions regulations are tightening. Every little bit helps and with the recent revelation that the "hybrids" can't come close to their gas milage claims in the real world(reports are 20 to 25% below reported EPA estimates in real world driving), it is obvious that these little gains are needed.

I am really curious on how somebody like GM does a cost analysis on something like GDI. That has to be an interesting discussion.

Todd

SWCDuke

I worked for GM early in my career, which was some time ago, but the basics of cost analysis have probably not changed. There have always been problems bringing lab techology to production, and these usually revolve around high volume production at a reasonable cost and the capital investment to pull it off. There is lots of risk! The issues constantly wax and wane, and nowadays, government regulation is as big driver as is the market.

Many higher cost Corvette ideas have been rejected over the years - most notably a mid-engine Corvette, because the high cost would have raised its price point in the market, and the marketeers did not want to do that. IMO that was a GOOD call!

GDI is a promising technology, but to bring it to mass production at the lowest possible cost requires HUGH investments and the decision process can be agonizingly convoluted and slow. In the past GM was very innovative, but some of their innovation got them in trouble. They then went through a period in the seventies and eighties when they just quit innovating, and it showed in the products.

They seem to have regained their sea legs, but they are still conservative about new technology. Low volume models are certainly helpful, especially if they can command a high price in the marketplace, as they allow new technologies to be tested with less risk than committing it to a high volume model.

The Corvette has traditionally been in this role with fuel injection, retractible headlights, IRS, disk brakes to name a few. The Corvette is a much more mainstream product, nowadays, with a much older and more convervative customer base than in the fifties and sixties. Just look at all the howling over the exposed headlights! I think GM would be more reluctant to use it as a "test platform" for new technology. Consider the newC6, which is a refined version of the C5. Since the C5 was such a success the redesign was conservative, and the basic marketing plan has not changed, but consider that C5 was very innovative in terms of both design (chassis architectue) and manufacturing (hydroformed frame rails.) You don't want to kill the goose that lays the golden eggs. That's what happened to the Vega because the engine technology wasn't quite fully baked before it went into mass production and a couple of decisions to save a few dollars of production cost turned the early models into rust buckets more quickly than most cars of the same era.

If you have other questions feel free to e-mail me.

Duke

Runge_Kutta

United States Patent Application 20040107695
Cho, Byong Kwon; et al.
June 10, 2004

Ultra low power plasma reactor system for automotive NOx emission control

Abstract

A sidestream located hyper-plasma reactor having an axially discrete pattern of
alternating regions of active and passive electric field along the axial direction.
The hyper-plasma reactor has great efficacy in terms of ultra low power consumption
and copious production of NO.sub.x converting aldehydes in the absence of NO by
applying plasma power only to an air and hydrocarbon mix sidestream gas flow.
Only a small fraction (1% to 2%) of plasma power is required as compared to that
for a conventional plasma reactor to treat the full exhaust gas stream. The
hyper-plasma reactor produces ozone which reacts subsequently with hydrocarbons
to produce aldehydes ("ozonolysis"). The sidestream location of the hyper-plasma
reactor allows for the full exhaust stream to bypass it, without significantly
affecting the overall NO.sub.x conversion performance in the catalytic converter.

Inventors:
Cho, Byong Kwon; (Rochester Hills, MI);
Schmieg, Steven Jeffrey; (Troy, MI);
Oh, Se Hyuck; (Troy, MI)

General Motors Corporation

Filed: December 6, 2002

BACKGROUND OF THE INVENTION

[0002] The removal of nitrogen oxides (NO.sub.x) from internal combustion exhaust
is an increasing concern, especially for lean-burn engines such as direct injection
gasoline engines and Diesel engines. One method for post combustion NO.sub.x
removal is the subjection of the exhaust gases to a non-thermal plasma process.
In this regard, the exhaust gas is passed through a plasma processing device
whereat a high voltage electric field imparts formation of a plasma. The plasma
has a large number of energetic electrons which collide with exhaust gas molecules
to form atoms, ions and radicals. These atoms, ions and radicals, in turn, react
either with the NO to make NO.sub.2 or with hydrocarbons to produce aldehydes.
The produced aldehhydes subsequently reduce NO.sub.x over suitable catalysts to
make harmless nitrogen. Thus, the major role of the plasma reactor is to produce
NO.sub.2 from NO and aldehydes from hydrocarbons in the combustion exhaust stream.
Among the aldehydes produced in the plasma reactor, acetaldehyde (CH.sub.3CHO)
is known to be the most effective for NO.sub.2 reduction over alkali-based catalysts.

Runge_Kutta

http://www.sae.org/automag/techbrief...1-112-7-52.pdf