Tenneco Acquires Mobile Emissions Business of CCA
Louisiana Gets Its First Retail E85 Station

GM Demonstrates Gasoline HCCI On the Road

GM has demonstrated homogeneous charge compression ignition (HCCI) for the first time in two driveable concept vehicles, a 2007 Saturn Aura and Opel Vectra. The HCCI  gasoline engines operate in mixed mode, using HCCI at lower loads, and reverting to spark-ignition at higher loads and speeds.

When combined with the enabling advanced technologies such as direct injection, electric cam phasing, variable valve lift and cylinder pressure sensing, HCCI combustion provides up to a 15% fuel savings over conventional gasoline engines, while meeting current emissions standards.

I remember debating the limits of combustion capability when I was in college. HCCI was just a dream then. Today, using math-based predictive analysis and other tools, we are beginning to see how we can make this technology real. By combining HCCI with other advanced gasoline engine and control technologies we can deliver a good fuel savings value for consumers.

—Tom Stephens, group vice president, GM Powertrain and Quality

In an integrated engine concept, HCCI, along with other enabling advanced technologies, approaches the engine efficiency benefit of a diesel, but without the need for expensive lean NOx after-treatment systems. Its efficiency comes from burning fuel at lower temperatures and reducing the heat energy lost during the combustion process. Consequently, less carbon dioxide is released because the vehicle’s operation in HCCI mode is more efficient.

The HCCI-powered concept vehicles—a production-based Saturn Aura and the Opel Vectra, both with a modified 2.2L Ecotec four-cylinder engine—drive like conventionally powered vehicles, but offer up to 15% improved fuel efficiency relative to a comparable port fuel-injected engine. The fuel efficiency improvement will vary depending on the vehicle application and the customer driving cycle.

The driveable concept vehicles represent some of the first tangible demonstrations of HCCI technology outside of the laboratory.

Highlights of HCCI technology include:

  • Diesel-like engine efficiency with substantially reduced after-treatment cost;

  • Builds off proven gasoline direct-injection and variable valve actuation technologies;

  • Adaptable to conventional gasoline engine architectures;

  • Requires only conventional automotive exhaust after-treatment;

  • Compatible with all commercially available gasoline and E85 ethanol fuels.

An HCCI engine ignites a mixture of fuel and air by compressing it in the cylinder. Unlike a spark ignition gas engine or diesel engine, HCCI produces a low-temperature, flameless release of energy throughout the entire combustion chamber. All of the fuel in the chamber is burned simultaneously. This produces power similar to today’s conventional gas engines, but uses less fuel to do it.

Heat is a necessary enabler for the HCCI process, so a traditional spark ignition is used when the engine is started cold to generate heat within the cylinders and quickly heat up the exhaust catalyst and enable HCCI operation. During HCCI mode, the mixture’s dilution is comparatively lean. The lean operation of HCCI helps the engine approach the efficiency of a diesel, but it requires only a conventional automotive exhaust after-treatment.

HCCI builds on the integration of other advanced engine technologies, some of which are already in production and can be adapted to existing gas engines. The cylinder compression ratio is similar to a conventional direct-injected gas engine and is compatible with all commercially available gasoline and E85 fuels.

GM demonstrated the adaptation of the HCCI technology in driveable concept vehicles based on conventional, production-based products like the Saturn Aura and Opel Vectra. The Aura features an automatic transmission; the Vectra, which is aimed at the European market, has a manual transmission.

Both vehicles are powered by a 2.2-liter Ecotec engine (180 horsepower [134 kW] and 170 lb.-ft [230 Nm] of torque) that features a central direct-injection system, with variable valve lift on both the intake and exhaust sides, dual electric camshaft phasers and individual cylinder pressure transducers to control the combustion as well as deliver a smooth transition between combustion modes.

A sophisticated controller, using cylinder pressure sensors and GM-developed control algorithms, manages the HCCI combustion process, as well as the transition between HCCI combustion and conventional spark-ignition combustion. The transition between the combustion processes is notable in the demonstration prototypes, but production versions are intended to deliver an imperceptible transition while driving, similar to the deactivation performance of GM’s Active Fuel Management system.

Currently, the GM demonstration prototypes can operate on HCCI up to approximately 55 mph, transitioning to spark ignition at higher vehicle speeds and during heavy engine load. An extended range for HCCI operation is intended as further refinements to the control system and engine hardware are made.

Perhaps the biggest challenge of HCCI is controlling the combustion process. With spark ignition, you can adjust the timing and intensity of the spark, but with HCCI’s flameless combustion, you need to change the mixture composition and temperature in a complex and timely manner to achieve comparable performance.

—Prof. Dr. Uwe Grebe, executive director for GM Powertrain Advanced Engineering

GM’s global HCCI team will continue to refine the technology in the wide range of driving conditions experienced around the globe, from extreme heat and cold to the thin air effects of driving at high altitude.

In 2005, GM announced a three-year, $2.5-million research program with supplier Robert Bosch and Stanford University to accelerate development of HCCI. (Earlier post.)



What makes this car GREEN? All I see is another ICE vehicle that has been improved by 15%. Big deal. It's almost as bad as the Audi V12 on here. It's not a hybrid or anything else related to a green vehicle. Why is it on this website?


HCCI up to 55 mph - does that mean they have HCCI up to the first ~15 hp of the 180 hp engine?

Rafael Seidl

@ Curtis -

it's on this web site because 15% improvement in fuel economy is definitely nothing to sneeze at. HCCI mode combustion will be incorporated into ICE and controller designs at acceptable cost in coming years and eventualy be applied to many millions of vehicles. And its the sheer number of vehicles sold that equipped with fuel-saving technology that makes all the difference. Small fry really does add up.

Full hybrids also feature ICEs and also save just 15%, so they are no cleaner than this. And saying that HCCI is almost as a V12 is just a complete non-sequitur. Maybe you need to think a little about cars in terms of black boxes with inputs and outputs rather than getting stuck on any particular definition of green technology.


"but offer up to 15% improved fuel efficiency relative to a comparable port fuel-injected engine. The fuel efficiency improvement will vary depending on the vehicle application and the customer driving cycle."

So what does this 15% represent? Drive cycle, single point or theoretically dreamed up number? The statement is meaningless without qualification.

Grebe is the honest one when he says: "Perhaps the biggest challenge of HCCI is controlling the combustion process..."

Note also the claim of Diesel like efficiency - but no turbocharger. The throw away line of only needing variable camshaft phasing, variable valve lift, cylinder pressure sensors and direct injection for HCCI is a bit of an understatement considering the cost, complexity and likely reliability issues.

It's great progress but you wouldn't be banking on it making production soon in large volume.

Rafael Seidl

@ Clett -

HCCI requires very high EGR rates, which obviously reduces the amount of fresh charge in the cylinder a.k.a. the effective displacement. The combustion noise levels and mechanical stresses due to near-isochoric combustion limit the torque levels up to which HCCI can be applied. The risk of misfires limits the engine speed up to which it can be applied.

Finally, gasoline HCCI is a lean burn concept, so engine-out emissions have to be ultra-low because the three-way catalyst cannot reduce NOx in the presence of free oxygen. At some power level (actually, close to the stability limits), HCCI can no longer meet NOx emissions targets.

In summary, yes, HCCI can only be used in low part load. In the lab, researchers can go higher than the ~10% of rated power that you suggest but we'll have to wait and see how aggressive car makers will be in series production. However, LDVs spend much of their lives puttering away in low part load. It's only when you accelerate hard, drive fast or climb a hill that you actually use more of your engine's power - typically, still just 30-50% of it, though.

@ extracomment -

you may be right for the US market, where gasoline is relatively cheap and CAFE next to useless because of the E85 loophole. In Europe and perhaps Japan, we could see initial applications (cp. M-B DiesOtto) in the 2010-2015 time frame. Ergo, "soon" is a relative term.

John Schreiber

This is an excellent development, and I am glad to see that GM is on it. Does anyone know how long an engine must operate before HCCI mode will operate, and will the heat requirement be compatible with idle stop strategies?


The only disappointing thing I see in this report is the 55 MPH limit in HCCI mode. A previous report on Green Car, Oct 2005, implied that the Honda engine could run in HCCI mode up to 4,000 RPM which would translate to vehicle speeds well above 70 MPH. Also the 2.2 liter engine is rather small for a Saturn Aura. There is an existing 2.4 liter Ecotec engine that would probably be a better choice for the Saturn Aura.
The Mercedes people stated that their Otto/Diesel engine would be production capable in the medium term which is believed to be about 5 years. Do you see that as a reasonable time period until production engines are available?


So what you're saying is: This engine is disapointing because it's not green enough. Oh and it's too small too. Because clearly increasing engine size by nearly 10% is very green. The whole point of these modern engine technologies is to improve part load efficiencies and to improve the specific power output, thus obviating the requirementt for a 2.4l engine in a "small" car.


Interesting, hopefully we'll see it in the Chevy Volt as the 55mph limit shouldn't be a factor in a generator scenerio.


Good news. Any idea when it will be out in production vehicles and how much it will add to the cost?

Rafael Seidl

@ rca -

the speed of the vehicle and the speed of the engine are coupled by the gear ratios of the transmission. It's entirely possible that the Japanese have better mastery of HCCI combusion control than GM does, but 4000RPM on the engine doesn't imply squat about vehicle speed.

It's time to stop judging engines primarily on the number of cylinders, total displacement and rated power. What matters at least as much for everyday driving is available low-end torque (incl. turbo lag, if applicable) and, the transmission type and gear ratios. A modern downsized SI engine with a small TC may be almost as driveable but cheaper to buy and run than a larger NA engine with similar rated power.

3-7 years sound like a reasonable timeframe for the first HCCI engine to go into series production, provided the price of oil stays high enough. The fundamental control problems have been solved. Now, it's a question of making the systems robust enough, e.g. wrt sensor malfunctions, fluctuations in fuel quality, ambient air conditions etc. Also, total system cost remains high, so expect this to be introduced in low-volume high-margin vehicles first.

hampden wireless

HCCI and hybrids are going to mix well. The engine computer should be able to maximize lean burn mode easily on a hybrid.


This research is interesting and has been a goal of mech' engineers for some time now. The idea of high compressive combustion using gasoline type fuels has been an appealing goal. GM deserves credit for also demoing it in a prototype car. There is lots that's positive here. Seeing as how it appears to run in its best clean mode and power at lower rpm, it might be suitable as a small efficient genset in a serial HEV or run as the cruise engine in a parallel HEV or PHEV. Toyota uses an Atkins cycle 1.5 liter engine for their HEV. With 15% more efficiency, the HCCI engine might be a good replacement as it could be made even smaller and thus lighter and could run in full HCCI mode because of the narrow rpm band needed. Ever how appealing all this is, I'm still standing my ground and waiting for the affordable BEVs to come to town.

Bob Bastard

This engine sounds like it should have sufficient power for just about any LDV, other than specialty high performance apps. My heavy duty pickup has a 5.7 liter V8, which actually has lower peak hp although the peak torque is slightly higher than this 2.2 liter 4 cylinder. It would be interesting to get an idea, based on the current technology used, what the cost differential for production would be compared to several diesel and gas hybrid designs offering similar levels of power and fuel economy, assuming for a minute, that the technical hurdles were overcome.

Rafael Seidl

@ hampden wireless, Lad -

the fuel economy gains from a mild or full hybrid stem mainly from load leveling. In situations - e.g. low-speed cruising - in which conventional drivetrains would be operating inefficiently in low part load, a hybrid will articially increase the load to recharge the battery. This energy is later used for a spell in all-electric mode or, to boost torque. The latter feature lets the drivetrain engineers use a smaller and/or phlegmatized (i.e. Atkinson- or Miller-cycle) ICE, which reduces pumping losses.

Ergo, the ICE in a mild or full hybrid spends little time in the are of the engine map in which HCCI could be used. Therefore, the two concepts actually do NOT complement each other well at all.


Rafael, I've heard you explain this before, but something just doesn't sound right here. It is apparent that your technical knowledge exceeds my own, but what you described does not seem to perfectly translate to what I have observed from hybrids such as the Lexus RX400h, which I have driven.

I understand this is only one particular sample, but in that case, the Lexus seems to be programmed to utilize only the minimium required amount of power from the ICE as is needed at that moment. That ICE requirement obviously increases as your speed increases or if the driver accelerates faster, but this artificial load that you refer to, while it probably exists, seems like a very small percentage increase.

This is especially evident when cruising at highway speeds over longer distances, where the ICE loads are already pretty evenly balanced. It seems to me that it would be unlikely that the hybrid would be programmed to increase those loads to a high degree. There is only a finite amount of electrical storage available. Additionally, wouldn't it be more efficient to use the direct mechanical connection to put the required power down on the road than generate surplus electricity and convert it back to mechanical energy?

Because of that, I would have to think that HCCI would still increase highway fuel economy in a parallel, full hybrid. Would it be fairer to say that HCCI just would not add as much value for stop and go, city driving? If all of this were true, it would address a major shortcoming that prevents many people from purchasing hybrids - their much smaller increase to highway fuel relative to the city increase.

Rafael Seidl

@ Angelo -

I was referring to hybrid drivetrains designed to save fuel, such as the one in the Prius. In those layouts, the generator load is quite significant, much more than a few percent of the traction load.

The RX400h is more of a performance hybrid featuring both a a powerful ICE and a powerful electric system. Plus, its a 4x4 with all-electric drive on the rear wheels. Since fuel economy isn't the primary consideration, other factors such as cabin noise levels may have prompted Lexus to pursue a different control strategy. Whether it would benefit from HCCI, I cannot say, I don't know enough of the details of the RX400h. I doubt Lexus would go that route, though. A full hybrid system is already very expensive.


Who knew: advances in HCCI engines coming from Detroit? Yes, it's still 3-7 years away from the mainstream consumer- but it is a giant leap in the right direction. Bravo GM!

I might have to buy some GM stock after all.


I can see where the load on the ICE can be considerable while underway and in charge mode. Given all the above information it begs the question: what is the best practical combination of power for a hybrid to achieve maximum efficiency?

Toyota has recently announced a mild increase in the Prius battery capacity with plug in capability, a rather anemic approach designed for evolutionary technical advances and maintaining profits, an interim move to the next model year. I'm sure there are many solutions alive under the patent laws cloud; what's your favorite? If I read the tea leaves correctly all the LDVs, HEVs and PHEVs still point to the long-range, quick-charge BEV or some form of Hydrogen driven vehicle as the most likely winner.

Hurry up with those batteries guys!

Max Reid

Any tech that helps in reducing fuel consumption will appear in this site, definitely this Aura is greener than gas-guzzling hybrids like Highlander, Rx400, GS450h and the coming LS600h.

Honda has taken a big beating with their Hybrid Accord V6 and pulled it out, no one know when Toyota is going to take the beating with thier gas-guzzlers.

Roger Pham

Don't underestimate the value of this HCCI 15% fuel-saving.
Most over-powered LDV's (probably 80% of vehicles sold in the market today) operates most of the time at below 10% of their rated power output. Most modern cars already has VVT on the intake side. Adding another one on the exhaust side and cylinder pressure transducer won't amount to much cost increase if mass produced. HCCI can obviate the need for drastic engine downsizing with turbocharging. A lot of fuel can be saved if most non-hybrid cars in the future will be capable of HCCI.

I don't understand why direct injection (more expensive) would be needed for HCCI to work. The high compression ratio needed for HCCI can be solved by variable compression ratio via VVT on intake side. This will reduce maximum output some because Atkinson cycle will be needed on the SI mode to prevent knock, but cost saving by using port-injection can be realized.

Rafael Seidl

@ Lad -

if - very big IF - battery technology advances to the point where it delivers adequate safety, performance and price, it will indeed be a disruptive technology and, a welcome one at that for urban commercial and commuter traffic. However, given that ICEs displaced BEVs a century ago and have dominated the entire road transportation market ever since, it would be foolish to bet the farm on innovations in batteries alone. The devil is always in the details.

Fuel cell vehicles sound enticing because they cause zero *tailpipe* emissions except water vapor. So far, however, the only economically viable processes for producing hydrogen - an energy carrier not source - are steam reformation of natural gas and electrolysis based on cheap electricity (i.e. coal or nuclear). Producing emissions somewhere else is NOT zero emissions, especially if their impact is global in nature. The technical problems associated with storing hydrogen on board a vehicle also remain unsolved. Plus, PEM fuel cells remain expensive and bulky, because their low operating temperature requires the use of very large radiators. Personally, I expect the whole FCV effort will turn out to be a huge white elephant.

Fwiw, I wouldn't trust anyone who feels they can say today what the 2 billion cars expected to be on the world's roads by 2050 will be powered by. There is only so much oil in the ground. Natural gas could take up much of the slack, either as CNG or in the form of liquid synthetic fuels, IFF anthropogenic GHG turn out to be less of a problem than scientists currently claim they are. That is an optimistic scenario, though.

For now, the only thing we can do is tinker on all fronts simultaneously and wait for the technology shakeout to occur later. Moreover, we need people to think more holistically about the transportation of data/information, people and goods and what it means to our economy and quality of life. Personal cars are great, but they do have serious downsides. Novel services and urban architecture could actually make NOT having to own a car something we may one day aspire to. Just not anytime soon ...

Bob Bastard

Roger, I'm not sure that DI is needed for HCCI, but I think it was already a feature of the Ecotec engines. I'm guessing that DI helps provide tighter control over the combustion event.


GM should be congratulated. Now with Mercedes-Benz DiesOtto engine HCCI looks like it is more than just a laboratory experiment.

Do have a little disagreement with Rafael concerning complementing HCCI and Hybrid motors. Certainly, the value of HCCI would not be a great value in a Prius or Plug-In Hybrid, but this would not necessarily be the case with the Honda Hybrid and it’s Integrated Motor Assist (IMA) which has several modes of operation that would work well with HCCI (see http://world.honda.com/news/2005/4050705_b.html for more details)

Also, have not seen any recent news about the latest developments on Honda’s HCCI on the Green Car Congress web site. I was able to “google” the Honda Japanese web site and found that Honda had been awarded on May 22, by the Society of Automotive Engineers of Japan an award for their 1.8 liter i-VTEC engine for:

“making new EGR supercharging concept in the gasoline compressed ignition engine which adopts electromagnetic type unrestricted valve timing mechanism, HCCI (*) it proved the fact that the operating range can be expanded with apparatus engine, brought close to utilization furthermore.”
(Details on http://www.honda.co.jp/environment/e_news2/070522.html).

Other information on this engine can be found in the Honda U.S. patent 6817349. Maybe Green Car Congress could add some more information on this also and when we might see it in the next Honda Civic Hybrid or Insight replacement.

Stan Peterson


I think your analysis of how the world proceeds, addressing the fuel problem for Ground Transport, is spot on. From my previous posts you know that my research indicates that the pop science of AGW is mostly political tommyrot. But that does not mean that there is enough fossil to waste burning it, or that the process of cleaning it, does not bear costs.

I agree with your technical viewpoint. "Let a thousand flowers bloom", and then use the "prettiest" technology(s) that are the "best" from an efficiency, cost, and emissions cleanliness standpoint.

I also agree that HCCI does not particulary lend itself to a Series Parallel hybrid like the Prius, unless the control algorithms are altered somewhat. But it is not particularly inimical to such usage either.

But it would be very complementary and lend itself to the Chevy Volt Series hybrid type. In that setup you have a very small sustainer engine operating at near constant RPM to spin a generator. An ICE could be planned and optimized to be in the HCCI band when doing so, as it would for all but a few percent of its operation. In addition, the stop-start issue would be reduced, so cylinder warm up and cool down, and accompanying catalytic converter heat up issues, would be largely negated and bypassed.

Hence the sustainer-generator ICE, of a Series Hybrid could be very clean and very efficient, operating in the HCCI band.

The comments to this entry are closed.